Create your own conference schedule! Click here for full instructions

Abstract Detail

Ecological Section

HOLSINGER , KENT E [1], Moore, Timothy E. [2], Mitchell, Nora [3].

Understanding whole-plant trait-environment associations using phylogenetically-corrected multiple response multiple regressions.

Plants adapt to environmental differences in many different ways. One way of identifying trait differences that are important in environmental adaptation is to look for associations between traits and environments. If leaf mass per area (LMA), for example, is negatively associated with mean annual precipitation, it suggest that dense leaves are favored by natural selection in areas of low rainfall. There are two challenges when exploring trait-environment relationships in a set of closely related species: (1) If traits are phylogenetically conservative, trait-environment relationships are confounded with trait-phylogeny relationships. (2) Plants respond to the environment as an integrated whole, trait-environment associations in one trait may amplify or diminish those in another trait. We describe a phylogenetically-corrected multiple response model that addresses both of these challenges. Building on comparative approaches deriving from Lynch’s work in the early 1990s, we illustrate how to use an independently-estimated phylogeny to examine trait-environment relationships in a suite of integrated characters. Specifically, we introduce a statistical model that treats multiple traits of an individual measured simultaneously as a single response variable in a multiple regression with environmental observations as the covariates. Phylogenetic correlations are introduced through species-specific random effects, in which the covariance structure of the random effects is constrained by the phylogeny. We illustrate the model by applying it to data from Protea and Pelargonium in the Cape Floristic Region of South Africa, and we show that our model performs significantly better than ones that treats individual traits as independent responses or that treat all species as equally related.

Log in to add this item to your schedule

Related Links:
Parallel Evolutionary radiations in Protea and Pelargonium in the Greater Cape Floristic Region

1 - University Of Connecticut, Department Of Ecology & Evolutionary Biology, 75 N. Eagleville Road, U-3043, STORRS, CT, 06269-3043, USA
2 - University of Connecticut, Department of Ecology & Evolutionary Biology, U-3043, Storrs, CT, 06269-3043, USA
3 - University Of Connecticut, Ecology & Evolutionary Biology, 75 North Eagleville Rd, U-3043, Storrs, CT, 06269, USA

Cape Floristic Region
Phylogeentic regression
trait-environment association.

Presentation Type: Oral Paper:Papers for Sections
Session: 24
Location: Clearwater/Grove
Date: Tuesday, July 29th, 2014
Time: 10:45 AM
Number: 24003
Abstract ID:548
Candidate for Awards:None

Copyright 2000-2013, Botanical Society of America. All rights reserved