Data from: Long-term data suggest jarrah-forest establishment at restored mine sites is resistant to climate variability

1. Global climate change is projected to increase the frequency and intensity of drought in dry regions due to warming temperatures and declining rainfall. Severe drought can trigger tree mortality and drive persistent vegetation change. 2. To date, most empirical studies have focused on drought-induced mortality of adult trees, but this needs to be matched by similar efforts to understand drought impacts on seedling establishment if we are to understand the resilience of the world's forests to projected climate change. 3. The Eucalyptus marginata (jarrah)-forest ecoregion of south-western Australia has experienced a 17% reduction in mean annual rainfall from 1975 to 2011 compared with mean rainfall from 1900 to 1974. Drought-induced mortality of adult trees has been documented for jarrah forest, but there is limited understanding of its capacity to recover, making it difficult to predict the likelihood of persistent effects. 4. Long-term records of jarrah-forest restoration following bauxite mining are available for the 19-year period between 1992 and 2010. Records include annual data on seedling establishment in restored mine sites for 587 species in 1938 plots during a period of climatic variability. 5. We built a structural equation model to discriminate the relative effects of climate, restoration practice, and their interactive effects on three response variables including species richness of the restored plant assemblages. Climate variability had a significant negative effect on richness, but the effect size was relatively small, being less than half that of varying restoration practice. We suggest this is due, in part, to the reliability of rainfall (i.e. no change in the coefficient of variation, seasonality or evenness, between recent and historical climates) despite a threefold difference in the absolute amount of wet-season rainfall. Importantly, there was no evidence that restoration practices were compromised by interacting effects of increased climate variability. 6. Synthesis. Jarrah-forest establishment in a restoration context appears resistant to recent changes in climate. Our research highlights the importance of deriving multiple metrics of climate change to understand community responses. In particular, rainfall reliability should be a focus of future research to determine its broader significance to seedling establishment in forests subject to a drying climate.1. 全球气候变化预计将因气温升高与降雨量减少，加剧干旱地区干旱的发生频率与强度。严重干旱可引发树木死亡，并推动植被发生持久性变化。 2. 迄今为止，多数实证研究聚焦于干旱诱导的成年树木死亡现象，但若要理解全球森林对预测中气候变化的恢复力，我们还需开展类似研究以探明干旱对幼苗定植的影响。 3. 与1900年至1974年的平均降雨量相比，1975年至2011年间澳大利亚西南部的Eucalyptus marginata（jarrah，红柳桉）林生态区的年平均降雨量下降了17%。已有文献记录了红柳桉林内干旱诱导的成年树木死亡现象，但学界对其恢复能力的认知仍较为有限，因此难以预测其产生持久性影响的可能性。 4. 1992年至2010年这19年间，存在铝土矿开采后红柳桉林修复的长期记录。该记录包含了此气候波动期内，1938个样地中587个物种的修复矿区幼苗定植年度数据。 5. 我们构建了结构方程模型（structural equation model, SEM），以区分气候、修复措施及其交互作用对三类响应变量的相对影响，其中包括修复植物群落的物种丰富度。气候波动对物种丰富度存在显著负向影响，但其效应量相对较小，仅为修复措施波动效应量的一半以下。我们认为，这一现象部分源于降雨量的稳定性（即近期与历史气候间，降雨量的变异系数、季节性或均匀度均未发生变化），尽管湿季降雨量的绝对量相差达三倍之多。重要的是，并无证据表明修复措施会因气候波动性加剧的交互效应而受到削弱。 6. 综合分析表明：在修复场景下，红柳桉林的幼苗定植似乎能够抵御近期的气候变化。本研究强调了获取多维度气候变化指标以理解群落响应的重要性。尤为关键的是，未来研究应聚焦于降雨量稳定性，以探明其对处于气候干旱化进程中的森林幼苗定植的更广泛意义。