摘要

Derivation of Water Vapor Winds (WVWs) using water vapor channel (5.7 mu m- 7.1 mu m) of Very High Resolution Radiometer (VHRR) payload onboard India's KALPANA-1 geostationary meteorological satellite started in March, 2008 on a routine operational basis at India Meteorological Department (IMD), New Delhi using upgraded INSAT-2E meteorological data processing system of IMD. Recently a modification has been introduced in the algorithm used for derivation of this product. The primary change introduced is making use of first guess forecast fields obtained from National Center for Environmental Prediction (NCEP) Global Forecast System (GFS), at a resolution of 1 degrees x 1 degrees instead of forecasts of operational Limited Area Model (LAM) of IMD. In the present study, the quality check of operational WVWs has been carried out using ECMWF Model forecast data, covering the area 40 degrees E-129 degrees E and 40 degrees S-40 degrees N. Results show a significant increase in the number of good quality WVWs. The RMSE of KALPANA-1 derived WVWs, as compared to ECMWF first guess forecast field has improved to 5 m/s which are generally comparable with the similar results of other geostationary meteorological satellites in the world. Finally, these WVWs of improved quality are transmitted on GTS for operational use by the end users in both SATOB as well in BUFR formats.
Detailed error analysis has also been carried out for validation of WVWs using collocated radiosonde observations for the period from May 2008 to July 2009 and the available mid-upper level winds derived from European METEOSAT-7 data for the period from October to December 2008. The analysis with radiosonde data shows that before modification the bias of KALPANA-1 WVWs was variable between -2 and +4 m/s and after modification it has become almost steady to a value of 4 m/s. The RMSE of derived winds between 100 and 500 hPa levels with reference to the radiosonde winds ranged from 4.2 to 9.2 m/s before modification. However, from October 2008 onwards when modifications were made it has reduced to almost steady value of approximately 4 m/s. This is comparable with similar results generally obtained by the other satellite operators such as GMS (Japan) and METEOSAT (European). However, the bias is slightly higher as compared to that of other satellites. One of the reasons for observed high bias is the fact that fine details of temperature structure are sometimes not brought out properly in the model generated first guess forecast field. The overall results show consistent improvements in KALPANA-1 derived WVWs after using model first-guess wind field from NCEP instead of LAM. Case studies of specific cyclones have also been done and it is found that the KALPANA-1 derived WVWs are useful for predicting the future track of the cyclones. It is also found that the improved KALPANA-1 WVWs are in general in good agreement with METEOSAT-7 derived winds as they bring out the synoptic scale flow patterns very well.

  • 出版日期2010-10