Population density and community size structure: Comparison of aquatic and terrestrial systems

Hélène Cyr, Robert H. Peters, John A. Downing

Research output: Contribution to journalArticlepeer-review

81 Scopus citations


The size structure of aquatic communities is generally measured using size spectra, an approach which is tedious or inapplicable in benthic and terrestrial communities. This has inhibited comparison of size structure of aquatic and terrestrial communities. This study uses an approach more common among terrestrial ecologists to develop a general density-body size relationship for lacustrine communities, based on mean annual population densities for dominant species of phytoplankton, zooplankton, zoobenthos and fish measured in 18 lakes worldwide. Overall, mean annual population density (D, individuals m-2) decreases log-linearly with increasing species body size (M, μg fresh mass) as D = 4 x 105 · M-0 .89 (n = 280, r2 = 0.92), although the exponent appeared smaller (-0.55 ± 0.04) within broad taxonomic groups (algae, invertebrates). We found that density-body size relationships for dominant species are quantitatively similar to size spectra, a pattern which suggests that density-body size relationships may provide an interesting alternative to size spectra for the prediction of ecosystem processes. These relationships also suggest that aquatic species reach, on average, 6-60 times higher densities than terrestrial species, depending on their body size and on their thermoregulatory system (ectotherms vs endotherms). The implications of these differences in size structure for size-related patterns of energy use and other processes depend on which physiological groups (unicells, ectotherms, endotherms) are being compared.

Original languageEnglish (US)
Pages (from-to)139-149
Number of pages11
Issue number1
StatePublished - Oct 1997
Externally publishedYes


Dive into the research topics of 'Population density and community size structure: Comparison of aquatic and terrestrial systems'. Together they form a unique fingerprint.

Cite this