Stephen B. Brandt
Great Lakes Ctr., Buffalo State College, 1300 Elmwood Ave., Buffalo, NY 14222
Acoustic data are rich in information but models or other data (e.g., sound speed) need to be applied to the raw, time-dependent voltages to extract information on fish sizes, abundances, and distribution. Information extraction is hierarchical in that the more models (or assumptions) used, the more information obtained from the data. Recent research combines spatially explicit acoustic data with other environmental measurements to provide a template for ecological modeling of fish feeding and growth. This modeling links the physical and biological structure of the habitat in a way that maintains the spatial integrity of the system. Data visualization, geographic information systems, and data animation are used to display model output. The approach makes more full use of the spatial information inherent in acoustic data and can be used to predict fish growth and production, predator--prey interactions, the potential success of species introductions, and the effects of environmental perturbations. [Work supported by NSE, Biological Oceanography.]