Correspondence analysis of functional groups in a riparian landscape

Abstract

We used multivariate analysis and ordinations to characterize thecomposition and distribution of woody vegetation within the Ozark NationalScenic Riverways (ONSR), Missouri, USA. The objectives of the study were to: 1)evaluate patterns of woody species distributions along existing environmentalgradients; 2) determine if different classes of woody plants (i.e., dominantoverstory trees, all trees, understory trees, and shrubs) responded similarlytothe same suite of environmental variables; and 3) determine if discreteecotonaland/or ecoclinal vegetation patterns were present across the landscape. Woodyvegetation was sampled from 94 plots along 35 transects positioned at rightangles to the river channel. Sample plots were analyzed with DetrendedCorrespondence Analysis (DCA), Canonical Correspondence Analysis (CCA), andTWINSPAN. Overall, woody vegetation was correlated with several environmentalgradients, including elevation of the plot above the river, soil pH, soilmoisture, and soil particle size. Responses to secondary gradients differedamong the four classes of plants analyzed, however. CCA biplots of understorytrees indicated that patterns of those species were strongly correlated withslope through the plot and sand content of soil. CCA biplots of shrubs showedthat CCA axes were most strongly correlated with soil organic matter content,soil moisture, and silt content. Further, there was limited evidence fordiscrete assemblages of woody species, with the exception of streamsidevegetation. Instead, mixing of woody species was observed across a broadtransition zone. Because there is little correspondence between vegetationlayers, our results demonstrate including plant classes other than a subset ofcanopy dominant trees can provide additional resolution in characterizingvegetation responses along complex environmental gradients.