Dry-wet abrupt alternation (DWAA) disasters are not merely atmospheric phenomena but are also significantly influenced by land-atmosphere interactions. Understanding the propagation of meteorological dry-wet abrupt alternation (M-DWAA) to terrestrial systems (T-DWAA) is crucial for assessing compound hydroclimatic risks, yet a global assessment is lacking. This study quantifies the global pattern and trends in the coincidence between MDWAA and T-DWAA events (1982-2015) using Event Coincidence Analysis. We find that approximately 17.4% of M-DWAA events coincide with a T-DWAA within the subsequent season globally, with strong spatial heterogeneity. High coincidence rates (>20%) are concentrated in humid, well-vegetated regions characterized by strong land-atmosphere coupling (e.g., northern Eurasia, monsoon regions), while arid and mountainous regions show lower rates (<20%). Critically, the alternation velocity of M-DWAA events, rather than their mere frequency or intensity, emerges as the dominant factor influencing this propagation likelihood. Furthermore, the linkage is strengthening in regions like the Atlantic Americas and northern Eurasia, primarily associated with increasing M-DWAA velocity, thereby escalating compound hydrological risks. Conversely, a weakening linkage (decreasing coincidence) in some dry regions correlates with reduced meteorological alternation velocity. These findings highlight the pivotal role of the transition velocity of meteorological DWAA in lagged coincidence with terrestrial extremes and provide a global framework to guide future mechanistic research and targeted adaptation strategies.

Traditional triangle or trapezoid methods in surface temperature (Ts)–vegetation index (VI) space for evapotranspiration (ET) estimation generally require three to four predefined feature points under assumed extremes, i.e., bare soil and fully vegetated surfaces in completely dry or saturated states. However, these stringent assumptions make such points difficult to observe or validate, potentially leading to distorted feature spaces. Moreover, as these methods are typically constructed at satellite overpass times, they introduce temporal limitations and require upscaling, further increasing uncertainty in estimated ET. To address this, we developed a two-feature-point (T2) method to estimate ET directly at a daily scale, without relying on vegetation indices or temporal upscaling, while remaining fully verifiable at every step. Compared with flux site observations, the estimated feature-point surface temperatures agree satisfactorily with measurements under both wet conditions (R2 = 0.74, RMSE= 6.92 ℃, BIAS= –5.65 ℃) and dry conditions (R2 = 0.81, RMSE = 3.78 ℃, BIAS = 2.74 ℃). By incorporating a reconstructed air temperature process, the model reduces distorted feature spaces to 8.64% and yields robust air temperature estimates (dry: R2 = 1.00, RMSE = 0.26 ℃; wet: R2 = 0.93, RMSE = 5.13 ℃). Global validation at 117 flux towers (53,710 site-days) shows that daily latent heat flux estimated by T2 model attains mean R2 = 0.65 and RMSE = 26.81 W/m2. Overall, the T2 model provides a structurally simple, testable framework for estimating daily ET without temporal upscaling, offering new insights into the temperature–evapotranspiration relationship despite potential overestimation under some conditions.

Study regionChina.Study focusMeteorological drought indices, such as the Standardized Precipitation Evapotranspiration Index (SPEI) have been widely used as a proxy for soil moisture to study agricultural droughts. However, the SPEI and soil moisture are driven by different factors, which may lead to inconsistencies or even contradictions under certain circumstances, and the utility of the SPEI to represent soil moisture remains poorly understood. This study systematically evaluated the performance of the SPEI to detect agricultural droughts using two recently developed high-resolution datasets over China.New hydrological insights for the regionThe SPEI and soil moisture exhibit a strong correlation across eastern China, though notable seasonal variability is present. For near-surface soil, the correlation time scale is typically within three months, which increases with soil depth and towards the northwest. The time scale is shorter in autumn and longer in spring. Over 60% of China shows consistent trends between the SPEI and soil moisture. However, the SPEI may fail to detect agricultural droughts in the northwest and may incorrectly identify droughts in the southeast. Consequently, the usability of the SPEI for detecting agricultural droughts varies by region and season, highlighting the need for future studies to evaluate its utility prior to its direct application.

