Part 2: The leading mode of North American winter circulation has changed, and March 2026 is what the new mode produces
In Part 1, I described the diabatic engine behind March 2026’s record-shattering western heat dome: upstream precipitation in the Pacific Northwest released latent heat that amplified an extraordinary ridge over the interior West. That post explained the mechanism. This one asks a different question: why does the ridges keep forming?
The March 2026 heat dome was not a bolt from the blue. It was the most extreme pulse in a winter-long pattern that locked the western United States into warmth and the East into cold for months. That warm-West/cool-East dipole has become a recurring feature of North American winters over the past decade, and a growing body of research suggests it reflects something deeper than internal variability. The leading mode of winter atmospheric circulation over North America appears to have shifted, driven by greenhouse-gas forced changes in the jet stream and the atmospheric waveguide. The atmosphere has been restructured in a way that favors extreme ridges over the West.
This is not a formal attribution analysis. It is an attempt to situate the March 2026 event within the growing literature on changing winter circulation over North America.
The winter that wouldn’t end (in the East) and barely started (in the West)
Winter 2025-26 was dominated by a persistent ridge parked over or just west of the Pacific coast, with a deep trough anchored over eastern North America. The ridge kept the West anomalously warm and dry for months. The trough delivered repeated Arctic outbreaks, blizzards, and ice storms to the Midwest and East.
NOAA confirmed in early March that winter 2025-26 was the warmest on record across the majority of the American West. Seven states set all-time warmest winter records, some by margins exceeding 2°F. Meanwhile, parts of the Upper Midwest and Great Lakes experienced their snowiest and coldest stretches in years. The continent was running two different seasons simultaneously, separated by a few hundred miles.
Daniel Swain, writing at Weather West in January 2026, explicitly named this the “Warm West/Cool East dipole pattern” and noted it was “a familiar feature during the 2013-2015 and 2018-2020 drought years.” BoulderCAST, the Colorado forecast outlet, was more blunt, describing “a stubborn North American dipole pattern locked into place: a deep, frigid trough dominating the eastern US while a warm, bloated ridge camped over the West” and calling it “a structural failure of winter itself.”
The March heat dome was the crescendo: the ridge amplified to the strongest cool-season level ever observed in the southwestern United States. Temperatures ran 20 to 30 degrees above normal across a vast region, and the jet stream split around a heat dome so intense it looked like a midsummer pattern transplanted into mid-March. But the ridge didn’t appear from nowhere. It had been there, in one form or another, all winter. March was when the diabatic amplification kicked it into record territory.
The North American Winter Dipole
The warm-West/cool-East pattern has a name in the research literature: the North American Winter Dipole, or NAWD. And the story of how it went from a secondary mode of variability to the dominant pattern of North American winters is one of the more consequential, and underappreciated, findings in recent climate dynamics.
For most of the 20th century, the leading mode of winter upper-tropospheric variability over the North Pacific and North America was the Pacific-North American pattern, or PNA. The PNA is a wave train of alternating ridges and troughs stretching from the subtropical western Pacific through the Gulf of Alaska to southeastern North America. It’s the pattern that introductory meteorology courses teach, and the index the Climate Prediction Center still tracks as a primary driver of North American winter weather.
But something started to change around 1980.
Chien, Wang, and Chikamoto (2019) performed EOF analysis on winter 250-hPa eddy geopotential height anomalies across two periods: 1948-1979 and 1980-2017. In the earlier period, the first EOF was spatially coincident with the PNA pattern, explaining about 19% of the variance. In the later period, the first EOF had shifted by roughly a quarter wavelength. It was no longer aligned with the PNA. Instead, it matched the centers of the NAWD: a ridge over the Gulf of Alaska and western Canada, and a trough over the Great Lakes and eastern North America.
Their running-EOF analysis tracked the transition in detail. The NAWD, which had been the second mode (EOF2), intensified during the 1990s until it overtook the PNA as the leading mode. The PNA didn’t disappear. It dropped to second position. But the pattern that now explains the most variance in winter upper-level circulation is the one that produces warm-West/cool-East dipoles.
Why the waveguide changed
Lee, Wang, Son, Kim, Jeong, Kim, and Yoon (2024) picked up where Chien et al. left off, asking what caused the shift. Their analysis, published in npj Climate and Atmospheric Science, examined three successive 30-year windows (1951-1980, 1971-2000, 1991-2020) and found a progressive strengthening of the climatological ridge over western North America. The most pronounced change was the development of positive eddy geopotential height anomalies east of Alaska, representing an eastward expansion of the climatological ridge.
