Integrated Project (IP)
"From local scale processes to regional predictions"
Importance, societal challenges and relevance
The terrestrial environment experiences different changes due to e.g. climate change and land use changes. The Earth surface is therefore “measured & observed (Big Data)" in never seen manner and dimensions. Reliable predictions about the behavior of the environment according to the aforementioned changes needs the development and enhancement of robust and skillful land surface models with model parameters that are logically related to the mass of data and measurements.
IP scientific hypotheses
The scientific hypothesis of the IP is that regionalization and scaling methods combined with “Big Data” and local model supported measurement and monitoring strategies allow for a scale robust parameterisation of land surface models. Hence, one goal of this integrated project is to develop a chain of model approaches from highly resolved and detailed process models to reductionist regional scale models. A main hypothesis is that by connecting these models consistently with each other and to data of different spatial support, the predictive capacity of terrestrial system models will be improved. Furthermore, the IP will develop systematic multi-scale modeling protocols and consistent measurement designs/strategies for “Big Data” problems. A special focus in the IP is on new approaches for combining hydrological and ecological dynamics and processes in multi-scale land surface models.
The IP incorporates two methodological and two integrative pilot project pillars. The methodological pillars inverse modeling/optimization and scaling are covered in diverse workshops and established working groups. These two methodological core subjects are identified as important common ground in three research disciplines at the UFZ: hydrology, ecology and geophysics.
The two integrative pilot projects are the starting point for in-depth analyses and new method testing with regard to the aforementioned aims of the IP. The research focal point in the first pilot project is on the CO2 Balance of the Amazonian catchment by combining the advantages of local and global vegetation models as well as finding optimal scaling and optimization methods for computations on a large scale. Eddy Covariance data is used in order to optimize model parameters. The expertise of hydrologist, land surface modelers and ecologist is combined to handle this challenge and optimize and effectively scale the local vegetation model ( ) to the Amazonian basin scale.
The research focal point of the second pilot project is on the subsurface processes in large scale hydrologic models. Here the IP combines regionalization and scaling methods with multi-scale monitoring strategies to accomplish robust predictions of surface runoff, soil moisture and subsurface pressure heads on a regional scale. The new approach will combine the multi-scale model mHM-OGS ( , ) with model-supported measurement methods.