Scientists have proposed a new model to detect rainfall over localized land regions in real time using satellite images. This data will help improve short term rainfall prediction.
Dr. J. Indu and Prof. D. Nagesh Kumar of the Department of Civil Engineering at Indian Institute of Science, Bangalore have developed a model for detecting rainfall from satellite microwave sensor data. Their work was recently published in the Hydrological Sciences Journal.
Images taken by meteorological satellites can be used to detect the amount of rainfall in a region. “An initial step towards calculating rainfall from the microwave data involves the detection of rainfall signature”, explains Dr. Indu. Analytical models and computational simulations are applied on the data to detect rainfall signatures.
Detection of rainfall over land regions is tricky. Land gets heated up by the sun and presents a warm background to the microwave sensors. This warm background obscures the detection of raindrops using microwave sensors. Distinguishing the presence of this weak "scattering" of ice-bearing rain clouds against the land surface poses many uncertainties in detection. Furthermore, land-form variations introduce greater uncertainties.
The team carried out their research in the Mahanadi basin area prone to large-scale flooding each year. “The basin has a typical monsoon climate with many types of land cover such as forest, cropland, wetlands and urban and industrial settlements”, adds Prof. Nagesh Kumar. He feels that the observations in such a diverse region can be used to improve the existing algorithms to detect rainfall from microwave satellite images, and provide an estimate of the uncertainty in rainfall detection.
The team used historical Microwave Imager and Precipitation Radar (PR) data (2011 to2012)in the Mahanadi basin obtained from the space based Tropical Rainfall Measuring Mission (TRMM). TRMM is a joint mission between NASA and the Japan Aerospace Exploration (JAXA) Agency to study rainfall for weather and climate research. Algorithms were run on this data based on the new model for detecting rainfall. These were compared with ground based rainfall data from the Indian Meterological Department (IMD). The model developed by the team was able to detect rainfall using satellite images to an accuracy of 95%.
The team anticipates that this work will help in a big way to improve monsoon prediction in India. The recent launch of the space based Global Precipitation Mission (GPM), which will provide global observations of rain and snow, is expected to contribute towards the development of accurate models for detection and prediction of rainfall. “This method will be useful in quantifying uncertainty of similar data that becomes available from GPM", signs off Prof. Kumar.
About the authors
Dr. J. Indu is a PhD graduate from the Department of Civil Engineering, Indian Institute of Science, Bangalore and at present she is working as Assistant Professor in Dept. of Civil Engineering, Indian Institute of Technology Bombay, Mumbai.
Prof. D. Nagesh Kumar is the Chairman of the Centre for Earth Sciences (CEaS), and Professor in Dept of Civil Engg., Indian Institute of Science, Bangalore.firstname.lastname@example.org
About the paper
Title: Rainfall screening methodology using TRMM data over a river basin
Publication: Hydrological Sciences Journal