Understanding the consecutive occurrence of hydroclimate whiplash (HCW) extremes is critical for assessing global climate risks. However, most studies have focused on individual points or pixel scales, which fails to reveal the joint evolution of the extreme events over space and time, leading to potential underestimation. Here, we investigate the spatiotemporal evolution of contiguous HCW extremes globally from 1982 to 2015 from a 3D perspective (latitude x longitude x time). Results show that global HCW extremes have been averagely underestimated by 20% in frequency and nearly half in affected areas using pixel-level analysis compared to the 3D scanning approach. The contiguous HCW extremes are dominated by drying events with higher transition velocity and frequency, while wet-dominant HCW extremes have significantly increased in frequency from 1982 to 2015. Spatially, monsoon regions in the Western Pacific exhibit the highest comprehensive magnitude of contiguous HCW extremes with high transition velocity and frequency. Increased precipitation plays a crucial role in the changes of global contiguous HCW extremes, while variations in vegetation coverage also significantly contribute to the intensification of these extremes.

Marine heatwave (MHW) can increase heat exchange between the land and the ocean, which may further develop into a consecutive marine and terrestrial heatwave (CMTHW). Despite their significance, the feedback mechanisms underlying these compound events remain inadequately understood. This study provides a comprehensive analysis of the interactions between terrestrial and marine heatwaves across China's coastal regions, leveraging multiple temperature datasets. Our findings reveal a marked increase in both the frequency and spatial extent of CMTHWs over the past four decades. Notably, longer lasting and more intense MHWs are more likely to trigger subsequent terrestrial heatwaves (THWs), indicating that CMTHWs are associated with more severe and prolonged MHWs compared to standalone MHWs. Atmospheric processes cause additional land surface warming relative to the ocean. Specifically, during CMTHWs, land surface latent heat flux anomalies are significantly larger than those over the ocean, highlighting the critical role of atmospheric feedback. These findings underscore the need for further investigation into the mechanisms linking marine and terrestrial heatwaves and the broader implications for coastal climate dynamics and ecosystem resilience.

Understanding the transition from meteorological to hydrological drought is essential for accurately predicting hydrological droughts. However, the factors driving this transition are intricate, and a comprehensive understanding of how direct human activities influence this shift in drought is lacking. In this study, we initially explored the spatiotemporal correlation between the occurrence of meteorological and hydrological droughts. Subsequently, we formulated multiple hydrological replenishment scenarios using the soil and water assessment tool (SWAT) model to assess the environmental impact of the transition from meteorological to hydrological droughts. The Xijiang River Basin (XRB), the primary tributary of the Pearl River basin, was selected as the study area. Our results identified 91 meteorological droughts in the XRB, and only 66 hydrological droughts from 1978 to 2018. The transition rate from meteorological to hydrological drought demonstrated large spatial variability, with a basin-average rate of 56% and the lowest transition rate of 45% in the headstream The transition from meteorological to hydrological drought was mostly rapid between November and December (similar to 2 months), but can be prolonged during spring (about 3-5 months) and winter (around 7-9 months). Additionally, analysis of regeneration scenarios indicated that human activity has mitigated drought severity over recent decades. The primary driver affecting drought duration and frequency during the transition from meteorological to hydrological droughts shows conspicuous spatial disparities in densely populated areas. Although human activity significantly contributes to drought duration and severity compared to climate change during the transition in the headstream, its effects are more pronounced downstream in terms of drought duration and frequency. One plausible explanation is that increased water consumption downstream has considerably prolonged drought progression, whereas water management has counteractive effects on drought progression due to climate change in the headstream. Our findings offer valuable insights into the transition process from meteorological to hydrological droughts in the presence of extensive human activities.

CONTEXT: Improving the productivity of smallholder farmers in sub-Saharan Africa is a key component in reducing poverty and increasing food security as crop production is a significant source of livelihood for the majority of the population. Still, crop yields show a huge variability in smallholder farming systems whose productivity is poorly measured and understood. OBJECTIVE: In this work, we estimate maize ( Zea Mays) ) yield gap in Southern Malawi (Phalombe district) and assess drivers of productivity gap under different socio-economic and biophysical contexts. METHODS: We use a mixed-method approach which integrates multi-source datasets (including primary ground truth data we collected in the maize growing season 2019-2020 and secondary remote-sensing data), empirical and process-based crop-growth models (AquaCrop) to calculate the water-limited yield gap. In addition, we analyse the relationship between the relative yield (defined as the actual yield observed at the farmers' plots normalised by the AquaCrop simulated water-limited potential yield) and possible socio-economic drivers which we collected through surveys administered to households iin the same season 2019-2020. RESULTSAND CONCLUSIONS: We obtained a water-limited potential yield for the maize hybrid SC649 of 9.5 t/ ha during the season 2019-2020 in the Malawian trial site. The observed actual yield at the households in the season 2019-2020 varied from 0.8 to 10.9 t/ha. The estimate of the yield gap ranged between 15% and 85% thus showing a large variability due to the high resolution, but low accuracy of the empirical model. Results suggest that with higher income and increased fertiliser application there is potential to increase the relative yield and that the marginal increase is spatially differentiated. SIGNIFICANCE: Our spatially-explicit approach to yield-gap analysis is valuable in identifying high-productive areas and differentiated policy interventions aimed at closing the yield and income gaps for smallholder farmers.

