TERENO International Conference 2018

Affiliation: Hydrology and Water Resources, University of Arizona, Personal website

Cosmogenic neutrons can be used to measure soil water content in the root zone of soil for applications in field and catchment-scale hydrology, land surface modelling, and agricultural water management. The method takes advantage of the strong sensitivity of neutrons with energies greater than 1 eV to hydrogen present in and above soil. Hydrogen removes those neutrons from the environment in the process called moderation resulting in their reduced density in air above the soil surface. The resulting inverse correlation between the neutron density in air and the hydrogen density in soil is used to convert measured neutron intensity to soil water content. Accounting for hydrogen sources other than soil water is necessary for accurate computation of soil water content.
Neutrons are measured with the instrument called cosmogenic neutron sensor or probe. The probe’s sensing element consists of a thermal neutron detector surrounded by a moderator. The probe’s horizontal footprint is measured in hectometers and the vertical footprint in decimeters. Both depend on soil water content and other hydrogen present in and above soil. The probe can be installed permanently (stationary probe), giving time series of neutron intensity and soil water content, or moving (rover), used for mapping of neutrons and soil water content along lines or over areas. Networks of stationary probes exist, the largest being in Germany (TERENO), Australia (CosmOz), the UK (COSMOS-UK) and the USA (COSMOS). The aim is to analyze data from all locations within the network together and link the results to large-scale hydrological, meteorological, land-surface and atmospheric models. That is yet to happen.
Recent work showed that the horizontal footprint is more complex than hitherto thought: the probe has very high sensitivity to meter-sized heterogeneities. That finding led to the development of two new instruments. One reduces the sensitivity to local neutrons and improves wide-area sensing of soil moisture. The other instrument increases the sensitivity to local neutrons, reduces the contribution by far-away neutrons and becomes a small-footprint sensor for high-resolution mapping of soil moisture.

Affiliations: 1 University of Rostock; 2 Leibniz Institute for Baltic Sea Research Warnemünde; 3 University of Kiel; 4 University of Bremen

Abstract: Coastal zones are unique environments connecting terrestrial and marine ecosystems. Here, we assess the mutual interactions in the land-sea interfacial coastal zone on subterrestrial water characteristics. The study area is located at the southern Baltic Sea, that was artificially drained by ditches for decades and is now in a rewetting stage. The peat is underlain by fine sands, forming a regional heterogeneous aquifer, and glacial till. The peat layer reaches out into the Baltic Sea. In the present study the peatland and the Baltic Sea coastal sediments were equipped with groundwater monitoring stations and deep permanent pore water lances that were sampled on a monthly basis. The salinity in the groundwater underneath the peatland was 4-5 compared to 0.3 in the surrounding forest, indicating that it is a mixture of fresh and seawater. The seawater component may originate from former flooding of the peatland, since subsurface intrusions of seawater have recently only been observed locally behind the dune upon storm surges. Dating using the tritium-nobel gas method showed that the groundwater age increased from the border of the peatland towards the beach, where complex mixing patterns occur. In the peatland, a sharp increase in salinity within the first top meter of the soil was observed, indicating a gradual freshening from surface waters. In geoelectric profiles, a high-resistivity layer was found underneath a low-resistivity layer, which probably depicts brackish water in the sand layer overlain by fresh water in the peat. In the shallow Baltic Sea, pore waters salinities down to 2 were observed due to submarine groundwater discharge. Strong differences in temporal and spatial dynamics along the coastline are caused by the complex stratigraphy. Our findings reveal a strong impact of land-sea interactions on pore water characteristics and a pronounced legacy effect in the peatland.

Affiliations: 1 Leibniz Institute of Freshwater Ecology and Inland Fisheries / Freie Universität / University of Trento; 2 Leibniz Institute of Freshwater Ecology and Inland Fisheries, IGB-Berlin; 3 University of Trento

Abstract: The complex spatial patterns of hyporheic exchange flows across the sediment-water interface of streams are a challenge for traditional measurement approaches. The present study applies Fiber Optic Distributed Temperature Sensing (FO-DTS) to identify groundwater discharge, interflow discharge and local down- /upwelling of surface water based on their distinct thermal patterns at the sediment-water interface. However, the similar thermal pattern of groundwater discharge and deep hyporheic flows requires additional investigations. Geophysics can provide evidence of subsurface structures. Geophysics can distinguish connected areas of permeable sediment from discontinuities in the streambed limiting groundwater discharge. In particular, electromagnetic induction (EMI) enables quick and economic recognition of streambed structure based on texture changes. Results of our study show that strong permanent temperature anomalies at the sediment-water interface are located over homogeneous coarse sediment structures. It is likely that groundwater discharge occurs preferentially in these areas. Thus, the combined interpretation of EMI and FO-DTS data has potential to provide evidence of groundwater-surface water connectivity and allows differentiation of different origin of groundwater-surface water exchange.

Affiliations: 1 GFZ German Research Centre for Geosciences

Abstract: Transpiration and its spatio-temporal variability are still not fully understood, despite its importance for the global water cycle. This is in part due to our inability to measure transpiration comprehensively. Transpiration is usually either estimated with empirical equations based on climatic variables and crop factors, by measuring sap velocities, estimating sap wood area and scaling up to the forest stand based on a number of assumptions or by measuring the integral signal across a footprint with eddy flux towers. All these methods are focused on the total loss of water to the atmosphere and do not provide information on where this water is coming from. In this study, spatio-temporal variability of root water uptake was investigated in a forest in the northeastern German lowlands (TERENO Observatory NorthEast). We estimated root water uptake from different soil depths, from 0.1 m down to 2 m, based on diurnal fluctuations in soil moisture content during rain-free days. We used the physically based model HYDRUS to test assumptions and estimate uncertainties of this soil moisture based estimation of plant water uptake. The 15 field sites cover different topographic positions and forest stands: 4 pure stands of both mature and young beech and pine and 9 mixed stands. The resulting daily data set of root water uptake shows that the forest stands differ in total amounts as well as in uptake depth distributions. Temporal dynamics of signal strength within the profile suggest a locally shifting spatial distribution of uptake that changes with water availability. This unique data set of depth-specific contributions to root water uptake allows for a much more detailed analysis of tree species specific response to water availability than the more common transpiration estimates generated by sapflow or eddy flux measurements, and therefore has the potential to improve models of the soil-plant-atmosphere continuum.

Affiliations: 1 Université Grenoble Alpes, IGE, 38000 Grenoble, France; 2 Université Grenoble Alpes, CNRS, IRD, Grenoble-INP, IGE, 38000 Grenoble, France; 3 IRSTEA, UR RiverLy, centre de Lyon-Villeurbanne, 69625 Villeurbanne, France

Abstract: Flash floods are amongst the most important natural hazards in Europe, particularly in the Mediterranean region creating economical damage and life risk. This study investigates the hydrological response to rainfall events over the period 2012-2015 on different spatial scales for 13 small to meso-scale partially nested Mediterranean catchments of 0.17 km2 to 42 km2 in the Ardèche region. The Claduègne catchment, which contains all the other catchments, is part of the OZCAR network. This catchment can be separated into two distinctively different parts with a basaltic plateau in the upstream part and sedimentary rock in the downstream part (each about 50%). The sub-catchments are distributed over these two parts and different land use forms (forest, cultivated soils and grasslands). For most of the sub-catchments, no discharge rating curve exists, so the focus of the analysis mainly lies on the time component of the hydrological response as well as on descriptors of the rainfall-runoff event and catchment properties. Several characteristic times were calculated for each catchment based on a set of 49 to 158 events depending on the catchment and various statistical methods were applied in order to estimate the importance of different factors on the hydrological response on two levels: (A) inter-comparison of the catchments with regards to their properties and averaged event characteristics. (B) comparison of the storm events for each catchment considered separately with regards to the rain characteristics, initial degree of wetness and saisonnality. Besides the catchment’s size, we found the geology to be an important factor influencing the catchments’ response, with the catchments in the sedimentary part reacting faster. The precipitation was shown to be typically more intense for short time steps (e.g. hourly) in the sedimentary part of the catchment. On the catchment-level, a strong seasonal variation of event characteristics was found. The most intense, shortest events take place in autumn, while the events in spring are particularly long and slow. The degree of wetness is generally much higher in winter than in summer. Furthermore, two sub-catchments have singular behaviors. One was shown to react much faster and one much slower than expected. This suggests in-situ investigations in order to understand the hydrological mechanisms. A simple estimate of lag time was shown to be able to characterize a catchment’s dynamic and provide an estimate of time of concentration, if calculated for a selection of impulsive events.

Affiliations: 1 CNRS; 2 Université de Rennes

Abstract: Aquatic ecosystems are essential components of the environment which contribute to biodiversity and provide various ecological services. In particular, wetlands are typical hotspots defined by the close interaction between different water sources (surface and ground water) controlling both ecosystem structure and processes. Therefore, quantifying water and element fluxes, the mixing of different water bodies and the impact of ecological processes is essential to understand the ecosystem dynamics and resilience. Here, we show how high spatial resolution of dissolved gas data can decipher the spatial heterogeneity of water sources and biogeochemical processes. Such datasets were acquired using a CF-MIMS (Chatton et al, 2017) designed and adapted to the field within the CRITEX project (Innovative equipment for the Critical Zone) and implemented on a classified coastal wetland (Lannenec pond) downstream a hydrogeological observatory (H+ Ploemeur). This innovative experiment allowed to localise and quantify groundwater discharges (oversaturation of He and CO2), and biogeochemical reactivity hotspots (oversaturation of O2 and undersaturation of CO2). Finally, this study approaches the concept of “operational hydrology” and contributes to the enhancement of both the spatio-temporal distribution and the quality of environmental data for a thorough exploration of the Critical Zone. Chatton, E., Labasque, T., de La Bernardie, J., Guiheéneuf, N., Bour, O. and Aquilina, L.; Field Continuous Measurement of Dissolved Gases with a CF-MIMS: Applications to the Physics and Biogeochemistry of Groundwater Flow, Environ. Sci. Technol. 2017, 51, 846−854, DOI: 10.1021/acs.est.6b03706

Affiliations: 1 IPG PARIS; 2 IPGP Paris; 3 IRSTEA Antony

Abstract: Concentration–discharge relationships (C-Q) of river water is one powerful tool to track the coupling between water flow and chemical reactions in the Critical Zone. In the framework of the CRITEX equipment of excellence program, we have developed and installed a prototype called River Lab (RL) measuring the major elements dissolved in river water every 40 minutes for long period of times. Confined in a bungalow, the RL performs an analysis of all major dissolved species (Na, Mg, Ca, Cl, SO4, K, and NO3) using ion chromatographs, through continuous sampling and filtration of the river water. We will present the concentration-discharge relationship (C-Q) recorded during the exceptional hydrological year 2015/2016 measured in the Orgeval CZO (OZCAR RI), France. As already observed in catchments around the world, observed C-Q relationships follow a power-law relationships (C=AQ-b) with concentrations decreasing when discharge increases, but not according to a dilution curve, a process known as the chemostatic behavior of rivers. Observed C-Q depend on the time resolution (sampling frequency) and span (e.g. flood vs. drought) over which data are acquired. In more details, the C-Q relationships describe hysteresis loops whose shape evolves with time. Our results show the specificity of each event, which is in depending upon the flood history. We show that fitting of C-Q relationships with a unique power-law fit presents some limitations. The exponent b calculated from the power-law fit is not constant over the year and varies during a single flood event. We propose an interpretation based on a ”grey box” approach such as already developed for stream flow (Kirchner, 2009) but here extended to solute fluxes. This model allows us to reproduce the diversity of C-Q relationships and requires that the contribution reservoirs are evolving with time. Kirchner, J.W., 2009. Catchments as simple dynamical systems: Catchment characterization, rainfall-runoff modeling, and doing hydrology backward. Water Resour. Res. 45, 1–34. doi:10.1029/2008WR006912

Affiliations: 1 Université Grenoble Alpes, CNRS, IRD, Grenoble-INP

Abstract: The RIPLE platform has been developed to allow a continuous monitoring at high temporal frequency (~10 min) of water and solid fluxes in mountainous rivers. The scientific context of this development is defined as follows: (i) the simultaneous measurements of water discharge, bedload, suspension load and river bed topography contribute to the establishment of comprehensive mass balance at the catchment scale; (ii) measurements of the physical properties of fine sediments (size, shape, composition) provide information on the spatial origin of sediments within the catchment, the conditions for erosion and sedimentation processes within the river and the potential to transport other substances such as contaminants, nutrients, microorganisms. For the design of the platform, priority has been given to non-intrusive instruments due to their robustness. The basic prototype of the platform integrates the following instruments: water level and surface velocity radars, turbidimeters, conductivity probe, hydrophone, cameras, automatic water sampler and depth sounder. Other instruments are progressively integrated, such as the SCAF (system characterizing the sediment’s settling velocity), an acoustic Doppler profiler and a scanning laser meter. A wireless telecommunication has been set up to allow remote interactions with the platform and data transmission. The RIPLE platform has been designed to facilitate its use and maintenance: user interface allowing data monitoring and remote configuration, sending alerts (SMS, mail) according to programmed conditions, flexibility of on-site installation and energy autonomy allowing to easily move the platform from one site to another site. In September 2016, the RIPLE platform was installed on a bridge across the Romanche river at Bourg d’Oisans (45.1159° N, 6.0135° E) for a testing period. After a presentation of the architecture of the platform, the first results derived from in situ measurements are discussed. The next steps in the exploitation of the results of the RIPLE platform are finally addressed.

Affiliations: 1 Forschungszentrum Juelich

Abstract: To investigate and monitor soil moisture dynamics in detail caused by infiltration of water in soil, we used horizontal subsurface and on-ground GPR at the experimental rhizontron facility in Selhausen test site. The GPR antennas can be positioned at the surface or specific depths in horizontal boreholes to investigate the soil properties and states in detail. By analyzing the propagation of direct wave for horizontal borehole GPR, the relative permittivity and electrical conductivity of the soil at this position can be obtained. However, horizontal borehole GPR data for very shallow depths (0.1 - 0.2m) cannot be reliably processed by ray-based techniques because of the interference of the direct wave and critical refracted air wave. During an infiltration process, the low-velocity surface waveguide resulted in dispersive GPR data, which can be inverted to obtain high-resolution shallow subsurface information. By combining the complementary borehole and surface GPR data analysis, the dynamic processes during an infiltration experiment have been investigated. The infiltration experiment was conducted on bare soil at the rhizontron facility in 2013. The Borehole Zero-offset-profiling (ZOP) data were measured at 0.1, 0.2, 0.4, 0.6, 0.8 and 1.2 m depth using 200 MHz antennas. The borehole data at 0.1m are not used for analysis. Wide-angle reflection and refraction (WARR) surface GPR data were acquired with 100, 200, 500, and 1000 MHz antennas. For the horizontal borehole ZOP data, the first arrival time of the signal is determined to calculate the velocity of the radar wave. Surface WARR measurement using 500 MHz antennas showed pronounced waveguide dispersion, because of a thin saturated layer on top that was generated by the infiltration events, and was acting as a waveguide. Single-layer waveguide dispersion inversion was used to invert the dispersive surface GPR data to obtain the thickness and average permittivity of the waveguide layer. Combining surface and borehole GPR data enables the investigation of the soil water content profiles at depths between 0-1.2m. The averaged soil water content at depths of 0.2, 0.4, 0.6, 0.8, 1.2 m are obtained by the ZOP surveys using horizontal borehole GPR. The dispersion curve inversion has been employed for processing the dispersive surface GPR date to obtain the averaged soil water content for the upper 0.2 m. The soil water contents obtained from surface GPR and borehole GPR data show a good consistency and the combined profiles clearly display the depth dependent soil water content changes.

Affiliations: 1 Forschungszentrum Jülich

Abstract: Ground penetrating radar (GPR) is a useful electromagnetic tool, which can be used for imaging the subsurface in a non- or minimal-invasively way. It can provide two electromagnetic parameters simultaneously: relative dielectric permittivity $\varepsilon$r and electrical conductivity $\sigma$, which both influence the velocity and attenuation of electromagnetic waves. Crosshole GPR has often been used in hydrogeological investigations to obtain cross-sectional information about porosity, soil water content and connectivity of structures. Conventional crosshole GPR tomographic inversions are generally based on geometrical ray theory that employs first-arrival travel times and maximum first-cycle amplitude to obtain the wave velocity and attenuation of the amplitudes. In recent years, full waveform inversion (FWI) has been developed which uses the entire waveform and therefore provides higher resolution images of subsurface properties. Several crosshole GPR measurements acquired at the Krauthausen test site were inverted using the FWI approach in order to obtain high resolution images for permittivity and conductivity. We compared FWI estimation of subsurface properties with those obtained from ray-based approaches. Permittivity can be converted to porosity using CRIM model. Following this conversion, the obtained permittivity models could be further investigated by comparing it to accurate porosity results derived from Cone Penetration Test (CPT) measurements performed at 0.1-0.3m distance from crosshole planes. To evaluate the reliability of the FWI images, the Complex Refractive Index Model (CRIM) was used to convert relative permittivity into porosity, which corresponds with water content in CPT data for fully saturated porous media. Along the CPT transect, we compared the low frequency information content of CPT porosity with FWI results and investigated the possibility to improve the resolution of FWI results images using the higher resolution CPT data.

Affiliations: 1 Ecole des Mines de Paris - Mines ParisTech; 2 Univ Rennes, CNRS, Géosciences Rennes – UMR 6118; 3 University Rennes 1; 4 Ecole Normale Supérieure de Paris

Abstract: Faults are present in many geological reservoirs that may be used for either fluid production (e.g. geothermal energy) or fluid storage (e.g. CO2 sequestration). Understanding the hydromechanical processes governing the behavior of a fault in response to transient stresses is a necessity for a safe and sound exploitation of subsurface resources. Here, we explore the in situ hydromechanical behavior of faults from three standpoints : 1) a fault zone is disturbed by a continuous and controlled increase in fluid pressure ; 2) a fault is disturbed by controlled sinusoidal sources of fluid pressure at various frequencies ; 3) a fault is disturbed by an uncontrolled multi-harmonic source of external stress (remote seismic waves). The study site is everytime the same: the Ploemeur hydrogeological observatory (OZCAR network), which is equipped with innovative monitoring tools such as ground surface tiltmeters, implemented in the framework of the CRITEX project. In each experiment, we jointly investigate and interpret hydraulic and surface deformation data to gain insight on the properties and the mechanical functioning of the fractured subsurface. We show that, while all standpoints bring information on subsurface processes, each have the potential of focusing on specific phenomena. In particular, when the source of disturbance is controlled (experiment 1), it is possible to determine the evolution in time of the system and we show that the method is sensitive to contrast in properties between fault and matrix. Adding periodicity in the forcing (experiment 2) allows for characterizing the scale dependency of fracture properties and enhances drastically the detectability of the smallest fault strains. The third experiment has the advantage of benefiting from a perturbation signal of extraordinary richness compared to the previous methods. This allows for detecting complex mechanical behaviors, that might never have been unraveled before at these time scales. This work shows the importance of mixing approaches and viewpoints when it comes to subsurface property and processes characterization.

Affiliations: 1 Technische Universität Berlin; 2 Technical University of Berlin; 3 UP Umweltanalytische Produkte GmbH; 4 Bohdan Dobrzański Institute of Agrophysics

Abstract: The objective of the PROFILE TDT project is to develop a robust, cost-efficient profile probe with high accuracy for longtime observations of soil moisture, salinity and temperature e.g. for the agricultural sector, dyke control or permanent observation of contaminated sites. The probe works with a modified Time Domain Transmission (TDT) technique, which was invented and patented by our cooperation partner the Bohdan Dobrzański Institute of Agrophysics. So far, the validation of the profile probe was mainly carried out on laboratory scale. The experiments included thorough material test on the plastic POM-C, which is used for the probe body. The material was tested regarding its water absorption and signal interferences, which could arise from the water absorption and affect the water content measurements. For this test a TDR probe was used, which operates with a method comparable to the TDT measurement technique. Furthermore, the material´s resistance to stress fractures and the robustness of the electronics was investigated in a stress test, using hitting methods. This experiment was carried out with regard to the probe installation. Further tests during the field experimental phase will show the probe´s suitability for permanent and multiple installation. The calibration of the water content measurements was done for middle sand, sandy loam, clayey sand, silty sand and raised bog peat. Measurements were taken with the PROFILE TDT probe, TDR laboratory probes and gravimetrically as reference. The experiment showed, that the measurement accuracy of the PROFILE TDT probe (R²=0.98-0.99) is in the range of the TDR probe (R²=0.96-0.99), with slightly better results for the middle sand. The calibration of salinity and temperature measurements is the next pending task and will be also presented at the TERENO conference. To increase confidence and receive independent opinions on our product, we will supply research institutes and projects with PROFILE TDT probes. If you are interested in collaborating with us, please meet us at the conference or contact us via Email (a.zackiewicz@tu-berlin.de).

Affiliations: IPGP

Abstract: Soil moisture plays an important role in the of water and energy exchanges between atmosphere and ground, i.e. in the critical zone. The quantity is highly variable even over an area such as a catchment and a period of one year. A method to capture this variability and study its importance is distributed sensors monitoring in-situ soil moisture, preferably at different depths over the first 50 cm of soil. These specifications necessitate a network of simple, rugged and autonomous sensors. They would have to measure representative quantities of the medium, humidity related, and readily available. Sensor achievements take advantage of recent progresses on electronic as well as on information and communication technology. For unambiguous data interpretation they must be accurate and reliable, i.e. not affected by instrument bias. We will present the concept and achievements of such sensors and their network. The measurement principle is based on the dependence of soil complex permittivity on its water content and salinity, more precisely on the determination of the admittance of electrodes inserted in the soil. The technique is that of the Wheatstone bridge with a self-balanced mechanism for a fast and continuous measurement. It permits a high resolution and accuracy of the admittance at its input in one reading (with very low phase error, offsets and sensitivity to electronic temperature). Hence, and taking into account electro-magnetic parasitic effects (such as electrode inductance), the technique offers a good precision on soil permittivity. Preliminary works – laboratory calibration and comparison with physical models - on the conversion from the permittivity of a medium to its water content and its salinity, depending on the medium temperature and sensor signal frequency are shown. Thermometer inserted in a electrode gives the temperature. The techniques adopted to obtain a standard and low cost network of wireless autonomous sensors are briefly described. Field measurements in a mountainous head catchment over several months with current sensors are as well presented. CHAVANNE X and J.-P. FRANGI, 2017. Autonomous sensors for measuring continuously the moisture and salinity of a porous medium. Sensors, 17(5) :1094. http://www.mdpi.com/1424-8220/17/5/1094 CHAVANNE X, BRUERE A and J.-P. FRANGI, 2018. Comments to: A Novel Low-Cost Instrumentation System for Measuring the Water Content and Apparent Electrical Conductivity of Soils, Sensors, 15, 25546--25563 : http://www.mdpi.com/1424-8220/18/6/1730 Sensors, vol 18, n° 6, 1730.

Affiliations: Forschungszentrum Jülich GmbH (IBG-3)

Abstract: Understanding the feedback mechanisms between soil moisture and biomass production is important for sustainable resources management. Here, we present a new method enabling simultaneous non-invasive measurements of SWC and biomass dynamics based on cosmic-ray neutron sensing (CRNS). In the last decade CRNS developed to an established method for field scale soil moisture estimation. Its non-invasive nature and footprint size rely on the measurement of secondary cosmic ray neutrons. In the fast neutron energy range (1eV – 100 keV), neutrons are inversely correlated with soil moisture. In the thermal neutron energy range (<1 eV) this dependence is less pronounced. We found that the ratio between thermal and fast neutrons contains information on additional hydrogen pools such as aboveground biomass, canopy interception and ponding water. This enables the simultaneous and non-invasive field scale measurement of soil moisture and biomass development. For testing the simultaneous soil moisture and biomass estimation with CRNS we instrumented an arable field cropped with sugar beet. Therefore five CRNS probes and for reference a wireless in-situ soil moisture sensor network consisting of more than 150 sensors were installed on the field site. Above- and belowground biomass were sampled in a monthly interval. We found a close linear relationship between the thermal-to-fast neutron ratio and aboveground biomass. Using this information we estimated aboveground biomass with a RMSE of 0.14-0.22 kg/m2, depending on the used method. For the correction of biomass impacts on soil moisture estimation with CRNS we summed the CRNS-based aboveground biomass estimation with in-situ measured belowground biomass. This reduced the RMSE from 0.046 m3/m3 to 0.013 m3/m3. We have now started a field experiment with corn with the aim to simultaneously measure soil moisture, biomass and canopy interception using CRNS. We expect that the additional information on canopy interception will improve soil moisture and aboveground biomass estimation.

Affiliations: 1 UFZ-Helmholtz Centre for Environmental Research; 2 University of Bristol; 3 Heidelberg University; 4 Free University of Bolzano-Bozen; 5 University of Potsdam; 6 UFZ Helmholtz-Zentrum für Umweltforschung

Abstract: The regional monitoring of storage, movement, and quality of water is of crucial importance to practical applications such as agricultural production, water resources management, and flood, drought and climate change predictions. A very promising method to cover soil moisture dynamics at scales relevant for land management or modeling purposes is the method of Cosmic Ray Neutron Sensing (CRNS). While the stationary use of instrument has been applied successfully to monitor soil moisture dynamics at a spatial scale up to a footprint of 500 m, a new development, the mobile CRNS rover, promises to cover soil moisture patterns at regional scales. Recently published insights on the physics of the CRNS sensor’s spatial response revealed a dynamical nature of the footprint radius depending upon the ambient sources of water as well as the fact that the sensor is highly sensitive to hydrogen in the immediate vicinity. The latter aspect is of particular critical relevance for the practical application of mobile CRNS. During CRNS rover campaigns the sensor is very often moved along existing roads and pathways and it can be assumed that the CRNS measurement could be biased towards road moisture conditions. Thus, precise knowledge about the underlying physics and the relationships between site heterogeneity and sensor response is an essential prerequisite for the successful enhancement of the CRNS rover method. The purpose of the presentation is to highlight this challenge. Here, we present results from dedicated experiments and neutron transport simulations, and will propose approaches to correct the CRNS signal depending on site-specific wetness conditions, road geometry, and road material.

Affiliations: 1 UFZ Helmholtz Centre for Environmental Research; 2 UFZ-Helmholtz Centre for Environmental Research; 3 Public Enterprise Sachsenforst

Abstract: For sustainable forestry, high resolution data about site conditions is crucial to ensure forest functionality as well as timber production and pest management. Rapid changes of these conditions are currently driven by element input, forest conversion and subsoil liming. These factors influence nutrients and water availability and can therefore endanger sustainable forest management. For the assessment of the impact, high-resolution and up-to-date information is needed. However, field sampling and laboratory analysis of soil samples is costly, and reasonable-priced approaches are preferred. We use methods of reflectance spectroscopy in the visual and near-infrared-region of the electromagnetic spectrum for indirect measurement and prediction of physical and chemical soil properties (ph-value, cation exchange capacity, base saturation, C/N ratio). For this purpose, different spectrometers are used under laboratory conditions to create a spectral database of forest soils. The correlation between measured spectra and soil properties is analysed with regression methods und then be used to calibrate a statistical model. This approach is currently mainly used in agriculture, and its functionality is now to be examined on forest sites. At first, the focus lies on forest stands dominated by Norway spruce and Scots pine trees, as these species are the most abundant ones in forests in Saxony. The project partner Saxony Forest provides the soil samples (retain samples of Oh and Ah horizons) that are used to build the spectral database. The properties of the samples have already been established with conventional methods and are the basis for modelling the correlation to the measured spectra. The results will be used to assess forest stands with respect to nutrients availability, to develop site specific management strategies and to enhance current methods of periodical forest site mapping.

Affiliations: 1 Helmholtz Centre for Environmental Research - UFZ; 2 Eberhard Karls University Tübingen; 3 Agricon GmbH

Abstract: Detailed knowledge on soil functions is an important aspect for food security and carbon cycling. In particular, temporal and spatial information on the variability and quantity of soil organic carbon (SOC) is essential for further management decisions. Geophysical platforms offer possibilities for mapping and monitoring SOC at the field and small catchment scale. We will present an example of a combined application of geophysical data measured with a towed platform and lab analysis by VNIR-spectroscopy at an agricultural field site in Saxony-Anhalt. Focus was on the deployment of a mobile tool for soil sampling and subsequent analysis by visible and near-infrared reflectance (VNIR) spectroscopy. VNIR-spectroscopy was performed with a standard procedure and compared to a procedure using the fresh (wet) soil samples. We observe a good performance of SOC prediction for standard procedure and for the fresh samples treated by the mobile tool for soil sampling. Furthermore we tested the use of gamma spectroscopy for spatial prediction of SOC with the help of machine learning approaches with promising results. The used approach is able to improve field scale characterization of soil properties and functions.

Affiliations: 1 CNRS - Geosciences Rennes; 2 OSUR- University of Rennes

Abstract: Dissolved volatiles compounds (DVC) in groundwaters and rivers result both from natural processes (atmospheric gases dissolution, rock weathering, biological reactions) and from anthropogenic activities (industrial, urban, agricultural,). Noble gases (He, Ne…) and reactive gases (N2, O2, CO2, CH4, N2O, H2) have been widely used as proxies of respectively recharge conditions or residence time (4He), and redox conditions or biological reactivity. However, dissolved VOC produced by human activities, have never been analysed while they could provide valuable information on anthropogenic impacts on water compositions. This is made possible using the newly developed membrane inlet mass spectrometer (MIMS) that can measure simultaneously all VOC on the field (Chatton et al., 2017). Industrial activities influences could be revealed by the presence of molar masses 61, 78, 91, 92, 95, 106, 166, which correspond to BTEX compounds. Urban inputs could be identified by masses 66, 73, 91 and 105 due to gasoline components (MTBE, Toluene, xylene and C2 alkyl-benzene). Agricultural influence could be revealed by masses corresponding to pesticides, herbicides, fumigants and their metabolites. The sensitivity of the MIMS is less than 1µg/l, which allows the detection of contaminations for human consumption. Normalized mass spectra of dissolved compounds in waters, measured with a MIMS, are used in this study as a fingerprint of water. This fingerprint, through the diversity of compounds detected and their relative concentrations could be an integrator of chemical and biological reactions as well as anthropogenic influences. We will present here water fingerprints from several watershed and rivers in Brittany (North West France) in different anthropic contexts. The diversity of the fingerprints provides information on water pathways, recharge zone and human influence. This approach could be very useful to link contamination and water dynamics at the watershed scale.

Affiliations: 1 Institut des Géosciences de l’Environnement

Abstract: Suspended sediment fluxes are an important process in the critical zone as they shape the landscape and the riverine ecosystem (Owens et al., 2005; Brantley et al., 2017). Besides other detrimental effects such as loss of fertile top soil and reservoir siltation, excess sediment export from headwater catchments can cause the degeneration of water quality. Suspended sediments are a preferential transport vector for nutrients and contaminants, contributing to eutrophication in downstream water bodies and to toxic impacts on aquatic organisms. Knowledge of suspended sediment provenance in meso-scale catchments is important for applying erosion control measures and best management practices on hillslopes on the one hand and for understanding the processes that lead to sediment transport in the critical zone on the other hand. As suspended sediment fluxes are found to be highly variable in time (Poulenard et al., 2012; Legout et al., 2013; Cooper et al., 2015) knowledge of sediment provenance at high temporal resolution is crucial. We studied within- and between event dynamics of suspended sediment fluxes at the outlet of a 42 km2 Mediterranean catchment belonging to the French critical zone observatory network (OZCAR). The spatial origin of the suspended sediments was analyzed at a high temporal resolution using different low-cost fingerprinting approach (Color tracers, X-ray fluorescence and magnetic susceptibility). In addition to the various tracer sets, three mixing models were tested. The comparison of the tracer sets and the mixing models allowed us to evaluate the uncertainty inherent in sediment fingerprinting studies and to assess the challenges and opportunities of using these low-cost tracer sets as fingerprinting properties. Both tracer sets and all mixing models could identify marly badlands as the main source of suspended sediments. However, the percentage of source contributions varied between 11 flood events in the catchment. While for some events the percentage contribution remained constant between different samples taken during the same event, others showed a high within-event variability of the sediment provenance. Considerable differences in the predicted source contributions were also observed when different tracer sets or mixing models were used. These results highlighted the importance of using multi-tracer-multi-model approaches for sediment fingerprinting studies in order to assess and decrease the uncertainty of the method. High sampling resolution that can only be realized with low-cost methods is important to reveal within- and between event dynamics of sediment fluxes and to obtain reliable information of main contributing sources.

Affiliation: Director of Strategic Development, Environment and Infrastructure, Battelle-National Ecological Observatory Network (NSF-NEON), Boulder, US, Personal website

Environmental Observatories (EOs) are emerging globally out of necessity for large-scale integrated data and scope to address both important science questions and societal imperatives. Some develop organically from bottom-up networks, others more top-down, formal infrastructure. They are designed to complement other research projects, programs, networks, innovation, and policy—not in competition with. However, developing EOs are themselves a frontier activity. No two EO development paths are alike, and there are lessons to be learned from all. Balancing top-down approaches and rich bottom-up scientific creativity continues to be a challenge. This is a pragmatic keynote in; (i) the NEON Science design and development approach is discussed, (ii) challenges to deliver NEON ‘Observatory Science’, (iii) overall common near-death experiences among EOs, and (iv) ongoing cultural barriers to broker top-down and bottom-up approaches.

Affiliations: 1 Helmholtz Centre For Environmental Research - UFZ; 2 Helmholtz Centre for Environmental Research - UFZ; 3 Senckenberg Research Institute and Natural History Museum; 4 Environment Agency Austria; 5 University of Rostock

Abstract: It is therefore essential to monitor glaciers and ice caps. The Global Terrestrial Network of Glaciers has therefore been defined as a framework for the international coordinated monitoring of glaciers and ice caps in support of the UN Convention on Climate Change (UNFCCC). The Essential Climate Variables (ECVs) for this monitoring comprise the glacier length, elevation and mass changes. In this study, we provide examples for a region to continent-wide mapping of glacier volume and mass changes in South America and High Mountain Asia using data from the Shuttle Radar Topography and the TanDEM-X missions. Glaciers and ice caps in South America have lost 18.9±1.7 Gt a-1 in the period (2000-2012/14) corresponding to sea level rise equivalent of 0.052±0.004 mm. This is considerably less than previously estimated by other studies, in particular for the tropical, outer tropical and southernmost Andes. The spatial pattern of glacier mass loss is very heterogeneous with regions showing only small changes to almost balance conditions while the main contributors are the Northern and the Southern Patagonian Icefield due to the dynamic adjustment of tidewater and lake calving glaciers after or during retreat. We will show respective analysis for High Mountain Asia and how these data sets can be used for improved estimates of glacier volumes.

Affiliations: 1 University of Napoli Federico II

Abstract: Within the TERENO observatories, mostly distributed in northern and central Europe, the “Alento” observatory has been recently established to provide specific hydrological data for Mediterranean environments. Across the Mediterranean region, there is an urgent need to obtain an in-depth understanding of the impacts that changes in both climate seasonality and land-use exert on water resources, contaminant transport, soil erosion, landslides and wildfires, namely on the provision of ecosystem services of hilly-mountain catchments. Soil moisture is a key-state-variable controlling energy and water fluxes between land and atmosphere and its values and patterns can help understand and infer dominant hydrological processes occurring within a seasonal climate regime. More than two years ago two study sites (MFC2, agricultural use on clay soil and GOR1, forested site on loamy andic soil) were instrumented with advanced ground-based sensor network platforms that provide hydro-meteorological variables and fluxes in the groundwater-soil-vegetation-atmosphere system. Measured soil moisture datasets evidence that winter season is dominated by wet atmospheric conditions and soil moisture data are normally distributed at both sites. Spring is a transition period from wet to dry and the response of soil moisture values is slightly different between MFC2 (multi-modal with predominance of dryer-than-average values) and GOR1 (normal with negative skewness). In summer, soil moisture in both sites is exponentially distributed with predominance of low soil moisture data and few occurrences of wet soil conditions given by episodic rainfall events. Fall represents again a transition phase, but from dry to wet, and prompts different responses between the two sites. Soil moisture values at MFC2 are normally distributed with negative skewness (tendency of preserving drier-than-average soil moisture in some positions). In contrast, the probability distribution at GOR1 shows bimodal features. These preliminary findings will be compared to areal soil moisture estimated by cosmic-ray neutron probes, calibrated by ongoing field campaigns.

