The surprising resistance of trembling aspen (Populus tremuloides) to experimental drought.

Under climate change, not only are droughts becoming more severe, but seasonal precipitation patterns are changing, as well as the timing of snowmelt. Especially in mountainous or northern forests, where trees may depend strongly on snowmelt, it is therefore crucial to understand how the timing of droughts affects tree health and performance. Trembling aspen (Populus tremuloides Michx.) is the most widespread tree species in North America, and, like many other species, it has experienced large-scale, drought-related diebacks over recent decades. Depending on where they grow, aspens rely to various degrees on snowmelt for growing season transpiration, however it remains unclear whether droughts during different seasons affect the trees differently. Using a field experiment, we tested the effect of spring and summer drought on aspen leaf gas exchange, growth and hydraulics.Over two years (2024 & 2025), we conducted a fully factorial experiment in a mature aspen forest in northern Utah, USA. In the year before the experiment, plots were trenched to ca 70cm depth. Each year, we induced a spring drought by removing most of the snowpack (ca 1.7m3 m-2; > 95 % of total snowpack) in early spring, while for the summer drought, starting in June, we covered ca 60 % of the forest area with rainout shelters. We measured stem growth and various physiological parameters including water potentials (Ψ) and leaf gas exchange up to weekly throughout the growing season. Additionally, we assessed hydraulic conductivity and vulnerability, in early and late summer each year.Snow removal led to slightly more negative predawn Ψ in early summer, but these effects were small and short-lived. Otherwise, to our surprise, neither drought treatment affected Ψ, leaf gas exchange, growth or hydraulic conductivity, despite extremely dry topsoil. Interestingly however, during the much drier growing season 2025, trees showed reduced predawn Ψ, stomatal conductance and assimilation compared to 2024 irrespective of experimental treatment. These findings suggest that our trees were more affected by regional-scale droughts than by plot-level precipitation manipulation. One possible explanation for this is that these trees may have access to the water table. This is surprising as aspens are typically thought to mostly rely on water from shallow soil layers.