Hurricanes and tropical cyclones (TCs) pose a significant threat to power systems, causing widespread outages and lasting disruptions. The 2017 hurricanes Maria, Irma, and Harvey, and several others in the subsequent years, exemplify the devastating impact on power grids, leaving millions without electricity for extended periods. This necessitates a thorough investigation into power system resilience and the potential need for infrastructure redesign. Climate change is expected to exacerbate the situation, intensifying hurricanes and increasing the frequency and severity of heatwaves. The combination of more intense TCs and more frequent, longer heatwaves creates a compound hazard, where power outages caused by TCs exacerbate the already dangerous impacts of heatwaves due to the loss of air conditioning. This compound hazard is particularly dangerous because it can extend the duration of life-threatening conditions for vulnerable populations. Earlier research has hinted at the future increase of this compound hazard, but it often neglected the changes in TC climatology and the role of power system resilience. This study addresses these shortcomings by coupling projections of TCs and heatwaves with an analysis of power outage and recovery processes to better quantify the evolving risk of TC-blackout-heatwave compound events.

Previous research has explored the individual impacts of tropical cyclones and heatwaves. Studies have documented the extensive power outages caused by hurricanes, highlighting the vulnerability of power systems to extreme weather events. The increasing intensity of hurricanes projected under climate change scenarios is well established. Similarly, the literature shows a strong link between climate change and more frequent and intense heatwaves, emphasizing their impact on public health and mortality. A pioneering study by [citation] connected TCs with heatwave impacts, but this study had limitations because it did not account for potential changes in TC climatology, possibly leading to an underestimation of the compound hazard. Moreover, the impact depends on power system reliability, which was not explicitly modeled in previous work. This research builds upon the previous literature by incorporating these crucial elements.

This study uses a coupled modeling approach to investigate TC-blackout-heatwave compound hazards. It combines statistically downscaled TC climatology and directly projected heatwave climatology from general circulation models (GCMs). A physics-based power system failure and recovery model is used to simulate wind-induced outages at the census-tract level. This model is validated against observations from Hurricanes Harvey (2017) and Ike (2008). Synthetic TC datasets generated by a deterministic-statistical TC model are used. These datasets, including 10,000 20-year simulations for both historical (1981-2000) and future (2081-2100 under RCP8.5) climates, are based on NCEP reanalysis and six CMIP5 GCMs. The GCM-simulated storm frequency and landfall intensity distribution are bias-corrected. A heat index (HI) exceeding 40.6°C is defined as a heat event, and the study focuses on heatwaves lasting longer than five days following a TC landfall. The model accounts for the interdependence between TCs and heatwaves in the HI calculations. The study also incorporates network analysis to investigate the relationship between local disruptions and global power failures. Three network enhancement strategies—randomly undergrounding transmission and distribution networks and a greedy undergrounding strategy focusing on distribution network root nodes—are evaluated to assess their effectiveness in mitigating compound hazard risk.

The results reveal a dramatic increase in the risk of TC-blackout-heatwave compound hazards under the RCP8.5 scenario. The probability of a heatwave lasting over 5 days following a TC increases significantly from the historical to the future climate (2.7% to 20.2%). The expected percentage of Harris County residents experiencing at least one longer-than-5-day compound hazard in a 20-year period is projected to increase from 0.8% to 18.2%, a 23-fold increase. Analysis of historical cases (Hurricanes Harvey and Ike) shows that the power outage and recovery model accurately captures both large and small outage events. The study finds that the largest 20% of local disruptions account for 72% of global power outages, indicating a disproportionate impact from localized failures. Network analysis reveals a strong correlation between the longer-than-5-day power outage risk and the length of local power distribution networks, suggesting that urban development patterns may significantly influence the spatial distribution of outage risks. The greedy undergrounding strategy, focusing on protecting a small portion of wires close to the root nodes of the local distribution networks, proves to be highly effective in mitigating the compound hazard risk. Undergrounding only 5% of the distribution networks close to root nodes reduces the expected percentage of affected residents from 18.2% to 11.3%. This highlights the cost-effectiveness of this targeted approach compared to randomly undergrounding distribution or transmission networks.

These findings highlight the significant and potentially devastating impact of climate change on coastal communities. The projected increase in TC-blackout-heatwave compound hazard risk underscores the urgent need for climate adaptation strategies. The results suggest that focusing on improving the resilience of local power distribution networks is crucial for mitigating the impacts of these compound events. The superior performance of the targeted undergrounding strategy compared to random approaches indicates the importance of strategically planning infrastructure upgrades to maximize their effectiveness. The study also emphasizes the importance of accounting for the interdependence between TCs and heatwaves, as well as other factors such as flooding and aging infrastructure, in future risk assessments.

This study demonstrates the dramatic increase in TC-blackout-heatwave compound hazard risk projected for coastal communities under high-emission scenarios. The findings highlight the urgent need for climate adaptation measures, particularly focusing on improving the resilience of local power distribution networks. Strategically undergrounding distribution networks near root nodes offers a cost-effective approach to mitigating this risk. Future research should focus on refining models to incorporate additional factors like flooding, aging infrastructure, and the impacts of backup power generation. Further investigation into the economic impacts of these events and the cost-benefit analysis of different mitigation strategies is also warranted.

The study's analysis focuses on wind effects on the power grid, neglecting the impacts of flooding and debris, which can also cause significant damage. The model simplifies the distribution networks using star-like networks and does not explicitly model protective devices, although sensitivity analysis indicates that these simplifications have limited effect on daily-scale outage predictions. The projections are based on a high-emission scenario (RCP8.5) and may not represent the full range of possible future climates. The study also assumes that power system operation and recovery strategies will remain unchanged in the future. Future research should incorporate more detailed modeling of the power system and consider other potential mitigation strategies.