Affiliations: 1 Helmholtz Centre for Environmental Research - UFZ; 2 Dead Sea and Arava Science Center, 3 Kiel University

Abstract: The concept of Ecosystem Integrity (EI) has been developed already some decades ago as a framework for measuring and assessing the state of ecosystems. It is focusing at certain key features of ecosystems like the capability for self-organisation, gaining complexity and being sustainable by following the precautionary principle. This concept has been further elaborated to fit the needs of the Long-Term Ecosystem Research and monitoring (LTER) networks. It is used as a rationale for choosing variables addressing qualities of ecosystems. However, as following the EI framework is quite demanding in terms of measurements which need to be taken, we explore whether focusing on one of the central concepts within EI, the self-organisation capability of (eco-) systems could serve as a proxy for EI. We understand self-organisation in this context as driven by organisms by building communities and creating even certain abiotic conditions (e.g. micro-climate). Self-organisation is most evident in succession of communities by pattern formation on spatial, taxonomic and functional levels. The potential for development of successions is always framed by external abiotic conditions like nutrient and water availability or temperature regimes. Using examples of different systems, we focus on the question whether the state of self-organisation can be assessed and evaluated. This has direct implications for the usability of this concept in monitoring schemes.

Affiliations: 1 University of Strasbourg; 2 CNRS - Université de Strasbourg; 3 INRA; 4 CNRS; 5 Ecolab UMR 5245

Abstract: The pollutant gaseous emissions into the atmosphere (SO2, NOx, CO2, metals, PCB, HAP…) have modified for decades the biogeochemical cycle of many elements, from the local to the global scale. Sulfur and nitrogen emissions have resulted in acid rain by the 60s, which have degraded the quality of springs, rivers, lakes, and soils resulting in nutrient imbalances, forest decline, acidification and eutrophication of surface water, among various nuisances. Fortunately, following clean air measures, acid atmospheric deposition has strongly diminished since the 80's both in Norh-America and Europe. This study reports on 30 years of continuous records of atmospheric deposition, spring and stream water composition in a mid-mountain forested critical zone. The Strengbach is a small granitic catchment (80ha, North East of France in the Vosges Mountains) where hydrological and geochemical data have been collected since 1985 (Observatoire Hydro-Géochimique de l'Environnement ; http://ohge.u-strasbg.fr). These long-term data series (including water fluxes, pH, Na, K, Mg, Ca, Cl, HCO3, SO4, NO3, and Si) have been investigated using mathematical and statistical tools. Results show classical statistical distributions, but associated with periodic and non-stationary signals that can be deciphered via specific signal processing. A clear annual periodicity but the lack of other significant low (e.g., multiannual) or high (e.g., seasonal) frequencies was evidenced for all chemical and hydrological parameters (from rain, spring, stream). Many data are of the self-averaging type (their means over large periods tend to constant values) with a few exceptions such as Cl, Ca, Mg and SO4 in spring and stream waters that are non-stationary over time. Taking for example the emblematic species H+ and SO42- associated with acid rain, the non-stationary behavior of concentrations in various compartments of the hydrosystem goes with the reduction of anthropogenic SO2 (proton precursors) emissions since the 80s in the North Hemisphere. These observations are the signature of the positive impact of the clean air measures in remote areas, which are no longer under the threat (for the above species) of long-range transport of gaseous emissions. However, the most recent data of bicarbonate, pH and sulfate indicate that complete recovery has not been reached yet. Moreover nitrogen pressure is still on the way.

Affiliations: 1 Institut de Recherche pour le Développement; 2 IWMI; 3 Cea; 4 DALaM

Abstract: The Multi-Scale Environmental Changes observatory (https://mtropics-fr.obs-mip.fr/), which belongs to the French Research Infrastructure OZCAR (Critical Zone Observatory, Applications and Research), was launched in 1998 and has been monitoring weather, land use, agricultural practices, stream discharge, and sediment yields in three agricultural catchments in Laos, Thailand and Vietnam. Soil erosion not only deplete soil resources on-site, but also degrades water resource quality and leads to the siltation of reservoirs off-site. This paper reviews the major outcomes of this long-term observatory in terms of sediment yields at the outlet of the catchment in Laos (60 ha) and soil detachment from 250 1-m² plots installed in a large variety of land uses, soil and slope conditions. The land use of these last twenty years was upland rice slash and burn cultivation with a declining fallow period. This last decade, this system has been gradually converted to teak plantations. The catchment runoff coefficient increased from 16 to 31% of the rainfall and sediment yield from 98 to 609 tons km-2. The mean annual runoff coefficient in 1-m² micro-plots raised, from 5% for 4-year fallow to 45% for 14-year teak trees plantation, with a very high soil detachment of 3 765 g m−2 y−1 under teaks. The virtual absence of understorey under teaks and rubber trees favoured the formation of impervious surface crusts due to high kinetic energy of througflow. For an intensity of 30 mm h-1, this kinetic energy measured with a disdrometer was 18.41 j m−2 mm−1 for free falling rainfall, 24.76 j m−2 mm−1 under teaks, 39.42 j m−2 mm−1 under rubber trees, but only 9.24 j m−2 mm−1 under grass. These data contradict the governmental policies that promote the shift from annual crops to commercial perennial crops (teaks, rubber trees) to improve soil conservation in montane environment. Policies should ban fire under teaks and herbicides under rubber trees to enable soil protection from aggressive througfall. These results illustrate that soil detachment data from 1-m² plots can explain the changes in sediments yields, even though other erosion processes must also be considered within the catchment including tillage erosion, gully erosion, landslides and bank erosion. Also they show that several components of the critical zone (canopy height and cover, understorey, crusts, farming practices) must be taken into account to model soil detachment and sediment yield.

Affiliations: Centre for Ecology & Hydrology

Abstract: In 2013 the UK’s Centre for Ecology & Hydrology (CEH) embarked on an ambitious project to install and operate a national soil moisture monitoring network based on the cosmic ray neutron sensor (CRNS). The CRNS was seen as a “game-changer” in soil moisture measurement particularly because of its large footprint, autonomous remote operation, and the ability to provide data in near-real time. By mid-2018, 47 COSMOS-UK sites have been established, with plans in place to install more sites during the second half of the year. In addition to the CRNS, each site is equipped with instruments to measure a comprehensive set of meteorological and soil variables, as this was seen as important to make the data useful to the widest possible user community. COSMOS-UK has addressed a wide range of issues from the identification of suitable sites and negotiation with site owners, through installation of instrumentation, to data processing, quality control and data management, and finally dissemination of information products. However, an unexpected challenge concerned the derivation of volumetric water content from the recorded neutron counts, especially at sites in the UK with higher rainfall and soils with high organic matter content. This forced a comprehensive review of what had been considered a standard “CRNS-method” and resulted in the development of an empirical method to better correct for variations in the incoming cosmic ray flux. This presentation will provide an insight into the many challenges faced by COSMOS-UK, and the progress that has been made towards the objective of providing near real time monitoring of the moisture content of the UK’s soils.

Affiliations: Brandenburg University of Technology Cottbus-Senftenberg

Abstract: The Huehnerwasser catchment was built 2005 in a post-mining landscape in Eastern Germany and was left for an unmanaged primary ecosystem succession. At the same time a comprehensive monitoring program was launched for observing the expected dynamic development of this 6 ha site. During the following years this artificially created system was subject to a series of fast changes with regard to morphology, hydrology or vegetation cover. Parallel to these clearly visible processes several hidden processes, such as soil forming processes or groundwater development, led to differentiations in the subsurface. In summary, these processes resulted in significant alterations of the original system within a short period of time. This development can be divided up into single phases with dominating processes or groups of processes. A first phase was mainly characterized by interactions between abiotic system components such as water and substrate. However, already in this very early developmental phase biotic components in the form of microorganisms were a crucial system forming factor by means of biological soil crusts. This period was quickly replaced by a second phase with dominating hydrological processes. During this phase the local groundwater body of the system developed to its full extent. With the beginning of the present phase biota gained control of the system behavior. Particularly plants and related processes such as water uptake and evapotranspiration are now responsible for new modifications of the system. The Huehnerwasser site and its monitoring program offer an example for the observation of highly dynamic ecological systems. This transitional development of a very young ecosystem contrasts with the behavior of mature ecosystems with slow and less pronounced changes unless the system is disturbed. Thus, the observations of this young system also requires dynamic adaptations of the monitoring design as new and in some cases unexpected processes occur resulting in new system changes to be observed. In this paper we present insights into a unique landscape observatory and its design for observing critical zone processes and interactions. Results of the ongoing monitoring program illustrate the dynamic nature of natural systems in fast motion. Even if the previous period of 13 years cannot be regarded as “long-term” monitoring, the results might be of interest for both critical zone observation and explanation. The fast reactions of the system offer analogues for reactions of mature systems triggered by global change. Hueherwasser can be regarded as a laboratory for adaptation processes of natural systems.

Affiliations: 1 Helmholtz-Zentrum Potsdam Deutsches GeoForschungsZentrum GFZ; 2 Julius-Maximilians-Universität Würzburg; 3 Deutsches Zentrum für Luft –und Raumfahrt e.V. (DLR); 4 Friedrich-Schiller-Universität Jena; 5 Friedrich-Schiller-Universität Jena, Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR); 6 Julius-Maximilians-Universität Würzburg, Martin-Luther-Universität Halle-Wittenberg

Abstract: Agricultural landscapes form the basis for the supply of food, textiles and increasingly also energy. Against the background of the high demand of a growing world population, it is necessary to make land use decisions, to optimize production and, at the same time, to guarantee the preservation or qualitative improvement of ecosystems and their services. The targeted use of remote sensing as well as field and laboratory measurements can be used for biophysical monitoring of agricultural landscapes. The DEMMIN large-scale facility operated by the German Remote Sensing Data Center of the German Aerospace Center (DLR) and the German Research Center for Geosciences (GFZ) collects agrometeorological data and soil parameters in an environmental monitoring network with over 40 climate and 63 soil moisture stations. The facility is part of the observation network TERENO (www.tereno.net). Within the project GLAM.DE (Global Agricultural Monitoring - the German experiment) all measurement network data collected in DEMMIN are now freely accessible. In 2017, DEMMIN was assigned to the international validation network Joint Experiment of Crop Assessment and Monitoring (JECAM, www.jecam.org). Together with a large number of other JECAM validation areas on different continents, the facility thus contributes to the JECAM SAR Inter-Comparison experiment. Up to now, ecosystem parameters (e.g. soil moisture, biomass production or yield level) on the agricultural land itself have only been recorded irregularly and with a low temporal resolution. To establish a permanent monitoring, a standardized sampling and measurement design was developed. The design is based on the approach of the JECAM network and enables global validation of remote sensing products. The field measurement campaign will be carried out for the first time in 2018 in cooperation with students from various universities. The aim is to collect high-quality reference data for the Copernicus validation component. Within the framework of the promotion of young scientists, students are offered the opportunity to participate in the implementation of current and future campaigns. An extension of the network of cooperation partners in data collection and use is aimed for. The presentation is intended to inform about current and future work as well as conceptual developments with regard to the sampling and measurement design at DEMMIN. The overarching goal is to establish DEMMIN as a reliable testing and validation area in an agricultural landscape representative for the temperate latitudes within the agricultural authorities and research associations.

Affiliations: 1 IPG PARIS; 2 IRSTEA Lyon

Abstract: We will present the French Critical Zone Research Initiative called OZCAR (Observatoires de la Zone Critique –Application et Recherche – Critical Zone Observatories – Application and Research), a National Research Infrastructure (RI). OZCAR is a network of instrumented sites monitoring over the long term different compartments of the Critical Zone. OZCAR-RI is built on existing monitoring sites and networks. These sites have their own initial scientific questions, monitoring strategies, databases, modeling activities, and are quite well representative of the heterogeneity of the critical zone and of the associated scientific communities studying this zone. Despite this diversity, all OZCAR-RI sites share the main overarching scientific question, which is: how to monitor, describe and simulate Critical Zone adaptation to a changing environment (climate, LU and LC change). OZCAR is articulated around a set of scientific questions, has an ambitious instrumental development program, and aims at working toward a better interaction between data and surface models. The ultimate aim of OZCAR-RI is to build a CZ community, in close interaction with LTSER network, able to share a common representation of CZ dynamics, and educate a new generation of scientists more apt to tackle the wicked problem of the Anthropocene.

Affiliations: University of Hohenheim (310d)

Abstract: In the frame of two large collaborative research projects on Regional Climate Change the University of Hohenheim (Institute of Soil Science and Land Evaluation, Institute of Physics and Meteorology) has performed extensive measurements in agricultural fields in two regions of Southwest Germany, Kraichgau and Swabian Alb. In each region we set up three eddy-covariance stations. Each station was placed in the middle of a field. Field sizes varied between 8.7 and 24 ha. We measured sensible heat, latent heat and CO2 flux in half-hourly resolution. In addition we measured radiation, air temperature and humidity, precipitation as well as soil temperature, water content and matric potential in 0.5, 0.15, 0.3, 0.45 und 0.75 cm depth. In addition, we measured phenological stages, biomass production and leaf area index (LAI). In the presentation, we will show the measured data with focus on evapotranspiration and the Bowen ratio as well as data variation across sites and years. By comparison with state-of-the-art crop models, a land surface model and a new, coupled land-surface-crop model I will demonstrate the importance of long-term measurements for crop and land surface modeling.

Affiliation: Civil Engineering and Geosciences, TU Delft, Personal website

Spaceborne SAR and scatterometer systems deliver radar backscatter data at a range of spatial resolutions from meters to tens of kilometers. Radar backscatter is sensitive to soil moisture, and vegetation water content as well as surface and vegetation geometry. Spaceborne radar data have been used operationally for soil moisture retrieval, land cover classification and above ground biomass monitoring. An implicit assumption in each of these applications is that the vegetation constitutes a relatively static dielectric medium.
In reality, vegetation is highly dynamic. Vegetation acts as an interface between the earth's surface and the atmosphere, modulating exchanges of water, carbon and energy and responding to environmental stressors. Potential gradients between the root zone and atmosphere drive moisture transport within the vegetation, influencing both the total amount of water in the vegetation and its internal distribution at sub-daily scales. Results from a recent field experiment, centered on a new L-, C- and X-band radar system, will be used to illustrate how these dynamics affect radar backscatter as a function of frequency, polarization and viewing geometry during a growing season. Spaceborne SAR and scatterometer data will be used to demonstrate that phenomena observed at field scale are also observed at footprint scale, yielding new opportunities to observe vegetation water dynamics in agricultural and natural ecosystems.

Affiliations: 1 Karlsruhe Institute of Technology, Institute of Meteorology and Climate Research – Atmospheric Environmental Research (IMK-IFU), Garmisch-Partenkirchen, Germany

Abstract: Pastures and natural grasslands are an important part of the pre-Alpine and Alpine landscape ranging from intensive grasslands in the lower regions to highly diverse seasonal mountain pastures and specialized natural ecosystems. Despite the economic value and the significant role of plants in grassland carbon (C) and nitrogen (N) cycling, spatially explicit information on grassland management and yields (biomass and quality) are rarely available. Within the SUSALPS project, we aim to use new Copernicus satellite data to derive spatially explicit information (maps) on grassland management and biomass production, including cutting frequencies and plant N contents. Satellite remote sensing will be complemented by unmanned aerial vehicle (UAV)-based data acquisition to better bridge the scale gap from field measurements to regional assessment, and to validate satellite-derived products on larger scales. In April 2018, we conducted a field campaign to investigate the potential to map biomass, leaf N content, and other vegetation parameters with two different UAV-based multispectral sensors (Parrot SEQUOIA; RedEdge-M by MicaSense). Data over ten different grassland sites in and close to the TERENO Bavarian Alps/ pre Alps Observatory was collected via UAV and field sampling (destructive and proximal sensing) and will be analyses for correlations between different multispectral bands/indices and vegetation parameters. This contribution will present the sampling design and first results of the field experiment.

Affiliations: Helmholtz Centre for Environmental Research - UFZ

Abstract: Human interactions with freshwater ecosystems, changes in climate and agriculture and other land-based activities generate major risks for inland water ecosystems and hence interfere with water quality. In recent years, capabilities in terms of spatial, radiometric and spectral resolution of available optical remote sensing sensors as well as methods to analyze such data have rapidly increased. An important requirement fostering the acceptance of environmental authorities and decision-makers worldwide is to provide reliable data at a high spatial and temporal resolution. The evaluation of data quality requires standards for the verification and assessment of the remote sensing data, and a common understanding of harmonized water quality properties derived by (passive) optical methods, which are capable to serve as consistent and accepted measures. To develop transferable concepts and methods for the validation and calibration of inland optical remote sensing products, a large-scale campaign was carried out in the TERENO observatory “Harz/Central German Lowland”. With the particular focus on the eutrophication of waters, five different water bodies with a strong trophic gradient on the regional scale were examined simultaneously for several days. For this purpose, in-situ measurement protocols were standardized and tested and the simultaneous cooperation of six measurement teams (40 participants in total from 9 institutions, of which 2 companies) was carried out operationally in 2017. The aim of the presentation is to discuss the potentials and challenges for the conception, organization and implementation of real-time, large-scale, remote sensing cal-val campaigns. In particular, the data management and the standardization of measurement protocols will be discussed and first results of the campaign will be presented.

Affiliations: Ludwig-Maximilians-Universitaet Muenchen

Abstract: Good knowledge about soil moisture over agricultural fields, especially with high spatial and temporal resolution, is important for several applications ranging from hydrology to precision agriculture. Microwave remote sensing technologies have proven the ability to derive soil moisture with high resolution over large areas. To derive soil moisture from microwave imagery, different approaches have been proposed in the literature, ranging from empirical (Water Cloud Model) over semi-empirical (Oh et al., Dubois et al.) to more physical based (Integral Equation Method) radiative transfer (RT) models. Those RT-models have been often used in the past, performing well for selceted sites and studies. Nevertheless, a clear validation site and tool for the comparison of the different models among themselves and in different model combination for the different backscatter contribution of soil and vegetation is missing. With the availability of the Sentinel mission, in particular Sentinel-1A/B, new opportunities for analyzing and validating soil moisture retrieval algorithms at high spatial and temporal resolution are now given. In this study, we present a performance analysis of different RT-model combinations for the retrieval of soil moisture over agricultural fields on the Munich-North-Isar test-site. The used RT-models consists of a vegetation scattering and a soil scattering component with different complexity. Therefore, different RT-models with different complexity for the estimation of the soil scattering contribution (Oh et al., Dubois et al., Integral Equation Method) are coupled with models which are calculating the vegetation scattering contribution (Water Cloud Model, Single Scattering Radiative Transfer) of the total radar backscatter. The different models are driven with in-situ measurements on field basis and the retrieved backscatter is compared to available Sentinel-1 time series. In addition, an outlook is given how these models can be used in a variational land-data-assimilation scheme.

Affiliations: 1 ZALF, Institute of Landscape Hydrology; 2 ZALF - Leibniz-Centre for Agricultural Landscape Research

Abstract: Unmanned aerial vehicles (UAV) borne remote sensing can be performed at high spatial resolution and at arbitrary timing, e.g., for assessing the pattern of actual evapotranspiration in arable fields with 10 cm resolution. On the other hand, various geophysical techniques provide valuable information about soil structure even at greater depth. But geophysical survey data hardly provide any information about plant behaviour beyond the lab scale. Merging information from remote sensing and geophysical data thus might be a way to assess plant root behaviour at low effort and with nearly arbitrary temporal resolution. We developed and applied an approach based on data from a set of field trials at Müncheberg, about 50 km east of Berlin, Germany, covering an area of 3.2 ha. In the year of the study two different varieties of maize were grown and have partly been irrigated. A soil electric resistivity survey gave some information about the soil texture patterns at different layers down to 1.5 m depth. Spatial patterns of actual evapotranspiration were assessed using UAV mounted multi-spectral and thermal sensors, combined with additional on-site measurements. UAV surveys were performed on 16 July and 31 August 2015. After the first survey a drought period developed with high air temperature and with hardly any rainfall which substantially affected maize growth and transpiration on the non-irrigated fields. We assumed that the spatial pattern of evapotranspiration partly reflected the spatial pattern of subsurface water availability in different soil layers that should be closely related to soil texture, and that the latter was partly at least reflected by the soil resistivity data. Our approach did neither require any detailed information about the nature of those relationships nor any additional information about other factors that affected the spatial pattern of evapotranspiration. Then a non-linear Support Vector Machine regression was performed using the spatial pattern of evapotranspiration as target variable and those of soil resistivity in different soil layers as predictors. The contribution of single soil layers, determined by stepwise pruning of the Support Vector Machine, was taken as a proxy for the share of total root water uptake during the day of the respective UAV survey. These depth profiles were compared to soil moisture data. There were clear differences between the two UAV campaigns and between irrigated and non-irrigated fields, indicating root adaptation to decreasing water availability on non-irrigated fields.

Affiliations: 1 Helmholtz-Zentrum Potsdam Deutsches GeoForschungsZentrum GFZ; 2 Royal Belgian Institute for Natural Sciences (RBINS); 3 Tartu Observatory (TO); 4 Consiglio Nazionale delle Ricerche (CNR); 5 Laboratoire d’Oceéanographie de Villefranche-sur-Mer, CNRS Universiteé Paris VI; 6 Instituto de Astronomía y Física del Espacio, Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET/IAFE); 7 National Physical Laboratory (NPL); 8 Consiglio Nazionale delle Ricerche (CNR) - IREA; 9 Consiglio Nazionale delle Ricerche (CNR) - ISAC

Abstract: For providing validation data for optical remote sensing mission the most effective was is the use of a network of automated instruments on unmanned platform, e.g. AERONET-OC and RADCALNET. The re-use of data from each site for many optical missions (S2AB, S3AB, PROBA-V, MODIS-AQUA/TERRA, VIIRS, Landsat8, Pléiades, ENMAP, PRISMA, HYPXIM, HYSPIRI, SABIAMAR, PACE, etc and commercial satellites e.g. Planet Doves) gives a huge economy of scale. Especially for multi-mission validation the existing AERONET-OC and RADCALNET networks that are based on multispectral instruments, do not cover all required spectral bands of all sensor and require spectral modelling with associated uncertainties to cover all spectral bands of all sensors. In addition the instruments are very expensive in terms of acquisition and maintenance The H2020/HYPERNETS project, started in February 2018, is to develop a new lower cost hyperspectral radiometer and associated pointing system with embedded calibration device for automated measurement of water and land bidirectional reflectance. The instrument will be tested in a prototype network covering a wide range of water and land types under different operating conditions. One of the test sites is the TERENO NE test site DEMMIN, where a 10m height measuring mast for the HYPERCAM instrument is planned on an agricultural field. In addition the measuring station will be equipped with sensors for climate and soil variables measurements and an Eddy Flux station. Quality controlled data with associated uncertainty estimates will be provided automatically for the validation of all optical satellite missions. Preparations will be made a) for the new instrument design (and associated calibration service) to be commercialized with an expected lifetime of at least 10 years and b) for the networks to be further expanded to become the main source of surface reflectance validation data for all spectral bands of all optical missions for at least the period 2022-2032. This presentation will put forward the project's current understanding of the required radiometer characteristics, auxiliary equipment and network infrastructure that is needed by validation scientists and by the European Space Agency and EU/Copernicus for validation of optical reflectance products. Furthermore the planned station at DEMMIN will presented. The HYPERNETS instrument will measure just a small spot at the earth surface, while the spectral variability of agricultural dominated area is much higher. This spectral, spatial and temporal variability will by analyse to optimise the location of the HYPERNETS station.

Affiliations: Middle East Technical University

Abstract: Soil moisture measurements are the key elements of water resources engineering, since knowing the soil moisture helps calculation of the other unknowns of the hydrological cycle such as evaporation and infiltration. From an agricultural point of view, soil moisture is the most important element for decision making processes of irrigation systems. Soil moisture measurements are also vital for prediction and assessment of water related disasters such as floods and droughts. Effective possible usages of soil moisture measurements require high spatiotemporal resolution. Satellite soil moisture products cover larger areas and can be used for soil moisture monitoring. In this study we want to present the potential use of the cosmic-ray probes (CRP) to validate the satellite soil moisture products. Data from the CRP installed at 1459 m in the south part of Turkey in October 2016 and data obtained from all the stations of the COSMOS database (http://cosmos.hwr.arizona.edu) are used. The water content reflectometer from ML3 ThetaProbe (CS 616) which was installed at the same location at a depth of 5 cm in the ground, which was calibrated for soil-specific conditions and soil temperature, is used to monitor the soil moisture at a point scale. Both CRP and CS616 sensors show consistent changes in soil moisture due to the storm events occurred during the observation period. Satellite soil moisture products obtained from the Soil Moisture and Ocean Salinity (SMOS), the METOP-A/B Advanced Scatterometer (ASCAT) and Soil Moisture Active Passive (SMAP) are compared with the measurements of the CRP for the period of March 2015-April 2018. CRP data obtained from the station located in Turkey and soil moisture obtained from SMAP Level 4 product give the highest correlation. It is observed that regions with large vegetation dynamics have lower correlation whereas regions having semi-arid climate has larger correlation. Standard validation scores present that SMAP is a high accuracy soil moisture product with low uncertainties as compared to CRNPs. The correlation coefficients are larger in semi-arid regions whereas low correlation coefficients are obtained for the regions with large vegetation dynamics. Modelled soil moisture obtained from Noah Land Surface Model (Noah LSM) are used to perform Triple Collocation (TC) for the station in Turkey.

Affiliations: 1 University of KwaZulu-Natal; 2 South African National Parks; 3 South African Environmental Observation Network; 4 Kansas State University

Abstract: The relative importance of groundwater (GW) to sustain terrestrial vegetation has been well documented. However, estimating the contribution of various sources to transpiration (T), while important, remains a challenge particularly with respect to understanding its variability in space and time. Furthermore, quantifying GW use by riparian vegetation in data scarce regions may prove to be even more challenging. Although, several methods for estimating evapotranspiration (ET) at multiples scales exist, these methods rarely offer a way to partition subsurface water contribution to T at scales appropriate for informing management decisions. For this purpose, we present a potentially useful approach to bridging this knowledge gap. In this study ET estimates from the satellite-based surface energy balance system (SEBS) model were coupled wit stable isotope analysis, to map and quantify the contribution of GW to transpiration (ETg), along the lower reaches of a perennial river system, in the semi-arid north-eastern region of South Africa. Plant stem, soil, stream and GW samples were collected on 3 sampling occasions during the 2016 dry season, which also coincided with a large El Nino induced drought period. D2H and D18O values of the respective samples were measured and analysed. We found that while GW use was prevalent and increased with aridity, overall ETg was fairly minimal. During the initial stages of the dry season ETg for the study area was extremely low, approximately 0.10 % of daily ET or 0.01 mm d-1. However, as aridity increased, ETg increased to approximately 10 % of daily ET or 0.30 mm d-1. The results of the various investigations undertaken in this study demonstrates the potential of coupling stable isotope analysis with satellite-derived ET estimates, as it can be used to provide a cost-effective and spatially representative means to quantify plant water use dynamics. This in turn can prove to be extremely beneficial to water resources managers, as it can be used to facilitate the improved allocation of water resources for human needs and environmental water requirements.

Affiliations: University of Oviedo

Abstract: The albedo is recognized as very important parameter in the Earth’s energy balance. The analysis of their values allows to diagnose the snow cover in the polar zones. For this purpose, satellite data and in-situ data are fundamental. In Antarctica there are fewer studies than in the Arctic, probably due to its more severe climatic conditions. The aim of this work is to study the albedo in the zone corresponding to Livingston Island, in the South Shetland Archipelago, specifically taking as a reference the location of the Spanish Antarctic Base Juan Carlos I (JCI), where there is a station of the State Meteorological Agency (AEMET). For the analysis, data were taken from two satellite sources: Moderate Resolution Imaging Spectroradiometer (MODIS), specifically the daily snow product, MOD10A1, and Landsat data. They will be validated by in-situ data from JCI station. There is a previous validation of the MOD10A1 data on that island based on data from the Johnson Glacier (JG) station (Calleja et al, 2017), which found a decrease in albedo in the summer and an increase in winter, when, in addition, the MOD10A1 data variability is greater. The present study covers the period between January 1, 2000 to March 6, 2017. Two filters were applied: the solar zenithal angle filter (SZA) to all data, selecting the dates with SZA <75º, and the MODIS quality filter (QA), selecting good quality data (QA = 0). The results obtained show albedo values notably lower than those reported in the mentioned study, probably because the JCI station is closer to the coast and is lower, and the presence of bare soil. The intercomparison of satellite data from two sources (MOD10A1 and Landsat) and in-situ data allows us to advance in the validation of the results obtained by remote sensing, and is part of the algorithm that we have proposed for the elaboration of albedo maps on Livingston Island.

Affiliations: 1 Palacky University Olomouc; 2 Leibniz Centre for Agricultural Landscape Research (ZALF); 3 Palacký University Olomouc

Abstract: Soil erosion is a natural phenomenon but it is often aggravated by anthropogenic interference in land uses. Spatiotemporally monitoring land cover status and estimating the vulnerability to potential soil erosion, especially for large geographical areas, have been a paramount task in terms of resource requirements and timely needed actions in curbing soil erosion. Utilization of satellite images for these purposes has become a growing trend. However, the usability of such images for large geographical areas relied on local knowledge or small (plot) scale extrapolation; and it is often done using single satellite observation data, overlooking the spatiotemporal aspect of land cover status. The aims of this study are therefore 1) to estimate and compare vegetation cover factor values (C-factor), one of the main factors in several soil loss estimation models, for large agricultural geographic region using time series satellite images (Landsat ETM+) coupled with ancillary land use data (the integrated administration and control system (INVEKOS); and 2) to examine the influence of spatiotemporal and cover type variations on satellite image derived C-factor values and its implication on subsequent soil erosion predictions. To this end the Uckermark district of the Brandenburg region, in the North Eastern Germany, was taken as a study area. Twelve Landsat ETM+ images, from the year ca. 2014 to ca.2016, were acquired from USGS representing four seasons of a year and the normalized vegetation index (NDVI) based C-factor values (Cndvi) were calculated. For each scene date the corresponding C-factor values (Ctab) were assigned according to DIN 19708:2017-08 using the INVEKOS land use data. Statistical analysis was performed using spatially balanced random data points. Results revealed that C-factor values derived from satellite images (Cndvi) correlate well (r=0.86) with INVEKOS assigned values (Ctab) particularly in spring and early summer season. Furthermore, Cndvi values were found to be influenced by variations in edaphic condition, in slope shapes, and in plant cover types and their phenological stages. As a concluding remark taking these parameters into account while estimating C factor values using remote sensing images could improve the accuracy of the estimate and help for appropriate interventions. Keywords: Remote sensing; Landsat ETM+; INVEKOS; NDVI, C-factor; erosion

Affiliations: 1 German Aerospace Center (DLR)

Abstract: Present and future spaceborne Synthetic Aperture Radar (SAR) sensors can acquire data in a high temporal and spatial resolution (independent of daylight and weather conditions). In particular, SAR measurements over agricultural fields are sensitive to dielectric properties of the soil and the vegetation and therefore provide key variables for monitoring biophysical parameters of agricultural vegetation. However, the characterization of scattering changes in relation to changes in the soil or plant parameters is challenging due to the high dimensionality of the problem. Using SAR polarimetry enlarges the observation space and allows an identification of different types of scattering mechanisms. Nevertheless, it is still difficult to differentiate between changes occurring on the ground or in the vegetation volume. By introducing SAR interferometry or further SAR tomography, i.e. sensitivity to height differences, it is possible to resolve scattering mechanisms along height. Since it is the wavelength which drives the sensitivity to different plant components an important aspect is to understand and investigate which frequency might be most suitable for monitoring a certain parameter. Within the CROPEX 2014 campaign, multi-baseline polarimetric SAR data was acquired by DLR’s airborne sensor F-SAR covering different dates of the phenological cycle at X-, C- and L-band. At the same time, soil moisture and plant measurements were collected. Based on this data set, the poster intends to give an overview about the current status about polarimetric and/or interferometric techniques which can be used to qualitatively or quantitatively characterize specific changes in agriculturally vegetated fields (soil moisture, vegetation water content, plant height, etc). Finally, the overview intends to provide a perspective on the next steps towards the development of biophysical products in agriculture from SAR measurements.

Affiliations: 1 UFZ Leipzig; 2 Hochschule Anhalt; 3 Heidelberg University; 4 University of Arizona; 5 University of Potsdam; 6 UFZ-Helmholtz Centre for Environmental Research; 7 Fraunhofer-Institut für Kommunikation, Information und Ergonomie FKIE

Abstract: While the detection of albedo neutrons from cosmic rays became a standard method in planetary space science, airborne neutron sensing has never been conceived for hydrological research on Earth. We assessed the applicability of atmospheric neutrons to sense root-zone soil moisture averaged over tens of hectares using neutron detectors on an airborne vehicle. Large-scale quantification of near-surface water content is an urgent challenge in hydrology. Information about soil and plant water is crucial to accurately assess the risks for floods and droughts, to adjust regional weather forecasts, and to calibrate and validate the corresponding models. However, there is a lack of data at scales relevant for these applications. Most conventional ground-based geophysical instruments provide root-zone soil moisture only within a few tens of m^2 , while electromagnetic signals from conventional remote-sensing instruments can only penetrate the first few centimeters below surface, though at larger spatial areas. In the last couple of years, stationary and roving neutron detectors have been used to sense the albedo component of cosmic-ray neutrons, which represents the average water content within 10-15 hectares and 10-50 cm depth. However, the application of these instruments is limited by inaccessible terrain and interfering local effects from roads. To overcome these limitations, we have pioneered first simulations and experiments of such sensors in the field of airborne geophysics. Theoretical investigations have shown that the footprint increases substantially with height above ground, while local effects smooth out throughout the whole area. Campaigns with neutron detectors mounted on a lightweight gyrocopter have been conducted over areas of various landuse types including agricultural fields, urban areas, forests, flood plains, and lakes. The neutron signal showed influence of soil moisture patterns in heights of up to 180 m above ground. We found correlation with ground-truthing data, using mobile cosmic-ray neutron sensors, local soil samples, TDR, and buried wireless soil moisture monitoring networks. The work opens the path towards further systematic assessment of airborne neutron sensing, which could become a valuable addition – or even an alternative – to conventional remote-sensing methods.

Affiliations: 1 Humboldt-Universität zu Berlin

Abstract: By adding attributes of space and time to the spectral traits (ST) concept we developed a completely new way of quantifying and assessing land use intensity and the hemeroby of urban landscapes. Calculating spectral traits variations (STV) from remote sensing data and regressing STV against hemeroby, we show how to estimate human land use intensity and the degree of hemeroby for large spatial areas with a dense temporal resolution for an urban case study. We found a linear statistical significant relationship (p = 0.01) between the annual amplitude in spectral trait variations and the degree of hemeroby. It was thereof possible to separate the different types of land use cover according to their degree of hemeroby and land use intensity, respectively. Moreover, since the concept of plant traits is a functional framework in which each trait can be assigned to one or more ecosystem functions,the assessment of STV is a promising step towards assessing the diversity of spectral traits in an ecosystem as a proxy of functional diversity.

Affiliations: Universität Kassel

Abstract: Grassland biomass is one of the most frequently used productivity measures for grasslands. Compared to destructive or field-based measurements, remote sensing images have the advantage to quantify grassland biomass over large areas in a time-efficient manner. In this study, we examined the potential of consumer-grade camera images acquired from an unmanned aerial vehicle (UAV) for estimating field-scale grassland biomass in a 0.1 ha area. Considering the allometric relationship between grass sward height and biomass, models were developed based on crop surface height (CSH), which was derived from the UAV-borne 3D point clouds. The objectives of the study were to estimate grassland fresh (FB) and dry (DB) biomasses using two different empirical models: 1) a simple linear regression model (LM) using the best-correlated crop height metric; 2) non-linear machine learning random forest regression model (RF) using all crop height metrics. The UAV-borne images and reference biomass data were collected from three grasslands with different botanical composition and mowing regimes in northern-Hesse, Germany. To develop models, 24 crop height metrics were computed based on height and colour information (R, G, B) in the derived CSH. From all crop height metrics, the 75th percentile (p75) height metric obtained the highest correlation coefficient with both grassland FB and DB (0.60 and 0.67 respectively). The LM model with p75 resulted in adjusted R² = 0.33 and 0.41 (nRMSE = 45.7%, 35.5%) respectively for FB and DB estimation. Comparably, the RF model including all crop height metrics resulted in 0.61 and 0.66 pseudo R² (nRMSE = 31.6%, 24.8%), respectively, for FB and DB. Both linear and random forest models predicted DB more accurate than FB. The results showed that the RF model explained approximately more than twice of the variance of FB and DB compared to the LM model. The UAV-borne images do not provide information from below the canopy level due to dense vegetation in the grassland and, which could be a limitation of this methodology. Overall, this study showed that consumer-grade camera images acquired from the UAV platform can be utilised for monitoring grassland biomass on large areas.

