Australia: The Land Where Time Began

A biography of the Australian continent 

Increasing Likelihood of Dry-Hot Extremes on a Continental Scale revealed by Century of Observations

It is shown by this paper, based on more than a century of ground-based observations over the contiguous US, that the frequency of compound dry and hot extremes has increased substantially in the past decades, together with an alarming increase in very rare dry-hot extremes. It is indicated by the results of this study that the area affected by the concurrent extremes has also significantly increased. Further, Alizadeh et al. explored homogeneity (i.e., connectedness) of dry-hot extremes across space. They have shown that dry-hot extremes have enlarged homogeneously over the past 122 years, which points to a spatial propagation of extreme dryness and heat and increased probability of continental-scale compound extremes. Last, they have shown an interesting shift between the main drivers of dry-hot extremes over time. While meteorological drought was the main driver of dry-hot events in the 1930s, the observed warming trend has become the dominant driver in recent decades. A deeper understanding of spatiotemporal variation of compound dry-hot extremes is provided by the results of this study.

Traditional climate risk analyses have focused on the hazardous/anomalous states of 1 variable at a time (Zscheischler & Seneviratni, 2017). E.g., it has been shown by studies that magnitudes of heatwaves, their frequencies, intensities and spatial extents are increasing over many regions (Schӓr et al., 2004), a trend that is projected to continue in a warming climate (Hansen, Ruedy, Sato & Lo, 2010). Moreover, analyses of historical precipitation, streamflow and soil moisture indices have shown an increasing trend of aridity over many regions around the globe (Dai, 2011) and an increasing drought hazard in the 21^st century is pointed to in model simulations. Individually, these extreme events can cause significant adverse impacts, however when they occur concurrently the effects can be even more devastating (Mueller & Seneviratni, 2012; Leonard et al., 2014). E.g., a significant increase in tree mortality can result from the compounding effects of drought and high temperatures, which, in turn, may cascade into other hazards, such as wildfires (Williams et al., 2013). The most damaging stressors to the production of wheat are concurrent drought and heatwaves with grave implications for global food security; they can also jeopardise the reliability of electric grids and a wide range of natural and built systems (Guerreiro, 2018). Between 2011 and 2013, 3 concurrent drought and heatwave events caused damage estimated at roughly 60 billion U.S. dollars (USD). Typically, compound extremes are characterised by a complex chain of independent processes at different special and temporal scales (Zscheischler et al., 2018). E.g., droughts and heatwaves are initiated typically by similar anomalies of synoptic circulation; local and regional scale land-atmosphere feedbacks drive the evolution of compound drought-heatwave events and intensify both extremes (Miralles et al., 2019). Climate change has altered the relationships systematically between drivers of natural hazard, which increases the probability of their concurrence as well as their severity and magnitude, while the probability of the multiple extremes occurring simultaneously or successively has historically been low. Background warming resulting from anthropogenic emissions, e.g., triggers initiation of and stronger land-atmosphere feedback loops, extending their spatial impacts across North America (Dirmeyer et al., 2013) which in turn, can intensify drought-heatwave extremes and spread their spatial extent. More frequent extremes across the entire globe are shown by the literature (Sarhadi et al., 2018). Concurrent droughts and heatwaves have increased with a shift that is statistically significant in their distribution between 1990 and 2010 and 1960 and 1980 across the contiguous US (CONUS) (Mazdiyasni & Kauchak, 2015). Additionally, concurrent moderate droughts and heatwaves have increased across India (Sharma & Mujumdar, 2017), and meteorological droughts are associated with a more rapid warming rate than average climate over southern and northwestern US (Chiang & Mazdiyasni, 2018).

