Simulation of heavy precipitation over Santacruz, Mumbai on 26 July 2005, using Mesoscale model

Abstract

An attempt has been made to simulate the unprecedented heavy precipitation of 94.4 cm in a day over Santacruz, Mumbai during 0300 UTC 26 July to 0300 UTC 27 July 2005. Three experiments have been conducted using Advanced Regional Prediction System model developed by Center for Analysis and Prediction of Storms of Oklahoma University, USA. In first experiment the model input at large domain size has been obtained using NCEP/NCAR reanalysis data at 2.5° × 2.5° lat.–lon. resolution. In other two experiments model input at large as well as at small domain sizes, have been obtained from NCEP/NCAR FNL data of 1° × 1° lat.–lon. resolution. In all three experiments model’s horizontal resolution is 40 km and integration period is 30 hours from 0000 UTC 26 July 2005. Based on the temporal distribution of observed rainfall rates it is considered that the rainfall of 38.1 cm during 0900–1200 UTC on 26 July could be due to cloud burst phenomenon and 56.3 cm from 1200 UTC of 26 July to 0300 UTC of 27 July has been due to continuous regeneration of thunderstorm activity under influence of mesoscale cloud complex. It is found that model forecast of rainfall in first experiment was qualitatively as well as quantitatively very poor. Among other two, experiment with large domain size has predicted better rainfall values and location compared to the experiment with small domain size. The larger domain has produced rainfall of 41 cm as against observed rain rate of 56.3 cm. during 1200 UTC of 26 July to 0300 UTC of 27 July. Divergence, vorticity, vertical velocity and moisture parameters are examined in relation with the various stages of the event. The maximum values of convergence, vorticity and moisture fluxes precede the initial phase of mature stage, however vertical velocity follows the later phase of mature stage. Vorticity budget over the location of maximum rainfall, revealed the significant role of tilting term in maintenance and dissipation of the cloud complex responsible for the event. The model has simulated mixing ratios of ice, snow and hail up to height of 15 km which matches with the observations that clouds reaching up to 15 km were present at the time of event of heavy precipitation.