IPredicting the Spatial Distribution of Dreissena polymorpha (Zebra Mussel) among Inland Lakes of Wisconsin: Modeling with a GIS

Previously developed models and avai lable l imnological data were used to predict ( i) absence or presence, ( i i) categorical populat ion density, and ( i i i) numerical abundance of Dreissena polymorpha (zebra mussel) for 194 in land Wiscons in lakes. A geographica l in format ion system (GlS) was used to tes t for assoc ia-t ions between predicted lake populat ion density classes and three landscape-scale characterist ics (surf icial deposits, bedrock type, U.S. Environmental Protection Agency developed ecoregions) that may affect l im-no log ica l parameters. A l though the models used d i f fered in the i r pred ic t ions o f spec i f ic lakes that would suppor t Dre issena, the overa l l spat ia l d is t r ibut ion o f lakes wi th in a g iven dens i ty c lass was s imi lar among the models. A signif icant associat ion was found between each landscape-scale characterist ic and Dreissena density classes. This study suggests that (1) avai lable lake monitoring data can be used to predict Dreissena dens i ty for groups of in land lakes, (2) more in format ion on Nor th Amer ican lakes wi th zebra mussel is requi red to reduce the uncer ta in ty o f the models, and (3) spat ia l ana lys is us ing CIS methods can prov ide valuable ins ight in to the overa l l pat terns o f the potent ia l spat ia l d is t r ibut ion o f Dre issena