Often discussed as an aspect of climate change is the idea that seasons will somehow be different than they have been in the recent past. How precisely they will differ, though, is exceptionally hard to pin down for analysis. For example, what defines 'spring'? Is it simply the transition between winter and summer? Is it a certain statistical distribution of temperature, precipitation, and cloudiness (not just their values but the combinations of them on certain days, the temporal sequences in which they occur, and so forth)? If May becomes substantially warmer, does that make it part of summer? How would we know?
It is these types of questions that have so far precluded meaningful seasonal-change analysis. One challenge is that seasons are so different from place to place -- both their first-order statistics, as noted, but also the experience of them. What would winter in New England be without abrupt cold fronts, or Great Plains summer with fewer thunderstorms? Such changes might not be readily apparent in temperature and precipitation averages, if compensated for example by increases in non-thunderstorm precipitation, but would lend the season a different feeling and influence impacts such as power outages. Similarly, particular combinations of weather patterns can be of immense cultural and/or agricultural significance. A moderate decrease in early-spring diurnal temperature variation may lead to a dramatically shorter maple-sugaring season in Vermont or Quebec, affecting sense of place. Crops that require chilling hours or moderate temperatures would necessarily see geographic shifts. Elsewhere, typical mid-latitude conceptions of distinct seasons do not apply, requiring a more careful multivariate depiction. Along the Southern California coast, for example, Santa Ana winds can cause the most intense heat of the year to occur in late autumn or even winter, leading to health impacts. This year, January, April, September, October, and November all saw higher temperatures in coastal Los Angeles than anything in June, July, or August.
Extreme (or indeed unprecedented) events can color impressions and shape impacts of an entire season, despite relatively short durations. The incredible Pacific Northwest heatwave this summer stands as a canonical example. Even if the rest of the summer conformed closely to expectations, the die of memory (for humans, vegetation, soil, etc.) was already cast. That raises the question of the proper scope and timescale for attribution analyses, which are coming fast and furious these days. Event-based attribution to anthropogenic climate change is the standard, but by omitting the seasonal context, important information may be being lost. This is particularly important for multi-season events such as summer drought preconditioned by spring heat, fire-following mudslides, or spatially propagating droughts.
One approach that holds promise for a more intuitive seasonal quantification involves a deeper investigation of 'weather types' -- in other words, reducing the dimensionality of weather by categorizing days into multivariate composite clusters. Changes in frequency of weather types could then indicate important shifts in seasonal behavior, perhaps concentrated at boundaries between seasons. But then here again several philosophical issues rear their head. First, whether to rely on actual values or values relative to an annual distribution -- warming, combined with a smaller winter/summer difference in a warmer world (continuing the pattern of greater winter warming that has occurred since the Last Glacial Maximum, also supported by longstanding model evidence), is the key underlying factor. Second, even an idealized conception of seasons has a 'ragged edge' where there is some alternation between weather typical of the preceding and the subsequent season.
Evidently, each of the Köppen climate 'types' contain multitudes, some more important than others but all posing complications to assumptions about what meaningfully distinguishes one season or weather type from another. The most material impact of warming-driven seasonal change will involve behavioral practices that will have to shift along with changing environmental suitability due to physiological constraints (those of humans, animals, and/or crops). A rigid climatic shift where all patterns are identical except warmer by 2 or 3 deg C would mean that a once-pleasantly-warm summer outing becomes uncomfortable or dangerous, potentially leading to large-scale societal changes in the timing of activities (avoiding the warmest part of the day and/or the warmest time of the year). Of course, the thermodynamics of a warmer and moister atmosphere also drive exponentially intensifying extreme events, which often have contributions from changes in dynamics as well.
