Dry season droughts may increasingly threaten Mediterranean forests under climate change. While plants employ three desiccation avoidance strategies to avoid or delay dehydration damage, including reduced water loss, enhanced tissue water storage, and improved root water access, resource allocation competition may lead to trade-offs among these strategies that are not yet fully understood. We investigated six Mediterranean woody species by analysing: (1) twig hydraulic capacitance (0.32 - 2.81 mmol m(-2) MPa-1) representing tissue water storage capacity; (2) twig residual conductance (g(res)) at 25 degrees C (1.23 -7.73 mmol m(-2) s(-1)) reflecting water loss rate; and predawn water potential (Psi(PD)) and its difference from midday water potential (triangle Psi) at the end of the dry season as root water access indicators. Significant trade-offs in plant desiccation avoidance strategies were observed as g(res) positively correlated with triangle Psi (R-2 = 0.78, P = 0.02) and twig hydraulic capacitance negatively correlated with Psi(PD) (R-2 = 0.68, P = 0.04). Consequently, species with greater root water access exhibited lower tissue water storage capacity and higher g(res), potentially increasing mortality risk when soil moisture becames limiting. By inverting a plant desiccation model, we also demonstrated that minimum survival-required hydraulic capacitance and a novel risk index were both positively correlated with Psi(PD), consistent with historical mortality records. Additionally, despite temperature-dependent g(res) patterns which revealed species-specific responses, elevated temperatures amplified the risk index for all species.

Due to the increasing frequency of extreme weather events, drought damage to trees threatens forestry production and forest ecosystems worldwide. Assessing the site conditions under which trees are vulnerable to drought damage provides key information for the establishment of countermeasures to prevent such damage. This study aimed to clarify the differences in drought vulnerability of young planted forests between regions and species by using forest insurance claims from all over Japan as a damage indicator. We targeted the two most damaged species in two of the most drought-affected regions from 2016 to 2021. Although landform and soil type were found to be influential factors in the Kamikawa Subprefecture of Hokkaido, these factors did not affect the drought damage in Yamaguchi Prefecture. In Kamikawa, the drought damage risk was high for Larix kaempferi on river terraces and for Abies sachalinensis on mountain areas with compacted brown forest soil. Clayey soil, which can prevent plants from absorbing water, has been known to distribute on the terraces and the mountains with compacted soil in Kamikawa. Therefore, our analysis identified clayey soil as a cause of drought vulnerability in Kamikawa. In addition, L. kaempferi was suggested to be especially vulnerable on flat terraces with less permeable clayey soil due to root damage associated with excessive soil moisture before drought. This study demonstrated that forest insurance can be used not only for damage compensation, but also as a source of information for identifying region- and species-specific risk factors for meteorological damage in forests.

Increased tree mortality rates have been observed worldwide in connection to climate warming-related processes, such as drought, heat, fire, and insect pest outbreaks. An understanding of the drivers of tree mortality during the Anthropocene is urgently needed to estimate forest vulnerability in a warmer climate. In this study, we assessed the drivers of tree mortality in an urban recreational boreal forest area in Helsinki, Finland, of approximately 830 ha, where increased tree mortality rates have been recently observed. A time series of aerial images was used to quantify tree mortality over the area to detect dead trees from 2005 to 2021 at seven timestamps. In total, 6008 dead trees were observed from the aerial images collected during the monitoring period. Forest environmental and climatic variables were used to explore the tree mortality drivers for individual trees and tree communities using logistic regression and correlation analysis. Our results showed that droughtrelated variables, i.e., the standardised precipitation evapotranspiration index and the Palmer drought severity index, were linked with increased tree mortality rates. We found that the stand-level basal area predicted tree mortality risk and was linked to site type; smaller basal area stands were located on rocky dry soils, resulting in a greater probability of tree mortality. We also observed that trees at high elevations or on steep slopes showed a greater mortality risk. Our results can increase the understanding of tree mortality in urban areas and help the planning of built and green areas in a changing climate.

Bark beetle outbreaks are a significant cause of high tree mortality rates, dramatically impacting the resilience of forests. Understanding the triggers and impacts of these outbreaks is critical for effective forest management strategies. In this context, we studied windfall and bark beetle outbreaks in the period 2015-2021 in the southern part of Kurilskiy Nature Reserve, North Pacific Ocean region. Massive bark beetle outbreaks on Kunashir Island were not previously studied. The dominant tree species are Yezo spruce (Picea jezoensis) and Sakhalin fir (Abies sachalinensis), which collectively form spruce -fir forests on Kunashir Island. Glehn spruce (Picea glehnii), although less common on the island, forms pure spruce forests. Typically, spruce bark beetle (Ips typographus L.) attacks Yezo and Glehn spruce, and fir bark beetle (Polygraphus proximus) attacks Sakhalin fir. Significant tree mortality was observed in the aftermath of a substantial bark beetle outbreak, induced by gale -force winds. The total disturbance area was 620.5 ha, which is about 4% of the study area, 72% of the windfall area, and 28% of the bark beetle -infested area. Utilising a forest loss dataset (Global Forest Change dataset) and Sentinel 2 imagery, we identified windfall areas and standing tree mortality through unsupervised classification, accompanied by field sampling. Subsequently, the authors analysed the main drivers of disturbances caused by wind and bark beetle outbreaks using datasets combined with forest inventory data. Field data showed a pattern of tree infestation by both bark beetle species at the tree level, and the potential infestation of Sakhalin fir by the spruce bark beetle. We used boosted regression tree (BRT) models to analyse the main drivers using the presence and severity of wind damage and bark beetle outbreaks by phases. As predictors, we used a set of forest characteristics (tree species percentage, height, diameter of trunk, age, growth class) and environmental characteristics (slope, elevation, potential solar radiation, soil pH). The bark beetle outbreak was split into two phases: the first phase (2017-2019) involved the transition of bark beetles from colonised downed trees to standing trees, and the second phase (2020-2021) occurred during the spreading of beetles in standing trees. Stand tree characteristics were of greater significance for the likelihood of a bark beetle outbreak than environmental characteristics, across both phases for the southern part of the reserve. The percentage and the age of Glehn spruce and Yezo spruce were the main influencing factors for the presence and severity of an outbreak.