A predictive model for water clarity following dreissenid invasion
Optical transparency, or water clarity, is a fundamental property of lake ecosystems which influences a wide range of physical, chemical and biological variables and processes. The establishment of non-native dreissenid mussels in lake and river ecosystems across North America and Europe has been associated with often dramatic, but highly variable, increases in water clarity. The objective of this study was to develop a predictive model for water clarity (Secchi depth, m) in lakes following the establishment of dreissenids. We compiled water clarity data before and after dreissenid invasion from North American lakes that varied in size and nutrient status. An AIC model averaging approach was used to generate post-invasion water clarity predictions based on pre-invasion water clarity and lake morphometric characteristics from a 53 lake dataset. The accuracy of the model was verified using cross-validation. We then extended this model to existing empirical models of lake mixing depth and Walleye (Sander vitreus) yield, to demonstrate that increased water clarity associated with dreissenid invasion may have far-reaching physical and ecological consequences in lakes, including deeper thermoclines and context-dependent changes in fish yields.
KeywordsDreissena Mussel Transparency Thermal stratification Sander vitreus Ecological modelling
Many individuals and organizations assisted with the provision of data for this project, including the US Environmental Protection Agency, Minnesota Pollution Control Agency, Michigan and Wisconsin Departments of Natural Resources, J. Young and H. Jarjanazi (OMOECC), S. Sandstrom and M. (OMNRF), J. Halfman (Hobart and William Smith Colleges), L. Rudstam and R. Jackson (Cornell University). Financial support was provided by Fisheries and Oceans Canada to M.D.R. and the Experimental Lakes Area Graduate Fellowship and Major G.E.H. Barrett-Hamilton Memorial Scholarship to M.E.G. We acknowledge the input from H. MacIsaac, D. Strayer and two anonymous reviewers in helping prepare this manuscript for publication.
- Akaike H (1973) Information theory as an extension of the maximum likelihood principle. In: Petrov BN, Ckaski F (eds) Proceedings of the second international symposium on information theory. Akademiai Kiado, pp 267–281Google Scholar
- Burnham KP, Anderson DR (2002) Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. Springer, BerlinGoogle Scholar
- Fisheries and Oceans Canada (2010) 2010 Survey of recreational fishing in Canada [online]. http://www.dfo-mpo.gc.ca/stats/rec/canada-rec-eng.htm. Accessed 10 Nov 2014
- Kinnunen RW (2003) Great lakes commercial fisheries. Michigan Sea Grant Extension, MarquetteGoogle Scholar
- Phosphorus Management Strategies Task Force (1980) Phosphorus management for the Great Lakes. Final report to the great lakes water quality board and the great lakes science advisory board. International Joint Commission, WindsorGoogle Scholar
- R Development Core Team (2012) R: a language and environment for statistical computing. http://www.R-project.org. Accessed 26 Oct 2012
- Schindler DW, Bayley SE, Parker BR, Beaty KG, Cruikshank DR, Fee EJ, Schindler EU, Stainton MP (1996) The effects of climatic warming on the properties of boreal lakes and streams at the Experimental Lakes Area, northwestern Ontario. Limnol Oceanogr 41(5):1004–1017. doi: 10.4319/lo.1918.104.22.1684 CrossRefGoogle Scholar
- Sokal RR, Rohlf FJ (1995) Biometry: the principles and practices of statistics in biological research, 3rd edn. W.H. Freeman and Company, New YorkGoogle Scholar
- Wetzel RG (2001) Limnology: lake and river ecosystems, 3rd edn. Academic Press, CambridgeGoogle Scholar
- Zuur A, Hilbe JM, Ieno EN (2013) A beginner’s guide to GLM and GLMM in R: a frequentist and Bayesian perspective for ecologists. Highland Statistics Ltd., NewburghGoogle Scholar