1. Introduction
The Atlantic meridional overturning circulation (AMOC), also referred to as the thermohaline circulation, is a large-scale ocean circulation that redistributes heat into the subpolar North Atlantic from the tropics and Southern Hemisphere (Weaver et al. 1999; Hu et al. 2010, 2012b; Weaver et al. 2012; Buckley and Marshall 2016; Smeed et al. 2016; Trenberth and Fasullo 2017; Ferreira et al. 2018; Weijer et al. 2019; Trenberth et al. 2019; Zhang et al. 2019). The influence of AMOC extends to North Atlantic sea surface temperatures (SSTs; Dong and Sutton 2002; Dahl et al. 2005; Zhang and Delworth 2005), El Niño–Southern Oscillation (ENSO; Timmermann et al. 2005, 2007; Dong and Sutton 2007; van Oldenborgh et al. 2009; Williamson et al. 2018), and past abrupt climate change events (Clark et al. 2002; Hu et al. 2007). A deep overturning circulation is absent in the present-day North Pacific because relatively freshwater hinders deep water formation (Burls et al. 2017). Studies have proposed that AMOC has an asynchronous, seesaw-like relationship with the Pacific meridional overturning circulation (PMOC), in which AMOC weakening potentially initiates Pacific deep convection (Seidov and Haupt 2003; Saenko et al. 2004; Okazaki et al. 2010). AMOC is projected to weaken under both low- and high-emission scenarios (Sigmond et al. 2020) and the asynchronous Atlantic–Pacific relationship suggests PMOC could develop. However, little is known about how changes to AMOC and PMOC may modulate global SSTs and ENSO.
AMOC and ENSO influence regional climates across the world (McPhaden et al. 2006; Zhang et al. 2021; Liu et al. 2020). For example, AMOC influences the atmospheric circulation and mean winds over the North Atlantic Ocean (Jackson et al. 2015), which in turn contribute to the present-day mild climate of western Europe (Seager et al. 2002; Sutton and Hodson 2005). ENSO teleconnections to extremes include tropical cyclones (Lin et al. 2020), Indian summer monsoons (Azad and Rajeevan 2016), and North American severe thunderstorms (Molina et al. 2018). AMOC also exerts an influence on ENSO; a weakened or collapsed AMOC can increase ENSO amplitude (Dong and Sutton 2007; Timmermann et al. 2007), potentially intensifying regional teleconnections. ENSO characteristics are strongly dependent on background climatic conditions, such as mean SSTs and thermocline depth (Fedorov and Philander 2001; Capotondi and Sardeshmukh 2017), which AMOC can change. We hypothesize that PMOC may also modify ENSO amplitude and global SSTs.
The influence of AMOC is global due to its redistribution of water and heat. AMOC transports warm and saline waters of the upper-ocean northward into the subpolar North Atlantic. The water then cools and becomes more dense, sinks (forming the North Atlantic Deep Water) and flows southward, and eventually upwells elsewhere in the World Ocean (Dickson and Brown 1994; Stocker and Broecker 1994; Hu et al. 2015). As the climate continues to warm due to anthropogenic greenhouse gas forcing, AMOC may slow down due to increases in freshwater flux from ice sheets, ice caps, and glaciers across the North Atlantic (Rahmstorf et al. 2015; Caesar et al. 2021). While long-term AMOC trends are still actively debated due to the limited observational record (Willis 2010; Parker and Ollier 2016), a slowdown of AMOC could result in consequential climate change impacts, including modulations to ENSO amplitude and initiation of Pacific deep convection (Okazaki et al. 2010; Williamson et al. 2018).
The proposed asynchronous, seesaw-like relationship between the Atlantic and Pacific meridional overturning circulations (Okazaki et al. 2010) is not settled science, however. Previous studies have found that an asynchronous relationship between AMOC and PMOC is unlikely with an open Bering Strait because it allows water mass exchange between the Pacific and Atlantic via the Arctic (Hu et al. 2012a). Under present-day conditions, water flows from the North Pacific into the Arctic through an open Bering Strait, and can eventually reach the subpolar North Atlantic as comparatively fresher water or sea ice (Woodgate and Aagaard 2005). When AMOC is strong, water flux into the subpolar North Atlantic from the North Pacific increases, which can weaken AMOC. In contrast, a weak AMOC can reduce the amount of freshwater flux from the North Pacific into the subpolar North Atlantic, allowing AMOC to strengthen on decadal to centennial time scales due to the high salinity content of subpolar North Atlantic waters (Hu et al. 2011). Significant AMOC weakening could even trigger a reversal of water flow through the open Bering Strait, allowing comparatively fresher water to flow into the North Pacific (Hu et al. 2012a,b), which could also prevent PMOC development. Thus, an open Bering Strait acts as a stabilizing mechanism for the strength of AMOC. Saenko et al. (2004) showed that the extraction of freshwater from the North Pacific can result in the development of PMOC when the Bering Strait is closed because of the cutoff Arctic pathway and resultant salinity increase. Uncertainty remains, however. In fact, to what degree the present-day salinity profiles of the Atlantic and Pacific basins are driven by oceanic (e.g., saline water advection) versus atmospheric processes (e.g., evaporation) remains an open question (Ferreira et al. 2018). Given this existing uncertainty regarding cross-basin interactions, we aim to understand how different states of Atlantic and Pacific deep water formation impact global SSTs and ENSO, which could amplify or exacerbate regional climate change impacts (Cai et al. 2014; Rahmstorf et al. 2015).
This paper is structured as follows. Model specifications and analyses of meridional overturning circulation experiments are explained in section 2. Changes to global SSTs and ENSO that occurred within the experiments are detailed in section 3, along with the potential driving physical mechanisms and processes. Key results and discussion of possible future avenues for research follow in section 4.
2. Methods
a. Model configuration and sensitivity experiments
This study used simulations created with the Community Earth System Model version 1 (CESM1; Hurrell et al. 2013) at 1° horizontal grid spacing across coupled model components, developed at the National Center for Atmospheric Research in collaboration with laboratories supported by the United States Department of Energy (DOE) and the university community. CESM1 contains the Community Atmospheric Model version 5 (CAM5; Neale et al. 2010), the Parallel Ocean Program version 2 (POP2; Danabasoglu et al. 2012), the Community Land Model version 4 (CLM4; Lawrence et al. 2011), and the Community Sea Ice Code version 4 (CICE4; Hunke et al. 2008). The freshwater sensitivity experiments used in this study were all initialized using the CESM1 preindustrial control simulation from January 1 of year 800. AMOC strength in the CESM1 preindustrial control simulation (20.0 Sv at 26.5°N; Sverdrups, where 1 Sv ≡ 106 m3 s−1; Hu et al. 2021) is comparable to observational estimates (17.5 Sv at 26.5°N; Smeed et al. 2016) and is defined as the maximum of the Atlantic Meridional Streamfunction below a 500-m depth.
Five sensitivity experiments (Table 1) were conducted to explore the effect of AMOC and PMOC on oceans globally (Hu et al. 2021). The first four experiments (Table 1) involved the addition of freshwater into the subpolar North Atlantic (50°–70°N) uniformly at rates of 0.2 and 0.4 Sv, which reduces deep convection due to the comparatively lower density (i.e., salinity) of freshwater. Added freshwater weakens AMOC and can subsequently collapse the overturning circulation. The added freshwater into the subpolar North Atlantic was compensated by uniformly draining freshwater from other regions. The first two experiments involved drainage of −0.05 and −0.10 mm day−1 uniformly from all other oceans, referred to as “0.2 Sv Global” and “0.4 Sv Global,” respectively. These experiments can also be thought of as salinity reduction in the subpolar North Atlantic and salinity increase in other ocean basins. The next two experiments involved drainage of −1.15 and −2.30 mm day−1 uniformly from the subpolar North Pacific between 40° and 60°N, referred to as “0.2 Sv Pacific” and “0.4 Sv Pacific,” respectively. The purpose of the removal of freshwater and increase in salinity across the subpolar North Pacific was to facilitate the generation of PMOC during AMOC collapse. Freshwater forcing was applied for a period of 500 years, which collapsed AMOC within the first 200 years during the four experiments (Fig. 1a). Figure 1 also shows that stronger freshwater forcing results in a faster collapse of AMOC as compared to weaker forcing. The 0.2 and 0.4 Sv Pacific experiments also show that PMOC develops by the year 200 (Fig. 1b). Freshwater forcing was then turned off and CESM1 was run for another 300 years in order to allow the recovery of AMOC in the four experiments. AMOC begins recovering in intensity shortly after the year 500 during the 0.2 Sv forcing experiments, in contrast to 0.4 Sv forcing experiments during which AMOC does not begin to recover until after the year 600 (Fig. 1a). PMOC also returns to a collapsed state shortly after year 500 in the 0.2 Sv Pacific experiment and shortly after year 600 in the 0.4 Sv Pacific experiment (Fig. 1b).
The five sensitivity experiments created using CESM1 and used in this study. Freshwater forcing for the experiments involved uniform forcing in the subpolar North Atlantic (50°–70°N), the subpolar North Pacific (40°–60°N), and all other oceans (Global).
