Compound drought and heatwave (CDHW) events have received considerable attention in recent years due to their devastating effects on human society and ecosystem. In this study, we systematically investigated the changes of CDHW events in multi‐spatiotemporal scales for historical period (1951–2014) and four future scenarios (2020–2100) (SSP1‐2.6, SSP2‐4.5, SSP3‐7.0, and SSP5‐8.5) over global land by using Coupled Model Intercomparison Project Phase 6 (CMIP6) models. The responses of the CDHW events to the changes of maximum air temperature and the climatic water balance variable are also examined. The results show that the multi‐model ensembles project a significant increasing trend in CDHW characteristics over almost all global lands under SSP2‐4.5, SSP3‐7.0, and SSP5‐8.5, especially across northern North‐America, Caribbean, Mediterranean and Russian‐Arctic, there is a stronger increasing trend. A significantly increasing CDHW risk will occur across most global land for the medium to long term future without aggressive adaptation and mitigation strategies. The results of path analysis suggest that temperature is the dominant factor influencing CDHW events. Additionally, higher sensitivity of CDHW events to global warming will occur in the future. Particularly, each 1°C global warming increases the duration of the CDHW events by 3 days in the historical period, but by about 10 days in the future period. Overall, this study improves our understanding in the projection of CDHW events and the impacts of climate drivers to the CDHW events under various future scenarios, which could provide supports about the risk assessment, adaptation and mitigation strategies under climate change.

Drought risk assessment can identify high‐risk areas and bridge the gap between impacts and adaptation. It is thus vital to investigate changes in drought risk and exposed social economy under climate change. Here, the future global meteorological drought risk is projected for the 2021–2100 period under four combined scenarios of Representative Concentration Pathways and Shared Socioeconomic Pathways (SSPs), i.e., SSP1‐2.6, SSP2‐4.5, SSP3‐7.0, and SSP5‐8.5. And then the exposed population and Gross Domestic Product (GDP) under high risk are analyzed. Drought risk will further strengthen in the future under SSP2‐4.5, SSP3‐7.0, and SSP5‐8.5, with large increases distributed in South Asia, the Mediterranean, East Asia, Southeast Asia, and Central America. The strongest increasing trends in drought risk are concentrated in equatorial regions including South Asia, West Africa, and Central America. A large percentage of the population and economy are exposed to high drought risk. The exposed population under the highest risk level is the largest under SSP5‐8.5 and the smallest under SSP3‐7.0 for regions in North America, while being the largest under SSP3‐7.0 for other regions. The exposed GDP under the highest risk level is the highest under SSP5‐8.5 and the lowest under SSP3‐7.0, while the disparity among diverse scenarios is larger for the rich regions than the developing ones.

Compound extremes of lethal heat stress-heavy precipitation events (CHPEs) seriously threaten social and ecological sustainability, while their evolution and effects at the global scale under climate warming remain unclear. Here we develop the global picture of projected changes in CHPEs under various scenarios and investigate their socioeconomic and ecosystem risks combining hazard, exposure, and vulnerability through the composite indicator approach. We find a high percentage of heat stress is followed by heavy precipitation, probably driven by atmospheric conditions. Global average frequency and intensity of CHPEs are projected to increase in the future under high-emission scenarios. Joint return periods of CHPEs are projected to decrease globally, predominantly driven by changes in heat stress extremes. In the long-term future, over half of the population, gross domestic product, and gross primary productivity may face high risk in most regions, with developed regions facing the highest risks under SSP5-8.5 and developing regions facing the highest risks under SSP3-7.0.

Accurate monthly runoff prediction is significant to extreme flood control and water resources management. However, traditional statistical models without multi-variable input may fail to capture runoff changes effectively due to the dual effect of climate change and human activities. Here, we used five multi-input machine learning (ML) models to predict monthly runoff, where multiple global circulation indexes and surface meteorological indexes were selected as explanatory variables by the stepwise regression or copula entropy methods. Moreover, four univariate models were adopted as benchmarks. The multi-input ML models were tested at two typical hydrological stations (i.e., Gaochang and Cuntan) in the Upper Yangtze River. The results indicate that the LSTM_Copula (long short-term memory model combined with copula entropy method) model outperformed other models in both hydrological stations, while the GRU_Step (gate recurrent unit model combined with stepwise regression method) model and the RF_Copula (random forest model combined with copula entropy method) model also showed satisfactory performances. In addition, the ML models with multi-variable input provided better predictability compared with four univariate statistical models, and the MAPE (mean absolute percentage error), RMSE (root mean square error), NSE (Nash–Sutcliffe efficiency coefficient), and R (Pearson’s correlation coefficient) values were improved by 5.10, 4.16, 5.34, and 0.43% for the Gaochang Station, and 10.84, 17.28, 13.68, and 3.55% for the Cuntan Station, suggesting the proposed ML approaches are practically applicable to monthly runoff forecasting in large rivers.

Anthropogenic climate change has significantly exacerbated the frequency and severity of Compound Drought and Heatwave (CDHW) events, increasing risks such as water shortages, wildfires, and heat-related mortality. Previous studies often use a single drought index, such as the Standardized Precipitation Index (SPI) or the Standardized Precipitation Evapotranspiration Index (SPEI), while our study uses both SPI and SPEI to elucidate the effect of different drought indices on the quantification of population exposure to CDHW events. Six General Circulation Models under four future Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5) are considered. Exposure is defined as the product of CDHW Magnitude (°C) and the population in each region (million people), providing a quantitative measure of how CDHW events affect populations. The role of potential evapotranspiration (PET) in CDHW population exposure projections is examined by comparing SPI, which considers only precipitation, with SPEI, which accounts for both precipitation and PET in drought measurements. Results show that after 2050, CDHW Magnitude population exposure diverges significantly across scenarios, with SSP3-7.0 exhibiting the largest increase, reaching 0.72 (SPI) and 1.78 (SPEI) million person-°C by the end of the century. Regions such as Western Africa (WAF), Southeastern Africa, and South Asia (SAS) experience the largest increases in population exposure under SSP3-7.0 with SPEI, reaching 6.93, 6.77, and 5.56 million person-°C, respectively. Additionally, regions like Western & Central Europe, the Mediterranean, WAF, Western Central Africa, Eastern Asia, and SAS display heightened sensitivity to PET, with discrepancies between SPEI and SPI projections exceeding 1 million person-°C. Attribution analysis reveals that climate change, particularly when drought is calculated using PET by SPEI, is the primary factor, followed by interaction change and population change. These findings emphasize the critical role of PET in CDHW projections and the need for region-specific adaptation strategies to manage escalating risks in highly vulnerable areas.