This study examined the transport network of global marine dry bulk carriers for agricultural trade during the period from 2018 to 2021. Firstly, the resilience of agricultural trade network is noteworthy throughout the COVID-19 pandemic. Agricultural trade initially plunged by 10.15% from 2019 to 2020 and bounced by a remarkable 11.45% in 2021, ultimately restoring trade volumes to the average level observed in the pre-pandemic year of 2019. However, the ports in Brazil and Argentina displayed less resilience in their agricultural trade with a continued decline in agricultural trade quantities in 2021. Additionally, the outbound trips increased in Ukraine, Canada, and Russia and decreased in Brazil and Argentina, leading to a more tightly knit agricultural network since 2020. Overall, this study provided evidence in comprehensively assessing the capacity and resilience of global food supply chains, especially in the context of constantly evolving circumstances and challenges.

Climate change has triggered more frequent drought occurrence, which can have devastating impacts on the ecosystem functions. Studies on vegetation behavior during droughts have mainly focused on arid/semi-arid regions, yet the ecological and vegetation responses during drought in humid regions remain unclear. Here we systematically evaluated the evolution of the historic drought occurred in the humid Pearl River Basin in 2021 and quantified the vegetation responses using a multitude of vegetation indicators. Analyses showed that the East River Basin and North River Basin were the most severely hit by drought, which enhanced surface temperature and evapotranspiration, and caused soil moisture and terrestrial water storage deficits. Mean vegetation response time was shorter based on solar-induced fluorescence (SIF, 2.7 months) and the water use efficiency (WUE, 2.8 months), followed by the gross primary productivity (GPP, 3.2 months), and longer using the normalized difference vegetation index (NDVI, 4.2 months) and the vegetation optical depth (VOD, 5.0 months). By contrast, over 90% of the ecosystems recovered to their normal states within 3 months using all indicators. The results implied that the NDVI lacks sensitivity to changes in water stress in humid regions, and revealed that vegetation in humid regions may respond slowly and recover rapidly under droughts, which may relate to the water availability that enhances the resistance and resilience of the ecosystems.

Water resources’ management at a high spatial and temporal resolution calls for data support at the relevant scales, which has long been hindered by the availability of high-resolution data [...]

Global warming alters the inherent variability of climate variables and induces more frequent extremes of hydroclimatic compound events with substantial impacts. The concurrence of hydrological drought and saltwater intrusion in the estuary can have devastating effects on agricultural production, water security, and ecological health. In this study, the compound drought and saltwater intrusion extreme indices, including the Blend Drought and Saltwater intrusion Index (BDSI), the Compound drought and Saltwater intrusion Index (CDSI) and the Standardized Drought and Saltwater intrusion Index (SDSI), were constructed to monitor compound extremes in the critical zone of Pearl River Estuary during the dry season from 1999 to 2016. The results reveal that the three compound indices can effectively depict different aspects of compound drought and saltwater intrusion extreme events in the critical zone of Pearl River Estuary. Among the three indices, the SDSI performed best in depicting compound drought‐saltwater intrusion events, while the BDSI outperformed the other indices in depicting high value of Standardized Runoff Index (SRI) and Standardized Saltwater intrusion Index (SSI), and the CDSI outperformed the other indices in depicting low values of SRI and SSI. Using those developed compound indices, we found that the frequency of abnormal and moderate compound drought‐saltwater intrusion events increased from 9.26% to 12.03% during the period of 1999–2016, while severe, extreme and exceptional compound drought‐saltwater intrusion events were rare.

