Beyond nitrogen: phosphorus – estimating the minimum niche dimensionality for resource competition between phytoplankton. (16th February 2021)

• Record Type: Journal Article
• Authors: Hofmann, Peter; Clark, Adam; Hoffmann, Petra; Chatzinotas, Antonis; Harpole, W. Stanley; Dunker, Susanne
• Editors: Jeyasingh, Punidan
• Is Part Of: Ecology letters. Volume 24:Number 4(2021)
• Page Start: 761
• Page End: 771
• Publication Date: 2021-02-16
• Subjects: Cyanobacteria -- green algae -- Hutchinson -- macronutrients -- microcosm -- micronutrients -- N:P -- phytoplankton -- Redfield -- trace elements; Ecology -- Periodicals; 577
• DOI: 10.1111/ele.13695
• Languages: English
• ISSNs: 1461-023X

Abstract

Abstract: The niche dimensionality required for coexistence is often discussed in terms of the number of limiting resources. N and P limitation are benchmarks for studying phytoplankton interactions. However, it is generally agreed that limitation by small numbers of resources cannot explain the high phytoplankton diversity observed in nature. Here, we parameterised resource competition models using experimental data for six phytoplankton species grown in monoculture with nine potential limiting resources. We tested predicted species biomass from these models against observations in two‐species experimental mixtures. Uptake rates were similar across species, following the classic Redfield ratio. Model accuracy levelled out at around three to five resources suggesting the minimum dimensionality of this system. The models included the resources Fe, Mg, Na and S. Models including only N and P always performed poorly. These results suggest that high‐dimensional information about resource limitation despite stoichiometric constraints may be needed to accurately predict community assembly. Abstract : We tested predicted species biomass from monocultures against observations in two‐species mixtures. Although uptake rates were similar across species, following the classic Redfield ratio, we found that model accuracy was maximised around three to five resources, including resources such as Fe, Mg, Na and S, and that models that included only N and P always performed poorly. Thus high‐dimensional information about resource limitation may be needed to accurately predict outcomes of community assembly.