Downscaling of NWP based meteorological variables for potential applications

Abstract

A general problem with output from weather forecasting models is that modeled variables often do not compare well with ground-based weather variables and complex downscaling procedures are necessary to convert Numerical Weather Prediction (NWP) weather variables into realistic weather variables on the ground. NWP models have now firmly established as major forecasting tool and there is increasing use of Ensemble Prediction Systems (EPS) based probabilistic information in support of decision-making by forecasters as well as other special users. However, the challenges in down scaling the model forecasts, normally applicable for a grid area commensurable with the model resolution, to a meaningful location-specific product are by no means trivial. An attempt has been made in this study to down scale coarse resolution NWP based meteorological parameters to fine resolution products. The meteorological parameters considered for this study are temperature and downward surface shortwave radiation (DSSR) as important for ecological studies at regional scale. The comparisons have been carried out for the DSSR and temperature from satellite with that of <i style="">in situ</i> data. The NWP based fine grid data were found to be in good agreement with satellite and <i style="">in situ</i> data as compared to that of coarse resolution NWP parameters. The variations of these remapped fine grid parameters have also been studied at local and regional scale over India.