Temporal evolution of (a) AMOC and (b) PMOC for the CESM1 control, Global and Pacific experiments, and the Pacific Salt experiment, as indicated in the legend. AMOC and PMOC are defined as the maximum of the Atlantic and Pacific meridional streamfunction (Sv) below a 500-m depth, respectively.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
The last experiment (Table 1) involved drainage of freshwater from the subpolar North Pacific (40°–60°N) at a rate of −0.4 Sv and freshwater addition at a rate of 0.1 mm day−1 uniformly across all other oceans. This experiment is referred to as “Pacific Salt” given the resultant increase in salinity and density across the subpolar North Pacific. The motivation for this experiment was to explore whether the development of deep convection over the North Pacific would weaken AMOC because of the proposed seesaw-like interaction between the two deep overturning circulations (Saenko et al. 2004; Okazaki et al. 2010). PMOC developed during the first 50 years of the simulation and did not result in an AMOC collapse, but did weaken AMOC by about 5 Sv (Fig. 1). Once PMOC and AMOC reached a quasi-steady state for a period of 100 years in the simulation, the freshwater drain was turned off to explore whether PMOC would persist. Figure 1 shows that upon stopping the freshwater drain, PMOC returned to a collapsed state by year 300 and AMOC reintensified. As a result, the experiment was stopped at year 350. The experiment showed that PMOC is not dependent on the state of AMOC (both can be active or weak) under present-day conditions with an open Bering Strait, as also previously shown by Hu et al. (2012a, 2015). Here we build upon past work and explore changes that AMOC and PMOC can have on global SSTs and ENSO. Additional configuration details for the experiments can be found in Hu et al. (2021).
b. SST postprocessing and analyses
Climatology characteristics of global SSTs (°C) and surface wind stress (N m−2) associated with AMOC and PMOC states were analyzed using monthly mean data. Spectral and wavelet analyses of SSTs were conducted to describe periodic variations in terms of frequency distribution within these time series [such as ENSO in Stuecker et al. (2015)]. Spectral analysis measures the tendency for oscillations of a particular frequency to appear in time series, which permits detection of dominant frequencies (Bloomfield 2004). Since a sinusoid cannot match oscillations that change in amplitude over time, time series were analyzed during the AMOC collapse period (years 201–500) for Global and Pacific experiments, and the PMOC active period (years 101–250) for the Pacific Salt experiment. These time series were detrended to remove long-term changes in scale and tapered at 10% with a split-cosine bell to reduce leakage. A fast Fourier transform (FFT) was then performed to convert from time variance to frequency variance and a periodogram was computed from the real and imaginary FFT coefficients. The periodogram estimates were smoothed using a modified Daniell kernel between t − m and t + m inclusive (where t is time and m = 7) and then normalized so that the area under the curve equaled variance after detrending and tapering. Various tapering and smoothing coefficients were tested, but the dominant frequency relative to all frequencies remained unchanged overall. All ocean points were considered for the spectral analysis and areas where sea ice (SST ≤ −1.8°C) was present for at least a part of the year were demarcated in the respective figures.
A limitation with spectral analysis is the loss of temporal information, which can mask dominant frequency changes. Wavelet analysis offers a useful alternative for time series that contain nonstationary variance at numerous frequencies (Torrence and Compo 1998). The analysis outputs a two-dimensional plane with time on the x axis and Fourier period on the y axis, displaying power of the wavelet transform from which the dominant frequency can be assessed over time. The Morlet wavelet (product of a sine wave and Gaussian envelope) was used for the wavelet transform with a parameter of six (Farge 1992). Numerous wavelet functions can be used for the analysis, but the Morlet wavelet is better suited for capturing oscillatory behavior and offered sufficient resolution in the time and period axes (Torrence and Compo 1998). Seasonal means1 (DJF, MAM, JJA, and SON) were used instead of monthly means to reduce the sample size of time and to better visualize changes in frequency over years for the full simulation time series. The smallest resolvable scale of the Morlet wavelet was one season for accurate reconstruction and variance computation and was also the amount of time between discrete scales. Statistical significance was computed at the 95% confidence level using a white noise background. Wavelet software was provided by C. Torrence and G. Compo and is available with more detailed methodology at: http://paos.colorado.edu/research/wavelets/.
While no index can capture the full diversity of ENSO (Capotondi et al. 2015), the Niño-3.4 region in the east-central tropical Pacific (5.0°S–5.0°N and 170.0°–120.0°W; Bamston et al. 1997) is closely related to the overall ENSO state (L’Heureux et al. 2015). The oceanic Niño index (ONI) is based on Niño-3.4 region SSTs and is the index used by numerous operational centers for monitoring ENSO (Kousky and Higgins 2007; Barnston et al. 2019). Thus, ONI was used to characterize ENSO during various phases of AMOC and PMOC. ONI was computed using a methodology similar to that used by the NOAA Climate Prediction Center, which is as follows: (i) a centered 30-yr rolling mean was used to create a monthly climatology of the Niño-3.4 region for each simulation, (ii) monthly Niño-3.4 region SST anomalies were computed from the respective climatology, (iii) Niño-3.4 region SST anomalies were area-weighted averaged, and (iv) SST anomaly averages were then smoothed using a 3-month running mean for the full time series. ONI was normalized using the standard deviation of the resultant time series. Additionally, for comparison to present-day observations, several statistical and index calculations were also applied to the 0.25° daily NOAA Optimum Interpolation SST (OISST) version-2.1 product, spanning the years 1982–2020 (Banzon et al. 2020).
3. Results
a. Changes to global SSTs
During AMOC collapse in Global experiments, SSTs cool by more than 5°C across the subtropical and subpolar North Atlantic as compared to the CESM1 control (Figs. 2b,c), a well-known feature simulated by climate models in response to AMOC decline (e.g., Drijfhout et al. 2012). Southwesterly surface winds also weaken across the subpolar North Atlantic during AMOC collapse, as evident by northeasterly surface wind stress anomalies exceeding approximately 0.04 N m−2 (Figs. 2b,c). SSTs also cool by about 3°C across the subtropical and subpolar North Pacific due in part to a lack of PMOC development in Global experiments. In addition to SSTs cooling during AMOC collapse, surface winds also intensify across the subtropical and subpolar North Pacific (Figs. 2b,c). SST and surface wind stress anomalies of Global experiments are very similar, although anomalies are of greater magnitude during the 0.4 Sv experiment as compared to the 0.2 Sv, likely due to stronger freshwater forcing used in the former. The addition of freshwater to the subpolar North Atlantic and increased salinity across other oceans also results in a temperature anomaly dichotomy between the Northern and Southern Hemispheres (Figs. 2b,c). Anomalous warming is present in the Southern Hemisphere, which is generally accompanied by weaker westerly surface winds along the region of the Antarctic Circumpolar Current (Figs. 2b,c). The hemispheric temperature anomalies do not balance each other; Global experiments result in global surface temperature cooling of approximately 2°C during AMOC collapse (Fig. 3a). Cooler SSTs in the Northern Hemisphere are also accompanied by a larger sea ice extent (Figs. 2b,c).
Annual mean SSTs (°C) and surface wind stress (N m−2) of (a) the CESM1 control and (b)–(f) anomalies computed as the difference between the sensitivity experiments and the CESM1 control. The CESM1 control climatology is an 800-yr average [years 800–1599 in (a)] and the sensitivity experiment climatology consists of 300-yr averages spanning the period of AMOC collapse [years 201–500 in (b)–(e)], except for the Pacific Salt experiment in (f), which is a 150-yr average spanning the period of an active PMOC (years 101–250). Light blue contours contain locations where sea ice was present at any point during the respective simulations.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
(a) Global annual mean surface temperature (°C), and oceanic northward heat transport (PW) at (b) 24°N in the North Atlantic, (c) 20°N in the North Pacific, and (d) 14°S in the South Pacific for the CESM1 control, Global and Pacific experiments, and the Pacific Salt experiment, as indicated in the legend.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
AMOC collapse in Pacific experiments also results in anomalous cooling of more than 5°C and a weakening of southwesterly surface winds across the subtropical and subpolar North Atlantic (Figs. 2d,e). However, the meridional north–south SST gradient weakens across the Pacific Ocean (Figs. 2d,e) as compared to Global experiments (Figs. 2b,c). Anomalous cooling is of comparatively lower magnitude across the North Pacific in the 0.2 Sv Pacific experiment and anomalous warming is present in the 0.4 Sv Pacific experiment (Figs. 2d,e). Thus, added salinity in the subpolar North Pacific limits cooling during AMOC collapse (Figs. 2d,e). Anomalous warming and weakening of northerly surface winds also occur in the region of the California Current during Pacific experiments (Figs. 2d,e). Anomalous warming in the North Pacific is accompanied by strengthening of westerly surface winds, but anomalous surface wind stress is comparatively weaker along the eastern branch of the subpolar North Pacific gyre (Figs. 2d,e). Observational proxies and redox tracers have shown that the warming signature across the subpolar gyre in the North Pacific occurred during the Last Glacial Maximum, when PMOC was active and transporting heat northward (Rae et al. 2020). The smaller magnitude SST anomalies present in Pacific experiments are also evident in global mean surface temperatures; cooling of approximately 1°C occurs during Pacific experiments (Fig. 3a).
AMOC did not collapse during the Pacific Salt experiment, but did weaken by about 5 Sv and PMOC developed (Fig. 1). As a result, warming in excess of 5°C occurs across the subpolar North Pacific during the Pacific Salt experiment (Fig. 2f). However, very little change in surface wind stress occurs across the North Pacific during the Pacific Salt experiment (Fig. 2f). Additionally, the persistence of AMOC (albeit weaker) limits SST and surface wind stress anomalies across the Atlantic and South Pacific Oceans (Fig. 2f). Overall, warming of the global mean surface temperature occurs when PMOC and AMOC are active (Fig. 3a).
Wind can modulate AMOC variability (Polo et al. 2014; Ruprich-Robert and Cassou 2015). However, wind stress anomalies are also partly driven by the ocean’s forcing on the overlying atmosphere that can result from changes to SSTs (Waldman et al. 2021). SST changes can be driven in part by transport and mixing processes, which impact heat, momentum, and water fluxes along the air–sea interface (Liu and Liu 2014; Buckley and Marshall 2016). On synoptic-to-planetary scales, cooling in the Northern Hemisphere and warming in the Southern Hemisphere in response to changes in AMOC and PMOC states can also modulate the thermal wind and trade winds (Waldman et al. 2021), as evident in Fig. 2. The north–south hemispheric temperature dichotomy is partly driven by changes to oceanic northward heat transport across the Atlantic and Pacific basins (Fig. 3). During Global and Pacific experiments, oceanic northward heat transport is substantially reduced across the North Atlantic (Fig. 3b), which coincides with AMOC collapse and partly explains SST cooling north of the equator as compared to the CESM1 control. Across the North Pacific, little change is observed in the oceanic northward heat transport during Global experiments, but an increase occurs during the Pacific Salt and Pacific experiments as compared to the CESM1 control (Fig. 3c), which coincides with PMOC development and partly explains Pacific basin SST warming as compared to Global experiments (Hwang et al. 2021). Interestingly, oceanic southward heat transport across the South Pacific decreases for all sensitivity experiments (Fig. 3d), suggesting that changes to heat transport in the South Pacific may not be reflected in surface temperatures across the South Pacific.
The depth of the 20°C isotherm can be used to approximate the thermocline depth and provides characterization of subsurface oceanic conditions (Kessler 1990). Here we analyze changes related to thermocline depth as compared to the CESM1 control due to different states of AMOC and PMOC (Fig. 4). Regions exhibiting negative SST anomalies across the North Atlantic during Global and Pacific experiments generally exhibit shallower thermocline depth, with the exception of the Caribbean Sea, Gulf of Mexico, and southern region of the Gulf Stream, which exhibit deepening in relation to the slowdown or collapse of AMOC (Figs. 4b–f). Deepening of the thermocline occurs across North Pacific tropical latitudes during Global experiments, despite anomalous surface cooling (Figs. 4b,c). However, across the tropical Pacific region often used to characterize ENSO, a weak zonal dipole is evident in thermocline depth anomalies during Pacific experiments, which is indicative of a reduced zonal thermocline tilt and also consistent with the weaker equatorial trade winds (Figs. 4d,e; 2d,e).
