Votre recherche
Résultats 4 ressources
-
Drought-induced tree mortality, which rapidly alters forest ecosystem composition, structure, and function, as well as the feedbacks between the biosphere and climate, has occurred worldwide over the past few decades, and is expected to increase pervasively as climate change progresses. The objectives of this review are to (1) highlight the likely ecological consequences of drought-induced tree mortality, (2) synthesize the hypotheses related to drought-induced tree mortality, (3) discuss the implications of current knowledge for modeling tree mortality processes under climate change, and (4) highlight future research needs. First, we emphasize the likely ecological consequences of tree mortality from ecosystem to biome to continental scales. We then document and criticize multiple non-exclusive tree mortality hypotheses (e.g., carbon starvation — carbon supply is less than carbon demand; and hydraulic failure — desiccation from failed water transport) from a more comprehensive ecological perspective. Next, we extend a forest decline concept model, Manion’s framework, by considering new emerging environmental conditions, for a more thorough understanding of the effects of climate change on forest decline. We find that an increase in drought frequency and (or) climate-change-type droughts may trigger increased background tree mortality rates and severe forest dieback events, accelerating species turnover and ecological regime shifts. The contribution of CO 2 fertilization, rising temperature within the optimal growth range, and increased nitrogen deposition may defer or reduce this trend in tree mortality, but such contributions will vary between locations, species, and tree sizes. Multiple hypotheses proposed for drought-induced tree mortality are discussed, but coupling carbon and water cycles could help resolve the debate. The absence of a physiological understanding of tree mortality mechanisms limits the predictive ability of current models from stand-level process-based models to dynamic global vegetation models. We thus suggest that long-term observations, experiments, and models should be tightly interwoven during the research process to better forecast future climate changes and evaluate their impacts on forests.