Stable isotope monitoring of benthic–planktonic coupling using salt marsh fish

Abstract

Salt marshes are important coastal ecosystems whose trophic function can be monitored with stable isotopes of abundant fish biosentinel species such as the mummichog Fundulus heteroclitus and the Atlantic silverside Menidia menidia. We compared movement patterns and feeding biology of these species in the summers of 1999 and 2000 in the Rowley River salt marsh estuary north of Boston, Massachusetts, USA. A 15N tracer addition experiment showed that fish of both species were more resident than transient, with mummichogs resident at scales of 1 km or less. Natural abundance stable isotope C, N, and S distributions showed that mummichogs feed more strongly in the benthic food web while silversides feed more in the planktonic food web, with % benthic feeding respectively averaging 58 ± 5 and 32 ± 3% (mean ± 95% confidence limit, CL). For both species, isotope results indicated considerable individual specialization in foraging behavior, likely related to use of channel habitat versus use of the marsh. Power analysis showed that measuring 3 composite samples each comprising 10 to 15 individual fish should provide relatively low errors of 0.5‰ (95% CL) or less around stable isotope averages. Use of such composite samples in monitoring programs will allow detection of significant temporal and spatial changes in benthic-planktonic coupling for salt marsh ecosystems, as recorded in average fish diets. Analyzing some individual fish also is recommended to obtain more detailed information on fish food sources, feeding specializations, and end-member isotope values used in estimating importance of benthic and planktonic food sources.