1. Introduction
From 26 to 30 June 2021 an exceptional heatwave affected the Pacific Northwest (PNW) region of North America with the U.S. states of Washington and Oregon and the Canadian provinces of British Columbia and Alberta experiencing the most extreme heat. Many all-time high temperature records were broken, some by several degrees. For example, Portland, Oregon, recorded a high of 46.7°C on 28 June shattering the previous all-time record high of 41.7°C. The state of Washington set a new state record of 48.9°C at Hanford (Miller and Bair 2022) while Oregon tied its previous record of 48.3°C at both Pelton Dam and Moody Farms (Vescio and Bair 2022). Lytton, British Columbia set a new national high temperature record for Canada on three consecutive days, peaking on 29 June at 49.6°C (ECCC 2021). The event was associated with numerous severe societal and environmental impacts including significant loss of human and marine life and damage to vegetation (Chang et al. 2021; Raymond et al. 2022). Here we diagnose some of the key meteorological drivers of this event.
Despite its reputation for having a generally mild climate, the PNW has historically experienced episodes of extreme heat. West of the Cascade Mountains, including the densely populated Willamette Valley of Oregon and the Puget Sound region of Washington, the underlying temperature frequency distribution is positively skewed, with a long warm-side tail, resulting from the occurrence of rare, but extreme (in an anomalous sense) heatwaves (Loikith and Broccoli 2012; Cavanaugh and Shen 2014; Loikith and Neelin 2015; McKinnon and Simpson 2022). Here, the moderating effects of the Pacific Ocean dominate summer climatology; however, occasional offshore wind events and/or strong suppression of the Pacific marine layer by large-scale subsidence can lead to substantial warm excursions from the mean (Catalano et al. 2021). East of the Cascades, the temperature distribution is negatively skewed as the cooling effect of the Pacific Ocean is diminished.
While no two events are exactly alike, heatwaves in the PNW are generally associated with a set of common meteorological features (Bumbaco et al. 2013; Brewer and Mass 2016a). Aloft, a strong ridge of high pressure is key, resulting in large-scale subsidence and associated adiabatic warming while also suppressing the Pacific Ocean marine layer, thus limiting the inland extent of marine cooling. At the surface, a thermally induced trough of low pressure is a critical component of heatwaves (Cramer and Lynott 1970; Brewer et al. 2012, 2013). The development, location, and evolution of the thermal trough strongly influences local heat magnitude, especially west of the Cascades where the axis of the trough determines whether a location receives cooler and cloudier onshore flow from the west or warmer offshore flow from the east. As the surface thermal trough shifts to the east of the Cascades, onshore flow resumes across western parts of the region, ending anomalous heat there (Brewer et al. 2013). East of the Cascades and the British Columbia Coast Mountains, the cooling effect of the Pacific Ocean is minimal or nonexistent. Here, subsidence under a strong ridge of high pressure with associated clear skies, maximizing solar heating, is key.
In the context of past PNW heatwaves, the June 2021 event was truly exceptional with no precedent in the observational record. It was so extreme that it is difficult to estimate a return period for an event of this magnitude (Philip et al. 2022). Using the Community Earth System Model Large Ensemble, McKinnon and Simpson (2022) estimate a return interval of a remarkable ∼100 000 years for temperatures that deviated from the mean by the magnitude of the most affected areas in June 2021. Evidence is clear that global warming is reducing the return interval of an event of this magnitude and very likely made the event hotter than if it had occurred in the past (McKinnon and Simpson 2022; Thompson et al. 2022). To improve our physical understanding of this event, here we provide a detailed meteorological analysis of the progression of the heatwave. We also compare the atmospheric conditions associated with the 2021 event with other historical, highly amplified warm-season ridging patterns to provide insight into why this event was so much more extreme than other events in the region.
2. Data and methods
Meteorological data are from the ERA5, provided on a 0.25° latitude–longitude grid at hourly temporal resolution spanning 1979–2021 (Hersbach et al. 2020). Anomalies, when presented, are computed using a 1991–2020 climatology. Historical analyses use ERA5 data extending to 1950 (section 3d) on a 0.5° latitude–longitude grid to maintain consistency with the available resolution of the 1950–1978 ERA5 “back extension” data (Bell et al. 2021). Station data are from the Global Historical Climatology Network–Daily (GHCN-D; Menne et al. 2012). Air parcel backward trajectories are from the NOAA Air Resources Laboratory’s Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model (Stein et al. 2015), run on the HYSPLIT online interface (https://www.ready.noaa.gov/HYSPLIT_traj.php). The North American Mesoscale Model (NAM) 12-km meteorological dataset was used to compute the back trajectories. We note that while we use ERA5 for other analyses, it is not an available dataset on the HYSPLIT online portal.
3. Results
Figure 1a shows the maximum temperatures recorded during the event (26–30 June) at stations that set or tied all-time high temperature records across the region. Only stations including at least 30 years of data with less than 10% of all days missing were included in the plot. All-time records were set across a very large region spanning Alberta to northern California. Many stations recorded temperatures exceeding 43°C in British Columbia, Washington, Oregon, and northern California with some exceeding 47°C. Figure 1b shows by how many degrees each station broke its previous all-time high temperature record. Several stations across northwestern Oregon and central and western Washington broke their record by more than 4°C. Remarkably, a handful of stations broke their previous all-time record high by more than 6°C with Kamloops, British Columbia, breaking its previous record by 6.5°C (period of record since 1951).
(a) Maximum temperature recorded between 26 and 30 Jun (°C) for stations that set or tied all-time high temperature records. (b) Degrees by which each station broke their previous record. For stations that set or tied their previous all-time high temperature record more than once during the event, as was common, the value plotted is the difference between the highest temperature within the span of the event and the all-time record high temperature from before 26 Jun 2021.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
a. Atmospheric circulation
The progression of the midtropospheric circulation leading up to the heatwave is shown in Fig. 2 using 500-hPa geopotential height (Z500) anomalies. The start of the event roughly coincides with Fig. 2g with positive Z500 anomalies strengthening dramatically from 25 to 27 June. However, the area of positive Z500 anomalies can be traced back in time by several days. Starting on 19 June (Fig. 2a), an area of positive Z500 anomalies associated with an atmospheric ridge was present over much of the western United States and southwestern Canada. This region of positive anomalies was nearly stationary over the course of the next several days while producing hot weather in the PNW, including daily record maximums of 36.1°C in Portland and 31.7°C in Seattle on 21 June. At the same time, an area of negative Z500 anomalies associated with a large low pressure trough developed and strengthened to the west of the anomalous ridge, peaking on 24 June (Fig. 2f). The large Z500 gradient that developed between the trough and ridge facilitated strong poleward moisture transport via a trans-Pacific atmospheric river, making landfall over southeastern Alaska with local daily precipitation amounts exceeding 50 mm (not shown). Mo et al. (2022) show that this atmospheric river transported moisture and sensible heat into the ridge. Diabatic heating within the developing associated cyclone excited downstream zonal wave activity flux, amplifying the Rossby wave train and strengthening the downstream ridge (Neal et al. 2022). These features have been further connected to variability within the East Asian monsoon system (Qian et al. 2022; Lin et al. 2022). In subsequent figures, we focus on the regional evolution of the meteorology associated with the event.
