Publications

Aravind, A; Srinivas, CV; Hegde, MN; Seshadri, H; Mohapatra, DK (2022). Sensitivity of surface roughness parameters on the simulation of boundary layer winds over a complex terrain site Kaiga in western India. METEOROLOGY AND ATMOSPHERIC PHYSICS, 134(4), 71.

Abstract
Kaiga, in the Western Ghats region, has a hilly topography and dense forest canopy. In this study, the sensitivity of surface roughness length (z(o)) on the boundary layer winds over Kaiga is investigated using the Weather Research and Forecasting (WRF) model for summer and winter seasons. Observational analysis shows high frequency of calm winds due to the flow blockage by the surrounding hills and surface drag from forest cover in the Kaiga valley. To realistically simulate the surface winds, surface roughness length is estimated using the sonic anemometer data and tower wind speed measurements employing the logarithmic profile relationship under neutral stability conditions for different months. The estimated z o varies in the range of 0.96-1.84 m in different seasons. To study the effect of forest cover on the wind field, WRF simulations are conducted using various roughness factors within the range of default value for evergreen broadleaf forest in the MODIS land-cover data used in the model and the highest estimated value, i.e., z(o) = 0.5 m, 0.75 m, 1.0 m, 1.25 m, and 1.50 m. Simulations show considerable overestimation of wind speed in control run (z(o)z(o) = 0.50 m) and experiments with increased roughness length reduced the bias in the surface wind speed, wind direction, and temperature. On average, the simulated winds are corrected by 2 m/s and 3 m/s for z(o) = 1.0 m and z(o) = 1.50 m, respectively. Increasing the surface roughness length also improved the prediction of the frequency of occurrence of calm winds to some extent. The assimilative capacity of the Kaiga valley atmosphere is evaluated by estimating the ventilation coefficient for the winter and summer seasons. It has been found that the model overestimated the surface winds and thus overpredicted the ventilation coefficient. By modifying the surface roughness length, the overestimation in the ventilation coefficient is corrected to an extent.

DOI:
10.1007/s00703-022-00912-7

ISSN:
1436-5065