As in Fig. 2, but showing the 20°C isotherm depth in meters for (a) the CESM1 control and (b)–(f) anomalies computed as the difference between the sensitivity experiments and the CESM1 control. Light blue contours contain locations where sea ice was present at any point during the respective simulations.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
To explore whether SST anomalies for the various experiments exhibited multiyear variability, specific regions of the Pacific and Atlantic were subset that were located across tropical and subtropical latitudes with different anomaly patterns (e.g., Fig. 2). Multiyear SST variability across the tropical Pacific is approximately double in magnitude during Global and Pacific experiments than in the CESM1 control (Fig. 5c). SST variance also doubles across the subtropical North Atlantic during Global and Pacific experiments (Fig. 5b). SST variance across the subtropical North Pacific differs from the CESM1 control mostly during Global experiments (Fig. 5a), during which SST variance doubles. Thus, while anomalous SST cooling occurs over the subtropical and subpolar North Atlantic and Pacific regions during Global experiments, and warming across the tropical Pacific during Pacific experiments, the signals have substantial multiyear variability, possibly related to decadal and multidecadal variability (e.g., Rashid et al. 2010; Russell and Gnanadesikan 2014; Levine et al. 2017; Moreno-Chamarro et al. 2020; Kim et al. 2020). Meehl et al. (2021) recently delineated the mutually interactive relationship between the tropical Pacific and Atlantic related to Pacific decadal variability and Atlantic multidecadal variability, whereby warmer tropical Atlantic SSTs can generate an opposite-sign response in tropical Pacific SSTs, and cooler tropical Pacific SSTs can produce a same-sign response in the tropical Atlantic. The processes and mechanisms driving cross-basin interaction include changes to the Walker circulation and midlatitude teleconnections (Meehl et al. 2021), which could potentially explain the decadal variability observed herein. Conversely, SST variance across the tropical Atlantic, subtropical South Pacific, and subtropical South Atlantic remain similar in magnitude across experiments and of comparatively smaller magnitude than other regions (Figs. 5d–f). The Pacific Salt experiment has comparable variance to the CESM1 control across the tropical Pacific and other regions considered (Fig. 5). However, the Pacific Salt experiment only has a 50-yr overlap of reduced AMOC strength with other experiments. Comparison to present-day observations (OISST) reveals consistent results (Fig. 5), although we note that the comparatively shorter temporal length of observations is a limiting factor in a robust analysis of multidecadal variability.
Variance of area weighted averages of annual SSTs (°C) computed using a 30-yr centered rolling mean across various regions of the Pacific and Atlantic basins, as indicated on the global map with black polygons. The temporal period of AMOC collapse is shown for the sensitivity simulations (years 201–500), except for the Pacific Salt experiment, which only contains the overlap years of an active PMOC (years 201–250) given the shortened duration of the experiment. The CESM1 control for the corresponding time period is also shown (years 1001–1300), as well as present-day observations (OISST; years 1982–2020), all indicated in the legend. The lower-left and upper-right coordinates for the oceanic regions of interest are indicated in the plot titles.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
The period and amplitude of the annual cycle remain consistent with the CESM1 control across the subtropical Pacific and Atlantic basins in both Northern and Southern Hemispheres during all sensitivity experiments (Figs. 6a,b,e,f). However, anomalous cooling occurs year-round across the subtropical North Pacific and Atlantic in the Global experiments (Figs. 6a,b), and conversely, anomalous warming occurs year-round across the subtropical South Pacific and Atlantic (Figs. 6e,f). Results for subtropical latitudes during Pacific experiments are similar to Global experiments but of weaker magnitude. Temperatures of the Pacific Salt experiment and OISSTs align with the CESM1 control, except across the subtropical North Pacific, which are warmer uniformly year-round (Fig. 6a). Thus, while there is substantial multiyear SST variability across the North Atlantic and Pacific regions, the annual cycle persists.
Annual cycle of area weighted averages of monthly SSTs (°C) across the same regions indicated in Fig. 5. The CESM1 control (800-yr mean), Global and Pacific experiments (300-yr mean), Pacific Salt experiment (150-yr mean), and present-day observations (OISST; 40-yr mean) are indicated in the legend.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
A reduction in seasonality occurs across tropical regions in Global and Pacific experiments (Figs. 6c,d). Anomalous warming occurs primarily during June–January across the tropical Atlantic in Global and Pacific experiments, which contributes to the reduced annual cycle amplitude (Fig. 6d). Temperatures vary about 1°–3°C across the tropical Atlantic throughout the year in Global and Pacific experiments as opposed to 4°C in the CESM1 control. Across the tropical Pacific, SSTs vary less than 1°C annually during Global and Pacific experiments as opposed to more than 1°C in the CESM1 control (Fig. 6c). The reduction in SST seasonality in Global experiments is accompanied by overall cooling as well during April–September (Fig. 6c). Timmermann et al. (2007) showed that intensified northeasterly trade winds weaken the annual cycle across the eastern tropical Pacific during AMOC collapse, which is also apparent during Global experiments in Fig. 2. However, Pacific experiments show intensified equatorial westerlies across the central-western Pacific (Figs. 2d,e), which can reduce the zonal slope of the equatorial thermocline, possibly weakening the annual cycle via warming of the cold tongue in the eastern Pacific (Fig. 6c). These conditions can also result in stronger eastern Pacific ENSO events, as occurred, for example, during 1977–2000 (An and Bong 2016; Capotondi and Sardeshmukh 2017). We also note that previous studies have identified an anticorrelation between seasonal cycle amplitudes and ENSO (Guilyardi 2006), suggesting that reduced seasonal cycles can result in amplified ENSO. Further analysis of changes to ENSO resulting from AMOC and PMOC states follow in a subsequent section.
b. A spectral perspective of SST changes
Spectral analyses of SSTs show that the annual cycle (0.08 cycles per month) is the dominant frequency (i.e., frequency of peak variance) for the CESM1 control and Pacific Salt experiment across the Pacific and Atlantic basins, in agreement with observations (Fig. 7). During Global and Pacific experiments, little change was observed in dominant frequency across subtropical latitudes (Figs. 7a–d). However, the dominant frequency changed from 0.08 cycles per month (annual cycle) in the CESM1 control to 0.02 cycles per month (about 4 years) in the Global and Pacific experiments across the tropical Pacific (Fig. 7f; Table 2). These results suggest that an AMOC collapse can amplify SST variance in a lower-frequency signal (0.02 cycles per month) and reduce the annual cycle signal (Table 2), regardless of whether PMOC develops. We note that changes to the magnitude of the semiannual cycle are not as substantial (Table 2).
(a)–(f) Spectral analysis of area weighted averages of monthly SSTs (°C) across regions indicated in Fig. 5. Years 201–500 were considered for the Global and Pacific experiments, years 1982–2020 were considered for OISST, and years 1001–1300 were considered for the CESM1 control for correspondence to the sensitivity experiments during AMOC collapse. Years 101–250 were considered for the Pacific Salt experiment, which were the years PMOC was active. Simulations and corresponding frequency of maximum variance for the tropical Pacific are indicated in the legend in (f). Star markers represent peak variance for each respective experiment and region.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
Spectral analysis of area weighted averages of monthly SSTs (°C) across the tropical Pacific (10.5°S, 170.5°W; 10.5°N, 120.5°W), as in Fig. 7. Years 201–500 were considered for the Global and Pacific experiments and years 1001–1300 were considered for the CESM1 control for correspondence to the sensitivity experiments during AMOC collapse. Years 101–250 were considered for the Pacific Salt experiment, which were the years the PMOC was active.
Spectral analysis was also conducted on each ocean pixel to visualize results across the global domain. The dominant frequency for most areas in the CESM1 control and observations is the annual cycle, including across the eastern equatorial Pacific (Figs. 8a,b). However, the west-central equatorial Pacific is characterized by a dominant frequency of comparatively lower magnitude (≤0.05). These regional differences in dominant frequency across the western Pacific region (Figs. 8a,b) could be related to the atmospheric–oceanic interaction resulting from SST perturbations in relation to ENSO (Lau and Nath 2003) and the equatorial warm pool (An et al. 2012). In contrast, the dominant frequency across the western and west-central Indo-Pacific is comparatively higher in magnitude (Figs. 8a,b; ≥0.15), possibly associated with the semiannual cycle from equatorial solar insolation (Wang and Wang 1999). Global and Pacific experiments show a decrease in dominant frequency of approximately 0.05 cycles per month across the equatorial central-eastern Pacific as compared to the CESM1 control (Figs. 8c–f). The depletion of the annual cycle frequency occurs across the Niño-3.4 region, suggesting that changes to ENSO can occur during AMOC collapse even if PMOC is active (Figs. 8c–f). Little change is observed in dominant frequency during the Pacific Salt experiment (Fig. 8g), consistent with earlier results.
Spectral analysis conducted per pixel on monthly SSTs (°C) for the same time periods as in Fig. 7. The dominant frequency (i.e., frequency of peak variance) for the (a) CESM1 control, (b) OISSTs, and (c)–(g) the difference in dominant frequency between the sensitivity experiments and the CESM1 control are shown. The Niño-3.4 region is indicated with a polygon. Light blue contours contain locations where sea ice was present at any point during the respective simulations.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
Changes in the dominant frequency across the tropical Atlantic are mixed during Global and Pacific experiments (Figs. 8c–f), with both a decrease and increase in dominant frequency observed related to regional shifts of the intertropical convergence zone (ITCZ; Fig. 9a), as has been observed during past AMOC collapses (McGee et al. 2014; Liu and Hu 2015; Yu and Pritchard 2019). The position of the ITCZ is closely aligned with the vertically rising branch of the thermally direct Hadley circulation and both generally migrate toward the warmer hemisphere (Schneider et al. 2014; Richter et al. 2017; Kang et al. 2018). ITCZ shifts toward the warmer hemisphere (south) also occur across the Pacific during Global experiments, when the interhemispheric temperature differences are more pronounced than during Pacific and Pacific Salt experiments (Fig. 9b).