Daily mean Z500 anomalies spanning 19–27 Jun.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
Figures 3–7 show key meteorological variables across the PNW and southwestern Canada for each day of the event. Starting at the beginning of the heatwave, Fig. 3a shows the Z500 field at 0000 UTC and daily Z500 anomalies for 26 June. A highly amplified ridge is evident, with the highest Z500 values centered roughly over Vancouver Island and the largest positive anomalies just to the north. A substantial region, outlined in magenta, recorded the all-time highest daily mean Z500 values (among all calendar days) in the ERA5 1979–2021 record on 26 June. Areas to the south of the ridge center, including the southern half of British Columbia and the PNW, were under the influence of large-scale offshore midtropospheric flow inferred from the geostrophic wind within the Z500 gradient. At the surface, Fig. 3b shows an area of high sea level pressure (SLP) over the northeast Pacific, extending into the northern periphery of the ridge center over British Columbia. A surface thermally induced trough is evident over California and southern Oregon. Contours in Fig. 3b indicate weak easterly wind anomalies at 850 hPa (U850) along the Washington coast with stronger anomalies offshore. Figure 3d shows that the greatest positive 850-hPa temperature (T850) anomalies occurred under and just to the south of the Z500 ridge center where parts of British Columbia were around 15°C warmer than the 1991–2020 mean. Most of the PNW was experiencing anomalous warmth, but not yet experiencing record-breaking heat at this point in the event, corresponding to the afternoon of 25 June local time.
Meteorological variables for 0000 UTC 26 Jun and daily anomalies. (a) Z500 total field shaded (in m). Z500 anomalies contoured every 20 m. The magenta outline highlights grid cells that recorded all-time high daily mean Z500 values, among all calendar days, in the ERA5 record (1979–2021). (b) SLP (shaded, in hPa). The dashed orange line depicts the approximate location of the surface thermally induced trough axis. U850 anomalies are contoured every 4 m s−1. Only negative anomalies are plotted to highlight easterly wind anomalies. (c) Z500 zonal anomalies contoured every 20 m. Zonal anomalies are computed by subtracting the daily mean Z500 value at each grid cell from the mean of all global grid cells along the same latitude band. Shading indicates where the Z500 zonal anomaly was in the top six highest for all June, July, and August days in the ERA5 record. A value of one means that at that grid cell, the Z500 zonal anomaly on 26 Jun was higher than on any other previous day in June, July, or August in the ERA5 record. (d) T850 anomalies in °C.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
Not only was the ridge notable in having the highest Z500 in the reanalysis record, but Fig. 3c shows that the ridge was also the most amplified, as defined by Z500 zonal anomalies, in the 1979–2021 ERA5 record when considering all June, July, and August days. The Z500 zonal anomalies plotted in Fig. 3c are computed by subtracting the latitudinal mean Z500 value from each grid cell as in Liu et al. (2018) and Loikith et al. (2022). Grid cells shaded in magenta in central British Columbia recorded their largest daily zonal anomaly in the ERA5 record (for meteorological summer) on 26 June with other shaded grid cells ranking between the second and sixth highest. At its center, the ridge was more than 240 m higher than the zonal mean Z500. That such extreme amplification occurred in late June is remarkable, as western North America tends to see more amplified ridging during winter months due to the interaction of seasonally stronger westerly flow with continental orography and greater baroclinicity (Blackmon et al. 1977; Branstator 1992). It is worth noting that zonal Z500 anomalies are not influenced by thermal expansion of the troposphere due to global warming while Z500 total field is. This means that while a warming atmosphere could have made a contribution to the record Z500 values in Fig. 3a (contribution not analyzed here), the record high zonal anomalies in Fig. 3c indicate that this ridge was also exceptional for summer over the region without a direct thermodynamic contribution from global warming.
Stepping forward in time, Fig. 4 shows the same meteorological variables for 27 June. At this point the ridge had reached its peak strength as measured by Z500 and Z500 anomalies (Fig. 4a). A large region spanning much of British Columbia, Alberta, Washington, and a sliver of northern Oregon recorded the highest Z500 values in the ERA5 record. Z500 anomalies exceeding +320 m were present over central British Columbia with large positive anomalies covering a substantial portion of western North America. Also evident in Fig. 4a is a cutoff Z500 low west of central California. The presence of this low increased the height gradient over western North America and contributed to large-scale offshore flow in the midtroposphere. A surface thermally induced trough was evident over western Oregon (Fig. 4b) with easterly U850 anomalies over parts of Oregon and Washington. These features coincided with a broad region of T850 anomalies between +10° and +20°C from Northern California to northern British Columbia. At this point, the most positive temperature anomalies were located over western British Columbia. The ridge continued to set records on 27 June for the largest Z500 zonal anomalies, exceeding 240 m, in the ERA5 record over northeastern British Columbia and northwestern Alberta.
As in Fig. 3, but for 27 Jun.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
On 28 June (Fig. 5), the ridge began to weaken while surface heating strengthened. While the Z500 ridge was still very strong, with anomalies exceeding 260 m over British Columbia and Alberta (Fig. 5a), only a very small area over southeastern British Columbia recorded the highest Z500 values in the full (all calendar days) 1979–2021 ERA5 record. The upper-level low was slightly northeast of its previous-day location, leading to a very strong Z500 gradient over the western United States. At the surface, a very prominent thermally induced trough is evident, extending from the Central Valley of California northward into western Washington (Fig. 5b). The solid contours in Fig. 5b indicate easterly wind anomalies, with some exceeding −8 m s−1 over Washington and Oregon, inhibiting the cooling effect that would be brought by onshore flow and aiding surface heating west of the Cascades through orographically induced adiabatic warming. While no grid cell recorded the highest summer Z500 zonal anomaly on 28 June, much of southern British Columbia and parts of western Alberta recorded values within the top 6 days in the ERA5 record.
