Recent advances in remote sensing have widened the platform for research in science and technology. Undoubtedly the estimation of geo-bio-physical properties of the land cover features like water, urban, vegetation, rocky and barren areas play an important role in environmental, transportation and region planning, natural disaster, industrial and agricultural production. Since the water transport is cheapest, extraction of the water body in hyper spectral images of remote areas for which we don’t have enough details of its terrain is inevitable. Till now, natural computation and bio-inspired intelligent techniques like DNA computing, membrane computing, genetic algorithms, neural computing have been used for demonstrating the applications of computational intelligence in the field of remote sensing. However, geo-science has never been used as a nature inspired intelligent technique for developing a computational model. This paper demonstrates the evolution of a new geo-science based approach for satellite image processing using an analogy between plate tectonics and biogeography based optimization. The paper presents biogeography and plate tectonics based optimization (BPBO) as a powerful paradigm for identifying water body area in the satellite image and hence, make a significant contribution towards the development of a new computational intelligence technique in the field of AI. Our major assumption is that it is the entropy which is the driving force leading to the formation of heterogeneous regions called as plates, similar to the convection force in the mantle of the Earth.