An algorithm is developed to identify precipitation affected pixels and quantitatively measure the precipitation using Megha-Tropiques humidity sounder (SAPHIR) channels around water vapor absorption line at 183 GHz. Based on observed brightness temperatures at all the six channels of the SAPHIR, a probabilistic rain identification algorithm is proposed. The rain thus identified is subjected to intensive testing using SAPHIR and PR collocated dataset, that showed that false alarm and missing rain is below 0.9 mm/h. Further a radiative transfer simulations supported rain retrieval algorithm is developed that explained a correlation of 0.7 and rmse of 0.81 mm/h. When both precipitation detection and retrieval algorithms are applied the correlation marginally deteriorates but rmse reduces to 0.55 mm/h. Further comparisons are made of monthly, daily and instantaneous rain over different geographical regions from SAPHIR with corresponding rain values from GSMap, TRMM-3B42 V7 and TRMM-TMI/PR, etc. The paper provides details of algorithm development and validation results.
The south-west monsoon seasonal variations of the rainfall and water vapor over the Indian subcontinent and oceans are studied using microwave (MW) and near-Infrared (NIR) satellite measurements on monthly scales. The total precipitable water (TPW) derived from multi channel imaging data acquired with the Moderate Resolution Imaging Spectrometer (MODIS) on the Terra Spacecraft and rainfall data from merged infrared estimates calibrated against Tropical Rainfall Measuring Mission (TRMM) microwave data respectively are used. Since TPW is an important link connecting the various components of the hydrological cycle, its variability with rainfall on monthly scales have been used to meet this objective in the present study during the three successive contrasting good (normal), bad (drought) and good (normal) south-west monsoon years of 2001 to 2003 respectively.
The forthcoming Indian satellite Oceansat-2 to be launched in 2007 will carry a microwave scatterometer and an
ocean colour monitor onboard. The scatterometer, a Ku-band pencil beam sensor similar to that onboard Quikscat
satellite, will provide surface vector winds over global oceans with a two days repetivity. An algorithm for retrieving
wind vector from scatterometer has been developed with a solution ranking criteria of minimum normalized standard
deviation (NSD) of wind speeds derived using backscatter measurements through a geophysical model function
(GMF). Using Quikscat observational geometry and QSCAT-1 GMF, simulation based evaluation of algorithm
performance under different noise conditions and its comparison with standard algorithm known as Maximum
Likelihood Estimator (MLE) algorithm have been performed. Besides having retrieval performance closely
comparable with MLE, the present algorithm has quality and rain flagging provisions. Moreover, it is
computationally efficient with least subjectivity on various retrieval related parameters. These features are equally
desirable for the operational implementation. Results of simulation studies related to retrieval, quality control and
rain flagging along with its implementation to limited Quikscat data are presented.