Southern Ocean Temperature Research Articles

Shifting Depth Distributions of Deep‐Sea Corals in the Southwest Pacific: Implications for Deglacial Dynamics of the Southern Ocean

AbstractWe compare depth and temporal distributions of sub‐fossil assemblages of two cold‐water scleractinian corals on seamounts in the Southwest Pacific to help define the temporal variations of water mass properties in the Southern Ocean (SO) during deglaciation. Peaks in the deep‐water abundance of the two species complement one another, with Desmophyllum dianthus peaking around the Antarctic Cold Reversal (ACR), and Solenosmilia variabilis briefly during the late Heinrich Stadial 1 (HS1) and during the Younger Dryas (YD). Environmental tolerances of the two species and the geochemistry of S. variabilis carbonate skeletons suggest that their secular distributions reflect complementary effects of temperature (higher at Antarctic Intermediate Water/Upper Circumpolar Deep Water depths during the YD and late HS1) and surface productivity (lower during the YD and HS1). Higher temperatures at depth we interpret as evidence of increased Zonal West Wind (ZWW)‐driven Ekman pumping during the late HS1 and YD, whereas coeval low surface production reflects poleward expansion of sub‐tropical water masses as a result of correlated poleward shifts of the ZWW belt and the Intertropical Convergence Zone. More broadly, a continuous deep coral population in the southwest Pacific that spans two species and three deglacial periods (HS1, ACR and the YD) and an early Holocene shift in coral distribution from deeper to shallower habitats appear to reflect large‐scale changes during deglaciation in SO temperature profiles and productivity.

Impacts of Stratospheric Ozone Recovery on Southern Ocean Temperature and Heat Budget

AbstractThe impacts of stratospheric ozone recovery on Southern Ocean surface and interior temperature, heat content, heat uptake, and heat transport are investigated by contrasting two ensemble chemistry‐climate model simulations in 2005–2099: one with fixed ozone depleting substances (ODSs) and another with decreasing ODSs. In our simulations ozone recovery significantly affects Southern Ocean temperature, with large latitudinal and vertical variations. Ozone recovery causes a dipole change of the full‐depth ocean heat content (OHC) with an increase south of 60°S and a decrease between 45°S and 60°S. Integrated over latitudes south of 40°S, OHC decreases in response to ozone recovery. This ocean heat loss is shown to be driven by weakened poleward ocean heat transport (OHT) across 40°S, which is partly canceled by enhanced heat uptake. The weakening of poleward OHT into the Southern Ocean is caused by the ozone‐induced equatorward shift of the meridional overturning circulation.

Roles of Surface Forcing in the Southern Ocean Temperature and Salinity Changes under Increasing CO2: Perspectives from Model Perturbation Experiments and a Theoretical Framework

Abstract A rapid warming and freshening of the Southern Ocean have been observed over the past several decades and are attributed to anthropogenic climate change. In this study, ocean model perturbation experiments are conducted to separate roles of individual surface forcing in the Southern Ocean temperature and salinity changes. Model-based findings are compared with results from a theoretical framework including three idealized processes defined on the θ–S diagram. Under the future scenario of CO2 doubling, the heat flux forcing dominates the large-scale warming, deepening of isopycnals, and spiciness changes along isopycnals, which can be captured by an idealized pure warming process to represent the subduction of surface heat uptake. The poleward-intensifying westerly winds account for 24% of the enhanced warming between 35° and 50°S and would have comparable contribution as the heat flux forcing after removing the global ocean warming effect. In contrast, the widespread freshening in the Southern Ocean driven by increased surface freshwater input is largely compensated by the wind-driven saltening. The response to freshwater forcing could not be approximated as a similar pure freshening process as the induced cooling and freshening have comparable effects on density. The wind-driven changes are primarily through the local heave of isopycnals, thus resembling an idealized pure heave process, but contain considerable spiciness signals especially in the midlatitude Southern Ocean, resulting from anomalous northward transport and subduction of heat and salt that are largely density-compensating. These distinct signatures of individual surface forcing help us to better understand observed and projected changes in the Southern Ocean. Significance Statement Considerable changes including a rapid warming and freshening have been observed in the Southern Ocean as it absorbs most of the extra heat from the anthropogenic climate change, receives increased surface freshwater input, and experiences a poleward shift and intensification of the westerly winds. The purpose of this study is to distinguish different contributions from surface heat flux, freshwater flux, and wind forcing to the Southern Ocean temperature and salinity changes, based on ocean model experiments and three idealized processes from a theoretical framework. Our study reveals distinct signatures of individual surface forcing that help us to understand linkages between changes seen at the surface and in the interior Southern Ocean.

