Linking planktonic biomass and metabolism to net gas fluxes in northern temperate lakes

Ecology, June, 1999 by Paul A. del Giorgio, Jonathan J. Cole, Nina F. Caraco, Robert H. Peters

INTRODUCTION

Ecologists have traditionally recognized that the biomass structure is one of the fundamental attributes of ecosystems (Odum 1971). It is thought that the shape of the biomass pyramid summarizes not only the structure of communities, but also integrates functional attributes of communities, such as patterns of organic carbon and energy flow, transfer efficiency and turnover of different components of the food web (Odum 197 I, Reichle 1981). Across a large-scale productivity gradient there is a general shift in the shape of the biomass pyramid, from normal pyramids with very broad bases in forests to pyramids with increasingly smaller autotrophic bases in plankton communities (Whittaker and Likens 1973, del Giorgio and Gasol 1995). These broad changes in biomass structure are coherent with systematic changes in the turnover times of the autotrophic base and thus agree with the ecological paradigm that relates the specific rates of production of autotrophs to the partition of total biomass into autotrophic and heterotrophic components (O'Neill and DeAngelis 1981).

Not all planktonic communities are characterized by inverted biomass pyramids. In fact, plankton communities exhibit almost the entire range of relations between biomass compartments that is found among all ecosystems. Recent comparative analyses have shown that there is a systematic decrease in the relative proportion of heterotrophic biomass in the plankton as the autotrophic component of the system becomes more productive. The same qualitative patterns have been observed in both marine communities (Dortch and Packard 1989, Gasol et al. 1997) and in lakes (Cole and Caraco 1993, del Giorgio and Gasol 1995). The plankton of oligotrophic oceans, for example, is often dominated by the biomass of heterotrophs, notably bacteria and colorless protozoans (Cho and Azam 1990), and so is the plankton of unproductive lakes (del Giorgio and Gasol 1995).

Within the context of current ecological theory, in which plankton communities are often viewed as self-supporting systems, the inverted biomass pyramids in oligotrophic plankton communities are assumed to be the consequence of two factors, acting individually or in unison: (1) high specific production of phytoplankton (Odum 1971) and (2) slow turnover of the heterotrophic component (Cho and Azam 1990). A systematic shift from inverted biomass pyramids in oligotrophic areas to normal pyramids in more productive areas must then be the result of (1) a general decline in the specific production of phytoplankton and (2) a general increase in the turnover rate of heterotrophs, along the same productivity gradient. The evidence supporting these two hypotheses is contradictory, however. An inverse relationship between phytoplankton growth rate and production has often been assumed (Harris 1984), but not supported by recent comparative studies which show that algal specific production rates do not tend to be higher in oligotrophic areas (Smith 1979, del Giorgio and Peters 1993). The turnover of heterotrophs, particularly bacteria in oligotrophic marine areas indeed seems to be low (Cho and Azam 1990).

A third, but seldom addressed hypothesis, is that the partition of total biomass into autotrophic and heterotrophic components is influenced by organic carbon imports to the community which serve as substrates for heterotrophic growth (Jones 1992). An inverted biomass pyramid could potentially occur in a community where bacteria and other heterotrophs are consuming both phytoplankton carbon and allochthonous organic matter (del Giorgio and Gasol 1995). The predominance of inverted biomass pyramids in the plankton of oligotrophic areas would suggest, according to this alternative hypothesis, a greater impact of allochthonous organic subsidies on these communities.

Each of these hypotheses suggests a fundamentally different relationship between the biomass distribution and the metabolism of the community. Although the falsification of this simple set of hypotheses is fundamental to our understanding of the function of planktonic communities, the patterns involved have seldom been assessed explicitly in natural systems. In this paper we test these hypotheses for lake plankton communities by integrating the results of a large-scale study of northern temperate lakes. We investigated (1) the relationship between plankton primary production, respiration, and the partitioning of planktonic biomass into autotrophic and heterotrophic components and (2) the relationship between planktonic metabolism, biomass structure, and net fluxes of [O.sub.2] and C[O.sub.2] in surface waters. In addition, we compare our empirical results to simple models constructed on the basis of the predictions of each of the three hypotheses.

MATERIALS AND METHODS

Study sites

This study was a comparative analysis of plankton metabolism and biomass structure in 20 temperate lakes that was carried out in 1992. Portions of the study have been published elsewhere: rates of phytoplankton production and plankton respiration, nutrients and DOC (dissolved organic carbon) concentration appear in del Giorgio and Peters (1994). Here we report the corresponding data on the biomass of autotrophic and heterotrophic planktonic components, and also on whole lake [O.sub.2] and C[O.sub.2] concentrations for these lakes during the same period. The study lakes are located in southern Quebec, Canada (45 [degrees] N, 72 [degrees] W). The morphometry of these lakes is diverse: mean depth = 2.9 to 48 m, lake area = 0.3 to 47 [km.sup.2], and water retention times range from 0.3 to 11 yr. These lakes also vary widely in trophic status, total phosphorus (4.9 to 46 [[micro]gram]/L) and chlorophyll (0.7 to 37 [[micro]gram]/L) (Table 1). Because of large differences in the ratio of basin area to lake area (3.2 to 30.5), humic color varies 13-fold (0.014 to 0.180, absorbance at 440 nm in a 10-cm cell) and dissolved organic carbon (DOC) varies from 2.7 to 8.0 mg C/L. Morphometric and chemical data for the sampling period appear in del Giorgio and Peters (1994).