VERIFICATION OF DISTRICT-LEVEL WEATHER FORECAST OF KOLKATA AND ITS SUBURBS DURING MONSOON ‘2019 & 2020 FOR COMPARATIVE STUDY OF THE PERFORMANCE OF MODEL BETWEEN PRE COVID NON-LOCKDOWN AND COVID LOCKDOWN PERIOD

India Meteorological Department has started issuing district-level weather forecasts for up to 5 days on an operational basis from 1st June 2008. The weather parameters related to agro, namely rainfall, maximum and minimum temperature, wind speed, and direction, relative humidity, and cloudiness were chosen for outputs from the model. The rainfall forecast is generated based on multi-model ensemble techniques ( MME ) and ECMWF forecasts ( presently IMDGFS) are used for forecasting other parameters. These forecast generated for the districts of West Bengal by the model is further moderated by State Agro Met. Centre, Kolkata, and forwarded to six Agro Met. Field Units ( created by six agro-climatic zones in West Bengal ) and seven District Agro Met. Unit ( DAMU ) for preparation of weather-based District as well as Block level Agromet advisory bulletin which benefits the farmers in their crop production. Thus forecast verification of the model as well as moderated value for the monsoon season of 2019 and 2020 has been done to make a comparative study of the model performance concerning Kolkata and its suburbs based on Probability of Detection, False alarm, Heidke Skill score, Missing rate, Critical Success Index, True Skill Score, Hanssen, and Kuipers Index, etc. The monsoon rainfall of 2019 and 2020 was chosen to study the performance of the model concerning the pre-covid non-lockdown and covid lockdown period so that the effect of pollutants on the performance of the model can be analyzed. The verification results show that the model forecast, as well as a moderated forecast of this region, has to be more refined by taking inputs of other parameters and index that has been computed by different recent research works on this region because this region is under the influence of tropical climate. Moreover, the comparative study between monsoon 2019 and monsoon 2020 reveals that there have been changes in the performance of the model.