Current approaches to modelling the environmental niche of eucalypts: implication for management of forest biodiversity
Austin, MP & Meyers, JA 1996, Forest Ecology and Management, vol. 85, pp. 95-106.
Due to cost, time and other constraints, forest managers cannot have optimum knowledge of all the species distributed throughout the ecosystem they administer. One of the next best options is having access to spatial models to predict the area’s biodiversity and its distribution. This paper, in the context of nine species of eucalyptus trees across 8377 sites in south-eastern Australia, examined the performance of two modelling techniques: Generalised Linear Modelling (GLM) and Generalised Additive Modelling (GAM). The authors chose trees to study because of their size and immobility, and their value in predicting the distribution of some other biota, including koalas. Results pertaining to seven environmental predictors (including temperature, nutrients and light) for one species, Eucalyptus cypellocarpa, are provided.
Building robust models requires selecting appropriate predictor variables, which sit on a continuum from indirect variables (such as, latitude or altitude) to direct causal variables (for example, temperature or the nitrate concentration at the root surface). Evidence has shown that predictors are effective if they are less based on location and more process-oriented with a higher relevance to biological processes. However, predictors are not always available for the particular region under study. A useful technique that may overcome some difficulties associated with a lack of direct variable data is linking indirect variables to direct ones. In this study, the researchers used rainfall, temperature, nutrient and light as predictor variables, and produced estimates relevant to these by accessing data from sources that included weather and rainfall stations and regional geology maps.
The researchers present another requirement for modelling the environmental niche of trees: complementary ecological theory, which will indicate the expected response function of organisms. The researchers discuss niche theory and that the traditional assumption for its hypotheses is the Gaussian response curve. They explain that for the realised niche (when the organism is in competition with others) the Gaussian response curve will display in a range of skewed or curvilinear shapes. They argue that to avoid wasting resources by adopting inefficient methodologies for predicting species distribution and performance, it is necessary to appreciate that a linear response will bias the response.
The researchers found that non-parametric GAM provided evidence of more informative response shapes than parametric GLM ‘due to the flexible nature of the non-parametric smoothing function’. Making incorrect assumptions about the shapes of response curves when utilising models to evaluate the conservation value of an area may result in significant distortion of the distribution and abundance of species, and the inefficient use of limited resources.
Summarised by Rosemary Shaw
Disclaimer: The summary of this report is provided for reference purposes only and does not represent the findings or opinions contained in the original report. Although every effort has been made to bring forward the main elements of the report, this review is no substitute for the full the report itself. Should you have any concerns or perceive any errors please contact us and we shall endeavour to rectify and improve the review.