This matters because the background ridge changes the waveguide. As the climatological anticyclonic circulation strengthens over western North America, it alters the magnitude and shape of the mean zonal wind, which in turn changes the stationary Rossby wavenumber. Lee et al. showed that the stationary Rossby wavenumber has decreased in the Pacific Northwest region. A lower stationary wavenumber means the waveguide favors longer-wavelength patterns, suppresses the propagation of shorter waves through the ridge region, and effectively narrows the corridor through which Rossby wave energy can pass.
With limited paths available for stationary Rossby waves, different teleconnection patterns are forced to increasingly resemble each other. Lee et al. found that the PNA and NAWD, historically weakly correlated, have developed a pronounced negative correlation in recent decades. The modes are no longer independent. The waveguide is channeling variability into a preferred geometry: ridge over the West, trough over the East.
Lee et al. argue that greenhouse gas forcing is the likely driver. Their attribution analysis identified GHG emissions as the probable cause of a northward drift of the Asia-Pacific jet core, which, amplified by orographic lifting over the Alaskan Range, strengthens the winter stationary wave across western North America. The topography hasn’t changed. But the jet flowing over it has shifted, and that shift changes how the mountains shape the downstream wave pattern.
The wavenumber-5 connection
There is a deeper structural reason why the western ridge keeps recurring in the same location. Teng and Branstator (2017) showed that extreme ridges over western North America, the kind that drive California droughts and western heat events, are preferentially associated with circumglobal wave patterns at zonal wavenumber 5.
The mechanism works through the jet stream acting as a waveguide. Strong, narrow jets can trap Rossby wave energy, channeling it zonally around the hemisphere rather than allowing it to disperse meridionally. When the jet is configured to act as an efficient waveguide, it preferentially selects wavenumber 5, a pattern with five ridges and five troughs arrayed around the hemisphere. One of those ridges naturally falls over western North America, anchored by the topography of the Rockies and the Alaskan Range.
Teng and Branstator found that tropical diabatic heating anomalies are not required to initiate these wavenumber-5 patterns. They can arise from internal midlatitude atmospheric dynamics alone. But tropical heating, particularly from the western Pacific warm pool, can double the probability of extreme ridging by projecting additional wave energy onto the waveguide at the right wavelength.
This connects directly to the 500-millibar height field during March 2026, which showed a visually striking five-lobed deformation of the polar vortex: five ridges and five troughs circling the Northern Hemisphere. The pattern was not a clean wavenumber 5, because the western North American ridge was grotesquely amplified relative to the others. But the underlying geometry was consistent with preferential excitation of a wavenumber-5 circumglobal pattern. The diabatic amplification (described in Part 1) then supercharged one particular lobe far beyond what the wave dynamics alone would produce.
The dipole is amplifying
The structural shift toward NAWD dominance has a measurable consequence at the surface. Singh, Swain, Mankin, and colleagues (2016) documented a significant increase in both the occurrence and severity of warm-West/cool-East temperature dipole events between 1980 and 2015. The positive trend is attributable to historical anthropogenic emissions.
This is the surface expression of the waveguide change. As the NAWD becomes the leading mode and the stationary wave strengthens, the ridge-trough couplet that defines the dipole intensifies. The West gets warmer. The East gets colder (relative to the warming trend). The temperature gradient across the continent during winter dipole events steepens. The pattern becomes more extreme in both directions simultaneously.
Winter 2025-26 was a textbook example. Seven western states set warmest-winter records while parts of the East experienced significant cold and snow. The 2-meter temperature anomaly field on March 17, 2026 showed a continent sliced in half: +14 to +18°C above normal in the West, -10 to -18°C below normal in the Upper Midwest and Great Lakes. That kind of continental-scale gradient is what a deeply amplified, quasi-stationary NAWD pattern produces.
The ocean remembers
The persistent ridging associated with the NAWD doesn’t just affect the atmosphere. It reshapes the ocean beneath it. When a strong, persistent anticyclone sits over the northeast Pacific, it weakens the surface wind field, suppresses coastal upwelling, and reduces evaporative cooling. The ocean surface warms. During March 2026, a Category 2 marine heat wave was active off the California coast, with SSTs running up to 5°F above average.
And the atmospheric forcing is only part of the story. The ocean itself has been independently restructured by warming. As the surface warms faster than the deeper ocean, thermal stratification increases and the mixed layer shoals (Liu et al. 2024). A thinner mixed layer is more responsive to solar heating and has less thermal inertia, meaning warm anomalies develop faster and persist longer. This forced oceanic change is happening globally, driven by greenhouse warming rather than any particular atmospheric pattern.
