Droughts exert far-reaching societal and economic impacts over the globe. A full recovery of drought-stricken areas depends on so many factors that a deterministic estimation of the timing is difficult. This study explores an alternative way to inform how likely a drought-prone region can recover from meteorological droughts over time. We focus on exploring the drought recovery patterns over Yangtze River Basin (YRB) with the development of a modified joint drought index (MJDI) model based on the principal component analysis (PCA) and vine copula. Regarding the seasonal difference in drought recovery, our results show that during April-August the required precipitation to recover from a drought is around 100–350% above the long-term averaged precipitation, and the probabilities of receiving such substantial excess precipitation exceed 0.4 in most regions. In contrast, less precipitation is required during the winter months, i.e., below 50% of their climatologies (i.e., long-term mean) in large parts of the basin. Moreover, we explored the recovery patterns for each drought category (i.e., slight, moderate, severe and extreme droughts) across the basin. It can be found that precipitation needed for terminating a slight or moderate drought is mostly 50–150% of their climatologies with the recovery probability between 0.5 and 0.7. Our results also reveal that most of the northwest regions are unlikely (probabilities close to zero) to end the extreme droughts in a short time span (e.g., 1 month), where the expected precipitation reaches nearly 200% above normal. We also examined variations in vegetation responses to drought recovery. Generally, the forests need an amount of near-normal precipitation to achieve a full recovery from the droughts, while the grasslands which tend to recover with the recipient of approximately 50% of their long-term averages. These findings could provide proactive information for drought managements to make decisions on replenishing water supplies.

干旱对全球产生深远的社会和经济影响。受旱灾地区的全面恢复取决于许多因素，因此很难确定时间。这项研究探索了另一种方法来告知易干旱地区随着时间的推移如何从气象干旱中恢复。我们着重于探索长江流域（YRB）的干旱恢复模式，并基于主成分分析（PCA）和葡萄树copula开发了改进的联合干旱指数（MJDI）模型。关于干旱恢复的季节性差异，我们的结果表明，在4月至8月期间，从干旱中恢复所需的降水量比长期平均降水量高出约100-350％，并且接受如此大量的过量降水的概率超过0。大多数地区为4。相反，在冬季，该盆地大部分地区所需的降水量较少，即低于其气候的50％（即长期平均值）。此外，我们探索了整个盆地中每种干旱类别（即轻度，中度，重度和极端干旱）的恢复模式。可以发现，终止轻度或中度干旱所需的降水大部分是其气候的50-150％，恢复概率在0.5到0.7之间。我们的研究结果还表明，西北地区的大部分地区（概率接近零）不太可能在短时间内（例如1个月）结束极端干旱，那里的预期降雨量比正常水平高近200％。我们还研究了植被对干旱恢复的响应变化。通常，森林需要一定数量的接近正常的降水量才能从干旱中完全恢复过来，而草地往往在接受者的恢复下达到其长期平均水平的50％。这些发现可以为干旱管理人员提供主动信息，以决定补充水源。