Control of zooplankton dependence on allochthonous organic carbon in humic and clear-water lakes in northern Sweden

We compared the stable carbon isotopic composition (d13C) of crustacean zooplankton with that of potential carbon sources in 15 lakes in northern Sweden with different dissolved organic carbon (DOC) concentrations (2–9 mg L21) to test the hypothesis that zooplankton depended more on allochthonous carbon in humic lakes than in clear-water lakes. Based on d13C signature, we found that the pool of organic matter in the lakes was dominated by carbon of allochthonous origin over the whole DOC gradient. Zooplankton were generally depleted in 13C compared to organic matter in the catchment, particulate organic matter in the lake water, and shallow surface sediment. However, the isotopic composition of zooplankton could not be explained without a significant contribution from both allochthonous and autochthonous carbon sources in all lakes. The relative importance of these two carbon sources did not relate to the concentration of, or proportion between, allochthonous and autochthonous organic carbon in the water. Instead, the proportion between allochthonous and autochthonous carbon in the crustacean zooplankton was consistent with a rather conservative use of the energy mobilized by bacterioplankton and phytoplankton in the lakes. Both allochthonous and autochthonous carbon sources can support secondary production in lakes (Hessen and Tranvik 1998). Photosynthesis is the dominant energy mobilization process for secondary production in many lakes, where organic carbon fixed by primary producers is consumed directly by grazing or recycled via the microbial loop (Wetzel 2001). Zooplankton thus utilize autochthonous carbon by grazing on phytoplankton or on bacteria and other organisms depending on phytoplankton production. However, bacteria can also utilize allochthonous organic matter as a source of energy (De Haan 1974, 1977; Tranvik 1988). Furthermore, allochthonous organic matter can enter the food web directly by zooplankton grazing of detrital particles (Hessen et al. 1990). Consequently, allochthonous sources may contribute a substantial portion of the zooplankton carbon in the lakes that have low primary production (Salonen and Hammar 1986; Meili et al. 1993, 1996, 2000; Grey et al. 2001). The relative importance of allochthonous and autochthonous carbon for pelagic biota can vary among lakes. Jones (1992) suggested that allochthonous organic carbon should be relatively more important in supporting planktonic food webs in humic lakes compared with clear-water lakes. A similar difference was shown by Jansson et al. (2000) in a Acknowledgments Thanks to Ludmilla Janeck and Thomas Westin for their assistance during the field work and to Bo Edlen, Sonja Lojen, and Håkan Wallmark for the stable isotopic analyses. EU structural funds and funding from regional and national governments financially supported this study. compilation of data from temperate unproductive lakes, where an increasing portion of the pelagic energy mobilization was covered by bacterial utilization of allochthonous organic carbon as the dissolved organic carbon (DOC) concentration of the lakes increased. In lakes with food webs fuelled by energy mobilized from allochthonous carbon (i.e., humic lakes), a large part of the zooplankton production could therefore depend on allochthonous carbon (e.g., Meili et al. 1993, 1996, 2000). No quantitative cross-lake studies have, however, been made to clarify the relative importance of allochthonous carbon versus autochthonous carbon utilization by zooplankton in humic lakes compared with clearwater lakes. To test the hypothesis that allochthonous carbon is more important as a base for zooplankton production in humic lakes as compared with clear-water lakes, we analyzed the stable carbon isotopic composition of crustacean zooplankton and their potential carbon sources in 15 subarctic lakes in northern Sweden with different concentrations of allochthonous DOC. Materials and methods Investigated lakes—Fifteen lakes in subarctic northern Sweden (688N, 188E) were included in the study. The lakes are situated along an altitudinal gradient (270 to 1,140 m a.s.l.) from the coniferous forest to the alpine belts (Table 1) in the Scandinavian mountains. For a more detailed description of the lake catchments, see Karlsson et al. (2001). Ear-