Ultimately, the question of how to treat future seasons crystallizes a few of the principal issues in our collective understanding of a changing climate. Surely our ancestors of 15,000 years ago wrestled with some of the same problems, although with a different vocabulary and presumably less guilt, because there was little objective understanding of the situation (rapid glacial melting) and no perceived control over it. Despite (or even perhaps because of) reams of data, the world almost unavoidably requires narratives to imbue it with meaning. The concept of seasons is one such narrative. But they are a simplistic approximation of a complex system, and a threshold-based approach to a continuous problem. Both of these often lead to dramatic conclusions and misleading impressions. A relativistic definition of seasons across changing climates is without doubt the most basic approach, yet this conflicts with our closely held associations. Bringing our mindsets to a physical system is bound to get ourselves tied up in knots. As elegantly described in the essential climate-philosophy book "Climate Conundrums", Earth (and indeed life itself) has no values. *We* have values, but we shouldn't get confused into thinking these are objective or inevitable. Technology often outpaces principles, and following that idea, maybe it will turn out to be the case that changing the atmosphere, ocean, and land surface is the easy part, and adjusting our ideas and expectations to match the new reality that follows will prove significantly harder.
Extreme events with decreasing trends in frequency or severity can still cause major disruptions to human or natural systems — as vividly illustrated by February’s winter storm in the south-central United States, which may ultimately become the costliest natural disaster in U.S. history. These impacts underscore that the path to comprehensive climate resilience runs through assessments of the structure and interconnections of systems, as well as narratives that emphasize the large variability around trends in extreme events.
In mid-February 2021, widespread extreme cold paralyzed power and water systems in Texas and neighboring states (Figure 1). Seeing areas known for hurricanes and summer heat shivering amid near-record cold and snow was especially striking in light of observed trends and future projections emphasizing increases in heat, drought, flood, and fire.(1) The concept of ‘global weirding’ has gained prominence for describing this surficial paradox of increasing extremes in both tails of a distribution, a paradox which is readily resolved by recognizing several critical factors, including the ongoing amplification of certain hydrological processes and the (possibly temporary) destabilization of the boreal-winter polar vortex.(2,3) Here, I argue that weirding — the variability component of climate change — and its potential impacts be investigated more closely and systematically, using the Texas event as a lens into some of the key issues and categories of potential solution.
Global-mean warming is causing the ‘leading edge’ of various hazards to shift, decade by decade, into locally unprecedented territory.(4) Appropriately, much effort and funding is being expended to assess the growing risks these hazards pose, and to determine optimal adaptation and mitigation strategies for them.(5) In fact, catastrophic risk theory argues that they perhaps merit even more attention, so as to fully understand low-probability, very-high-risk scenarios for the remainder of the 21st century and beyond.(6)
In this context, the disaster in Texas serves as a reminder that extreme winter cold — and other ‘trailing-edge’ hazards such as cold wet springs (7) with observed or projected decreasing trends in frequency or severity — remain a critical source of risk, more than capable of overwhelming fragile societal systems. That this is possible, despite rapid warming, is due in part to a lack of generalized resilience in the affected systems and in part to anthropogenic networks which scramble, and in some cases amplify, the direct link between hazard and impact which would exist in a simpler world.(8) As illustration, extreme cold between February 10th and 17th spiked power demand across Texas to levels higher than the state’s electric grid could handle, and its isolation from interstate network connections, along with the inadequate winterization of natural-gas power plants and other critical infrastructure, led to multi-day power outages. Poorly insulated water pipes indoors and outdoors then froze and burst, causing serious damage to buildings and leaving millions of people without running water, exacerbating a humanitarian crisis.
