Meta-analysis of meta-analyses in plant evolutionary ecology

Abstract

After two decades of meta-analyses on plant traits, we can now look for global emergent patterns in plant evolutionary ecology. Hundreds of meta-analyses have focused on the effects of specific selection pressures on plant fitness, and the buildup of such results allows us to ask general questions regarding selection pressures and plant responses, a major focus of evolutionary ecology. Plant traits are affected by both abiotic and biotic factors. For example, biotic pressures like herbivory may affect physiological (i.e. secondary defences) and reproductive (i.e. seed predation) traits. Similarly, abiotic pressures such as increased CO₂ may affect both plant physiology and reproduction. We tested whether biotic or abiotic selective pressures are more important for plant traits, and if the strength of the response to those pressures depends on the plant trait studied by meta-analyzing published meta-analyses on plant responses. We classify meta-analyses according to the type of response variable studied (fitness and non-fitness traits) and the type of selective pressure examined (biotic or abiotic). Our database showed biases in the meta-analysis literature, for example that the majority of studies are focused on non-fitness traits, i.e. on traits that are not directly related to reproduction or survival, and furthermore, on non-fitness traits under abiotic selection pressures. The meta-meta-analysis showed that the strength of responses to selection depends on the nature of selection (stronger for biotic than for abiotic factors) but, unexpectedly, not on the type of trait under study as previously found. The stronger responses to biotic factors can be explained if biotic selection is more variable in space and time, driven by interactions with other organisms. The relative importance of biotic versus abiotic factors on plant traits has been little studied in the past, and would benefit from more studies and reviews that fill the under-represented combinations of selective pressures and plant traits (i.e. abiotic factors on fitness traits).