ITCZ median latitude across the tropical (a) Atlantic and (b) Pacific Oceans for the CESM1 control and various experiments. The median ITCZ latitude was estimated using maximum wind convergence at 850 hPa between 15°S–20°N and 35°–15°W for the Atlantic and between 3°S–20°N and 170°E–90°W for the Pacific.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
SST variance at the annual cycle frequency (0.08 cycles per month) is highest across subtropical and subpolar latitudes in the Northern and Southern Hemispheres of the CESM1 control (Fig. 10a), as would be expected given the amplified seasonal cycle as compared to the tropics. The meridional gradient across the North Pacific may be related to the presence of sea ice part of the year, as demarcated with the red contour line (Fig. 10a). The spatial variance pattern of the CESM1 control is similar to observations, but observations are approximately half the magnitude, possibly due to the normalization of periodogram coefficients or the shorter duration of observations (Figs. 10a,b). Global and Pacific experiments show regional changes in variance at the annual cycle frequency, with both increases and decreases across high-latitude regions in the Northern Hemisphere (Figs. 10c–f) that could be associated with changes to surface wind stress (e.g., Figs. 2b–e). A reduction in variance at the annual cycle frequency occurs across the eastern equatorial Pacific and Atlantic (Figs. 10c–f), which helps explain the dominant frequency changes during AMOC collapse shown in Figs. 8c–8f. During the Pacific Salt experiment, SST variance decreases across most latitudes (Fig. 10g). An active PMOC and AMOC may result in reduced SST variance across most regions because of water overturning in the subpolar Atlantic and Pacific.
As in Fig. 8, but showing monthly SST variance for the (a) CESM1 control, (b) OISSTs (doubled due to comparatively low variance magnitude), and (c)–(g) the variance difference between the sensitivity experiments and the CESM1 control, all at the annual cycle frequency (0.08 cycles per month). Red contours contain locations where sea ice was present at any point during the respective simulations.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
Wavelet analysis of Niño-3.4 seasonal SSTs shows power consistently throughout a period of 1 year during the CESM1 control, which suggests that an SST annual cycle signal is persistent within the Niño-3.4 region throughout the time series (Fig. 11a). The 2–7-yr oscillations also show power throughout the time series of the CESM1 control, representative of ENSO. While more temporally sparse, an approximately 16-yr oscillation is also evident in the CESM1 control, as well as weaker signals at periods exceeding 20 years, which relate to longer modes of variability (e.g., Capotondi et al. 2020). Global experiments show a loss of power at the annual cycle at approximately 100–600 years (Figs. 11b,c) as compared to the CESM1 control, supporting earlier spectral analysis results. Power at 2–7-yr oscillations increases (significant at the 95% confidence level) during Global experiments, illustrating an intensification of ENSO amplitude. Power also appears at the 16-yr period and at periods exceeding 20 years during Global experiments, but the signals do not appear to differ substantially from the CESM1 control. This result could be an indication that the interdecadal Pacific oscillation (Power et al. 1999) may not be the low-frequency ENSO residual. Results for Pacific experiments (Figs. 11d,e) are similar to Global experiments (Figs. 11b,c). The Pacific Salt experiment shows that the annual cycle and ENSO remain comparatively similar to the CESM1 control (Fig. 11f), which is surprising given the changes to the North Pacific circulation.
The local wavelet power spectral for the Niño-3.4 region seasonal SSTs (area weighted average) using the Morlet wavelet (as in Torrence and Compo 1998) for the (a) CESM1 control and (b)–(f) sensitivity experiments. 100 years of the CESM1 control were added to the start of each time series to pad years prior to experiment responses.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
c. Changes to ENSO frequency and amplitude
In this subsection, we analyze changes to SSTs specific to the tropical Pacific and Niño-3.4 region. The SST climatology across the Niño-3.4 region was computed using a 30-yr rolling mean applied to each month and shows temporal homogeneity in the CESM1 control, with an annual cycle centered at 26°C, seasonally varying approximately ±2°C (Fig. 12a). The SST climatology for Global experiments differs from the CESM1 control during AMOC collapse, characterized by reduced seasonality and a cooler mean temperature (Figs. 12b,c). However, when PMOC is active during the Pacific experiments, the decline in seasonality as compared to the Global experiments is less pronounced at approximately ±1°C and overall warming of the mean temperature occurs (Figs. 12d,e). Reduced seasonality is related to AMOC collapse, but the differing responses between cooling (Global experiments) and warming (Pacific experiments) are related to cascading influences from an inactive or active PMOC, respectively. In contrast, the Pacific Salt experiment shows little change in the Niño-3.4 region SST climatology (Fig. 12f) as compared to the CESM1 control. These results suggest that the annual cycle of SSTs across the Niño-3.4 region can strengthen or weaken depending on how AMOC and PMOC evolve, but the effect appears to be more sensitive to the evolution of AMOC than PMOC.
Niño-3.4 region monthly SSTs (°C) for the (a) CESM1 control and (b)–(f) sensitivity experiments. The gray line indicates the SST climatology for the respective simulations, which were computed using a rolling 30-yr monthly mean applied to each month individually. Thus, the 30-yr monthly rolling mean was applied to January, then February, and so forth. The December and June rolling means are shown with blue and red lines, respectively. SSTs and climatology as indicated in the legend.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
ONI time series for the various sensitivity experiments are shown in Fig. 13. Results show that ENSO amplitude is large with greater frequency in Global and Pacific experiments (Figs. 13b–e) during AMOC collapse as compared to the CESM1 control and Pacific Salt experiment (Figs. 13a,f). Quantitatively, ONI variance values of Global and Pacific experiments during AMOC collapse are approximately 0.2°C2 greater than variance of the CESM1 control (20% increase; Table 3). As AMOC increases in intensity during years 551–650 due to the stopped freshwater hosing at year 500, the 0.2 Sv Global and Pacific experiments show a decrease in ONI variance of about 0.4°C2 (35% decrease; Table 3). The 0.4 Sv Global and Pacific experiments also show a decrease in ONI variance once the freshwater hosing ends at year 500, but not until years 651–800 (Table 3). This delay in the reduction of ONI variance for the 0.4 Sv experiments is likely due in part to AMOC remaining in a collapsed state longer than the 0.2 Sv experiments. Comparison of the Pacific Salt experiment and CESM1 control reveals that when PMOC and AMOC are both active, ENSO variability decreases. This result is evident in Fig. 13f during the years 101–250, when PMOC is active. Table 4 also shows that ONI variance for the Pacific Salt experiment is lower in magnitude (7% decrease) than that of the CESM1 control during years 101–250. This result suggests that active overturning circulations across both basins can reduce oscillatory SST behavior in the tropical Pacific.
ONI for the (a) CESM1 control and (b)–(f) sensitivity experiments. Red shading indicates ONI ≥ +0.5 (El Niño conditions) and blue indicates ONI ≤ −0.5 (La Niña conditions). 100 years of the CESM1 control were added at the start of all time series for illustration of ONI prior to AMOC and PMOC modulations, delineated with the solid black vertical lines. The dashed vertical lines show the approximate years when AMOC was collapsed for the Global and Pacific experiments in (b)–(e) and when PMOC was active for the Pacific Salt experiment in (f).
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
ONI variance for different time periods of the Global and Pacific freshwater hosing experiments as compared to the CESM1 control. Years 201–500 include the time period of AMOC collapse for the sensitivity experiments. Years 551–650 contain the time period when the 0.2 Sv experiments overshoot AMOC strength in the CESM1 control and 0.4 Sv experiments contain a mostly collapsed AMOC state. Years 651–800 contain the AMOC overshoot period for the 0.4 Sv experiments.
ONI variance for different time periods of the Pacific Salt experiment as compared to the CESM1 control. Years 101–250 include the time period of PMOC intensification, years 251–300 contain the time period of rapid PMOC weakening, and years 301–350 contain the time period of a PMOC collapsed state.
After a boreal summer or autumn onset, El Niño and La Niña usually peak in amplitude during DJF months (Capotondi et al. 2015). This property is typically referred to as ENSO phase locking. Multiple mechanisms have been proposed to explain the phase preference of ENSO to peak during boreal winter months (Neelin et al. 2000). For example, using a simplified coupled atmosphere–ocean model (Zebiak and Cane 1987), Tziperman et al. (1997) found that one of the most dominant mechanisms by which the equatorial Pacific seasonal cycle influences ENSO is through the evolution of background wind divergence, contingent on the seasonal motion of the ITCZ. In contrast, McGregor et al. (2012) found weakening of climatological wind speeds south of the equator to be one of the key factors in the seasonal termination of strong El Niño events using an intermediate complexity atmosphere model, while Stein et al. (2014) found seasonal modulation of ENSO’s coupled stability to be a dominant factor for ENSO synchronization to the annual cycle using a parametric recharge oscillator. Despite specific hypotheses with observational and modeling support, some uncertainty remains regarding the specific underlying physical processes that explain ENSO phase locking (Chen and Jin 2020). Strong equatorial asymmetry of SSTs between the western and eastern equatorial Pacific is also characteristic during boreal winter, related to the eastern Pacific cold tongue, low-level convergence partly driven by the position of the ITCZ, and coastline influences (Choi et al. 2015). These DJF climatological features across the equatorial Pacific are captured by the CESM1 control, albeit with a cold bias in the mean SST climatology (Zhang et al. 2017), a limitation noted in numerous coupled modeling systems (Zhang and Sun 2014; Sun et al. 2016).
During DJF El Niño in Global and Pacific experiments, anomalous warming is homogeneous across the Niño-3.4 region (Figs. 14b–e) and of higher magnitude than the CESM1 control (Fig. 14a) related to previously shown changes in background climatic conditions. Beyond the Niño-3.4 region during DJF El Niño events in Global and Pacific experiments, anomalous SST warming across the eastern equatorial Pacific is also of higher magnitude than the CESM1 control (Figs. 14b,c). Easterly trade winds also gain a southward component across the western Niño-3.4 region during Global experiments (Figs. 14b,c) partly related to southward shift of the ITCZ (e.g., Fig. 9). During Pacific experiments, trade winds weaken across the western Niño-3.4 region (Figs. 14d,e) as compared to the CESM1 control (Fig. 14a). These results show that DJF El Niño becomes warmer during AMOC collapse irrespective of PMOC state, with anomalies approximately 1°C warmer than the CESM1 control across the central-to-eastern tropical Pacific. In contrast, Pacific Salt experiment SSTs are comparable to the CESM1 control (Fig. 14f). SST anomalies for DJF La Niña during Global and Pacific experiments show intensified cooling as compared to CESM1 control DJF La Niña events (Figs. 14g–k). Global experiments also show a stronger northeasterly wind component associated with trade winds east of the Niño-3.4 region and across the southern Niño-3.4 region, with overall strengthening of the westward zonal wind. Increased divergence of trade winds is present across the Niño-3.4 region during La Niña events in Pacific experiments, and convergence of trade winds east of the Niño-3.4 region (Figs. 14j,k). Anomalies related to Pacific Salt experiment DJF La Niña events are comparable to CESM1 control La Niña events (Figs. 14g,l).
