Advancing polar prediction: Enhancing short-range forecasting accuracy with Polar WRF at Bharati and Maitri stations in Antarctica
摘要
Polar regions are characterized by extreme weather and climate conditions, making accurate weather forecasting essential for ensuring the safety of outdoor operations, particularly in Antarctica. The Polar-optimized weather research and forecasting model (Polar-WRF) has emerged as a key tool, benefiting from modelling advancements and data coverage. India’s Maitri and Bharati research stations in Antarctica rely on Polar-WRF forecasts. Utilizing GFS T1534 data, Polar-WRF generates 72-hr forecasts with a 3-km grid resolution, accurately depicting Antarctic weather dynamics. This study compares forecasts with observed data to evaluate the recently upgraded Polar-WRF’s short-range performance over Maitri and Bharati from August to October 2021. Results show improvement in the forecast of mean sea level pressure from the previous model, with root mean square errors (RMSE) reduced from 3.4 to 2 hPa (24 hr) and 4.1 to 1.5 hPa (48 hr) at Maitri. The correlation coefficient improved for wind speed at 10 m from 0.7 (24 hr) to 0.9 and 0.67 to 0.9 (48 hr) at Maitri. The study underscores Polar-WRF’s reliability in its model domain, supporting safe operations during Indian Scientific Expeditions to Antarctica, contributing to improved decision-making, and enhancing understanding of Antarctic weather dynamics.
HighlightsImprovement in Polar WRF for Antarctic short-range forecasting. Enhanced resolution and accuracy for detailed forecasts in Polar weather predictions. Comparative analysis with ERA5 and MERRA2 datasets. Investigation of extreme weather events like blizzards by models. Evaluation of forecast lead times: 24, 48, and 72 hr. Rigorous assessment using correlation coefficients and RMSE. Reliability of reanalysis datasets as benchmarks. Foundation for future Polar WRF model refinements.