Editor: Naia Morueta-Holme Abstract Aim: Climate change has impacted forest ecosystems, leading to species-level tree migration. However, the impacts of climate change on forest communities are mostly unknown. Here, we assess changes to forest communities at three scales: withincommunity changes in species composition, individual community spatial shifts and changes across all communities. Location: Eastern USA. Major taxa studied: Forest tree species. Methods: Using a region-wide forest inventory dataset from the United States Department of Agriculture Forest Service's Forest Inventory and Analysis Program with > 70,000 plots, we identified forest communities using the latent Dirichlet allocation method. We analysed changes in species composition within communities and assessed community-level spatial shifts over the last three decades to quantify the responses of individual communities to climate change. We used the distribution of forest communities across climate conditions to predict where communities could migrate to during the study period and compared climate-predicted shifts with observed community shifts. Changes across all communities were modelled as a function of climatic and non-climatic variables using generalized linear mixed-effects models. Results: We identified 12 regional forest communities of the eastern USA, which varied in their stability of species composition over the study period. All communities experienced relatively short yet significant shifts in their spatial distribution (median = 8.0 km/ decade). Historical climate and changes in seasonal temperature variability were the best predictors of change across all communities. However, the distance and direction of individual community migration were poorly predicted by climate change, and the observed direction was often the opposite of the predicted direction. Main conclusions: Forest communities shifted their distributions over the last three decades, but climate change outpaced the rate of community migration. Continued lags between climate change and forest community responses and the lack of migration in the direction predicted by climate change might lead to the inability of forests to keep up with changing climate.

中文翻译：

---

社区层面对美国东部森林气候变化的响应

编辑：Naia Morueta-Holme 摘要目的：气候变化影响了森林生态系统，导致物种级树木迁移。然而，气候变化对森林群落的影响大多是未知的。在这里，我们在三个尺度上评估森林群落的变化：群落内物种组成的变化、个体群落空间变化和所有群落的变化。地点：美国东部。研究的主要分类群：森林树种。方法：使用来自美国农业部林务局森林清查和分析计划的区域范围的森林清查数据集，其中包含超过 70,000 个地块，我们使用潜在狄利克雷分配方法确定了森林群落。我们分析了社区内物种组成的变化，并评估了过去三年中社区层面的空间变化，以量化各个社区对气候变化的反应。我们使用不同气候条件下森林群落的分布来预测研究期间群落可以迁移到哪里，并将气候预测的变化与观察到的群落变化进行比较。使用广义线性混合效应模型将所有社区的变化建模为气候和非气候变量的函数。结果：我们确定了美国东部的 12 个区域森林群落，它们在研究期间的物种组成稳定性各不相同。所有社区的空间分布都经历了相对短暂但显着的变化（中位数 = 8.0 公里/十年）。历史气候和季节性温度变化的变化是所有社区变化的最佳预测指标。然而，气候变化对个体群落迁移的距离和方向的预测较差，观测方向往往与预测方向相反。主要结论：森林社区在过去三年中改变了它们的分布，但气候变化的速度超过了社区迁移的速度。气候变化与森林社区响应之间的持续滞后以及缺乏向气候变化预测的方向迁移可能导致森林无法跟上气候变化的步伐。气候变化对个体社区迁移的距离和方向的预测很差，观察到的方向往往与预测的方向相反。主要结论：森林群落在过去三年中改变了它们的分布，但气候变化的速度超过了群落迁移的速度。气候变化与森林社区响应之间的持续滞后以及缺乏向气候变化预测的方向迁移可能导致森林无法跟上气候变化的步伐。气候变化对个体社区迁移的距离和方向的预测很差，观察到的方向往往与预测的方向相反。主要结论：森林社区在过去三年中改变了它们的分布，但气候变化的速度超过了社区迁移的速度。气候变化与森林社区响应之间的持续滞后以及缺乏向气候变化预测的方向迁移可能导致森林无法跟上气候变化的步伐。