Affiliations: 1 University of Padova; 2 Cambisol BV

Abstract: Hillslope terracing is an ancient and widespread practice, mostly used to facilitate agriculture by controlling water flow and sediment loss. In Europe, terraces are mostly present in the Mediterranean for grape and olive production, where this production system is both economically and culturally vital. Nowadays, vineyard terraces are a substantial source of soil erosion due to land abandonment, unsuitable terrace designs and expansion, worsened by growing climate aggressiveness, and increased pressure by machinery. There is a need for accurate terrain modelling of such landscapes, to allow hydro-geological assessment and sustainable redesign of landforms, and their drainage functioning. Unmanned Aerial Vehicle (UAV) offers a suitable opportunity for a digital reconstruction of terrace geomorphology through Structure-from-Motion photogrammetry. The high-resolution digital elevation model enables a suitable representation of the terrace micro-topography and its impact on surface water flow and critical concentration points. The flexibility of UAV deployment and flight planning allows adaptive data acquisition of complex landscape features, e.g. by oblique imagery or a higher photographic overlap. Furthermore, the low costs related to UAV surveys may offer a unique acquisition method to land managers and owners of small-medium sized farms. Surveys with a micro-UAV were carried out in vineyard terrace landscapes in the Veneto region of Italy, resulting in 0.2m DTMs. A geomorphologic index Relative Path Impact Index (RPII; Tarolli et al. 2013) was applied to identify preferential flow paths, and estimate runoff concentration by the terrace morphology. High RPII values could be related to the high probability of surface instability in the field (e.g. soil erosion, landslides, and collapse of terraces). The high resolution topography allowed also the digital implementation of drainage networks and a preliminary assessment of its effectiveness to mitigate critical surface water concentration. Using a cost-estimation of different design scenarios, an optimum of costs and benefits could be assessed prior to any field interventions. These case studies illustrate an improved efficiency in vineyard terrace maintenance, which is merely one example of the broad potential of UAV in landscape restoration projects. REFERENCES Tarolli, P., Calligaro, S., Cazorzi, F., Fontana, G.D., 2013. Recognition of surface flow processes influenced by roads and trails in mountain areas using high-resolution topography. Eur. J. Remote Sens. 46, 176–197. https://doi.org/10.5721/EuJRS20134610.

Affiliations: 1 Esri Germany; 2 Technische Universität München

Abstract: Due to global warming, natural disasters such as thunderstorms are becoming more frequent (Brooks, 2013). Such events have a strong impact on forest health, wildlife habitat and inflicts economic losses. Methods to quickly assess damage and to manage disaster in the aftermath of storm events are of high importance for the management. Due to the large areas affected by storms, assessing the damage in a forest is time consuming, if done manually using aerial imagery and GIS tools. As time is a crucial factor in disaster management, integrating deep learning with remote sensing data into a GIS environment holds a high potential for accelerating and improving damage assessment. Convolutional Neural Networks (CNN), a supervised learning architecture, is extensively used in many areas (e.g. interpretation of medical images, 3D reconstructions, objects detection and classification in self-driving cars…). They are efficient in finding patterns within a large amount of data and are widely used in computer vision and recognition in images (LeCun et al., 1989). We develop an algorithm based on CNNs that are trained on labeled damaged areas visible in after-storm aerial orthophotos (RGB and NIR, 0.2 m spatial resolution) of a large forest area in Bavaria (~ 109 square km). Integration into ArcGIS was achieved with the Python API and Jupyter Notebooks. We investigate two CNN architectures: Considering the four bands (RGB + NIR) as independent and trying to find features in each band independently (2-Dimensional Convolution); and looking for patterns in of the image space of all four bands (3-Dimensional Convolution). In preliminary results, the CNN trained using only 100 000 pixels of labeled data could perform, within seconds (depending on the size of the input raster), a detection with a score of over 90% based on the intersection over union metric. These results are promising, considering the complexity of detecting areas of fallen trees. A next step will be to use a second CNN architecture to quantify damage in terms of volumetric loss of timber in addition to areal damage. We will also consider further optimization of the algorithm by assessing the optimal ratio of true positives to false negatives, based on a decision analysis. Overall, our results highlight the potential of deep learning on high resolution imagery for damage assessment following disasters.

Affiliation: German Center for Integrative Biodiversity Research (iDiv), Halle-Jena-Leipzig, Personal website

Global change factors, such as climate and land use change have the potential to alter the diversity and composition of our ecosystems, with important consequences for humanity. Understanding biodiversity change requires sampling communities in space and time using consistent methodology. First, this talk will focus on the value of historic datasets, by illustrating an example using Hermann Müller's data on 8000 interactions between plants and insect pollinators in the Alps from 1874-1879 and another example from UFZ TERENO plant biodiversity sampling. Second, this talk will discuss key findings from other projects that use distributed networks for observing biodiversity change. Third, this talk will present new results on plant-pollinator interactions in the UFZ TERENO site and discuss opportunities for expansion to other sites.

Affiliations: 1 Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences; 2 ecostrat GmbH Berlin; 3 Heinz Sielmann Foundation

Abstract: Earth observations from air- and spacecraft have become an important source of information about the distribution of living organisms, the status of the biosphere and hence, the functioning of ecosystems. To this effect, different remote sensing sensors can be used by nature conservation to evaluate and monitor the diversity of life on earth. However, biodiversity itself is understood as a complex phenomenon, integrating multiple habitats and species interactions that are targeted by different sensor systems. Such versatile information content needs to be structured for interpretable monitoring systems. We developed a multiscale approach from satellite imagery, airborne imaging spectroscopy towards high resolution drone cameras that is realized on a former military training area Kyritz-Ruppiner Heathland in the north-west of Brandenburg, Germany. For each recording system a unique set of biodiversity indicators was defined for spatially explicit mapping purposes. Thus, we propose a terrestrial sampling scheme over the spatial scale of a whole sensor cascade. For this purpose, broad scale habitat units such as Natura 2000 habitat types and biotope types were aggregated for multispectral satellite imagery. On that basis a set of indicators for an assessment of habitat type conservation status was derived. In that respect, imaging spectroscopy was related to individual plant species records and structural plant traits such as senescence and species turnover. Unmanned aerial vehicles could be used to determine plant flowering status and shoot extension. On every scale different machine learning algorithm were trained to calibrate sensor specific spectral responses that serve as predictive models for mapping purpose. Our study shows that a remote sensing based biodiversity monitoring of open grass- and heathlands on former military training areas needs to be conducted at specific sensor levels. As a function of data processing capabilities and feasible expenses, a suitable set of biodiversity indicators will be selected a priori and mapped using standard machine learning models. Vegetation status and habitat encroachment can thereby be evaluated a posteriori on the basis of specific indicator combinations. We propose a scale specific provision of robust sensor models for different biodiversity indicators that should be based on spectral libraries to support the repetitive monitoring of natural inventories in habitats for selected areas.

Affiliations: 1 Helmholtz Centre for Environmental Research - UFZ; 2 Helmholtz Centre for Environmental Research UFZ

Abstract: Wild bees play an important role as pollinators in many habitat types. They are rich in species but difficult to access in terms of monitoring and identification. Moreover, long-term data about insect communities are scarce, although these are nowadays highly requested in the light of recent insect decline. About 60 % of the area of Saxony-Anhalt (Germany) is covered by agricultural fields. This emphasizes the importance to focus on insect communities in agricultural dominated landscapes. We monitored wild bees in Saxony-Anhalt in six landscapes of 4x4 km varying in local weather conditions and land use. In each landscape 16 combined flight traps were installed. The time series started in 2001-2002 and was continued from 2010 onwards. These activities are embedded in ongoing work within the long term monitoring initiative TERENO (TERrestrial ENviromental Observatoria) which is part of the LTER (Long Term Ecosystem Research and Monitoring) network. We interpret our findings with respect to (a) temporal trends of composition and dominance structures of wild bee communities and (b) the capacity of wild bee communities to provide pollination services within the selected landscapes by analyzing overall bee abundance and the diversity of species traits relevant for pollination.

Affiliations: 1 Martin Luther University of Halle-Wittenberg; 2 Martin Luther University Halle-Wittenberg

Abstract: Communities of Apidae bees and Syrphidae flies were sampled at six different sites in Sachsen-Anhalt, Germany that are a part of the Terrestrial Environmental Observatories network (TERENO). Two different sampling methods—yellow pan traps and sweep netting—were used to compare the differences in species richness and abundance. Altogether 3,812 individuals of 102 Apidae species and 226 individuals of 39 Syrphidae species were collected during a month in early spring. We found that the bees species caught using sweep netting had lower richness and abundance than those bees caught in the pan traps, while the opposite was true for the syrphid flies. Larger bee species were more abundant in both methods, whereas smaller syrphid fly species were found more frequently in the pan traps. Bee species that were specialized foragers were more likely to be caught in the pan traps. Sweep netting caught more non-migratory syrphid flies than pan traps. Our results demonstrated the despite thorough sampling, different method provide different community data. In using both methods for insect sampling, the complete pollinator community can be assessed.

Affiliations: 1 Museum für Naturkunde - Leibniz Institut for Evolution and Biodiversity Science

Abstract: The recent biodiversity research and monitoring activities in Germany are diverse and highly fragmented (Geschke et al., 2017; Schliep et al., 2016). Due to the federal structure of Germany, many monitoring activities are funded and managed at federal state level and are mostly not sufficiently connected with each other. However, both the National Strategy on Biological Diversity and the “Naturschutz-Offensive 2020” include the objective to establish a biodiversity monitoring system at the national level (BMUB, 2007, 2015). Therefore, the establishment of a national, long-term oriented and scientifically well-founded independent institution or organizational body is needed (Geschke et al., subm.; Marquard et al., 2013). This shall bring together the diverse biodiversity monitoring activities in Germany and coordinate the collaboration between the respective institutions. Only this way, Germany will be able to establish a comprehensive biodiversity monitoring scheme, which integrates data from different scales and disciplines also covering biodiversity components so far underrepresented (e.g. freshwater and soil ecology, some groups of animal species (Braeckevelt et al., subm.)). Further, a standardization of monitoring methodologies and applied indicators is needed (Geschke et al., subm.; Marquard et al., 2013; Schmeller et al., 2015). Such efforts would additionally support national as well as international reporting obligations (Schmeller et al., 2015). All such needs and recommendations pose the need for collaboration. With our poster, we aim to illustrate the status quo of institutionalized collaboration throughout Germany. We show which research networks are already established, and discuss the status of these networks in the light of recent efforts within the German biodiversity monitoring landscape, including both political and scientific developments. Further, we map ongoing monitoring initiatives in Germany in order to identify overlaps, gaps and requirements for streamlining. We emphasize the need for collaboration between research institutions as well as academics, NGOs and citizen science and aim to develop and provide policy recommendations ready to be further discussed with the conference participants.

Affiliations: Institute of Biology, University of Latvia

Abstract: Nowadays more and more people are concentrated in cities. To make their lives more comfortable and healthier, implementation of the green infrastructure is important in the city. The main prerequisite of functionally sound green infrastructure is the soil. The condition of urban soils directly affects the structure and development of vegetation and costs of its maintaining. Urban soils are subjected to multiple stress, such as fragmentation, mechanical disturbance, trampling, and pollution. Soil mesofauna is considered to be sensitive bioindicator of soil ecosystem state, but its use for assessing the soil condition in the city is associated with two problems. First, regular long-term sampling by standard methods may destroy small fragmented soil patches. Second, the amount of collected material from multiple sites usually is too large to be processed in foreseeable period of time. Solution to this problem could be the development of more robust approach, which requires the use of only one mixed soil sample per site consisting of (3-5) small random soil cores (surface area of 10-20 cm2). By this approach soil assessment is based not on species level but on ecological group and/or family level of dominant components of soil mesofauna which are detectable already at a small number of samples and which ultimately determine the ecological processes in the given habitat. We examined this approach in studies on changes of soil mesofauna on the urban gradient A – E (arbor – eremus) in Riga city comprising street edge lawns in the city center, central park grasslands, private garden lawns, cemetery lawns, urban forests, suburban forests, in total 21 sample site. Presence of plant litter on the sample site, structure of O horizon, organic matter content and soil pH was measured. NMS Ordination of sites by Collembola, Mesostigmata and Oribatida mites identified to the family level showed very good distribution of along the NMS axis characterizing changes in soil pH and organic matter content.

Affiliation: National Ecological Observatory Network (NSF-NEON), Boulder, US, Personal website

by Metzger, S., Desai, A. R., Durden, D., Florian, C., Hartmann, J., Kohnert, K., Luo, H., Pingintha-Durden, N., Sachs, T., Serafimovich, A., and Xu, K.

Understanding evapotranspiration through space and time requires the joint use of observations across various scales, and from disparate measurement platforms. Present-day examples include large-scale satellite remote sensing missions to monitor the earth system such as the forthcoming NASA ECOSTRESS. These in turn rely on distributed in-situ observations for calibration and validation, such as from eddy-covariance flux tower networks the likes of AmeriFlux, ICOS, NEON and TERENO, or intensive flux aircraft campaigns. However, oftentimes the different types of observations overlap only fractionally in space and time at best, posing a substantial and frequently ignored challenge for their joint use.

Here, the Environmental Response Function (ERF) integrative data processing can come to the rescue. This is achieved by approximating an information continuum through catalyzing the strengths of process and artificial intelligence concepts: ERF complements existing mathematical descriptions of better-understood processes such as atmospheric dispersion with observations from towers, aircraft, and satellites about less well-understood phenomena. It then mines the joint information content and yields the most complete solution that is possible from the processes and observations provided.

Here, we investigate evapotranspiration maps at high spatial (10 – 100 m) and temporal (0.5 h) resolutions as one such solution: ERF can map extensive areas around individual flux towers (order 1 – 10 km2) and flux aircraft missions (order 1000 km2) with spatial coverage in excess of 80%. Equipped with this information hot spots and hot moments can be clearly located, ecohydrological and ecophysiological processes can now be analyzed in space and time, and traditional scale gaps can be bridged. This opens the door not only for greatly improved calibration, validation and data assimilation, but ultimately for harnessing the full complement of available information for modeling at the Earth system scale.

Affiliations: 1 Helmholtz-Zentrum Potsdam Deutsches GeoForschungsZentrum GFZ; 2 Leibniz Institute for Agricultural Engineering and Bioeconomy (ATB)

Abstract: The potential yield of agricultural land is an important parameter for optimizing agricultural management strategies in the context of global population growth, food security and the increasing scarcity of arable land. The project "AgriFusion" aims to combine various data sources such as elevation models, remote sensing data, yield surveys or soil maps in order to create potential yield maps and to make them directly available for farmers. The data are combined using a “transferable belief model”, which enables the farmer to incorporate his local expert knowledge. Due to their high spatial and temporal resolution, the remote sensing data of the ESA Copernicus missions Sentinel-1 and Sentinel-2 are highly suitable for crop monitoring. Using these satellite data, yield-relevant vegetation parameters such as dry and moist biomass, leaf area index, chlorophyll content, plant height and crop coverage can be estimated and mapped from the field-scale up to a regional scale. The combined use of optical and radar remote sensing data is in focus, whereat the weather-independent radar amplitude data serve as a supplement to any missing optical data due to cloud cover. To understand the different information of the two sensors during the vegetation period, the temporal development of vegetation indices and SAR backscattering is analyzed. Furthermore, regression models are used to test how crop parameters measured on the ground and satellite data are related. Numerous in-situ field measurements on winter wheat and winter barley fields during the growing seasons 2017 and 2018 in two study areas in Mecklenburg-Western Pomerania and Brandenburg serve as reference data for calibration and validation of the models.

Affiliations: 1 Deutsches GeoForschungsZentrum GFZ Potsdam; 2 Hochschule Neubrandenburg

Abstract: The challenges for the agricultural sector in countries all over the world are surely intensified by a growing world population, an increasing demand of sustainable farming and the need for efficient management due to cost pressure. Precision farming offers ideas, technology and applications to tackle problems within the range of difficulties in this complex system. Precision agriculture can profit from the parallel uprising big data field. Analysis of vast and different data needs algorithms, able to evaluate and interpret. At the same, applications should not take every decision from the farmer, who has years of experience and knowledge about his fields. We therefore present a method of data fusion, which aims to delineate fields into yield zones, by incorporating the farmers knowledge into the fusion method. We use the Transferable Belief Model (TBM) to interpret different data types (e.g. satellite images, soil maps, elevation models) and combine them by applying “beliefs”, rather than probabilities, on the combination metrics. Input data for the model can be any kind of relevant spatial data, which are classified into comprehensible classes. Every class of every “source of evidence” is assigned one or more “hypotheses” – which are the target classes, e.g. yield. Every of those assignments is accompanied by a degree of belief into this assumption of the expert. After applying “Dempsters Rule of Combination” on a pixel basis, the sources of evidence are combined two-by-two. Any new source of evidence, for example a satellite image taken on a different day, may lead to a more precise result. The model is tested on different fields in the TERENO-NE region Demmin and produces promising results for a precision farming application, which is understandable and accessible for the farmer himself. For the development of the method, RapidEye time series, the soil map “Bodenschätzung” and a 5m-DGM was used. The result is validated with dense yield data taken by the farmer’s harvester. The ongoing cooperation with farmers in the area delivers us with farm data and feedback for the developed model.

Affiliations: 1 Ludwig-Maximilians-Universitaet Muenchen

Abstract: The latent heat flux (LE) plays an important role in surface-atmosphere interactions. Traditionally, LE or the actual evptranspiration (Eta) is measured by means of lysimeters, eddy-covariance systems or fiber optics, providing estimates which are spatially restricted to a footprint from a few square meters up to several hectares. In the past, several methods have been developed to derive LE by means of multi-spectral remote sensing data using thermal and VIS/NIR satellite imagery of the land surface. As such approaches do have their justification on coarser scales, they do not provide Eta information on the fine resolution plant level over large areas which is mandatory for the detection of water stress or tree mortality. In this study, we present a comparison of a drone based assessment of LE with eddy-covariance measurements over two different forest types - a deciduous forest in Alberta, Canada and a tropical dry forest in Costa Rica. Drone based estimates of LE were calculated applying the Triangle-Method proposed by Jiang and Islam (1999). In our study, the Triangle-Method estimates LE by means of the Normalized Difference Vegetation Index (NDVI) and land surface temperature (LST) provided by two camera systems (MicaSense RedEdge, FLIR TAU2 640) flown simultaneously on an octocopter. Results indicate a high transferability of the original approach from Jiang and Islam (1999) developed for coarse to medium resolution satellite imagery to the high resolution drone data, leading to a deviation in LE estimates of 10% compared to the eddy-covariance measurements. In addition, the spatial footprint of the eddy-covariance measurement can be detected with this approach, by showing the spatial heterogeneities of LE and thus Eta due to the spatial distribution of different trees and understory vegetation.

Affiliations: 1 Friedrich-Schiller-University of Jena; 2 German Aerospace Center (DLR)

Abstract: Biomass retrieval from SAR is one of the main applications of radar remote sensing. Different sensor configurations have been tested in recent years. They are ranging from single polarization SAR data to fully polarimetric and tomographic SAR data sets either at one radar band or at different radar bands. However, it must be said that especially fully polarimetric or multi-band sensors are mostly limited to experimental campaigns. Operational space borne SAR missions offer SAR data at one radar band with mostly on like (HH/VV) and one cross-polarization (HV/VH). Algorithms to retrieve biomass information from these data sets comprise simple regression models as well as sophisticated semi-theoretical radar backscatter models. Since the 1990s, SAR data were regularly acquired by the space agencies (ESA, JAXA, CSA). Latest with the advent of the ESA SAR sensors using the ScanSAR technique (ENVISAT mission with the ASAR instrument and the Copernicus Sentinel-1 satellite series with the C-SAR instrument), C-band data coverage of land surfaces increased substantially. Thus, multi-temporal to hyper-temporal data sets became available. They have been exploited successfully for biomass mapping from regional to global scales using change detection methodologies. Even the SAR data acquired by the latest SAR sensors are still containing ambiguous information. Besides the geo-physical variable of interest (e.g. biomass), other environmental factors may have an influence on the backscattered signals. Soil moisture is one of these environmental factors. There are authors, who assume that even at C-band (radar band used by ESA SAR satellites since the 1990s) the water content of the soil layer below forests may have an effect on the radar backscatter at C-band. Since 2015, the Department of Remote Sensing at the University of Jena maintains a measurement network with 240 sensors measuring soil moisture and soil temperature in a managed forest in Thuringia, Germany. In parallel, all Sentine-1a and b data sets acquired over the region are collected. Additional meteorological information will be used for data interpretation and analysis. We intend to present the results of our ongoing research activities and try to answer the question, if an influence of soil moisture below forest on C-band radar backscatter has been found for our test site region. This may also help with the interpretation and analysis of biomass maps produced by the ESA DUE GlobBiomass and the ESA CCI Biomass projects, where the department has been and is involved.

Affiliations: 1 KIT- Karlsruher Insitut für Technologie; 2 German Aerospace Center (DLR); 3 UFZ Leipzig

Abstract: Consistent determination of soil moisture across scales is a persistent challenge of hydrological science. In situ measurements within soil profiles are trusted at the point scale and spatially distributed sensor networks capture field scale extents. Cosmic-ray neutron sensing (CRNS) has become an established method to derive volume-averaged soil moisture at the hectometer scale. Airborne remote sensing provides coverage on regional scales, but is limited to the topmost layer. All of these methods are challenging in terms of scaling and data processing, which includes the removal of unwanted vegetation and surface effects and the calibration with independent data. We studied the performance and synergistic potential of these complementary observational methods for the determination of soil moisture in a 55 km^2 Alpine foothill river catchment of Southern Germany. The individual approaches are evaluated and synergistically optimized for a 9 ha grassland test site and thereafter spread to several other locations of the catchment. The results indicate that in situ sensor network data provide valuable information to calibrate the CRNS rover, and to optimize the vegetation removal within the polarimetric synthetic aperture radar retrieval algorithm. Furthermore, the CRNS soil moisture product is improved by taking into account the remotely sensed cross-polarized backscatter product as a proxy for biomass water. The study illustrates the great potential of interdisciplinary research to advance the methodologies and individual geoscientific instruments and how this can lead to improved hydrologically relevant products.

Affiliations: University of Venda

Abstract: Veld fire is an essential natural hazard in South Africa, fires can be a resultant of nature or anthropogenic activities. This study assesses and maps veld fire hazards on the environment in Limpopo Province, South Africa. The study employed GIS and remote sensing techniques on MODIS satellite data to produce veld fire risk maps in Limpopo Province. Enhanced Vegetation Index (EVI) and Land Surface Temperature (LST) variables were used as a surrogate to determine vegetation attributes, behavior and conditions, estimating the Fuel Moisture Content (FMC) using regression analysis and facilitate the classification of fuel types. The estimated FMC was used to generate the veld fire risk maps using Improved Fire Susceptibility Index (IFSI) in ArcGIS 10×. The MODIS satellite images were used to map the burnt areas using the Maximum Likelihood Classifier (MLC) and to estimate the burn severity using difference Normalized Burn Ratio (dNBR). For this reason, the study used Uni-temporal and bi-temporal approach to discriminate severity levels of veld fires in the study area. The classical multiplicative model of time series analysis was used to analyse the veld fire trend from 2002 – 2017 and make a forecast of one year into the future (2018). Based on the results, EVI is strongly related to LST and showed an inverse relationship in the study area. Fuel type 1, fuel type 2, fuel type 5 and no fuel were classified. EVI and LST positively estimated the FMC of each classified fuel type. The burnt areas and burn severity were spatially distributed in the study area across slopes facing Northeast (22.5 – 67.5) and Southeast (112.5 – 157.5) characterized by fuel type 1 and fuel type 2. The regression output between IFSI and historical fire data in the study area a significance F of 0.079 indicating a probability of 7.9% that the correlation between IFSI and historical fire data is a random occurrence. The time series analysis showed a decreasing trend of veld fires since 2002.

Affiliations: Helmholtz-Zentrum Potsdam Deutsches GeoForschungsZentrum GFZ

Abstract: Local parameters for climate modeling are highly dependent on crop types and their phenological growth stage. The land cover change of agricultural areas during the growing season provides important information to distinguish crop types. Knowing the current crop type is advantageous for many applications, therefore an automated crop-type classification was developed. The progressive classification algorithm identifies crop types based on their phenological development and their corresponding reflectance characteristics in multitemporal satellite images of the four sensors Landsat-7 and -8, Sentinel-2A and RapidEye. It distinguishes crop types not only retrospectively but progressively during the growing season starting in early spring. Binary fuzzy c-means clustering differentiates seven crop types in eight decisions at particular time periods. These decisions are based on expert knowledge about plant characteristics in different phenological stages. The unsupervised classification approach and previously defined decisions enable the algorithm to work independently of training data. The fuzzy approach provides certainties of crop type existence and generates first results in early spring. Accuracy and reliability of the classification results improve with increasing time. Due to the automation, daily crop-type classification results are available. The method is developed at the German Durable Environmental Multidisciplinary Monitoring Information Network (DEMMIN). The study area is an official test site of the Joint Experiment of Crop Assessment and Monitoring (JECAM) and is located in an intensely agricultural used area in Northern Germany. Until now, classification results were produced for the growing seasons 2015 and 2016 to develop the algorithm and to test its performance. The overall accuracy in 2015 was amounted to 89.49 %. A challenge remains the distinct separation of wheat and rye, whereas barley, rapeseed, potato, corn and sugar beet are classified with high accuracies. The overall accuracy in 2016 was lower (77.19 %) due to unfavourable weather conditions. The implementation of a daily automatic execution is established for the following years.

Affiliations: Bundesanstalt für Gewässerkunde (BfG)

Abstract: Operational hydrological applications like monitoring of water resources and extreme events, such as floods and drought, as well as forecasting and climate change prediction services require among others spatial and temporal high resolution information about soil water content. This information can be used for multi criteria calibration and validation of rainfall-runoff modelling or by use of data assimilation techniques for updating of initial conditions at relevant forecast times. There is a rapidly increasing number of remote sensing products with a potential application in operational hydrology. Remotely sensed images offer an opportunity to supplement ground measurements for runoff prediction during dry periods and flood events. The EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (H-SAF) operationally provides high quality precipitation, soil moisture and snow products derived from Earth Observation Satellites (EPS), together with their continuous quality assessment.The vision for the ongoing third phase of the initiative (CDOP-3; 2017-2022) maintains the footprints of previous phases. Besides the development, generation and consolidation of products derived from primary EUMETSAT geostationary and polar orbiting missions (MTG and EPS-SG) the use and benefits of H-SAF products are promoted and evaluated in collaboration with the hydrological community. The H-SAF soil moisture products are available as European or global products. The soil moisture product SM-DAS-3 offers a global view of soil moisture in the root zone and is produced by assimilating satellite data with a Land Data Assimilation System. The data record covers approximate thirty years, which allows a use in climatological research questions. The features of the data record will be presented. The usability and the concept for the validation of soil moisture products will be discussed. Examples are given using Tereno SoilNet soil moisture monitoring, lysimeter data, as well modelling results for selected river basins located in Europe. Analysis and hydro-validation of H-SAF soil moisture products gives promising results for different test sites with different topography and catchment characteristics.

Affiliations: GFZ German Research Centre for Geosciences

Abstract: Evapotranspiration (ET) is one of the key flows in the global water cycle and current meteorological, climatological and hydrological models need to make approximate assumptions for covering this effect. Due to its numerous influencing variables, as water availability and transport, vegetation and weather, ET is difficult to monitor, especially on large and heterogeneous areas. Remote sensing offers spatially distributed information about land cover and the top layer of the earth, as well as indirect measurements of physical properties of this layer. Combined with additional information of the surface conditions and standard equations from meteorology this can be used for modelling ET. An adapted algorithm based on Allens METRIC methodology (Allen, 2007) is used for this comparison. This method was successful applied to mainly homogenous areas and is now adapted and applied to heterogeneous landscapes, characterized by changing elevation, land cover types, land use and strongly varying soil conditions. Agrometeorological models for estimating soil moisture and ET at point scale were calibrated to site specific conditions using soil moisture measurements and the ET was compared to the ET derived from remote sensing data. Current results show promising results, however a detailed investigation of single parameters is not yet finished and further model improvements are expected, together with using more specific input data, e.g. crop type information.

Affiliations: 1 GFZ German Research Centre for Geosciences; 2 Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research

Abstract: The Earth surface is coupled to the global climate system by vertical fluxes of energy and matter, which vary greatly in time and space. Understanding the underlying processes and both the effects of changes in environmental or climatic conditions on those fluxes as well as their feedback on the climate system requires investigations of the temporal and spatial dynamics. Here, we focus on the spatial patterns of surface fluxes as detected by airborne eddy covariance measurements above parts of the Northeast German Lowland TERENO Observatory. For this purpose we analyzed an eddy-covariance data set obtained by the research aircraft POLAR 5 as part of the AIRMETH campaign performed in 2011. After thorough data pre-processing, Reynolds averaging is used to derive spatially integrated fluxes. To increase spatial resolution, we then use wavelet transforms of the original high-frequency data. This enables much improved spatial discretization of the flux observations and determine biophysically relevant land cover properties in the flux footprint. In order to improve the spatial coverage of surface fluxes we derived footprint flux maps combining the flux values along the flight tracks and the corresponding footprints. These maps allow to asses energy partitioning and to determine dominant exchange processes across heterogeneous landscapes.

Affiliations: University of Hohenheim

Abstract: Water-vapor and temperature profiles with high accuracy and vertical resolution from the surface to the lower troposphere are fundamental for accurate weather forecasts, process studies, and validation of satellites. Within the ACROSS project (Advanced Remote Sensing - Ground Truth Demo and Test Facilities) of the Helmholtz Alliance of German Research Centers, a new remote sensing system fulfilling these requirements are is currently being developed by the Institute of Physics and Meteorology (IPM) at the University of Hohenheim (UHOH). The aim of the ACROSS project is close the gap in the data assimilation of satellite-based earth-observing systems, offering reference data of environmentally essential parameters in different spatial and temporal resolution, in order to enable its interpretation in the field of environmental science. This new remote sensing system is a robust, portable, high-power, scanning rotational Raman lidar. It is based on the knowledge acquired at the IPM in the development of different generations of rotational Raman lidar systems in the recent years. The system measures atmospheric temperature profiles with high resolution, even in daytime due to the use of strong UV laser radiation and an optimized receiving chain in each one of the implemented channels It is well known that different parts of a pure rotational Raman backscatter spectra show different temperature dependence. Therefore, the ratio from two backscattered signals from two of this parts can be used to obtain a temperature profile of the atmosphere. Besides, an elastic and water-vapor profile are measured. First atmospheric measurements have been obtained during the Land-Atmosphere Feedback Experiment (LAFE) at the ARM Southern Great Plains (SGP) central facility, OK, USA, in August, 2017. During 2018 the system is being tested at the UHOH site. Both results will be presented at the conference. The new lidar will enhance the Terrestrial Platform planned inside the ACROSS project, showing the potential of remote sensing systems gathering ground-truth information about the land-surface-atmosphere feedback, the behavior of the atmospheric boundary layer and the lower troposphere. Furthermore, it may serve as prototype for a future network of automatic thermodynamic lidar profilers.

Affiliation: Dept. of Geography, Environment and Society, University of Minnesota, USA, Personal website

Even after more than a century of coordinated monitoring, instrumental weather observations are still too short to adequately constrain decadal or multidecadal behavior in the Earth’s climate system. Leading climatologists and climate modelers have called for the wider application of high-resolution proxy records to decadal variability and prediction studies, and our community has responded by producing new paleoclimate products that specifically target this type of ‘intermediate-term’ behavior. But we now also know our medium changes that message: the biological and geological systems that encode climate information into natural archives often also alter the original ‘input’, usually due to either seasonal filtering or non-climatic persistence. In this talk, we’ll discuss some of the challenges inherent to the use of high-resolution proxies to study decadal or multi-decadal climate variability, and suggest strategies that might clarify how climate acts on those timescales. And we’ll also present a new theoretical framework that could help paleo-scientists evaluate competing ideas about the causes of decadal- or multi-decadal events known to have occurred during the past one or two millennia.

Affiliations: 1 Belgian Nuclear Research Centre (SCK∙CEN); 2 BIAX consult; 3 ONDRAF/NIRAS; 4 VUhbs archeologie

Abstract: Hydro(geo)logical systems are routinely described using state variables (groundwater head or river discharge), obtained from instrumental observations which cover a time period of a few decades only. This time window is too short to capture the full variability of a hydrological system that might be expected under different boundary conditions and driving forces. In an attempt to extend this time window, we present preliminary results of an integrated study of small-scale buried depressions (buried fens) in the northeastern Belgian sand belt, aiming at deciphering the observed archive in terms of water table dynamics. Several proxies were used to estimate the elevation of the past groundwater table, including podzol soil morphology, texture and depositional style of the sediments in the depressions, and their palynological content. Age control was derived from OSL dating of quartz and radiocarbon dating of plant remains. The depressions, studied through shallow trenching, are located on an interfluve reaching several meters above the adjacent valley bottom. Their size varies between 100-5000 m², while they are between several dm up to 2 m deep (relative to the ancient surface). We observe the following: (1) soil horizon morphology (hydromorphic soil features) and peat development are related with depression depth, (2) peat in the deepest depression contains pollen of aquatic plants, (3) the pollen content reflects alternating periods of eutrophic and oligotrophic conditions, and (4) the subsequent sandy infill shows sedimentological features typical for deposition in a standing water body. Combining the evidence from proxies with age control leads to the following water table reconstruction: (i) an intial phase of podzol soil formation and landscape stabilization under drainage conditions that were sufficiently good to allow the soil to be formed, and (ii) a subsequent phase during which the deepest depressions were permanently filled with water to allow peat to be developed and drifting sand to be deposited in a standing water body. As a preliminary conclusion we put forward that the rising water table which was initiated at the latest around ca. 3000 years ago is probably a combined effect of climate and land use change (less evapotranspiration due to lower temperatures and different vegetation). Furthermore, the correlation between proxy characteristics on the one hand, and fen depth on the other hand strongly suggests that the observed water table fluctuations reflect groundwater table dynamics. Overall, we find second-order sub-millenial variations superimposed on a first-order supra-millenial signal.

Affiliations: 1 Leibniz-Institute for Baltic Sea Research Warnemünde (IOW), Marine Geology, 18119 Rostock, Germany; 2 Department of Geology, Quaternary Sciences, Lund University, 22362 Lund, Sweden; 3 Physics Institute, Climate and Environmental Physics, University of Bern, 3012 Bern, Switzerland; 4 GFZ-German Research Centre for Geosciences, Section 5.2 Climate Dynamics and Landscape Evolution, 14473 Potsdam, Germany; 5 Max Planck Institute for the Science of Human History, 07743 Jena, Germany; 6 Polish Academy of Sciences, Institute of Geography and Spatial Organization, Warszawa 00-818, Poland; 7 Polish Academy of Sciences, Institute of Geography and Spatial Organization, Torun 87-100, Poland; 8 Department of Geology, United Arab Emirates University, 15551 Al Ain, UAR; 9 Tandem Laboratory, Uppsala University, 75120 Uppsala, Sweden

Abstract: Time-scale uncertainties between paleoclimate reconstructions often inhibit studying the exact timing, spatial expression and driving mechanisms of climate variations. Detecting and aligning the globally common cosmogenic radionuclide production signal via a curve fitting method provides a tool for the quasi-continuous synchronization of paleoclimate archives. In this study, we apply this approach to synchronize 10Be records from varved sediments of Lakes Tiefer See and Czechowskie covering the Maunder-, Homeric- and 5500 a BP grand solar minima with 14C production rates inferred from the IntCal13 calibration curve. Our analyses indicate best fits with 14C production rates when the 10Be records from Lake Tiefer See were shifted for 8 (-12/+4) (Maunder Minimum), 31 (-16/+12) (Homeric Minimum) and 86 (-22/+18) years (5500 a BP grand solar minimum) towards the past. The best fit between the Lake Czechowskie 10Be record for the 5500 a BP grand solar minimum and 14C production was obtained when the 10Be time-series was shifted 29 (-8/+7) years towards present. No significant fits were detected between the Lake Czechowskie 10Be records for the Maunder- and Homeric Minima and 14C production, likely due intensified in-lake sediment resuspension since about 2800 a BP, transporting ‘old’ 10Be to the coring location. Our results provide a proof of concept for facilitating 10Be in varved lake sediments as novel synchronization tool required for investigating leads and lags of proxy responses to climate variability. However, they also point to some limitations of 10Be in these archives mainly connected to in-lake sediment resuspension processes.

