Ecological Indicators (Jan 2021)
Tree vitality indicators revealed a rapid response of beech forests to the 2018 drought
Abstract
Changes in regional and global hydrological cycles and temperature regimes will affect forest health, functioning and growth, and therefore their role in carbon (C) sequestration. A severe spring-summer drought hit large parts of central Europe in 2018 leading locally to premature leaf senescence in deciduous tree species. In late summer 2018 and 2019, we carried out a special survey on selected plots (n = 75) of the Swiss National Forest Inventory to detect whether beech (Fagus sylvatica L.) trees (n = 271) responded to the drought, and to evaluate possible lasting effects in the subsequent year. We concentrated on three vitality indicators: severe crown transparency (SCT), severe leaf browning (SLB), and growth in terms of basal area increment (BAI). In late summer 2018, we observed a frequency of beech trees affected by SLB and SCT significantly higher than in long-term baseline data. The most significant predictors of the occurrence of SLB and SCT were previous growth, species composition, tree size and precipitation deficit. Overall, BAI was significantly lower in 2018 compared to the 2010–2017 average. In particular, trees with both SLB and SCT grew significantly less than unaffected trees, with 40–100% lower median C increment. SCT, SLB and BAI largely recovered after more favorable weather conditions in 2019: almost no instances of SLB were recorded, and SCT was comparable to long-term baseline data. BAI in 2019 was significantly higher than in 2018 and 2010–2017, and weakly related to the crown condition recorded in late summer 2018. We conclude that SCT and SLB recorded in late summer 2018 indicated rapid, drought-induced canopy deterioration of beech, associated with an immediate reduction in BAI and tree-level carbon sequestration. These effects were, however, mainly temporary for most of the investigated beech trees. Since the standard long-term forest health assessment was not able to capture the above signals, we discuss opportunities for adapting its design without jeopardizing data comparability over time.
