Uvod

In this investigation, climate changes related to two Pacific Ocean climate shifts in 1976/77 and 1998/99 are studied, where Period 1 (P1) is defined as 1977–98 and Period 2 (P2) is defined as 1999–2022. The shifts have been associated with opposite phases of the interdecadal Pacific oscillation (IPO; Meehl et al. 2016a; Folland et al. 2002) and possibly also with a transient response to greenhouse gas and aerosol emissions (Heede and Fedorov 2023) related to the ocean thermostat hypothesis. This would help explain the difference between the current climate, represented here by P2, and the climate predicted by climate models, which is closer to P1 (Heede et al. 2021).

The IPO is also known as interdecadal Pacific variability (IPV) since the mode may not be an oscillation; it could be a red noise process with a 3–5-yr decay time scale. Tung et al. (2019) even dispute the existence of an interdecadal Pacific SST mode, preferring the well-known Pacific decadal oscillation (PDO/PDV) based on temporal filtering and rotated EOFs. Meehl et al. (2021) studied large member composites of model IPO transitions but did not characterize the type of temporal variability. We will use the above acronyms interchangeably, somewhat dependent on their use in the reference being cited.

After the first climate shift in 1976/77, the Indo-Pacific warm pool (IPWP) expanded eastward (Roxy et al. 2019; Hartmann and Wendler 2005) creating a favorable ocean state for basinwide El Niño (warm) events involving the eastern Pacific cold tongue including the strong 1982/83 and 1997/98 events. Related to this warm pool expansion the positive phase of the IPO/IPV/PDO dominated much of 1977–98 (Meehl et al. 2016a; Henley et al. 2015).

After the second climate shift in 1998/99, the IPWP center moved northwest and expanded poleward as the negative phase of IPV/IPO prevailed (Dong and Dai 2015). There is the suggestion that “global warming” pauses when IPV is in its negative phase (i.e., P2) leading to a “warming hiatus” (Medhaug et al. 2017) such as was observed during 2000–15. After the 2015/16 El Niño, Su et al. (2017) and Hu and Fedorov (2017) proposed global surface temperature had exited the hiatus with a transition to positive IPO. Several models (Henley et al. 2017; Meehl et al. 2016b; Thoma et al. 2015) also predicted that the IPO would return to a positive phase during 2015–19 and resume an accelerated warming. The 3-yr La Niña that developed in early 2020 leaves that result in doubt. Although variability has increased, little additional warming in global temperatures has occurred since the 2015/16 El Niño.

Fasullo et al. (2023) postulate transient forcing from the 2019/20 Australian wildfire smoke helped force the La Niña and shifted the climate back to a negative IPO. A diagnosis of AAM and outgoing longwave radiation (OLR) anomalies during the onset of the 2019/20 La Niña (section 7) suggests that typical ENSO variability was at play. Less typical were the effects from the ozone hole in Southern Hemisphere circulation, especially since ∼1993 when it reached its maximum areal extent and remains nearly that large through 2022. The downward coupling from the strong jet in the stratosphere that accompanies the decreased heating in the southern polar stratosphere reaches the surface in December–February (Orr et al. 2021) and interacts with the internal (IPO/IPV) and forced (greenhouse gases and aerosols) variability discussed earlier.

Sklepi in diskusija

While climate models successfully predicted a transition to IPO positive during 2015–19, the climate now appears stuck in IPO negative, especially given the increase in SST observed in the west Pacific after the strong 2015/16 and weak 2019/20 El Niños (Fig. 2b). La Niña developed in spring/summer 2020, the start of an amplified, “La Niña–like” P2 pattern that lasted through the 2022/23 boreal winter. As convection shifts seasonally between Southeast Asia and the southwest Pacific Ocean (Meehl 1987), a La Niña–like residual is now more likely given the meridional expansion and shift northwest by the center of the IPWP and its connected circulation anomalies.

As part of the seasonal change in zonal AAM (Fig. 5) and 500-hPa heights (Fig. 6), large and intense subtropical-midlatitude anticyclones have dominated NH continental regions during P2 leading to severe droughts especially across southwestern North America (Williams et al. 2020). These ridges have also provided ideal atmospheric conditions for the recent intense wildfire seasons across the western United States as well as other parts of the globe. It is the same ridges, amplified by Rossby wave dispersion linked to subseasonal tropical convective anomalies over the eastern end of the IPWP that have contributed to Arctic amplification since the late 1990s (Gong et al. 2020).