As in Fig. 3, but for 28 Jun.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
The highest temperatures of the event in most places west of the Cascades were recorded on the afternoon of 28 June, coinciding roughly with the 0000 UTC 29 June conditions in Fig. 6. The Z500 ridge continued to gradually weaken and shift to the east between 28 and 29 June (Fig. 6a), although it was still 260 m above the long-term average over central British Columbia and west-central Alberta. The upper-level low was farther north, closer to the ridge, and although it weakened compared with the previous day, its presence resulted in a very strong Z500 height and anomaly gradient over the PNW, driving strong offshore flow in the midlevels. At the surface, thermally induced low pressure was evident across the region with a clear trough axis extending northward from the Central Valley of California into western Oregon with a northeastern extension into the Columbia basin of Oregon and Washington (Fig. 6b). Very strong easterly wind anomalies were also present over Oregon and Washington, with anomaly values exceeding −8 m s−1 over much of the region. This easterly flow at lower levels continued to enhance surface warming west of mountain ranges, especially the Cascades, as the air descended the topography and warmed adiabatically. Z500 zonal anomalies continued to be ranked in the top six daily summer values in the ERA5 record over southeastern British Columbia and southwestern Alberta, emphasizing the persistence of this extraordinarily amplified ridge. Widespread T850 anomalies near or exceeding +20°C were common at this time.
As in Fig. 3, but for 29 Jun.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
As the ridge gradually shifted east and weakened (in an absolute strength sense) into 30 June, record high Z500 values were recorded over northeastern Alberta and far western Saskatchewan (Fig. 7a). A strong Z500 gradient was still present over the northwestern United States, enhanced near the coast by the weakening upper-level low. Interestingly, the upper-level low was characterized by positive Z500 anomalies on 30 June, even though it was a closed low pressure system. At the surface, the thermally induced surface trough shifted east compared to the previous day, with the lowest SLP over the Columbia basin. This shift of the thermally induced trough, combined with the approach of the cutoff low pressure allowed an inland surge of cooler marine air into western Oregon and Washington. Dramatic cooling resulted, with overnight temperature falls of 31.1°C in Salem, Oregon (from 47.2°C on 28 June to 16.1°C on the morning of 29 June) and 29.5°C in Portland, Oregon (from 46.7° to 17.2°C), ending the extreme heat there. However, extreme heat persisted over and east of the Cascades. The highest surface temperatures of the event were recorded around 0000 UTC 30 June (corresponding to the afternoon of 29 June) over interior parts of Washington, Oregon, and British Columbia, including the state-record maximums (for Washington and Oregon) and the national-record maximum for Canada. This is evidenced by the reduced magnitude of the positive T850 anomalies over western Oregon but with very high positive anomalies inland and to the north (Fig. 7d). The ridge, while exhibiting lower zonal anomalies than before, was still resulting in zonal anomalies ranking in the top six within the entire ERA5 summer record over Alberta and Saskatchewan (Fig. 7c).
As in Fig. 3, but for 30 Jun.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
b. Air parcel back trajectories
Figure 8 shows 48-h backward trajectories for air parcels originating at Seattle (nearest grid point to KSEA) for seven different origin times throughout the event. The origin times are colored as shown in the legend in Fig. 8a and represent the point at which the air parcel arrived at the Seattle grid cell. For example, the first time step, shown in red, traces an air parcel backward in time and space from 0000 UTC 27 June at Seattle for 48 h. This parcel was located over central British Columbia 48 h prior to the origin time and traveled to the south and across Puget Sound before arriving in Seattle at 0000 UTC 27 June. Figure 8c shows the temperature and potential temperature of each air parcel throughout the 48-h trajectory. This 0000 UTC 27 June red trajectory had a large difference between its temperature and potential temperature 48 h prior to the origin time. The temperature then warmed while the potential temperature cooled slightly. The warming of the temperature was due to adiabatic processes as shown in Fig. 8b. The parcel was roughly 3000 m above sea level 48 h prior to the origin time and descended to roughly 500 m above sea level within 12 h. Around 20 h before the origin time, the temperature and potential temperature of the parcel cooled, likely as it traveled over open water. In the last 8 h before arrival in Seattle, the parcel underwent considerable diabatic warming, shown by the increase in temperature during the final 10 h, even though it was over open water near sea level. This warming coincides with daylight hours and likely occurs over water due to large-scale subsidence suppressing the marine air layer. Additionally, downward mixing of warmer air above the surface may have contributed to further warming of the parcel under a deep atmospheric boundary layer (Schumacher et al. 2022). This is further supported by plotting back trajectories for the same time starting just above the surface which arrive at Seattle from the east (not shown).
The 48-h backward trajectories for air parcels originating at Seattle. (a) Map of parcel locations going backward in space and time from the origin city (identified by the yellow star) with each symbol on the lines representing a 6-h increment. All trajectories originate at ground level except for the cyan line with × symbols, which originates at 400 m above ground level to highlight near-surface offshore flow at 0000 UTC 29 Jun. The lines are colored according to the origin time of the back trajectory. For example, the 0000 UTC 27 Jun red line shows the path of the parcel going backward in time from 0000 UTC 27 Jun at the yellow star to 0000 UTC 25 Jun at the end of the red line. The inset map in the upper left is zoomed in on the Puget Sound and southern Vancouver Island. (b) The parcel elevation above sea level along its path to the origin location. (c) Parcel temperature (solid lines) and potential temperature (dashed lines) along the trajectory. Hour 0, on the left side of the x axis is the point where the parcel arrives at its origin (Seattle in this case). Blue horizontal lines mark 12-h increments.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
The next two timesteps show some similarities to the red one, although the two parcels traveled much shorter distances during 48 h. The 0000 UTC 28 June parcel (green) experienced a similar daytime diabatic warming prior to arrival at Seattle despite being over the Puget Sound. The 1200 UTC 28 June parcel (black) shows a completely different trajectory, traveling from northern Alberta and covering a long distance during the 48-h period. Additionally, this parcel went through dramatic descent starting around 30 h before the origin time, descending from over 4000 m to roughly 500 m before ascending and then descending the Cascade Mountains and arriving in Seattle. Notable adiabatic warming is clearly evident with the descent while the potential temperature does not change. The parcel reached remarkable temperatures of nearly 50°C over eastern Washington during the daytime hours approximately 12 h before the origin time. The parcel then cooled overnight before arriving in Seattle early in the morning of 28 June. If this parcel trajectory were to have occurred 12 h prior or after, when the parcel would arrive during maximum daytime heating, it is plausible that temperatures in Seattle would have been even hotter than actually observed during this event.