Subglacial precipitates record Antarctic ice sheet response to late Pleistocene millennial climate cycles

Ice cores and offshore sedimentary records demonstrate enhanced ice loss along Antarctic coastal margins during millennial-scale warm intervals within the last glacial termination. However, the distal location and short temporal coverage of these records leads to uncertainty in both the spatial footprint of ice loss, and whether millennial-scale ice response occurs outside of glacial terminations. Here we present a >100kyr archive of periodic transitions in subglacial precipitate mineralogy that are synchronous with Late Pleistocene millennial-scale climate cycles. Geochemical and geochronologic data provide evidence for opal formation during cold periods via cryoconcentration of subglacial brine, and calcite formation during warm periods through the addition of subglacial meltwater originating from the ice sheet interior. These freeze-flush cycles represent cyclic changes in subglacial hydrologic-connectivity driven by ice sheet velocity fluctuations. Our findings imply that oscillating Southern Ocean temperatures drive a dynamic response in the Antarctic ice sheet on millennial timescales, regardless of the background climate state.

Wilkes subglacial basin ice sheet response to Southern Ocean warming during late Pleistocene interglacials

The response of the East Antarctic Ice Sheet to past intervals of oceanic and atmospheric warming is still not well constrained but is critical for understanding both past and future sea-level change. Furthermore, the ice sheet in the Wilkes Subglacial Basin appears to have undergone thinning and ice discharge events during recent decades. Here we combine glaciological evidence on ice sheet elevation from the TALDICE ice core with offshore sedimentological records and ice sheet modelling experiments to reconstruct the ice dynamics in the Wilkes Subglacial Basin over the past 350,000 years. Our results indicate that the Wilkes Subglacial Basin experienced an extensive retreat 330,000 years ago and a more limited retreat 125,000 years ago. These changes coincide with warmer Southern Ocean temperatures and elevated global mean sea level during those interglacial periods, confirming the sensitivity of the Wilkes Subglacial Basin ice sheet to ocean warming and its potential role in sea-level change.

Compiled Southern Ocean sea surface temperatures correlate with Antarctic Isotope Maxima

The magnitude and spatial variability of millennial-scale changes in Southern Ocean temperature during Marine Isotope Stage 3 (MIS-3) are poorly constrained. Here we present a compilation of 14 previously published high-resolution sea surface temperature records from 30°S to 70°S. At each site we re-calibrate radiocarbon dates and synchronise records with the Antarctic Ice Core Chronology 2012 and then determine the presence, amplitude and duration of millennial-scale warming that correspond to Antarctic Isotope Maximum (AIM) events. Individual sediment cores recorded warming during an average of 7 out of 10 AIM events. These warming events were then used as tie-points to refine the age models at each site before combining records into Southern Ocean and basin-wide averaged temperature anomaly records or “stacks”. The resulting stacks of Southern Ocean, Atlantic and Indo-Pacific Ocean basin temperature anomalies show a signal consistent with AIM events during MIS-3. Stacked amplitudes of warming are also positively correlated with the amplitude of temperature increases in the EPICA Dome C ice core (n = 10, r2 = 0.65, p = 0.001). Additionally, rates of warming in the Southern Ocean are consistent across all warming events. These findings are consistent with the thermal seesaw hypothesis, where a reduction in Atlantic Meridional Overturning Circulation causes decreased cross-equatorial heat transport in the Atlantic resulting in heat accumulation in the south. Our results solidify the link between northern high latitude climate variability, Antarctic temperature and Southern Ocean surface temperature variability during MIS-3. This compilation of records with updated age models tied to the same ice core provides the first basin-scale synthesis of the sea surface temperature changes in the Southern Ocean during the rapid climate changes of MIS-3.