The unfolding of this disaster points to two problems: a failure of knowledge (or awareness of that knowledge) and a failure of action. Much attention has focused on the latter issue, and justifiably, because recommendations following disruptive cold and snow events in Texas in 1989 and 2011 were mostly not followed through upon.(9) Designed to fit short-term climatologies and encompassing a spatial area too small to include meaningfully offsetting temperature anomalies, the state’s power grid has also seen strain from summer heat waves.(10) The issue of knowledge is somewhat more subtle but nonetheless crucial for optimal risk assessment and mitigation. Consider two episodes that affected Houston: the recent cold, and the flooding from 2017’s Hurricane Harvey. Both overwhelmed the city’s infrastructure, and both had impacts strongly shaped by investment, development, and policy decisions that often aggravated pre-existing socioeconomic inequalities.(11) However, although temperatures in Houston fall below -3C about annually, with -10C not uncommon across North Texas, freezing temperatures of sufficient duration and extent to cause serious impacts went unmentioned in a recent report on extreme-weather risks in the state (in contrast to tropical cyclones).(12) There are even indications that this ‘trailing-edge’ event could ultimately be the costliest natural disaster in US history, surpassing Harvey.(13)
Beyond the recent headlines and regional anecdotes, the argument that a more complete picture of climate risk necessitates a deeper appreciation of trailing-edge-type events has several pillars. For one, there is evidence that the tails of the distribution of climate impacts are often thicker than models would suggest.(14) If true for the trailing edge as well as the leading edge, even large decreases in trailing-edge hazards might result in only small decreases in their impacts, for example due to downstream risk cascades.(15) At a more theoretical level, while the question of how much to value the future relative to the present (i.e., the economic discount rate) is a subjective one, this rate is nearly always negative — that is, avoiding harm now is worth the most. Skimming over events with decreasing trends in favor of preparation for the medium- and long-term future also requires a degree of presumption that societies are reasonably well-prepared for the set of extremes physically possible with (for example) 20th-century climate forcings. This may be true in isolated cases, but neglects the various resource impediments to optimal adaptation measures, as well as the crucial and dynamic role of societal modulation in shaping both events and their impacts.(16)
The most direct way to ensure that trailing-edge risks are not underestimated lies in the framing of the narrative around extreme events, such as emphasizing how in many cases awareness and preparedness matter more than the magnitude of a hazard in determining its impacts. In both public-facing and intra-scientific discourse, this narrative could be furthered by portraying trailing-edge-type hazards over the next several decades as a threat that is very much present in most extratropical locations due to inherent climate-system variability. The emerging concept of multidimensional risk provides a helpful and widely applicable perspective, particularly relevant for its emphasis on the importance of specific combinations of extreme events rather than broad trends in any one variable. For example, the frequency of false springs (a period of warmth followed by a late frost) is expected to increase in some regions despite significant warming,(17) while rapid ‘whiplash’ between anomalous hydroclimatic states of opposite type is likely to accentuate both drought- and flood-related impacts in California.(18) Multi-hazard dependencies can also act to decrease risk compared with independent events.(19) Relative to a top-line emphasis on trends, this type of framing aids in placing into appropriate context the occurrence of any extreme events involving those variables, whether they are overall increasing or decreasing. A variety of complex relationships can be usefully explored through the framework of compound events (20), which considers the interaction of distinct variables or hazards as an integral driver of risk, and which can be extended to incorporate the substantial modulating effect of human systems like policies and transport networks.(8)
Holistically evaluating climate risks also includes determining how much stands to be gained at the trailing edge in cases where risk is increasing at the leading edge. These benefits are often overshadowed, and rightly, by ramifications of the rapid and unrelenting increase in temperature to levels outside the range in which humans evolved.(21) But in a world where extreme cold still causes more mortality than extreme heat (22), the likely reduction of this harm by mean warming is worth exploring in greater detail. Similarly, that cities tend to be warmer than their surroundings is typically investigated as a summertime-heat problem, but it is equally valid to consider this urban-heat-island effect in moderating winter cold.(23)
Many types of extreme events consistent with the current state of the climate system can lead to major harm when societies are not sufficiently resilient. In the recent Texas winter storm and cold spell, serious deficiencies in infrastructure were paired with critical socioeconomic vulnerabilities, as has been true for other recent disasters.(24) Although most focus deserves to remain on extreme events that are growing in prominence, due to their looming and poorly constrained impacts, trailing-edge-type risks cannot be neglected in refining scientific understanding or implementing resilience measures. Such a balance will help ensure that future climate extremes are not so disastrous, no matter what their location within statistical distributions or which combinations of systems they affect.
1. Seneviratne, S. I., et al. (2012). A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press.
2. Held, I. M., & Soden, B. J. (2006). J. Clim. doi:10.1175/jcli3990.1.
3. Cohen, J., et al. (2020). Nat. Clim. Change. doi:10.1038/s41558-019-0662-y.
4. Coumou, D., & Rahmstorf, S. (2012). Nat. Clim. Change. doi:10.1038/nclimate1452.
5. AghaKouchak, A., et al. (2020). Ann. Rev. Earth Planet. Sci. doi:10.1146/annurev-earth-071719-055228.
6. Lenton, T. M., et al. (2019). Nature. doi:10.1038/d41586-019-03595-0.
7. Faust, E., & Strobl, M. (2018). https://www.munichre.com/topics-online/en/ climate-change-and-natural-disasters/climate-change/ heatwaves-and-drought-in-europe.html
8. Raymond, C., et al. (2020). Nat. Clim. Change. doi:10.1038/s41558-020-0790-4.
9. Federal Energy Regulatory Commission and North American Electric Reliability Corporation. (2011). https://www.ferc.gov/sites/default/files/2020-04/08-16-11-report.pdf.
