Analysis of the resolution of precipitation data required to obtain robust results from a hydrodynamic sewer network model

Abstract

The need for precipitation data for calibrating hydrodynamic sewer network models is often compromised by using the nearest available rain gauges to study area. Due to the scarcity and irregular locations of the rain gauges, this way of satisfying the need for precipitation data can lead to incorrect conclusions with respect to the temporal and spatial patterns of precipitation, depending on the location of the rain gauges in the study area. Recent developments in the field of precipitation measurement by means of weather radar data open up new possibilities for the use of such data sources in hydrodynamic sewer network models. Even though weather radar provides precipitation information with a high temporal and spatial resolution, the raw radar data contains several sources of error and is inaccurate. The radar data are therefore often corrected and merged with ground measurements. The main objective of this study is to investigate the resolution of precipitation data required to obtain robust results in a hydrodynamic channel network model. The study area is a small catchment close to Munich in Bavaria, Germany. Data from the Isen weather radar station of the German Weather Service (DWD), which is located around 33 km from the study area, was used. Following the objectives of this study, various weather radar data products were processed in order to be used as input for a hydrodynamic sewer network model. The data with a temporal resolution of 5 minutes to 1h and a spatial resolution of 250 m x 250 m up to 1.000 m x 1.000 m form the basis for creation of datasets to be investigated. It has been observed that the use of high-resolution precipitation data leads to better model results, especially when the data is merged with rain gauges. However, it should be noted that the quality of the model results does not decrease linearly when the resolution of the precipitation data is reduced.