Drought-heatwave extremes can be intensified by land-atmosphere feedbacks through 2 mechanisms: self-intensification and self-propagation (Miralles et al., 2019). Self-intensification refers to droughts and heatwaves being intensified by each other, and self-propagation refers to the spreading of droughts and heatwaves from a region to regions that are downwind (Miralles et al., 2019; Herrera-Estrada et al., 2019). Examination of hot-dry (as well as their derivatives, i.e., compound drought-heatwave events have previously focussed on patterns of synoptic circulation that initiate these extremes and self-intensifying land-atmosphere processes that drive the evolution of these events (Zscheischler & Seneviratni, 2017; Sarhadi, Ausin, Wiper & Touma, 2018; Mazdiyasni & Kouchak, 2015; Chiang, Mazdiyasni, Kouchak, 2018). In this study a mechanism that has been less explored of land-atmosphere feedbacks that explains the propagation of atmospheric moisture deficit that has been terrestrially sourced from one region to its neighbouring areas, i.e., self-propagation. As manifested as trends in the spatial homogeneity – connectedness – this process of concurrent dry-hot extremes, has received less attention in past studies. Spatial homogeneity of compound dry-hot extremes over CONUS was analysed by the use of more than 100 years of climate data from ground observations. Further, most of the literature analysed only the concurrence of droughts and heatwaves after the 1950s (Mazdiyasni & Kouchak, 2015; Sharma & Mujumdar, 2017; Hao, Kouchak & Phillips, 2013), which overlooks the megadrought of the 1930s (Sarhadi, Ausin, Wiper & Touma, 2018). Alizadeh et al. extended the analysis to 1896-2017 (122 years) and provided a new perspective into temperature trends of compound dry-hot events.

They used observations of monthly precipitation and average temperature at the climate division scale to derive annual precipitation (water year – WY: October to September), spring-summer (March to August) average annual WY temperature, and average spring-summer (March to August) temperature, which are, in turn, used in an empirical copula framework to calculate the joint probability and return period of compound hot-dry years, as well as various subannual events. Compound dry-hot extremes were defined by Alizadeh et al. as years that have joint return periods of deficit of precipitation and excess of heat of more than 25 years (probability of joint annual exceedance of less than 0.04), unless otherwise stated. They estimated that the joint return period using the “AND” hazard scenario (drier than a threshold) in a multivariate framework (Corbella & Stretch, 2012). It is shown by the results of Alizadeh that the frequency of dry-hot extremes is increasing across CONUS, which is a significant trend at the 5% level over the western US and parts of the northeastern and southeastern US. It was demonstrated by spatial correlation analysis, i.e., the connectedness of the areas that are impacted is increasing, that these compound extremes are enlarging homogeneously. This has significant environmental repercussions that are as large and extreme events that are more intense can deplete rapidly regional and national relief resources. This knowledge can help in the assessment of regional-to-continental vulnerabilities of natural and built systems under climate change and inform adaptation and mitigation efforts to curb the grave compounding impacts of multiple extremes (Field et al., 2014).

An increasing trend for the mean annual temperature between 1896 and 2017, which is statistically significant, across much of the CONUS, with the exception of portions of the Southeast, east of the Southern Great Plains, and the southern part of the Midwest, was shown by the nonparametric Mann-Kendall analysis. There is a trend that is relatively similar for mean spring-summer (March to August) temperatures, though less pronounced for portions of the Northern Great Plains and Midwest. Over much of CONUS annual precipitation has not changed significantly, though patterns of precipitation and intensities may have shifted. There is only a strip of land that extends from eastern Texas to the Great Lakes and the Northeast that has a statistically significant increasing trend (at the 5% level) for annual precipitation. When there are persistent multiyear dry periods (Bayazit & Önöz, 2007), trend analysis of precipitation at the annual scale, however, might be contentious in the presence of autocorrelations among successive annual precipitation values. It is shown by further investigation, however, that lag-1 autocorrelation values between annual precipitation – as well as spring-summer precipitation – do not generally reach statistical significance to justify removal of autocorrelation before trend analysis. Similar behaviour is observed for high values of lag. The interdependence of precipitation and temperature can intensify the impacts state of each driver, while trends in both temperature and precipitation have significant socio-environmental implications (Kouchak, Cheng & Mazdiyasni, 2014).

