Ensemble lagged forecasts of a monsoon depression over India using a mesoscale model

摘要

A monsoon depression formed over the Bay of Bengal, India, during 27 July 1999 and crossed the east coast of India on 28 July 1999. The system caused copious rainfall over the east coast of India and adjacent regions and is investigated in this study using ensemble lagged methods with the Pennsylvania State University/National Center for Atmospheric Research (PSU/NCAR) Fifth Generation Mesoscale Model (MM5). Two sets of experiments are designed with five members of ensembles in each set starting at 12 UTC 25 July 1999 and at preceding times separated by 3 hour intervals. In the first experiment, the National Center for Environmental Prediction (NCEP) reanalysis data is utilized in the forecast of a coarse grid spacing domain and subsequent nest down to a finer (30 km) grid spacing domain. In the second experiment, the NCEP reanalysis data is directly utilized in the 30 kin domain as initial/boundary conditions. The results of the ensemble average of both experiments are compared with the analysis and observations. It is found that at the initial times of verification, the ensemble average of the sea level pressure field corresponding to the second experiment has a larger horizontal structure and is more closer to NCEP reanalysis. However at later times of verification, the ensemble average of sea level pressure field corresponding to the first experiment is found to be better. The area averaged 24 hour accumulated precipitation of all the ensemble members have higher values corresponding to the first experiment as compared to the second experiment. Also, the spread of the area averaged 24 hour accumulated precipitation of all the ensemble members with respect to their respective ensemble average are higher for the first experiment as compared to the second experiment. The results of the study would be useful to operational weather forecasting centers in India as it would provide them different evaluating ways to develop and test the importance of ensembles.