Abstract Drylands are the homes to over one-third of the world’s population, and are vulnerable to anthropogenic climate change. Based on climate projections, recent studies reported a substantial expansion of global drylands in the coming decades and attributed that expansion to future warming. However, the expansion of drylands contradicts a widespread vegetation greening and a slight runoff increase in dryland ecosystems in the same climate models. Here we re-examine changes in global drylands for the coming century and at two future warming targets (1.5 °C and 2 °C warming relative to the preindustrial level) based on outputs of climate models who participated in the Fifth Coupled Model Intercomparison Project (CMIP5). In addition to aridity index (AI) that has been widely used to measure the atmospheric aridity, we also assess changes in drylands from the hydrologic and agro-ecological perspectives, using runoff (Q) and leaf area index (LAI) as indicators, respectively. Our results show that when the impact of elevated atmospheric CO2 concentration ([CO2]) on vegetation water consumption is considered in the estimation of potential evapotranspiration (EP) and AI, the expansion of atmospheric drylands is at a much slower rate (~0.16% per decade under RCP4.5 and ~0.30% per decade under RCP8.5) than previously reported. Moreover, the additional 0.5 °C warming does not lead to an evident further expansion of atmospheric drylands. In terms of hydrologic and agro-ecological drylands, both of them show significant shrinks over the coming decades, suggesting reduced hydrologic and agro-ecological aridity in the region. Finally, contrasting with previous perceptions, our results demonstrate that warming only plays a minor role in altering global drylands from all three perspectives. Increases in net radiation are primarily responsible for the expansion of atmospheric drylands, and increases in P and [CO2]-induced increases in vegetation water use efficiency are the key drivers of changes in hydrologic and agro-ecological drylands.

中文翻译：

---

气候预测中未来变暖下全球旱地变化的多维评估

摘要 旱地是世界三分之一以上人口的家园，并且容易受到人为气候变化的影响。根据气候预测，最近的研究报告称，未来几十年全球旱地将大幅扩张，并将这种扩张归因于未来的变暖。然而，旱地的扩张与同一气候模型中旱地生态系统的广泛植被绿化和径流略有增加相矛盾。在这里，我们根据参与第五次耦合模型比对项目的气候模型的输出，重新审视下个世纪全球旱地的变化以及两个未来升温目标（相对于工业化前水平升温 1.5 °C 和 2 °C）（ CMIP5)。除了被广泛用于测量大气干旱的干旱指数（AI）外，我们还分别使用径流 (Q) 和叶面积指数 (LAI) 作为指标，从水文和农业生态的角度评估旱地的变化。我们的结果表明，当在估算潜在蒸散量 (EP) 和 AI 时考虑大气 CO2 浓度升高 ([CO2]) 对植被耗水量的影响时，大气旱地的扩张速度要慢得多（~0.16% RCP4.5 下每十年和 RCP8.5 下每十年约 0.30%）比以前报告的要多。此外，额外的 0.5 °C 变暖不会导致大气旱地明显进一步扩大。就水文和农业生态旱地而言，它们在未来几十年都显示出显着萎缩，表明该地区的水文和农业生态干旱减少。最后，与之前的看法相反，我们的结果表明，从所有三个角度来看，变暖在改变全球旱地方面只发挥了次要作用。净辐射的增加是大气旱地扩张的主要原因，而磷和 [CO2] 引起的植被水分利用效率的增加是水文和农业生态旱地变化的关键驱动因素。