Extreme temperatures, e.g., can induce “flash droughts” that have devastating impacts such as causing large wildfires (Hoerling et al., 2014; Mo & Lettenmaier, 2015). It is shown by Pearson linear correlation analysis that there is significant negative association between annual precipitation and mean annual temperature over much of the Great Plains and the Southwest. However, this correlation is not statistically significant at the 5% level for the west coast for much of the Pacific Northwest, and Midwest. The interdependence is further pronounced between annual precipitation and mean spring-summer (March to August) temperature as well as a maximum spring-summer temperature. The dependence structure between precipitation and temperature at different special and temporal scales can be altered by climate change (Hao, Kouchak & Phillips, 2013). The change in their association might, arguably, be more important than the change (e.g., increasing trend) in each variable. Given that the mean annual temperature shows a statistically significant trend across much of the CONUS, unlike annual precipitation, Alizadeh et al. linearly detrended the mean annual (as well as mean annual and spring-summer) temperature and reanalysed the linear association between temperature and precipitation. A more pronounced Pearson correlation coefficient between annual precipitation and mean annual temperature is pointed to by the results of Alizadeh et al., when temperature time series are detrended. There is a similar conclusion when temperature is exponentially detrended. This shows that the association between temperature and precipitation has been weakened by climate change at the annual scale across CONUS. At first glance it seems that this finding is counterintuitive, as it is believed background warming has strengthened land-atmosphere feedbacks (Dirmeyer, Jin, Singh & Yan, 2013). It is shown by a closer look, however, that detrending strengthens the correlation between annual precipitation and mean spring-summer temperature, as well as maximum, which confirms the strengthening of land-atmosphere feedbacks. The land-atmosphere feedback effects that are activated in the warm season are overwhelmed by the more pronounced warming rate of winters, as a result of anthropogenic background warming, at the annual scale. Nevertheless increasing temperature trends increases the probability of dry and hot extremes.

Temporal trends in dry-hot extremes

Alizadeh et al. used return period as a statistical measure of the severity and likelihood of an extreme event. The expected recurrence of a phenomenon is signified by return period (Sadegh et al., 2018); e.g., a 25-year event is expected to occur once in 25 years, on average, which is associated with the exceedance probability of 0.04 (nonexceedance probability of 0.96). For univariate extremes, the definition is intuitive, such as dryness. The concept is complex, however, for multivariate cases (Corbella & Stretch, 2012). In this study the AND hazard scenario was used, which determines the probability (or frequency) of a compound dry-hot event, i.e., drier than a threshold AND hotter than a threshold event (Grӓler et al., 2013). The empirical copula was used and the entire 122 years of the record to estimate the joint exceedance probabilities of dryness and heat excess. The annual precipitation and mean annual temperature was focused on in this paper. Mean annual temperature was selected to include winter temperature, which, together with spring, is warming at a higher rate compared to other seasons (Mello, Richmond & Yohe, 2014). There are important environmental implications associated with winter temperatures that range from snowmelt to the availability of water to phenology and the health of wildlife. Moreover, an increase in mean annual temperature is associated with a pronounced likelihood of extreme heat events (Diffenbaugh et al., 2017).

