A decision tree approach modelling functional group abundance in a pasture ecosystem

Decision tree, one of the data mining approaches, was used to model the relative abundance of five functional groups of plant species, namely high fertility response grasses (HFRG), low fertility tolerance grasses (LFTG), legume, moss and flatweeds in a New Zealand hill-pasture ecosystem using aboveground biomass. The model outputs were integrated with a geographic information system (GIS) to map and validate the predictions on a pasture. The decision tree models clearly revealed the interactions between the functional groups and environmental and management factors, and also indicated the relative importance of these factors in influencing the functional group abundance. Soil Olsen P was the most significant factor influencing the abundance of LFTG and moss, while soil bulk density, slope and annual P fertiliser input were the most significant factors influencing the abundance of legume, HFRG and flatweeds, respectively. Generally, slope and soil Olsen P were the two key factors underlying the patterns of abundance for these five functional groups. For the five functional groups studied, there was an overall predictive accuracy of 75%. Modelling functional group abundance simplified the investigation of the complex interrelationship between species and environment in a pasture ecosystem. The integration of the decision tree with GIS in this study provides a platform to investigate community structure and functional composition for a pasture over space, and thus can be applied as a tool in pasture management.     

Modelling the abundance of three key plant species in New Zealand hill-pasture using a decision tree approach

Decision tree models were developed to investigate and predict the relative abundance of three key pasture plants [ryegrass (Lolium perenne), browntop (Agrostis capillaris), and white clover (Trifolium repens)] with integration of a geographical information system (GIS) in a naturalised hill-pasture in the North Island, New Zealand, and were compared with regression models with respect to model fit and predictive accuracy. The results indicated that the decision tree models had a better model fit in terms of average squared error (ASE) and a higher percentage of adequately predicted cases in model validation than the corresponding regression models. These decision tree models clearly revealed the relative importance of environmental and management variables in influencing the abundance of these three species. Hill slope was the most significant environmental factor influencing the abundance of ryegrass while soil Olsen P and annual P fertilizer input were the most significant factors influencing the abundance of browntop, and white clover, respectively. Soil Olsen P of approximately 10 μg/g, or a slope of about 10.5° was critical points where the competition between ryegrass and browntop tended to come to an equilibrium. Integrating the decision tree models with a GIS in this study not only facilitated the model development and analyses, but also provided a useful decision support tool in pasture management such as in assisting precision fertilizer placement. The insights obtained from the decision tree models also have important implications for pasture management, for example, it is important to maintain a soil Olsen P higher than 10 μg/g in order to keep the dominance of ryegrass in the hill-pasture.