10. Krauss, C., Fernandez, M., Penn, I., & Rojas, R. (2021). New York Times. https://www.nytimes.com/2021/02/21/us/texas-electricity-ercot-blackouts.html.
11. Smiley, K. T. (2020). Environ. Res. Lett. doi:10.1088/1748-9326/aba0fe.
12. Nielsen-Gammon, J., Escobedo, J., Ott, C., Dedrick, J., & Van Fleet, A. (2020). https://climatexas.tamu.edu/products/texas-extreme-weather-report/ClimateReport-NOV2036-2.pdf.
13. Ivanova, I. (2021). CBS News. https://www.cbsnews.com/news/texas-winter-storm-uri-costs.
14. Schewe, J., et al. (2019). Nat. Commun. doi:10.1038/s41467-019-08745-6.
15. De Ruiter, M. C., et al. (2020). Earth’s Future. doi:10.1029/2019ef001425.
16. United Nations Office for Disaster Risk Reduction (2019). eISBN: 978-92-1-004180-5.
17. Allstadt, A. J., et al. (2015). Environ. Res. Lett. doi:10.1088/1748-9326/10/10/104008.
18. Swain, D. L., Langenbrunner, B., Neelin, J. D., & Hall, A. (2018). Nat. Clim. Change. doi:10.1038/s41558-018-0140-y.
19. Hillier, J. K., & Dixon, R. S. (2020). Environ. Res. Lett. doi:10.1088/1748-9326/abbc3d.
20. Zscheischler, J., et al. (2020). Nat. Rev. Earth Environ. doi:10.1038/s43017-020-0060-z.
21. Xu, C., Kohler, T. A., Lenton, T. M., Svenning, J.-C., & Scheffer, M. (2020). Proc. Nat. Acad. Sci. doi:10.1073/pnas.1910114117.
22. Gasparrini, A., et al. (2015). Lancet. doi:10.1016/S0140-6736(14)62114-0.
23. Yang, J., & Bou-Zeid, E. (2018). J. Appl. Meteorol. Climatol. doi:10.1175/jamc-d-17-0265.1.
24. Kishore, N., et al. (2018). N. Engl. J. Med. doi:10.1056/nejmsa1803972.
This Halloween and election season follow right on the heels of some of the most dramatic weather-related challenges that the Western US has seen in a long time. Enormous fires burned major portions of the Pacific Northwest down through central California, including the coastal ranges, causing orange skies and historic air pollution in Portland and San Francisco, while Phoenix set a variety of extreme-heat records (that no one was cheering on) due in part to a nearly absent North American Monsoon. Drought has covered much of the West the last few months, also resulting in intense autumn fires in the Colorado mountains, despite a preceding and record-breaking early-September snowstorm.
These weather extremes highlighted the slow climatic drivers that stress systems and, in combination with human-behavioral factors, can cause them to behave in intense or unexpected ways. Case in point: warming plus natural variability led to the extreme region-wide drought of 2012-16; the drought led vegetation to be weaker and more prone to beetle infestations; the combination of dryness and insects killed millions of trees; those trees provided ready kindling for any fires that sparked in the next dry year, either from lightning or human activities. That the Western-US landscape is dynamic and heavily shaped by human influence makes facing the constraints of the naturally dry (but strongly variable) climate a daunting psychological and political challenge. Having lived in California now for a bit over a year, my observation is that this state epitomizes how climate can serve as a revealing lens into the conflicted soul of a place. Below, I elaborate on this thought and what it means.