There has been a significant increase in the frequency of more than 25-year compound dry-hot extremes (joint return period levels that exceed 25 years) over the last 3 quarter-century periods. While most of the climate divisions across CONUS in the period 1943-1967 observed only 1 more than 25-year compound dry-hot extreme (as expected per the definition of such an event), and some climate divisions not observing any this slightly increased to 1 to 3 bivariate events in 1968-1992 for almost all climate divisions in CONUS.  However, there is a spike in the number of >25-year compound dry-hot extremes in the most recent period (1993-2017), with several climate divisions observing more than 5 such compound events. The increase is most pronounced for the Pacific Northwest, southern portions of the Southeast, Florida, and portions of the Northeast. This escalation in the frequency of dry-hot extremes extends beyond the randomness of climate phenomena that is expected. A statistically significant trend at the 5% level for the western CONUS, Florida, the eastern portion of the Northeast, as well as some climate divisions in Michigan, Minnesota, Mississippi, and Alabama, is shown by the non-parametric Mann-Kendall trend analysis of the return period level of compound events over the past 122 years. According to Alizadeh et al., similar inferences about more frequent compound dry-hot extremes could be made for >50-year and >75-year events. Multiple >75-year compound events are observed in the coastal Pacific, Northwest, inland Southern California, Florida, Maine, and several climate divisions in Texas, which point to the intensification of compound dry-hot extremes over many portions of CONUS with significant social-environmental repercussions, such as causing very large wildfires (Abatzoglou & Williams, 2016).

These results show an increasing frequency of concurrence of precipitation and temperature extremes over the globe and CONUS, respectively, which accords with the findings of Hao et al., (Hao, Kouchak & Phillips, 2013) and Mazdiyasni & Kouchak, (Mazdiyasni, 2015). In this study the compound the spatial distribution of compound extremes, however, is not in complete agreement with that of Mazdiyasni & Kouchak (Mazdiyasni & Kouchak, 2015), especially in the case of California and the Pacific Northwest. This discrepancy results from differences in the definition of compound events, as well as differences in the study period. In this study the definition of compound events as years when there is <0.04 exceedance of being dry AND hot (>25-year return level), whereas they use meteorological drought and various definitions of heatwaves as compound events. Also, the study of Mazdiyasni & Kouchak spans 1960-2010, whereas this study spans 1896-2017. Significant information on the frequency of compound dry-hot extremes can be added by this longer time period, as it includes the megadrought of the 1930s (discussed in the next section).

Spatial trends in compound dry-hot extremes

The analysis of Alizadeh et al. points to a substantial increase in the number of climate divisions where the were >25-year compound extremes after the 1950s across all climate regions in the CONUS. In many regions if a longer record is used (1896-2017) this increasing trend is, however, not present.  There is, more specifically, no increasing trend observed in the number of climate divisions that observe >25-year compound dry-hot extremes for the Great Plains, Midwest, and Southeast, and to some extent, the Northeast. There is, however, an interesting shift in the dominant driver for these compound events. A long, severe dry event that engulfed ⅔ of CONUS in the 1930s was the dominating driver of the joint probability of compound dry-hot extremes. The infamous dust storms of the Southern Great Plains were contributed by this drought (Shubert et al., 2014). Precipitation deficit and an excess of heat contributed to the compound events across many climate regions in the mid-2000s, and since 2010, hotter years became the main driver of the compound events across all climate regions. This observation accords with the findings Mo & Lettenmaier (Mo & Lettenmaier, 2015), who reported flash droughts driven by heatwaves have displayed an increasing trend across CONUS since 2011. It is implied by this that the dominant triggering driver of the land-atmosphere feedback has shifted from dryness in the earlier half of the study to excess heat in recent decade(s).

The number of climate divisions over the entire CONUS with >25-, >50-, >75-year events compound dry-hot extremes has shown a trend that is increasing, with >75-yer events having the highest rate of increase (i.e., longer slope of linear regression). It is also confirmed by these results that the shift in the main driver of the compound extreme events from dry years in the 1930s to hot years in recent decade(s). The concurrence of droughts and heatwaves, and, in general terms, dry and hot years, over the past 3-6 decades, has been focussed in by many compound event studies. According to Alizadeh et al., climate analysis studies should use longer time series in order detect low frequency climate cycles (Overpeck, 2013). They argued that the literature might underestimate the risk of compound dry-hot extremes by not accounting for longer climate cycles and events of lower frequency, such as the drought in the 1930s. While both megadrought (Cook et al., 2016) and megaheatwaves (Dole et al., 2011) can result from internal stochastic – not forced – variability of the atmosphere, their probability of occurrence (Steiger et al., 2019) and, most importantly, their occurrence (Sarhadi et al., 2018), has been increased considerably by anthropogenic emissions.