Anthropogenic Temperature and Salinity Changes in the Southern Ocean

AbstractIn this study, we compare observed Southern Ocean temperature and salinity changes with the historical simulations from 13 models from phase 5 of the Coupled Model Intercomparison Project (CMIP5), using an optimal fingerprinting framework. We show that there is an unequivocal greenhouse gas–forced warming in the Southern Ocean. This warming is strongest in the Subantarctic Mode Waters but is also detectable in denser water masses, which has not been shown in previous studies. We also find greenhouse gas–forced salinity changes, most notably a freshening of Antarctic Intermediate Waters. Our analysis also shows that non–greenhouse gas anthropogenic forcings—anthropogenic aerosols and stratospheric ozone depletion—have played an important role in mitigating the Southern Ocean’s warming. However, the detectability of these responses using optimal fingerprinting is model dependent, and this result is therefore not as robust as for the greenhouse gas response.

Coupled Southern Ocean cooling and Antarctic ice sheet expansion during the middle Miocene

The middle Miocene climate transition (~14 million years ago) was characterized by a dramatic increase in the volume of the Antarctic ice sheet. The driving mechanism of this transition remains under discussion, with hypotheses including circulation changes, declining carbon dioxide in the atmosphere and orbital forcing. Southern Ocean records of planktic foraminiferal Mg/Ca have previously been interpreted to indicate a cooling of 6–7 °C and a decrease in salinity that preceded Antarctic cryosphere expansion by up to ~300,000 years. This interpretation has led to the hypothesis that changes in meridional heat and vapour transport along with an early thermal isolation of Antarctica from extrapolar climates played a fundamental role in triggering ice growth. Here we revisit the middle Miocene Southern Ocean temperature evolution using clumped isotope and lipid biomarker temperature proxies. Our records indicate that the Southern Ocean cooling and the associated salinity decrease occurred in phase with the expansion of the Antarctic ice sheet. We demonstrate that the timing and magnitude of the Southern Ocean temperature change seen in previous reconstructions can be explained if we consider pH as an additional, non-thermal, control on foraminiferal Mg/Ca ratios. Therefore, our new dataset challenges the view of a thermal isolation of Antarctica preceding ice sheet expansion, and suggests a strong coupling between Southern Ocean conditions and Antarctic ice volume in times of declining atmospheric carbon dioxide. Antarctic ice volume expansion in the middle Miocene coincides with Southern Ocean cooling, according to biomarker and clumped isotope temperature records from south of Tasmania.

Sea surface temperature in the Indian sector of the Southern Ocean over the Late Glacial and Holocene

Abstract. Centennial- and millennial-scale variability of Southern Ocean temperature over the Holocene is poorly known, due to both short instrumental records and sparsely distributed high-resolution temperature reconstructions, with evidence for past temperature variations in the region coming mainly from ice core records. Here we present a high-resolution (∼60 year), diatom-based sea surface temperature (SST) reconstruction from the western Indian sector of the Southern Ocean that spans the interval 14.2 to 1.0 ka (calibrated kiloyears before present). During the late deglaciation, the new SST record shows cool temperatures at 14.2–12.9 ka and gradual warming between 12.9 and 11.6 ka in phase with atmospheric temperature evolution. This supports the evolution of the Southern Ocean SST during the deglaciation being linked with a complex combination of processes and drivers associated with reorganisations of atmospheric and oceanic circulation patterns. Specifically, we suggest that Southern Ocean surface warming coincided, within the dating uncertainties, with the reconstructed slowdown of the Atlantic Meridional Overturning Circulation (AMOC), rising atmospheric CO2 levels, changes in the southern westerly winds and enhanced upwelling. During the Holocene the record shows warm and stable temperatures from 11.6 to 8.7 ka followed by a slight cooling and greater variability from 8.7 to 1 ka, with a quasi-periodic variability of 200–260 years identified by spectral analysis. We suggest that the increased variability during the mid- to late Holocene reflects the establishment of centennial variability in SST connected with changes in the high-latitude atmospheric circulation and Southern Ocean convection.

Changes in positive associations among vertebrate predators at South Georgia during winter

We studied positive associations among seabirds and marine mammals at South Georgia on research cruises during the Austral winters of 1985, 1991 and 1993 and found statistically significant differences. We collected data on abundance and distribution, providing a critical reference for sub-Antarctic conservation in anticipation of future environmental changes. We found significant changes in the abundance of 29% of species surveyed and a consequent change in species diversity. We postulate that the resulting altered community composition may have previously unanticipated population effects on the component species, due to changes in positive interactions among species which use each other as cues to the presence of prey. We found a near threefold reduction in spatial overlap among vertebrate predators, associated with warming sea temperatures. As the strength and opportunity for positive associations decreases in the future, feeding success may be negatively impacted. In this way, environmental changes may disproportionately impact predator abundances and such changes are likely already underway, as Southern Ocean temperatures have increased substantially since our surveys. Of course the changes we describe are not solely due to changing sea temperature or any other single cause—many factors are important and we do not claim to have removed these from consideration. Rather, we report previously undocumented changes in positive associations among species, and argue these changes may continue into the future, given near-certain continued increases in climate-related changes.