Seasonal variation of vegetation profoundly affects the water cycle. However, the seasonal divergence of evapotranspiration (ET) sensitivity in response to vegetation variations has not been fully understood. Here we derived an analytical solution to examine the impact of seasonal vegetation changes on ET with an extended Budyko framework based on an improved ET algorithm with improved water balance constraints. Results reveal a clear seasonal divergence of ET sensitivity to vegetation coverage changes across climate regimes and biomes. Generally, the high ET sensitivity to vegetation coverage has a clear north-south shift trajectory from spring to winter. For moderate-humid regions (0.7 < aridity index < 1.0), vegetation exhibits higher importance in altering ET in March-September. While for moderate-dry regions (1.0 < aridity index < 1.4), the sensitivity of ET to vegetation changes is the highest in September-November. Moreover, the spatial-temporal pattern of ET sensitivity to seasonal vegetation changes is different between short vegetation cover and forest. Additionally, negative ET sensitivity to vegetation coverage changes was discovered in regions with seasonal precipitation of less than 500 mm and sparse vegetation coverage (predominant land cover types of grassland, scrubland, and savannas). In summary, our study provides an analytical solution to estimate ET sensitivity to seasonal vegetation changes within the extended Budyko framework. The results highlight the difference in hydrological response to vegetation dynamics across seasons and vegetation types.

Compound Drought and Saltwater intrusion Events (CDSEs) refer to hydrologic drought and saltwater intrusion occurring simultaneously or consecutively in estuaries, and exacerbate the negative impacts resulting from an individual extreme event. CDSEs have been drawing increasing attention due to their potential adverse impacts on water resources, crop production, and food security. A new Standardized compound Drought and Saltwater intrusion Index (SDSI) was developed in this study to systematically detect changes in the severity of CDSEs in six estuaries (Little Back, Ebro, Rhine, Orange, Pearl River and Murray). The results illustrated that (1) compared to the Standardized Runoff Index (SRI), SDSI effectively characterizes and quantifies the occurrences and severity of CDSEs in major river estuaries worldwide. (2) Temporally, the SDSI trend varied across estuaries. Specifically, a decreasing trend was observed in the Little Back, Ebro, and Orange estuaries, with corresponding Zs values of -2.43, -3.63, and -3.23. (3) Spatially, moderate CDSEs occurred more frequently among different estuaries, and their frequency, duration and severity varied in different estuaries. Notably, Ebro, Rhine and Murray River estuaries had the highest probability of CDSEs, nearing 60%. Among them, the Murray Estuary had the longest average duration, spanning 7.68 months, and the highest severity was 5.94. (4) According to the contributions analysis, saltwater intrusion plays a dominant role in influencing SDSI severity, accounting for a substantial percentage (54%-95.30%) compared to runoff. Notably, the Orange Estuary experienced the greatest impact from saltwater intrusion (81.54%-95.30%), while the Murray Estuary had relatively equal contributions from hydrological drought and saltwater intrusion.

Global warming is anticipated to largely impact drought dynamics. Therefore, a comprehensive projection of global drought under various future warming levels is critical for water resources management. This study investigated the changes in the severity, intensity, area, and duration of droughts at 1.5 degrees C, 2 degrees C, 3 degrees C, and 4 degrees C of global warming above pre-industrial levels based on the latest simulations under climate change scenarios from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The Standardized Precipitation Evapotranspiration Index (SPEI) at 3-month time scale was adopted to characterize seasonal droughts. The results show that: 1) Global land is projected to experience a significant nonlinear increasing trend of drought severity, intensity, areal extent, and duration during the 21st century, and drought characteristics have the highest growth rate under the 3 degrees C warming scenario, followed by the 4 degrees C, 2 degrees C, and 1.5 degrees C warming scenarios. 2) Under the 1.5 degrees C and 2 degrees C warming levels, drought characteristics in most land areas change slightly compared to the baseline period of 1995-2014. 3) Hotspots defined as robust increases in drought severity appear in the 3 degrees C and 4 degrees C warming world, including northern South America, Europe, northern and southern Africa, western and central Asia, and central Australia, whereas eastern North America and the Russian Arctic exhibit substantial reduction in drought severity under different global warming levels. 4) Global warming is projected to change the timing of droughts, leading to more frequent summer droughts in Europe. Overall, the findings of this study provide a scientific basis for articulating future global drought adaption and highlight the need for mitigation measures to limit greenhouse gas emissions to achieve the 2 degrees C target of the Paris Agreement.