By the afternoon of 28 June local time (0000 UTC 29 June), the cyan-colored surface-based trajectory shows a return of northwesterly flow. However, this was the hottest parcel upon arrival in Seattle. A parcel originating at the same time at 400 m above ground level (cyan with × symbols in Fig. 8a) shows a continuation of easterly flow. This helps explain the high surface temperature at Seattle despite low-level onshore flow. Vertical mixing of hot air from just above the surface promoted by daytime heating and strong suppression of cooler marine air combined to heat the air at the surface in Seattle. By the morning of 29 June, a wind reversal is evident in the final two trajectories. Air travels northward along the Pacific coast, just offshore, nearly at sea level before moving inland and arriving in Seattle. The daytime parcel (gray) still experiences considerable warming over land before arriving in Seattle, but the marine origin of the parcel limited the magnitude of this warming. These two trajectories clearly illustrate the coastally trapped wind reversal phenomenon that typically ends PNW heatwaves, coinciding with the surface thermally induced trough moving eastward (Brewer et al. 2013).
Backward trajectories for parcels originating at Portland, Oregon (grid cell closest to KPDX) are shown in Fig. 9. All air parcels traveled in an anticyclonic pattern from 0000 UTC 27 June to 0000 UTC 29 June. These parcels broadly followed the large-scale geopotential height gradient as shown in Figs. 3–7. With the exception of the parcel with origin time of 1200 UTC 27 June (blue), these parcels traveled long distances to reach Portland. Figures 8b,c show that these parcels underwent considerable adiabatic warming along these trajectories, descending from 4000 to 5000 m to sea level at Portland. Potential temperature is relatively constant throughout the trajectories with air temperature warming along with the descent, demonstrating the adiabatic warming contribution to the heat at Portland. By 1200 UTC 29 June a southerly wind reversal had occurred causing parcels to follow a path along the coast from Northern California to Portland, bringing an end to the extreme heat. This coincides with the eastward shift of the surface thermally induced trough in Fig. 7b.
As in Fig. 8, but for parcels originating at Portland, Oregon.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
Figure 10 shows backward trajectories for Pendleton, OR (grid cell nearest to KPDT) starting with 1200 UTC 27 June. Similar to Portland, air parcels followed an anticyclonic pathway to Pendleton, descending from over 4000 m and warming adiabatically. As the ridge and thermally induced trough shifted eastward on 30 June, air parcels originating over the Pacific Ocean were transported into the Pendleton area bringing an end to the heatwave there, roughly 24 h later than at Portland.
As in Fig. 8, but for Pendleton, Oregon. Trajectory origin times are shifted forward in time by 12 h compared with Seattle and Portland.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
The highest temperatures of the four trajectory examples were recorded at Lytton, British Columbia (grid cell closest to 50.23°N 121.58°W; Fig. 11). Lytton was more closely collocated with the ridge center than Seattle, Portland, or Pendleton and therefore the trajectories exhibited different behavior. Near the ridge center, pressure/height gradients were weaker, resulting in lighter winds and air parcels that traveled much shorter distances in the days prior to arrival in Lytton. For this reason, Fig. 11 displays 72-h back trajectories. Parcels traveled from the north and northwest to Lytton for the first four origin times and then the trajectories shifted to the southeast for the later part of the event. However, this change in trajectory direction had little apparent effect on the parcel temperatures. Rather, the trajectories show that the air parcels warmed consistently during the 72 h prior to arrival with each subsequent parcel being warmer than the previous. For all parcels, the regular temperature oscillations reflect the diurnal cycle as adiabatic changes were small. As opposed to locations farther south where large-scale subsidence caused strong adiabatic warming in the 48 h prior to the origin times, here strong diabatic heating under the record-breaking ridge, without any advection of cooler air, led to the extreme high temperatures. These examples illustrate that depending on location relative to the ridge, different physical processes were important for driving temperatures to such extreme levels throughout the geographic span of this event.
As in Fig. 8, but for parcels originating at Lytton, British Columbia. Trajectories are 72 h long instead of 48.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
c. Role of large-scale horizontal temperature advection
The backward trajectories strongly support subsidence under the exceptionally strong midtropospheric ridge as a major driver of the extremely warm temperatures experienced at the surface to the south of the ridge center. Critically for locations west of the Cascades, offshore flow at most levels induced additional adiabatic warming as air parcels descended the Cascades and further prevented marine cooling until the surface thermally induced trough shifted eastward. In all locations, maximized solar heating inevitably played a major role as the event occurred in a cloudless sky near the summer solstice. Here we investigate whether large-scale horizontal transport of warmer air into the region, for example, from the U.S. Southwest, which also experienced a record-breaking heatwave before this event (Osman et al. 2022) also played a role.
Figures 12 and 13 show daily mean temperature and wind direction at 850 and 700 hPa, respectively, for the 4 days spanning the core of the event. At the beginning of the event on 27 June, there is no evidence of meaningful large-scale warm-air advection at either level into Oregon, Washington, or British Columbia as flow at both levels is generally from the north, thus not originating from locations with climatologically hotter lower-tropospheric air (i.e., the U.S. Southwest) (Figs. 12a, 13a). As the lower troposphere warmed in the following days, there continued to be little evidence of broad transport of warmer air into the region or even a source of hotter air available for transport. Some localized warm-air advection at 700 hPa is apparent on 29 and 30 June over parts of northwestern Oregon and southern British Columbia (Fig. 13c), although because 700-hPa winds were light the actual advection magnitude is minimal (Figs. S1 and S2 in the online supplemental material). On 30 June, Fig. 13d suggests some larger-scale warm-air advection at 700 hPa over northern and central Alberta (Fig. 13d), although again very light winds resulted in minimal advection magnitude here (Fig. S2). Overall, Figs. 12 and 13 largely point to a lack of any large-scale transport of a warmer air mass in the lower troposphere to areas affected by the extreme heat. Localized contributions, mostly confined to 29 June and over Alberta on 30 June are apparent and may have been somewhat influential within the heatwave region.
T850 (shaded, in °C) and 850-hPa wind direction (vectors).