Affiliations: University of Greifswald

Abstract: Besides climate, tree growth in general depends on a multitude of other factors, of which several are heavily influenced by human activities especially in the temperate areas of Europe and North America. Atmospheric depositions of sulphur and nitrogen, rising CO2 levels, drainage and changing management practices have shaped growth conditions that are unprecedented in the past and together create a so called “no-analogue problem” in dendroclimatology. If these factors are incompletely acknowledged in the calibration stage of tree-ring based climate reconstructions, a strong bias is likely added to the trends and amplitudes of the reconstructed climate parameters. Further bias might occur if different wood sources are incorporated in the tree-ring chronologies. The older parts of such chronologies usually contain samples from construction wood or of subfossil origin preserved and recovered from lakes, rivers or peatlands. Often the exact origin of the historic samples is unknown and due to wood-trade, different provenances are mixed in a chronology. Even if the samples are taken in situ, specific growth conditions at sites like peatland margins potentially lead to very different climate-growth relationships of the subfossil material compared to the modern material usually sampled in forests of differing site conditions. Using example tree-ring datasets from the northern central European lowlands and Alaska we highlight these potential biases and show that at least parts of the no-analogue conditions can be statistically addressed. We demonstrate that a better separation of non-climatic noise from climatic signals in tree-ring data is possible, which significantly changes reconstructed climatic amplitudes.

Affiliations: 1 TU Dresden; 2 Deutsches GeoForschungsZentrum GFZ Potsdam; 3 Landesamt für Denkmalpflege Baden-Württemberg; 4 Friedrich-Schiller-Universität Jena; 5 Uninversität Greifswald

Abstract: Land-use history of Central European medium-range upland regions, e.g., the Black Forest, is controversially debated throughout the last decades. Archaeological investigations reveal cultivation and settling not before medieval times (ca. 900-1000 cal BP). In contrast, findings of cultural indicators in many pollen records from lake sediments and mires indicate a much longer human land- use history in this region reaching back as far as the Bronze Age (ca. 4000 cal BP). However, pollen records might be biased by transport from distal sources pointing to the need of additional sediment proxies that are sensitive to local human impact. Here, we compare Holocene pollen and geochemical records derived from µXRF core scanning of a 740 cm long lacustrine sediment core recovered from Titisee (Southern Black Forest, 846 m asl). Clustering of elements revealed two contrasting sources: (i) clastic-detrital input of soil particles as reflected by the elements Ti, Si and K, and (ii) redox cycling reflected by Fe, Mn, S and As. A major shift towards a dominance of redox-sensitive elements occurred between 6000 and 1800 cal BP. Levels of As follow this general trend but lead these changes by 300-700 years indicating further environmental control on As. A shift to higher As levels at 6300 cal BP coincides with a major change in forest composition, i.e., decline of deciduous trees and spread of Abies alba (silver fir). Increased As levels could have been induced by As mobilization following soil acidification through accumulation of heavily degradable Abies needles. Decreasing As levels at 2550 cal BP coincide with slightly increasing cultural pollen amounts, indicating the spread of pasture land that could have led to stabilization of soil As. These results suggest As a sentinel for past environmental changes evidencing the onset of man-made landscape transformation in the Titisee area at around 2550 cal BP. The suggested control by coniferous trees further indicates that the recent trend of increasing As concentrations over the last 50 years will proceed into the future following recent reforestation with Picea abies (Norway spruce).

Affiliations: 1 GFZ German Research Centre for Geosciences, Section 5.2 – Climate Dynamics and Landscape Evolution, Potsdam, Germany; 2 Max Planck Institute for the Sciences of Human History, Jena, Germany; 3 Institute of Earth and Environmental Sciences, University of Potsdam, Potsdam, Germany

Abstract: Annually laminated (varved) lake sediments represent unique archives in continental areas providing both, precise chronologies and seasonally resolving proxy data. Monitoring of physical, chemical and biological processes influencing lake sediment formation are a suitable approach for detailed proxy understanding of varved sediment records. Lake Tiefer See (NE Germany) indicates deposition of varved sediments today as well as millennia ago (Dräger et al., 2017; Kienel et al., 2013). Therefore, the lake provides the possibility to trace current seasonal layer formation in the lake and to pair these data to climate and lake conditions (Kienel et al., 2017). Lake Tiefer See was formed during the last glaciation and is part of the Klocksin Lake Chain, a subglacial channel system that crosses the Pomeranian terminal moraine. The lake is a mesotrophic hard water lake with a maximum depth of 63 m and a surface area of 0.75 km². During six consecutive years (2012-2017) the particulate matter deposition was trapped at bi-weekly to monthly resolution at two different water depths (12 and 50 m). The sediment trap material was analysed for sediment flux and organic matter and calcite content. In addition, we monitored limnological parameters (e.g. temperature, pH, oxygen content) as well as the meteorological conditions (e.g. temperature, wind speed and direction, precipitation) with a monitoring and climate station installed on the lake. These data describe strength and duration of lake mixing and lake stagnation phases. Our results show distinct seasonal peaks in sediment formation, which correspond to the spring and summer productivity phases comprising of diatom blooms and calcite precipitation. This observation is in line with microfacies results from surface sediment cores. The content of biogenic calcite content decreases in the trapped material with increasing water depth indicating dissolution processes. However, the strength of calcite dissolution varies between seasons and years. We will discuss the depositional processes in relation to conditions in the water column and to meteorological data. This study is a contribution to the Virtual Institute of Integrated Climate and Landscape Evolution Analysis –ICLEA– of the Helmholtz Association; grant number VH-VI-415.

Affiliations: 1 University of Greifswald; 2 Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences

Abstract: Fossil pollen is a powerful proxy for land cover reconstructions. However, quantitative interpretation of pollen data is, among others, hampered by differential representation of plant taxa in the pollen record. Plant taxa produce pollen in very different amounts. Due to the different size and shape of pollen grains, also their dispersal pattern differs. Recent attempts to account for that bias in pollen data mainly rely on taxon specific correction factors, such as relative pollen productivity estimates (PPEs). PPEs are calculated in laborious calibration that involves extensive surface pollen and vegetation sampling. Large discrepancies among the existing studies show that the methodology is still problematic. High resolution pollen analysis from partly annually laminated Lake Tiefer See (NE-Germany) has highlighted further limitations of that approach. PPEs are commonly applied under the assumption that pollen productivity of plant taxa is constant. Comparing pollen data from Lake Tiefer See with historic records has shown that pollen productivity of grasses and other herbs has changed significantly due to changes in land management during the 19th and 20th century. The modern PPEs may therefore not be representative for the past. Using the Lake Tiefer See record we developed the alternative ROPES approach, which does not require calibrated correction factors. The approach, for the first time, involves pollen accumulation rates (PAR), i.e. annual pollen deposition. PAR values may change largely across a single lake or between lakes, and thus appear too noisy for quantitative reconstructions. However, in stable sedimentary environments, the changes in pollen deposition at a given point in a lake are expected to linearly represent changes in the abundance of the respective plant taxon. ROPES builds upon this relationship by assuming that in the most probable reconstruction, the reconstructed changes in the abundance of each taxon are similar to changes in the respective PAR values. The approach searches for this most probably reconstruction in an iterative optimization algorithm. By that it not only produces a landcover reconstruction but also pollen productivity estimates suitable for the given study period. The approach can hence be applied to study changes in pollen productivity in time and space. ROPES requires high resolution, well dated pollen records with sufficient variation in each pollen type. Wherever such records are available, the approach enables landcover reconstructions devoid of additional calibration.

Affiliations: 1 German Centre for Geoscience GFZ, Section 5.2 Climate Dynamics and Landscape Evolution, Potsdam, Germany; 2 German Archaeology Institute, Scientific Department of the Head Office, Berlin, Germany; 3 University of Arizona, Tree-Ring Laboratory, Arizona, United States; 4 Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Birmensdorf, Switzerland

Abstract: Millennial length temperature reconstructions developed from tree-ring archives remain rare in Europe. Where records do exist, they originate from high elevation sites along the European Alps, a region much smaller and ecologically distinct in comparison to the far-reaching temperate lowlands of the greater region. Consequently, our knowledge of long-term temperature variability across lowland Europe according to the tree-ring record largely relies on deterministic factors governing the interaction between tree productivity at high elevation and alpine climate. To distinguish similarities and differences in temperature variability between the European lowlands and the neighbouring alpine region, we developed a millennial length proxy record of annual temperature using dendro-dated wood anatomical measurements of European oak earlywood vessels. The chronology was built using living tree material collected from Hinnensee (northeastern Germany) covering the period AD 1696-2016, and extended to AD 1000 with historic building material sourced from the surrounding region. To remove age related growth trends and better capture intra-annual and decadal-length climate variability, vessel size measurements were standardized by applying linear models to juvenile, intermediate and adult growth years. Correlations between vessel size and instrumental temperature data were consistently positive with an optimal correlation period lasting the previous growing season (April) until the month preceding the current growing season (March). Since 1901 intra-annual correlations with climate data were 0.62 and 0.9 after decadal smoothing (Pearson correlation coefficient; p < 0.001). A strong spatial relationship between the vessel chronology and gridded climate records extend across and beyond Europe indicating that during the calibration period the proxy shares a high degree of common variability with regional climate. Furthermore, comparisons with other well-known climate reconstructions from the European alpine region and Fennoscandinavia will be conducted and new results presented and discussed at the TERENO International Conference.

Affiliations: 1 German Research Centre for Geosciences, Section 5.2: Climate Dynamics and Landscape Evolution

Abstract: Mechanistic understanding of tree-ring formation and its modelling requires a cellular-based and spatially organized characterization of a tree ring, moving from whole rings, to intra-annual growth zones and individual cells. A tracheidogram is a radial profile of conifer anatomical features, such as lumen area and cell wall thickness, of sequentially- and positionally-ranked tracheids. However, its construction is tedious and timeconsuming since image-analysis-based measurements do not recognize the position of cells within a radial file, and present-day tracheidograms must be constructed manually. Here we present the R-package RAPTOR that complements tracheid anatomical data obtained from quantitative wood anatomy software (e.g., ROXAS, WinCELL, ImageJ), with the specific positional information necessary for the automated construction of tracheidograms. The package includes functions to read and visualize tracheid anatomical data, and uses local search algorithms to ascribe a ranked position to each tracheid in identified radial files. The package also provides functions to ensure that tracheids are adequately aligned for identifying the first tracheid in each radial file, and obtaining the correct ranking of tracheids along each radial file. Additional functions allow automating the analyses for multiple samples and rings (batch mode) and exporting data and plots for quality control. RAPTOR allows tracheidogram users to take advantage of the latest generation of cell anatomical measuring systems. With this R-package we aim to facilitate the construction of more robust and versatile tracheidograms for the benefit of the research community.

Affiliations: 1 Humboldt-Universität zu Berlin; 2 DeLaWi-TreeRing Analyses; 3 Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences

Abstract: The project “Trees as Indicators of the Urban Heat Island” examines different tree species in Berlin, the city with the most pronounced Urban Heat Island (UHI) effect in Germany. Using dendroecological techniques, the main objective of the project is to model and reconstruct the distribution and spatiotemporal evolution of UHI using tree ring-width, pointer year analysis, and additional wood anatomical features. In a first step, climate-growth relationships were analyzed for urban tree sites (n = 13) along a UHI gradient in the Berlin city district of Neukölln. These results were then compared with climate-growth relationships from a selection of ecologically distinct control sites (n = 10) in the Berlin hinterland. The control sites range from near natural (Müritz National Park) to sub-urban (Potsdam-Telegrafenberg, and Stadtforst Köpenick). The tree sites UHI intensity was characterized by local microclimatic conditions. Both urban (acer, ash, lime) and forest tree species (beech, oak, pine, etc.) were considered in the analysis, in order to exclude species-specific differences and to emphasize visible urban-rural diversities. Two cores each from ~350 trees were collected with increment borers, the surfaces prepared, ring-widths dated and measured, and data processed following standard dendrochronological techniques. Standardization and detrending of all mean tree curves used 5- or 13-year moving averages resulting in tree-ring indices with eliminated long-term growth trends and enhanced year-to-year variations. Linear correlations between tree ring-width and temperature and precipitation data derived from climate stations at Berlin-Dahlem, Berlin-Tempelhof, Potsdam and Carpin were used to identify climate-growth relationships. Trees of urban sites correlated stronger with climate parameters compared to trees in the Berlin hinterland. Therefore, urban trees hold a great potential to differentiate UHI-signals. Deciduous trees (oak, beech, and ash) can be used for analyses of urban summer signals, whilst pines are suitable for the analyses of late winter-early spring signals. To determine the spatiotemporal resilience of species and sites to extreme weather conditions, such as drought, flood, heat waves and frost, an extreme year analysis was also performed within the UHI study area. This pilot-study offers a unique perspective to analyze retrospective evidence for the evolution of the Berlin UHI with respect to regional climate change, demographic and urban planning in the German capital.

Affiliations: 1 Helmholtz Centre Potsdam, German Research Centre for Geosciences GFZ, Section 5.2  Climate Dynamics and Landscape Evolution, D-14473 Potsdam, Germany; 3 German Archaeological Institute DAI, D-14195 Berlin, Germany; 4 Humboldt University, Department of Geography, D-12489 Berlin, Germany

Abstract: Tree rings of pine (Pinus sylvestris L.) from roof truss and storey ceiling construction wood of Boitzenburg palace in north-eastern Germany (53°12`2” N,13°33`13” E) have been utilised to establish an annually-resolved stacked multi-parameter tree-ring chronology of tree-ring width and stable isotope ratios of carbon and oxygen dating back to 1570AD. With this study we contribute to the growing body of tree-ring studies using historical and archaeological timbers from long been populated central European lowlands for better understanding hydroclimatic regimes. The study region is characterized by sub-oceanic to sub-continental climate with average annual air temperatures between ~8.3 °C (Angermünde 1961–1990) to 9.3°C (Falkenberg 1981–2010) and a mean annual precipitation sum of 520-552mm (Prenzlau, Angermünde) (Wulf et al. 2017). We have compared our records with numerous hemispheric, European and regional proxy and historical climate records. The results show some agreement between our regional tree-ring width chronology and the tree-ring width derived spatial reconstruction of the Palmer drought index (scPDSI) of the “Old World Drought Atlas (OWD)” by (Cook et al. 2015). However, our climate sensitive isotope records do only partly agree with the existing reconstructions pointing to uncertainties, some lack of understanding and/or specific regional climate features that will be discussed. References: Cook, E. R., Seager, R., Kushnir, Y., Briffa, K. R., Buntgen, U., Frank, D., Krusic, P. J., Tegel, W., van der Schrier, G., Andreu-Hayles, L., Baillie, M., Baittinger, C., Bleicher, N., Bonde, N., Brown, D., Carrer, M., Cooper, R., Cufar, K., Dittmar, C., Esper, J., Griggs, C., Gunnarson, B., Gunther, B., Gutierrez, E., Haneca, K., Helama, S., Herzig, F., Heussner, K. U., Hofmann, J., Janda, P., Kontic, R., Kose, N., Kyncl, T., Levanic, T., Linderholm, H., Manning, S., Melvin, T. M., Miles, D., Neuwirth, B., Nicolussi, K., Nola, P., Panayotov, M., Popa, I., Rothe, A., Seftigen, K., Seim, A., Svarva, H., Svoboda, M., Thun, T., Timonen, M., Touchan, R., Trotsiuk, V., Trouet, V., Walder, F., Wazny, T., Wilson, R. & Zang, C. 2015: Old World megadroughts and pluvials during the Common Era. Science Advances 1, 9. Wulf, M., Jahn, U., Meier, K. & Radtke, M. 2017: Tree species composition of a landscape in north-eastern Germany in 1780, 1890 and 2010. Forestry 90, 174-186

Affiliations: 1 Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences

Abstract: We implemented a comprehensive long-term monitoring of sedimentological, meteorological and hydrological parameters at Lake Tiefer See – a hardwater lake in Northeastern Germany – in 2012. An increasing interest in paleoclimate research is the better understanding and calibration of δ18O values of seasonally dependent precipitated calcium carbonate in lake sediment records. For the study of modern controls of the lake δ18O system we analyze the stable isotopic composition on the one hand of the lake water in different water depths and on the other hand of the rain water on a monthly basis. These monitoring results will allow the improvement of paleoclimate reconstructions of the varved lake sediments in Lake Tiefer See.

Affiliations: Freie Universität Berlin

Abstract: Within the Virtual Institute of Integrated Climate and Landscape Evolution Analyses (ICLEA) high-resolution geo-archives (e.g. lakes as natural data loggers) of the northeastern german and northern polish lowlands are investigated to identify influences of land-use on the landscape evolution. For the annually laminated lake sediment record of Lake Tiefer See, situated in rural environments in NE-Germany, we present a detailed heavy metal enrichment history for Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, and Zn for the last two hundred years at 1 cm intervals. Heavy metal concentrations are compared to the so-called index of geoaccumulation (Müller 1979), which is based on the average global metal content in shales (Turekian and Wedepohl, 1961). Only Pb, Zn and Cd show a clear parallel pattern of enrichment starting around 1850 according to mainly atmospheric input due to increasing industrialization within the framework of the Industrial Revolution. Highest input for Cd, Zn, and Pb occur around 1960 to 1980 and thereafter a clear pattern of declining anthropogenic input is registered. On the base of heavy-metal analysis of pre-industrial sediments and different sediment types (e.g. calcareous gyttja, organic gyttja etc.) the local and specific geogenic background values for various metals are determined. These results provide means to calculate and quantify with sub-decadal resolution anthropogenic heavy metal accumulations and enrichment factors as well as to define regional measures for a state of reference, reflecting natural conditions without human impact. Müller, G. (1979): Schwermetalle in den Sedimenten des Rheins – Veränderungen seit 1971. Umschau 79: 778-783. Turekian, K. and Wedepohl, K. (1961): Distribution of the elements in some major units of the earth’s crust. Bull.Geol.Soc.Am. 72: 175-192.

Affiliations: Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences

Abstract: Flood deposits in annually laminated lake sediments provide highly resolved records of past hydrological extreme events. However, numerous sedimentary processes during the erosion, transport and deposition of sediments affect the sedimentary structure of flood layers, complicating the reconstruction of flood magnitudes from records of flood deposits. Here, we show that the deposition of flood layers in the record of Lake Ammersee is followed by characteristic changes in sediment composition of subsequent varves, indicating a time-dependent adjustment of catchment processes to hydrological extremes. We present a novel methodological approach to quantify catchment response times based on inverse modelling of 2D µ-XRF maps of thin section slabs. Comparing our results to instrumental time series of hydrological data implies a link between inferred response times and flood magnitude in the record of Lake Ammersee. We discuss implications for landscape responses to extreme climate events and the applicability of long-term flood records for quantitative paleoclimatic reconstructions.

Affiliations: ¹ Czech University of Life Sciences; ² Czech University of Life Sciences Prague; ³ Forestry and Game Management Research Institute, v.v.i

Abstract: The region of northern Bohemia and adjacent areas of Saxony and Poland are one of the most polluted areas in central Europe, mainly for the frequent occurrence of high concentrations of SO2 and NOX originating from anthropogenic activities. The Klínovec study area was selected as a model one considering the combination of harsh climate and long-term air pollution. Anatomical parameters of wood can be used as a sensitive indicator of environmental changes with even better than annual resolution. Thus methods of quantitative wood anatomy were used to study the dynamics of xylem traits for spruce growing in this area. The series of anatomical features such as lumen area and cell wall thickness were examined separately for earlywood and latewood for the entire tree life span and correlated with climate and pollution time series. Cell wall thickness in latewood part was detected as the most responsive parameter. Keywords: Black Triangle, air pollution, quantitative wood anatomy, earlywood, latewood, lumen area, cell wall thickness, environmental change

Affiliation: Swedish University of Agricultural Sciences, Personal website

Our conceptual model of the processes controlling water quality has been in large part driven by infrequent (biweekly or monthly) and relatively short term (less than a decade) observations, many of which have been collected by discrete sampling of river mouths. Furthermore, those of us in the water quality modelling community have adopted model structures that work well for hydrological simulations but are not consistent with either measurements or biogeochemical process conceptualization.

While simulations developed on the basis of mesoscale infrequent, relatively short term measurements and model structures designed for simulating flood discharge peaks are widely used for catchment management, there are increasing concerns about their fitness for purpose. Predictions based on obsolete process understanding and questionable model structures are likely to result in misleading conclusions and poor management decisions.

The process understanding incorporated in water quality models is being challenged at both ends of the temporal spectrum. On the one hand, high frequency and other in-situ measurements offer new awareness of short term temporal variability, while on the other hand, ever longer observational time series raise questions about stationarity and the “elusive baseline”. Here, I will discuss how observations from both ends of the temporal spectrum can improve our conceptualization of catchment biogeochemical processes, and how this new understanding can inform the structure of a new generation of water quality models.

Affiliations: 1 Université de Rennes 1; 2 Univ Rennes, CNRS, Géosciences Rennes – UMR 6118; 3 Swedish University of Agricultural Sciences, Uppsala; 4 Univ Rennes, CNRS, Ecobio; 5 IRSTEA Lyon

Abstract: The evolution of ground- and surface water quality is a key-point for future generations and its prediction is crucial to advise public policies. Yet, while some catchments are highly instrumented, in most areas very little information is known. This work identifies a few crucial parameters that are needed to make predictions, and investigates a way to evaluate these parameters at a large scale. Predictions of nitrate concentrations following several future nitrate input scenarios were made using a three-dimensional groundwater flow model of a 35 km2 catchment located in Brittany, Western France. The flow model was calibrated with age tracer data (CFC) collected in wells. Oxygen and nitrates degradation rates were further inferred using reactive tracer data (NO3-, O2, N2 excess). To test the sensitivity of the predictions to natural characteristics (groundwater residence time, biological reactivity) and anthropic factors (space, time and magnitude of nitrogen inputs), we degraded step by step the information contained in the model. This allowed us to determine the key parameters controlling the water vulnerability to nitrate pollution. Evaluating these parameters based on in-stream measurements should give a spatially integrative information, as streams appear to be aquifer outlets. Thus, we performed spatially distributed radon, silica, nitrate, O and N isotopes, and reactive dissolved gas (N2, N2O, O2) measurements over the catchment. This allowed us to infer groundwater circulation and biogeochemical activity in the aquifer. However, using river measurements rises several questions, especially when it comes to gaseous tracers, that are subject to exchanges with the atmosphere. To quantify these gaseous exchanges, we performed continuous helium injections and measured the resulting concentrations using membrane inlet mass spectrometry (MIMS). This showed that fine-scale stream-bed morphology controls degassing rates, inducing a high spatial heterogeneity of stream-atmosphere exchanges. Moroever, continuous acetate and nitrate injections showed that denitrification is not occurring in the stream channel, which is consistent with the high oxygen concentration of stream water. This approach gives an integrative view of the processes controlling water vulnerability to nitrate pollution. In an upscaling perspective, we aim to define a measurement protocol as simple as possible, suitable for studies at regional scale.

Affiliations: 1 Université de Rennes; 2 CNRS; 3 BRGM

Abstract: Predicting matter fluxes in the Critical Zone relies on our ability to observe and characterise the physical and biogeochemical processes governing solute migration and attenuation. Owing to the variety of these processes in the Hydrosphere and to the complexity of reproducing natural conditions in laboratory, in situ experiments accounting for both the physics and the biogeochemical reactivity of solute transport are essential to acquire an integrated understanding of matter fluxes at field scale. For instance, the reduction of nitrates by autotrophic denitrification is a common biologically mediated reaction in groundwater which results from the widespread release of high nitrate concentrations in natural waters stemming from the intensive application of nitrogen fertilisers in agriculture. Characterising nitrate transport and denitrification kinetics is therefore essential to manage the environmental and human health issues. In this study, we propose to characterise simultaneously the physical and the biogeochemical processes controlling solute transport in a fractured porous media using a forced-gradient tracer experiment. In this respect, a radially convergent tracer test was conducted using a combined slug injection of dissolved noble gas tracers (He, Ar, Xe) for the characterisation of physical transport processes, and nitrates (NO3-) for the investigation of subsurface reactivity. This approach is complemented by an intensive sampling strategy which allows the continuous monitoring of reactive dissolved gases (N2, O2, CO2, CH4, N2O, H2), groundwater chemistry (field parameters, anions, cations & trace elements), stable isotopes of nitrate and sulphate (15N/14N, 18O/16O, 34S/32S) as well as microorganism communities (metagenomic analyses). Dissolved noble gas data enabled the characterisation of the dominant physical transport properties of a fractured media. These findings enabled the subsequent modelling of the nitrate breakthrough curve and the derivation of a reactive transport model based on a first order kinetic reaction of denitrification. The continuous monitoring of groundwater chemistry and microbiology allowed the identification of pyrite and biotite as reactive minerals involved in the denitrification reaction which is mediated by autotrophic microorganisms. The sudden availability of nutrients in the anoxic fractured media resulted in significant modifications in microorganism communities particularly a fast and ample growth of denitrifying and sulphur-oxidising bacteria. Therefore, this study consist in an original experiment opening new opportunities for the simultaneous characterisation of the physical, chemical and biological controls on matter fluxes in the Hydrosphere.

Affiliations: 1 KU Leuven; 2 KULeuven

Abstract: The intensive agricultural sector in the Flemish region of Belgium does so far not fully achieve the objectives of the European Water Framework Directive. In 2016, the agricultural sector was still the main emitter of nitrates (66%) and phosphates (40%) into the surface and groundwater bodies. This observation lead to strengthened limitations on the application of organic and mineral fertilisers through a renewed Manure Action Plan which targets that not more than 5% of the 760 measurement stations would exceed the 50 mg NO3-/l concentration in 2018. To estimate diffuse losses to surface and groundwater of nitrates and phosphates from both mineral and organic fertilizers, as well as from atmospheric deposition, and hence to enable the ex-ante evaluation of regulatory measures, a spatially distributed, mechanistic Nutrient Emission Model for agriculture (NEMO) was developed (Van Opstal et al., 2014). NEMO operates on a grid with a spatial resolution of 50m × 50m where each cell is either a ‘land cell’ or a ‘river cell’. The objectives of this research were to analyse how the consideration of agricultural drainage ditches, so far not accounted for by NEMO, would affect the accuracy of the modelled runoff and nitrate loads whereby ditch trajectories and dimensions were extracted from LiDAR point clouds. Incorporating the drainage ditches in the classification of cells as land or river led to an increase in the ratio of river cells to land cells from a 1:10 to a 1:4. Extracted cross-sectional profiles of all water courses including ditches made it possible to substitute the arbitrary pre-set drainage depth of the river cells by a spatially-distributed value derived from LiDAR data. As a result of modified hydraulic gradients, NEMO attributed less water to drainage runoff and more to groundwater flow. As the travel time of the groundwater to the river cells decreased, lower levels of denitrification were calculated leading to higher modelled nitrate loads in the surface water with more catchments exceeding the target of 50 mg NO3-/l, however not always improving the model accuracy. Given the observed sensitivity of the NEMO output to the extent of the drainage network and the drainage depths, there is a potential to obtain more accurate runoff and load predictions, but this requires adjusting NEMO to work with fine grid cells. References Van Opstal et al. (2014). In: EGU General Assembly Conference Abstracts. Vol. 16.

Affiliations: Helmholtz Centre for Environmental Research - UFZ

Abstract: Streams and rivers not only deliver nutrients to the catchment outlets or estuaries, but also transform and store them during the traveling time in the river network. The in-stream processes are strongly impacted by hydrological, morphological and biogeochemical characteristics, which results in high spatiotemporal variability of in-stream nutrient dynamics throughout the river network. Moreover, in-stream nutrient uptake is also impacted by factors driven by terrestrial processes. With such high level of complexity, nutrient uptake at river network scale remains challenging. The nutrient spiraling concepts have widely been accepted in lotic biogeochemical research. Tracer experiments (e.g., using 15N isotope) have revealed advanced understanding of in-stream processes. However, those experiments were mostly conducted in small headwater streams due to the experimental constraints and thus the representativeness for large rivers is dubious. Instead of direct tracer experiments, stream metabolism is highly correlated to nutrient uptake. Benefiting from high-frequency sensor monitoring, stream metabolic data open a door for nutrient uptake investigation, especially in larger streams and rivers. As a part of the TERENO-Bode observatory, since 2010 multi-parameter high-frequency sensor monitoring has been deployed at two stations (Meisdrof and Hausneindorf, representing forest and agricultural streams, respectively) in the main Selke River. We analyzed five years’ continuous data in a 15min interval, including discharge, temperature, pH, dissolved oxygen and nitrate concentration, from which nitrate diel amplitude and daily metabolism rates (gross primary production, GPP) were obtained. Moreover, daily assimilatory nitrate uptake rate (Ua) was calculated based on the Ua and GPP correlation. The Ua of forest and agricultural streams observed totally different seasonal patterns due to different riparian conditions. Based on the daily Ua data, a parsimonious approach of Ua calculation was derived considering the global radiation, riparian shading effect and the potential uptake rate (a general parameter). The data were used to validate the performance of the new approach and to determine the potential uptake rate. Results showed that the approach well reproduced seasonal patterns in both streams (R2 = 0.45 and 0.47 at Meisdorf and Hausneindorf, respectively). We further integrated the approach to a fully distributed catchment nitrate model (the mHM-Nitrate model). mHM-Nitrate can provide detailed information for each stream and therefore is a promising tool to upscale the Ua calculation to the whole river network. Results showed that the spatiotemporal variabilities of Ua and the uptake percentage (uptake amount/nitrate load) were both high, but the latter was more pronounced.

Affiliations: Technical University of Crete

Abstract: Koiliaris CZO (Critical Zone Observatory) is a Mediterranean watershed with intensive agricultural activities over many centuries. This human-induced land use change plays a significant role in the hydrologic response, water quality and water management of the watershed. It is mainly driven by socioeconomic pressures (livestock farming, tourism and population growth) exhibiting significant spatial variation. Additionally, since many areas in the Mediterranean are under imminent threat of desertification due to climate change, one cannot overlook the effect of precipitation and temperature variations through periods with water scarcity or flash flood events. The objective of our proposed methodology was to assess Karst-SWAT (a modification of SWAT model for simulating the hydro-chemical response of karstic systems) over a period of 58 years (1961 – 2018) by conducting a historical reconstruction of land use change of Western Crete for that period and relate it to climatic, hydrologic and agricultural data in order to identify the interrelationships between land and climate change and the effect on water quality and availability. The results of the hindcasting simulation show a very good fit between the observation data and produced karst and surface runoff. The system response presents a positive delayed trend due to the large retention time of the Karst. Incorporating the changes in land use did not alter the hydrologic simulation of the model. On the other hand, the impact of land use change was apparent in the nitrate simulation. The impact of livestock grazing was significant on the quality of the karstic groundwater and surface runoff. This methodology can be extended in large areas in order to simulate the impacts of land use change.

Affiliations: 1 Helmholtz-Centre for Environmental Research - UFZ

Abstract: Increased anthropogenic input of nitrogen (N) to the biosphere during the last decades resulted in increased groundwater and surface water concentrations of N (primarily as NO3-N) causing exceeded drinking water limits and eutrophication as a global problem. Although political measures were implemented to reduced N input mainly in agricultural land use, these decreases were not directly reflected in decreasing riverine NO3-N concentrations and fluxes. Limited riverine responses can be caused by the accumulation of the bios- and hydrosphere as legacies as well as by long time lags between implementation and potential success of measures. Here we analyze a 42-year time series of nitrate concentration in a mesoscale catchment in Central Germany to jointly assess the catchment scale N budget, the potential and characteristics of an N-legacy and on the apparent travel time of N. We show that the catchment has a retained up to 84 % of the N-input, which can either be a still stored as a legacy within the system or be denitrified in the terrestrial or aquatic system. Elevated riverine nitrate concentrations also nowadays as well as a chemostatic export behavior rather hint to an N-legacy than to denitrification. Log-normal travel times of N that links input to the riverine export history differed seasonally, with modes spanning 7 – 17 years: Under low-flow conditions, travel times where systematically longer than with high discharges. We assume that the migration of the temporally varying N-input through the subsurface explains seasonal shifts in nitrate concentrations and concentration variability in the surface water: Inter-annual concentration changes are controlled by the seasonal activation and deactivation of subsurface flow paths with differing ages and nitrate loadings. Long time-lags and the lag of a dominant denitrification in the catchment let us conclude that catchment management needs to focus on a reduction of N-input to favor a depletion of the cumulated legacy. However, measures addressing the recent N leaching are highly recommended. Long-term investigations to understand the spatial and temporal variation of legacy build-up, denitrification and travel times as well as their controlling factors are needed to improve water-quality in anthropogenically impacted catchments.

Affiliations: 1 Lhyges

Abstract: Natural solid flux in rivers can be strongly disturbed by anthropogenic activities including urbanization, mining, agriculture, forestry and more generally land-use planning. Impact of forestry management on erosion flux can be due to trees logging but also to forest roads, skid trails, stream-crossings required for silvicultural operations. The impact of forestry management on solid exports in mountainous environment has been studied in a small watershed (0,8Km²) located in the Vosges massif. The Strengbach catchment (Observatoire Hydro-Géochimique de l’Environnement) was concerned by a partial trees harvesting between July and August 2014. This small extended forestry operation (2,3% of the catchment area) involved trees logging and implementation of a skid trails network including poor stream-crossings. The bedload flux was estimated since April 2009. The suspended sediment (SS) flux was evaluated on the basis of stream water samples collected every 16 hours and during high-flow events since December 2012. Before forestry operation, the mean bedload flux was 2,5T/yr+/-8,3% for a mean outlet runoff of 730mm/yr, although the SS flux was 7,7T/yr+/-9,7% for an outlet runoff of 948mm/yr. The forestry operation occurring in 2014 has involved a significant and quasi-immediate impact on SS concentrations and fluxes. As an illustration, the mean SS concentration of the stream was 129mg/L (outside high-flow periods) after the forestry operation beginning, whereas it was only 6,2mg/L just before. In addition, the forestry operation led to a SS flux approximately 5 to 6 times larger than expected for the July-August 2014 period. The impact on annual SS flux was even significant during two hydrological years, with an increase of +100% and +50% for 2013-2014 and 2014-2015, respectively. This relatively high disturbance is mainly due to the occurrence of non-improved stream-crossings and skid trails, responsible to the introduction of a huge amount of fine soil particles into the stream. At the opposite, the bedload export was slightly lower than usual in 2013-2014 whereas it was 2 times higher the following year. This delay of trees harvesting impact on coarse sediment export can be explained by the trapping of bedload by logs constituting stream-crossings during forestry operation. After the logs removal, the trapped bedload needed several flood events to reach outlet, explaining the delay. Overall, a post-logging recovery time of approximatively 10 months can be assumed for solid exports following 2014 forestry operation.

Affiliations: Helmholtz Zentrum für Umweltforschung GmbH - UFZ

Abstract: The challenge of the presented project is to characterize and quantify large (regional) scale dynamics and trends in water and solute fluxes from the mesoscale Holtemme River catchment located in the TERENO (Terrestrial Environmental Observatory) test site of the Bode region in the Harz Mountains, Germany. Spatially and temporally variable and often superimposing processes like mobilization and turnover of N-species strongly affect nitrate fluxes at catchment outlets. It remains thus challenging to determine dominant nitrate sources to derive an effective river management. Here, we combine data sets from two spatially highly resolved key-date monitoring campaigns of nitrate fluxes along a mesoscale catchment in Germany with four years of monitoring data from two representative sites within the catchment. The study area is characterized by a strong land use gradient from pristine headwaters to lowland sub-catchments with intense agricultural land use and wastewater sources. Flow conditions were assessed by a hydrograph separation showing the clear dominance of base flow during both investigations. However, the absolute amounts of discharge differed significantly from each other (outlet: 1.42 m³ s-1 versus 0.43 m³ s-1). Nitrate concentration and flux in the headwater was found to be low. In contrast, nitrate loads further downstream originate from anthropogenic sources such as effluents from wastewater treatment plants (WWTP) and agricultural land use. The agricultural contribution did not vary in terms of nitrate concentration and isotopic signature between the years but in terms of flux. The contrasting amounts of discharge between the years led to a strongly increased relative wastewater contribution with decreasing discharge. This was mainly manifested in elevated δ18O-NO3- values downstream from the wastewater discharge. The four-year monitoring at two sides clearly indicates the chemostatic character of the agricultural N-source and its distinct, yet stable isotopic fingerprint. Denitrification was found to play no dominant role only for controlling nitrate loads in the river. The spatially highly resolved monitoring approach helped to accurately define hot spots of nitrate inputs into the stream while the long-term information allowed a classification of the results with respect to the seasonal N-dynamics in the catchment.

