The Publications Warehouse does not have links to digital versions of this publication at this time
Data from several sources were collated and analyzed by correlation, regression, and principal components analysis to define surrrogate variables for use in the brook trout (Salvelinus fontinalis) habitat suitability index (HSI) model, and to evaluate the applicability of the model for assessing habitat in high elevation streams of the southern Blue Ridge Province (SBRP). In all data sets examined, pH and alkalinity were highly correlated, and both declined with increasing elevation; however, the magnitude of the decline varied with underlying rock formations and other factors, thereby restricting the utility of elevation as a surrogate for pH. In the data sets that contained biological information, brook trout abundance (as biomass, density, or both) tended to increase with elevation and decrease with the abundance of rainbow trout (Oncorhynchus mykiss), and was not significantly correlated (P >0.05) with the abundance of most benthic macroinvertebrate taxa normally construed as important in the diet of brook trout. Using multiple linear regression, the authors formulated an alternative HSI model A? based on point estimates of gradient, pH, elevation, stream width, and rainbow trout density A? which explained 40 to 50 percent of the variance in brook trout density in 256 stream reaches. Although logically developed, the present U.S. Fish and Wildlife Service HSI model, proposed in 1982, seems deficient in several areas, especially when applied to SBRP streams. The authors recommend that the water quality component in the model be updated and reevaluated, focusing on the differential sensitivities of each life stage, the stochastic nature of the water quality variables, and the possible existence of habitat requirements that differ among brook trout strains.
Additional Publication Details
Federal Government Series
Habitat suitability index model for brook trout in streams of the Southern Blue Ridge Province: Surrogate variables, model evaluation, and suggested improvements