Last Abundant Appearance Datum of Hemidiscus karstenii driven by climate change

Abstract The Last Abundant Appearance Datum of the endemic Southern Ocean diatom Hemisdiscus karstenii Jouse is dated at around 190,000 years Before Present, which provides a robust and convenient biostratigraphic marker to Late Pleistocene paleoceanographic studies. However, no prior studies have sought to understand what processes may have driven H. karstenii to extinction. We here present the first morphometric investigation of H. karstenii over Marine Isotope Stage (MIS) 10 to 6 (355,000 years to 150,000 years). The combination of these data to records of H. karstenii absolute abundances, Fragilariopsis kerguelensis biometry and absolute abundances, and southern high-latitude air and ocean temperatures over the past 355,000 years suggests that H. karstenii populations reduced their mean size and productivity in response to the rapid drop in temperatures across the MIS 7 to MIS 6 glacial inception. This prolonged cooling event may have pushed H. karstenii to its adaptive limit until it could not sustain physiological viability. The repetitive occurrence of very cold events during the early MIS 6 may have additionally prevented surviving populations of H. karstenii from re-establishing abundant communities leading to extinction late in MIS 6.

Importance of wind and meltwater for observed chemical and physical changes in the Southern Ocean

The Southern Ocean south of 30° S represents only one-third of the total ocean area, yet absorbs half of the total ocean anthropogenic carbon and over two-thirds of ocean anthropogenic heat. In the past, the Southern Ocean has also been one of the most sparsely measured regions of the global ocean. Here we use pre-2005 ocean shipboard measurements alongside novel observations from autonomous floats with biogeochemical sensors to calculate changes in Southern Ocean temperature, salinity, pH and concentrations of nitrate, dissolved inorganic carbon and oxygen over two decades. We find local warming of over 3 °C, salinification of over 0.2 psu near the Antarctic coast, and isopycnals are found to deepen between 65° and 40° S. We find deoxygenation along the Antarctic coast, but reduced deoxygenation and nitrate concentrations where isopycnals deepen farther north. The forced response of the Earth system model ESM2M does not reproduce the observed patterns. Accounting for meltwater and poleward-intensifying winds in ESM2M improves reproduction of the observed large-scale changes, demonstrating the importance of recent changes in wind and meltwater. Future Southern Ocean biogeochemical changes are likely to be influenced by the relative strength of meltwater input and poleward-intensifying winds. The combined effect could lead to increased Southern Ocean deoxygenation and nutrient accumulation, starving the global ocean of nutrients sooner than otherwise expected. Physical and biogeochemical changes in the Southern Ocean over the past decade are largely due to growing meltwater input and intensifying poleward winds, according to observations from ships and floats and model simulations.

Southern Ocean temperature records and ice-sheet models demonstrate rapid Antarctic ice sheet retreat under low atmospheric CO2 during Marine Isotope Stage 31

Over the last 5 million years, the Earth’s climate has oscillated between warm (interglacial) and cold (glacial) states. Some particularly warm interglacial periods (i.e. ‘super-interglacials’) occurred under low atmospheric CO2 and may have featured extensive Antarctic ice sheet collapse. Here we focus on an extreme super-interglacial known as Marine Isotope Stage 31 (MIS31), between 1.085 and 1.055 million years ago and is the subject of intense discussion. We reconstructed the first Southern Ocean and Antarctic margin sea surface temperatures (SSTs) from organic biomarkers and used them to constrain numerical ice sheet-shelf simulations. Our SSTs indicate that the ocean was on average 5 °C (±1.2 °C) warmer in summer than today between 50 °S and the Antarctic ice margin. Our most conservative ice sheet simulation indicates a complete collapse of the West Antarctic Ice Sheet (WAIS) with additional deflation of the East Antarctic Ice Sheet. We suggest the WAIS retreated because of anomalously high Southern Hemisphere insolation coupled with the intrusion of Circumpolar Deep Water onto the continental shelf under poleward-intensified winds leading to a shorter sea ice season and ocean warming at the continental margin. In this scenario, the extreme warming we observed likely reflects the extensively modified oceanic and hydrologic system following ice sheet collapse. Our work highlights the sensitivity of the Antarctic ice sheets to minor oceanic perturbations that could also be at play for future changes.

