How future compound drought-heatwave events would affect our society and ecosystem?

By combining hydrology, vegetation remote sensing and atmospheric dynamics, we attempted to understand the shifts in compound hazards as well as their impacts on our socio-ecosystem.
Published in Earth & Environment
Like

Water and weather-related extremes such as drought and heatwaves are intensifying as the Earth warms, posing major threats to infrastructure and ecosystem resilience. Driven by similar synoptic weather systems, these hazards are increasingly likely to occur simultaneously, disproportionately amplifying their adverse socio-ecological consequences. Compound drought-heatwave (CDHW) events can for example exacerbate vegetation mortality, which, in turn, may cascade into other hazards, such as wildfires and crop yield losses; they can also jeopardize electric grid reliability and adversely affect a wide range of natural and human-made systems. After severe CDHWs, plant recovery usually lags owing to reduced growth, non-reversible losses in hydraulic conductance or depletion of carbon reserves. This lagged growth may in turn increase vulnerability to another CDHW if it occurs before complete recovery, potentially limiting the capacity of continents to act as a carbon sink. 

With growing evidence about these damages, CDHWs are increasingly regarded as one of the worst climatic stressors to global socioeconomic sustainability and ecosystem health. Understanding CDHW dynamics in a warming Earth is thus essential for the implementation of the UN Sustainable Development Goals (SDGs), in particular SDG13 that aims to combat climate change and its impacts. Yet, how to describe CDHW remains an open question, particularly in terms of defining a fully representative stress index. Previous studies have assessed droughts through a variety of indices such as the (self-calibrated) Palmer Drought Severity Index (PDSI) and the soil moisture (SM) drought index. More recently, terrestrial water storage (TWS), a key determinant of global water and energy budgets, has been employed to reveal large-scale drought impacts on hydrologic systems and plant growth. TWS represents the vertically integrated water storage as opposed to conventional indices that only capture partial water storages or fluxes. When we started conceiving our paper in early 2021, the physical mechanisms behind CDHW remained poorly understood, especially from the perspective of atmospheric dynamics under an intensified hydrologic cycle. Furthermore, the effects of changing TWS on CDHW and the resulting impacts on socio-ecosystem productivity were not clear.

To our knowledge, our study was the first to disentangle the physical mechanisms of global CDHW as well as their socioeconomic and ecological impacts, under both current and future climates. We first analyzed the association between daily maximum near-surface temperature and TWS from satellite observations, field measurements, Gravity Recovery and Climate Experiment (GRACE)-constrained reconstruction and reanalysis data during 1979-2020. We detected strong multi-temporal-scale coupling during the warm season, highlighting the high likelihood of concurrent drought and heat extremes. To assess the physical mechanisms behind CDHW, we measured the responses of large- and local- scale atmospheric dynamics to heat stress, drought and their temporally compounding extremes (Figure 1). We then evaluated the effects of climatic extremes on the terrestrial carbon budget by using net ecosystem productivity (NEP) as well as its partitioning into photosynthesis (i.e., GPP) and respiration (i.e., total ecosystem respiration, TER). To achieve this, we combined in situ eddy-covariance flux tower observations, a recent satellite-based machine-learning-generated solar-induced chlorophyll fluorescence (SIF) dataset, and a light use efficiency theory-based GPP dataset. Moreover, we assessed future shifts in CDHWs for various socioeconomic and ecological subgroups using a large ensemble (96 scenarios) of climate-hydrology simulations under the ISIMIP2b (Inter-Sectoral Impact Model Intercomparison Project phase 2b), and 15 members of TWS simulations by driving the H08 global hydrological model with bias-corrected CMIP6 (Coupled Model Intercomparison Project Phase 6) ensemble outputs. Finally, we examined the changes in joint return period (JRP) using a “AND” hazard scenario of CDHWs under a bivariate non-stationary framework, and systematically quantified the associated uncertainty by using the multivariate analysis of variance (ANOVA) method.

Figure 1: Anomalies of composite water-heat-carbon variables during extreme climatic events. a-h, Anomalies of water and energy variables during extreme heat events. i, Pearson’s correlation coefficient between daily GLADS TWS and ERA5 Tmax. j, Mean probability of each percentile bin of Tmax and daily GLADS TWS across 73 flux tower sites. k-o, Anomalies of GPP, TER, and NEP for each percentile bin of Tmax and TWS (or SM) across 73 flux tower sites.  

Our large climate-hydrology model ensemble under both CMIP5/6 projected that CDHW magnitudes (occurrence, duration and severity) might quadruple over 70% of global land areas under medium and high emission scenarios. The CDHWr (i.e., ratio of CDHW to heatwave events) is also increasing globally, indicating that the interdependence between heatwave and droughts is strengthening as the climate warms. The compound events would pose a disproportionate impact on the global vegetation and human population in the future, unlike in the past. For instance, an additional 17-21% (18-25%) of the global population (GDP) is projected to experience CHDW by the end of the century, which could translate to an additional ~1.4 to ~1.7 billion people (~13 trillion to 20 trillion U.S. dollars at 2015 PPP) per year. Importantly, we found that the frequency of extreme CDHW (historical 50-year event) would increase by ten-fold under the highest emission scenario, and over 90% of the world population and GDP is projected to be exposed to increasing bivariate CDHW risks in the future climate under all SSPs/RCPs (Figure 2). 

Figure 1: Projected JRP of historical 50-year bivariate CDHW and socioeconomic exposure. a-c, Temporal dynamics of the fraction of global average exposed land area (a), population (b) and GDP (c) due to increasing CDHW risk. d, Boxplot of updated JRP of the historical 50-year CDHW in different Giorgi climate regions under RCP 8.5.

Figure 2: Projected JRP of historical 50-year bivariate CDHW and socioeconomic exposure. a-c, Temporal dynamics of the fraction of global average exposed land area (a), population (b) and GDP (c) due to increasing CDHW risk. d, Boxplot of updated JRP of the historical 50-year CDHW in different Giorgi climate regions under RCP 8.5.

As poor people often live in risky areas and have limited capacity to adapt, they might be more exposed or more vulnerable to natural disasters than wealthier people. We therefore examined whether the CDHW risks and corresponding socioeconomic exposure are different between poorer and richer subgroups. The poorer areas have a higher fraction of the population and GDP (0.44 and 0.44, respectively) exposed to CDHW than the richer areas (0.39 and 0.38, respectively) by 2070-2099 under SSP585 (Figure 3). In addition, we found that rural areas are more vulnerable to CDHW than urban areas in terms of higher CHDWr and socioeconomic exposure under climate change across all SSPs.

Figure 3: CDHW coincidence rate and socioeconomic exposures to CDHW in rich versus poor areas. a-f, Temporal dynamics of the global average coincidence rate (a-b), and exposed GDP fraction (c-d) and population fraction (e-f) to CDHW.

In summary, our findings provided a firm conclusion that future CDHW hazards are projected to intensify significantly and challenge the sustainable development of future socio-ecosystem system. This work thus calls for stark mitigation and adaptation actions to reduce the adverse impacts of warming on societies and to sustain ecosystem productivity, easing the growing pressures on global sustainable development, particularly for poorer and rural areas in the Anthropocene.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in