Observing Mother Nature

A look at severe weather events in a changing climate Stephanie C. Herring

Statements of fact and opinions expressed herein are those of the individual author(s) and are not necessarily those of the Society of Actuaries or the respective authors’ employers.


In 2017, the United States experienced record flooding from Hurricane Harvey, Hurricane Irma set record wind speeds, extreme wildfires spread across the Western United States, massive ocean heat events off the coast of Alaska continued to impact the U.S. fishing industry, flooding occurred in California and hail storms hit Colorado, just to name a few extreme events. In total, the United States experienced 16 weather and climate disaster events in 2017, with losses exceeding $1 billion each (see Figure 1). In addition to the high frequency of last year’s extreme events is the cumulative cost, which exceeded $300 billion in 2017, a new record.

Figure 1: 2017 Billion-Dollar Weather and Climate Disasters in the United States

Source: NOAA National Centers for Environmental Information (NCEI) “U.S. Billion-Dollar Weather and Climate Disasters” (2018). https://www.ncdc.noaa.gov/billions.

Extreme events can have extraordinary socioeconomic and environmental consequences that can alter communities and the landscape for years after the event. Of course, these events aren’t new. Extremes have been affecting the Earth since the beginning of time and will continue to be a fact of life. Given that extreme events are here to stay, the next question for our society is, what should we do about them? How do we prepare and build resilience to these events and mitigate their impacts on people and property? The answer to this depends in part on what we think the future holds for extreme events. Will they become more or less likely? More or less intense? Occur in new locations? Scientists have long studied these questions, and  evidence is mounting that a new driver is impacting how extreme events will change in the future. The evidence that human-caused climate change is affecting the intensity, frequency and geographic distribution of some extreme event types is now undeniable.1,2,3,4 Quantitatively connecting the impact of human-caused climate change on extreme events is the focus of the field of climate change event attribution. Event attribution examines whether climate change played a role in altering the risk of a specific event through a probabilistic approach similar to what is used in public health. This article explores the state of the knowledge for how climate change is impacting extreme events and how this information can be used to reduce risk exposure.

Not All Extreme Events Are Created Equal

A discussion of extreme events must first begin by clarifying the term “extreme events.” They represent a category of the most extreme weather and climate phenomena, but are individual types of events (i.e., heat, precipitation, drought, hurricanes). Each event type is unique, and our scientific understanding varies by event type. Much like the term “cancer” is a catchall for a group of diseases involving abnormal cell growth, the science, treatment and impacts of different cancer types vary greatly.

Climate change made the storm surge from Hurricane Sandy worse in 2012, and it also has increased floods caused by high tides by more than 500 percent off the coast of Miami during the past two decades.

Extremes related to temperature and sea level rise are ripe for event attribution because the impact of human-caused climate change on these is well established, and scientists have a strong understanding of how climate change is expected to impact sea levels in the future. For example, it was found that climate change made both the storm surge from Hurricane Sandy worse in 2012 and has increased floods caused by high tides by more than 500 percent off the coast of Miami in the past two decades.5,6 Daily tidal flooding is accelerating in more than 25 Atlantic and Gulf Coast cities.7 Storm surge and tidal flooding, and the economic costs associated with these events, are expected to worsen under future climate change. Heat extremes are regularly found to be exacerbated by climate change. For more on the public health impacts, see “The Vulnerable.”

Precipitation is another area where evidence is mounting that extremes are on the rise due to climate change. Because a warmer atmosphere holds more water, precipitation events are on average expected to increase in intensity by about 6 to 7 percent for each degree Celsius of temperature increase, but this varies by location. In the United States this has translated into a rise in extreme precipitation events, especially in the central and eastern states. For example, the number of extreme two-day precipitation events has increased by 74 percent in the Northeast over the past century.

The number of extreme two-day precipitation events has increased by 74 percent in the Northeast United States during the past century.

The annual Bulletin of the American Meteorological Society report “Explaining Extreme Events” has now looked at more than 130 extreme events from around the world over the past six years. Events examined range from the most common land and ocean temperatures, precipitation, and drought, to events such as forest fires, melting sea ice, extreme sunshine, hurricanes and winter storms. About 65 percent have found a role for climate change on the associated event, while 35 percent did not. Even when a role for climate change is not found, it doesn’t necessarily mean that climate change didn’t contribute in some way. It could mean that the tools currently available don’t allow for the signal to be identified, and as our scientific understanding grows stronger, future work could reveal additional information and understanding.

How Does Event Attribution Work?

The National Academies of Sciences, Engineering and Medicine recently published a detailed analysis of event attribution methodologies.8 One of the most common methods to calculate the change in risk of an event due to human-caused climate change is to employ a statistical approach originally developed in public health called the fraction of attributable risk (FAR).

In public health, for example, this would be used to assess how smoking increases your chances of lung cancer. By comparing cancer risk in a cohort of patients who smoke to cancer rates in a control group of nonsmokers, researchers can determine how much smoking increases the risk of lung cancer.