Simulating the 128‐ka Antarctic Climate Response to Northern Hemisphere Ice Sheet Melting Using the Isotope‐Enabled HadCM3

AbstractWarmer than present Antarctic and Southern Ocean temperatures during the last interglacial, approximately 128,000 years ago, have been attributed to changes in north‐south ocean heat transport, causing opposing hemispheric temperature anomalies. We investigate the magnitude of Antarctic warming and Antarctic ice core isotopic enrichment in response to Northern Hemisphere meltwater input during the early last interglacial. A 1,600‐year HadCM3 simulation driven by 0.25 Sv of meltwater input reproduces 50–60% of the peak Southern Ocean summer sea surface temperature anomaly, sea ice retreat, and ice core isotope enrichment. We also find a robust increase in the proportion of cold season precipitation during the last interglacial, leading to lower isotopic values at the Antarctic ice core sites. These results suggest that a HadCM3 simulation including 0.25 Sv for 3,000–4,000 years would reconcile the last interglacial observations, providing a potential solution for the last interglacial missing heat problem.

Asynchrony between Antarctic temperature and CO2 associated with obliquity over the past 720,000 years

The δD temperature proxy in Antarctic ice cores varies in parallel with CO2 through glacial cycles. However, these variables display a puzzling asynchrony. Well-dated records of Southern Ocean temperature will provide crucial information because the Southern Ocean is likely key in regulating CO2 variations. Here, we perform multiple isotopic analyses on an Antarctic ice core and estimate temperature variations at this site and in the oceanic moisture source over the past 720,000 years, which extend the longest records by 300,000 years. Antarctic temperature is affected by large variations in local insolation that are induced by obliquity. At the obliquity periodicity, the Antarctic and ocean temperatures lag annual mean insolation. Further, the magnitude of the phase lag is minimal during low eccentricity periods, suggesting that secular changes in the global carbon cycle and the ocean circulation modulate the phase relationship among temperatures, CO2 and insolation in the obliquity frequency band.

Asynchronous warming and δ18O evolution of deep Atlantic water masses during the last deglaciation

The large-scale reorganization of deep ocean circulation in the Atlantic involving changes in North Atlantic Deep Water (NADW) and Antarctic Bottom Water (AABW) played a critical role in regulating hemispheric and global climate during the last deglaciation. However, changes in the relative contributions of NADW and AABW and their properties are poorly constrained by marine records, including δ18O of benthic foraminiferal calcite (δ18Oc). Here, we use an isotope-enabled ocean general circulation model with realistic geometry and forcing conditions to simulate the deglacial water mass and δ18O evolution. Model results suggest that, in response to North Atlantic freshwater forcing during the early phase of the last deglaciation, NADW nearly collapses, while AABW mildly weakens. Rather than reflecting changes in NADW or AABW properties caused by freshwater input as suggested previously, the observed phasing difference of deep δ18Oc likely reflects early warming of the deep northern North Atlantic by ∼1.4 °C, while deep Southern Ocean temperature remains largely unchanged. We propose a thermodynamic mechanism to explain the early warming in the North Atlantic, featuring a strong middepth warming and enhanced downward heat flux via vertical mixing. Our results emphasize that the way that ocean circulation affects heat, a dynamic tracer, is considerably different from how it affects passive tracers, like δ18O, and call for caution when inferring water mass changes from δ18Oc records while assuming uniform changes in deep temperatures.

