Disentangling the antagonistic effects of development and temperature on the autumn phenology of european beech using manipulative experiments

Phenological shifts are crucial indicators of ecosystem responses to climate change. Complex feedbacks exist between the timing and length of temperate deciduous forest growing seasons and water, energy and carbon cycling. Understanding these feedbacks is essential for accurate future climate and forest productivity predictions. Despite knowing that warmer late-season temperatures postpone autumn senescence, extending the growing season, the specific drivers of autumn phenology remain unclear. Recent studies indicate that early-season warming may counteract these growing season extensions through developmental constraints. The compensatory point between these two antagonistic effects, i.e., the date when the effect of temperature on senescence reverses, may be governed by day length. However, the relative roles of early-season development and late-season temperature in driving autumn phenology need to be clarified. Our climate manipulation experiments on European beech (Fagus sylvatica) aimed to address this complexity. We show that cooling at different times of the day has inverse effects on beech's primary growth cessation, affirming that warming can either postpone bud set or advance it by speeding up development. By manipulating leaf-out timing we found that reduced early-season development delayed bud set (mean = 3.56 days, p = 0.002). The effect of cooling in July was more pronounced for the reduced-development (late-leafing) trees (mean = 5.27 days, p < 0.001). This indicates that the effects of temperature later in the season depend on early-season development, which, in turn, points to a flexible effect reversal date. However, cooling in August advanced bud set regardless of leaf-out timing (mean = -3.08 days, p < 0.001), indicating that temperature becomes increasingly important as a driver of autumn phenology as the late-season progresses. This study underscores the importance of accounting for both developmental and warming factors when predicting autumnal phenological shifts.