In the northeast Pacific, both processes coincide. The ocean is already primed to produce larger SST anomalies because of forced stratification changes. When the NAWD ridge then suppresses winds on top of an already-shallower mixed layer, the SST response is amplified beyond what either change alone would produce.
Once in place, the warm SSTs can feed back into the atmosphere by stabilizing the marine boundary layer and reducing the marine layer that normally moderates coastal temperatures, potentially reinforcing the ridge from below. This creates a compound feedback in which forced changes in both the atmosphere and the ocean independently increase the sensitivity of their coupling.
What this means for events like March 2026
The Part 1 blog post described the diabatic engine: how upstream storms build downstream heat domes through latent heat release and ridge amplification. That mechanism explains why the March 2026 ridge reached record-shattering intensity on any given day.
This post describes the structural context: why the atmosphere was already set up with a ridge over western North America before the diabatic amplification even began. The NAWD provides the template. Greenhouse-gas driven changes in the jet and the waveguide select this pattern preferentially. The topography of western North America anchors it. And once the ridge is in place, it conditions the ocean surface in ways that may further reinforce its own persistence.
The March 2026 heat dome was extreme because multiple processes aligned. The background waveguide, reshaped by decades of warming, favored a western ridge. The NAWD, now the leading mode of winter variability, locked the pattern in for months. And then the diabatic mechanism, fueled by a Kona low near Hawaii and an atmospheric river slamming into the Pacific Northwest, amplified the ridge to levels never before observed in the cool season.
None of these layers alone would have produced the event. The waveguide shift doesn’t cause record heat on its own. The diabatic mechanism needs a pre-existing ridge to amplify. And the NAWD can persist all winter without ever reaching the intensity that March 2026 achieved. It’s the stacking, the convergence of a slowly shifting background state with a fast dynamical amplification mechanism, that produces something truly unprecedented.
The atmosphere hasn’t fundamentally changed its physics. But the background state in which those physics operate has shifted. The waveguide has narrowed. The leading mode has changed. The ridge keeps forming in the same place because the jet, the topography, and the warming pattern all conspire to put it there. And when the right upstream trigger comes along, the same machinery that produced the 2021 Pacific Northwest heat dome, the 2020-2021 western drought, and the 2013-2015 Ridiculously Resilient Ridge is ready to amplify the next event to record-shattering levels.
This is the new structural reality of North American winter climate. The question is less whether the western ridge will form than how often the background pattern and the short-term amplification mechanisms will align to produce the next record.
This post is Part 2 of a series on the March 2026 western US heat dome. Part 1 covered the diabatic ridge-building mechanism. Claude Opus (Anthropic) contributed to this analysis through iterative analysis of synoptic patterns and literature review.
Further reading:
Bond, N. A., Cronin, M. F., Freeland, H. J., & Mantua, N. (2015). Causes and impacts of the 2014 warm anomaly in the NE Pacific. Geophysical Research Letters, 42. doi:10.1002/2015GL063306
Chien, Y.-T., Wang, S.-Y., & Chikamoto, Y. (2019). North American Winter Dipole: Observed and simulated changes in circulations. Atmosphere, 10(12), 793. doi:10.3390/atmos10120793
Lee, J., Wang, S.-Y. S., Son, S.-W., Kim, D., Jeong, J.-H., Kim, H., & Yoon, J.-H. (2024). Evolving winter atmospheric teleconnection patterns and their potential triggers across western North America. npj Climate and Atmospheric Science, 7, 63. doi:10.1038/s41612-024-00608-2
Liu, C., et al. (2024). Human-induced intensified seasonal cycle of SST. Nature Communications, 15, 3948.
Singh, D., Swain, D. L., Mankin, J. S., Horton, D. E., Thomas, L. N., Rajaratnam, B., & Diffenbaugh, N. S. (2016). Recent amplification of the North American winter temperature dipole. Journal of Geophysical Research: Atmospheres, 121(17), 9911-9928. doi:10.1002/2016jd025116
Teng, H., & Branstator, G. (2017). Causes of extreme ridges that induce California droughts. Journal of Climate, 30, 1477-1492. doi:10.1175/JCLI-D-16-0524.1
Wang, S.-Y., Hipps, L. E., Gillies, R. R., & Yoon, J.-H. (2014). Probable causes of the abnormal ridge accompanying the 2013-2014 California drought. Geophysical Research Letters, 41(9), 3220-3226. doi:10.1002/2014gl059748