The phenomenon of sudden drought after pluvial (or heavy pluvial after drought) is known as the Dry–Wet Abrupt Alternation (DWAA). This kind of disaster has been evaluated globally based on various DWAA indices. However, few validations have been performed to compare the various DWAA indices. Given that directly validating the identified DWAA events remains difficult because of the limited historical records of DWAA events, in this study, we developed a simple weighting analysis method to cross‐validate the detected DWAA events using three widely‐used indices. The weighting analysis method can be used to estimate the effectiveness of the DWAA indices by calculating the ratio of the synchronous DWAA events (the DWAA events could be synchronously detected by different indices) to the total number of the detected DWAA events. Results suggest that the standard weighted average of precipitation (SWAP) index has better performance for detecting DWAA events in the Pearl River Basin (PRB) compared to other indices. The soil moisture synchronization, meteorological concentration and real DWAA records further testify that the SWAP is more capable of detecting the DWAA events in the PRB. Based on the validation, the SWAP‐derived DWAA events show that both dry–wet (D–W) events and wet–dry events are severe in the upstream sub‐basins in the PRB, while D–W events are more frequent in the downstream PRB. Our study provides a simple but effective method to validate the detected DWAA. The finding underscores the need to validate DWAA detection generated from multiple indicators before further research, and it could help develop adaptation strategies to improve resilience to combined extremes under changing climate.

Modern problems in agricultural management require non-traditional solutions, one of which is by utilizing domain adaptive machine learning models for crop yield prediction which are able to perform reliably in different temporal or spatial domains. However, most studies have focused on the application of domain adaptation to classification tasks such as crop type identification, while the application to regression tasks such as crop yield prediction have been limited. In this study, we explore the generalisability and transferability of ordinary Deep Neural Network (DNN) and domain adaptive neural network models created using three domain adaptation algorithms, namely Discriminative Adversarial Neural Network (DANN), Kullback-Leibler Importance Estimation Procedure (KLIEP), and Regular Transfer Neural Network (RTNN). These three algorithms represent feature-based, instance-based, and parameter-based domain adaptations, respectively. Maize yield records, weather variables, and remotely sensed features from 11 states in the US corn belt acquired in 2006-2020 were compiled and segregated into classes according to temporal (year) and spatial characteristics (annual growing degree days [GDD], vapor pressure deficit [VPD], soil organic content [SOC], and green chlorophyll vegetation index/GCI). We found that models trained using datasets from temperate regions with medium-high GDD and moderate VPD perform well whereas SOC does not significantly affect the generalisability. It is not advisable to train models with datasets constrained by GCI as this feature correlates significantly with the maize yield, and adaptation between two domains that rarely intercept will not work well. We also demonstrate that KullbackLeibler divergence computed using features from source and target domains can be used to justify the feasibility of domain adaptation. Based on the divergence, a model trained in the US (or another region with sufficient data) is expected to work reliably in other regions through domain adaptation, especially feature-based adaptation.

Remote sensing has proven to be an invaluable and effective tool for mapping urban areas. However, further efforts are required to fully utilize remote sensing data in mapping the Chinese urban villages (Chengzhongcun), which are characterized by high population density, irregular shapes and a limited view of the sky. In this study, we integrated deep neural network architectures with GaoFen-2 satellite images at a spatial resolution of 1 metre to investigate the feasibility of using FCN, Unet and ResUnet models for monitoring two categories of urban villages. The results indicate that all three deep learning algorithms demonstrate superiority in identifying and classifying urban villages, surpassing outlier identification. The urban village patches in the study areas adhere to the Zipf’s law rule. Particularly, the ResUnet model outperforms FCN and Unet, achieving an overall classification accuracy of over 91% for urban villages. This study suggests that ResUnet is superior to FCN and Unet as it effectively avoids fragmentation and overfitting of individual irregular buildings within urban areas, yielding faster and more accurate results. Overall, this study underscores the significance of the Gaofen-2 dataset as an influential data source for analysing intra-urban structures. Additionally, the implementation of the ResUnet learning algorithm proves to be exceptionally valuable in extracting information across various scales and accurately capturing the intricate nature of irregular urban village patches, even when using satellite images with metre-scale resolution.

