A Multistage Approach to Estimate Fish Abundance in Streams Using Geographic Information Systems

We present a multistage approach for estimating the abundance of stream fishes using geographic information systems (GIS) technology. Stream habitat types (i.e., channel units), classified in terms of their quality (optimal, suitable, or unsuitable) for a fish species, were used to stratify the stream for population sampling. Population abundance estimates and habitat quality were then used with the GIS to predict abundances in areas that were not sampled and to calculate an abundance estimate for the entire stream or stream segment. We present the basic method and illustrate its use with a case study of the leopard darter Percina pantherina from Big Eagle Creek in southeastern Oklahoma. Our method is flexible, accounts for variation in habitat quality and its influence on fish abundance, and incorporates GIS technology as an aid for quantitatively sampling large areas that are logistically difficult to sample.