In event attribution, scientists start by determining the probability of an event (such as a heatwave) occurring in the presence of human-caused climate change. This represents the world we live in today, and the event’s probability is usually established by our observations of Earth’s system. This is then compared to the probability of the event occurring if human-caused climate change had not been present in the world. Since we only have one planet and cannot do a true “control,” this alternative world is based on model runs of a world that only includes natural climate forcing mechanisms and ignores changes driven by human greenhouse gas emissions. By comparing the probability of the event in the existing world with a world that might have been, the change-in-event probability can be quantified.

Events “Not Possible” Without Climate Change

The most recent, and perhaps notable, development in event attribution has been the emergence of events found “not to be possible” without the influence of human-caused climate change. In the report “Explaining Extreme Events of 2016 from a Climate Perspective,” which was released in December 2017, researchers examining three different events concluded they were not possible in a world without human influences on the climate.

In a paper analyzing the 2016 global heat record, researchers concluded that the record global warmth “was only possible due to substantial centennial-scale human-caused warming.”9 A second study investigating the record heat over Asia found that the extreme warmth across Asia in 2016 “would not have been possible without climate change.”10And, finally, researchers studying a large, persistent area of anomalously warm ocean water off the coast of Alaska found “no instances of 2016-like anomalies in the preindustrial climate” for sea surface temperatures in the Bering Sea.11

While remarkable, these results are not surprising. Scientists have long predicted we would eventually reach a point where human-caused climate change altered Earth’s system to such a degree that we would begin to see weather and climate events that would not have been possible without human contributions. It was also predicted that these first events would be related to high temperatures in the ocean and atmosphere, where the impact of human activity on our climate is most strongly observed.

What does it mean when stated that an event was not possible without human-caused climate change? It is important to remember that every event is still built on a foundation of natural variability. This includes the types of drivers that your local broadcast meteorologist highlights on maps during the evening news, including atmospheric circulation patterns of highs, lows, blocking events, La Niña and El Niño, sea surface temperature and so on. So it is likely that 2016 would still have been warm globally, but with the added influence of human-caused climate change, the temperatures surpassed a threshold they otherwise couldn’t have. The building blocks for the event were laid by the types of drivers that normally cause heat events to occur, but climate change pushed the event to reach temperatures it otherwise could not have.

Understanding Impacts and Informing Risk: The Emerging Discipline of “Impacts Attribution”

In recent years, interest has been growing in the area of “impacts attribution,” which aims to draw a line connecting climate change to the altered risk of an extreme event and the subsequent impact of the event. Impacts attribution increasingly is being recognized as the next major area of advancement for attribution science to enhance the field’s ability to connect to indices relevant to people. Impacts from extremes are determined not just by the event itself, but also by the vulnerability of the people and assets exposed to the event. By improving our understanding of how extremes are changing, governments, planners, businesses, communities and individuals can improve their preparedness and minimize the costs of these events.

The first paper to do this for human health was published in 2016 when attribution scientists partnered with public health officials to assess the role climate change played in increased mortality from a specific event—the 2003 European heatwave.12 Their results concluded that in the summer of 2003, “out of the estimated ~315 and ~735 summer deaths directly attributed to the heatwave event in Greater London and Central Paris, respectively, 64 (+/–3) deaths were attributable to anthropogenic climate change in London, and 506 (+/–51) in Paris.” For the first time, a methodology had been established for connecting climate change to the mortality of a specific heat event. Clearly, multiple approaches could be taken to address these questions and the paper lays out just one, but it was an important first step in quantifying the impact of climate change on health outcomes.

An area of active research has been to link specific heat events to health indicators such as insurance claims and hospital admissions. The goal is to help us better understand which heat events are having significant health impacts, and then to assess whether these events are expected to increase due to climate change—that way we can inform public health responses to heat events in the future. For more information about how climate change is expected to affect public health, see “Double Threat.”

Drawing connections to environmental impacts is also an area where research is beginning to emerge. For example, in a study of the 2016 extreme Great Barrier Reef bleaching event, researchers concluded the risk of the extreme event was increased through anomalously high sea surface temperature and the accumulation of thermal stress as a result of human-caused climate change.13 Two other papers authored by NOAA scientists showed how rising ocean temperatures, caused in part by human-caused climate change, impacted living marine resources and were linked to coral bleaching, reduced fish stocks and a decrease in seabird counts in the California current and the equatorial Pacific.14,15

While these papers represent early approaches, they are important first steps in improving our understanding of how climate change has, and is expected to, impact important socioeconomic outcomes including public health, the environment, infrastructure and the economy more broadly. This work will continue to be important in exploring ways to enhance our resilience to future extreme events and to minimize the impacts of extremes on our economy and society.

Stephanie C. Herring, Ph.D., is a climate scientist at NOAA’s National Centers for Environmental Information, Center for Weather and Climate, in Boulder, Colorado.