The statistics on California's success are everywhere: its economy is larger than that of France or India, and its companies shape the daily lives of billions of people. Yet despite its image as a paragon of a gentler sort of capitalism than is practiced in Texas or Florida, California encompasses gaping inequities: it is a land of dazzling success and glamour in beautiful surroundings, but also of grueling low-paid work in hot dusty fields or sprawling industrial zones. These poles of desirability are highly correlated with the climate, much more so than in other places, even ones that are more unequal. The wealthy are concentrated along the coast, monopolizing access to cool weather and fresh clean air off of the ocean, while poorer people (sadly, often this means people of color) live disproportionately in the Central Valley, Inland Empire, and interior deserts, with their hot summers and perennial smog. The relatively more pleasant climate of some minority neighborhoods, like those in South LA, is offset by longstanding pollution and other environmental issues there. Where the wealthy do own property or spend time in sometimes-adverse areas, like the Coachella Valley (home of Palm Springs), they tend to visit at the best times of year and have permanent dwellings elsewhere -- whereas workers in those communities stay there year-round and suffer through the entire summer at 40-45 C, often while working outside in sectors such as agriculture or property maintenance.
A similar story plays out on a more-local level, for example with respect to street shade. In a climate where trees often require conscious planting or tending, the differences among neighborhoods in terms of resources (and in city priorities) are on stark display. As in the below two images I've captured of a urban swath south of downtown LA, the relative wealth of an area can be ascertained with reasonable accuracy even from space. These features lead to localized heat stress and the associated impacts. While a given heat wave in California may have a larger impact along the coast, due to the rarity of extreme heat there, this must be weighed against the much higher absolute temperatures inland.
People with means have always sought out the most naturally desirable parts of a metropolitan region, whether that is the western (upwind) neighborhoods of mid-latitude cities or elevated neighborhoods away from docks, factories, and mosquito breeding areas. In California, the topography and coastline create large differences in climate over small distances, heightening the contrast. Restrictive zoning has further increased property values and reinforced this climate/income correlation by pushing lower-cost housing to the urban fringes, where heat, drought, and fire are more serious concerns.
However, a countervailing trend that has become apparent in recent years is that of the owners of expensive properties who would rather invest large sums for protection against climate-related hazards than sell and move elsewhere. This is happening in Malibu (despite recent wildfires), Montecito (despite recent mudslides), and Newport Beach (despite ongoing sea-level rise). Evidently, it will take repeated disasters to change mindsets, if at all, regarding the overall desirability of the coast. Another component of the calculation is that within urbanized areas (as opposed to the wildland-urban interface) fires are quite rare, and mudslides and sea-level rise threaten a relatively small number of properties, whereas heat, drought, and air pollution affect large areas all at once. And even when exposures are similar, socioeconomic factors cause greater vulnerability among poorer communities, such as through their lower levels of insurance coverage, resulting in very long shadows of impacts from disasters.
Many of California's challenges, from homelessness to climate change, hinge on affordable and equitable housing. Achieving this will require a certain innovativeness, graciousness, and progressiveness that is hard to realize anywhere. There must first be a broader recognition of the endemic de facto environmental/climatic segregation here, and how it worsens overall health and economic outcomes. Erasing such distinctions is of course impossible, but taking a completely laissez-faire approach is itself an active choice, and one whose short-sightedness climate-related stressors will likely continue to demonstrate over the coming years.
The ongoing, decades-long integration of non-atmospheric components into climate science takes many forms. This reflects increasing recognition of the two-way interactions by which climate impacts both modulate and are modulated by culturally based decisions, demographic vulnerabilities, societies' financial resources and technical capabilities, and so on. Simultaneously, the emphasis on mean change from the early IPCC reports has given way to a diverse bouquet of illustrations of how rare and geographically patchy conditions cause the lion's share of impacts in just about every way that we care about: health, ecosystems, economies, and societies. Even among extreme events, it is often multifaceted ones (or combinations of them) that are responsible for the most severe damages. These 'compound' events can result from intrinsic physical processes (as with the tight relationship between heat and drought in many places, or a single storm affecting coastal water levels in several ways), or from the more or less random co-occurrence of two events in close spatial or temporal proximity (e.g. wildfires followed at a lag of weeks or months by heavy rain).
