Compared to individual hot days/nights, compound hot extremes that combine daytime and nighttime heat are more impactful. However, past and future changes in compound hot extremes as well as their underlying drivers and societal impacts remain poorly understood. Here we show that during 1960-2012, significant increases in Northern Hemisphere average frequency (~1.03 days decade −1 ) and intensity (~0.28°C decade −1 ) of summertime compound hot extremes arise primarily from summer-mean warming. The forcing of rising greenhouse gases (GHGs) is robustly detected and largely accounts for observed trends. Observationally-constrained projections suggest an approximate eightfold increase in hemispheric-average frequency and a threefold growth in intensity of summertime compound hot extremes by 2100 (relative to 2012), given uncurbed GHG emissions. Accordingly, endof-century population exposure to compound hot extremes is projected to be four to eight times the 2010s level, dependent on demographic and climate scenarios.

Weather and climate extremes are critical drivers for deadly and costly natural disasters (IFRC, 2020; UN-DRR, 2020). Understanding their changes and causes have been staying high on the agenda of climate science and risk management sectors (Chen, Moufouma-Okia, et al., 2018). There is rising awareness that the impact of spatially and/or temporally correlated events tends to be disproportionately larger than that of singular hazards as well as the sum of them (Zscheischler et al., 2018). A new paradigm that moves beyond isolated extremes is therefore strongly advocated (Leonard et al., 2014). Compound events were for first time defined in a special report from the Intergovernmental Panel on Climate Change (SREX, Field et al., 2012) as multiple extreme or nonextreme events occurring (1) simultaneously at the same place; (2) concurrently across multiple regions; or (3) in rapid sequence, in the same location. The definition was further refined (Leonard et al., 2014; Zscheischler et al., 2018), with a typology recently proposed (AghaKouchak et al., 2018; Zscheischler et al., 2020). As this field burgeons, a growing body of studies has explored statistical frameworks, observed and projected changes, and attribution of diverse compound events, such as drought-heatwave (Alizadeh et al., 2020; Hao et al., 2013), costal floods combining heavy precipitation, high water level and storm surge (Wahl et al., 2015), and concurrent heatwaves across global breadbaskets

Urban areas are experiencing strongly increasing hot extremes. However, these events have seldom been the focus of traditional detection and attribution analysis designed for regional-to-global changes. Here, we show that compound (day-night sustained) hot extremes are more dangerous than solely daytime or nighttime heat, especially to female and older urban residents. Urban compound hot extremes across Eastern China have increased by 1.76 days decade -1 in 1961-2014, with fingerprints of urban expansion and anthropogenic emissions detected by a stepwise detection and attribution method. Their attributable fractions are estimated as 0.51 (urbanization), 1.63 (greenhouse gases) and -0.54 (other anthropogenic forcings) days decade -1 . Future emissions and urbanization would make these compound events two-to-five times more frequent (2090s vs. 2010s), leading to a three-to-sixfold growth in urban population exposure. Our findings call for tailored adaptation planning against rapidly growing health threats from compound heat in cities.