Resources supporting the food web of a naturally productive lake

A Bayesian mixing model and stable isotopes of carbon, nitrogen, and hydrogen were used to evaluate the extent to which six consumers (three fishes, two zooplankton, and a snail) in a naturally productive lake used terrestrial resources, epilimnetic and metalimnetic phytoplankton, benthic algae, and macrophytes. Resource use varied with consumer habitat use and feeding ability, but allochthony was consistently low (averaging 15% among consumers). The pelagic invertebrates Skistodiaptomus oregonensis and Chaoborus spp. relied on phytoplankton from the epilimnion (59% and 49%, respectively) and to a lesser extent from the metalimnion (28% and 26%, respectively); terrestrial resources comprised 9% and 18% of the diet of these consumers, respectively. The snail Helisoma trivolvis relied mainly on littoral resources (floating-leafed macrophytes; 68% of diet), but terrestrial resources also constituted a substantial portion of its diet (21%). The fishes integrated among habitats more evenly than the other consumers, but pelagic resources formed the largest portion of their diets (Pimephales promelas 5 64%, Lepomis gibbosus 5 47%, and Perca flavescens 5 47%). L. gibbosus was the fish with the most allochthonous diet (23%). The consumers of this productive lake were not highly dependent on allochthonous materials and tended to rely most heavily on local resources, including macrophytes. The organic resources supporting a consumer often originate from a mix of habitats within an ecosystem or even multiple ecosystems within a landscape. This spatial integration of organic resources by consumers is a longrecognized feature common to many food webs (Lindeman 1942; Polis et al. 1997) and influences ecosystem stability, dynamics, and metabolic balance (Cole et al. 1994; McCann and Rooney 2009). The importance of crosshabitat subsidies to a food web is influenced both by the relative availability of resources and by the extent to which its consumers use these subsidies (Marcarelli et al. 2011), making both availability and use relevant to ecosystem functioning. The structure of resource subsidy and consumption is not constant among ecosystems and is particularly interesting in lakes because their position in the surrounding landscape can cause them to receive large subsidies of terrestrial carbon (Kratz et al. 1997). Several studies have experimentally manipulated the d13C of aquatic primary producers and found that the use of terrestrially derived (allochthonous) resources by lake consumers can be substantial (Pace et al. 2004, 2007; Carpenter et al. 2005). However, the extent to which allochthonous resources are used by consumers changes with the types of resources available (Caraco et al. 2010; Marcarelli et al. 2011), the physiology and natural history of the consumers (Jansson et al. 2007), and qualities of the lake ecosystem and its constituent habitats (Francis et al. 2011). As a result, the importance of allochthonous resource use (hereafter allochthony) is contingent on other resource use, which can vary widely both among lakes and among consumers within a given food web. The use of terrestrial carbon by consumers is known to be variable among aquatic systems, and past studies have suggested that the magnitude of autotrophic production can explain a portion of this variability in both lotic (Huryn 1996; Nakano and Murakami 2001) and lentic (Cole et al. 2002; Maguire and Grey 2006) systems. However, few studies have examined allochthony in the food webs of productive lakes (Cole in press). An important early effort by Bunn and Boon (1993) used ambient levels of the stable isotopes of carbon and nitrogen to examine the use of terrestrial carbon, phytoplankton, macrophytes, and epiphytes by consumers in several eutrophic billabongs. Although they found that many of the consumers in these productive systems relied primarily on aquatic resources, isotopic analysis was not able to resolve resource use for all consumers, in part because of isotopically indistinct end members. Other studies of productive lentic systems have either used isotopic labeling experiments or natural abundances of 2H to circumvent the issue of isotopically indistinct aquatic and terrestrial resources. Cole et al. (2002) and Carpenter et al. (2005) enriched aquatic primary producers with 13C in two experimentally fertilized lakes and found that the zooplankton in these systems were highly autochthonous. Carpenter et al. (2005) also found that benthic consumers and fish relied heavily on algal resources. However, the fishes in this study had slow rates of biomass turnover, and because the nutrient fertilization lasted only one season, their long-term allochthony is unclear (Carpenter et al. 2005). Moreover, Babler et al. (2011) used ambient 2H and found that gizzard shad (a detritivorous fish) exhibited a wide range of allochthony among several eutrophic and hypereutrophic impoundments. Together, these studies suggest that while fishes in eutrophic lakes rely * Corresponding author: [email protected] Limnol. Oceanogr., 57(5), 2012, 1443–1452 E 2012, by the Association for the Sciences of Limnology and Oceanography, Inc. doi:10.4319/lo.2012.57.5.1443