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
As in Fig. 12, but for 700 hPa.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
Bartusek et al. (2022) performed a temperature tendency budget analysis at 850 hPa over the region, further quantifying the role of adiabatic and nonadiabatic processes (see their extended data; Fig. 5). Their findings are consistent with our analysis showing essentially no contribution from horizontal advection at the regional scale, with more substantial contributions over smaller subregions within the heatwave (especially over central Washington and southern British Columbia during the latter part of the event). Across the heatwave region itself, they show that adiabatic and diabatic warming processes which are not accounted for in the temperature tendency budget, and which may be due, at least in part, to land–atmosphere processes, play a dominant role. We also note that upstream heating resulting from a strong atmospheric river prior to the heatwave increased potential temperature in the vicinity of the source regions of many of the air parcels, which, while different than large-scale horizontal advection of warmer air along a pressure surface, could be argued as a mechanism for transporting heat into the affected region (Mo et al. 2022; Schumacher et al. 2022).
d. Comparison with historical events
The above analysis helps diagnose the meteorology that drove this particular extreme heat event. Many of the meteorological features were qualitatively similar to those known to be associated with other heatwaves in the PNW (i.e., large-scale subsidence, offshore flow, and solar heating under the influence of a strong ridge), yet temperatures were far warmer than any prior heatwave in the observational record. One likely contributing explanation for this is the strength and placement of the record breaking Z500 ridge, both in the sense of total field geopotential height and zonal anomalies. However, the degree to which the driving meteorology for this event differed from previous, less extreme heat events is not immediately clear.
To investigate more closely why this event was so different, we compare several key meteorological variables at the time of the peak of the Z500 ridge (27 June) to those present during other notable, highly amplified ridging events centered over a similar region during the 1950–2021 period. We restrict this analysis to 15 June–15 August, as the most intense heatwaves in PNW history have occurred during this time of year with the greatest impacts to human health and the environment (Bumbaco et al. 2013). Here, we use maximum Z500 zonal anomalies within a box spanning 49°–55°N, 127°–116°W, roughly encompassing central and southern British Columbia (Fig. 14a). To account for multiple consecutive days of highly amplified ridging conditions, we extract only the peak date of each amplified ridging event, defined as the day with the highest positive zonal anomalies in 15-day windows.
(left) The difference between 27 Jun 2021 and the average of the other 49 most amplified ridge events as defined by maximum Z500 zonal anomalies over a box bounded by 49°–55°N, 127°–116°W (shown in the red box in the top-left panel). For multiday events, only the day with the maximum zonal anomaly is included in the difference calculation. Differences are (from top to bottom) for SLP, T850, U850, and Z500. (right) As in the left column, but only the other nine most amplified ridge events are included in the differences calculation.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
Using these criteria, 27 June 2021 represents the peak date of the third-most amplified ridge in the period of record with the only instances of more amplified ridging occurring in 1969 and 1985 (Table 1). However, these ridges were not nearly as strong in an absolute sense with peak Z500 of 5850 and 5884 m, respectively, compared to 5981 m in the June 2021 event, and their large amplitude was therefore a by-product of lower hemispheric Z500 in their latitude bands from which zonal anomalies were calculated (Table 1). This table further demonstrates that the June 2021 ridge was exceptional in its combination of strength (Z500 total field) and amplitude (Z500 zonal anomalies) over this region. Even compared to another exceptional ridge that peaked on 27 July 2009, and produced numerous all-time record highs (Bumbaco et al. 2013; Stewart et al. 2017), including Seattle’s previous record of 39.4°C (broken in the June 2021 event), the June 2021 ridge was much stronger (+45 m) and even more amplified (Table 1).
Top-10 most amplified ridging events between 15 Jun and 15 Aug 1950–2021, over the box bounded by 49°–55°N, 127°–116°W, as measured by Z500 zonal anomalies. Dates are ranked according to the maximum Z500 zonal anomaly. For multiday events, only the date with the maximum Z500 zonal anomaly is shown. The maximum daily averaged absolute value of Z500 within the box is also listed, along with its daily rank among all days in the period (n = 4464). The event in boldface font is the focus of this paper.
Figure 14a shows the difference between the Z500 field on 27 June and the average of the other 49 most amplified ridge days. The Z500 values were remarkably higher in the June 2021 event, by as much as 150 m over much of British Columbia. Compared with the other nine most amplified days (Fig. 14b), 2021 still had much higher Z500 values. This provides perspective on just how much stronger the 2021 ridge was compared with other highly amplified ridges in the historical record. SLP values were also lower for the 2021 event over the western portion of the PNW and off the Pacific coast indicating a stronger thermally induced surface trough in 2021 (Figs. 14c,d). T850 was also much warmer for 2021 than the historical analogs by as much as 10°C over British Columbia, Oregon, and Washington. Last, Figs. 14g and 14h show the difference in U850 indicating much stronger easterly flow in the 2021 event over large areas of Washington and Oregon compared with previous events. This is especially prominent in the comparison to the other 49 most amplified ridges.
Figure 15 highlights one particularly similar event in regard to the synoptic setup and the multiday, trans-Pacific progression of the large-scale circulation features leading up to the heatwave peak. To identify the best matching event to June 2021, we use 2D pattern correlation for total field Z500 encompassing the entire North Pacific basin poleward of 20°N latitude, and compute daily pairwise correlations for the 7-day period preceding both the June 2021 event and each candidate historical analog. A total of eight pairwise correlation coefficients are computed, one for each day in the preceding 7-day period plus the day of event peak. Here, we consider the entire calendar year in order to find the closest historical match in terms of pattern progression and downstream amplification to what was observed in June 2021, regardless of surface temperature impacts in the PNW. We do this because extreme pattern amplification is rare during the time of year when most historical PNW heatwaves occur (i.e., 15 June–15 August) but considerably more common at other times of the year thereby greatly increasing the chance of finding suitable historical analogs. When compared with 20–27 June 2021, only the period from 9 to 16 May 2008 produced daily correlation of >0.9 for all 8 days in the paired time series (i.e., 9 May 2008 matched with 20 June 2021; 10 May 2008 matched with 21 June 2021; and continuing until the event peaks of 16 May 2008 and 27 June 2021). The 2008 event therefore represents the best historical match based on our criteria. Here we discuss the last 4 days leading up to the event peaks (12–15 May 2008 and 23–26 June 2021).
Maps of Z500 (in m) for (left) 23–26 Jun 2021 and (right) 12–15 May 2008.