DJF mean SST anomalies (°C) and surface wind stress (N m−2) during (a)–(f) El Niño (ONI ≥ +0.5) and (g)–(l) La Niña (ONI ≤ −0.5) as defined using ONI. The Niño-3.4 region is indicated with a black polygon. Time periods considered are the same as in Fig. 2.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
d. Influence of oceanic processes
In addition to atmospheric and surface processes, ENSO precursors also include oceanic memory processes associated with subsurface temperature anomalies, whereby variations in equatorial thermocline depth and observed warm water volume can precede tropical Pacific SST variability (Meinen and McPhaden 2000; Rustic et al. 2020). Variations in subsurface temperatures and thermocline depth can be communicated to the surface by mean upwelling (i.e., “thermocline feedback”) across the central and eastern equatorial Pacific (Wen et al. 2014). Previous studies have also put forth the “recharge–discharge oscillator” theory, by which warm water volume (≥20°C) of the tropical Pacific increases prior to El Niño (i.e., recharge) and decreases during El Niño (i.e., discharge), contributing to the timing of ENSO events (Jin 1997a,b; Meinen and McPhaden 2000; Chen et al. 2015). Here we analyze the depth of the 20°C isotherm during DJF months across the equatorial Pacific (Table 5), which can be used to represent the thermocline (Kessler 1990), along with mean ocean temperatures between 10°N and 10°S to a depth of 285 m (Fig. 15), which may help explain ENSO’s amplification.
DJF mean thermocline depth in meters, represented as the depth of the 20°C isotherm (Kessler 1990), across the tropical Pacific region (10°S–10°N, 160°E–80°W) during climatology, El Niño, and La Niña. ENSO events defined using ONI and the Niño-3.4 region SSTs. ENSO events and climatology are derived from the years 800–1599 for the CESM1 control, 201–500 for Global and Pacific experiments, and 101–250 for the Pacific Salt experiment.
DJF mean vertical temperature anomalies (°C) to a depth of 285 m across the tropical Pacific during (a)–(f) El Niño (ONI ≥ +0.5) and (g)–(l) La Niña (ONI ≤ −0.5) for the CESM1 control and sensitivity experiments. Dashed lines represent 20°C (black), 15°C (dark gray), and 12°C (gray) isotherms. Time periods considered are the same as in Fig. 2.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
All sensitivity experiments show limited changes in thermocline depth across the tropical Pacific during El Niño and La Niña relative to the CESM1 control (Table 5; Fig. 15). However, Global and Pacific experiments show positive and negative temperature anomalies of greater magnitude during both El Niño and La Niña (Figs. 15b–e,h–k). These results show that the steep temperature gradient (i.e., thermocline) occurs at a similar depth for the CESM1 control and all experiments, but the amplified positive and negative temperature anomalies during Global and Pacific experiments can help explain the increase in ENSO variance, given that increased warming and cooling results in stronger El Niño and La Niña events. These results also show that amplified oscillatory behavior associated with ENSO extends into the subsurface (down to approximately 285 m), which in effect counteracts any substantial thermocline changes (Figs. 15b–e,h–k) across the tropical Pacific.
Thermocline depth across the Niño-3.4 region can also be modulated by mass exchange of tropical and subtropical waters driven by subtropical cells (STCs; Zeller et al. 2019). In contrast to deep overturning circulations (AMOC and PMOC), STCs are shallow (e.g., ≤300 m) overturning circulations that serve as an oceanic pathway between subtropical latitudes of the Northern and Southern Hemispheres (such as 30°N and 30°S) and the equator (Wen et al. 2014). STCs transport water poleward near the surface and equatorward at deeper levels, which can modulate thermocline depth via volume transport and thermal anomalies that can affect equatorial heat content and ENSO (Schott et al. 2004; Feng et al. 2018). STCs are present at latitudes away from the deep tropics (e.g., beyond approximately 5°–7°N and 5°–7°S), as the equatorial current is present near the equator. Studies have found that a slowdown of equatorward pycnocline flow is associated with warming of equatorial SSTs, related to a reduction of equatorial upwelling (Capotondi et al. 2005). Warming associated with a slowdown of STC equatorward flow can reduce the seasonal cycle of tropical Pacific SSTs and amplify ENSO as a result of changes to the background state. Here we consider changes to STCs given the influence on ENSO.
STCs in the CESM1 control differ substantially during DJF and JJA months, and differences are also based on which hemisphere they are located in (Figs. 16a,g). During DJF, the Northern Hemisphere STC is comparatively stronger, while the Southern Hemisphere STC is comparatively stronger during JJA (Figs. 16a,g). While an overall slowdown of STC related volume transport would weaken the seasonal cycle near the equator, the asymmetry in STC strength across hemispheres complicates the response that could be inferred as a result of STC weakening or strengthening. Global and Pacific experiments show strengthening of the Northern Hemisphere STC during DJF as compared to the CESM1 control (Figs. 16b–e). This result illustrates that more volume transport occurs in the Northern Hemisphere during boreal winter, potentially resulting in increased poleward flow of warm water at and near the surface (and relative equatorward flow of deeper, cooler water). In contrast, the Southern Hemisphere STC weakens during JJA as compared to the CESM1 control for both Global and Pacific experiments (Figs. 16h–k). This result shows that less volume transport occurs in the Southern Hemisphere during austral winter, potentially resulting in reduced equatorward pycnocline transport and warmer equatorial background conditions. Given the possible several month lag between changes to STC strength and conditions at the equator, we hypothesize that this asymmetrical response in STC strength due to changes in AMOC and PMOC contributes to the weakening of the tropical Pacific seasonal cycle. We further explore this in Fig. 17, which focuses on changes to the mixed layer depth. During the Pacific Salt experiment, positive anomalies appear at deeper levels (≥500 m; not shown) when the deep meridional overturning circulation is active across the Pacific basin (i.e., PMOC), which is also the case for the Pacific experiments below 500 m.
(a)–(f) DJF and (g)–(l) JJA mean Pacific meridional streamfunction (Sv) to a depth of 450-m shown with filled contours (color bar at lower left). Anomalies for sensitivity experiments, as computed from the CESM1 control during the same years as Fig. 2, are shown with nonfilled contour lines, where solid lines represent positive anomalies and dashed lines represent negative anomalies [in (b)–(f) and (h)–(l); color bar at lower right]. The Northern Hemisphere STC flows from south-to-north in shallower levels and north-to-south in deeper levels, with red filled contours indicating a clockwise circulation. The Southern Hemisphere STC flows from north-to-south in upper levels and south-to-north in deeper levels, with blue filled contours indicating a counterclockwise circulation. Positive anomaly values (color bar at lower right) within the Northern Hemisphere show enhanced northward movement across the shallower branch of the STC relative to the CESM1 control [e.g., in (b) and (c)]. Positive anomaly values (color bar at lower right) within the Southern Hemisphere show reduced southward movement across the shallower branch of the STC [e.g., in (h)–(k)]. Schematic illustration of circulation flow is contained in (a) and (g).
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
Mean mixed layer depth (meters) shown for the CESM1 control during (a) DJF and (g) JJA, with the color bar shown at the lower left. Anomalies for sensitivity experiments computed from the CESM1 control, for (b)–(f) DJF and (h)–(l) JJA, with the color bar shown at the lower right. Time periods considered are the same as in Fig. 2.
Citation: Journal of Climate 35, 1; 10.1175/JCLI-D-21-0172.1
Here we analyze mixed layer depth to further explore our hypothesis, which is that asymmetrical STCs contribute to a weakened seasonal cycle across the tropical Pacific (Fig. 17). Mixed layer depth is deeper during DJF across the tropical Pacific and Niño-3.4 region (Fig. 17a) and shallower during JJA (Fig. 17g). Mixed layer depth is deeper during cooler season months (e.g., Fig. 6) because negative buoyancy related to surface cooling mixes temperature to deeper layers. However, during Global and Pacific experiments, deepening of the mixed layer depth occurs during JJA (particularly evident during Global experiments; Figs. 17h,i). In contrast, mixed layer depth becomes more shallow during DJF, particularly for Pacific experiments (Figs. 17d,e). Thus, results show an overall reduction in mixed layer depth differences between DJF and JJA, which leads to a reduced temperature contrast within the mixed layer and a reduced SST seasonal cycle across the tropical Pacific, likely partly related to STC changes, which can result in ENSO amplification.
4. Conclusions
Results from this study show that AMOC and PMOC can strongly influence global SSTs and ENSO. Some key results are summarized below:
The tropical Pacific annual cycle amplitude decreases and ENSO amplitude increases as a result of an AMOC shutdown irrespective of PMOC development.
Active overturning circulations in both the Atlantic and Pacific basins reduce ENSO amplitude and variance of SSTs globally.
Increased magnitudes of SST anomalies (both positive and negative) occur across the Niño-3.4 region and extend into the eastern equatorial Pacific during AMOC collapse and ENSO events.
AMOC- and PMOC-related changes to ENSO amplitude could be physically explained in part by local changes to surface wind stress across the tropical Pacific. Increased northeasterly trade winds reduced SST seasonality during AMOC collapse when PMOC was inactive (in agreement with Timmermann et al. 2007; Dong and Sutton 2007). Timmermann et al. (2007) proposed that changes to ENSO as a result of AMOC collapse were due to the following causal chain of events: (i) North Atlantic cooling generates a Caribbean anticyclone that enhances cooling in the northeastern North Pacific, (ii) northeasterly trade winds then intensify, which weaken the SST seasonal cycle in the tropical Pacific, and then (iii) ENSO intensifies. A causal inference analysis (such as Granger causality; McGraw and Barnes 2018) could help further quantify contributions from proposed precursor mechanisms and disentangle whether the signals were due to direct causal relationships or autocorrelations.
There are several opportunities to gain further insight and address potential limitations with future work. While the several multicentennial simulations used in this study provide evidence of changes related to AMOC and PMOC, it is possible that a larger ensemble of simulations could provide a more robust assessment of the influence of overturning circulations on the tropical Pacific. Additionally, this study could be repeated using transient climate scenarios (e.g., Hu et al. 2015) instead of the CESM1 preindustrial control to determine if the response is consistent. It is possible that positive feedbacks associated with increased greenhouse gases could change the response of physical mechanisms associated with AMOC- and PMOC-driven ENSO modulations. Associated downstream impacts to landmasses, such as extreme precipitation and drought, should also be examined to estimate potential environmental and societal consequences related to AMOC and PMOC. Information gained from studies of interactions among modes of variability, characterized by high and low frequencies, can better prepare society as the climate continues to change.
