The life-cycle of leaves, from budburst, leaf unfolding, maturation and senescence, is what we call leaf phenology. In temperate and boreal regions, these phenological stages are highly sensitive to temperature, among a few other factors. Temperature act as a direct signal for trees to know if conditions are good or not to grow or shed their leaves. This is the reason why the observation of leaves is widely used as an indicator of climate change. Over the last decades rising temperatures have lengthened the annual growth cycle of trees in the Northern hemisphere by advancing leaf unfolding in spring and generally delaying leaf fall in autumn.

The problem is that these shifts in leaf phenology result in several drawbacks for ecosystems and feedbacks to climate. For example, earlier budburst in spring exposes trees to late frost risks and potential mismatches with pollinators and herbivores. A longer growing season also results in longer metabolic activity during which trees will transpire water over a longer period and hence amplify drought stress during summer. Leaf phenology is also the starting point of land carbon cycle models, driving carbon allocation and partitioning between organs, and thus controls in large parts carbon budget estimates. Leaf phenology is therefore often quoted as the second largest uncertainty in future climate predictions. Understanding if phenology will keep pace with climate warming is thus more than just a hobby for gardeners.

Behind the paper: This is during a coffee break in spring 2019, trying to get fresh air outside, away from my computer, that I realized the weather was hot for this period of the year. However, my weather station was clearly disagreeing with me on the perceived temperature outside. The thing is that I was simply dressed in black, fully exposed to the sun. This is when I realized that trees might experience the same feeling. As a forest modeler I am studying and simulating phenology from air temperature, as it is the common practice for centuries now (Yes, the first pheno-climatic model was developed during the 18^th century). But in fact, tissue temperature and air temperature differ by several degrees because of solar radiation, wind or rain, as we experience this feeling ourselves all the time. The difference is that a tree has to face it.

In our paper we discuss the potential key role of microclimate, i.e. climate conditions sensed by the plant, and bud traits in controlling phenology. Even if we know for a long time that temperature sensed by plant tissues are different than air temperature, all phenology studies, models and interpretation of phenology response to climate change are still based on average air temperature acquired at distant weather stations. By applying well known energy budget theories (i.e., the amount of energy that an object absorbs or emits) to existing phenological observations, we managed to approximate the temperature of buds and how this temperature evolves over the period preceding leaf unfolding compared to air temperature. This approach allowed us to explore and discuss how using air instead of bud temperature can impact our interpretation of phenology sensitivity to warming, but also what are the factors that can affect the temperature sensed by trees. Bud temperature is highly sensitive to bud traits that will affect the amount of solar radiation that will be absorbed. Thus, different species with different bud colour or shape might respond differently to climate condition. Bud temperature is also highly sensitive to the amount of incoming solar radiation and wind. All these combine factors resulted in bud temperature being more variable than air temperature, and more importantly, non-linear, and non-proportional. Our results thus suggest that air temperature might be a biased predictor of leaf phenology in certain conditions, affecting our interpretation and forecasting of phenology response to climate change.

Published in Nature Plants here: https://www.nature.com/articles/s41477-022-01209-8

Peaucelle, M., Peñuelas, J., & Verbeeck, H. (2022). Accurate phenology analyses require bud traits and energy budgets. Nature Plants, 1-8.