The accumulative area (km²) affected by compound dry-hot years also shows an increasing trend at the 10% level that is statistically significant as determined by the nonparametric Mann-Kendall trend analysis. The highest increase rate (slope of linear regression of cumulative impacted area as a function of time) is associated with the Southeast, Southwest and Northeast, respectively. However, this does not account for the total area of each climate region, and care is advised when interpreting these results for the purpose of comparison. Also, the area that is affected by >25-year hot years is also associated with an increasing trend across all of the climate regions, which is determined to be nonsignificant at the 10% level. The cumulative area where 25-year dry years are observed is, however, not associated with an increasing or decreasing trend.

It was revealed by further analysis that the cumulative distribution of the percent of CONUS that were observing >25-year compound dry-hot years for 1993-2017 (the past period of 25 years) diverges from that of 1896-1920 (the first period of 25 years). This divergence is also visible for >25-year hot years, though between the periods 1896-1920 and 1993-2017 it is less marked. Moreover, for the Kolmogorov-Smirnov, Cramér-von Mises, and Anderson-Darling tests all point to shifts that are statistically significant in the cumulative distributions in the percent of CONUS that observed >25-yeaar compound hot-dry and univariate hot years between 1993 and 2017 and 1896 and 1920, which correspond to the last and first years of this study. This distribution for >25-year dry years does, however, not differ statistically between the 2 periods at the 5% (and the 10%) level using the Kolmogorov-Smirnov and Cramér-von Mises test, though their significant divergence is pointed to by the Anderson-Darling test. In this study these tests were repeated for different 25-year periods compared to 1993-2017; generally, the results pointed to significant changes in cumulative distribution of the percent of CONUS affected by >25-year compound hot-dry extremes, with the exception of the 1918-1942 period) and more than 25-year hot years, though do not generally identify significant changes in distributions of >25-year dry years. Rather, if the 2 periods 1896-1956 and 1957-2017 are used (61-year periods), similar results are observed.

Spatial homogeneity of dry-hot extremes

Analysis of the homogeneity of the area that is affected by compound dry-hot extremes is also important. Large compound events that are spatially homogeneous can endanger natural and built system services (Fischer, Beyerle & Knutti, 2013). For natural systems, heterogeneous habitats that are connected are resilient to synchronous and large-scale populations of aquatic species and collapse of ecosystems. Fragmentation of this connectedness by homogeneous compound dry-hot extremes and result in population collapse. Homogeneous compound extremes, in the context of built systems, can damage harvests across a wide range of agricultural lands and rapidly deplete federal and local relief resources.

This study assess the connectedness of climate systems that experience > 25-year compound dry-hot extremes. The spatial correlation was calculated in terms of area that is impacted (km²) through Moran’s I as a proxy for spatial homogeneity for compound dry-hot as well as univariate dry and univariate hot extremes for each year. The results of the study show that spatial homogeneity of >25-year dry years is associated with a nominal slope that fluctuates between negative and positive values for different climate regions as well as the entire CONUS over the past 122 years. An increasing trend with linear slope of Moran’s I that ranges between 0,04 and 0.07 across various regions is shown by the spatial homogeneity of about >25-year hot years. Similarly, >25-year compound dry-hot extremes are also growing homogeneously with a slope of linear regression of Moran’s I ranging from 0.1 to 0.18 across different regions. Homogeneous enlargement of compound events is steeper than each of the drivers alone, with the largest differences being observed in the Southwest (0.18 versus 0.02). Moran’s I analysis was performed on less (>5-year) and more (>75-year) intense extreme events, in order to investigate the potential impact of the threshold of extreme events (>25-year) on the observed connectivity trend (Western, Blöschl & Grayson, 2001). Further, the cumulative distribution of annual Moran’s I for >25-year compound dry-hot years between 1896 and 1956, and 1957 and 2017, is determined to be different, statistically, based on the Kolmogorov-Smirnov, Cramér-von Mises, and Anderson-Darling tests at the 5% level. These tests are also applied to the distribution of Moran‘s I for various 25-years periods.