Acknowledgments
Portions of this study were supported by the Regional and Global Model Analysis (RGMA) component of Earth and Environmental Systems Modeling in the Earth and Environmental Systems Sciences Division of the U.S. Department of Energy’s Office of Biological and Environmental Research (BER) via National Science Foundation IA 1947282. This work also was supported by the National Center for Atmospheric Research (NCAR), which is a major facility sponsored by the National Science Foundation (NSF) under Cooperative Agreement 1852977. Computing resources (ark:/85065/d7wd3xhc) were partially provided by the Climate Simulation Laboratory at NCAR’s Computational and Information Systems Laboratory, sponsored by NSF and other agencies. This research also used resources of the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract DE-AC02-05CH11231. We also acknowledge Dennis Shea (NCAR) and Richard Valent (NCAR) for assistance with software resources pertaining to the spectral analysis.
Data availability statement
The software developed for this study, which includes the creation of data products (e.g., ONI time series), spectral and wavelet analyses, and figures, is available open-source at https://github.com/mariajmolina/climatico. Postprocessed CESM model simulations shown in the figures are available on Zenodo (https://zenodo.org/record/5683537 and https://zenodo.org/record/5683569). Inquiry for access to experiment data output that were not used in this study should be forwarded to Aixue Hu (NCAR; ahu@ucar.edu). The OISST v2.1 product is freely available from the NCAR Research Data Archive (NCEI 2007).
REFERENCES
An, S.-I., and H. Bong, 2016: Inter-decadal change in El Niño-Southern Oscillation examined with Bjerknes stability index analysis. Climate Dyn., 47, 967–979, https://doi.org/10.1007/s00382-015-2883-8.
An, S.-I., J.-W. Kim, S.-H. Im, B.-M. Kim, and J.-H. Park, 2012: Recent and future sea surface temperature trends in tropical Pacific warm pool and cold tongue regions. Climate Dyn., 39, 1373–1383, https://doi.org/10.1007/s00382-011-1129-7.
Azad, S., and M. Rajeevan, 2016: Possible shift in the ENSO-Indian monsoon rainfall relationship under future global warming. Sci. Rep., 6, 20 145, https://doi.org/10.1038/srep20145.
Banzon, V., T. M. Smith, M. Steele, B. Huang, and H.-M. Zhang, 2020: Improved estimation of proxy sea surface temperature in the arctic. J. Atmos. Oceanic Technol., 37, 341–349, https://doi.org/10.1175/JTECH-D-19-0177.1.
Bamston, A. G., M. Chelliah, and S. B. Goldenberg, 1997: Documentation of a highly ENSO-related SST region in the equatorial Pacific. Atmos.–Ocean, 35, 367–383, https://doi.org/10.1080/07055900.1997.9649597.
Barnston, A. G., M. K. Tippett, M. Ranganathan, and M. L. L’Heureux, 2019: Deterministic skill of ENSO predictions from the North American Multimodel Ensemble. Climate Dyn., 53, 7215–7234, https://doi.org/10.1007/s00382-017-3603-3.
Bloomfield, P., 2004: Fourier Analysis of Time Series: An Introduction. John Wiley & Sons, 288 pp.
Buckley, M. W., and J. Marshall, 2016: Observations, inferences, and mechanisms of the Atlantic Meridional Overturning Circulation: A review. Rev. Geophys., 54, 5–63, https://doi.org/10.1002/2015RG000493.
Burls, N. J., A. V. Fedorov, D. M. Sigman, S. L. Jaccard, R. Tiedemann, and G. H. Haug, 2017: Active Pacific Meridional Overturning Circulation (PMOC) during the warm Pliocene. Sci. Adv., 3, e1700156, https://doi.org/10.1126/sciadv.1700156.
Caesar, L., G. McCarthy, G. Thornalley, N. Cahill, and S. Rahmstorf, 2021: Current Atlantic Meridional Overturning Circulation weakest in last millennium. Nat. Geosci., 14, 118–120, https://doi.org/10.1038/s41561-021-00699-z.
Cai, W., and Coauthors, 2014: Increasing frequency of extreme El Niño events due to greenhouse warming. Nat. Climate Change, 4, 111–116, https://doi.org/10.1038/nclimate2100.
Capotondi, A., and P. D. Sardeshmukh, 2017: Is El Niño really changing? Geophys. Res. Lett., 44, 8548–8556, https://doi.org/10.1002/2017GL074515.
Capotondi, A., M. A. Alexander, C. Deser, and M. J. McPhaden, 2005: Anatomy and decadal evolution of the Pacific subtropical–tropical cells (STCs). J. Climate, 18, 3739–3758, https://doi.org/10.1175/JCLI3496.1.
Capotondi, A., and Coauthors, 2015: Understanding ENSO diversity. Bull. Amer. Meteor. Soc., 96, 921–938, https://doi.org/10.1175/BAMS-D-13-00117.1.
Capotondi, A., C. Deser, A. Phillips, Y. Okumura, and S. Larson, 2020: ENSO and Pacific decadal variability in the Community Earth System Model version 2. J. Adv. Model. Earth Syst., 12, e2019MS002022, https://doi.org/10.1029/2019MS002022.
Chen, H.-C., and F.-F. Jin, 2020: Fundamental behavior of ENSO phase locking. J. Climate, 33, 1953–1968, https://doi.org/10.1175/JCLI-D-19-0264.1.
Chen, H.-C., C.-H. Sui, Y.-H. Tseng, and B. Huang, 2015: An analysis of the linkage of Pacific subtropical cells with the recharge–discharge processes in ENSO evolution. J. Climate, 28, 3786–3805, https://doi.org/10.1175/JCLI-D-14-00134.1.
Choi, K.-Y., G. A. Vecchi, and A. T. Wittenberg, 2015: Nonlinear zonal wind response to ENSO in the CMIP5 models: Roles of the zonal and meridional shift of the ITCZ/SPCZ and the simulated climatological precipitation. J. Climate, 28, 8556–8573, https://doi.org/10.1175/JCLI-D-15-0211.1.
Clark, P. U., N. G. Pisias, T. F. Stocker, and A. J. Weaver, 2002: The role of the thermohaline circulation in abrupt climate change. Nature, 415, 863–869, https://doi.org/10.1038/415863a.
Dahl, K. A., A. J. Broccoli, and R. J. Stouffer, 2005: Assessing the role of North Atlantic freshwater forcing in millennial scale climate variability: A tropical Atlantic perspective. Climate Dyn., 24, 325–346, https://doi.org/10.1007/s00382-004-0499-5.
Danabasoglu, G., S. C. Bates, B. P. Briegleb, S. R. Jayne, M. Jochum, W. G. Large, S. Peacock, and S. G. Yeager, 2012: The CCSM4 ocean component. J. Climate, 25, 1361–1389, https://doi.org/10.1175/JCLI-D-11-00091.1.
Dickson, R. R., and J. Brown, 1994: The production of North Atlantic Deep Water: Sources, rates, and pathways. J. Geophys. Res., 99, 12 319–12 341, https://doi.org/10.1029/94JC00530.
Dong, B.-W., and R. T. Sutton, 2002: Adjustment of the coupled ocean–atmosphere system to a sudden change in the thermohaline circulation. Geophys. Res. Lett., 29, 1728, https://doi.org/10.1029/2002GL015229.
Dong, B.-W., and R. T. Sutton, 2007: Enhancement of ENSO variability by a weakened Atlantic thermohaline circulation in a coupled GCM. J. Climate, 20, 4920–4939, https://doi.org/10.1175/JCLI4284.1.
Drijfhout, S., G. J. Van Oldenborgh, and A. Cimatoribus, 2012: Is a decline of AMOC causing the warming hole above the North Atlantic in observed and modeled warming patterns? J. Climate, 25, 8373–8379, https://doi.org/10.1175/JCLI-D-12-00490.1.
Farge, M., 1992: Wavelet transforms and their applications to turbulence. Annu. Rev. Fluid Mech., 24, 395–458, https://doi.org/10.1146/annurev.fl.24.010192.002143.
Fedorov, A. V., and S. G. Philander, 2001: A stability analysis of tropical ocean–atmosphere interactions: Bridging measurements and theory for El Niño. J. Climate, 14, 3086–3101, https://doi.org/10.1175/1520-0442(2001)014<3086:ASAOTO>2.0.CO;2.
Feng, J., D. Hu, F. Jin, F. Jia, Q. Wang, and S. Guan, 2018: The different relationship of Pacific interior subtropical cells and two types of ENSO. J. Oceanogr., 74, 523–539, https://doi.org/10.1007/s10872-018-0478-9.
Ferreira, D., and Coauthors, 2018: Atlantic-Pacific asymmetry in deep water formation. Annu. Rev. Earth Planet. Sci., 46, 327–352, https://doi.org/10.1146/annurev-earth-082517-010045.
Guilyardi, E., 2006: El Niño–mean state–seasonal cycle interactions in a multi-model ensemble. Climate Dyn., 26, 329–348, https://doi.org/10.1007/s00382-005-0084-6.
Hu, A., G. A. Meehl, and W. Han, 2007: Role of the Bering Strait in the thermohaline circulation and abrupt climate change. Geophys. Res. Lett., 34, L05704, https://doi.org/10.1029/2006GL028906.
Hu, A., and Coauthors, 2010: Influence of Bering Strait flow and North Atlantic circulation on glacial sea-level changes. Nat. Geosci., 3, 118–121, https://doi.org/10.1038/ngeo729.
Hu, A., G. A. Meehl, W. Han, and J. Yin, 2011: Effect of the potential melting of the Greenland Ice Sheet on the Meridional Overturning Circulation and global climate in the future. Deep-Sea Res. II, 58, 1914–1926, https://doi.org/10.1016/j.dsr2.2010.10.069.
Hu, A., G. A. Meehl, W. Han, A. Abe-Ouchi, C. Morrill, Y. Okazaki, and M. O. Chikamoto, 2012a: The Pacific-Atlantic seesaw and the Bering Strait. Geophys. Res. Lett., 39, L03702, https://doi.org/10.1029/2011GL050567.
Hu, A., and Coauthors, 2012b: Role of the Bering Strait on the hysteresis of the ocean conveyor belt circulation and glacial climate stability. Proc. Natl. Acad. Sci. USA, 109, 6417–6422, https://doi.org/10.1073/pnas.1116014109.