Affiliations: 1 Helmholtz Zentrum für Umweltforschung GmbH - UFZ; 2 Helmholtz Centre for Environmental Research ˗ UFZ; 3 Helmholtz Centre for Environmental Research ˗ UF

Abstract: Nitrogen models on catchment scale have key value identifying terrestrial nitrogen transport and transformation within river networks and evaluating water quality status under current and future land and climate changes. An unresolved problem in applying catchment water quality models is their often unknown spatial validation because of spares spatial measured data. For this reason, the semi-distributed hydrological water quality model HYPE (Hydrological Predictions for the Environment) is used for the evaluation of its spatial capability to represent the measured nitrogen loads at internal gauging stations, where originally the model was not calibrated. The nested meso-scale Selke catchment (463 km2, central Germany) was used as a demonstration site in this study. First, the runoff and inorganic nitrogen (IN) concentrations were simulated at three main stream gauging stations, (Silberhuette at the upper part mainly forest, Meisdorf in the middle covering two types of forest and Hausneindorf at the outlet covering the agriculture part), using multi-objective and multisite calibration approach. In addition, the DREAM (differential evolution adaptive Metropolis) algorithm, which is based on Markov chain Monte Carlo (MCMC) approach, was used for multi objective and multi-site calibration and uncertainty analysis. Then, the model was validated at seven internal stations (1994-2014), where the IN concentrations were biweekly to monthly monitored. Results showed that runoff was well reproduced for both calibration (1994-1998) and validation (1999-2014) periods at all three main gauging stations, with lowest Nash-Sutcliffe efficiency (NSE) of 0.75 and percentage bias (PBIAS) of less than 18%. There is a slight decrease in the model performance during the validation period, which can be explained by reduction of number of precipitation stations compared to the calibration mode. The dynamics of IN loads were well represented by the model during the whole simulation periods with the lowest NSE and PBIAS of 0.71 and 9%, respectively. Results confirmed that the seasonal dynamics of IN concentrations were controlled by the combination of both effects of hydrological and biogeochemical processes. Results of spatial validation of the HYPE model at internal stations indicated that the IN loads were well represented by the model during the period from 1994 to 2014 with NSE and PBIAS (%) for the seven internal stations of about: 0.92, 0.91, 0.87, 0.841, 0.84, 0.81, 0.79 and -18, 3.1, -12.8, -6.8, 13, -15 -27, respectively. The model could represent better the IN concentrations in the diffuse source-dominant sub-catchments compared to the point-source dominants sub-catchments.

Affiliations: 1 University of Potsdam, 2 GFZ

Abstract: The awareness of plastic debris in surface waters as environmental problem has been growing though a detailed quantification of its occurence in inland waters and understanding of underlying processes is still lacking behind. A major role is expected by urban areas that potentially emit a diverse range of microplastic particles from dynamic anthropogenic sources and in response to precipitation events. In our study, jointly by the Environmental Science Unit of the University of Potsdam together with the Helmholtz-Centre Potsdam - GFZ German Research Centre for Geosciences, an urban watercourse in Berlin, with several possible anthropogenic sources of microplastics, was investigated by surveys during several months. This was facilitated by a novel approach applying short-wave infrared imaging spectroscopy that allows for automated detection of filtered microplastic particles and recognition of different plastic types by their spectral absorption features. The method was refined by using environmental samples for testing and then used to detect number, size, and type of larger microplastic particles in the water samples taken. Concentrations of microplastic particles at the very upper end of reported ones were found and identified mainly as polyethylen material. The results show significant increases in microplastic particle concentrations downstream of urban areas and after precipitation events as well as a significant increase in microplastic particle concentrations for the wastewater treatment plant located furthest upstream, though being statistically non-significant for the two downstream ones.

Affiliations: 1 IRSTEA - UMR TETIS

Abstract: This work examines a methodology for hydrograph separation from high frequency water quality data (CRITEX - ORACLE Observatory). The hypothesis behind the work is that, at small time scales, data can be considered as the result of the mixing of two water systems : * A steady and stable system linked to slow transfers (baseflow) * A variable system linked to fast transfers (quick flow) This two water systems can be characterized by a regression between concentrations and flow repartition. With the high-frequency measurements, which give exhaustive and complete chemical information, we have shown that this simple two-system model allows us to effectively perform a hydrograph separation.

Affiliations: Helmholtz Centre for Environmental Research - UFZ

Abstract: River networks play important role in nutrient retention before nutrient export to ocean. Stream restoration changes the geomorphic features, the hydrologic properties and biochemical activities in the river networks thereby influence nutrient retention in river network. However, there is little quantitative information about how stream restoration impact on nutrient retention in whole river networks. Based on spatially distributed mesoscale hydrological model (mHM), coupled with nitrite module, the influence of stream geomorphology on nutrient retention at river network scale was investigated in Bode catchment. Thus, knowledge of that is essential for more effective restoration and water quality management at river network scale.

Affiliations: 1 Institute for Bio and Geosciences - Agrosphere, Research Centre Juelich GmbH, Juelich, Germany, 2 Center of Development Research, University of Bonn, Bonn, Germany

Abstract: WASCAL (West African Science Service Center on Climate Change and Adapted Land Use) is a large-scale research-focused program designed to help tackle the climate change challenge and thereby enhance the resilience of human and environmental systems to climate change and increased variability in West-Africa. Within WASCAL a large number of heterogeneous data are collected. These data are coming from different initiated research activities and hydrological, meteorological, remote sensing, biodiversity and socio economic observation networks within WASCAL, the activities of the WASCAL Competence Center in Ouagadougou, Burkina-Faso, which carries out research and provides science-based advice to policymakers and stakeholders on climate change impacts, mitigation, and adaptation measures, and other external research activities within West-Africa. The WASCAL Data Management Infrastructure (WADI) was implemented and setup to manage, store, publish of and provide online-access to data directly connected to WASCAL, like the data from the individual observation data networks, basic data sets and research results. It consists amongst others of a file storage system, a central metadata catalogue, a time series management system and a web-portal to enter, query and access data. Special attention was laid on the usage of open formats, open source software, OGC compliant web services, which supply standards and interfaces to search in and to access to spatial (and non-spatial) data. Consequent usage of these standards guarantees an interoperable access and allows in addition the integration of data from other data holders without bigger expenditure but also ensures retrievability, readability and interpretability for the long term preserved data. Most of institutions supplying data to WASCAL face a lack of technical and human resources to host individual data infrastructures, able to manage and to exchange “own” data and metadata with WASCAL as well as with third parties by using up-to-date, OGC-compliant web-services. Therefore, most data are currently being uploaded directly into the data infrastructure located at the WASCAL Competence Center. This classifies WADI as a centralized spatial data infrastructure (SDI), with the WASCAL Competence Centre as central node. In the further development of WASCAL, the vision of a distributed SDI in the region might be achieved, giving the participating institutions an extended autonomy in managing and distributing their data. Research to find out the most suitable SDI model for the region is a necessary precondition.

Affiliations: 1 Université Grenoble Alpes, CNRS, IRD, Grenoble-INP; 2 Université Grenoble Alpes, CNRS, IRD, Météo-France, Irstea; 3 IRSTEA

Abstract: OZCAR, the French Critical Zone Research Infrastructure (RI) gathers observatories sampling various compartments of the Critical Zone. Given their long history, most of them have developed their own data/metadata portals. There is however a need to gather all the data in a common Information System (Theia-OZCAR IS) to make the data visible, findable and easy to explore; allow their preservation and their citation; favor their reuse and sharing; be interoperable with European infrastructures; respect the international standards. The project team visited the 22 observatories and 6 data centers to compile their needs. The consultation pointed out heterogeneous levels of data management maturity, variable names nomenclature and data exposition; and that finding data by variables is the point of interest of future users. The starting postulate is that databases stay next to data producers, which ensures the best data quality. A continuous information flow between observatories and the Theia-OZCAR IS will keep information up to date. Suppliers push data and metadata to the Theia-OZCAR IS using an extraction script maintained locally. To build the system, the following tasks were undertaken: 1. Work on variables names and categories that will serve as data search criteria • List all the variables measured by the observatories and match them with the GCMD Earth Science hierarchical vocabulary. About 300 in-situ measured variables of which half are chemical measurements were listed. • Identify the relevant published thesauri for the field (AGROVOC, EARTh Thesaurus, GACS Core Beta 3.1, NAL Thesaurus, UNESCO Thesaurus, ANAEE, EnvThes) to promote semantic interoperability, • Build a controlled vocabulary based on the GCMD hierarchical vocabulary and linked to the previous thesaurus concepts via SKOS semantic relations; 2. Work on metadata needed to set up interoperability and citation services • Identify the metadata proposed by different standards (ISO 19115 / Inspire, DataCite, DCAT, schema.org, SensorML, O&M, DOI) in order to establish a minimum set of metadata to be exchanged with the data providers; 3. Work on data flux between system • Define a pivot format that can convey all the metadata identified and adapted to the uses; The pivot format is currently discussed with data bases managers and will be progressively implemented in order to get a first prototype of the data portal by the end of 2018. The communication will present the approach that can be reused by other disciplines/data portals.

Affiliations: Helmholtz Centre Potsdam – GFZ German Research Centre for Geosciences, eScience Centre

Abstract: The Helmholtz research infrastructure MOSES (Modular Observation Solutions for Earth Systems) is a novel „system-of-systems“ for Earth Observation, designed to decipher the interactions of short-term events and long-term trends in Earth and environmental systems. The Helmholtz Observatories provide essential data and concepts to monitor long-term trends. MOSES now expands these capabilities towards short-term events and their interaction with long-term trends across different compartments. Data management platforms play an increasing role in the advancement of knowledge and technology in many scientific disciplines. Through high quality services they support the establishment of efficient and innovative research environments. Well designed research environments can facilitate the sustainable utilization, exchange, and re-use of scientific data and functionality by using standardized community models. Together with innovative visualization, these concepts provide added value in improving scientific knowledge-gain, even across the boundaries of disciplines. For MOSES, a conceptual design has been devised that comprises an architectural blueprint for a data management platform based on common and standardized data models, protocols, and encodings. It also fosters the re-use of existing and proven software systems and components provided by individual Helmholtz Centres. Following the principles of event-driven service oriented architectures, the design enables novel processes by sharing an reusing functionality. Through the use of standards, this blueprint supports integrated approaches to include long-term sensor networks and observatories (e.g., TERENO, COSYNA) that provide complementary processes for dynamic and event-driven monitoring.

Affiliations: Lund University

Abstract: Integrated Carbon Observation System, ICOS, is a pan-European research infrastructure with a mission to provide standardized, long term, high precision and high quality observations on the Carbon Cycle and Greenhouse Gas budgets and their perturbations. ICOS is built around over 130 observation stations, each related to one of the three domains Atmosphere, Ecosystem and Ocean, and operated by ICOS 12 member countries. All collected data are processed and quality controlled at Thematic Centres (one for each domain), before being distributed via the Carbon Portal, which is the data centre of ICOS. The Carbon Portal has been designed as FAIR data system, ensuring that all ICOS data are easy to find, available for open access, fully traceable and complete with all relevant metadata, and interoperable with other (environmental) data and services. The Carbon Portal is also built as community platform for working on elaborated data products that use the ICOS data together with other data, using sophisticated computer models. Our main end user communities are scientific users from inside and outside the ICOS community - including other research infrastructures - as well as ICOS station managers and technicians; but also citizen scientists, students, the general public and policy makers are important target users. We believe that ICOS observational data and elaborated data products building on these, will play an important role in informing stakeholders and society on the progress in mitigation of Climate Change due to emission reductions of greenhouse gases and the development of the natural and managed land related GHG fluxes under the influence of ongoing climate change that feedbacks into the Earth System. The poster will summarize the main functionalities of the Carbon Portal operations, including: * the web-based services for ICOS data discovery, including an advanced search interface, on-the fly interactive visualization and data download * our RESTful services for data uploaders, including automated metadata collection and allocation of persistent identifiers * the ICOS cataloguing service & implementation of the RDF/OWL-based metadata ontology for ICOS * provisioning of dynamically created “landing pages” for data, data content types and measurement stations - formatted to be interpretable by both humans and computer-based processes * Support services for modelers and other end users, including provisioning of on-demand cloud data processing and virtual research environments

Affiliations: 1 Museum für Naturkunde - Leibniz Institut for Evolution and Biodiversity Science

Abstract: The field of biodiversity research in Germany is diverse (Schliep et al., 2016). A great variety of institutions is involved in several different biodiversity research and monitoring activities (Geschke et al., 2017). The integration of distributed data from various sources, which often is collected for different purposes, is crucial for assessing the current status and trends of biodiversity and supporting relevant policy processes (Geschke et al., subm.; Hoffmann et al., 2014). In this regard, considerable developments within biodiversity data management have been taking place recently. There are several projects running and under development that deal with concepts and infrastructures to better manage and integrate biodiversity data (Diepenbroek et al., 2014; Geschke et al., 2017). Conceptually important are the FAIR principles for data management that aim at guiding towards findable, accessible, interoperable and reusable scientific data sets (Wilkinson et al., 2016). Also, the concept of Essential Biodiversity Variables (EBVs) can support an efficient biodiversity monitoring through the integration of biodiversity data. As an intermediate abstraction layer between primary observations and indicators, EBVs should break down biodiversity as a whole into manageable state variables (Geijzendorffer et al., 2015; Pereira et al., 2013; Proença et al., 2017). For EBVs, remote sensing data can play a critical role providing global data with relatively high spatial and temporal resolutions (Pettorelli et al., 2016; Skidmore et al., 2015). The aim of our poster is to improve the knowledge on the status quo of practices and the challenges of the biodiversity research community as well as its stakeholders regarding biodiversity integration. To achieve this, a questionnaire was developed and distributed to stakeholders in Germany (https://survey.naturkundemuseum-berlin.de/en/biodiv-data-integration). First, we present what fundamental understanding of biodiversity data integration the participants have; then we focus on the ability to integrate remote sensing data in biodiversity research. Subsequently, we identify what parts of biodiversity integration in German biodiversity research already work well and what issues need to be improved. Ultimately, we present a set of recommendations for research, policy and other interest groups that could improve the situation of biodiversity data integration in Germany.

Affiliations: ¹ Environment Agency Austria; ² Consiglio Nazionale delle Ricerche; ³ BioSense Institute (BSI); ⁴ Research Centre Juelich GmbH; ⁵ Centre of Ecology and Hydrology (CEH)

Abstract: Providing long term, quality assured and reliable ecological data as input for scientific analysis workflows is one of the major aims of ecosystem monitoring and research. This allows the impact of ecological trends from climate change and/or land management policies to be detected and evaluated. In order to enable automation of the underlying scientific workflows, internet service based access to metadata and data needs to be developed using international standards. Sensor Web Enablement (SWE) and geospatial services provide tools for the provision of data that can be worked into science workflows such as R scripts. This allows greater access to data by linking data services from distributed data nodes based on computing cloud services to user workflows as well as the application of FAIR data exchange principle focusing on documentation, access and interoperability. LTER is currently developing the main components of a Common Data Infrastructure (CDI) in the eLTER H2020 project focusing on the FAIR data principles. The CDI is developed to share and publish information resulting from the LTER network in Europe. This focuses on (a) adopting standards for documentation of research objects (observation facilities and datasets), (b) the use of controlled vocabularies, (c) the provision of time series data in a standardised format, and (d) a catalogue of datasets from different environmental monitoring resources. The eLTER Information system is based on a range open source tools and applies common community standards for the integration. A common catalogue has been developed allowing the discovery of data as well as the visualisation of available time series data. The documentation of observation and experimental facilities using DEIMS-SDR acts as the basic framework and is playing a central role for file-based data while use of OGC SOS and CSW standards allow discovery, visualisation and access to ongoing environmental time-series data services. The presentation will focus (a) on the challenges resulting from a network of distributed observation sites, (b) on the strategies to harmonise and facilitate the provision and integration of data, and (c) on the applied tools and standards that could be proposed also for other infrastructure. Keywords: DEIMS-SDR, LTER, data management, metadata, time series, EnvThes, common vocabulary, data infrastructure

Affiliation: Luxembourg Institute of Science and Technology, Personal website

While bedrock controls on catchment mixing, storage, and release functions have been increasingly investigated in recent years, comparative analyses across different neighboring lithologies remains a major challenge. In the Alzette River basin (Luxembourg) we have examined 9 years’ worth of precipitation and discharge data (across 16 nested catchments), and 6 years of fortnightly stable isotope data in streamflow (for a subset of 12 catchments), to investigate catchment physiographic controls on:

• Streamflow – we studied the relationship between catchment bedrock geology and streamflow regime metrics (i.e. winter and annual runoff/precipitation ratios and average summer/winter discharge ratios);
• Storage – relying on catchment storage as a metric for catchment comparison, we derived catchment storage deficits from water balance calculations and quantified how storage and storage-discharge relationships differ between catchments and scales;
• Isotope response and catchment mean transit time (MTT) – we investigated how the standard deviation in streamflow dD and ratios of d18O amplitudes in streamflow and precipitation (AS/AP), as proxies for catchment averaged damping of isotopic signatures, and catchment MTT, relate to bedrock geology, storage and catchment area.
  

From our investigations we were able to conclude (for our region of interest) that bedrock geology controls (1) streamflow regime metrics, (2) catchment storage, and (3) isotopes response (and subsequently catchment MTT).

Affiliations: 1 University of Strasbourg; 2 EOST, University of Strasbourg, France; 3 UMR 7517 Laboratoire d'Hydrologie et de GEochimie de Strasbourg

Abstract: As gravity intensity at a given time and location depends on the surrounding mass distribution, gravimetric measurements are sensitive to water storage changes in the critical zone. Interestingly, their spatial-integrative nature makes them sensitive to the hydrological balance fluctuations of sub-catchments. In this way, gravimetry may be a key method to constrain water storage changes in catchments with complex hydrology such as mountainous catchments, where local measurements of water level in boreholes may differ from the global water storage due to local heterogeneity. Here we propose a joint analysis of hydrological balance and relative gravity measurements fluctuations in the well instrumented Strengbach catchment in the Vosges Mountains, France. This study provides new constraints on the water mass transfers within the catchment. Data consist in two years (from 2014 to 2016) of continuous meteorological monitoring provided by the OHGE observatory (Observatoire Hydro-Géeochimique de l’Environnement de Strasbourg, France) and 19 micro-gravimetric time-lapse surveys (using a Scintrex CG5® spring gravimeter) of a gravity monitoring network of 8 stations. We show that the choice of the reference station is a key parameter to maximize the coherence of observed relative gravity variations. We present two case studies indicating a fast water storage decrease under the base station (monthly temporal scale), and a longer scale (several years) water storage redistribution within the Strengbach catchment. Finally, we compare measured gravity variations to the gravimetric response predicted from an hydrological model developed by the LHyGeS (Laboratoire d’Hydrologie et de GEochimie de Strasbourg, France) which is currently only constrained by measured catchment outflow and precipitation as well as evapotranspiration estimation. Finally, we provide a first characterization of the gravity signal at the summit area of the Strengbach catchment surrounding the newly installed superconducting gravimeter iGrav® located at the gravity network reference station. The continuous gravity monitoring is operational since June 2017 and several hydrological contributions have been already identified in this dataset.

Affiliations: 1 ZALF, Institute of Landscape Hydrology; 2 Forschungszentrum Jülich GmbH

Abstract: In this study we aimed at determining the effect size of soil hydrological processes at various points in the catchment directly, based only on measured data. That information can then be used for sound model setup and calibration. For that approach a comprehensive data set of the Wüstebach catchment of the TERENO Eifel/Lower Rhine Valley Observatory was used. Here shallow soils with mostly silty clayey loamy texture developed from fractured Devonian shale and sandstone, comprising Cambisols and Planosols on the hillslopes and Gleysols and Histosols in the riparion zone close to the stream. Within the 38.5 ha headwater catchment a sensor network had been installed in a mature Norway Spruce stand, comprising numerous soil moisture probes installed at 5 cm, 20 cm and 50 cm depth. For this study soil moisture time series from 170 sensors were used, comprising daily data for a 2.5 year period without longer data gaps. Principal Component Analysis was used to characterize mean behaviour at single sites over the whole observation period, where the loadings on the components were used to assess the effect size of the respective process at single sites. In addition, Wavelet Coherence allowed for more detailed analysis during single periods. About 87% of the spatial variance of the observed behaviour at 170 sites was explained by three principal components. Thereof 69% was due to the first principal component. The time series of the scores of that component were nearly identical with the mean behaviour at all sites. Another 15% of the variance reflected the effect of clear cutting in a part of the catchment in autumn 2014. It indicated a step-wise increase of soil moisture during the first heavy rainstorms after clear cutting, nearly constant higher soil moisture in the clear cut area thereafter, and substantially reduced soil moisture depletion during the growing season in the subsequent two years compared to the remaining forest stand. The effect size of clear cutting differed considerably even between adjacent sensors and between depths. The damping of the hydrological input signal during seepage through the soil accounted for another 3% of the variance. Time series of soil moisture measured at different depths clearly differed in that regard, although exhibiting wide overlap between different depths. Loadings on that component did not exhibit any clear spatial pattern and were not related to patterns of the soil map, indicating very short correlation lengths.

Affiliations: 1 Forschungszentrum Juelich

Abstract: Understanding the dynamic response of soil moisture to rainfall is critical for water resources management in arid and semi-arid basins. However, little is known about rainfall-related soil moisture dynamics in arid high-altitude mountainous areas due to the absence of long-term, high-resolution soil moisture observations. In this study, we investigated the dynamic response patterns of profile soil moisture using data from a soil moisture monitoring network in the Qilian Mountains established in 2013 covering altitudes from 2,000 - 5,000 m above sea level. To investigate the effects of different land covers on soil moisture response, we selected data from eight soil moisture stations with similar soil texture and slope, but different land covers (scrubland, meadow, high coverage grassland (HCG), medium coverage grassland (MCG) and barren land). Several indices were evaluated to quantitatively describe soil moisture dynamics during the growing seasons of 2014 - 2016. Our results show that scrubland, meadow and HCG sites have significant different soil moisture response amplitudes, while MCG and barren land sites show similar moisture response characteristics along profile. The rate of soil moisture increment decreased significantly with depth for all land covers, except for the HCG. This pattern coincides with the distribution of the soil properties (especially hydraulic conductivity) of different land covers. The temporal pattern of soil moisture increase was highly variable along the soil profiles depending on the land cover type. In particular, the difference of response time between adjacent layers varied from negative values to 280 hours, and the duration of soil moisture increase varied from 6.5 hours to 170 hours along depth. Our results show that the soil moisture response indices are useful to quantitatively characterize patterns in soil moisture profile dynamics and provide new insights into the different soil hydrological process in the Qilian Mountains, which helps to support sustainable management of water resources in arid high-altitude mountainous areas.

Affiliations: 1 Leibniz Centre for Agricultural Landscape Research (ZALF)

Abstract: Small, non-permanent ponds of glaciofluvial origin, called cattle holes, are widely spread in the younger Pleistocene landscapes of the northern hemisphere. New investigations show that much more of the kettle holes in the Uckermark region of NE Germany are more closely connected to the groundwater than expected before. Thus kettle holes reflect a free groundwater surface at the interface between the aquifer and the topography. They are not isolated hydrological depressions and can be viewed as linked storage components of a hydrologic continuum. Therefore, these kettle holes have a high informative value regarding changing behavior of the groundwater system functioning as a suitable indicator for changes of the fundamental catchment functions. The unsolved challenge of this approach is the complexity and high abundance of kettle holes which requires an elaborate hydrological monitoring of a large number of small lakes. Therefore, an alternative approach was used to record the dynamic behavior of the hydrological system. Measurements of the stable isotopes 18O and Deuterium enables the quantitative estimation of the individual water flux and evapotranspiration rates. An isotope-mass-balance model was used to quantify lake water balances during a one year sampling period. The approach after Skrzypek et al. 2015 based on the global relationship between the 18O and Deuterium values of the precipitation – described by the Global Meteoric Water Line (GMWL) and the kinetic isotopic fractionation during evaporation which leads to a deviation from the GMWL indicated by a decrease of the slope of this relationship. Assuming that the lake is hydrostatically connected to the groundwater the used isotope mass-balance model accounts for the quantification of the evapotranspiration rate considering the groundwater inflow compensating the evaporation loss. Due to the low effort of isotopic sampling, the isotopic monitoring of a large number of kettle holes and lakes is possible, even regarding a longer period of time. First results clearly show distinct patterns of the temporal dynamics of the groundwater/surface water interaction reflecting the regional system behavior. They provide the basis for a better model calibration anticipating future development of the hydraulic system under climate change.

Affiliations:

Abstract: The dynamics of water storage in a catchment provides a fundamental insight into the interlinkages between input and output fluxes, and how these are affected by environmental change. Such dynamics also mediate, and help us understand, the fundamental difference of the rapid celerity of the rainfall-runoff (minutes to hours) response of catchments and the much slower velocity of water particles (months to decades) as they are transported through catchment systems. In this contribution we report from intensive, long-term (>10year), multi-scale isotope studies in the Scottish Highlands and Brandenburg, Germany, that have sought to better understand these issues. We integrated empirical data collection with diverse modelling approaches to quantify the dynamics and residence times of storage in different compartments of the hydrological system (vegetation canopies, soils, ground waters etc.) and their relationship between the magnitude and travel time distributions of output fluxes (stream flow, transpiration and evaporation). Use of conceptual, physically-based and probabilistic modelling approaches gave broadly consistent perspectives on storage-discharge relationships and the preferential selection of younger waters in runoff, evaporation and transpiration; while older waters predominate in groundwater. The work also highlighted the importance role vegetation plays in regulating storage and fluxes in evaporation and transpiration and how this contributes to the differential ageing of water in mobile and bulk waters in the soil compartment.

Affiliations: 1 BTU Cottbus - Senftenberg; 2 The James Hutton Institute; 3 BTU Cottbus-Senftenberg; 4 Forschungszentrum Landschaftsentwicklung und Bergbaulandschaften (FZLB) / BTU Cottbus

Abstract: Experimental catchments with well-known boundaries and characteristics may contribute valuable data to hydrological, critical zone and landscape evolution research. One of the most well-established and largest constructed catchments is the Chicken Creek catchment (6 ha area including a 0.4 ha pond, Brandenburg, Germany) representing an initial ecosystem undergoing a highly dynamic ecological development starting from clearly defined starting conditions. The water balance dynamics of the catchment was calculated using a simple mass balance approach to reveal the impact of ecological development during 12 years. Water storage in the catchment was calculated from a 3D-model of groundwater volumes, soil moisture measurements and water level recordings of the pond. The catchment water balance equation was resolved for evapotranspiration, the only part that was not measured directly. Due to the known boundary conditions and the inner structure of the catchment, we were able to quantify the different storage compartments and their role in hydrologic response. Our results indicate that for small catchments with a highly dynamic ecological development like the Chicken Creek, the knowledge about saturated and unsaturated storage volumes enables a good estimate and closure of the water balance using a rather simple approach, at least in annual resolution. We found a significant relationship between vegetation cover in the catchment and calculated ET. Time series of meteorological, hydrological, soil and vegetation data over 12 years enabled us to characterize the transient development of the catchment and to evaluate the effect of different feedback mechanisms on catchment hydrology. The dataset from the Chicken Creek catchment indicate at least three phases in ecosystem development, where initial abiotic feedbacks (e.g. erosion) were followed by more and more biotic controls (e.g. biological soil crusts, vegetation succession and growth). Data from Chicken Creek in high spatial and temporal resolution provide a valuable database underlining the high importance of abiotic/biotic feedback effects that change the hydrologic functioning and response of the catchment more than the water balance itself revealed and thus have to be included in catchment models.

Affiliations: 1 Georg-August Universität Göttingen; 2 Georg-August-Universität Göttingen

Abstract: Climate simulations suggest that Mediterranean groundwater resources, in particular, shall considerably decrease due to a projected increase in temperature and the change in temporal precipitation patterns. The importance of solid water management based on quantitative approaches involving flow and transport modelling is evident, considering the anticipated population growth and climate change effects. The quantification of recharge rates can be considered as a primary objective for the sustainable management of the Mediterranean karst aquifers. Here, we employ the fully-integrated surface-subsurface and physically-based hydrogeological modeling tool HydroGeoSphere on a high-performance-computing platform to simulate the terrestrial-based hydrological cycle of the Western-Mountain-Aquifer, located in Israel and the Palestinian Territories. The objectives of this study are (1) the estimation of groundwater recharge based on a rigorous implementation of the surface-hydrological processes, the particularities of rock-soil landscape, the characteristics of ephemeral streams and the specific climate conditions as well as the different precipitation patterns, and (2) the simulation of the effect of infiltration through a thick (several hundreds of meters) vadose zone on groundwater flow dynamics. The Western-Mountain-Aquifer catchment is subject to an autogenic recharge in the Judean and Samarian Mountains towards the west. Exposed karst features (i.e. sinkholes) may locally constitute areas of fast focused recharge. Orogenic rainfall in this region occurs at a high annual variability mainly from December to February. HydroGeoSphere simultaneously solves the 3-D Richards equation for subsurface flow for variably-saturated conditions and a 2-D diffusion-wave equation for surface routing. The surface-subsurface flow regimes are coupled via a first-order exchange term based on a thin boundary layer, allowing to naturally account for the partitioning of rainfall into diffuse and rapid direct recharge, e.g. along dry valleys or sinkholes. In addition a double-continuum approach is implemented for subsurface flow to adequately represent the duality of flow in the karst aquifers. Preferential flow paths are characteristic for karstic rocks, triggering rapid flow through fractures and conduits, while the rock matrix provides substantial storage and slow seepage. This results in a high space and time dependent variation of infiltration and subsequently a high variation of groundwater dynamics.

Affiliations: 1 University of Strasbourg

Abstract: The characterization of vadose zone processes is a primary goal for understanding, predicting, and managing water resources. In this study, the issue of soil water monitoring on a vertical profile in the small forested Strengbach catchment (France) is investigated using numerical modeling with the long-term sequences 1D-Richards’ equation and parameter estimation through an inverse technique. The forested Strengbach catchment is a small watershed of 0.8 km2 located in the Vosges Mountains (northeastern France). It has become a completely equipped environmental observatory with permanent sampling and measuring stations that have been in operation since 1986. This watershed (http://ohge.unistra.fr) belongs to the French and European networks of Critical Zone observatories. The monitoring devices have been installed in a spruce experimental stand located in the northern upper portion of the Strengbach catchment. This parcel is representative of a large part of the domain, which is 90% covered by forest, which consists of 80% spruce and 20% beech. For hydrological instrumentation, a pit of approximately 1.30 m in depth and 0.70 m in width has been implemented between the trees during September 2010. The present study is based on measurements achieved by three matric potential sensors. The Watermark probes have been installed at 42 cm, 71 cm, and 116 cm depth. The meteorological data, i.e. rainfall, temperature, humidity, wind velocity, and radiation, are monitored using an automatic weather station. All these data variables are monitored every 10 min. During the probe installation operations, many soil samples were collected and subjected to laboratory analysis, mainly hydraulic retention curves, determination of saturated conductivity and soil composition. Previous investigations in the catchment and local sampling at the bottom of the pit with a soil auger led to the consideration of a vertical profile of 150 cm depth which has been discretized into 3 layers. The scientific questions address the selection of the calibration sequence, the initial starting point for inverse optimization and monitoring frequency used in the inverse procedure to estimate van-Genuchten parameters of the 3 layers. As expected, our results show that the highly variable data period used for the calibration provides better estimations when simulating the long-term sequence. For the starting point of the initial parameters, handmade iterative initial parameters estimation leads to better results than a laboratory analysis or set of ROSETTA parameters. Concerning the frequency of monitoring, weekly and daily datasets provide efficient results compared to hourly data.

Affiliations: 1 Institut de Physique du Globe de Paris; 2 Sorbonne Université, UMR 7619 METIS; 3 Université de Rennes 1, CNRS, UMR 6118 Géosciences Rennes

Abstract: Active seismic methods are commonly used to characterize the near-surface structure of the Critical Zone (CZ). More recently, they have also proved their ability to monitor the seasonal dynamics of water in the CZ, both in fractured and sedimentary aquifers. Through the joint estimation of pressure (P) and shear (S) wave velocities, mechanical properties indeed provide innovative means of tracking down CZ water content along the continuum between the saturated and unsaturated zones. More particularly, one of the main interests of seismic methods is the ability to capture lateral variations of water storage with a single acquisition, thus providing spatial constraints to point continuous observations. To follow on with these encouraging results, we carried out in June 2018 a time-lapse seismic experiment in an alluvial aquifer connected to the Sélune river (Normandy, France) in order to investigate the short term reactivity of the system during a pumping and recovery test. 14 seismic acquisitions were repeated every 30-45 minutes along a 72-m-long profile centered on the active pumping well, with piezometric levels measured regularly next to the well during the pumping and recharge stages. Each acquisition consisted in 25 shots spread every 3 m along the line and lasted about 10 minutes so as to minimize the variations of piezometric level from the beginning to the end of the acquisition. We picked first arrival times associated with refracted P-waves and extracted surface-wave dispersion curves along the 14 different profiles. The overall 1-m variation of piezometric level occurring throughout the whole experiment is expected to lead to significant variations in observed traveltimes and phase velocities, associated with variations of water content in the CZ. Further processing steps will include inversions of traveltimes and dispersion curves to retrieve P- and S-wave velocity models, respectively, followed with rock physics modelling to estimate porosity and saturation variations in the aquifer system. Our results should thus provide interesting perspectives for the monitoring of water storage variations, both in the saturated and non-saturated areas, and at relatively elevated temporal and lateral scales.

Affiliations: 1 Leibniz Centre for Agricultural Landscape Research (ZALF); 2 Institut für Pflanzenernährung und Bodenkunde, Christian-Albrechts-Universität zu Kiel

Abstract: Landfill capping systems in its function as engineered barriers are purposed to prevent or minimize the contact of precipitation or melting water infiltration with the waste body to limit the generation of leachate or gas emissions, which may reach the aquifer system or the atmosphere. Mineral capping systems without polymers can be a cost-effective and sustainable alternative to the standard engineering constructions that must exhibit hydraulic stability during drier periods to successfully avoid crack formation resulting in bypass flow through the contaminated waste body. The soil water characteristics of the mineral capping system of the Rastorf landfill (Northern Germany) consisting of a highly compacted boulder marl of 100 cm thickness were determined with tensiometers to understand the interactions between drying and rewetting cycles while the shrinkage characteristics as function of matric potentials were also quantified to control the in situ shrinkage potential of the chosen substrate. The sealing layer in 70 to 100 cm depth as last barrier above the waste body was characterised by a proportional shrinkage behaviour or rather critical shrinkage-crack formation occurred between -300 and -500 hPa while the in situ matric potential courses varied between -70 and -100 hPa, even during drier periods between 2012 and 2016. By comparison the shrinkage potential with the continuous in situ measurements, the analysis showed that critical crack formation in the sealing layer combined with bypass flow of water and solutes has not yet been inserted. Taking account of soil physical knowledge, landfill capping systems without critical crack formation could be permanently established.

Affiliations: 1 Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences

Abstract: In spite of the fundamental role of the landscape water balance for the Earth’s water and energy cycles, monitoring the water balance and related storage dynamics beyond the point scale is notoriously difficult due to the multitude of flow and storage processes and their spatial heterogeneity. We present recent advancements in the field of terrestrial gravimetry for integrative monitoring of water storage changes, i.e., including all involved storage compartments at the field to landscape scale. In particular, we present the first outdoor deployment of an iGrav superconducting gravimeter (SG) in a minimized field enclosure on a wet-temperate grassland site for integrative monitoring of water storage changes at the Geodetic Observatory Wettzell, Bavarian Forest. About 99% and 85% of the recorded gravity variations caused by local to regional water storage variations change within a radius of 4000 and 200 meter around the instrument, respectively. Gravity variations observed by the field setup are almost independent of the depth below the terrain surface where water storage changes occur, and thus the field SG system directly observes the total water storage change in an integrative way. We furthermore give a glimpse on the recent deployment of a similar gravimeter system in the Müritz National Park within the TERENO-Northeast observatory, located directly in a mixed beech and pine forest. A framework to single out the water balance components precipitation, actual evapotranspiration or lateral subsurface discharge from the gravity time series on annual to daily time scales is presented, and related uncertainties are assessed. In addition, prospects for hydrological applications in the view of ongoing technological development of gravimeters, in particular atom quantum gravimeters, are outlined.