Satellite data provide high potential for estimating crop yield, which is crucial to understanding determinants of yield gaps and therefore improving food production, particularly in sub-Saharan Africa (SSA) regions. However, accurate assessment of crop yield and its spatial variation is challenging in SSA because of small field sizes, widespread intercropping practices, and inadequate field observations. This study aimed to firstly evaluate the potential of satellite data in estimating maize yield in intercropped smallholder fields and secondly assess how factors such as satellite data spatial and temporal resolution, within-field variability, field size, harvest index and intercropping practices affect model performance. Having collected in situ data (field size, yield, intercrops occurrence, harvest index, and leaf area index), statistical models were developed to predict yield from multisource satellite data (i.e., Sentinel-2 and PlanetScope). Model accuracy and residuals were assessed against the above factors. Among 150 investigated fields, our study found that nearly half were intercropped with legumes, with an average plot size of 0.17 ha. Despite mixed pixels resulting from intercrops, the model based on the Sentinel-2 red-edge vegetation index (VI) could estimate maize yield with moderate accuracy (R2 = 0.51, nRMSE = 19.95%), while higher spatial resolution satellite data (e.g., PlanetScope 3 m) only showed a marginal improvement in performance (R2 = 0.52, nRMSE = 19.95%). Seasonal peak VI values provided better accuracy than seasonal mean/median VI, suggesting peak VI values may capture the signal of the dominant upper maize foliage layer and may be less impacted by understory intercrop effects. Still, intercropping practice reduces model accuracy, as the model residuals are lower in fields with pure maize (1 t/ha) compared to intercropped fields (1.3 t/ha). This study provides a reference for operational maize yield estimation in intercropped smallholder fields, using free satellite data in Southern Malawi. It also highlights the difficulties of estimating yield in intercropped fields using satellite imagery, and stresses the importance of sufficient satellite observations for monitoring intercropping practices in SSA.

Predicting within-field crop yield early in the season can help address crop production challenges to improve farmers' economic return. While yield prediction with remote sensing has been a research aim for years, it is only recently that observations with the suited spatial and temporal resolutions have become accessible to improve crop yield predictions. Here we developed a yield prediction framework that integrates daily high-resolution (3 m) CubeSat imagery into the APSIM crop model. The approach trains a regression model that correlates simulated yield to simulated leaf area index (LAI) from APSIM. That relationship is then employed to determine the optimum date at which the regression best predicts yield from the LAI. Additionally, our approach can forecast crop yield by utilizing a particle filter to assimilate CubeSat-based LAI in the model APSIM to generate yield maps at 3 m several weeks before the optimum regression date. Our method was evaluated for a rainfed site located in the US Corn belt, using a collection of spatially varying yield data. The proposed approach does not need in situ data to rain the regression, with outcomes reporting that even with a single assimilation step, accurate yield predictions were provided up to 21 days before the optimum regression date. The spatial variability of crop yield was reproduced fairly well, with a good correlation against in situ measurements (R-2 = 0.73 and RMSE = 1.69), demonstrating that high-resolution yield predictions early in the season have great potential to meet and improve upon digital agricultural goals.

Global urbanization significantly impacts the thermal environment in urban areas, yet urban heat island (UHI) and urban heat wave (UHW) studies at the mega-region scale have been rare, and the impact study of urbanization is still lacking. In this study, the MODIS land surface temperature (LST) product was used to depict the UHI and UHW in nine mega-regions globally between 2003 and 2020. The absolute and percentile-based UHW thresholds were adopted for both daily and three-day windows to analyze heat wave frequency, and UHW magnitude as well as frequency were compared with UHI variability. Results showed that a 10% increase in urban built-up density led to a 0.20 °C to 0.95 °C increase in LST, a 0.59% to 7.17% increase in hot day frequency, as well as a 0.08% to 0.95% increase in heat wave number. Meanwhile, a 1 °C increase in UHI intensity (the LST differences between the built-up and Non-built-up areas) led to a 2.04% to 92.15% increase in hot day frequency, where daytime LST exceeds 35 °C and nighttime LST exceeds 25 °C, as well as a 3.30% to 33.67% increase in heat wave number, which is defined as at least three consecutive days when daily maximum temperature exceeds the climatological threshold. In addition, the increasing rates of UHW magnitudes were much faster than the expansion rates of built-up areas. In the mega-regions of Boston, Tokyo, São Paulo, and Mexico City in particular, the increasing rates of UHW hotspot magnitudes were over 2 times larger than those of built-up areas. This indicated that the high temperature extremes, represented by the increase in UHW frequency and magnitudes, were concurrent with an increase in UHI under the context of climate change. This study may be beneficial for future research of the underlying physical mechanisms on urban heat environment at the mega-region scale.