Hu, A., G. A. Meehl, W. Han, B. Otto-Bliestner, A. Abe-Ouchi, and N. Rosenbloom, 2015: Effects of the Bering Strait closure on AMOC and global climate under different background climates. Prog. Oceanogr., 132, 174–196, https://doi.org/10.1016/j.pocean.2014.02.004.
Hu, A., G. A. Meehl, N. Rosenbloom, M. J. Molina, and W. G. Strand, 2021: The influence of variability in meridional overturning on global ocean circulation. J. Climate, 34, 7697–7716, https://doi.org/10.1175/JCLI-D-21-0119.1.
Hunke, E. C., W. H. Lipscomb, A. Turner, N. Jeffery, and S. Elliott, 2008: CICE: The Los Alamos Sea ice model documentation and software user’s manual version 4.0 LA-CC-06-012. Tech. Rep. LA-CC-06-012, 500 pp.
Hurrell, J. W., and Coauthors, 2013: The Community Earth System Model: A framework for collaborative research. Bull. Amer. Meteor. Soc., 94, 1339–1360, https://doi.org/10.1175/BAMS-D-12-00121.1.
Hwang, Y.-T., H.-Y. Tseng, K.-C. Li, S. M. Kang, Y.-J. Chen, and J. C. Chiang, 2021: Relative roles of energy and momentum fluxes in the tropical response to extratropical thermal forcing. J. Climate, 34, 3771–3786, https://doi.org/10.1175/JCLI-D-20-0151.1.
Jackson, L., R. Kahana, T. Graham, M. Ringer, T. Woollings, J. Mecking, and R. Wood, 2015: Global and European climate impacts of a slowdown of the AMOC in a high resolution GCM. Climate Dyn., 45, 3299–3316, https://doi.org/10.1007/s00382-015-2540-2.
Jin, F.-F., 1997a: An equatorial ocean recharge paradigm for ENSO. Part I: Conceptual model. J. Atmos. Sci., 54, 811–829, https://doi.org/10.1175/1520-0469(1997)054<0811:AEORPF>2.0.CO;2.
Jin, F.-F., 1997b: An equatorial ocean recharge paradigm for ENSO. Part II: A stripped-down coupled model. J. Atmos. Sci., 54, 830–847, https://doi.org/10.1175/1520-0469(1997)054<0830:AEORPF>2.0.CO;2.
Kang, S. M., Y. Shin, and S.-P. Xie, 2018: Extratropical forcing and tropical rainfall distribution: Energetics framework and ocean Ekman advection. npj Climate Atmos. Sci., 1, 20172, https://doi.org/10.1038/s41612-017-0004-6.
Kessler, W. S., 1990: Observations of long Rossby waves in the northern tropical Pacific. J. Geophys. Res., 95, 5183–5217, https://doi.org/10.1029/JC095iC04p05183.
Kim, D., S.-K. Lee, H. Lopez, and M. Goes, 2020: Pacific mean-state control of Atlantic multidecadal oscillation–El Niño relationship. J. Climate, 33, 4273–4291, https://doi.org/10.1175/JCLI-D-19-0398.1.
Kousky, V., and R. Higgins, 2007: An alert classification system for monitoring and assessing the ENSO cycle. Wea. Forecasting, 22, 353–371, https://doi.org/10.1175/WAF987.1.
Lau, N.-C., and M. J. Nath, 2003: Atmosphere–ocean variations in the Indo-Pacific sector during ENSO episodes. J. Climate, 16, 3–20, https://doi.org/10.1175/1520-0442(2003)016<0003:AOVITI>2.0.CO;2.
Lawrence, D. M., and Coauthors, 2011: Parameterization improvements and functional and structural advances in version 4 of the Community Land Model. J. Adv. Model. Earth Syst., 3, M03001, https://doi.org/10.1029/2011MS00045.
Levine, A. F., M. J. McPhaden, and D. M. Frierson, 2017: The impact of the AMO on multidecadal ENSO variability. Geophys. Res. Lett., 44, 3877–3886, https://doi.org/10.1002/2017GL072524.
L’Heureux, M. L., M. K. Tippett, and A. G. Barnston, 2015: Characterizing ENSO coupled variability and its impact on North American seasonal precipitation and temperature. J. Climate, 28, 4231–4245, https://doi.org/10.1175/JCLI-D-14-00508.1.
Lin, I.-I., and Coauthors, 2020: ENSO and tropical cyclones. El Niño Southern Oscillation in a Changing Climate, Geophys. Monogr., Vol. 253, Amer. Geophys. Union, 377–408, https://doi.org/10.1002/9781119548164.ch17.
Liu, W., and Z. Liu, 2014: Assessing the stability of the Atlantic meridional overturning circulation of the past, present, and future. J. Meteor. Res., 28, 803–819, https://doi.org/10.1007/s13351-014-4006-6.
Liu, W., and A. Hu, 2015: The role of the PMOC in modulating the deglacial shift of the ITCZ. Climate Dyn., 45, 3019–3034, https://doi.org/10.1007/s00382-015-2520-6.
Liu, W., A. V. Fedorov, S.-P. Xie, and S. Hu, 2020: Climate impacts of a weakened Atlantic Meridional Overturning Circulation in a warming climate. Sci. Adv., 6, eaaz4876, https://doi.org/10.1126/sciadv.aaz4876.
McGee, D., A. Donohoe, J. Marshall, and D. Ferreira, 2014: Changes in ITCZ location and cross-equatorial heat transport at the Last Glacial Maximum, Heinrich Stadial 1, and the mid-Holocene. Earth Planet. Sci. Lett., 390, 69–79, https://doi.org/10.1016/j.epsl.2013.12.043.
McGraw, M. C., and E. A. Barnes, 2018: Memory matters: A case for Granger causality in climate variability studies. J. Climate, 31, 3289–3300, https://doi.org/10.1175/JCLI-D-17-0334.1.
McGregor, S., A. Timmermann, N. Schneider, M. F. Stuecker, and M. H. England, 2012: The effect of the South Pacific convergence zone on the termination of El Niño events and the meridional asymmetry of ENSO. J. Climate, 25, 5566–5586, https://doi.org/10.1175/JCLI-D-11-00332.1.
McPhaden, M. J., S. E. Zebiak, and M. H. Glantz, 2006: ENSO as an integrating concept in Earth science. Science, 314, 1740–1745, https://doi.org/10.1126/science.1132588.
Meehl, G. A., and Coauthors, 2021: Atlantic and Pacific tropics connected by mutually interactive decadal-timescale processes. Nat. Geosci., 14, 36–42, https://doi.org/10.1038/s41561-020-00669-x.
Meinen, C. S., and M. J. McPhaden, 2000: Observations of warm water volume changes in the equatorial Pacific and their relationship to El Niño and La Niña. J. Climate, 13, 3551–3559, https://doi.org/10.1175/1520-0442(2000)013<3551:OOWWVC>2.0.CO;2.
Molina, M. J., J. T. Allen, and V. A. Gensini, 2018: The Gulf of Mexico and ENSO influence on subseasonal and seasonal CONUS winter tornado variability. J. Appl. Meteor. Climatol., 57, 2439–2463, https://doi.org/10.1175/JAMC-D-18-0046.1.
Moreno-Chamarro, E., J. Marshall, and T. Delworth, 2020: Linking ITCZ migrations to the AMOC and North Atlantic/Pacific SST decadal variability. J. Climate, 33, 893–905, https://doi.org/10.1175/JCLI-D-19-0258.1.
NCEI, 2007: NOAA Optimum Interpolation 1/4 Degree Daily Sea Surface Temperature Analysis. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory, accessed 18 June 2021, https://doi.org/10.5065/EM0T-1D34.
Neale, R. B., and Coauthors, 2010: Description of the NCAR Community Atmosphere Model (CAM5.0). NCAR Tech. Note NCAR/TN-486+STR, 268 pp., www.cesm.ucar.edu/models/cesm1.1/cam/docs/description/cam5_desc.pdf.
Neelin, J. D., F.-F. Jin, and H.-H. Syu, 2000: Variations in ENSO phase locking. J. Climate, 13, 2570–2590, https://doi.org/10.1175/1520-0442(2000)013<2570:VIEPL>2.0.CO;2.
Okazaki, Y., A. Timmermann, L. Menviel, N. Harada, A. Abe-Ouchi, M. Chikamoto, A. Mouchet, and H. Asahi, 2010: Deepwater formation in the North Pacific during the last glacial termination. Science, 329, 200–204, https://doi.org/10.1126/science.1190612.
Parker, A., and C. Ollier, 2016: There is no real evidence for a diminishing trend of the Atlantic meridional overturning circulation. J. Ocean Eng. Sci., 1, 30–35, https://doi.org/10.1016/j.joes.2015.12.007.
Polo, I., J. Robson, R. Sutton, and M. A. Balmaseda, 2014: The importance of wind and buoyancy forcing for the boundary density variations and the geostrophic component of the AMOC at 26°N. J. Phys. Oceanogr., 44, 2387–2408, https://doi.org/10.1175/JPO-D-13-0264.1.
Power, S., T. Casey, C. Folland, A. Colman, and V. Mehta, 1999: Inter-decadal modulation of the impact of ENSO on Australia. Climate Dyn., 15, 319–324, https://doi.org/10.1007/s003820050284.
Rae, J. W., and Coauthors, 2020: Overturning circulation, nutrient limitation, and warming in the Glacial North Pacific. Sci. Adv., 6, eabd1654, https://doi.org/10.1126/sciadv.abd1654.
Rahmstorf, S., J. E. Box, G. Feulner, M. E. Mann, A. Robinson, S. Rutherford, and E. J. Schaffernicht, 2015: Exceptional twentieth-century slowdown in Atlantic Ocean overturning circulation. Nat. Climate Change, 5, 475–480, https://doi.org/10.1038/nclimate2554.
Rashid, H. A., S. B. Power, and J. R. Knight, 2010: Impact of multidecadal fluctuations in the Atlantic thermohaline circulation on Indo-Pacific climate variability in a coupled GCM. J. Climate, 23, 4038–4044, https://doi.org/10.1175/2010JCLI3430.1.
Richter, I., S.-P. Xie, Y. Morioka, T. Doi, B. Taguchi, and S. Behera, 2017: Phase locking of equatorial Atlantic variability through the seasonal migration of the ITCZ. Climate Dyn., 48, 3615–3629, https://doi.org/10.1007/s00382-016-3289-y.