Sector-specific analyses, e.g. for food, hydrology, or insurance, have gone a step beyond physical hazards, by diving deep into relevant locations, actors, and relationships that affect risks. Along with work by the natural-hazards community, these studies have refined the concept of 'risk' to now encompass a nuanced portrait of how individual and sociocultural characteristics ('societal drivers') and physical characteristics ('climatic drivers') combine and interact to produce the observed impact from an observed event. Inspired by such ideas, in May 2019 I and others hosted a workshop at Columbia University exploring the dual frontiers of (1) characteristics of interacting extreme events and (2) how they affect human societies.
Our new paper builds on some of the main threads of the workshop, but adds in what had been an overlooked element: how human decisions can amplify impacts relative to what they could have been, and in some cases can even feed back onto the severity of the event itself. In the terminology that we propose, these societal drivers transform an event (or event sequence) from being 'compound' to 'connected'; that is, interactions between events or event components can occur regardless of whether their physical hazards compound or not. The example we describe in the most detail -- because it was recent, well-documented, and catastrophic for those who were shortchanged -- revolves around Hurricane Maria's impacts on Puerto Rico in 2017. The island's existing vulnerabilities in terms of infrastructure and emergency-response systems were already laid bare when Hurricane Irma struck weeks earlier; Category 5 Maria added to the misery. But beyond that, bureaucratic inefficiencies and resource limitations on the part of the Federal Emergency Management Agency made for a slow and painful recovery, ultimately magnifying the long-term costs. Once the (admittedly complex) prototype is understood, examples abound. Multiple-breadbasket failure, for instance? Its likelihood is implicitly defined by, and thus modulated by, the places where crops are grown (and this could be changed in the future). Large wildfires primed by naturally dry years and climate-change-aggravated summertime heat? They're also primed by untrimmed vegetation, aging equipment, and land-use policies that allow or even encourage population growth in canyons and hills. In fact, the below figure from the paper illustrates how peaks in the timeseries of damages from four major types of extreme events are all straightforward to explain using this framework.
Paper figure 2: Major losses caused by extreme climate events over 1980-2019 and their connective elements, among tropical cyclones (green), floods (blue), droughts (orange), and wildfires (red). Annotations in high-loss years state some of the (first row) climatic and (second row) societal drivers that shaped the total impacts. Loss data are from Aon, Catastrophe Insight Division.
In the course of writing the paper, we realized that intricacies of the type described above are not at all rare. They may only seem so because data limitations, and the intimidating nature of the multiple feedback loops that can be present, have restricted the number of reports that directly reference such interactions. However, it does not take much supposition to conclude that connected extreme events are an important and fertile area spanning the intersection of climate science, engineering, and the social sciences, and one for which we recommend some principles for effective research based on our author team's researcher and practitioner expertises. Chief among these is the recognition that, to be actionable, climate information must be provided in a way that is directly congruous to existing decision-making pathways (for whatever the intended application). Studies focusing on even a slightly different area, timescale, or variable are nearly impossible for stakeholders to integrate, considering that they have many responsibilities and considerations outside of the climate domain, as well as legal mandates and financial limitations. In other words, collaborations that lead to important impact-relevant research share many characteristics with collaborations among scientists: they must be carefully and deliberately constructed; not suck up all of either party's time; rest on a solid foundation of trust; and be flexible enough to recognize each other's interests and strengths, yet firm enough to mutually shape them. Storylines, stress testing, and ethnographic surveys are some of the promising yet still sparsely employed techniques for identifying research questions that sit at the junction of what can happen, what can be found, and what a decision-maker needs to know when it comes to connected extreme events.
Notwithstanding the topic's early stage, we conclude the paper on a note of optimism, relating that similar challenges have been successfully met in other contexts, and that it takes only having a sufficient incentive for people to come together and devise improvements to a functional but suboptimal status quo. The increasing severity and frequency of many types of climate extremes, resulting in a continual stream of new combinations of impactful and hard-to-predict idiosyncratic connections, hopefully provide that basis. Even without improved forecasts or projections per se, we are convinced that investing in better awareness about and allocation of resources for connected extreme events is more than worthwhile if it enables responses to them that result in recovery, innovation, and increased resilience going forward, rather than degradation of human and financial capital and a perverse widening of inequalities.