A Large-Scale MaxEnt Model for the Distribution of the Endangered Pygmy Madtom

Abstract

Aquatic resource managers often need detailed knowledge of the distributional patterns of imperiled species to facilitate conservation and recovery actions. The Pygmy Madtom Noturus stanauli is a rare catfish in family Ictaluridae that is federally listed as endangered under the U.S. Endangered Species Act. To better understand and forecast its distributional patterns, we assembled Pygmy Madtom occurrence records from past collections in the Clinch and Duck rivers in Tennessee, the only two waterways known to support populations of this species. We entered these presence-only coordinates into the maximum entropy species distribution model integrated with layers from geographic information systems. This approach produced suitability score maps and response curves for each environmental variable: flow rate, water velocity, stream order, gradient, air temperature, precipitation, canopy cover, and drainage area. The variables flow rate, drainage area, and canopy cover were important in predicting the distribution of the Pygmy Madtom throughout its range. The maximum entropy model predicted a high suitability score of Pygmy Madtom occurrence at new sites throughout the Upper Clinch River and the lower middle reaches of the Duck River. Our analytical approach gives managers a large-scale tool to better delineate the Pygmy Madtom’s distributional range by identifying and prioritizing locations in the field for sampling at a later date to verify species presence and absence.