Large-scale hydrological modeling using the ECHSE* framework - Test of computational concepts and verification of satellite-based precipitation estimates*) Eco-hydrological simulation environment

在PROGRESS项目的背景下，申请人开发了ecse水文建模框架。ECHSE是一个通用的工具，有效地创建（生态）水文模拟模型。到目前为止，该框架已被用于建立中尺度流域的降雨径流模型。在拟议的项目中，计划测试基于ecse的水文模型在大规模应用中的可用性。研究将在印度的马哈纳迪河流域进行（大约100英里）。130000 sqkm)。拟议项目的第一个目标是获得在大空间尺度上应用基于ecse的水文模型的经验。研究的重点是空间离散化和并行化对计算效率的影响。计划中的实验旨在确定ecse通用模型核心的可能限制。另一个目标是为未来的大规模应用程序和计算昂贵的应用程序推导指导方针。第二个目标是研究将雨量计信息与基于卫星的TRMM- 3b42每日数据（TRMM：热带降雨测量任务）合并而成的降水估计的适用性。降水估计将通过交叉验证以及通过分析其在水文建模中的性能来验证。理想情况下，开发了一种具有稀疏雨量计网络的热带地区定量降水估计的空间可转移解决方案。这样的解决方案将对许多其他项目的实践和研究产生兴趣。利用空间连续协变量进行降雨区划是目前研究的一个活跃领域。预计将与位于Kharagpur的印度理工学院合作进行这项研究。特别是，计划与R. Singh教授和C. Chatterjee（农业与食品工程系）的小组合作。这个小组专门研究水文模型，包括预测。他们进行了几项与Mahanadi河流域水管理有关的研究。近年来，印度代表团的成员已经访问了申请人所在的合作研究和教学机构。

In the context of the PROGRESS project, the ECHSE hydrological modeling framework has been developed by the applicant. The ECHSE is a generic tool to efficiently create (eco)-hydrological simulation models. So far, the framework has been used to set up rainfall-runoff models for meso-scale river basins. In the proposed project, it is planned to test the usability of an ECHSE-based hydrological model for large-scale applications. Studies are to be conducted in the Mahanadi River Basin, India (approx. 130000 sqkm).The first goal of the proposed project is to gain experience with the application of ECHSE-based hydrological models at large spatial scales. In the focus of interest are the effects of spatial discretization and parallelization on computational efficiency. The planned experiments aim at the identification of possible limits of the ECHSE's generic model core. Another aim is the derivation of guidelines for future large-scale applications and computationally expensive applications in general.The second goal is to study the suitability of precipitation estimates created by merging rain gage information with satellite-based TRMM-3B42 data on a daily basis (TRMM: Tropical Rainfall Measuring Mission). Precipitation estimates are to be verified by cross-validation as well as by analyzing their performance in hydrological modeling. Ideally, a spatially transferable solution for quantitative precipitation estimation in tropical regions with sparse rain gage networks is developed. Such a solution would be of interest to many other projects in practice and research. Rainfall regionalization using spatially continuous co-variables is currently an active field of research.It is envisaged to conduct this study in cooperation with the Indian Institute of Technology at Kharagpur. In particular, collaboration is planned with the group of Prof. R. Singh and C. Chatterjee (Department of Agricultural & Food Engineering). This group is specialized on hydrological modeling, including forecasting. They conducted several studies related to water management in the Mahanadi River basin. Members of the Indian group already visited the applicant's home institute for collaborative research and teaching in recent years.