The homogenous spatial growth of >25-year compound dry-hot years shows a slope that is even steeper if the past 50 years alone are analysed, which is driven mainly by the homogeneous enlargement of hot years. For different regions the slope of Moran’s I for compound extreme ranges between and 0.1 and 0.6 and 0.7 for the entire CONUS. Conversely, not much change is exhibited in dry years in terms of spatial homogeneity. In this period, though the growth rate (slope of linear regression to Moran’s I) for the compound events and the univariate hot extremes for the Northwest, Northern Great Plains, and Midwest are rather similar (slightly higher for compound events) the increase in homogeneity for compound events over the remainder of the CONUS occurs at a much higher rate than the univariate hot extremes increase.

Discussion

Persistent anomalies in the large scale circulation are generally known to initiate drought and heatwave events and their co-occurrence; land-atmosphere feedbacks can, however, also intensify and propagate those anomalous climate events (Schumacher et al., 2019). E.g., megaheatwaves are generally preceded by persistent anticyclones, which enable cloud free conditions and advection of hot air (Schumacher et al., 2019), but heatwaves are intensified by drier soils by partitioning the incoming solar radiation into heat that is more sensible and less latent heat. It is noted that natural atmospheric variability can generate megaheatwaves even with land-atmosphere feedbacks (Dole et al., 2011). Land-atmosphere feedbacks, however, increase the probability of heatwaves occurring (Fischer, 2014). The probability of the Russian megaheatwave event in 2010, e.g., increased 13-fold as a result of the self-intensification feedback of drought and heatwave (Hauser, Orth & Seneviratni, 2016). Evaporation decreases as does partitioning of solar radiation into latent heat, with a lack of soil moisture; therefore, a larger fraction of the incoming solar radiation is translated into sensible heat, which, in turn, warms the environment (Fischer et al., 2007). Soils that are desiccated contribute to increase in temperature, heat entrainment, and deepening of the atmospheric boundary layer. In turn, the latter increases the evaporative demand and desiccates soils further and increases the temperature. The formation of clouds is inhibited by the cycle of drying and warming and, in turn, restrains local convective precipitation, which intensifies the drought still further (Fischer et al., 2007).

Self-propagation is a mechanism that has been explored less as a mechanism of land-atmosphere feedback (Fischer et al., 2007). Atmospheric moisture deficit and heat can propagate from a single location to locations downwind, in a Lagrangian perspective (Herrera-Estrada et al., 2019). Though heat advection and its impact on the formation and expansion of heatwaves have been explored in the literature, especially in the cases of the European megaheatwave in 2003 and the Russian megaheatwave in 2010 (Miralles et al., 2014), terrestrially sourced advection of atmospheric moisture has been explored in much less detail, but it is receiving more attention in the recent literature (Herrera-Estrada et al., 2019). The concept of “teleconnected land-atmosphere feedbacks” has been promoted by intracontinental transport of moisture that has been sourced terrestrially (Miralles, Gentine, Seneviratni & Teuling, 2019). It is believed that these teleconnected land-atmosphere feedbacks help propagate droughts to neighbouring regions, though this concept is still in its infancy (Miralles, Gentine, Seneviratni & Teuling, 2019; Herrera-Estrada et al., 2019). For regions that depend on precipitation that is sourced terrestrially, which includes large parts of North America, the propagation of droughts is more important (van der Ent et al., 2010). It is shown by the analysis of Alizadeh et al. that compound dry-hot extremes have enlarged homogeneously, i.e., areas that have been impacted are increasingly becoming connected, which points to the propagation of atmospheric moisture deficit AND heat from a region to its neighbouring regions. Significant natural and societal repercussions can result from spatial growth, which is connected, of compound dry-hot events. The summer drought of 2010 and heatwave in Russia that decreased crop production by 25%, caused more than 500 wildfires that burned more than 1 million hectares, and induced economic loss that was estimated to be 15 billion USD, is an example of a spatially large and severe compound extreme (van der Ent et al., 2010).