Affiliations: 1 Leibniz Centre for Agricultural Landscape Research (ZALF)

Abstract: For erosion-affected soils, it was unknown whether the dynamics of water retention properties depended also on erosional soil profile modifications. The objective was to separate shorter-term hysteretic from longer-term seasonal dynamics in field-measured water retention data of eroded Luvisols. Water retention data were obtained using tensiometers and TDR sensors of six lysimeter soil monoliths from two field sites. Differences in retention data were observed for differently eroded Luvisols. Retention data from 2012 to 2014 allowed identifying initial and maximum drying and wetting periods, which were fitted to obtain separate sets of parameters of the van Genuchten retention function. For field water retention curves of the initial drying, steepness increased during the intra-seasonal drying-wetting cycles, and returned to a similar level only for one site in the two following years. For the soil monoliths moved from a neighboring site, water retention was found to increase for inter-annual drying curves in the topsoil with time. Soil water retention dynamics could be identified at three temporal scales. Intra-seasonal retention dynamics suggested that re-wetting was increasingly limited drying-wetting cycles perhaps due to soil water repellency and thereby limiting the moisture content for the following drying period.

Affiliations: GFZ German Research Centre for Geosciences

Abstract: Rainfall redistribution by forest canopies differs between tree species and can play an important role for soil moisture patterns and groundwater recharge. A thorough understanding of these relationships will improve our ability to predict future impacts of climate and forest structural changes on the water balance of forest stands. In the TERENO observatory in the Müritz National Park (north-eastern Germany), throughfall was continuously measured at 7 sites with different dominant tree species and ages: young and old beech, young oak, and young and old pine. To this end, trough-based throughfall monitoring systems with a total collecting area of 6.6 m² per site were installed within 2000 m²-plots. Furthermore, stemflow was measured at 5-10 trees per site with a temporal resolution of 1 min, providing a unique high-temporal resolution data set of stemflow response. The measured time series extend over 4 years, offering a good data base for detailed event analyses. This study gives a first overview of the different response patterns observed in the different forest stands, both for throughfall and stemflow

Affiliation: Aquatic Ecosystems Analysis and Management, UFZ Leipzig, Technical University of Dresden, Personal website

We have to consider that climate change may lead to an increase in global average temperature of at least 2°C in the near future. There is substantial evidence that this climate change will be associated with an increase in the frequency, intensity or a shift in timing of extreme events such as rainfall extremes and dryness, elevated flooding and extended low water periods, all with as yet unforeseeable environmental and socioeconomic consequences and feedbacks. Such meteorological and hydrological events are restricted in time and spatially distinct, but their ultimate impact may be significant for much larger regions (e. g. downstream catchment areas flooded from run-off generation in upstream headwaters) and with delayed effects (e.g. algal blooms in the vegetation periods in coastal zones triggered by legacy nutrient pulses from inland sources). The presentation will exemplarily analyze such event chains with emphasis on hydrological extremes and the processes they trigger: the mobilization of nutrients, carbon and harmful substances, their passage from land based sources into the aquatic environment and how these are transported or retained. We present a new modular and event driven observation concept that captures events from their origin to their fading and that complements existing data and models from long-term monitoring and observatories. This concept systematically combines mobile and high resolution event monitoring with stationary integrative observation and as such it aims at unraveling the (potentially decisive) role of the increasing frequency and intensity of extreme meteorological and hydrological events on the status of our tightly coupled hydrologic, ecological and socio-economic environment.

Affiliations: Helmholtz Centre for Environmental Research - UFZ

Abstract: Although it is well known that the Global Change affects the Earth and environment at many different time and length scales, currently, only very limited knowledge is available on the importance of distinct dynamic events for the long-term development of the environment. We will present the MOSES (Modular Observation Solutions for Earth Systems) observing system, which is being developed to close this gap. From 2017 till 2021, the Helmholtz Association invests approximately €28 million to implement the modular infrastructure, which is designed to capture dynamic events such as heavy rains and floods or heatwaves and droughts and to investigate their long-term impacts. The mobile facility is set up as a ‘system of systems’: All together nine research centres develop sensor and measuring systems to record energy, water, greenhouse gas and nutrient cycles on the land surface, in rivers and lakes, at the coastal zone and in the atmosphere. Because MOSES is developed for event-driven, short-term field campaigns, it complements and extends existing monitoring networks, which are mostly designed for long-term observation. Examples include ICOS (Integrated Carbon Observation System), LTER (Long-Term Ecological Research) and TERENO (Terrestrial Environmental Observatories). First MOSES test campaigns start in summer 2018 at several TERENO sites.

Affiliations: Helmholtz Zentrum für Umweltforschung UFZ

Abstract: In the talk we present results of the TERENO - Platform and present our research hypothesis for the new Helmholtz MOSES- Initative. Over 120.000 of data points from 133 sites were sampled over a period from 2012 to 2017 representing small catchment areas in the Bode catchment. The aim was to do a long-term survey to quantify potential sources of main and trace chemistry compounds and determine nitrate process dynamics with its drivers as well as distribution patterns of nitrate. It becomes apparent that nitrate isotopic variations increase with decreasing discharge. So a huge data basis exists now in the Bode Catchment for springs, surface water quality and quantity, involving stable and radioactive isotopes (H-2, H-3, O-18, N-15). As a result of the study, we were able to differentiate which catchment areas (1) have a high surface water dynamics while extreme weather, (2) are overstrained agriculturally,and (3) have a high variability (N, P, DOC). The results will be the base as part for the new Helmholtz-MOSES initiative started in May 2018. Here we will focus on precipitation / surface water / soil moisture/ groundwater interactions during extreme events (drought and heavy storm events) in two smaller parts of the Bode catchment, the Holtemme and Ilse catchments. In connection with the new fertilization ordinance (2017), land use, slope and extreme weather conditions, the groundwater will be included in the monitoring program. The aim is to describe the chemical loads, residence times and flow dynamics in surface water and groundwater catchments under altered climatic conditions. The study will be supported actively by a network of farmers, agricultural science as well as local citizens.

Affiliations: Karlsruhe Institute of Technology (KIT)

Abstract: Deep convective systems and thunderstorms often lead to heavy precipitation events characterised by a considerable local variability. Especially in urban areas the sewage systems often cannot handle precipitation peaks fast enough and flooding of roads and basements in parts of the cities occurs. Therefore, even short-term forecast of precipitation is highly desirable. In order to investigate the ability of radar systems to capture these spatial precipitation pattern, in summer 2017 an X-band radar was deployed close to the city of Stuttgart to measure convective precipitation events with high spatio-temporal resolution in the city and its surroundings. The goal is to evaluate down to which scale finer resolved measurements lead to additional information on the precipitation field. Stuttgart is suitable for this kind of study because the city is embedded in a valley. Additionally, the C-band radar positioned at KIT in Karlsruhe was able to measure precipitation with lower spatial resolution but for a greater area. Results from the summer period will be presented highlighting the spatially inhomogeneous precipitation pattern in Stuttgart and its environment caused by convective precipitation.

Affiliations: 1 KARLSRUHE INSTITUTE OF TECHNOLOGY

Abstract: Non-closure of the surface energy balance is a frequently observed phenomenon of hydrometeorological field measurements, when using the eddy-covariance method, which can be ascribed to an underestimation of the turbulent fluxes. Several approaches have been proposed in order to adjust the measured fluxes for this apparent systematic error. However, there are uncertainties about partitioning of the energy balance residual between the sensible and latent heat flux and whether such a correction should be applied on 30-minute data or longer time scales. The data for this study originate from two grassland sites in southern Germany, where measurements from weighable lysimeters are available as reference. The adjusted evapotranspiration rates are also compared with joint energy and water balance simulations using a physically-based distributed hydrological model. We evaluate two adjustment methods: the first one preserves the Bowen ratio and the correction factor is determined on a daily basis. The second one attributes a smaller portion of the residual energy to the latent heat flux than to the sensible heat flux for closing the energy balance for every 30-minute flux integration interval. Both methods lead to an improved agreement of the eddy-covariance based fluxes with the independent lysimeter estimates and the physically-based model simulations. The first method results in a better comparability of evapotranspiration rates, and the second method leads to a smaller overall bias. These results are similar between both sites despite considerable differences in terrain complexity and grassland management. Moreover, we found that a daily adjustment factor leads to less scatter than a complete partitioning of the residual for every half-hour time interval. The vertical temperature gradient in the surface layer and friction velocity were identified as important predictors for a potential future parameterization of the energy balance residual.

Affiliations:

Abstract: Forests provide an important ecological service by partially balancing the global carbon budget, sequestering about one quarter of anthropogenic emissions (2.4 GT C per year). However, several forest biomes are subject to increasing stress and tree mortality due to invasive pests, drought and fire, and these appear to be exacerbated by climate change. A question arising for forest managers and policy makers is how to anticipate and deal with the acceleration of forest stress and mortality with on-going warming climate. To provide a baseline to which anticipated changes may be compared. We attempt to answer questions: (1) how can we quantify and measure forest resilience? (2) what kind of climate trend or pattern is the key control to forest tipping point? And what are tipping point behaviours of forests? We have explored these questions by using tree-ring data, remote sensing images, eddy flux tower data, and nonlinear stability theory. Building/bridging linkages amongst these three abundant databases is extremely challenging but would be powerful to understand forest resilience to warming climate and key control to the tipping transition events. Tree-ring is the final product of tree-climate interaction. Tree-ring data serve as virtual “weather stations” and can be used to study forest-climate interactions going back thousands of years with high resolution. The disadvantage of tree-ring data is the lack of spatial coverage. Satellite images have excellent spatial coverage and well resolved seasonal variability in vegetation but remote sensing data records cover just a few decades. The linkage between tree-ring data and remote sensing images can translate the tree-ring view from spot to regional maps and further extend them to thousands of years in the past. Forest-climate interaction is a stochastic process or a noise-driven process. High frequency eddy-covariance flux measurements provide datasets to measure the noise strength, which play a critical role in a transition stage. We have formulated concept frameworks for these ideas. Here we report some initial results and some examples along with the pilot-studies.

Affiliations: 1 Helmholtz Centre For Environmental Research - UFZ; 2 Johann Heinrich von Thünen Institute; 3 Helmholtz Centre for Environmental Research - UFZ

Abstract: Gap probability along with canopy shape and leaf area index, are some of the most important structural parameters used to directly interpret the transfer of radiation and related processes in forest systems. Knowledge of the seasonal course of these parameters is critical for the estimation of a number important ecosystem processes. The gap probability is the structural parameter quantifying the probability of a direct beam of radiation passing through the canopy without being intercepted by the foliage and controls amongst others the energy distribution between soil surface and plant surfaces as well as within the plant. Multiple techniques exist and have been widely used to measure gap probability, but these are usually not suitable for measuring gap probability continuously. In general, they deliver the precise results but are very labor intensive, and multiple observations during the year are often not feasible. A simple approach of a photosynthetically active photon flux density (PPFD) sensor network is a promising tool to monitor continuously the light conditions in between conventional gap probability sampling with high temporal resolution. Moreover, modeling gap probability in high temporal resolution for the whole stand by means of the observed points would enable the determination of growth and development of the forest. The aim of this study is to compare the performance of the established methods for gap probability estimation like digital cover photography against PPFD sensor monitoring system and illustrate the limiting factors of the approximations. In addition, analyses of the time series point out not only seasonal changes of gap probability but also rapid events like leaf fall.

Affiliation: Geosciences and Natural Resource Management, Copenhagen University, Personal website

Greenhouse gas (GHG) fluxes from Arctic have attained increasing interest. This as a consequence of global climatic changes anticipated to increase soil temperature and thawing permafrost in vast regions, with significant implications for land-atmosphere GHG exchange. This lecture outlines characteristics of Arctic GHG fluxes with a main focus on the powerful non-CO2 gas species methane (CH4) and nitrous oxide (N2O). The dominant processes and their anticipated interactions with climatic conditions are outlined, and perspectives for future GHG fluxes in a changing Arctic discussed. Northern latitudes contribute in the range 4-5% to global CH4, whereas the contribution to N2O remains unresolved. Meanwhile, research has demonstrated that global warming and permafrost thaw may significantly increase emission rates of these two powerful GHGs. The presentation will also include results from recent and ongoing research in Disko, Western Greenland. This work includes several seasons of in situ measurement of GHG fluxes in large-scale climate manipulation experiments with snow fences, to manipulate winter-time snow cover, combined with open-top chambers to increase temperature.

Affiliations: 1 Botswana Institute for Technology Research and Innovation; 2 University of California Los Angeles

Abstract: Climate models indicate that climate change is likely to affect carbon (C) cycling in drylands, particularly savannas, but the magnitude and direction of change are not fully understood. In this study, we used the Century model to analyze how net primary productivity (NPP), soil respiration and soil C sequestration would respond to an increase in atmospheric CO2 and soil temperature in a savanna ecosystem. We also assessed the coupled effects of precipitation and temperature change on C dynamics under future climatic conditions, as well as the decoupled effects of each of the climate variables under three IPCC climate scenarios; historical, Representative Concentration Pathway 2.6 (RCP2.6) and RCP8.5. An increase in soil temperature results in loss of soil organic C (SOC), whereas doubling atmospheric CO2 concentration causes an increase in SOC. The increase in air temperature causes soil respiration to increase, while it causes NPP to decrease. We calculated the total SOC in the Kalahari savannas to be 0.9 Pg C (1 Pg = 1015 g) in the top meter, and the rate of SOC loss due to anthropogenic climate change to be ~1.1 Tg C yr-1 (1 Tg = 1012 g) and ~2.0 Tg C yr-1 under RCP2.6 and RCP8.5, respectively until the end of this century. If extrapolated to the global extent of savannas, our results imply net SOC loss of at least ~28.4 Tg C y-1 and 64.1 Tg C yr-1 under RCP2.6 and RCP8.5, respectively. The rapid loss of C from dryland soils predicted by Century could accelerate global warming and strengthen positive feedback mechanisms between climate change and processes controlling SOC. Our results vividly support the positive feedback between the SOC and atmospheric C cycles and further indicate that these feedbacks are not adequately accounted for in existing Earth System Models (ESM) that are part of CMIP5. Revisions to these ESM would appear necessary to adequately account for this positive feedback.

Affiliations: 1 University of Innsbruck, Institute of Ecology; 2 HBLFA Raumberg-Gumpenstein

Abstract: As climate change proceeds, extreme climatic events (ECEs) such as drought are expected to increase in intensity and in frequency, with consequences for the carbon cycle. Soil respiration (Rs) is the biggest flux of CO2 from terrestrial ecosystems to the atmosphere. While effects of drought on Rs have been repeatedly studied, less is known how a future warmer climate under elevated CO2 will modify drought responses. While climate warming is expected to enhance drought induced reduction in soil respiration, elevated CO2 has been suggested to enhance soil respiration and slow down the drying of soil. As contribution to the ClimGrass-project we assessed Rs-dynamics during and after a drought event under ambient conditions (C0T0) and under a +3°C warming scenario with a CO2 increase of 300ppm (C2T2). For each of the four treatments three replicate plots were equipped with an automated soil respiration system to assess high resolution Rs fluxes before, during and after drought. Additionally we investigated soil CO2 concentrations with a multiplexed membrane tube system. Our results show that Rs fluxes were generally higher in the C2T2 treatment than in C0T0 and were generally reduced during drought-treatments. All drought and post-drought effects on Rs were more pronounced in a future (C2T2) compared to a current climate. This included Rs reduction during drought and CO2 pulses after rewetting, which led to a transient overshooting the magnitude of corresponding control-fluxes after the drought. An in-depth analyses of soil CO2 gradients and fluxes across the soil profile showed unexpected modifications on soil-CO2 production and transport processes, concerning the lag and severity of drought induced effects.

Affiliations: 1 Hebrew University of Jerusalem

Abstract: Decomposition of organic matter is a key process in the cycling of carbon and nutrients in ecosystems, and in the exchange of carbon and nutrients between the biosphere and the atmosphere. Decomposition is generally considered to be the direct consequence of microbial activity, as regulated by temperature, moisture and the chemical composition of the organic material. However, as the climate gets warmer and drier, with a higher frequency of extreme events, such as prolonged droughts, other mechanisms, and their driving factors and feedbacks, become dominant. Such mechanisms have recently been described for carbon and nutrient cycling in drylands, but with the climate changing, they are expected to contribute to biogeochemical processes in more humid regions as well. Photochemical degradation (photodegradation) is an important abiotic mechanism of decomposition of dead plant material (litter) and surface soil organic matter. High flux densities of ultraviolet and shortwave visible (blue-green) radiation induce photochemical processes, which degrade organic materials exposed to solar radiation. Thermal degradation operates at temperatures above 30°C, which are common conditions at and near the soil surface and cause emission of carbon and nitrogen compounds. In addition to abiotic mechanisms, biotic degradation can operate in the absence of precipitation. Such humidity-enhanced microbial degradation is enabled by moisture from non-rainfall water sources, such as water vapor, dew and fog that is absorbed by plant litter, predominantly at night. Those and other emerging mechanisms of decomposition do not only affect organic materials directly, but can modify organic matter and nutrient dynamics beyond periods of their operation. For instance, photodegradation during daytime facilitated humidity-enhanced microbial degradation at night, and vice versa. These emerging mechanisms should be considered in more humid environments to better understand and predict carbon and nutrient cycling between soil, litter and atmosphere in a warmer and drier climate.

Affiliations: 1 Karlsruhe Institute of Technology; 2 University of Hamburg; 3 Helmholtz Centre for Environmental Research - UFZ

Abstract: An improved regional assessment of the productivity of grasslands depends on in-depth knowledge of the interactions between climatic drivers, nutrient cycles, vegetation properties and human activity. Managed grasslands in Europe display considerable potential for dynamic growth and phenological change, which contribute to the challenge in making representative observations in the field, as well as with remote sensing and models. We have investigated the relationships between vegetation state changes and productivity of meadow grasslands by comparing three sites in Southern Germany (DE-Fen, DE-RbW, DE-Gwg) that are characterised by different management intensities and elevations. Weekly observations of vegetation height, leaf area, above-ground biomass and plant species composition were compared to estimates of the surface exchange of carbon dioxide. This revealed that the productivity of these grasslands correlated positively with management intensity (negatively with elevation) at the seasonal scale. However, at the scale of harvest periods the relationship between (gross) productivity and vegetation dynamics appeared to follow unified patterns for all sites. The variability in above-ground vegetation properties was most pronounced during the spring period and contributed to significant differences in carbon and nitrogen biomass yield between the sites. Furthermore, the study highlighted a substantial potential for bias based on the techniques used to quantify vegetation properties and we discuss a mitigating approach. The outcomes may serve as a reference for model studies on the seasonal allocation of carbon (and nitrogen) in managed grassland systems. The study built on TERENO/ScaleX multi-disciplinary cooperation.

Affiliations: 1 University of Hohenheim; 2 National Oceanic and Atmospheric Administration (NOAA)

Abstract: The Land-Atmosphere Feedback Experiment (LAFE) deployed several state-of-the-art scanning lidar and remote sensing systems to the ARM SGP site during August 2017. A novel synergy of remote sensing systems was applied for simultaneous measurements of land-surface fluxes and horizontal and vertical transport processes in the atmospheric boundary layer (ABL). The impact of spatial inhomogeneities of the soil-vegetation continuum on LA feedback was studied using the scanning capability of the instrumentation as well as soil, vegetation, and surface flux measurements. Thus, both the variability of surface fluxes and CBL dynamics and thermodynamics over the SGP site was studied for the first time. This is essential for advanced observation and understanding of L-A feedback. The objectives of LAFE are as follows: I. Determine turbulence profiles and investigate new relationships among gradients, variances, and fluxes II. Map surface momentum, sensible heat, and latent heat fluxes using a synergy of scanning wind, humidity, and temperature lidar systems III. Characterize land-atmosphere feedback and the moisture budget at the SGP site via the new LAFE sensor synergy IV: Verify large-eddy simulation model runs and improve turbulence representations in mesoscale models. In this presentation, it will be demonstrated how these objectives can be achieved using the LAFE sensor synergy. First results are presented, including a) first 2D measurements of the transition between the convective ABL and the stable nighttime land-atmosphere system, b) unique measurements of L-A feedback during the solar eclipse in the US on August 21, 2017, and c) advanced studies of surface layer stability and turbulence parameterizations taking advantage of the simultaneous measurements of gradients and turbulent properties of atmospheric variables in heterogeneous terrain. It will be demonstrated that these measurements will make excellent contributions to the parameterization of the next generation of high-resolution weather forecast, climate, and earth system models.

Affiliations: 1 Helmholtz Zentrum München

Abstract: N2O is the most important ozone depleting molecule and the third most important greenhouse gas. The mole fraction of N2O rises with 0.25% per year. From all sectors, agricultural soils and food production contribute with approximately 60% most. At the TERENO reserach farm Scheyern, N2O-flues were measured for at least 25 years. Most of the flux measurements were conducted with the automated chamber method, however, recently, also eddy covariance measurements using quantum cascade lasers were used. Here we present modelling approaches and measurements over the last 7 years with the automated chamber method and measurements and simulations of 1 year eddy covariance measurements. The focus of the work is on comparing simulations of both measurement methods and the measurement methods themselves. The eddy covariance measurements were conducted in the year 2015 which was the warmest year in the history of the farm (Tmean = 10.1°C) and also a very dry year (617mm). An annual sum (0.2 kg-N ha-1 a-1) was obtained by using a fitted neuronal net as gap filling procedure. The described sum is very low but fits to the obtained chamber measurements as long as measurements are available. The reason for the low N2O-fluxes is mostly seen in the warm winter, where freeze-thaw cycles did not occur and the very dry and warm summer with unideal conditions for denitrification. The findings with the QCL-EC-method are consistent with the behaviour of fluxes according to the steering parameters temperature, soil moisture and fertilisation. Moreover an influence of atmospheric turbulence on nitrous oxide fluxes can be seen which is not possible with the chamber method as the chamber excludes the atmosphere during the measurement process. In our case this has the consequence, that a diurnal cycle is visible especially during the fertilisation period with the eddy covariance method but not with the automated chamber method. Simulation results of the ecosystem model Expert-N showed, that the automated chamber measurements can be simulated better than QCL-EC measurements. The reason is – amongst other things – seen in the missing atmospheric impact in the N2O-flux model. The findings according to the diurnal cycle and their consequence for the annual balance agree with recent publications using the eddy covarince measuring. The low fluxes in the extreme summer 2015 are amonts the smallest fluxes that have been ever measured in Scheyern.

Affiliations: 1 Laboratoire d’Hydrologie et de Géochimie de Strasbourg (LHyGeS), Université de Strasbourg; 2 INRA de Nancy

Abstract: Forests are subject to several natural and/or anthropogenic perturbations, over short and long terms time. A better understanding of how soil/water/plant interactions can react and response to modifications become more and more crucial for a better management of natural and renewable resources. Since 1970, major symptoms of forest decline had been observed worldwide. In granitic environment, as in the Vosges Mountains (NE France) forest decline can be related to soil nutrient leaching by acid rain. Indeed acidity from atmospheric pollution can totally modify the availability of nutrients in soil, and the equilibrium between different nutrient pools. Traditionally, the quantity of exchangeable cations, and among them nutrient as Ca and Mg, are quantified with extracting agent such as cobaltihexamine. But these measurements are not enough to fully characterize and quantify the different processes involving the mobilization of nutrients for plants. Further investigations are needed to identify the different pool of nutrient, their reactivity, their stock and their availability over time. With this study we develop an experimental approach in order to reproduce and analyze in laboratory natural phenomena at smaller scale. The studied site is the Strengbach catchment, a forest ecosystem of 80 ha located in the Vosges Mountain where severe forest decline symptoms have been observed since 1980, especially on spruces. The main parameters of this site (climate, geochemistry, hydrology…) are monitored since 1986 by the Observatoire Hydro-Géochimique de l’Environnment (OHGE: http://ohge.unistra.fr). Experiments have been realized with natural soils from two different plots located in beech and spruce stands. Soil samples from 6 horizons (until 80 cm depth) are disposed on batch series (triplicate) with two different acids (hydrochloric acid and oxalic acid). The solutions are analyzed at different time from 24h to 150 days. DRX, SEM, chemical and isotopic (Sr) characterization are determined on soil and solution. Strontium isotope data aimed to decipher mineral weathering, atmospheric deposit and biological activity. The results show highly variable chemical behavior versus element, type of soil, horizons and type of acid. In addition we observed unexpected results for calcium. Indeed the total quantity of Ca extracted after 150 days correspond to a very low proportion of the calcium exchangeable obtained with cobaltihexamine, underlining a complex bioavailability of this nutrient in soil.

Affiliations: Luxembourg Institute of Science and Technology (LIST)

Abstract: There is an urgent need for the quantification of water availability in forest ecosystems exposed to global change. A major challenge in this respect lies in the reduction of uncertainties in water mass balance calculations at ecosystem or catchment scales via a better constraining of transpiration fluxes. While plant transpiration plays a major role in the water cycle, the mechanisms that regulate its contribution within the critical zone remain poorly understood. More specifically, pressing open questions relate to where water available for plants is stored and where transpiration fluxes actually occur at the regolith-plant interfaces. An important way forward in this respect lies in the efficient sampling of water that is taken up by the trees during the growth period. Recent studies relied on a multi-tracer approach – combining geochemical and isotopic compositions of tree compartments – for fingerprinting water sources and assess its availability in space and time. Here, we propose to focus on the temporal variability in the geochemical and isotopic composition of tree sap during the leaf-out period. We present proof-of-concept work on the potential for a multi-tracer monitoring protocol carried out on biological samples for sampling tree water sources. We collected sap from 120 beech trees, 15 soil water solutions and 15 groundwater samples over two consecutive years in a deciduous forest of the Weierbach catchment in Luxembourg. We carried out a complete geochemical (pH, alkalinity, DOC, total N, Cl-, NO3-, SO42-, NH4+, Na+, K+, Ca2+, Mg2+, 30 trace elements) and isotopic (O-H stable isotopes) characterization on each collected sap, soil and ground water sample. Our preliminary results highlight the spatial variability of the evolution of sap geochemical and isotopic composition during the four sampling campaigns of the first year and the geochemical and isotopic similarities between soil solutions, groundwater and sap during eight sampling campaigns during the second year. We also showed that several trace elements could be considered as complementary tracers of water uptake dynamics during transpiration processes that occur in spring, at the beginning of the vegetative period.

Affiliations: 1 Forschungszentrum Jülich/IBG-3; 2 Jülich Research Center GmbH/IBG-3; 3 Forschungszentrum Juelich/IBG-3

Abstract: Soil greenhouse gas (GHG) emissions from agriculture contribute to global warming. To support mitigation measures against global warming, it is important to understand the controlling processes of GHG emissions. Previous studies focusing mainly on paddy rice fields or wetlands showed a close relationship between soil redox potential and GHG emission (e.g., N2O). However, the interpretation of redox potentials as one of the controlling factors of GHG emission is limited due to the low number of continuous redox measurements in most ecosystems. Recent sensor developments open the possibility for the long-term monitoring of field-scale soil redox potential changes. We performed laboratory lysimeter experiments to investigate how changes in the redox potential, induced by changes in the water level, affect GHG emissions from agricultural soil. Under our experimental conditions, we found that N2O emissions followed closely the changes in redox potential. The dynamics of redox potential were induced by changing the water-table depth in a laboratory lysimeter. Before fertilization during saturated conditions, we found a clear negative correlation between redox potentials and N2O emission rates. After switching from saturated to unsaturated conditions, N2O emission quickly decreased, indicating denitrification as the main source of N2O. In contrast, the emissions of CO2 increased with increasing soil redox potentials. After fertilization, N2O emission peaked at high redox potential, suggesting nitrification as the main production pathway, which was confirmed by isotope analysis of the 15N site preference in the emitted N2O. Recently, we started a more advanced experiment in which the lysimeter was irrigated until full saturation. In addition, N fertilizer has been applied to the soil surface. The data of this fertilization experiment will enable a more detailed analysis of the relationship between soil redox potential and N2O emissions under more natural conditions. Preliminary results (Fig. 1) are promising and show that the redox potential decreased earlier in the lower part of the soil column, while the water potential decreased earlier in the upper part of the soil. This analysis will help to better understand the controlling factors of GHG emissions and to better differentiate between different source processes in order to improve process-based GHG models.

Affiliations: Helmholz-Zentrum München

Abstract: Intensive agricultural land use is considered to be the major source of anthropogenic contribution to the increase in atmospheric N2O concentration during the last decades. A necessary reduction of N2O emissions from agricultural soils requires changes of agricultural management practices. Mathematical models can help to identify adapted land use and improved production systems by providing knowledge about the interaction between the determining processes of N2O production and the dynamics of state variables affecting N2O emissions. In particular the impact of climate change on N2O emissions can be better analyzed. The aim of this study was to test modeling approaches for their ability to describe and quantify the long-term development of N2O emissions from agricultural fields observed at the TERENO Research farm Scheyern situated 40 km north of Munich, Bavaria. Data for model evaluation were obtained during 25 years (1992-2017) mainly by the closed chamber method. We applied two different modeling approaches, where one model assumes a fixed N2O/N2 ratio for N2O production and neglects the transport of N2O in the soil profile; whereas the other model explicitly considers N2O transport and assumes a dynamic reduction of N2O to N2. Generally, the modeling approaches were able to describe the observed long-term and seasonal dynamics of N2O emissions and events of higher N2O emissions due to increased denitrification activity after heavy precipitation and during thawing after soil freezing. It is concluded that the decrease of frost thaw-events due to higher temperatures during the cold season is the main reason for the decrease of N2O from the agricultural fields at the research farm Scheyern.

Affiliations: 1 CNRS; 2 University Rennes 1

Abstract: The impact of microorganisms on the groundwater biogeochemical cycles of carbon, nitrogen and sulfur is poorly constrained. Recent studies evidenced microbial activity up to hundreds of meters, driven by mixing of waters that differ in redox states and hence in the availability and abundance of possible electron acceptors and donors. However, in order to predict where hotspots of microbial activity occur, it is crucial to understand and predict the spatial distribution of the most important electron acceptor, oxygen. This study aims at investigating the mechanisms and processes that govern the distribution of oxygen in groundwater, in particular the factors that control its vertical distribution. The isolation of bore sections, with an inflatable packer, allowed obtaining vertically discrete profiles of oxygen and microbial communities at different times in a fractured aquifer bore. The seasonal occurrence of biofilms was related to an intermittent availability of oxygen, which could be a result of changing hydrological conditions. A literature review of oxygen concentrations measured in groundwater is analysed in respect of the geological context of aquifers as well as residence time of waters, in order to link groundwater dynamics with reactive zone locations and potential enhanced biogeochemical activity.

Affiliation: Institute of Hydrology, Freiburg University, Personal website

Phosphorus (P) is an essential element for primary productivity of an ecosystem. Natural forests are known to be limited in P supply and therefore develop tight P-recycling strategies. The P availability is, however, significantly affected by P-losses due to hydrological fluxes in the surface and subsurface during rainfall or snowmelt events. Since typical observations of runoff and nutrient losses sample baseflow conditions and average rainfall-runoff events, possible tipping points for extreme events are generally not known. However, climate change may increase the probability of such extreme events and the loss of essential nutrients like phosphorous may disproportionally increase.

Affiliations: Technical University of Crete

Abstract: The Koiliaris River watershed is a Critical Zone Observatory (CZO) located in Crete, Greece and represents severely degraded Mediterranean soils due to heavy agricultural impact such as grazing, over many centuries. The area is under imminent threat of desertification (soil carbon loss) due to climate change that is predicted for the region over the next century. The main type of soil degradation in the basin is water erosion, which is due to the clearing of forests and natural vegetation for cropping and livestock grazing. De-vegetation and inappropriate cultivation practices induces soil organic matter losses making soils susceptible to erosion and desertification with global consequences for food security, climate change, biodiversity, water quality, and the agricultural economy. The total area is 130 km2 and the total length of the river is 36 km. Intensive hydrologic and geochemical monitoring of surface and groundwater has been conducted since 2004 while the site has historical data since the ‘60s. There are 3 meteorological stations within the watershed and several outside. Experimental plots have been established to assess the impact of land management on soil functions and ecosystem services. Key research areas investigated are Sustainable Water and Soil Resources Management, Hydrologic Modelling of Complex Terrains (Karstic Systems), High Frequency Environmental Monitoring, Soil Degradation and Soil Formation, Stream and Ground Water Chemistry. The objective of the present study is the geochemical characterization of karstic, groundwater and surface runoff at Koiliaris CZO and the identification of the factors influencing its quality. Sampling was designed as to identify the chemical characteristics of karstic water and the impacts of highland degraded soils by livestock on water quality. The analysis was comprised of measurements of several chemical species obtained from wells and springs at Koiliaris CZO. The chemical species analyzed included anions (NO3 -, SO4 2-, Cl- , etc.) and cations (K+, Na+, Ca2+, Mg2+, Si4+, Sr2+, Ti2+ etc.). Principal component analysis (PCA) was used for the categorization of water bodies according to their chemistry. The temporal variability in karstic, groundwater and surface water chemistry showed that it was related to both micro-climatic, tectonic factors and anthropogenic impact. The rates of chemical weathering were estimated for both the karst and schist geologic areas of the watershed and were compared to other CZOs. High-frequency hydrologic and water quality monitoring stations have been used for the characterization of the hydrologic and biogeochemical processes with varying process response-times.

Affiliations: 1 University of Bonn; 2 Research Centre Jülich; 3 DeLaWi-TreeRing Analyses

Abstract: Tree water use is a key variable in forest eco-hydrological studies and is often monitored by sap flow measurements. Upscaling these point measurements to the stand or catchment level, however, is still challenging. Due to the spatio-temporal heterogeneity of stand structure and soil water supply, extensive measuring campaigns are needed to determine stand water use from sap flow measurements alone. Alternatively, water balance models can be used to estimate stand transpiration. Such models basically calculate transpiration as a function of atmospheric boundary conditions, soil water supply and vegetation characteristics from which leaf area index and stomatal conductance are the most important. Under non-limiting conditions, transpiration generally follows the dynamic of atmospheric demand. However, the response of plants to variabilities in soil water supply is highly species specific. To account for these species-specific feedbacks in the soil-vegetation-atmosphere system, water balance models commonly include plant water stress functions, where critical limits of soil water supply can be adapted according to the vegetation characteristics. For many agricultural crops, such thresholds have already been defined and included into models that help to optimize irrigation scheduling and crop yields. However, little is known about the species-specific water-stress-response of forest trees. Since forests play an important role in regional and trans-regional water cycles, there is evidence that an improved knowledge of tree species specific water stress thresholds would greatly enhance the simulations of water fluxes from the forest stand to the watershed and landscape scale. The aim of this study was: (1) to parameterize the Feddes water stress model for Norway spruce, which is a wide-spread and economically important species in Europe; (2) to investigate if the species-specific calibration of the Feddes parameters can improve water balance simulations in the soil hydrological model HYDRUS-1D. After successful calibration and validation of HYDRUS-1D, we combined root-zone water potential simulations with a new plant water stress factor derived from sap flow measurements at two plots of contrasting soil moisture regimes. By calibrating HYDRUS-1D against our sap flow data, we determined the critical limits of soil water supply. Drought stress reduced the transpiration activity of mature Norway spruce at root-zone pressure heads <−4100 cm, while aeration stress was not observed. Using the recalibrated Feddes parameters in HYDRUS-1D also improved our water balance simulations. We conclude that the consideration of sap flow information in soil hydrological modeling is a promising way towards more realistic water balance simulations in forest ecosystems.