Citation: Monthly Weather Review 151, 5; 10.1175/MWR-D-22-0284.1
On 12 May 2008, a similar wave train pattern was evident in the Z500 field as 23 June 2021. Notably, a broad Z500 trough was present over the Aleutian Islands with a ridge to the west of coastal North America. One day later (24 June 2021 and 13 May 2008), the trough deepened and progressed eastward, strengthening the downstream ridge over western North America. This feature was stronger with higher heights in 2021, but the wave pattern was quite similar between the two. By 25 June 2021 and 14 May 2008, the wave pattern had amplified further, with strong ridging over western North America in both cases. Strong south/southwesterly flow along the height gradient on the east side of the North Pacific trough likely contributed to the downstream ridge in 2008 as has been documented for 2021 (Neal et al. 2022). By the fourth day, the wave pattern exhibited a strong dipole between a highly amplified ridge over the PNW and a deep trough south of Alaska, setting the stage for peak ridge amplification on the following day in both cases. One difference between the two events that is worth noting is the presence of the cutoff low south of the ridge peak in June 2021.
Although outside the scope of this study, we further note that both events appear to have been forced via tropical energy infusion into the jet stream from the western Pacific Ocean, with downstream wave amplification resulting in extremely anomalous ridging and record-breaking extreme heat several days later over the study region. For example, Archambault et al. (2013) identified Typhoon Rammasun (which reached peak intensity on 11 May 2008) as a top-10 event in a climatology spanning 1979–2009 in terms of energy infusion into the extratropical large-scale flow from a recurving western Pacific typhoon, with large downstream effects on the jet stream. Although not associated with typhoon recurvature, the extreme 2021 wave train has been linked to anomalous moisture fluxes and heavy precipitation over the western Pacific Ocean associated with intraseasonal variations in the East Asian monsoon, that similarly infused the jet stream with energy and led to an extreme ridging configuration farther downstream (Lin et al. 2022; Qian et al. 2022; Schumacher et al. 2022).
While these two events had similar synoptic-scale features and evolution, they were very different events as far as impacts. For example, although many daily and monthly high temperature records were broken across the PNW on 16–17 May 2008, including maximums as high as 42.2°C in Oregon (NOAA 2008)—observationally unprecedented for mid-May—the occurrence of this ridge during spring precluded all-time records from being broken and reduced heat-related impacts to both human health and the environment. This illustrates that while the 2021 meteorological setup did have some historical analogs, with May 2008 being one shown here as an example, differences in the details of the meteorology and timing of occurrence compared with past events set it apart. Using May 2008 as an example, the ridge in 2021 was much stronger and pushed much farther north. This is evident when compared with all historical analogs in Figs. 14a and 14b as well and contributed not only to the warmth of the air mass, but the strong subsidence that led to exceptional heating of air parcels as they descended to the surface. The May 2008 event also did not have an upper-level low pressure pushing into the ridge from the southwest, a feature which likely further enhanced the large-scale offshore flow during the 2021 event. Additionally, the June 2021 event occurred near the time of peak insolation, allowing for the maximum number of daylight hours for surface heating. We note that recent studies have demonstrated the role of widespread negative soil moisture anomalies in further enhancing surface heating during the 2021 event (e.g., Bartusek et al. 2022); however, an investigation into land-atmosphere interaction is beyond the scope of this current study.
4. Discussion and conclusions
In late June 2021, an exceptionally severe heatwave affected a large swath of western North America. Numerous all-time high temperature records were broken across the region, many by several degrees, and for two or three consecutive days. The impacts were profound on both people and the environment. It is therefore critical to understand what led to such an extreme event, how it fits into the historical context, and what role global warming played in its occurrence. Toward these goals, we have provided a regional-scale meteorological diagnosis of the event and compared it to previous similar events to understand what set it apart. We note that we use the word “similar” in a general sense to describe the large-scale meteorological setup as no event in the observational record was similar in the magnitude of heat or severity of impacts.
This heatwave was in many ways associated with the same meteorological features and processes as other heatwaves in the PNW: a strong ridge of high pressure in the midtroposphere, large-scale subsidence and suppression of the Pacific Ocean marine layer, offshore flow, and a surface-based thermally induced trough parallel to the Pacific coast. However, as we have demonstrated, many of those features were much stronger, even record breaking, compared with previous similar setups (Fig. 14). At the root of the event was a massive midtropospheric ridge comprising record breaking 500-hPa geopotential heights and breaking the record for the largest summer Z500 deviation from the zonal mean in the ERA5 record (dating to 1979) in some places (Figs. 3–7). This facilitated both strong large-scale subsidence leading to notable adiabatic warming of air parcels and broad offshore flow across much of the region as shown by the backward trajectories for Seattle, Portland, and Pendleton (Figs. 8–10). We do note, however, that under the core of the ridge in British Columbia where some of the highest temperatures of the event were recorded, backward trajectories show that synoptic-scale flow was weaker than over Washington and Oregon and collocation with the center of the exceptional warm air mass was key (Fig. 11). For western Oregon and Washington, a very strong surface based thermally induced trough modulated low-level airflow and when it shifted east of the Cascades brought an abrupt end to the heat. Another unique feature of this event, which likely further enhanced its severity, was a cutoff low pressure in the midtroposphere to the southwest of the ridge center. This low increased height gradients and associated offshore flow, a key ingredient to the heat south of the ridge center. Last, we do not find evidence of large-scale horizontal transport of warm air playing an important role in the occurrence of the extreme temperature anomalies (Figs. 12, 13).
The role that global warming likely played in magnifying this heatwave has received considerable attention from the general public and the scientific community both during and since the event. While we do not assess this in a specific way here, we offer some perspective based on our findings. Overall, this event was primarily driven by a near-perfect meteorological setup for extreme heat over the region. The record-breaking ridge amplifying over just the right geographic location to bring maximum subsidence and offshore flow to most of the region was critical. The coincidence of the ridge with the time of maximum solar heating, limiting nighttime cooling, further contributed to the extreme heat, as did the unusual multiday persistence of such extreme ridge amplification and strong subsidence which further warmed the lower atmosphere as the event progressed. In locations downwind of topography during the event, such as Portland and Salem, Oregon, additional adiabatic warming of lower-level air parcels as they descended mountain slopes under unusually strong easterly flow (Figs. 14g,h) added an additional boost to the extreme heat. However, summer temperatures have also warmed by between 1° and 2°C over much of the PNW (Vose et al. 2017) since the first half of the twentieth century, with most of that warming attributable to anthropogenic forcing. Similarly, previous studies have suggested that the most extreme heatwaves across this region have warmed by approximately 1°–2°C over the past century (Wehner et al. 2018). Summers are projected to continue to warm more than other seasons across the PNW in the coming decades (Mass et al. 2022). Thompson et al. (2022) estimate that under a high-end emissions scenario, the chance of an event of this magnitude in this region could increase from 1 in 1000 years to 1 in 6 years by the end of the current century and Philip et al. (2022) estimate that anthropogenic warming has added about 2°C to a 1-in-1000-yr heatwave compared to the late 1800s. At the present time, the simplest assumption one could make about the role of global warming is that temperatures were warmer than they would have been had this event occurred in the past by the magnitude of observed anthropogenic warming over that time span.
