Compositional variation in grassland plant communities

Authors: BAKKER, JONATHAN - University Of Washington; PRICE, JODI - Charles Sturt University; HENNING, JEREMIAH - University Of Minnesota; BATZER, EVAN - University Of California, Davis; OHLERT, TIMOTHY - University Of New Mexico; WAINWRIGHT, CLAIRE - University Of Washington; ADLER, PETER - Utah State University; ALBERTI, JUAN - Consejo Nacional De Investigaciones Científicas Y Técnicas(CONICET); ARNILLAS, CARLOS - University Of Toronto; BIEDERMAN, LORI - Iowa State University; BORER, ELIZABETH - University Of Minnesota; BRUDVIG, LARS - Michigan State University; BUCKLEY, YVONNE - Trinity College Dublin; BUGALHO, MIGUEL - University Of Lisbon; CADOTTE, MARC - University Of Toronto; CALDEIRA, MARIA - University Of Lisbon; CATFORD, JANE - King'S College; CHEN, QINGQING - Peking University; CRAWLEY, MICK - Imperial College; DALEO, PEDRO - Consejo Nacional De Investigaciones Científicas Y Técnicas(CONICET); DICKMAN, CHRIS - University Of Sydney; DONOHUE, IAN - Trinity College Dublin; DUPRE, MARY - Mpg Ranch; EBELING, ANNE - University Of Jena; EISENHAUER, NICO - University Of Leipzig; Fay, Philip; GRUNER, DANIEL - University Of Maryland; HAIDER, SYLVIA - Martin Luther University; HAUTIER, YANN - Utrecht University; JENTSCH, ANKE - University Of Bayreuth; KIRKMAN, KEVIN - University Of Kwazulu-Natal; KNOPS, JOHANNES - Jiaotong University; LANNES, LUCIOLA - Sao Paulo State University (UNESP); MACDOUGALL, ANDREW - University Of Guelph; MCCULLEY, REBECCA - University Of Kentucky; MITCHELL, RACHEL - Northern Arizona University; MOORE, JOSLIN - Monash University; MORGAN, JOHN - La Trobe University; MORTENSEN, BRENT - Benedictine University Of Illinois; OLDE VENTERINK, HARRY - Vrije Universiteit Brussel; PERI, PABLO - Consejo Nacional De Investigaciones Científicas Y Técnicas(CONICET); POWER, SALLY - Western Sydney University; PROBER, SUZANNE - Commonwealth Scientific And Industrial Research Organisation (CSIRO); ROSCHER, CHRISTIANE - Helmholtz Centre For Environmental Research; SANKARAN, MAHESH - National Centre For Biological Sciences; SEABLOOM, ERIC - University Of Minnesota; SMITH, MELINDA - Colorado State University; STEVENS, CARLY - Lancaster University; SULLIVAN, LAUREN - University Of Missouri; TEDDER, MICHELLE - University Of Kwazulu-Natal; VEEN, G - Netherlands Institute Of Ecology; VIRTANEN, RISTO - University Of Oulu; WARDLE, GLENDA - University Of Sydney

Submitted to: Ecosphere
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/17/2023
Publication Date: 6/8/2023
Citation: Bakker, J.D., Price, J.N., Henning, J.A., Batzer, E.E., Ohlert, T.J., Wainwright, C.E., Adler, P.B., Alberti, J., Arnillas, C.A., Biederman, L.A., Borer, E.T., Brudvig, L.A., Buckley, Y.M., Bugalho, M.N., Cadotte, M.W., Caldeira, M.C., Catford, J.A., Chen, Q., Crawley, M.J., Daleo, P., Dickman, C.R., Donohue, I., Dupre, M.E., Ebeling, A., Eisenhauer, N., Fay, P.A., Gruner, D.S., Haider, S., Hautier, Y., Jentsch, A., Kirkman, K., Knops, J.M., Lannes, L.S., MacDougall, A.S., McCulley, R.L., Mitchell, R.M., Moore, J.L., Morgan, J.W., Mortensen, B., Olde Venterink, H., Peri, P.L., Power, S.A., Prober, S.M., Roscher, C., Sankaran, M., Seabloom, E.W., Smith, M.D., Stevens, C., Sullivan, L.L., Tedder, M., Veen, G.F., Virtanen, R., Wardle, G.M. 2023. Compositional variation in grassland plant communities. Ecosphere. 14(6). Article e4542. https://doi.org/10.1002/ecs2.4542.
DOI: https://doi.org/10.1002/ecs2.4542

Interpretive Summary: Human activities alter the diversity and species composition of plant communities in intentional or unintentional ways, and there is considerable debate about the utility and interpretation of the statistics used to measure the composition of plant communities globally. This study furthers that debate by evaluating different measures of change in plant communities using experimental manipulations of nutrients as a test case. ARS and university collaborators provided data from over 50 grassland sites from around the globe, each conducting nutrient manipulations and sampling plant communities with identical methods. Analysis of these data revealed that different measures of community change were affected by different site-level properties related to community composition and diversity, among them the number of surrounding species, site productivity, and climate and their variation in space and time. The findings highlight how grasslands exhibit substantial variation in composition, and multiple measures of plant composition provide a more comprehensive evaluation of community change than any one measure. Using multiple measures of plant community composition can improve detection of the impacts of human activities.

Technical Abstract: Human activities alter the diversity and composition of ecological communities globally, but a growing body of evidence demonstrates the importance of multiple-site experimental tests of limiting factors to assess generality. In addition, our ability to detect treatment effects will be affected by the spatial and temporal variation at a site. Composition reflects changes in abundance and identity and can be quantified using a variety of metrics; conclusions may therefore be sensitive to how composition is expressed. We used a globally distributed experiment to explore compositional variation within 55 grassland sites spanning 6 continents. Each site had an identical experimental and sampling design, with treatments manipulating consumer pressure and nutrients. We expressed compositional variation within each site using four metrics: an abundance-based measure of the magnitude of dissimilarity (Bray-Curtis), an incidence-based measure of the magnitude of dissimilarity (Sorensen), and abundance- and incidence-based measures of the relative importance of turnover (balanced variation and species turnover, respectively). We quantified each metric per site, related variation to site-level biotic and climatic variables, and partitioned variation related to spatial, temporal, and treatment related sources. Compositional variation was large for all metrics, even in untreated plots. Variability in composition at a site was related to gamma diversity, local variation, and climate, but each metric responded to a different set of explanatory variables. Much of the compositional variation was related to spatial and temporal sources and surprisingly little to nutrient addition. We highlight four implications: (i) grasslands exhibit substantial compositional variation both spatially and temporally, even in control plots; (ii) differences in compositional variation among sites are predictable based on characteristics of the biotic environment (gamma diversity, biomass, local variation in productivity) and climate; (iii) using multiple compositional metrics provides unique ecological insights; and (iv) spatial and temporal sources of compositional variation need to be accounted for to detect actual treatment effects. Programs that do not account for these sources, both when designing sampling strategies and analyzing data, will have a weaker ability to detect the impacts of human activities.