THE SOUTHWESTERN NATURALIST 30(2):225-238 MAY 31, 1985 EVALUATION OF SPATIAL PATTERNING IN OKLAHOMA FURBEARER POPULATIONS GARY D. SCHNELL, JOSEPH A. GRZYBOWSKI, DANIEL J. HOUGH AND THOMAS M. MCKENNA ABSTRACT.-TO evaluate furbearer populations and harvest in the state, the Oklahoma Department of Wildlife Conservation compiles data for each of the 77 counties from scent station surveys and pelt purchases. Similar kinds of data are collected in other geographic regions, and appropriate statistical techniques would be useful for analyzing these types of information. Mantel test values and matrix correlations were used as statistical measures to determine the extent of geographic patterning in these data. Two types of patterning-regional and local-were differentiated and operationally defined. Mantel tests and matrix correlations were also used to compare patterns of abundance (as reflected by scent station visits or pelts purchased) between years, between species, and for the same species but based on different data sets. For scent post visits, regional and local patterning were found for coyotes (Canis latrans) in 1978, gray foxes (Urocyon cinereoargenteus) in 1979, and combined data for gray and red foxes (Vulpes vulpes) in both years. Close concordance occurred between years (1978 and 1979) for coyotes, foxes (combined), bobcats (Felis rufus), and domestic dogs. In 1978, positive correspondences between distributions of scent post visits across the state were identified for four pairs of species: raccoons (Procyon lotor) and bobcats; coyotes and raccoons; coyotes and bobcats; and raccoons and dogs. In 1979, with more species being evaluated, 10 such comparisons of species pairs were statistically significant. Marked associations were between: dogs and cats; Virginia opossums (Didelphis virginianus) and striped skunks (Mephitis mephitis); and red foxes and raccoons. The pelt harvest data showed both local and regional patterning for most species. The only exceptions were: striped skunks for both years and both types of patterning; and coyotes in 1979 for regional patterning. Only for the gray fox was there a substantive association of scent post and harvest figures indicating, not surprisingly, that the two types of data reflect rather different aspects of furbearer populations. Furbearer populations vary geographically in numbers, and geographic variation is evident in the harvesting of these animals by sportsmen. Effective management of furbearer resources should be based on the population status of each species. Sound analytical procedures are needed to evaluate furbearer populations and harvest, both of which can vary in time and space. These procedures need to be robust and designed to deal with what may be relatively inexact data; they should also provide statistical confidence for statements concerning spatial and temporal population variation. While several types of relevant information are gathered by various agencies concerning furbearer populations over large geographic areas, these data have seldom been subjected to statistical evaluations for spatial patterning. Harvest regulations in many states are typically defined by regions; often counties are grouped by managers using biogeographic criteria. In practice, once established these regions are seldom redefined. Areas of transition between ecotypic regions may be difficult to assign, and land-use practices within ecotypic regions can modify conditions substantially. Wildlife populations might vary for a number of reasons within biogeographically This content downloaded from 129.15.14.53 on Thu, 07 May 2015 02:26:25 UTC All use subject to JSTOR Terms and Conditions