Using trophic structure to reveal patterns of trait‐based community assembly across niche dimensions. (13th March 2017)

• Record Type: Journal Article
• Authors: Fitzgerald, Daniel B.; Winemiller, Kirk O.; Sabaj Pérez, Mark H.; Sousa, Leandro M.
• Editors: Seymour, Colleen
• Other Names: Seddon Philip guestEditor.
• Is Part Of: Functional ecology. Volume 31:Number 5(2017)
• Page Start: 1135
• Page End: 1144
• Publication Date: 2017-03-13
• Subjects: α‐niche -- community assembly -- fish -- functional diversity -- null model -- stable isotope -- Xingu River; Ecology -- Periodicals; 574.505
• DOI: 10.1111/1365-2435.12838
• Languages: English
• ISSNs: 0269-8463

Abstract

Summary: Trait‐based approaches for studying community assembly have improved understanding of mechanisms; however, the challenge of interpreting process from pattern is complicated by the possibility of multiple mechanisms operating simultaneously. Different traits may influence the assembly process in different ways. Analysing patterns of functional diversity among co‐occurring species for each trait individually may aid interpretation of complex assembly processes; yet, few studies have tested whether patterns vary depending on trait function. We used tropical fish assemblages from the Xingu River, Brazil to test whether traits associated with resource acquisition play a stronger role in niche segregation relative to other traits. First, a null modelling approach was used to determine how trait distributions within local assemblages deviated from expectations under random assembly. Then, correlations between functional traits and stable isotope ratios (δ 13 C, δ 15 N) were used as a measure of a trait's association with trophic structure. Finally, we used mixed effects models to test whether traits having higher correlations with trophic structure also had greater deviation from null expectations. In addition, we explored how well stable isotopes explain multivariate functional trait variation and compared our correlation‐based approach for organizing traits with previous categorical approaches. A significant relationship was found between a trait's deviation from null expectations and its correlation with isotopic patterns. Traits strongly associated with trophic structure had greater dispersion from the assemblage mean and were more evenly spaced than weakly associated traits. Traits strongly associated with trophic structure also were more clustered because trophic diversification tended to occur around some basic feeding strategies, such as benthic grazing or capturing food items from the water column. Based on redundancy analysis, isotopic ratios explained a low (11·6%) but significant portion of trait variation. In this study system, traits strongly associated with trophic ecology were more influential in niche differentiation among coexisting species compared to weakly associated traits. These results suggest that certain traits may respond to assembly mechanisms in predictable ways despite the complex, multidimensional nature of the assembly process. Methods designed to identify differential trait response will be critical to developing a general theory of trait‐based community assembly. A lay summary is available for this article. Abstract : Lay Summary