Observational analysis and numerical simulation of sea breeze using WRF model over the Indian southeast coastal region

Sea-breeze (SB) is a predominant mesoscale atmospheric circulation over coastal areas and its detection sometimes is complicated mainly due to change in weather pattern. In this study the SB frequency between Chennai and Pondicherry over the east coast of India is studied for the period 2016-2019 using an objective selection method based on background winds, surface wind direction shifts and land-sea temperature gradient. Comparison of selection method with a conventional method based on changes in surface meteorological variables for the year 2016 indicated that the conventional method overestimates SB occurrence (152 days) by nearly 25% compared to the selection method (129 days). A detailed intercomparison for the year 2016 showed that the selection method provides more realistic detection of SB days with improved correlation (r = 0.75) than the conventional method (r= 0.27) due to consideration of the most relevant factors such as large-scale winds (> 11.0 m/s), surface temperature gradient (> 2 degrees C) and surface wind direction shift (> 22.5 degrees) over the study domain. Results of analysis for 2016-2019 period using selection method showed that SB occurs about 20% during winter, 40% during summer, 22% during SW monsoon, and 18% during NE monsoon, with more predominance in summer. The performance of Weather Research and Forecasting (WRF) model for the selection scheme in simulating the SB during 2016 is evaluated. WRF is able to simulate and capture the SB occurrence for 77 days of 129 days given by selection method. WRF is unable to simulate the upper air and surface flow characteristics sometimes resulting in less (similar to 20%) number of SB days simulation. Sensitivity experiments with two land surface physics schemes Noah and Noah-MP for three rejected cases showed improvement with Noah-MP in the simulation of temperature and winds resulting in detection of SB events.