Affiliations: 1 Univ Rennes, CNRS, Géosciences Rennes – UMR 6118; 2 INRA, Agrocampus Ouest, UMR 1069 SAS

Abstract: Soil organic matter (SOM) destabilization produces gaseous and dissolved species and plays a key-role toward environmental and climatic issues. This process can be viewed as a combination of abiotic (solubilization) and biotic (biodegradation) processes. Both of them are controlled by the water occurring in soils since this molecule is at the same time the solvent and the living media. The soil water content controls the emission of soil carbon dioxide and the water-table dynamic controls the temporal variations of dissolved organic carbon in soil solutions. The aim of this study is to explore how the fluctuations of water-table levels can impact the gaseous and dissolved components of SOM destabilization. This was performed on the Kervidy-Naizin catchment (AgrHys critical zone observatory) localized in Brittany and that is part of the OZCAR Research Infrastructure. Measurments of CO2 fluxes and soil solution species (dissolved organic and inorganic carbon DOC/DIC and ferrous cation FeII as a) sampled in the macroporosity were performed fortnightly during the hydrologic year 2014-2015 from October to May at two sampling sites. They were chosen along a topographic transect equipped to monitor water-table levels with one downslope (Down) and one midslope (Mid). The difference in water-table fluctuations were measured by the mean residence time and mean return time. Down the residence and return times were 47 and 4 days, respectively, while Mid they were 2 and 12 days, respectively. Along the sampling period, the flux of CO2 was higher at Mid (1.5 ± 0.5 µmol m-2 s-1; mean ± standard deviation) than at Down (1.0 ± 1.0 µmol m-2 s-1). At both sites CO2 decreased from October to February and then increased up to May, however the kinetics were different. DIC and DOC were differently correlated in the two sites. At Mid, they were inversely correlated, with a steady DOC decrease from 14 to 8 mg/l from October to March and a steady DIC increase from 9 mg/l to 30 mg/l. At Down two periods were defined by the occurrence of FeII, a marker of reducing conditions. Before the reducing period DIC and CO2 were inversely correlated, while during reducing conditions DIC, DOC and CO2 were positively correlated and increased by 4, 3 and 8 respectively, highlighting reducing period as a hot moments for SOM biodegradation. The results of this study, together with literature survey, are used to propose a conceptual model of SOM destabilization.

Affiliations: 1 ILR, iFZ, Justus Liebig University Giessen, 2 Institute of Meteorology and Climate Research - Atmospheric Environmental Research (IMK-IFU), Garmisch-Partenkirchen, 3 Centre for International Development and Environmental Research (ZEU), Justus Liebig University Giessen, Giessen

Abstract: This study presents the results of a combined measurement and modelling strategy to analyse N2O and CO2 emissions from adjacent arable land, forest and grassland sites in Hesse, Germany. The measured emissions reveal seasonal patterns and management effects, including fertilizer application, tillage, harvest and grazing. The measured annual N2O fluxes are 4.5, 0.4 and 0.1 kg N ha-1 a-1, and the CO2 fluxes are 20.0, 12.2 and 3.0 t C ha-1 a-1 for the arable land, grassland and forest sites, respectively. An innovative model-data fusion concept based on a multi-criteria evaluation (soil moisture at different depths, yield, CO2 and N2O emissions) is used to rigorously test the LandscapeDNDC biogeochemical model. The model is run in a Latin Hypercube based uncertainty analysis framework to constrain model parameter uncertainty and derive behavioural model runs. The results indicate that the model is generally capable of predicting trace gas emissions, as evaluated with RMSE as the objective function. The model shows a reasonable performance in simulating the ecosystem C and N balances. The model-data fusion concept helps to detect remaining model errors, such as missing (e.g., freeze-thaw cycling) or incomplete model processes (e.g., respiration rates after harvest). This concept further elucidates the identification of missing model input sources (e.g., the uptake of N through shallow groundwater on grassland during the vegetation period) and uncertainty in the measured validation data (e.g., forest N2O emissions in winter months). Guidance is provided to improve the model structure and field measurements to further advance landscape-scale model predictions.

Affiliations: Helmholtz Centre for Environmental Research - UFZ

Abstract: Environmental research strives to understand how we can provide clean drinking water, mitigate climate change, and optimise the agricultural management to increase the soil’s long-term potential for crop production. Soils are complex systems that control the environment with their filter, reactor and storage functionality. The magnitude of this control and the resulting environmental impact largely depend on the site-specific soil characteristics. The spatial soil variation within a landscape is caused by the site-specific interaction of the soil forming factors which drive particular soil processes and result in a double differentiation: the soils’ vertical sequence of horizons and their variation in geographical space. Today, the vast amount of data from remote sensing but also from expert-based spatial geo-information provides a good basis to approximate the soil forming factors. Relating this information with that of a large soil database, opens up the striking prospect that we may succeed in understanding and modelling spatial soil variation at national scale (Germany). The state of the art methodology to do so belongs to the field of data science. It explores the development and application of mathematical algorithms that can derive knowledge from big data in order to understand complex systems. To succeed with this ambition, high importance has to be given to the information content of the data. Soil formation takes thousands of years. In contrast, environmental data collection is rather young. Hence, the derivation of predictors that approximate soil formation is not trivial. A similar issue concerns the soil data, which are unequally distributed in space, and are heterogeneous in quality and their inherent uncertainties. Moreover, the numerous soil properties, processes and driving factors are interrelated. This complex system involves multiple non-linear dependencies which need to be addressed by advanced modelling algorithms. Yet, there are more aspects that influence the model building procedure and the corresponding uncertainty assessment. These relate to the algorithms themselves, model tuning, predictor selection, uncertainty propagation, spatial autocorrelation, and the inclusion of expert knowledge. Overall, the powerful machine learning and optimisation algorithms can only derive knowledge immanent to the data. Furthermore, the high amount of data which makes it feasible to address this research question is also one of its biggest challenges; it requires advanced computation and programming solutions. In this spirit, the need for a soil-landscape model providing soil process units at national scale in order to understand and control environmental impact will promote this research field.

Affiliations: Research Institute for Nature and Forest (INBO)

Abstract: On 11 ICP Forests Level II plots in Flanders, Northern Belgium, the forest floor and the mineral soil was sampled and analysed in 1989, 1991/1992, 2004 and in 2014. The data were compiled in one database and subjected to a comparisons over time. The objective of this study was to evaluate whether we are able to detect significant changes in forest soil properties applying the sampling and laboratory analytical protocols of the ICP Forests manual. The following chemical and physical forest soil properties were studied: • Soil reaction (pH-H2O and pH-CaCl2) • Exchangeable cations including cation exchange capacity and base saturation • Total contents of carbon, nitrogen, macro-nutrients, heavy metals and C:N ratios • Forest floor thickness and mass First results showed statistically significant differences in a number of chemical soil properties between the sampling in 2004 and 2014, where exactly the same field sampling protocols were used. Though reanalysis of the samples taken in 2004 together with the samples taken in 2014 did not longer reveal statistical significant differences (except for changs in soil pH). Several explanations can be postulated. Either no real changes occurred in most of the soil properties or the changes were less than the minimum detectable difference determined by spatial and interlaboratory analytical variation. In case of the latter, sampling and analytical protocols should be revised.

Affiliations: 1 Helmholtz Centre for Environmental Research

Abstract: Lysimeters are considered to be a suitable tool for studying the water and material balance of soils. Conventional gravitational lysimeters allow an accurate quantification of seepage-related material fluxes, whose knowledge is important for compliance with environmental quality objectives. As part of the TERENO SoilCan program, a lysimeter station was built at the Sauerbach site (Bode catchment, Germany) in 2010, consisting of six monolithically filled lysimeter vessels. These lysimeters have a controlled lower boundary condition in which the matrix potential is controlled by using bi-directional pumps according to the comparison of matrix potential (actual value) measured in the lysimeter and the matrix potentials (target value) determined outside by 3 field tensiometers. In the case of drainage, water is pumped out of the lysimeter via the suction rake installed at the bottom of the lysimeter vessel into the leachate tank. In this case, a defined sample volume is taken from the leachate stream via a bypass pump and collected in a separate collecting flask. In the watering case, water is taken from the leachate tank and pumped back into the lysimeter. In critical phases (high evapotranspiration rates, low rainfall), the water volume available in the leachate tank can be completely used up, so that additional water (accumulated lysimeter leachates or drinking water) is required here. We show to what extent the material concentrations in the bypass collection flasks are influenced by the control of the lower boundary conditions based on the comparison of measured material concentrations of water samples from the leachate tank and the bypass collection flasks. Conclusions for the operation of the lysimeter station are derived.

Affiliations: 1 University of Hohenheim; 2 Universisty of Hohenheim

Abstract: The Richards equation is a standard process model for numerical simulations of transient water fluxes in variably saturated soils. For example, the Richards equation is part of many agroecosystem models that simulate vertical water and solute fluxes in the unsaturated zone as well as soil organic matter turnover and crop growth. For reliable predictions of these coupled models, effective models describing the soil hydraulic properties are mandatory. This is particularly important when crop growth is reduced due to limited water availability. To reach at soil hydraulic functions that represent the full moisture range, i.e. from saturated to completely dry soils, recently a model for the water retention curve was developed as the sum of any given parametric capillary saturation function and a new model for the non-capillary saturation function. With it, a continuously differentiable, flexible, and physically coherent representation of the water retention curve is at hand. The expressions for the capillary and non-capillary saturation function are then used to calculate the respective hydraulic conductivity curve. This is achieved by adopting Mualem’s integral for the capillary part of the hydraulic conductivity curve only and calculating the non-capillary hydraulic conductivity curve directly from the new non-capillary saturation function. The new hydraulic conductivity curve leads to consistent description of measured data, including the often observed change of slope beyond pF = 2. In a previous study, the new soil hydraulic properties model framework was shown to adequately and coherently describe measured water retention and hydraulic conductivity curve data of soil samples with a wide range of textures and origins at a wide range of moisture states. In this study, we quantify the effect of the non-capillary part of the hydraulic conductivity curve on simulated crop growth, root water uptake, seepage, and nitrate leaching under field conditions with a focus on years and locations with rather low precipitation rates. For this, the new model of the water retention curve was implemented into the agroecosystem modelling framework Expert-N. The new model is compared to simulations based on the classical van Genuchten-Mualem approach.

Affiliations: 1 Max-Planck-Institute for Biogeochemistry Jena; 2 Universität für Bodenkultur Wien; 3 Luxembourg Institute of Science and Technology (LIST); 4 University of Hohenheim; 5 Max-Planck-Institute for Biogeochemistry; 6 BOKU, University of Natural Resources and Life Sciences Vienna

Abstract: While many different modeling approaches of evapotranspiration (ET) perform rather well when evaluated at daily or monthly time scales, they can show systematic deviations at the sub-daily time scale, which results in potential biases in ET to global climate change. Here we use a metric which measures the linear and non-linear relationship in terms of a phase lag between a response variable (heat and mass fluxes, meteorological states) to incoming solar radiation. We illustrate this approach with data from an Eddy-Covariance site at a temperate grassland, which experienced a pronounced summer drought. We employ three structurally different modeling approaches which are used in remote sensing retrievals and quantify how well these models represent the observed diurnal cycle under clear sky conditions. We find that energy balance residual approaches, which use the surface-air temperature gradient as input are able to reproduce the reduction of the phase lag from wet to dry conditions. However, approaches which use the vapor pressure deficit (Da) as driving gradient (Penman-Monteith) show significant deviations from the observed phase lags, which is found to depend on the parameterization of surface conductance to water vapor. This is due to the typically strong phase lag of 2-3h of Da, while the observed phase lag of ET is only in the order of 15 min. In contrast, the temperature gradient shows phase differences in agreement with the sensible heat flux and represents the wet-dry difference rather well. We conclude that these differences in phase lags can be understood by the different mechanisms of diurnal heat exchange and thus allow a process-based insight to improve the representation of land-atmosphere interactions in models.

Affiliations: 1 Institute of Environmental Physics (IUP), Heidelberg University, Heidelberg, Germany; 2 HGS MathComp, Heidelberg University, Heidelberg, Germany; 3 Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Heidelberg, Germany

Abstract: The mathematical representation of soil water movement exhibits uncertainties in all model components, even when using process based descriptions like the Richards equation. Knowledge fusion aims at the consistent aggregation of all information pertinent to some reality. This means that the representation of uncertainties in each model component becomes an integral part of the model formulation. The goal then is an optimal consistent representation with minimal uncertainties. Data assimilation methods are a key tool for this task. However, they are still limited with respect to the reduction of uncertainties in the dynamics itself. We demonstrate how data assimilation methods can be employed for knowledge fusion on a small hydraulic example. We assessed the key uncertainties for the specific hydraulic situation of a 1-D soil profile with TDR (time domain reflectometry)-measured water contents. These key uncertainties are initial condition, soil hydraulic parameters, small-scale heterogeneity, upper boundary condition, and the local equilibrium assumption by the Richards equation. We employ an ensemble Kalman filter (EnKF) with an augmented state to represent and estimate all key uncertainties, except for the intermittent violation of the local equilibrium assumption. To bridge this time, we employ a closed-eye period, which pauses the parameter estimation and only guides the states through this time. During this time, unrepresented model errors are compensated through an inflation method. The consistent representation of the uncertainties ensured constant parameters throughout the whole estimation, suggesting that these characterize correct material properties. This improves predictions during periods when the local equilibrium assumption is met, but consequently worsens predictions when the assumption is violated. The descriptions of the dynamics during local non-equilibrium phases remains an open challenge. Still, the approach shows a way to limit the incorporation of errors into parameters, which is of great importance in knowledge fusion.

Affiliations: 1 Technische Universität Berlin; 2 Bundesanstalt für Gewässerkunde; 3 BearingPoint GmbH

Abstract: The German Federal Ministry of Transport and Digital Infrastructure (BMVI) expects a significant increase of inland navigation traffic by the year 2030; therefore, a higher efficiency of the entire shipping logistics and management will be required. In this context, the mFund project Digital Skipper Assistant (DSA) has the objective to develop a mobile application, to support inland navigation, e.g. calculating routes and Estimated Time of Arrival (ETA). Being part of this exiting project, this work has the aim to explore the capabilities of artificial neuron networks (ANNs) to predict water levels up to ten-days ahead at some crucial gauges along the German part of the Rhine River Basin. A single- and multiple-outputs model based on long short-term memory (LSTM) networks was implemented, exploring the LSTM promising performances in identifying long-term dependencies. The model developed has been tested and several sensitivity analyses have been conducted (e.g. varying the number of hidden neurons or the batch size), using first as input data only the water level measurements at certain upstream gauges, to predict a downstream one. Afterwards, the forecasts of the hydrological model chain of the Federal Institute of Hydrology (BfG) have been integrated in the better performing LSTM architecture as additional predictors/input data. The most relevant results have been compared regarding three different input scenarios for the water level forecasting at the gauge Oestrich: 1) 6-hours water levels; 2) daily water levels; 3) daily water levels integrated with the BfG forecasts. The results of the latter scenario outperformed the longer-term predictions (leading time of about 2-10 days), delivering forecasts with a higher accuracy than the ones of the BfG hydrological model chain. On the other hand, up to 2-days forecasting, the BfG physically based model series or the single-output LSTM networks provide superior performances. In further research, the developed multiple-outputs network will be consolidated regarding water-level dependent evaluations and further refined in the gauges downstream to Oestrich, such as Cologne and Emmerich. Further ANN architectures such as echo state networks (ESN) or support vector machine (SVM) will be investigated for longer-term water levels predictions (longer than 14 days) and compared systematically with the current approaches developed at BfG.

Affiliations: 1 Esri Germany

Abstract: Deep learning has been used successfully in computer vision problems, e.g. image classification, target detection and so on. We use deep learning in conjunction with ArcGIS to implement a model with advanced convolutional neural networks (CNN) for lithological mapping in the Mount Isa region (Australia). The area is ideal as there is only sparse vegetation and besides freely available Sentinel-2 and ASTER data, several geophysical datasets (radiometric data and magnetometry) are available from exploration campaigns. There are first studies based on Sentinel-2 data for geological applications (Ge, 2018) and ASTER has been used for a long time for lithological mapping. By fusing the data and thus covering a wide spectral range as well as other geophysical properties of rocks, we aim at significantly improving classification accuracies. Based on a proposed 3-D CNN-based feature extraction model (Ghamisi, 2016) , we developed an end-to-end deep learning model with 1-D, 2-D and 3-D convolutional, pooling and deconvolutional layers using Tensorflow. Our model was also inspired by the family of U-Net architectures, where low-level feature maps or encoders are concatenated with high-level ones or decoders, which enables precise localization. This type of network architecture was especially designed to effectively solve pixel-wise classification (segmentation) problems, which is very similar to lithological classification. By spatially resampling and fusing remote sensing data with different bands and geophysical data into a larger imagery with tens of bands, we build up a cube of images which is the input data of our model. The connection between ArcGIS and the deep learning libraries was achieved by using the Python API for ArcGIS and implementing the workflow into Jupyter Notebooks. Our model classifies each pixel of the multiband imagery into different types of rocks according to a defined probability threshold. Preliminary results based on the Sentinel-2 bands alone are very promising with accuracies of around 60%. By including geophysical data and ASTER spectral bands that perfectly capture major absorption features of clay minerals and mafic minerals such as pyroxenes and carbonates we should be able to improve significantly using the same deep learning architecture. Ge, W. a. (2018). Lithological Classification Using Sentinel-2A Data in the Shibanjing Ophiolite Complex in Inner Mongolia, China. Remote Sensing, 10. Ghamisi, Y. C. (2016). Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks. IEEE Transactions on Geoscience and Remote Sensing, 54(10):6232--6251.

Affiliations: 1 CNRS, Sorbone University; 2 Université Grenoble Alpes and CNRS; 3 IRSTEA Lyon

Abstract: One of the main objectives of the French OZCAR (Critical Zone Observatories: Research and Applications) Research Infrastructure (RI) is a better understanding of the terrestrial compartments of the Earth System and an integrated representation of the critical zone coupled water, energy and matter cycles, including biogeochemical cycles; covering various spatial and temporal scales and incorporating the heterogeneity of the critical zone and human influence on its environment. OZCAR CZOs offers the opportunity to document all those coupled processes in selected sites/catchments and to progress in process understanding and modeling. However, as revealed by a survey, disseminated in the OZCAR-RI community, the modeling activity is currently fragmented amongst the disciplines. The challenge is thus to increase the level of integration of Critical Zone models able to address OZCAR scientific challenges and provide feedbacks to the society on the status of its environment. To this aim, strong interactions between data science and modeling approaches are necessary. Such an approach is already widely used by the atmosphere community to provide, through data assimilation, reanalyses of the state of the atmosphere and of the components of the water cycle at the global scale. Such approach could be developed for the Critical Zone, providing valuable and novel information for any research or planning activities; and feedbacks on the necessary variables to be measured in CZOs. The presentation will give an overview of the modeling approaches developed independently at the CZO scale within the OZCAR community. This state of the art highlights a strong complementarity between these models in terms of i) compartments of the critical zones concerned by the model; ii) investigated time and space scales, iii) how the CZOs observation data are used in the modeling chain and iii) modeling objectives addressed so far by the community. Based on this state of art, we will discuss the future challenges that will be addressed in OZCAR, in synergy with international initiatives, in order to propose novel integrated modeling approaches of the dynamics of the critical zone. This implies performing coupling between processes described by models running at their own space and time scales and coming from different disciplines; strengthening the links between databases, GIS layers and models; and developing data assimilation tools able to integrate all the available knowledge.

Affiliations: 1 FZJ

Abstract: In the context of the TERENO research infrastructure, a large amount of experimental data is collected for the Rur catchment in Germany. However, the dataset still shows large spatial gaps as only a limited number of sites is intensively monitored. In order to reconstruct terrestrial system variables of interest on a spatial grid that covers the complete catchment, spatial (and temporal) interpolation of the variables is needed. A (bio)physically based interpolation is possible by terrestrial system models in combination with sequential data assimilation. The question is whether this approach improves the assessment of the net water, energy and carbon balances of the catchment and improves our understanding of the catchment functioning. A series of data assimilation studies with different models was carried out to address this research question. Three studies investigated the impact of assimilation of soil moisture data (from SoilNet) at the hillslope scale into different models like the Community Land Model and VIC (land surface models) and the integrated land surface-subsurface model ParFlow-CLM, and found some improvement in characterization of hydrological variables. Three other examples illustrate the transfer of information from intensively monitored sites, to the rest of the catchment. Two of the these studies were carried out with a land surface model (the Community Land Model), assimilating soil moisture data measured by cosmic ray probes respectively net ecosystem exchange monitored by eddy covariance stations. In both studies a clear improvement could be detected for the characterization of hydrological resp. ecological variables at the scale of the complete catchment. Another study was carried out with the integrated land surface-subsurface model ParFlow-CLM and also showed the positive impact of assimilation of cosmic ray probe data. It can be concluded that a limited number of monitoring sites allows the improved characterization of water, energy and carbon balances at the catchment scale. A current study focuses on the impact of groundwater level assimilation on the characterization of hydrological variables at the catchment scale.

Affiliations: 1 UFZ Leipzig; 2 Helmholtz Centre for Environmental Research GmbH - UFZ

Abstract: The 2003 drought event in Europe had major implications on many societal sectors, including energy production, health, forestry and agriculture. The reduced availability of water accompanied by high temperatures led to substantial economic losses in the order of 1.5 billion Euros in the agricultural sector alone. Furthermore, soil droughts have considerable impacts on ecosystems, forest fires and water management. Monitoring soil water availability in near real-time and at enables water managers to mitigate the impact of these extreme events. The German drought monitor (Zink et al. 2016) was established in 2014 as an online platform. It uses an operational modeling system that consists of four steps: (1) a daily update of observed meteorological data by the German Weather Service, with consistency checks and interpolation; (2) an estimation of current soil moisture using the mesoscale hydrological model mHM (www.ufz.de/mHM); (3) calculation of a quantile-based soil moisture index (SMI) based on a 60 year data record; and (4) classification of the SMI into five drought classes ranging from abnormally dry to exceptional drought. Finally, a map is produced and published on a daily basis on www.ufz.de/droughtmonitor. Recently, the simulation results have been validated using historical drought damage data from the European Drought Impact Report Inventory (EDII) for Germany (Herber et al. 2018). This analysis revealed that the accepted soil moisture threshold used to define drought events (the 20th percentile) even underestimates the observed drought damage in Germany. The integration of existing local moisture observations can help to further improve the soil moisture predictions and to learn about limitations of the operational system. We present a comparison of simulated soil moisture from the operational drought monitor against various observations derived from soil moisture monitoring networks, cosmic-ray neutron stations, and other instruments at different locations in Germany including but not limited to the TERENO observatories.

Affiliation: Gembloux Agro-Bio Tech, Université de Liège, France, Personal website

by Sarah Garré, Bernard Longdoz, Vincent Leemans

Finding solutions to adapt to the impact of global change is still one of the most important challenges of the 21st century. While huge modelling advances have been made and experimental methods to study the earth system are diversifying, a major problem to find those solutions remains: the unclosed gap between spatial and temporal scales. Results from controlled experiments in pots or greenhouses are sometimes difficult to relate to observations at the field and landscape scale due to the complex interplay of various processes. In addition, it has been hard to reproduce the future effects of climate change experimentally, leading to an inevitable extrapolation of model predictions to zones where no validation is possible.

Ecotrons provide unprecedented possibilities to study the soil-vegetation-microbiome-atmosphere continuum in highly controlled conditions and this under any climate scenario of interest. In this presentation, we will present the new Ecotron facility of the TERRA Research Center in Gembloux, Belgium designed to submit six mesocosms (small ecosystems) installed on lysimeters (2m² X 1,5m), to any climatic conditions desired and see how it can reproduce (or not) the current and future in-nature reality. In addition to the controlled variables, instrumentation is set up to monitor energy and water budget and ecosystem functions (yield and production quality, GHG mitigation and C storage, water retention capacity and leaching quality, soil fertility,…)

In the first experiment planned, we will compare the behavior of winter wheat plots submitted to present and future climatic conditions. The present conditions will mimic the situation already recorded on the Lonzee ICOS site. The future scenario will represent a typical year of the 2050-2070 period (drift of the mean annual value and increase of the temporal variabiity) and chosen for its ability to produce a contrasting crop yield.
The results will also give us information about the ability of ecotrons to reproduce a realistic ecosystem functionning and expose which aspects of plant growth and soil behavior are well and not well reproduced in the artificially generated climatic conditions. This phase is essential to demonstrate the actual potential and shortcomings of this cross-cutting research infrastructure.

Affiliations: 1 AREC Raumberg-Gumpenstein; 2 Universitity of Graz

Abstract: Grassland comprises 16% of the total area and more than half of the total agricultural land in Austria. Particularly in alpine areas, the impact of climate change on soil hydrology and vegetation will also have an enormous influence on grassland ecosystems. Moreover the interactions between local climate, vegetation and soil are still yet poorly investigated and understood. For the quantification of climate-induced changes in the soil water balance as well as in plant productivity, data of six weighable lysimeters from the ClimGrass-experiment at AREC Raumberg-Gumpenstein were used. In this multifactorial outdoor experiment future climatic conditions in form of increasing temperatures and increasing CO2-concentrations are simulated on an alpine site within a managed permanent grassland ecosystem. As the main research focus at AREC is put on applied science, parameters such as water productivity under future climatic conditions are of interest. First analyses of lysimeter data covering all three cuts from the experimental year 2015 showed the highest water productivity in the treatment with increased CO2-concentration. Correspondingly, water productivity was of 8 % lower in the CO2 and temperature enhanced treatment compared to the ambient conditions. Overall, the impact of the modified climatic factors was higher on biomass production than on evapotranspiration. To gain better understanding of the underlying processes, more data of different years would be desirable.

Affiliations: 1 Leibniz-Centre of Agricultural Landscape Research (ZALF) e.V.; 2 Federal Institute for Geosciences and Natural Resources (BGR)

Abstract: The water management of shallow water table sites is often the subject of discussion between different interest groups. The stakeholders often have different ideas concerning the target water levels at these sites. A good level of knowledge about the effects of the different water levels on the water budget components is a precondition for the development of compromises in a well‐balanced water resource management. A lysimeter station located at a grassland site within the Spreewald wetland was used to analyse the impact of increased target water levels in early spring on the development of the water budget components during the vegetation period from April to September. The higher target water levels in spring should help to increase the water storage capacity of the wetland and to buffer the subsidence of the water levels in summer. We have compared the measured water balance components and groundwater levels of four water management variants from 2014 to 2017. One variant represent the actual situation of a typical grassland site of the Spreewald wetland. Two variants represent a regime with higher groundwater levels in early spring, as discussed between water managers, nature conversation groups and farmers of the region and one variant simulates constant, slightly lower water levels, optimal for agricultural production. The higher water tables resulted in an increase of evapotranspiration. The additionally stored water buffered the effect of higher evapotranspiration for some weeks, but not for the whole vegetation period. The varying meteorological conditions of the investigation period from 2014 to 2017 and especially periods with extreme precipitation had a much greater effect on the water budget than raising target water levels in spring. In the long term, higher water tables lead to a change in the vegetation composition, being the main reason for increasing evapotranspiration values.

Affiliations: 1 Leibniz Centre for Agricultural Landscape Research (ZALF); 2 Forschungszentrum Jülich GmbH; 3 Helmholtz Centre for Environmental Research — UFZ; 4 Helmholtz-Centre for Environmental Research - UFZ; 5 Helmholtz Centre for Environmental Research - UFZ

Abstract: Future food production is expected to be affected by climate change, because it will alter the crop water balance components, such as water storage, evapotranspiration and drainage. Variations in weather conditions could explain more than 50% of the variability of wheat yield. Higher temperatures and lower rainfall amounts mainly limit the actual evapotranspiration and reduce soil water storage, which in turn affect crop yield and water use efficiency. To study these effects, lysimeters soil were moved to sites with contrasting climatic conditions (space for time concept) and monitored. In this contribution, yield, evapotranspiration, and changes in soil water storage from lysimeters soils for a period from 2011 until 2017 were analyzed. Data were obtained from a German wide monitoring network of lysimeter stations (TERENO-SOILCan), which was established across a rainfall and temperature transect, and lysimeters were transferred between the stations to subject them to different climate regimes. A uniform crop management (crop type, fertilizer, growth regulator, tillage, and use of pesticides) and crop rotation allows investigating the response of soil water storage and cropping water use efficiency for different soil types to a change in climate conditions.

Affiliations: Swiss Federal Institute of Technology Zurich (ETH Zurich)

Abstract: Weighing lysimeters are a well-established means of accurately obtaining local-scale estimates of actual evapotranspiration (ET), precipitation, and seepage within soils. At the Rietholzbach research catchment in northeastern Switzerland, two weighing lysimeters are in operation. One is a recently installed state-of-the-art monolithic mini-lysimeter with a tension-controlled lower boundary; the other is a large backfilled free-drainage lysimeter in operation since 1976. For this study, the mini-lysimeter measurements were processed using the Adaptive-Window and Adaptive-Threshold (AWAT) filter. The resulting water-balance estimates were then compared with those of the lower resolution large lysimeter, whose processing has remained unchanged since its installation. A number of additional, retrospectively applicable processing steps for the large lysimeter were then tested to mitigate the main sources of error for this instrument. Those found to be most beneficial were the application of a 10-min moving average to the mass measurements and the setting of ET and condensation to zero in hours with liquid precipitation. In spite of the differences in design, a generally close agreement between the two lysimeters was observed, which was further improved with the optimized large-lysimeter processing. A comparison of the lysimeter mass increases associated with liquid precipitation further revealed, however, that the large lysimeter experiences a previously unknown under-catch of about 11.5%, which could also be important for other lysimeter facilities. This under-catch led to a reduction in large-lysimeter seepage during the analyzed period, although the ET flux was not found to be affected.

Affiliations: JR-AquaConSol Ltd.

Abstract: Although lysimeters supply data at the point scale, lysimeters are an indispensable tool for calibrating and validating numerical models for water movement, solute transport and heat flow in soil and unsaturated zone. Coupling these models with groundwater models we get a tool to predict the impact of different land use practices on water resource management. Lysimeters are operated at the agricultural test site Wagna, Austria. The technical lysimeter design consists of: high-resolution weighing cells, a suction controlled lower boundary condition for sucking off seepage water, thus emulating undisturbed field conditions, comparative soil temperature, water content and matrix potential measurements inside and outside the lysimeter at different depths, an installation of the lysimeters directly into test plots and a removable upper lysimeter ring enabling machinery soil tillage. The aquifer "Westliches Leibnitzer Feld", is a significant resource for regional and supraregional drinking water supply for more than 100,000 inhabitants, but the region also provides excellent agricultural conditions. Sandy soil conditions and groundwater depths beneath the root zone allow the use of a unidirectional sequential coupling of the unsaturated water flow and nitrate transport model SIMWASER/ STOTRASIM – calibrated and validated on lysimeter measurements - with FEFLOW. Considering separated inputs of water and nitrogen into groundwater out of surface water bodies, agricultural, residential and forested areas, first simulation results match observed groundwater tables, but underestimate nitrate concentrations in general. Thus, multiple scenarios assuming higher nitrogen inputs at the surface are simulated to converge with measured nitrate concentrations. Preliminary results indicate that N-input into the groundwater is strongly dominated by contributions of agricultural land. For modeling the fate of the herbicide S-metolachlor (2-chloro-N-(2-ethyl-6-methylphenyl)-N-[(1S)-2-methoxy-1-methylethyl]acetamide; SMET) and the main metabolite metolachlor ethane sulfonic acid (MESA) at the Westliches Leibnitzer Feld (WLF) aquifer a modeling approach based on coupling the one-dimensional vadose zone model PEARL and the two-dimensional groundwater flow and solute transport model FEFLOW was developed. To calibrate the one-dimensional pesticide fate model, we used leachate concentrations of SMET and MESA from lysimeter experiments. Additionally, samples of representative soil types in the WLF aquifer were analyzed to infer SMET- and MESA-specific fate parameters, which were used for the PEARL model. The results show that using SMET fate parameters derived from the lysimeter data considerably improved the fit of the simulation results with the field observations compared with the application of standard laboratory-derived fate parameters accounting for soil type differences.

Affiliations: 1 UHasselt; 2 UAntwerp; 3 Vrije Universiteit Brussel

Abstract: Ecotrons are large scale research infrastructures offering a compromise between high level of control and recording ecosystem processes at a relevant spatial scale. Since ecosystem functioning is strongly influenced by climatic variables, simulating weather conditions and stochasticity accurately is crucial to ensure that results are not disconnected from reality. Here, we describe how to use input from an ICOS ecosystem tower to simulate in real time weather conditions in an ecotron facility. The ecotrons are filled with lysimeters containing dry heathland ecosystem. The parameters are controlled are not only atmospheric (temperature, relative humidity, precipitation, CO2 concentration) but also belowground (temperature and water tension). We illustrate the accuracy at which our ecotron-ICOS tower design can reproduce these parameters. We also describe how we plan to alter the ICOS input for two climate-change experiments.

Affiliations: 1 LIEC UMR 7360 CNRS UL - GISFI; 2 GeoRessources UMR 7359 CNRS UL CREGU - GISFI; 3 GISFI; 4 INERIS

Abstract: In industrial sites historically contaminated by coal-tar (abandoned coking plants), other families of organic pollutants than the 16 PAHs (polycyclic aromatic hydrocarbons) classified by the US-EPA can occur and induce potential risk for groundwater resources. Polar PACs (polycyclic aromatic compounds), especially oxygenated PACs (O-PACs), are present in the initial pollution and can also be generated over time. Their aqueous solubility is much greater than those of the PAHs. For these reasons, we need to increase our knowledge on polar PACs in order to better predict their behavior and the potential on-site risk. Two different column experiments were carried out on coking plant soils under saturated (laboratory scale) and unsaturated (lysimeter column on the GISFI experimental station) conditions. On the laboratory scale, different conditions were applied on two different columns: a column with raw coking plant soil and a column with the same soil previously preheated (100°C) under inert atmosphere in order to increase the availability of the contaminants. Different nature (ionic strength, organic matter concentration) of the percolating solutions as well as the flow of the solution were tested in order to evaluate consequences on PAC mobilization. On the field scale (lysimetric column), leaching tests were performed on a lysimeter column (2 m3) filled with a contaminated soil sampled on a former coking plant site. An artificial rain was applied during 4 consecutive days (6 hours per day at 10L/h) twice. The solutions were collected at the bottom of the lysimeter as well as at several depths of the column. Our works confirm a preferential release of polar PACs compared with PAHs. Polar PACs and low molecular weight PAHs (LMW-PAHs) are mainly released according to a coal-tar dissolution mechanism (Raoult’s law). In the case of laboratory column tests, satisfactory predictions are obtained for polar PAC release when the pollution is “fresh” (available), it is not the case for the raw soil, suggesting that availability is a major parameter that needs to be taken into account for historical contamination impacted by aging. Leaching tests performed on the lysimeter column confirm a release of PACs according to a coal-tar dissolution mechanism, although PAC concentrations are much lower compared to those predicted by the Raoult’s law. Whatever the experimental scale (laboratory or lysimeter columns), a clear influence of the biological compartment has been evidenced. Once the PAHs and polar PACs are dissolved into water, they are highly biodegraded.

Affiliations: 1 CNRS; 2 GISFI; 3 Forschungszentrum Jülich GmbH

Abstract: Due to past industrial activities, there are large areas of soil contaminated by metallic and /or organic pollution. Remediation strategies are conducted, especially when there is a high risk for human health and for the environment, or when there is land use pressure. However, large areas of soil are abandoned and unused, while they can still be contaminated. Natural attenuation is a strategy leading to a reduction of contamination with time due to biotic and abiotic processes. It is therefore important to follow the long-term fate of contamination under natural attenuation. Lysimeter technology can be useful to follow the solute transfer, including pollutants, from the soil to the solutions. Within the frame of a collaboration between TERENO SOILCan and GISFI lysimeter station, a series of six lysimeters were set up with an industrial polluted soil from a former coking plant site. The objectives are to follow the fate and fluxes of metallic and organic pollutants in the leachates under two slightly differing climatic conditions. Most of the time, historically polluted soil has been disturbed when the factory was dismantled and is very heterogenous, therefore, undisturbed monoliths cannot be used. The soil contaminated with metallic trace elements and polycyclic aromatic hydrocarbons was excavated, homogenized using a mechanical shovel and transported to Homécourt (France) and to Jülich (Germany). Then, at both sites, 3 lysimeters were filled step by step in a procedure to limit heterogenous compaction at Jülich and at Homécourt. Soil samples were taken at each step of the lysimeter filling and analysed. Technical questions regarding sample collection and saving before analysis and flux determination are discussed. The results of the first year of the study will be presented and discussed.

