Publication Details

Category Text Publication
Reference Category Journals
DOI 10.1007/s12665-014-3327-1
Document Shareable Link
Title (Primary) Inverse determination of groundwater inflow using water balance simulations
Author Körner, P.; Pluntke, T.; Sachse, A.; Böttcher, N.; Naumov, D.; Kolditz, O. ORCID logo ; Bernhofer, C.
Journal Environmental Earth Sciences
Year 2014
Department ENVINF
Volume 72
Issue 12
Page From 4757
Page To 4769
Language englisch
Keywords Ukraine; Water balance; Western Bug; BROOK90; Groundwater; Karst; LAI; OpenGeoSys
UFZ wide themes RU5;
Abstract The IWAS project Ukraine deals among others with the water balance in the headwaters of the river Western Bug. During investigations in the sub-catchment Sasiv, a gap in the water balance was detected by Pluntke et al. (Environ Earth Sci, this issue, 2014). The runoff of this sub-catchment is significantly higher than that of any other catchment in the region. Unexpectedly, in 2011 the runoff was observed to be higher than precipitation. We hypothesize additional subsurface flow in the Sasiv catchment from adjacent aquifers. The hydrological model BROOK90 was used to model water flows. In addition to publicly available meteorological input data from the NOAA and ECA&C databases, own precipitation measurements in Sasiv were used. The mean seasonal leaf area index (LAI)—cycle of forest, grass, and agricultural land was determined using satellite data. To improve the output quality, the model was adapted for the superior catchment Kamianka-Buzka/Western Bug (2,500 km2). It was calibrated for the period 2001–2006 and validated for 2007–2011. Infiltration and drainage parameters were unknown, and were therefore calibrated applying Monte Carlo simulations. It was not possible to calibrate the sub-catchment Sasiv (107 km2) due to high deviations of the natural water balance. Thus, the model was calibrated for the superior catchment and afterwards applied to the Sasiv catchment. The monthly estimated inflow was calculated based on, first, a constant term, which is a fixed monthly inflow with its minimum in October (22 mm) and its maximum in May (41 mm). Secondly, a dynamic term was added. Here, different dependencies were tested, e.g., precipitation, soil moisture, groundwater level, and modelled runoff. The best coefficient of determination was achieved using a monthly modelled runoff from the 4 months before the current time. The model showed a good performance in the validation period (R 2 = 0.88). On average, the runoff observed in Sasiv consists of 60 percent groundwater inflow. At the gauge Kamianka-Buzka, this fraction is still about 5 %. The determined LAI cycles and the usage of an additional precipitation station near Sasiv significantly improved the quality of the water balance modelling in the Sasiv sub-catchment. The presented hydrological model approach was extended by hydrogeological simulations to enhance the understanding of processes in groundwater and surface water interactions. Therefore, it was a special challenge to model with scarce data the heterogeneous hydrogeological character of the catchment. The multi-level calibration enables a comprehensible estimation of annual groundwater recharge rates (30–700 mm) of the shallow quaternary aquifer, which is drained by the river Bug system.
Persistent UFZ Identifier
Körner, P., Pluntke, T., Sachse, A., Böttcher, N., Naumov, D., Kolditz, O., Bernhofer, C. (2014):
Inverse determination of groundwater inflow using water balance simulations
Environ. Earth Sci. 72 (12), 4757 - 4769