Ruprich-Robert, Y., and C. Cassou, 2015: Combined influences of seasonal East Atlantic Pattern and North Atlantic Oscillation to excite Atlantic multidecadal variability in a climate model. Climate Dyn., 44, 229–253, https://doi.org/10.1007/s00382-014-2176-7.
Russell, A. M., and A. Gnanadesikan, 2014: Understanding multidecadal variability in ENSO amplitude. J. Climate, 27, 4037–4051, https://doi.org/10.1175/JCLI-D-13-00147.1.
Rustic, G. T., P. J. Polissar, A. C. Ravelo, and S. M. White, 2020: Modulation of late Pleistocene ENSO strength by the tropical Pacific thermocline. Nat. Commun., 11, 5377, https://doi.org/10.1038/s41467-020-19161-6.
Saenko, O. A., A. Schmittner, and A. J. Weaver, 2004: The Atlantic–Pacific seesaw. J. Climate, 17, 2033–2038, https://doi.org/10.1175/1520-0442(2004)017<2033:TAS>2.0.CO;2.
Schneider, T., T. Bischoff, and G. H. Haug, 2014: Migrations and dynamics of the intertropical convergence zone. Nature, 513, 45–53, https://doi.org/10.1038/nature13636.
Schott, F. A., J. P. Mccreary, and G. C. Johnson, 2004: Shallow overturning circulations of the tropical-subtropical oceans. Earth’s Climate: The Ocean–Atmosphere Interaction, Geophys. Monogr., Vol. 147, Amer. Geophys. Union, 261–304, https://doi.org/10.1029/147GM15.
Seager, R., D. S. Battisti, J. Yin, N. Gordon, N. Naik, A. C. Clement, and M. A. Cane, 2002: Is the Gulf Stream responsible for Europe’s mild winters? Quart. J. Roy. Meteor. Soc., 128, 2563–2586, https://doi.org/10.1256/qj.01.128.
Seidov, D., and B. J. Haupt, 2003: Freshwater teleconnections and ocean thermohaline circulation. Geophys. Res. Lett., 30, 1329, https://doi.org/10.1029/2002GL016564.
Sigmond, M., J. C. Fyfe, O. A. Saenko, and N. C. Swart, 2020: Ongoing AMOC and related sea-level and temperature changes after achieving the Paris targets. Nat. Climate Change, 10, 672–677, https://doi.org/10.1038/s41558-020-0786-0.
Smeed, D., G. McCarthy, D. Rayner, B. Moat, W. Johns, M. Baringer, and C. Meinen, 2016: Atlantic meridional overturning circulation observed by the RAPID-MOCHA-WBTS (RAPID-Meridional Overturning Circulation and Heatflux Array-Western Boundary Time Series) array at 26N from 2004 to 2015. British Oceanographic Data Centre, Natural Environment Research Council, accessed 1 January 2021, https://doi.org/10.5285/35784047-9b82-2160-e053-6c86abc0c91b.
Stein, K., A. Timmermann, N. Schneider, F.-F. Jin, and M. F. Stuecker, 2014: ENSO seasonal synchronization theory. J. Climate, 27, 5285–5310, https://doi.org/10.1175/JCLI-D-13-00525.1.
Stocker, T. F., and W. S. Broecker, 1994: Observation and modeling of North Atlantic deep water formation and its variability: Introduction. J. Geophys. Res., 99, 12 317–12 317, https://doi.org/10.1029/94JC00956.
Stuecker, M. F., F.-F. Jin, and A. Timmermann, 2015: El Niño-Southern Oscillation frequency cascade. Proc. Natl. Acad. Sci. USA, 112, 13 490–13 495, https://doi.org/10.1073/pnas.1508622112.
Sun, Y., F. Wang, and D.-Z. Sun, 2016: Weak ENSO asymmetry due to weak nonlinear air–sea interaction in CMIP5 climate models. Adv. Atmos. Sci., 33, 352–364, https://doi.org/10.1007/s00376-015-5018-6.
Sutton, R. T., and D. L. Hodson, 2005: Atlantic Ocean forcing of North American and European summer climate. Science, 309, 115–118, https://doi.org/10.1126/science.1109496.
Timmermann, A., S. An, U. Krebs, and H. Goosse, 2005: ENSO suppression due to weakening of the North Atlantic thermohaline circulation. J. Climate, 18, 3122–3139, https://doi.org/10.1175/JCLI3495.1.
Timmermann, A., and Coauthors, 2007: The influence of a weakening of the Atlantic meridional overturning circulation on ENSO. J. Climate, 20, 4899–4919, https://doi.org/10.1175/JCLI4283.1.
Torrence, C., and G. P. Compo, 1998: A practical guide to wavelet analysis. Bull. Amer. Meteor. Soc., 79, 61–78, https://doi.org/10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2.
Trenberth, K. E., and J. T. Fasullo, 2017: Atlantic meridional heat transports computed from balancing Earth’s energy locally. Geophys. Res. Lett., 44, 1919–1927, https://doi.org/10.1002/2016GL072475.
Trenberth, K. E., Y. Zhang, J. T. Fasullo, and L. Cheng, 2019: Observation-based estimates of global and basin ocean meridional heat transport time series. J. Climate, 32, 4567–4583, https://doi.org/10.1175/JCLI-D-18-0872.1.
Tziperman, E., S. E. Zebiak, and M. A. Cane, 1997: Mechanisms of seasonal–ENSO interaction. J. Atmos. Sci., 54, 61–71, https://doi.org/10.1175/1520-0469(1997)054<0061:MOSEI>2.0.CO;2.
van Oldenborgh, G. J., L. A. te Raa, H. A. Dijkstra, and S. Y. Philip, 2009: Frequency-or amplitude-dependent effects of the Atlantic meridional overturning on the tropical Pacific Ocean. Ocean Sci., 5, 293–301, https://doi.org/10.5194/os-5-293-2009.
Waldman, R., J. Hirschi, A. Voldoire, C. Cassou, and R. Msadek, 2021: Clarifying the relation between AMOC and thermal wind: Application to the centennial variability in a coupled climate model. J. Phys. Oceanogr., 51, 343–364, https://doi.org/10.1175/JPO-D-19-0284.1.
Wang, B., and Y. Wang, 1999: Dynamics of the ITCZ–equatorial cold tongue complex and causes of the Latitudinal climate asymmetry. J. Climate, 12, 1830–1847, https://doi.org/10.1175/1520-0442(1999)012<1830:DOTIEC>2.0.CO;2.
Weaver, A. J., C. M. Bitz, A. F. Fanning, and M. Holland, 1999: Thermohaline circulation: High-latitude phenomena and the difference between the Pacific and Atlantic. Annu. Rev. Earth Planet. Sci., 27, 231–285, https://doi.org/10.1146/annurev.earth.27.1.231.
Weaver, A. J., and Coauthors, 2012: Stability of the Atlantic meridional overturning circulation: A model intercomparison. Geophys. Res. Lett., 39, L20709, https://doi.org/10.1029/2012GL053763.
Weijer, W., and Coauthors, 2019: Stability of the Atlantic Meridional Overturning Circulation: A review and synthesis. J. Geophys. Res. Oceans, 124, 5336–5375, https://doi.org/10.1029/2019JC015083.
Wen, C., A. Kumar, Y. Xue, and M. McPhaden, 2014: Changes in tropical Pacific thermocline depth and their relationship to ENSO after 1999. J. Climate, 27, 7230–7249, https://doi.org/10.1175/JCLI-D-13-00518.1.
Williamson, M. S., M. Collins, S. S. Drijfhout, R. Kahana, J. V. Mecking, and T. M. Lenton, 2018: Effect of AMOC collapse on ENSO in a high resolution general circulation model. Climate Dyn., 50, 2537–2552, https://doi.org/10.1007/s00382-017-3756-0.
Willis, J. K., 2010: Can in situ floats and satellite altimeters detect long-term changes in Atlantic Ocean overturning? Geophys. Res. Lett., 37, L06602, https://doi.org/10.1029/2010GL042372.
Woodgate, R. A., and K. Aagaard, 2005: Revising the Bering Strait freshwater flux into the Arctic Ocean. Geophys. Res. Lett., 32, L02602, https://doi.org/10.1029/2004GL021747.
Yu, H., and Coauthors, 2015: The fertilizing role of African dust in the Amazon rainforest: A first multiyear assessment based on data from Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations. Geophys. Res. Lett., 42, 1984–1991, https://doi.org/10.1002/2015GL063040.
Yu, S., and M. S. Pritchard, 2019: A strong role for the AMOC in partitioning global energy transport and shifting ITCZ position in response to latitudinally discrete solar forcing in CESM1.2. J. Climate, 32, 2207–2226, https://doi.org/10.1175/JCLI-D-18-0360.1.
Zebiak, S. E., and M. A. Cane, 1987: A model El Niño–Southern Oscillation. Mon. Wea. Rev., 115, 2262–2278, https://doi.org/10.1175/1520-0493(1987)115<2262:AMENO>2.0.CO;2.
Zeller, M., S. McGregor, and P. Spence, 2019: Hemispheric asymmetry of the Pacific shallow meridional overturning circulation. J. Geophys. Res. Oceans, 124, 5765–5786, https://doi.org/10.1029/2018JC014840.
Zhang, J., and Coauthors, 2021: On the connection between AMOC and observed land precipitation in Northern Hemisphere: A comparison of the AMOC indicators. Climate Dyn., 56, 651–664, https://doi.org/10.1007/s00382-020-05496-9.
Zhang, R., and T. L. Delworth, 2005: Simulated tropical response to a substantial weakening of the Atlantic thermohaline circulation. J. Climate, 18, 1853–1860, https://doi.org/10.1175/JCLI3460.1.
Zhang, R., R. Sutton, G. Danabasoglu, Y.-O. Kwon, R. Marsh, S. G. Yeager, D. E. Amrhein, and C. M. Little, 2019: A review of the role of the Atlantic meridional overturning circulation in Atlantic multidecadal variability and associated climate impacts. Rev. Geophys., 57, 316–375, https://doi.org/10.1029/2019RG000644.
Zhang, T., and D.-Z. Sun, 2014: ENSO asymmetry in CMIP5 models. J. Climate, 27, 4070–4093, https://doi.org/10.1175/JCLI-D-13-00454.1.
Zhang, T., X. Shao, and S. Li, 2017: Impacts of atmospheric processes on ENSO asymmetry: A comparison between CESM1 and CCSM4. J. Climate, 30, 9743–9762, https://doi.org/10.1175/JCLI-D-17-0360.1.
Computed as 3-month means, such as December–February (DJF) and June–August (JJA).
