Determining Dynamic Biogeographic Regions using Remote Sensing Data

One of the most important topics in natural resource management is the determination of biogeographic regions (BGR). Each BGR is the result of a particular combination of physical, chemical and biological factors and the variation of one of these factors can have different effects on each region. This characteristic trait makes biogeographic regionalization a basic tool of environmental modeling. Topics such as the carbon cycle and global climate change, and it effects on fisheries (IOCCG, 2008; 2009a) can be addressed using BGRs to bring together in situ data and data obtained from remote sensor observations (IOCCG, 2009b). Defining a BGR is simpler in terrestrial ecosystems than in marine environments because ecotypes such as forests, jungles and deserts can be used to define a BGR. The highly-dynamic nature of the marine environment makes defining BGRs a great challenge, which requires different approximations. There are two basic approximations that can be used to define marine BGR. The first uses a large, in situ database (direct approximation) and the second uses data obtained from remote sensor observations (indirect approximation). Applying the direct approximation requires a database that is robust in both space and time. Such a database is generated by monitoring programs such as CalCOFI (California Cooperative Oceanic Fisheries Investigations) within the California Current systems. Millán-Núñez et al. (1997) surveyed the area using the CalCOFI database and determined six BGR. Not all areas are candidates for direct approximation, because of the high costs. For areas that do not have continuous monitoring stations, the only way to determine BGR is to use indirect approximation. The basic idea of this technique is to use satellite images as virtual maps to generate a database of the surface layer of the oceans. A BGR can be determined by the association patterns