Meta-analysis and its application in phenological research: A review and new statistical approaches

Abstract

Climate change is one of the greatest challenges to sustainable development. Results from many individual studies show significant variation in response to climate change and human activities. Given the scope and variability of these trends, global patterns may be much more important than individual studies in assessing the effects of global change. There is a need to synthesize quantitatively, existing results on ecosystems and their responses to global change, in order to reach a general consensus or summarize the differences. Meta-analysis provides such a quantitative synthetic method, in that it statistically integrates results from individual studies to find common trends and differences: so called fingerprints of change. Only a small number of studies to date have performed a formal statistical meta-analysis of species' responses or have synthesized independent studies to reveal emergent and globally coherent patterns of ecological changes in physical and in biological systems in the phenology and distribution of plants and animals. There are very few reports available on the use of meta-analysis to examine global climate change in Australia. The unequivocal conclusion across the global syntheses, to date, is that twentieth-century anthropogenic global warming has already impacted on the Earth's biota. This chapter reviews the general methodology of meta-analysis, assesses its advantages and disadvantages, synthesizes its use in global climate change phenology and discusses future directions and proposes new statistical methods, as yet not applied to phenological research, and only recently applied, only in part, in the health-climate epidemiological literature.