When weather becomes less predictable, so do reproductive traits

Small variations in a plant’s environment can cause changes in the way traits are expressed in the growing plant both at the individual and population level. While these trait changes may be merely developmental noise, theory suggests that they may also be under selection pressure. Until now, evidence supporting this hypothesis has been sparse.

In a new paper published in Annals of Botany, Martí March-Salas and colleagues investigated whether differences in the predictability of a plant’s environment affect the variability of reproductive traits within both individuals and populations, as well as whether intra-individual variability is subject to selection pressure. The authors grew Onobrychis viciifolia under field conditions for four years, using treatments of more or less predictable precipitation. They then quantified variation at the individual and population levels, as well as testing for maternal environmental effects on offspring.

The researchers found that less predictable precipitation, particularly during late growth, led to greater variation in reproductive traits at both the individual and population levels, suggesting that the predictability of a plant’s environment can affect its reproductive allocation strategy. The within-plant variation of all the studied traits was under a consistent stabilizing selection, meaning that within-plant variability is a functional trait rather than simply phenotypic noise, and should therefore be taken into account in eco-evolutionary research.

No consistent effects on variability were passed from parent to offspring in the three generations studied, which helps to explain how the potential for variability is maintained in the face of stabilizing selection pressure. “This finding may be explained by the climatic unpredictability prevailing in most parts of the natural distribution of O. viciifolia, under which maintaining intra-individual variability may be highly beneficial and rapid responses may lead to important fitness costs,” explain the authors.

These findings have relevance to climate change, which in addition to shifting averages for weather-related metrics, will also increase their variability. “These results indicate that effects of climate change on the intrinsic predictability of the weather, rather than changes on weather means, may entail greater shifts in trait variation, in line with studies suggesting that differences in environmental predictability may have greater effects on plant performance than differences in environmental averages,” write the authors.