One might also, however, consider whether global warming influenced the event through less direct means than background warming. For example, was the record breaking Z500 ridge made more likely or more amplified as a result of global warming? Some evidence from recent studies suggests the answer to this question may be no. For example, Loikith et al. (2022) found little or no significant increase in summer ridging frequency or amplitude over most of the region influenced by this ridge under a high-end emissions scenario using a suite of global climate models (note they did find an increase in ridge days to the east and southeast of the most affected areas). These results are consistent with Brewer and Mass (2016b). Widespread anomalously low soil moisture, which has been shown to be a contributing factor to extreme heat and ridging in other parts of the world (e.g., Fischer et al. 2007), has also been implicated as a contributing factor to the severity of the heat in this event, which itself is likely influenced by anthropogenic warming (Bartusek et al. 2022; Schumacher et al. 2022). Finally, the role of anomalously large quantities of atmospheric moisture over the North Pacific Ocean in strengthening and heating the June 2021 ridge through latent heat release (Mo et al. 2022; Neal et al. 2022; Schumacher et al. 2022), and therefore leading to a more extreme heatwave, begs the question of whether a wetter atmosphere in a warmer future could be more favorable to the formation of similarly extreme midlatitude ridges. However, our results are most suggestive of highly unusual natural variability driving the event and the severity of the heat, with a contribution from background anthropogenic warming, consistent with the findings of McKinnon and Simpson (2022) and Thompson et al. (2022) although complex nonlinear effects may play roles that are not readily apparent (Bartusek et al. 2022).
It is critical to learn from exceptionally severe weather events like the June 2021 PNW heatwave. Our modern observational record is too short to capture the full range of weather events that can occur, making events like this a valuable scientific opportunity to understand extremes with very long return intervals. Furthermore, understanding the range of complex processes that are required to come together for such an extreme event to arise in the first place improves our understanding of meteorology and can be extensible to better modeling and forecasting. Last, extreme events like this provide an opportunity to advance our understanding of the role of anthropogenic forcing in the occurrence and severity of extreme weather.
Acknowledgments.
Support for this work was provided by the U.S. National Science Foundation through Grant AGS-2206997 (P.C.L.) and by NASA FINESST Award 80NSSC21K1603 (D.A.K.).
Data availability statement.
ERA5 data are provided at https://climate.copernicus.eu/climate-reanalysis. Hysplit data and back trajectories are from https://www.arl.noaa.gov/hysplit/.
REFERENCES
Archambault, H. M., L. F. Bosart, D. Keyser, and J. M. Cordeira, 2013: A climatological analysis of the extratropical flow response to recurving western North Pacific tropical cyclones. Mon. Wea. Rev., 141, 2325–2346, https://doi.org/10.1175/MWR-D-12-00257.1.
Bartusek, S., K. Kornhuber, and M. Ting, 2022: 2021 North American heatwave amplified by climate change-driven nonlinear interactions. Nat. Climate Change, 12, 1143–1150, https://doi.org/10.1038/s41558-022-01520-4.
Bell, B., and Coauthors, 2021: The ERA5 global reanalysis: Preliminary extension to 1950. Quart. J. Roy. Meteor. Soc., 147, 4186–4227, https://doi.org/10.1002/qj.4174.
Blackmon, M. L., J. M. Wallace, N.-C. Lau, and S. L. Mullen, 1977: An observational study of the Northern Hemisphere wintertime circulation. J. Atmos. Sci., 34, 1040–1053, https://doi.org/10.1175/1520-0469(1977)034<1040:AOSOTN>2.0.CO;2.
Branstator, G., 1992: The maintenance of low-frequency atmospheric anomalies. J. Atmos. Sci., 49, 1924–1946, https://doi.org/10.1175/1520-0469(1992)049<1924:TMOLFA>2.0.CO;2.
Brewer, M. C., and C. F. Mass, 2016a: Projected changes in heat extremes and associated synoptic- and mesoscale conditions over the northwest United States. J. Climate, 29, 6383–6400, https://doi.org/10.1175/JCLI-D-15-0641.1.
Brewer, M. C., and C. F. Mass, 2016b: Projected changes in western U.S. large-scale summer synoptic circulations and variability in CMIP5 models. J. Climate, 29, 5965–5978, https://doi.org/10.1175/JCLI-D-15-0598.1.
Brewer, M. C., C. F. Mass, and B. E. Potter, 2012: The West Coast thermal trough: Climatology and synoptic evolution. Mon. Wea. Rev., 140, 3820–3843, https://doi.org/10.1175/MWR-D-12-00078.1.
Brewer, M. C., C. F. Mass, and B. E. Potter, 2013: The West Coast thermal trough: Mesoscale evolution and sensitivity to terrain and surface fluxes. Mon. Wea. Rev., 141, 2869–2896, https://doi.org/10.1175/MWR-D-12-00305.1.
Bumbaco, K. A., K. D. Dello, and N. A. Bond, 2013: History of Pacific Northwest heatwaves: Synoptic pattern and trends. J. Appl. Meteor. Climatol., 52, 1618–1631, https://doi.org/10.1175/JAMC-D-12-094.1.
Catalano, A. J., P. C. Loikith, and J. D. Neelin, 2021: Diagnosing non‐Gaussian temperature distribution tails using back‐trajectory analysis. J. Geophys. Res. Atmos., 126, e2020JD033726, https://doi.org/10.1029/2020JD033726.
Cavanaugh, N. R., and S. S. P. Shen, 2014: Northern Hemisphere climatology and trends of statistical moments documented from GHCN-daily surface air temperature station data from 1950 to 2010. J. Climate, 27, 5396–5410, https://doi.org/10.1175/JCLI-D-13-00470.1.
Chang, H., P. Loikith, and L. Messer, 2021: The June 2021 extreme heat event in Portland, OR, USA: Its impacts on ecosystems and human health and potential adaptation strategies. J. Extreme Events, 8, 2175001, https://doi.org/10.1142/S2345737621750014.
Cramer, O. P., and R. E. Lynott, 1970: Mesoscale analysis of a heatwave in western Oregon. J. Appl. Meteor. Climatol., 9, 740–759, https://doi.org/10.1175/1520-0450(1970)009<0740:MAOAHW>2.0.CO;2.