It is sown by the results of this study that the frequency of compound dry-hot extremes in CONUS has increased substantially over the past 50 years, though this trend is less pronounced if a longer period analysis is used (1896-2017). Background warming that result from anthropogenic emissions has strengthened, caused by earlier start of, and, extended the spatial impact of land-atmosphere feedbacks in North America, though anomalous synoptic circulation patterns are recognised for initiation of compound dry-hot events (Dirmeyer, Jin, Singh & Yan, 2013). It was reported by Alizadeh et al. that there is a shift in the dominant driver of compound dry-hot extremes from precipitation deficit in the 1930s to excess heat in recent decade(s). I.e., there has been a change in dominant driver that triggers the land-atmosphere feedbacks from meteorological drought to excess heat. This is similar in concept to the precipitation deficit-driven and heatwave-driven flash drought categorisation of Mo & Lettenmaier (Mo & Lettenmaier, 2015; Mo & Lettenmaier, 2016), in spite of temporal scale differences. It was shown by Mo & Lettenmaier (Mo & Lettenmaier, 2015) that the frequency of flash droughts that are driven by heatwaves was associated with a trend that has been decreasing over the last century, which rebounded after 2011, though flash droughts of all categories are shown to have increased in frequency around the globe over the last century. This is in agreement with the argument of Alizadeh et al. of the changing nature of the dominant driver of compound dry-hot events in the recent decade, i.e., from precipitation deficit to an excess of heat. Further, anthropogenic emissions have enhanced significantly the probability of concurrent drought and heat waves, and aggressive emissions reduction is the only strategy that can mitigate the risks associated with their increasing frequency (Sarhadi et al., 2018). Last there is no significant increasing trend in the frequency of dry years, though there are no notable variations over the different regions of CONUS; however, univariate hot years are becoming more frequent as well as more intense.

It has been argued by Alizadeh et al. that the recent literature may underestimate the risks of dry and hot episodes, as they only study the post-1950s period with no recourse to the meteorological drought of the 1930s that engulfed almost ⅔ of CONUS for almost a decade. Alizadeh et al. argue that if meteorological droughts of the length and severity that were observed in the 1930s occur during the hot years that are increasingly common in recent decades due to global warming, their concurrence can have devastating impacts (Mueller & Seneviratni, 2012; Overpeck, 2013; Diffenbaugh & Ashfaq, 2010). It has been shown by the recent literature that no major region in the U.S. is immune to the multi-decadal continental-scale megadroughts that occurred in the 12^th and 13^th centuries (Cook et al., 2014), and their return has been markedly increased by global warming (Steiger et al., 2019). Moreover, water demand is increased by a hotter climate (Das et al., 2011), concurrence of which, with dry years, would strain social, built and natural systems (Williams et al., 2013) and might push them to unprecedented states (van den Pol et al., 2017). The results of this study contribute to a deeper understanding of the spatiotemporal patterns of compound events to help with reliable risk projections in the context of climate change. The consequences of increased frequency, intensity and spatial homogeneity of climate extremes for compound events with multiple drivers are far graver than the effect of each driver individually (Zscheischler et al., 2018), and risk assessment frameworks need to consider the compounding effects of multiple extremes rather than addressing a single driver at a time within the traditional univariate framework.

Alizadeh, M. R., et al. (2020). "A century of observations reveals increasing likelihood of continental-scale compound dry-hot extremes." Science Advances 6(39): eaaz4571.