Performance of High Resolution Rainfall Forecast from Weather Research Forecast Model Over Southern Agro Climatic Zone of Tamil Nadu

Abstract

Weather forecasting become important tool for the planning of everyday farming activities, particularly the medium range weather forecast is highly required to reduce both input and production loss. Tamil Nadu Agricultural University is providing block level medium range weather forecast for next six days with Weather Research Forecast (WRF) Model and having accuracy of 50 -70 per cent, varies with season. An attempt was made at Agro Climate Research Centre, Tamil Nadu Agricultural University, Coimbatore during 2019-21 to improve the accuracy of medium range rainfall forecast over different Agro Climate Zones of Tamil Nadu state by increasing the resolution from block level (9 km) to village level (3km) and altering the microphysics options in WRF v 4.2.1. The performance study tested four microphysics schemes viz., Kessler, WSM3, WSM5 and WSM6 for all the four seasons of 2020. The results obtained for the Southern Zone (SZ) is discussed in this paper and study concluded that the performance of high resolution village level (3km) rainfall forecast at Southern Agro Climatic Zone of Tamil Nadu generated from WRFv4.2.1 model was best with the WSM3 microphysics scheme followed by Kessler scheme, WSM5 and WSM6. Between the major rainfall seasons, Forecast Accuracy Index (FAI) was higher in NEM (0.56 to 0.73) than SWM (0.38 to 0.70), whereas the Forecast Usability Percent (FUP) was almost similar during both the South West Monsoon (SWM, 53.0 to 81.4) and North East Monsoon (NEM, 52.9 to 85.9). There was overestimation observed irrespective of season and microphysics options.