As addressed in considerable detail in IPCC AR6 (IPCC 2021, p. 989), large uncertainty exists of anthropogenic greenhouse gas and aerosol emissions to the future of the planet including whether La Niña– or El Niño–like states will dominate the response. The current overall assessment has “medium confidence that equilibrium warming in response to elevated CO₂ will be characterized by a weakening of the east–west tropical Pacific SST gradient.” This does not preclude the possibility of a transient response to greenhouse gases that involves the strengthened SST gradient and expanded IPWP observed during P2 (Heede and Fedorov 2023; Heede et al. 2021).

The other prominent long-term climate feature that shows up in P2–P1 is the negative SST anomaly along 60°S (Fig. 3). It coincides with a negative frictional torque anomaly, especially during DJF, that is linked with deep westerly wind anomalies. These westerlies have been traced to the ozone hole whose zonal mean dynamics was investigated in reanalysis datasets by Orr et al. (2021). Hartmann (2022) even suggests a connection between the ozone hole and the negative SST anomalies in the eastern tropical Pacific Ocean as an alternative to the ocean thermostat hypothesis.

In the SH zonal AAM budget, the negative frictional torque anomaly centered at 60°S is approximately balanced by positive torque anomalies north of 40°S; this includes a positive mountain torque along 10°–30°S that comes primarily from the Andes (Fig. A1a). The torques in the two regions are linked by poleward AAM transports produced by ozone hole dynamics that “saturates” in the early 1990s and the transition to a negative IPO/IPV that occurs abruptly during 1998. Saturates refers to the area of the ozone hole; it climbs rapidly to a maximum from 1979 to 1992, peaks in the early 1990s, and then decreases very slowly thereafter (https://www.eea.europa.eu/en/topics/in-depth/climate-change-mitigation-reducing-emissions/current-state-of-the-ozone-layer). The 60°S friction torque (not shown) also becomes persistent in the early 1990s, that is, when the ozone hole saturates. The positive mountain torque anomalies (Fig. A1a) start to develop in the early 1990s and become persistent in 1998, consistent with a role for both the ozone hole and the transition to P2 or negative IPO.

A positive mountain torque anomaly implies a negative surface pressure anomaly on the west side of the Andes giving enhanced southerly flow west of the mountains and conversely a positive surface pressure anomaly on the east side of the Andes giving enhanced northerly flow east of the mountains. Given the steep terrain from 10° to 30°S, anomalous southerly winds could extend west over the adjacent east Pacific Ocean. In concert with an enhanced Walker circulation and easterly wind anomalies over the tropical Pacific during P2, this provides an additional mechanism for upwelling and advection of negative SST anomalies toward the equatorial east Pacific. It is unclear how well climate models capture the observed mountain torque during 1977–2022 and its related circulation anomalies.

Povzetek

Atmospheric angular momentum (AAM) is used to study the variability of Earth’s atmospheric circulation during the past 45 years, a time of considerable climate change. Using global AAM, two interdecadal states are defined covering the periods 1977–98 (hereinafter P1) and 1999–2022 (P2). Global AAM decreased from P1 to P2 and was accompanied by weakened subtropical jet streams in both hemispheres, strong convection around the northern Maritime Continent, and a strengthened sea surface temperature (SST) gradient across the tropical Pacific Ocean. The period differences project onto 1) internal interdecadal Pacific variability (IPV), 2) a postulated transient ocean thermostat response to greenhouse gas and aerosol emissions, and 3) circulation anomalies related to the ozone hole. During 1977–2023, the first two processes are forcing the climate toward larger Pacific Ocean SST gradients and a poleward expansion of the Indo-Pacific warm pool (IPWP), especially into the Northern Hemisphere. The ozone hole produces its own distinct pattern of anomalies in the Southern Hemisphere that tend to become persistent in the early 1990s. The zonal and vertical mean AAM variations during P1 have frequent westerly wind anomalies between 40°N and 40°S with poleward propagation on interannual time scales. During P2, the circulation is dominated by subtropical easterly wind anomalies, poleward-shifted jets, and weaker propagation. Locally, the zonal mean anomalies manifest as midlatitude ridges that lead to continental droughts. Case studies illustrate the weakened subtropical jet streams of P2 and examine the factors behind a transition to La Niña in early 2020 that maintains the P2 pattern.