Increasing forest mortality and its drivers: Simulating central European forests under climate change

English. Increasing tree growth and mortality rates in Europe are still poorly understood and have been attributed to a variety of drivers. This study aimed to relate increasing forest mortality rates in six central European countries to climate drivers (CO2 concentration, temperature and precipitation) from 1985-2015, using a process-based vegetation model. For this, the direct (e.g. mortality due to water scarcity) and indirect (e.g. enhanced competition leading to increased mortality) effects of changes in climate drivers on mortality were assessed. Using the LPJ-GUESS dynamic vegetation model (DVM), a dataset showing increased canopy mortality rates was aimed to be reproduced. Factorial simulations excluding changes in individual drivers were run to identify causes of simulated trends. The lack of relationship between simulated and observed canopy mortality rates were suggested to be linked to simplifications of the model. Also, a link to land use changes and increased harvest intensity, which cannot be captured by a model simulating natural vegetation is suggested. In most countries examined, an increased tree mortality could partially be attributed to increased competition, caused by faster tree growth and crowding. An exception was Switzerland, where a negative trend in canopy mortality was associated with its vegetation being simulated at a higher altitude and treeline advance was suggested to be responsible for decreasing mortality rates and increasing NPP. The main driver of increased competition mortality was identified to be the increase in CO2 concentration. Interestingly, changes in seasonal precipitation patterns caused an increase in water availability and a small increase in competition. Increasing nitrogen deposition partially increased competition trends, but it remains uncertain at which scale. Temperature (mainly increasing stress mortality) was the largest driver of mortality followed by CO2 concentration (mainly increasing competition mortality due to higher productivity induced by ‘CO2 fertilisation’), however, this assessment includes a large uncertainty. Further studies distinguishing between vitality mortality caused by direct and indirect resource stress (through neighbourhood crowding) are recommended to decrease that uncertainty. This study contributes to the understanding of the current changes in the global carbon cycle and sink, can help to adapt forest management practices to those changes and improves the understanding of the LPJ-GUESS model.

English. European forests grow and die faster than in the past. This has been caused by many different drivers that have changed because of climate change. In this study it was investigated if the increase in tree mortality in European forests from 1985-2015 was caused by higher stress among trees because resources decreased (stress mortality) or because resources increased causing trees to grow and compete with each other for limited resources. The role of CO2 concentration, temperature and precipitation was assessed for the changes in resource availabilities in forests. A dynamic vegetation model (LPJ-GUESS), that simulates vegetation based on climate conditions, was run to be able to link changes to individual drivers. By comparing simulations with change in individual driver ‘switched off’ to a simulation aiming to reproduce observed vegetation trends, one could identify causes of tree mortality trends. The trends in simulated tree crown death (canopy mortality) were shown to not be related to the trends revealed by an observational dataset. This discrepancy was attributed to simplifications of the model and by land use changes and increased timber harvest intensity in the dataset that are not captured by a model not simulating human impact through e.g., harvest. The increase in tree mortality was partially shown to be related to increased competition, since trees were displayed to grow faster, which leads to trees being outcompeted faster (competition mortality). A decrease in canopy mortality in Switzerland was the exception and was suggested to be caused by its vegetation being simulated at a higher altitude. There, vegetation is generally more stressed and only vegetation that can cope with those conditions can grow there (usually smaller, herbaceous vegetation). Due to climate change the conditions are becoming more favourable for different vegetation so that more species start to grow in higher altitudes (tree line advance). Decreasing tree mortality rates were suggested to be caused by growth conditions improving. The main driver of increased competition mortality was identified to be the increase in CO2 concentration, which positively impacts vegetation productivity through ‘CO2 fertilisation’. Changes in precipitation during the different seasons caused an increase in water availability and a small increase in competition. Temperature (mainly increasing stress mortality) was shown to be the largest driver of mortality followed by CO2 concentration (mainly increasing competition mortality), however, this assessment includes a large uncertainty. This study contributes to the understanding of the current changes in the global carbon cycle and sink, can help to adapt forest management practices to those changes and improves the understanding of the LPJ-GUESS model.