Affiliations: 1 Karlsruhe Institute of Technology KIT

Abstract: The study analyzes water balance components and nitrogen leaching measured at different lysimeters of the TERENO pre-alpine observatory. The weighable grassland lysimeters sites are located at three elevations (Graswang: 860, Rottenbuch: 770 and Fendt: 600m a.s.l.) which differ in mean annual temperature and annual precipitation. Mean annual air temperature ranges from 6.5°C at the highest elevation site Graswang to 8.6°C at the lowest elevation site Fendt; mean annual precipitation is highest at the Graswang site (1359 mm) and lowest at the Fendt site (982 mm). The lysimeters are split into intensively and extensively managed treatments. i.e. intensive refers to a slurry application of approx. 250 kg N ha-1 year-1 and four harvests per year, extensive refers to approx. 80 kg N ha-1 year-1 and two harvests. The study analyzes, how the amount and / or intensity of precipitation events trigger upward and downward water fluxes and coupled N transport. Precipitation measured by the weighable lysimeters is compared with precipitation observed by pluviometer or composite data from the respective lysimeter station site. The study analyzes the reasons and magnitude of differencesbetween those two observation methods. Furthermore, climatic and management effects on evapotranspiration and leaching are analyzed. The results show that at the lowest site, evapotranspiration losses are higher than at the highest elevation site. At the same time, leaching (measured at 1.4 m depth) is lower at the lowest site and highest at the highest lysimeter station. Impacts of agricultural management on water balance components are negligible compared to climate effects. Leaching rates of N are, in general, very low and suggest that N uptake by plants is very efficient. N leaching is dominated by nitrate, while ammonium and DON leaching rates are much lower. However, total nitrogen leaching rates are significantly higher under intensive management across sites as compared to extensive management. The results indicate a low risk of nitrate losses due to leaching or surface runoff. These finding are of particular interest with respect to practical applications in montane grassland management and in the light of the specifications set by the German Fertilizer Ordinance.

Affiliations: 1 Universität Tübingen; 2 University of Hohenheim; 3 University of Tübingen

Abstract: The application of pesticides is an essential component of modern agriculture. The formation of a pool of pesticide residues in arable soils is widely observed due to the application of numerous compounds with different physico-chemical characteristics. This poses a potential risk for agro-ecosystems. Unlike pesticide fate in controlled lab studies with their regular variations of experimental factors (e.g. temperature, soil moisture), their fate in the field is controlled by on-site weather variations and farmers’ activities. Field studies better reflect the real environment and thus may be used to test the reliability of pesticide authorization data. To examine the dislocation of pesticides in pace with prevailing agricultural practices, we installed two lysimeters (tension-controlled suction plate field lysimeter, UGT, Müncheberg, Germany) in the Ammer catchment in South Germany. Soil types on the study sites are Luvisol and Cambisol-Pelosol. The set-up of our lysimeters allows regular soil tillage including ploughing and management operations throughout the whole monitoring campaign. Soil water is extracted by tension-controlled suction plates in two depths, below the ploughing horizon (50cm) and across the rooting zone in 100cm depth. In addition, we investigate the spatial and temporal distribution of freshly applied pesticides using core samples down to 50cm depth. Results of a previous pollutant inventory of the regarded study site soils shows residues of carboxamide, neonicotinoide, pyridazinone, strobilurine, triazine, triazole and urea compound group in a range between 0.1 – 21.1 ng/g even years after their last documented application as well as their dislocation from the surface to the subsoil. We will present first data from lysimeter and field sampling campaigns.

Affiliations: 1 Forschungszentrum Jülich GmbH (IBG-3); 2 Leibniz Centre for Agricultural Landscape Research ZALF; 3 University of Bayreuth Germany; 4 Forschungszentrum Jülich GmbH - IBG3

Abstract: To understand water fluxes and matter transport instruments are required, which precisely depict changes in soil hydraulic properties and soil solution. The acquisition of detailed measurements that determines these properties has been challenging to date, since monitoring and control of experimental parameters in the field is difficult already. Heterogeneous soil hydraulic properties affect soil chemical and hydrological variables at the field scale. Lysimeters represent observations from an intermediate scale between laboratory and field experiments. In order to transfer the gained lysimeter data to field scales, it requires representative data for the field. Hence, a lysimeter should cover the heterogeneity of the soil at field scale. Accordingly, the aim is to study if (1) a lysimeter with a surface of 1 m2 and height of 1.4 m represents all essential processes of the water balance and mass transfer of an extensively managed grassland and (2) how climate change conditions impact the magnitude and temporal dynamics of soil chemical and hydrological variables from an extensively managed grassland? The investigated SOILCan lysimeters in Rollesbroich and Selhausen are part of the TERENO Observatory Eifel/Lower Rhine Valley. Working with lysimeter replicates allows verifying the representativeness of lysimeters. The temporal, spatial and spatiotemporal variability of soil chemical and hydrological observations will be examined with statistical tests.

Affiliation: Mazingira Centre, International Livestock Research Institute (ILRI), Nairobi, Kenya, Personal website

Quantifying greenhouse gas (GHG) exchange between ecosystems and the atmosphere is essential to identify potential sinks and source and furthermore to evaluate potential changes in GHG fluxes caused by climate change and/or land use change. Besides a variety of existing methods that are capable of deriving GHG measurements from ecosystems in-situ, biogeochemical process models are particularly powerful to simulate deviations in GHG exchange in the future. Thereby the models could go as far as testing specific interventions to mitigate GHG emissions a priori. Neither individual measurement approaches nor models should be used independently as each method focuses on an individual scale or process and only the combination of as many approaches as possible will allow to understand an ecosystem and particularly ecosystem GHG exchange comprehensively. Here we give an example from a Central European grassland where soil greenhouse gas measurement, eddy covariance GHG flux measurement, isotopic measurements as well as several biogeochemical process models were used successfully in order to quantify GHG exchange in a first stage and reduce the GHG emissions in a second stage. The study not only evaluated the GHG exchange between the grassland system and the atmosphere but further included a full assessment of the productivity of the system in order to derive GHG emission intensities.

Affiliations: University of Innsbruck

Abstract: Nitrous oxide (N2O) is a strong greenhouse gas and an important ozone-depleting substance released from natural and anthropogenic sources, in particular microbial production in soils via nitrification and denitrification. The extent of these pathways – controlled by many competing factors, including moisture and climate, land use, and soil properties – is a key uncertainty in the nitrogen cycle. Future climate scenarios predict increased summer drought for European mountain grasslands, which has generally been observed to reduce N2O production, with large pulses upon rewetting that may even lead to overall increased emissions. However, the effects on specific N2O production and consumption pathways through both drought and rewetting events are relatively unknown, complicating efforts to understand observations, and model and mitigate emissions. We hypothesise that the relative importance of N2O release from nitrification will increase during drought treatment, however strong denitrification will cause an emission pulse upon rewetting that will increase the overall contribution of denitrification to N2O emissions. These changes will be evident in a changing N2O isotopic composition, in particular site preference, throughout drought and rewetting. This study uses cutting-edge isotopic techniques to investigate the effects of drought and rewetting on grassland N2O emissions, to gain a detailed understanding of N2O production and consumption pathways. 16 soil monoliths from an alpine grassland site (Kaserstatt Alm, Stubaital, Austria) are being subjected to different treatments affecting soil moisture: Drought followed by rewetting, drought followed by flooding, high precipitation, and control. A LICOR soil flux measurement system using automated chambers is directly interfaced with a Picarro spectrometer (model G2301-f) to monitor CO2, CH4 and H2O fluxes in a closed recirculation loop, with subsampling to a second Picarro spectrometer (model G5131-i) for measurements of N2O fluxes and isotopic composition. This innovative set up allows automated monitoring of monolith emissions at a time resolution of <2 hours throughout the drought, with a particular focus on N2O emission dynamics following rewetting. Soil moisture and temperature measurements, as well as soil sampling for the δ15N of NO3 and NH4, will be combined with automated isotope data in a modelling approach to allow a new depth of understanding of the effects of drought and rewetting on grassland N2O emissions.

Affiliations: 1 University of Rostock; 2 GFZ German Research Centre for Geosciences; 3 University of Göttingen

Abstract: Rewetting is considered as an appropriate measure to stop aerobic peat decomposition and to re-establish the net natural C sink function of peatlands. In this long-term study, we followed the development of a degraded fen peatland from drainage to year-round flooding. Based on CH4 and CO2 flux measurements, remote sensing-based vegetation monitoring, and a characterization of the microbial community composition, we develop a comprehensive perspective on the processes that control atmospheric carbon exchange in rewetted peatlands. Methane emissions peaked in the year after rewetting by increasing 100 fold compared to pre-rewetting levels. Methane emissions lowered in the following years but were still considerably higher than in pristine peatlands. These elevated methane emissions occur in parallel to high abundances of methane producing archaea whilst, at the same time, the number of methane oxidizing bacteria is relatively low. High CH4 emissions in the initial rewetting phase are considered to be (at least partially) compensated because CO2 release by aerobic respiration decreases. However, our results indicate that flooding does sustainably affect peatland vegetation, causing a lower plant coverage and a shortening of the growing season. As a result, not only CO2 release by ecosystem respiration decreases, but also CO2 uptake by canopy photosynthesis is diminished to the same degree. Our study highlights the importance of a multi-year monitoring to cover the dynamic ecosystem development within the initial rewetting phase. We further emphasize the relevance of interdisciplinary approaches to understand the complex interactions between ecosystem compartments as basic controls for GHG exchange.

Affiliations: 1Karlsruhe Institute of Technology, Institute of Meteorology and Climate Research, Garmisch-Partenkirchen, 2Senckenberg Biodiversity and Climate Research Centre (BiK-F), 3Vietnam Academy for Agricultural Sciences

Abstract: Rice production significantly contributes to global anthropogenic GHG emissions. Particularly CH4, which is produced under anaerobic conditions, is estimated to contribute to roughly 10% of total anthropogenic CH4 emissions (IPCC 2014). However, regional GHG inventories from agricultural production systems in general are still subject to great uncertainties due to mostly scarce information of field management (e.g., duration of flooding, harvest residues and manure application) and the complexity of environmental factors that determine soil C and N cycling and the resulting production and transport of GHGs. Here we present inventories of the two GHGs CH4 and N2O from major rice producing areas in Vietnam and the Philippines. The inventories are based on an IPCC Tier 3 approach using the process based biogeochemical model LandscapeDNDC, which has been linked to newly developed spatial land use and land management databases (e.g., weather, soil properties, and detailed field management data). Although the inventories likely represent the best of current knowledge regarding GHG emissions from those regions, we also show that uncertainties regarding field management are large suggesting that the bottleneck of more precise GHG inventories depends on the improvement of regional information of field management.

Affiliations: 1Karlsruhe Institute of Technology, Institute of Meteorology and Climate Research, Garmisch-Partenkirchen, 2Senckenberg Biodiversity and Climate Research Centre (BiK-F), 3Vietnam Academy for Agricultural Sciences, 2Empa, Swiss Federal Laboratories for Materials Science and Technology, Laboratory for Air Pollution / Environmental Technology, Dübendorf, Switzerland

Abstract: The excessive use of Nitrogen (N) fertilizers has boosted global population growth, but also has had detrimental effects on the environment, such as increased soil emissions of nitrous oxide (N2O), or Nitrate (NO3-) leaching. Process-based biogeochemical models are increasingly used to assess the fate of N compounds in the environment, to track soil N2O emission sources and to develop targeted strategies to mitigate soil N2O emissions as well as nitrate leaching from agricultural systems. Although sophisticated process based models consider the relevant N cycling processes, only a fraction of the modeled quantities can be determined with reasonable effort, and some processes are even inaccessible with direct measurements. As a consequence, the improvement and validation of biogeochemical models requires information N cycle processes. The isotopic composition of large N pools, such as the soil N pool, reflect the dominant N cycle processes in a specific environment and, thus, may be used as additional integrative validation quantity. While soil 15N changes on decadal time scale, N2O emitted to the atmosphere carries process information on a time scale of hours to days. Progress in instrument development allows on-line determination of, for instance, N2O isotopic composition in daily resolution. Especially the isotopic quantity site preference (SP), which reflects the relative abundance of N2O in which the central N of N2O has been substituted by 15N compared to the terminal N atom, has been identified as a powerful tool for N2O source partitioning. Here we present a module (SIMONE) that uses model output of a biogeochemical model (here LandscapeDNDC) and isotopic fractionation factors to track changes in the isotopic composition of the various N compounds. The performance of SIMONE will be discussed based on a comparison of simulated and measured 15N-N2O, 15N-NO3- and ammonium (15N-NH4+) using datasets obtained during field campaigns conducted at the grassland sites Chamau and Beromünster, central Switzerland. In addition, we will show how SP can be used to refine process parameterizations in biogeochemical models.

Affiliations: Key Laboratory of Mountain Surface Processes and Ecological Regulation, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, 610041 Chengdu, China

Abstract: Anthropogenic activities such as land use change and management are altering soil biogeochemical processes and have altered terrestrial biogenic greenhouse fluxes, which results in increases of CH4 and N2O emissions that can contribute to climate change. Overall, the terrestrial biosphere has evident to be a net source of greenhouse gases to the atmosphere. Particularly, subtropical and tropical Asian countries have been identified as hot spot of CH4 and N2O emissions by both top-down and bottom-up approaches most likely due to agricultural production. Thus it is critical to mitigating CH4 and N2O emissions in subtropical and tropical Asian countries. Although numerous studies individually investigated CH4 and N2O emissions from most cropping systems at various spatial and temporal scales in these regions, the comprehensive measurements of CH4 and N2O emissions from different land uses and management intensities within one agricultural landscape is lacking in particular for the subtropical and tropical region. Therefore, we conducted (are conducting) multi-year and site field measurements of CH4 and N2O emissions from agricultural systems (wheat-maize, wheat-rice, rape-rice, permanent flooding rice and vegetable systems) with transect of management intensities and forest ecosystems from afforestation of upland cropland with chronosequences of stand age and vegetation recovery stage (i.e. abandoned cropland, grassland, grassland-shrub, shrub and secondary forest systems) in subtropical montane agricultural landscape of southwest China. Preliminary results showed that there were significant differences in CH4 and N2O emissions between agricultural systems and the response patterns of CH4 and N2O emissions to land management intensity were inconsistent among different agricultural systems. Afforestation significantly decreased N2O emissions but not significantly affected CH4 emissions. The average annual N2O fluxes for forest ecosystems found in the present study represent as the lowest values recorded for forest ecosystem worldwide. Moreover, the pronounced spatiotemporal variability of N2O and CH4 emissions were observed across different experimental sites and years, which might be well explained by dynamics of environmental variables of soil temperature and moisture as well as availability of soil C and N substrates related to production and consumption processes of N2O and CH4 in soils. Our study highlights the importance of proposing mitigation practices at landscape scale for avoiding trade-offs.

Affiliations: 1 GFZ German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam; 2 Landscape Ecology, University of Rostock, Justus-von-Liebig Weg 6, 18059 Rostock

Abstract: Drained and degraded peatlands are significant and longterm carbon dioxide sources and thus contribute to further climate warming. In the northeast German state of Mecklenburg-Vorpommern, an estimated 20-30 % of the statewide carbon dioxide emissions are attributed to drained peatlands and one tool in reducing greenhouse gas emissions is therefore the re-wetting of peatlands to restore their natural carbon sink capacity. This is expected to relatively quickly lead to a substantial decrease in carbon dioxide emissions or even CO2 uptake at the expense of an initially strongly increased emission of methane. Within the Terrestrial Environmental Observatory Network (TERENO), surface-atmosphere fluxes of CO2 and CH4 were measured at the TERENO Northeast German Lowland Observatory eddy covariance site Polder Zarnekow (Fluxnet ID: DE-Zrk), a formerly drained and rewetted rich fen located in the Peene River valley. Draining of the fen began in the 18th century and was intensified between 1960 and 1990, when the water table was lowered to >1 m below the surface. Mineralization of the peat caused the surface to subside to levels below the adjoining Peene River. The site was rewetted by inundation during the winter of 2004/2005 and a shallow lake with depths up to 1.2 m and an area of about 7 ha developed. A layer of organic sediment formed at the bottom of the lake, which originated from the fen’s former vegetation and has since been annually replenished by dying aquatic plants and helophytes. A first observation period predating the establishment of TERENO infrastructure covers the years 2007–2009 and continuous monitoring has been ongoing since 2013. Methane emissions have remained high and even a net CO2 loss was reported for the 2013-2014 study period despite massive vegetation growth. We present a re-processed time series of surface-atmosphere carbon fluxes as well as the annual carbon balances (excluding lateral transport) for at least five years. Drivers of the turbulent fluxes are analysed on various time scales and the role of observed temporary summertime drying of the lake in the unexpected net CO2 emission is analyzed.

Affiliations: 1Karlsruhe Institute of Technology, Institute of Meteorology and Climate Research, Garmisch-Partenkirchen

Abstract: Montane Grassland soils support environmental key functions such as carbon storage, nutrient and water retention, erosion control and biodiversity. At present, these soils functions are jeopardized by climate change and rapid land use/ management changes, which both are likely to be accelerated in coming decades. In the TERENO Bavarian Alps / Pre-Alps Observatory KIT, IMK-IFU is operating three grassland monitoring sites differing in elevation, and thus climatic conditions (E860: Graswang 860 m.a.s.l., E770 Rottenbuch 770 m.a.s.l., E600: Fendt 600 m.a.s.l.) and grassland management (intensive vs. extensive). This study revealed that already under current climate conditions, N uptake by plants is in the range of total N fertilization rates, thus, N losses to the environment such as microbial driven N2O emissions (< 1.0 kg N ha-1 yr-1, which is far below IPCC EF estimates) and nitrate leaching (<5 kg N ha-1 yr-1) of the montane grassland soils are comparable low. If considering other ecosystem N losses of NH3 (15 kg N ha-1 yr-1) and N2 emission (28 kg N ha-1 yr-1), it gets obvious that also under present climate conditions substantial N needs to be provided from mineralization of soil organic N, indicating soil N (and C) mining. Since the latter is associated with negative effects on soil fertility/productivity, C sequestration and GHG exchange and filter functions protecting water bodies, this trend is putting risks on central soil functions in the long term. Considering investigations from long term monitoring sites and scenario simulations with the process based LandscapeDNDC model, SUSALPS results also show that C and N sequestration dynamics of pre-alpine soils also depend on the level of former and current C and N soil stocks which are highly influenced by grassland management such as frequency, amount and quality of organic fertilization.

Affiliations: Uni Trier

Abstract: The EC-PeT * software presented encompasses a eddy-covariance calculation scheme which is a fork of the well-known EC-PACK and the quality control and assessment (QC/QA) scheme developed in course of the TERENO project. EC-PeT can be run either via a graphic user interface (GUI), via a command line interface, or completely headless, which makes it suitable both for novice users and for fully automated use. Unlike many other comparable suites, EC-PeT is completely parallelized to take advantage of multiple processors, which reduces the time needed to process huge amounts of data. EC-PACK was originally developed at the Dutch weather service (KNMI) and the universities of Utrecht and Wageningen and is one of the most well established eddy-covariance softwares. It follows the approach of Lenschow et al. (1994) to assign an error value to each flux value calculated. The process QC/QA scheme incorporates the calculation of error flags after Vickers & Mahrt (1997), quality classes after Foken & Wichura (1996), a few other QC tests, assessment of the original EC-PACK errors (van Dyke,1992), calculation errors after Finkelstein & Sims (2001), as well as target area matching algorithm, currently using the footprint model (Korman & Meixner, 2001). Configuration All tests can be individually enabled or disabled and even reconfigured by the user, if desired. The classification of the output, depending on the QC test results, is also fully configurable by the user. Extensive comparisons show that EC-Frame delivers results that are in accord with other major eddy-covariance software packages (e.g. TK3). * acronym of Latin "elaboratio concursuum perturbationum Treverensis" which means "eddy-covariance software from Trier".

Affiliation: Eawag - Swiss Federal Institute of Aquatic Science and Technology, Dept. Systems Analysis, Personal website

The challenge of building a process based model in a small catchment can be alleviated by fieldwork experience. But when catchments start being too big for experimental hydrology, the model building process becomes even more challenging. Our goal is to build a distributed process based model for simulating streamflow within the Mosel basin (30.000 km2) at 26 nested subcatchments. We propose a 2-stages approach. In a first “data-analysis” stage, similar to catchment classification studies, we seek whether signatures of catchment function can be explained by signatures of climate variability or indicators of geophysical characteristics of the catchment. These relations are then used to inform a distributed “perceptual” model of catchment function. In a second “hypothesis-testing” stage, the model is subject to a phase of testing and further improvement. The first “data-analysis” stage outlines the importance of geology in controlling quickflow vs. baseflow partitioning, of topography in controlling the shape of the hydrograph, as well as of precipitation in controlling the overall water balance. These insights were used to inform model decisions such as how to break-up the landscape, which structure to assign to each landscape element, and how to distribute model parameters in space. The second “hypothesis-testing” stage was used to confirm the validity of model decisions, and test model alternatives. Overall, we show that large part of the variability in the observed streamflow at the 26 catchments can be well simulated by a parsimonious model, with a few spatially distributed parameters. Our approach is a step towards a more systematic distributed model development approach.

Affiliations: 1 TU Berlin; 2 Politecnico di Milano; 3 Stadtwerke Düsseldorf

Abstract: River bank filtration is an important source of drinking water production in Germany. Due to high concentrations pathogenic bacteria and viruses in wastewater and surface waters linked with short travel times at river bank filtration sites these pathogens are often found in the groundwater. Depending on the properties of the porous media and the pathogens themselves they can be removed from the mobile water phase during migration through the subsurface. A 2D pathogen transport model was created for the river bank filtration site at the waterworks Flehe in Düsseldorf (Germany) for an optimal monitoring for viruses and pathogenic bacteria. A sensitivity analysis was performed employing 37 model parameter. These parameters are characterizing the subsurface heterogeneity and also pathogen properties. An initial parameter screening was performed using Morris Indices for local sensitivity to assess the most important parameters. The 10 most sensitive parameters, which accounted for 76% of the total sensitivity over all observation locations and timesteps, where used for a more thorough global sensitivity analysis using Quasi-Monte-Carlo-sampling and Sobol Indices. This methodology gives valuable insight for the model calibration allowing to assess which parametric uncertainties affect the result uncertainty most and if parameters interact with each other. Furthermore, the obtained sensitivities can be used to reduce calibration complexity through estimating the result uncertainty lost by setting parameters constant, which are less sensitive or have a small pre-calibration uncertainty. These analyses are used to design an optimized monitoring concept for the ongoing monitoring of virus transport at the waterworks Flehe. Finally, the sensitivity and uncertainty analyses allows to design a simplified new model that can be used with a reduced complexity as risk assessment tool.

Affiliations: 1 UFZ Leipzig; 2 UFZ-Helmholtz Centre for Environmental Research

Abstract: Numerical groundwater models have become increasingly sophisticated and can now handle highly complicated problems. This capability comes at a price: the data demand has grown substantially. Obtaining reliable data of the subsurface is notoriously difficult. Hydrogeological maps with thickness, slope, and hydraulic conductivity or transmissivity exist, but the level of their reliability is not always clear. The rampant spatial variability of hydraulic properties further muddies the waters. In view of these complexities, much simpler analytical groundwater models might still deliver acceptable results for practical applications. We therefore embarked on a comparative study in which we pitch a state-of-the-art numerical groundwater model against a recently developed highly simplified analytical solution of the linearized Boussinesq equation for a strip aquifer. On one hand, we will examine under what conditions the analytical model loses its validity. On the other, we will examine when the numerical model performance breaks down under increasingly scarce and error-prone data. We hope to develop criteria that guide the choice between these two extremes for practical applications. We will present initial results of our study.

Affiliations: 1 Laboratory of Hydrology and Geochemistry of Starsbourg

Abstract: Application of a low-dimensional integrated model at the catchment scale – case study of the Strengbach catchment Sylvain WEILL,Benjamin JEANNOT,and Frederick DELAY Université de Strasbourg, CNRS, ENGEES, LHyGeS UMR 7517, F-67000 Strasbourg, France An integrated hydrologic model that couples a low-dimensional subsurface model with a diffusive wave surface model is applied to the Strengbach catchment, located in the mid-mountain Vosges Massif (elevation 880-1150 m) in France. The Strengbach catchment is the site of the Observatoire Hydro-Géochimique de l’Environnement (OHGE), a French Critical Zone observatory labeled and founded by the French National Research Institute CNRS, and part of the French research infrastructure OZCAR (the French network of Critical Zone observatories - applications and research). The aim of this application is to demonstrate the ability of low-dimensional approaches – i.e. physically based approaches with moderate level of complexity – to reproduce streamflow generation in a complex environment with steep topography and very transient hydrologic behavior. The model is calibrated using streamflow measurements. Residence time distribution and evolution of water storage are also computed as by-products associated with flow simulations. These variables should link the hydrological response of the catchment to the geochemical signature of waters and geophysical gravimetric investigations locating groundwater bodies in the subsurface and their evolution over time.

Affiliations: 1 Research Centre Jülich; 2 Gent University

Abstract: In order to evaluate the impact of climate change and human water use on the terrestrial water and energy cycle, a climatology of terrestrial water and energy states, fluxes and storages over large spatial scales is required. Arguably, a consistent terrestrial climatology is impossible to derive from measurements alone, because of scarce and often inconsistent observations, which are dispersed between many different national and international research and governmental institutions. Here, models may be valuable auxiliary tools to close information gaps in measurements in space and time. While these models highly simplify the natural complexity of the terrestrial system and may not agree with point measurements to an arbitrary accuracy, bulk mass fluxes and dynamics may be well-reproduced. In this study, an attempt was made to derive a physically consistent terrestrial climatology from integrated simulations of the terrestrial system from groundwater across the land surface into the atmosphere applying the Terrestrial Systems Modeling Platform (TerrSysMP) over continental Europe. In TerrSysMP, groundwater-to-atmosphere (G2A) closure of the coupled water and energy balances results in a physically consistent, dynamic equilibrium of the coupled subsurface-land surface-atmosphere system honoring two-way, non-linear feedback processes. Following the protocol of the Coordinated Regional Climate Downscaling Experiment (CORDEX), we present simulation results from a validation time period commencing in 1989 spanning almost 30 years. The resulting time series constitute a first terrestrial G2A climatology including groundwater, which can be used to assess future potential trends and anomalies. In addition, we will discuss the limitations of our simulation approach and the challenges of assessing the results utilizing local measurements and regional modeling results from previous studies.

Affiliations: 1 Wageningen University; 2 German Research Center for GeoSciences - GFZ

Abstract: Droughts are caused by low precipitation or a specific temperature regime which impacts snow melt. In catchment perspective, droughts can be classified into meteorological (no/low precipitation), soil moisture (low soil water) and hydrological (low discharge) droughts. From the view of a hydrologist, soil moisture and hydrological droughts are of major interests. This requires understanding the storage and release of water in times of droughts. In terms of modelling, drought representation is highly challenging. While flood modelling is strongly controlled by precipitation observation (errors), drought modelling largely depends on the model structure. In this context, models may vary in the way model parameters control water storage and release. Up to now, the relationship between model structure and simulations of soil moisture and hydrological droughts has not been investigated in detail. To fill this knowledge gap, a parameter sensitivity analysis (DELSA approach) was conducted for three frequently-used hydrological models (HBV, SAC, VIC) for 605 basins in the US with a high variability in hydroclimatological conditions. We applied DELSA to two drought indicators (duration and deficit) and used modelled soil moisture and discharge as target variables for the sensitivity analysis. Moreover, the relationship between parameter sensitivity and climate characteristics, e.g. temperature, seasonality, was investigated. Our results show that different model parameters control drought characteristics. Hereby both flux and storage parameters are sensitive to drought duration and deficit. The way in which soil moisture and hydrological droughts are represented varies between the three models. In HBV and SAC, snow parameters are dominant in cold regions, while soil and evapotranspiration parameters are most sensitive in warm regions. In contrast, groundwater parameters are dominant in VIC. The spatial variability in dominant model parameters is hereby higher for drought duration than for drought deficit. Thus, different processes control drought representations in these three models. These controversial results might impact long-term simulation and prediction of droughts.

Affiliations: 1 Ecole des Mines de Paris - Mines ParisTech; 2 Université de Rouen-Normandie

Abstract: Hydrological variables, such as river flow rate or groundwater levels, have only little resemblance with the effective precipitation signal that feeds continental hydrosystems. Even more disturbing is the fact that catchments undergoing similar climatic conditions do not produce similar hydrological responses. Therefore, deciphering how a climatic signal is altered and re-transcribed into hydrological signals is a difficult task. Nevertheless, an important challenge for water management is to understand how watersheds react to different time scales of climatic forcing, as it is key to predicting extreme hydrological events (severe droughts and floods). In the last decades, several authors have used the spectral analysis framework to tackle this challenge. Indeed, some fundamental 1D flow equations have been solved in the frequency-domain and applied to the interpretation of Fourier-transformed data in order to describe hydrosystems, as well as characteristic time scales of groundwater flow. However, these approaches either neglected surface processes or transfer through the unsaturated zone, which may have a considerable role in shaping the spectral content of hydrological variables. Here, we build upon previous mathematical developments to derive a new transfer function (TF) for an idealized 1D hydrosystem, incorporating diffuse surface runoff, flow through the unsaturated zone and through a Dupuit aquifer. By comparing the new theoretical TF to the outputs of a distributed pseudo-3D hydrological model, we show that, with the appropriate processing, experimental TF of river discharge and hydraulic heads can be explained with a high degree of reliability. In addition, we observe that the impact of one compartment on climatic signal transformation is mainly dependent on its characteristic time scale of transfer, relative to the one of the other compartments. For instance, the impact of the unsaturated zone is typically highest when the aquifer low inertia and when the fraction of infiltration is elevated. Having validated the new approach, we apply it on actual hydrological data from several catchments in Northern France. We thereby demonstrate that we have refined the interpretation of such data in the spectral domain, and that we are able to estimate key properties of each functional compartment in the system. Finally, we conclude that our approach is a practical way to understand which processes are responsible for the observed variability in catchment responses, and that the frequency-domain analysis of data has potential in assisting calibration procedures of more realistic 3D distributed models.

Affiliations: Belgian Nuclear Research Centre (SCK∙CEN)

Abstract: Due to computational limitations, simulating dynamic subsurface processes at high spatial resolution over large domains often requires coupling a coarser scale with a finer scale groundwater model. The underlying idea is that the coarse-scale model (CSM) provides the general patterns while the fine-scale model (FSM) simulates the considered processes in those parts of the simulation domain where high spatial resolution is needed. When there is a large spatial resolution gap between CSM and FSM, this can be very challenging. Here we propose a one-way nested grid coupling between MODFLOW-2005 and MODFLOW-NWT to simulate groundwater-induced flooding. We apply our approach to the Nete catchment (Belgium), which covers 1700 km2 and is innerved by a dense network of streams and drains. The CSM has a spatial extent of 96 x 146 400-m width square grid cells and 16 layers of various thicknesses in the vertical direction which goes down to -300 m a.s.l. The FSM is nested within the CSM of which it occupies floodplains with an additional buffer of 400 m. These zones consist of about 38,500,000 active 10 m x 10 m grid cells distributed over 5 horizontal layers with thickness ranging from 1 m to 11 m. Both the CSM and FSM are ran in steady-state mode at each time step (∆t) of the simulation, as the focus is on long-term flooding and ∆t > 50 years. The proposed coupling is as follows. The lateral boundary conditions (BCs) of the FSM consist of fixed heads interpolated from the CSM-simulated heads. To prevent over-constraining the FSM by using fixed heads for the bottom BC, we use fixed fluxes partially derived from the CSM-simulated fluxes. For every 400 m x 400 m block of 10 m x 10 m FSM cells that corresponds to a CSM grid block, the mean bottom BC flux honors the associated CSM-simulated flux. Yet the inter-cell bottom BC flux variability is made proportional to the difference between starting and simulated heads obtained from a first FSM run parameterized with bottom BC fluxes directly interpolated from the CSM. This requires 2 FSM runs per CSM run, but produces consistent results. In addition, runoff in the FSM is simulated by means of an artificial top layer with very high conductivity while a custom nonlinear function based on channel geometry is used to model the relationship between recharge and water levels in streams or drains.

Affiliations: 1 IRD IGE; 2 University Grenoble Alpes; 3 IRD

Abstract: West Africa undergoes an intensification of the hydrological cycle with more extreme precipitation events together with an intense land-use change due to increasing food demand. This lead the Critical Zone hydrological responses through new trajectories that have never been experienced before in term of flood intensity/frequency, water table dynamics, … To take into account these global changes and propose relevant modeling tools to these vulnerable areas, we build a physically based nested modeling approach spanning scales from small (few ha) headwater catchments to the whole West-Africa, with various resolutions from 100m2 to 1km2. To reduce uncertainties at regional scales we build on the AMMA-CATCH CZO located in Benin (Sudanian Climate) and Niger (Sahelian Climate) which provides long term data series and a knowledge base for local processes. This study proposes an example in north Benin of this modeling strategy to show how small scales processes associated with a huge heterogeneity that cannot be represented in a km² mesh help to design relevant regional scale models. We first used the Parflow/CLM physically based model on small headwater inland valleys known as Bas-fond which have been identified as the main producers of streamflow. We used such scales to identify the relevant features for runoff generation. We then transposed those results to a Parflow/CLM mesoscale model to reproduce mesoscale behavior.

Affiliations: Technical University of Crete

Abstract: The island of Crete is characterized by an uneven spatial and temporal rainfall distribution, which in combination with the intensive agricultural activity and the developed tourism, have made it one of the most drought-prone regions of Greece. The Koiliaris river basin, located in west Crete, is a Critical Zone Observatory, with a composite flow system: karstic springs which provide permanent low flows throughout the year are combined with surface intermittent flows. Climate scenarios, which mainly predict a rise in temperature and an upcoming decrease in flow for the Mediterranean region, are likely to provide important information, especially concerning the low flows, for the conservation of water resources in the area for the coming years. However, the uncertainty that exists in the input time series can significantly affect the model outputs, to an extent that they may become unreliable. The aim of this study is to estimate the predictions of climatic scenarios and their uncertainty in the hydrology of the Koiliaris river basin, using a modified version of the deterministic Soil Water Assessment Tool (SWAT model) which incorporates the karstic properties (i.e. the Karst-SWAT model). The model uses daily precipitation data from five stations and temperature data from two stations located in the area. Three climate change scenarios (from the regional climate model REMO) which portray the bandwidth of potential hydrologic states of the area were selected: a “favorable”, an “intermediate” and an “extreme”, in terms of predicted precipitation decrease. The precipitation and temperature time series were analyzed and stochastically modeled using the LARS-WG generator (Semenov and Barrow, 1997) which provides a better fit for time series with long inter-arrival times compared to Markovian generators. 90 Monte Carlo simulations (30 for each scenario) covering the 1974-2098 time period were used to drive the Karst-SWAT hydrological model and examine the range of uncertainty of the model output. Results provided confidence intervals for the predicted mean values of both karstic and surface flow according to the selected scenarios. The uncertainty estimates of the 3 scenarios indicated that even though there is a large uncertainty in the predictions of each storyline, the results suggested that there is more than 15% likelihood that runoff will significantly decrease after 2060 for the case of the “favorable” scenario, 75% for the “intermediate” scenario and 92% for the “extreme” scenario, imposing in this way additional constrains in managing the water supply and demand of the Region.

Affiliations:

Abstract: Understanding the spatio-temporal variability of the chemical composition of surface waters is a major issue for the scientific community, especially given the prospect of significant environnemental changes for the next decades. Up until now, the study of concentration-discharge relationships has been intensively used to assess the spatio-temporal variability of the water chemistry at watershed scales, however, the lack of independent estimations of the water transit times within catchments limits our ability to model and predict the water chemistry with the only use of geochemical approaches. This study demonstrates the potential of coupling mathematical hydrology and hydrochemical modeling to better understand the temporal and spatial variability of the composition of surface waters. In a first step, a hydrological and integrated depth model (NIM) has been used to constrain the distribution of the flow lines that are feeding the springs. In a second step, hydrochemical simulations with the code KIRMAT have been performed to calculate the evolution of the water chemistry along the flow lines. The results indicate that the concentrations in dissolved silica (H4SiO4) and in basic cations (Na+, K+, Mg2+, and Ca2+) in the spring waters are correctly reproduced with a simple integration along the flow lines. The results also show that the modest variability of the flow line distribution and of the flow velocity imply that the water transit times only vary from about 2 to 3.5 months from floods to drought events. These findings demonstrate that the chemostatic behavior of the spring chemistry, the modest variations of the solute concentrations (±15%) over a wide range of water discharges (×15), is a direct consequence of the strong hydrological control of the water transit times within the catchment. Overall, this work shows that the hydrochemical functioning of an elementary watershed such as the Strengbach catchment is relatively simple. The acquisition of the water chemistry can be explained by water-rock interaction processes that are spatially relatively homogeneous within the catchment.