In the context of human-induced warming climate, the atmosphere is expected to hold a greater amount of water vapor, leading to heavier precipitation on a global scale. However, the extent to which changes in extreme rainfall can be attributed to human influences varies at regional scales. Here we conduct attribution analyses on 40 extreme precipitation events in different seasons during 1998–2018 over the Pearl River Delta (PRD), by using the Weather Research and Forecasting (WRF) model and applying the pseudo global warming (PGW) method. The model was integrated with the factual and counterfactual conditions separately, with the latter derived from differences between the Coupled Model Intercomparison Project Phase 5 (CMIP5) historical and historical-natural runs. By comparing parallel experiments, extreme daily rainfall (>95th percentile) in PRD enhanced by 8%–9.5% for 0.9–1.1 K near-surface warming (nearly Clausius-Clapeyron, or CC scaling) in the May-to-September (MJJAS) and 12.4% at most for 0.6–0.8 K warming (∼2 × CC rate) in non-MJJAS seasons; the probability of those extremes increased by 10%–30% during MJJAS (20%–40% in other seasons). While moisture-related thermodynamic effects play a similar role in modulating rainfall, the wind circulation-related dynamic effects act differently in different seasons. Changes in MJJAS extremes are related to stronger low-level southerly winds, while non-MJJAS rainfall is exacerbated by strengthened low-level wind convergence and updrafts. Moisture budget analysis suggests that thermodynamic effects associated with the increased moisture amount account for the mean rainfall increase, whereas dynamic effects related to wind circulation changes are responsible for extreme precipitation, regardless of seasons.

中文翻译：

---

基于拟全球变暖的华南珠三角地区极端降水特征归因于人为影响

在人为引起的气候变暖的背景下，预计大气中将含有更多的水蒸气，从而导致全球范围内的降水量增加。然而，极端降雨量的变化在多大程度上归因于人类影响，在区域范围内有所不同。本文利用天气研究与预报（WRF）模型，应用伪全球变暖（PGW）方法，对1998-2018年珠江三角洲（PRD）不同季节的40次极端降水事件进行归因分析。该模型分别与事实和反事实条件相结合，后者源自耦合模型比对项目第五阶段（CMIP5）历史和历史自然运行之间的差异。通过比较平行实验，5 月至 9 月，珠江三角洲的极端日降雨量（>95%）在近地表变暖 0.9-1.1 K（接近克劳修斯-克拉佩龙，或 CC 尺度）的情况下增加了 8%-9.5% (MJJAS ），非 MJJAS 季节升温 0.6–0.8 K（∼2 × CC 率）时最多 12.4%；在 MJJAS 期间，这些极端事件的概率增加了 10%–30%（其他季节为 20%–40%）。虽然与湿度相关的热力学效应在调节降雨方面发挥着类似的作用，但与风环流相关的动态效应在不同季节的作用有所不同。MJJAS 极端事件的变化与较强的低层南风有关，而非 MJJAS 降雨则因低层风辐合和上升气流的加强而加剧。水分收支分析表明，与水分含量增加相关的热力学效应导致了平均降雨量的增加，而与风环流变化相关的动态效应则导致了极端降水，无论季节如何。