Trends in coastal storm impacts are not only about changes in hurricanes

Coastal storms are among the most important natural hazards affecting the densely populated eastern seaboard of the United States. Such storms are capable of producing property damage and even loss of life through their strong winds, heavy precipitation, and storm surge flooding. Tropical cyclones (i.e., hurricanes and tropical storms) are well known for producing such impacts. For example, the Great Hurricane of 1938 devastated southeastern New England with a prodigious storm surge, with water estimated to reach 18 feet above normal at Fall River, MA (1). But extratropical cyclones (ETCs), commonly called “nor’easters” in New England and the mid-Atlantic states because winds blow from the northeast as they move offshore of those regions, can bring similar impacts on a more frequent basis. Chen et al. (2) analyzed over 80 y of nor’easter activity to determine whether there were any trends in their frequency or intensity. They found statistically significant trends in the maximum wind speeds of the most intense nor’easters, along with an increasing trend in area-averaged precipitation intensity.

Given the magnitude of the impacts of such storms and the possibility that their characteristics may change in a changing climate, the work by Chen et al. is an important step in understanding how strong ETCs may respond to anthropogenic climate change.

The findings of Chen et al. are important in the context of a changing climate. There is evidence that warming oceans will lead to stronger tropical cyclones (3). The connection between oceanic warmth and tropical cyclone intensity is not surprising, given that these storms derive their energy from heat stored in the upper ocean (4). The relationship between climate change and extratropical cyclones is more complicated, because the primary energy source for ETCs is the contrast in temperature that typically exists in middle latitudes (5). Climate change is weakening that contrast as the Arctic warms more rapidly than lower latitudes (6), and it may also be influencing the position of storm tracks (i.e., the regions with the highest density of extratropical storm activity) (7). Because of these complexities, there is not a straightforward expectation of how nor’easters affecting the Atlantic coast will respond to a changing climate.

Coastal flooding is a significant hazard associated with nor’easters. Although rising sea levels have led to more frequent “high tide flooding” or “sunny day flooding,” the most impactful coastal flooding events occur when high winds drive storm surge events. At coastal locations in the northeastern United States, more than 70% of the largest storm surges were associated with ETCs (8). Some of the most impactful storm surge events in New England and the mid-Atlantic states have occurred when slow-moving or nearly stationary ETCs persisted across multiple high-tide cycles.

In March 1962, a devastating storm surge affected the mid-Atlantic region when a moderately deep ETC with multiple centers combined with a blocking anticyclone over eastern Canada to establish a strong pressure gradient, producing a long fetch of onshore winds, occasionally gusting to hurricane force, that continued through 4 to 5 high tide cycles (Fig. 1). Combined with the effects of high astronomical tides (perigean new moon), the Great Atlantic Storm of March 1962 caused extensive damage and loss of life from Virginia through New York and the offshore waters (9).

Fig. 1.

Surface weather map for the Great Atlantic Storm of 1962 (1200 UT, March 7, 1962) from ERA5 reanalysis (10). Contour lines are isobars of sea level pressure (hPa). Winds blow approximately parallel to the isobars with strength inversely proportional to the spacing between them. Note the large area of strong winds implied by the tightly spaced isobars north of the low-pressure system, indicating a long fetch of onshore winds directed at the mid-Atlantic coast.

The Northeast Blizzard of 1978 is another example of a nor’easter that became quasistationary and extensively impacted both inland and coastal regions. The synoptic evolution of this storm was unusual. The storm originated as a weak system off the southeast US coast and did not intensify until a strong cyclone in the upper atmosphere originating in high latitudes moved south to southeastern Virginia. Although the exceptional snowfall produced by this February 1978 storm paralyzed a large area and accounted for the most notable of the cyclone’s impacts, the storm also induced a powerful storm surge that continued through several high tide cycles, accompanied by peak wind gusts above hurricane force and massive breaking waves. Extensive destruction occurred along coastal areas, particularly in southern New England (11, 12).

The Great Atlantic Storm of 1962 and the Northeast Blizzard of 1978 serve as examples of especially powerful nor’easters that can produce high winds, heavy precipitation (often in the form of snow), and storm surge flooding. Given the magnitude of the impacts of such storms and the possibility that their characteristics may change in a changing climate, the work by Chen et al. is an important step in understanding how strong ETCs may respond to anthropogenic climate change. The concepts of detection and attribution make up a well-established paradigm for evaluating whether a particular aspect of climate has changed (13). In this paradigm, detection is the process by which some climatic property is shown to have changed in a statistically significant way. The process of detection does not require the identification of the underlying cause of the change. In the context of this paradigm, which has been widely applied to other climatic variables, Chen et al. have detected an increase in the intensity of the strongest nor’easters along the US East Coast, in terms of both their maximum wind speeds and area-averaged precipitation rate.

But is this change in nor’easter intensity a consequence of increasing greenhouse gas concentrations in the atmosphere? Answering this question requires the other component of the detection-attribution paradigm. Attribution seeks to determine the most likely cause of the detected change with a specified level of statistical confidence (13). For example, the landmark work of Santer et al. (14) established an anthropogenic influence on the vertical temperature structure of the atmosphere. The trend in nor’easter intensity found by Chen et al. has not yet been attributed to a specific cause in this formal definition of attribution. Indeed, the attribution of increased ETC intensity to human-caused climate change could be regarded as counterintuitive, as the equator-to-pole temperature gradient in the lower atmosphere has been reduced as the Arctic warms more rapidly than the tropics, a phenomenon referred to as “Arctic amplification” (15).

Chen et al. suggest two mechanisms that could induce the observed increase in nor’easter intensity, both related to increasing oceanic warmth. One mechanism is an increase in temperature contrast between land and ocean when Arctic air drives south over eastern North America in proximity to a warmer ocean. This mechanism would require a smaller warming of the Arctic air relative to the warming of the offshore waters, which may be difficult to reconcile with Arctic amplification. The second mechanism is increased latent heating that is realized as water evaporated from the warmer ocean is released as it condenses in the overlying atmosphere. Latent heat release has been shown to be an important source of energy in high-impact nor’easters (16). Furthermore, simulations in which higher-resolution atmospheric models are conditioned with large-scale environmental changes taken from coupled atmosphere–ocean models driven by increases in greenhouse gases (so-called “pseudo-global warming” experiments), have shown increases in ETC intensity resulting from enhanced latent heating (17).

There is still much to learn about the potential impacts of climate change on ETCs. As sea levels continue to rise, the impacts of coastal flooding from these surge-producing storms, which occur more frequently than tropical cyclones, will grow. The study by Chen et al. should inspire future work on detecting changes in ETC frequency and intensity in other parts of the world, and it should also inspire future efforts to attribute such changes to specific causes.