Prediction of Grassland Plant Responses to Global Change

Global change is challenging the predictive power of biology. Increasingly, biologists are being asked to draw together existing knowledge and new experimental data, in order to provide predictions of the effects of global change at regional, national and global scales. In this paper we examine how well global change effects on grassland plants can be predicted, and what avenues might be used in future to improve predictive capabilities. Broad predictions of the direction of change in grassland vegetation can be made quite simply and from a limited set of data. Predictions of the rate and magnitude of change are more difficult to make but are of much greater potential value. In order to achieve the latter, it is necessary to synthesise a vast array of relevant biological information, extending from short-term physiological effects of temperature, water, carbon dioxide and UV­D, through to environmental cues for phenology, reproductive growth and succession in vegetation. This highlights the need for reductionist models which can integrate short-term physiological data with knowledge of the life-history, resource demands, climatic tolerance and reproductive biology of the plants and the downstream effects of environmental changes over several seasons. We explore the value in this process of recognising recurrent specialisations (or strategies) and distinct functional types. We also examine how models can incorporate land-use effects when making predictions on the resistance and resilience of grassland communities, The potential of these approaches is illustrated using some preliminary results from recent laboratory screening and field experiments on New Zealand and British grassland plants.