WRF- SWMM Coupled Model Performance Assessment With LCZ Classifications
Abstract
Spatial and temporal information plays a critical role in short-term weather forecasting using Weather Research and Forecast (WRF) models. Information on landuse features helps to identify their impact on the exchange of energy, moisture, and momentum between Earth's atmosphere and regional landscape. These features act as a major input to meteorological models and are responsible for the robustness of the models. In this study, a WRF model is developed to evaluate the impact of land use on rainfall patterns in Bangalore city, India. The city is a complex urban landscape with 77.5 % built-up, 18% vegetation, 3.83% water, and 9.7 % open space as observed in the year 2020. The city has witnessed a rapid and rather haphazard growth of urbanization over the past few years, making it difficult to develop a robust WRF model for weather forecasting. A total of 17 Local Climate Zone (LCZ) classifications are performed for Bangalore city using the World Urban Database and Access Portal Tools (WUDAPT) database. WRF simulations are performed for landuse classification developed using Advanced Wide Field Sensor (AWiFS) and WUDAPT database. The results showed a significant variation between landuse classification from AWiFS and LCZ classification from WUDAPT databases. Further, the WRF model is run operationally for various heavy rainfall events under "Default (Df)", "Updated with AWiFS (UA)", and "Updated with WUDAPT (UW)" scenarios. These scenarios are compared with the observed station data for heavy rainfall events in the years 2020-22. Hydrological models with WRF 1km output are simulated and validated for flood depth using Personal Computer Storm Water Management Model (PCSWMM). Amongst these scenarios, UW showed a more realistic and coherent behavior when compared with ground truth. The results for UA showed a spatial shift in rainfall patterns as compared to UW. Thus, updating the land surface information helps to improve the quality of WRF outputs as well as the flood forecasting system.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.A55P1334G