A satellite remote sensing case study of the hydrological cycle and oceanic response in the Bay of Bengal

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Meteorology and Physical Oceanography

First Committee Member

Peter J. Minnett - Committee Chair


Aspects of the hydrological cycle over the Bay of Bengal, the Andaman Sea, and their respective catchment areas are analyzed with a focus on seasonal and inter-annual variability. Taking an Earth System Sciences approach, this study examines the coupled terrestrial, oceanographic, and atmospheric processes involved in the region using various satellite remote sensing data sets. The Bay of Bengal was selected due to its unique combination of forcing mechanisms at work: (1) low latitude - high insolation regime, (2) monsoonal reversal of winds and currents, (3) immense quantities of freshwater input from river runoff and precipitation leading to strong surface stratification in the ocean, (4) occasional tropical cyclones and low pressure systems, and (5) equatorial oceanic forcing. The performance of two satellite-derived precipitation products were compared to weather station observations for 2002 and 2003 and evaluated for their potential as input for hydrological land surface models. Despite certain limitations these products reproduced well the monsoonal progression of rainfall and the natural variability of daily rainfall accumulation. They were found to be quite adequate for large, continental scale watershed modeling. River discharge estimates were generated for 2001 and 2002 using NASA's Land Information System, a University of Washington river routing model and a University of New Hampshire artificial river network. The routed model output performed well against measured observations for the Ganges/Brahmaputra combined river basin, but underestimated peak discharge periods at the height of the summer monsoon. Results for the other major river basins compared favorably with the available, but limited climatology. The oceanic response to the large riverine flux was examined using SeaWiFS ocean color imagery. A time series of bio-optical properties such as chlorophyll concentration, absorption by colored dissolved organic material, and backscatter from river sediments tracked the zone of the river influence into the central Bay. High, positive correlations of these properties with river discharge were found to be limited to the east coast of India down to the mouth of the Godavari and Krishna rivers and in the Gulf of Marataban off the mouth of the Irrawaddy and Salween rivers. Scatter plots and imagery enhancement techniques were employed to classify specific bio-optical provinces in terms of riverine, coastal shelf, and open ocean water masses. It was concluded that the spatial and temporal distribution of the inorganic component can be used to trace river plumes and fronts while the distribution of the organic component does not necessarily mirror the inorganic component as they are influenced by different processes. Within this Earth System Science framework, several avenues are available for further study.


Hydrology; Physical Oceanography; Physics, Atmospheric Science; Remote Sensing

Link to Full Text


Link to Full Text