G2’s Research



Breaking news — here you can watch my testimony to the US House of Representatives Committee on Natural Resources in the hearing on “The Costly Impacts of Predation and Conflicting Federal Statutes on Native and Endangered Fish Species

Our laboratory conducts all three major types of ecological research: descriptive field studies, field and laboratory experiments and modeling. Regardless of the approach, all research in the lab is “question-driven” (i.e., we test specific hypotheses even in descriptive studies). In general, we conduct long-term field studies to understand the dynamics of natural systems and to provide information for the scientific management of these systems. Typically, field studies suggest potential mechanisms for observed processes that are tested experimentally, both in the field and laboratory. At heart, I am an experimentalist, because I believe that experiments yield the most rigorous insights into ecological phenomena. Finally, we attempt to build predictive models for the systems that we study. I am not a big math jock so most of these studies are done in collaboration with modelers or with my graduate students. Conceptually, the lab’s focus is linking individual behavior (e.g., habitat selection) to population-level phenomena using fitness correlates such as energy intake/individual growth. A second focus involves the relationship between interspecific-interactions, environmental variation, and patterns of assemblage structure. Finally, we are also interested in the effects of invasive species on native fishes.

ecent and Current Research Projects

Effects of trout stocking on native non-game fishes in Georgia

Trout have been stocked in streams for sport fishing for decades. Nonetheless, little is known about the impact of this practice on native fishes, especially in North America. We are utilizing a combination of field and lab experiments to determine whether stocked rainbow trout (Oncorhynchus mykiss) affect microhabitat use, foraging success or density of native non-game fishes in North Georgia streams. Our design for the field experiment involves introducing trout using a BACI approach, and we hope to replicate this experiment in two different drainages (Little Tennessee and Etowah). The objective of the laboratory experiment is to evaluate the effects of rainbow trout on microhabitat use, foraging success and social interactions of a common native fish, warpaint shiner (Luxilus coccogenis). We are conducting this experiment in an artificial stream in our laboratory at UGA. Funding for this project comes from the Georgia DNR and Warnell School of Forest Resources.

Effects of turbidity on foraging success and optimal habitat use by native stream fishes

Perhaps the biggest problems affecting aquatic ecosystems today is increasing sediment loads. In these projects, we are examining the effects of turbidity on: reactive distance, capture success and ultimately optimal habitat use by several water-column fishes (rosyside dace Clinostomus funduloides), and (yellowfin shiner, Notropis lutipinnis). . We would like to use these data to build a model that will predict population-level effects of turbidity changes on these species. Experiments were conducted in our artificial stream systems at UGA  Funding for this project came from NSF-LTER program, EPA and the Warnell School of Forest Resources.An example of a river chub foraging in clear water, one prey is released every minute, four captures are shown. Here, a river chub is foraging in water with low turbidity, notice the difference in position of the fish (it holds at the front of the tank) and how prey are captured (the fish turns and captures prey that have passed, rather than capturing them from straight ahead).


Predicting optimal habitat use in stream fishes

We have developed two general models for predicting habitat use in stream fishes both based on energy maximization principles. The model for water-column fishes uses prey-capture success data to predict focal-point velocities that maximize energy intake for water-column feeders (see Grossman et al. 2002. A new optimal foraging model predicts habitat use by drift-feeding stream minnows. Ecol. Freshwat. Fish 11:2-10). Similarly, our model for benthic feeders predicts that the patchy distribution of prey will be the dominant factor influencing patch-choice in these species (see Petty, T. and G. D. Grossman. 1996. Patch selection by mottled sculpin (Pisces: Cottidae) in a southern Appalachian stream. Freshwat. Biol. 35: 261-276. Thompson,et al. 2001. Multi-scale effects of resource patchiness on foraging behaviour and habitat use by longnose dace, Rhinichthys cataractae. Freshwat. Biol.46: 145-161. Petty, T. and G. D. Grossman. 2004. Restricted movement by mottled sculpin (Pisces: cottidae) in a southern Appalachian stream. Freshwat. Biol. 49:631-645). We are continuing tests of these models using new species resident to the southern Appalachian streams. Funding for this project comes from USDA Forest Service, NSF-LTER program and the Warnell School of Forest Resources. Our most recent and exciting work involves building a cognitive-based model of optimal microhabitat selection for three Alaskan fishes Chinook salmon (Oncorhynchus tshawytscha), Dolly Varden charr (Salvelinus malma) and Arctic Grayling (Thymallus arcticus) and this project has its own web page http://www.driftmodelproject.org/


Assemblage organization and population regulation of stream fishes along a longitudinal gradient in the Coweeta Creek drainage, N.C.

We have been conducting long-term fish population censuses in three 100m sites for the past 13 years (3-4X mean generation time of most species) and collecting environmental data from the same sites. During this period, the drainage has experienced extensive environmental variation including high water years and a four-year drought. We will examine assemblage and population-level dynamics of fishes in this system and compare them to variations in flow , depth, substratum composition, etc., to test hypotheses regarding the roles of environmental variation and inter- and intraspecific biotic interactions (see Grossman et al. 1998. Ecol. Monogr. 68:395-420, and 2006 Ecol. Monogr. 76:217-234) on both assemblage and population dynamics . We are using information theoretic and artificial intelligence approaches to test these hypotheses. Funding for this project comes from NSF-LTER program, USDA Forest Service and the Warnell School of Forest Resources.