The Transient Response of the Southern Ocean to Stratospheric Ozone Depletion

Abstract Recent studies have suggested that the response of the Southern Ocean to stratospheric ozone depletion is nonmonotonic in time; consisting of an initial cooling followed by a long-term warming. This result may be significant for the attribution of observed Southern Ocean temperature and sea ice trends, but the time scale and magnitude of the response is poorly constrained, with a wide spread among climate models. Furthermore, a long-lived initial cooling period has only been observed in a model with idealized geometry and lacking an explicit representation of ozone. Here the authors calculate the transient response of the Southern Ocean to a step-change in ozone in a comprehensive coupled climate model, GFDL-ESM2Mc. The Southern Ocean responds to ozone depletion with an initial cooling, lasting 25 yr, followed by a warming. The authors extend previous studies to investigate the dependence of the response on the ozone forcing as well as the regional pattern of this response. The response of the Southern Ocean relative to natural variability is shown to be largely independent of the initial state. However, the magnitude of this response is much less than that of natural variability found in the model, which limits its influence and detectability.

Ocean temperature thresholds for Last Interglacial West Antarctic Ice Sheet collapse

AbstractThe West Antarctic Ice Sheet (WAIS) is considered the major contributor to global sea level rise in the Last Interglacial (LIG) and potentially in the future. Exposed fossil reef terraces suggest sea levels in excess of 7 m in the last warm era, of which probably not much more than 2 m are considered to originate from melting of the Greenland Ice Sheet. We simulate the evolution of the Antarctic Ice Sheet during the LIG with a 3‐D thermomechanical ice sheet model forced by an atmosphere‐ocean general circulation model (AOGCM). Our results show that high LIG sea levels cannot be reproduced with the atmosphere‐ocean forcing delivered by current AOGCMs. However, when taking reconstructed Southern Ocean temperature anomalies of several degrees, sensitivity studies indicate a Southern Ocean temperature anomaly threshold for total WAIS collapse of 2–3°C, accounting for a sea level rise of 3–4 m during the LIG. Potential future Antarctic Ice Sheet dynamics range from a moderate retreat to a complete collapse, depending on rate and amplitude of warming.

Southern Ocean control of glacial AMOC stability and Dansgaard-Oeschger interstadial duration

Glacial periods exhibit abrupt Dansgaard-Oeschger (DO) climatic oscillations that are thought to be linked to instabilities in the Atlantic meridional overturning circulation (AMOC). Great uncertainty remains regarding the dynamics of the DO cycle, as well as controls on the timing and duration of the individual events. Using ice core data we show that the duration of the warm (interstadial) phase is strongly correlated with Antarctic climate, and presumably with Southern Ocean (SO) temperature and the position of the southern hemisphere (SH) westerlies. We propose a SO control on AMOC stability and interstadial duration via the rate of Antarctic bottom water formation, meridional density/pressure gradients, Agulhas leakage and SO adiabatic upwelling. This hypothesis is supported by climate model experiments that demonstrate SO warming leads to a stronger AMOC that is less susceptible to freshwater perturbations. In the AMOC stability diagram, SO warming and strengthening of the SH westerlies both shift the vigorous AMOC branch towards higher freshwater values, thus raising the threshold for AMOC collapse. The proposed mechanism could provide a consistent explanation for several diverse observations, including maximum DO activity during intermediate ice volume/SH temperature, and successively shorter DO durations within each Bond cycle. It may further have implications for the fate of the AMOC under future global warming.

Response of Southern Ocean Convection and Abyssal Overturning to Surface Buoyancy Perturbations

AbstractThis study explores how buoyancy-driven modulations in the abyssal overturning circulation affect Southern Ocean temperature and salinity in an eddy-permitting ocean model. Consistent with previous studies, the modeled surface ocean south of 50°S cools and freshens in response to enhanced surface freshwater fluxes. Paradoxically, upper-ocean cooling also occurs for small increases in the surface relaxation temperature. In both cases, the surface cooling and freshening trends are linked to reduced convection and a slowing of the abyssal overturning circulation, with associated changes in oceanic transport of heat and salt. For small perturbations, convective shutdown does not begin immediately, but instead develops via a slow feedback between the weakened overturning circulation and buoyancy anomalies. Two distinct phases of surface cooling are found: an initial smaller trend associated with the advective (overturning) adjustment of up to ~60 yr, followed by more rapid surface cooling during the convective shutdown period. The duration of the first advective phase decreases for larger forcing perturbations. As may be expected during the convective shutdown phase, the deep ocean warms and salinifies for both types of buoyancy perturbation. However, during the advective phase, the deep ocean freshens in response to freshwater perturbations but salinifies in the surface warming perturbations. The magnitudes of the modeled surface and abyssal trends during the advective phase are comparable to the recent observed multidecadal Southern Ocean temperature and salinity changes.