Abstract
Abstract. Vegetation patterns arise from the interplay between intraspecific and interspecific biotic interactions and from different abiotic constraints and interacting driving forces and distributions. In this study, we constructed an ensemble learning model that, based on spatially distributed environmental variables, could model vegetation patterns at the local scale. The study site was an alluvial floodplain with marked hydrologic gradients on which different vegetation types developed. The model was evaluated on accuracy, and could be concluded to perform well. However, model accuracy was remarkably lower for boundary areas between two distinct vegetation types. Subsequent application of the model on a spatially independent data set showed a poor performance that could be linked with the niche concept to conclude that an empirical distribution model, which has been constructed on local observations, is incapable to be applied beyond these boundaries.
Highlights
Ecosystems are complex, evolving structures whose characteristics and dynamic properties depend on many interrelated links between direct gradients, their environmental determinants and potential natural vegetation, and the processes that mediate between the potential and actual vegetation cover (Baird and Wilby, 1999)
Vegetation patterns arise from the interplay between intraspecific and interspecific biotic interactions and from different abiotic constraints and interacting driving forces and distributions
The random forest distribution model was assessed in terms of: (i) its classification accuracy, (ii) its applicability on a similar alluvial floodplain, and (iii) its potential to model vegetation distributions based on a reduced number of important environmental variables in groundwater-dependent ecosystems
Summary
Exploring vegetation patterns is a central goal in ecology. Numerous studies examined environmental gradients in relation to vegetation type distributions in various ecosystems (Schulze et al, 1996; Famiglietti et al, 1998; Molina et al, 2004; Rudner, 2005), and different techniques have been developed to quantify vegetation-environment relationships. Canonical ordination (Jongman et al, 1995) for example, is widely applied in ecological studies to detect patterns of variation in vegetation data and quantify the main relations between vegetation and environmental variables. Distribution models tend to predict spatial distributions of species based on environmental variables (Guisan and Zimmerman, 2000; Guisan and Thuiller, 2005). An ensemble learning technique named random forests (Breiman, 2001; Prasad et al, 2006), is applied to a spatially distributed data set containing information on environmental conditions and vegetation type distributions. The random forest distribution model was assessed in terms of: (i) its classification accuracy, (ii) its applicability on a similar alluvial floodplain, and (iii) its potential to model vegetation distributions based on a reduced number of important environmental variables in groundwater-dependent ecosystems
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