ECCC, 2021: Canada’s top 10 weather stories of 2021. Environment and Climate Change Canada, accessed 9 September 2022, https://www.canada.ca/en/environment-climate-change/services/top-ten-weather->stories/2021.html.
Fischer, E. M., S. I. Seneviratne, D. Lüthi, and C. Schär, 2007: Contribution of land-atmosphere coupling to recent European summer heat waves. Geophys. Res. Lett., 34, L06707, https://doi.org/10.1029/2006GL029068.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Lin, H., R. Mo, and F. Vitart, 2022: The 2021 Western North American heatwave and its subseasonal predictions. Geophys. Res. Lett., 49, e2021GL097036, https://doi.org/10.1029/2021GL097036.
Liu, P., and Coauthors, 2018: Climatology of tracked persistent maxima of 500-hPa geopotential height. Climate Dyn., 51, 701–717, https://doi.org/10.1007/s00382-017-3950-0.
Loikith, P. C., and A. J. Broccoli, 2012: Characteristics of observed atmospheric circulation patterns associated with temperature extremes over North America. J. Climate, 25, 7266–7281, https://doi.org/10.1175/JCLI-D-11-00709.1.
Loikith, P. C., and J. D. Neelin, 2015: Short‐tailed temperature distributions over North America and implications for future changes in extremes. Geophys. Res. Lett., 42, 8577–8585, https://doi.org/10.1002/2015GL065602.
Loikith, P. C., D. Singh, and G. P. Taylor, 2022: Projected changes in atmospheric ridges over the Pacific–North American region using CMIP6 models. J. Climate, 35, 5151–5171, https://doi.org/10.1175/JCLI-D-21-0794.1.
Mass, C. F., E. P. Salathé, R. Steed, and J. Baars, 2022: The mesoscale response to global warming over the Pacific Northwest evaluated using a regional climate model ensemble. J. Climate, 35, 2035–2053, https://doi.org/10.1175/JCLI-D-21-0061.1.
McKinnon, K. A., and I. R. Simpson, 2022: How unexpected was the 2021 Pacific Northwest heatwave? Geophys. Res. Lett., 49, e2022GL100380, https://doi.org/10.1029/2022GL100380.
Menne, M. J., I. Durre, R. S. Vose, B. E. Gleason, and T. G. Houston, 2012: An overview of the global historical climatology network-daily database. J. Atmos. Oceanic Technol., 29, 897–910, https://doi.org/10.1175/JTECH-D-11-00103.1.
Miller, R. J., and A. Bair, 2022: Washington maximum temperature. State Climate Extremes Committee Memorandum, 28 pp., https://www.ncei.noaa.gov/monitoring-content/extremes/scec/reports/20220210-Washington-Maximum-Temperature.pdf.
Mo, R., H. Lin, and F. Vitart, 2022: An anomalous warm-season trans-Pacific atmospheric river linked to the 2021 western North America heatwave. Commun. Earth Environ., 3, 127, https://doi.org/10.1038/s43247-022-00459-w.
Neal, E., C. S. Y. Huang, and N. Nakamura, 2022: The 2021 Pacific Northwest heat wave and associated blocking: Meteorology and the role of an upstream cyclone as a diabatic source of wave activity. Geophys. Res. Lett., 49, e2021GL097699, https://doi.org/10.1029/2021GL097699.
NOAA, 2008: Climatological data, Oregon, May 2008. Vol. 114, NOAA, 36 pp., https://www.ncdc.noaa.gov/IPS/cd/cd.html.
Osman, M., B. F. Zaitchik, and N. S. Winstead, 2022: Cascading drought‐heat dynamics during the 2021 Southwest United States heatwave. Geophys. Res. Lett., 49, e2022GL099265, https://doi.org/10.1029/2022GL099265.
Philip, S. Y., and Coauthors, 2022: Rapid attribution analysis of the extraordinary heat wave on the Pacific coast of the U.S. and Canada in June 2021. Earth Syst. Dyn., 13, 1689–1713, https://doi.org/10.5194/esd-13-1689-2022.
Qian, Y., P. Hsu, J. Yuan, Z. Zhu, H. Wang, and M. Duan, 2022: Effects of subseasonal variation in the East Asian monsoon system on the summertime heat wave in Western North America in 2021. Geophys. Res. Lett., 49, e2021GL097659, https://doi.org/10.1029/2021GL097659.
Raymond, W. W., and Coauthors, 2022: Assessment of the impacts of an unprecedented heatwave on intertidal shellfish of the Salish Sea. Ecology, 103, e3798, https://doi.org/10.1002/ecy.3798.
Schumacher, D. L., M. Hauser, and S. I. Seneviratne, 2022: Drivers and mechanisms of the 2021 Pacific Northwest heatwave. Earth’s Future, 10, e2022EF002967, https://doi.org/10.1029/2022EF002967.
Stein, A. F., R. R. Draxler, G. D. Rolph, B. J. B. Stunder, M. D. Cohen, and F. Ngan, 2015: NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bull. Amer. Meteor. Soc., 96, 2059–2077, https://doi.org/10.1175/BAMS-D-14-00110.1.
Stewart, R. E., and Coauthors, 2017: A multi-perspective examination of heat waves affecting Metro Vancouver: Now into the future. Nat. Hazards, 87, 791–815, https://doi.org/10.1007/s11069-017-2793-7.
Thompson, V., and Coauthors, 2022: The 2021 western North America heat wave among the most extreme events ever recorded globally. Sci. Adv., 8, eabm6860, https://doi.org/10.1126/sciadv.abm6860.
Vescio, M. D., and A. Bair, 2022: Oregon maximum temperature record. State Climate Extremes Committee Memorandum, 25 pp., https://www.ncei.noaa.gov/monitoring-content/extremes/scec/reports/20220210-Oregon-Maximum-Temperature.pdf.
Vose, R. S., D. R. Easterling, K. E. Kunkel, A. N. LeGrande, and M. F. Wehner, 2017: Temperature changes in the United States. Climate Science Special Report: Fourth National Climate Assessment, D. J. Wuebbles et al., Vol. I, U.S. Global Change Research Program, 185–206.
Wehner, M., D. Stone, H. Shiogama, P. Wolski, A. Ciavarella, N. Christidis, and H. Krishnan, 2018: Early 21st century anthropogenic changes in extremely hot days as simulated by the C20C+ detection and attribution multi-model ensemble. Wea. Climate Extremes, 20, 1–8, https://doi.org/10.1016/j.wace.2018.03.001.
