More on this visualization

 Scientists use satellites to map land cover to understand how humans alter ecosystems. When sunlight hits the Earth's surface, some of the light bounces back into space. Different land surfaces, such as forests, bare ground, roads, and water, reflect different wavelengths of the sunlight differently. Satellites can measure the reflected radiation to image distinct land features.

 A new computer technique called "object-oriented analysis" is making it easier to identify details in satellite images of river channels, such as sandbars, tributaries, shallow water, and whitewater. Previously, researchers either mapped water features directly via fieldwork, or interpreted differences in reflected radiation in satellite images by eye. Now, computer programs can automatically classify nuances of radiation to define the shape and size of objects, differentiate them from adjacent objects, and analyze their context in the larger landscape. Using this technology, ecologists can characterize the habitat types that these objects denote and map habitats precisely. Automated analysis of satellite data can provide information about large river basins much more quickly, cheaply, and consistently than field surveys or manual interpretation.

 AMNH ichthyologists Melanie Stiassny and Bob Schelly are examining the distribution of closely related fish species in the lower Congo River in west-central Africa. They hypothesize that physical barriers in the river may have led to the evolution of new species. Stiassny and Schelly have noted that several fish species in the genus Lamprologus occupy distinct portions of the river. Scientists at the American Museum of Natural History and the University of Georgia are using object-oriented analysis to map the river channel via satellite to discover if regions of whitewater have acted as a barrier to fish movement over tens of thousands of years, thus allowing species to diverge. If such barriers coincide with the ranges of fish species, it will show that the mapping techniques featured in this visualization can be used to understand fish evolution, habitat conservation, and other ecological questions in river basins worldwide.