Evaluation of the performance of satellite products and microphysical schemes with the aim of forecasting early flood warnings in arid and semi-arid regions (a case study of northeastern Iran)

Abstract

Flood early warning requires rainfall data with a high temporal and spatial resolution for ood risk analysis to simulate ood dynamics in all small and large basins. However, such high-quality data are still very scarce in many developing countries. In this research, in order to identify the best and most up-to-date rainfall estimation tools for early ood forecasting in arid and semi-arid regions, the northeastern region of Iran with 17 meteorological stations and four rainfall events was investigated. The rainfall products of satellites (PERSIANN-CDR and GSMaP, ERA5, GPM CHIRPS) along with the most widely used microphysical schemes of Weather Research and Forecasting (WRF) model (Purdue-Lin (Lin), WRF Single-Moment class 3, 6 (WSM3, WSM6), and WRF Double-Moment class 6 (WDM6).were used for rainfall modeling. The e ciency of each of these models to forecasting the amount of rainfall was veri ed by four methods: Threat Scores (TS), False Alarm Ratio (FAR), Hit Rate (H), and False Alarm (F). Analysis of research ndings showed that the WRF meteorological model has better accuracy in rainfall modeling for the next 24 hours. In this model, Lin's microphysical scheme has the highest accuracy, and its threat score (TS) quantity is up to 98% e cient in some stations. The best accuracy of satellite products for estimating the amount of rainfall is up to 50%. This accuracy value is related to the satellite product (ERA5). In this method, an 18 km distance from the ground station is the best distance for setting up the space station, which is used for input to hydrological/hydraulic models. Based on the results of this research, by using the connection of the WRF model with hydrology/hydraulic models, it is possible to predict and simulate rainfall-runoff up to 72 hours before its occurrence. Also, by using these space stations, the amount of rainfall is estimated for the entire area of the basin and an early ood warning is issued.