Predicting plant responses to global environmental change

SUMMARY Predicting the future responses of plants and ecosystems to further changes in the CO 2 concentration of the atmosphere and to the possibility of global warming are important current concerns. Predictions have been most frequently attempted using short‐term, single‐factor experiments in controlled environments. However, these experiments have failed to indicate the outcome of field experiments at larger spatial and temporal scales. Some of this failure is due to ignorance of environmental conditions and interactions while some is due to the use of inappropriate short‐cuts, such as the addition of fertilizers for simulating enhanced mineralization, and some is due to ignorance of the processes involved in scaling‐up from individual plants to populations. Long‐term observations on plants in ecosystems nevertheless indicate that community‐scale experiments may provide a useful but imperfect capacity to predict ecosystem responses. Although difficult to implement in practice, it is concluded that catchment‐scale experiments offer the best opportunity to predict plant, community and ecosystem responses to environmental change. CONTENTS Summary 239 I. Introduction 239 II. CO 2 ‐enrichment experiments 240 III. Experiments with applied nitrogen fertilizer 243 IV. Population ecophysiology 245 V. Ecosystem predictions 248 Acknowledgements 249 References 250