Reconstructing the evolution of the Benguela and Agulhas currents throughout the Plio-Pleistocene and understanding the controls on productivity and sea surface temperature at their confluence is of wide interest. This period of Earth's history is marked by a multi-million-year global cooling trend that may be reflected in the local climates influenced by the two systems. We present alkenone based sea surface temperature (SST) and biological productivity (C37 total/MAR) records at ODP Sites 1085 and 1087 located at the southern end of the modern Benguela Upwelling System (BUS) and under the influence of the Agulhas Current, respectively. At both sites we find SST evolution mirrors the evolution of other upwelling systems across the globe, while productivity was more susceptible to local precession-driven climate change. The SST gradient between the two sites tracks the intensity and location of the currents and shows that the Agulhas leakage was present consistently at Site 1087 for the past 3.5 Ma. The gradient oscillates at orbital time scales with its strongest peaks coinciding with the glacial cycles. Through a comparison to other Benguela alkenone records we propose that prior to the Mid-Pleistocene Transition (MPT) the BUS extended further south and slightly muted the presence of Agulhas leakage at Site 1087. Our results suggest that post-MPT the southern boundary of the BUS shifted northward resulting in the increased influence of the Agulhas at Site 1087 and strengthened upwelling in the northern and central cells of the BUS.
Oxygen concentrations in seawater are sensitive to both changes in carbon storage and surface processes that set the preformed oxygen concentrations of different water masses. New and developing oxygen proxies have led to (semi-)quantitative estimates of glacial oxygen concentrations, which suggest the glacial deep Atlantic had similar to 50 mu mol/kg lower oxygen than in the Holocene. However, the water depth structure of oxygen concentrations in the past is poorly constrained, and so attribution of oxygen variability to specific water masses and parsing the potential role of preformed variability is difficult. In this study, we present the most highly resolved depth transect of glacial oxygen concentrations, which comes from the western tropical Atlantic at the Demerara Rise (7-9 degrees N, 53-54 degrees W, 350-2,550 m water depth). We use the novel quantitative proxy U/Ba to generate oxygen reconstructions in a setting where the foraminifera needed for Delta delta 13C, the only other feasible proxy, are rare. Glacial oxygen content along the Demerara Rise was consistently lower than modern, with the largest change (similar to 100 mu mol/kg) occurring at 1,500-2,500 m water depth. We combine these new oxygen reconstructions with published data to optimize an endmember mixing model that provides estimates of the preformed delta 13C and oxygen content of three water masses in the Atlantic Ocean. In contrast to glacial model simulations, our results suggest lower oxygen concentrations throughout the depth range of our cores, which we suggest was predominately driven by lower preformed oxygen in Antarctic Intermediate Water and North Atlantic Deepwater.
The transition from the Last Glacial Maximum was punctuated by several abrupt cooling periods, the last of which occurred similar to 8.2 ka BP. These cooling events are thought to have occurred when collapse of the impounding ice along the Laurentide Ice Sheet margin released massive volumes of freshwater into the Labrador Sea. Ice core records from the Greenland Ice Sheet (GIS) show modest cooling of similar to 2 degrees C-4 degrees C over ca. 200 years during the 8.2 ka event. Here we present an alkenone-based temperature reconstruction from Limnaea S & oslash;, a West Greenland lake proximal to the GIS margin that indicates a cooling as much as 8 degrees C, although potential biases in the seasonality of production or non-stationarity in the alkenone-water temperature relationship may account for a significant portion of the signal (similar to half). Carbonate isotope records indicate concurrent hydroclimate changes, while clumped isotope temperatures provide important insight into seasonal timing of carbonate accumulation. These data indicate that the amount of cooling along the GIS margin may have been 2-3 times larger than the GIS summit, which is likely attributable to sea ice changes in the Labrador Sea, as predicted by climate models. Resumption of the Atlantic Meridional Overturning Circulation following the 8.2 ka slowdown brought abrupt warmth back to Greenland within similar to 140 years, raising lake temperatures by potentially up to 11 degrees C and resulting in a greater contribution of 18O-depleted meltwater into proximal lakes that produced a marked delta 18O reduction in Limnaea S & oslash; carbonates. The considerable cooling associated with the 8.2 ka event in this terrestrial environment indicates that relatively small climate forcing can have an outsized impact on high-latitude lake systems due to regional feedbacks related to snow/ice cover, meltwater input, and shifts in climate seasonality.
Insolation forcing at the top of the atmosphere is widely used in paleoceanographic and paleoclimatological studies. In paleoclimate simulations, it must reflect Earth's orbital history. General Circulation Models (GCMs) typically use a modern orbital configuration to allow direct comparison to modern and near-future scenarios, often with time slice orbital experiments (DeepMIP, PlioMIP, etc.). However, especially in transient simulations (e.g., of the past 3 Myr), orbital parameters from outdated astronomical solutions (e.g., those labeled Ber78 and BL91) are still used, potentially due to their integration in widely used models such as the Community Earth System Model (CESM). These routines are particularly inappropriate for deep-time simulations, for example, those targeting the Eocene, when the eccentricity is incompatible and the obliquity amplitude is too large compared to modern solutions. Furthermore, prior to Myr, the frequently used La04 solution should not be used. Our orbital solutions, ZB18a and ZB20a, best match geologic data to 58 and 71 Ma, respectively. In this study, we show that outdated astronomical solutions yield incorrect top-of-atmosphere insolation. The difference between insolation from Ber78 and our solution ZB18a(1,1) increases periodically with increasing age, reaching significant offsets (up to 24 at kyr and at 2.1 Myr)-of the same order of magnitude as the orbital forcing itself. To facilitate the use of updated astronomical solutions (AS) in paleoclimate simulations, we provide ZB18a(1,1) and ZB20a(1,1) and Fortran subroutines that interpolate the astronomical parameters and compute insolation. We provide drop-in replacements for existing CESM routines, hoping that proper AS inputs be used in future studies.
Branched glycerol dialkyl glycerol tetraethers (brGDGTs), which consist of intact polar lipids (IPLs, with head group, from living microorganisms and recently deceased cells) and core lipids (CLs, without head group, normally from the degradation of IPLs), are proposed as proxies for temperature reconstruction in peat deposits. To date, most brGDGT-based temperature reconstructions have focused on core-brGDGTs, which are thought to be degraded from IPL brGDGTs. For peat brGDGT-based paleotemperature reconstructions, the fundamental assumption is that core-brGDGTs in surface peats can record the environmental temperature in real-time and are then well preserved after burial. However, whether newly synthesized IPL brGDGTs might mask geological core-brGDGTs signals—crucial for paleotemperature reconstruction—needs further investigation. Here, we analyzed both core- and IPL brGDGTs from a peat core in the Sahara Sand peatland (Northwest China) to investigate this issue. Our theoretical results indicate that, for any sample in the Holocene, more than 87.2% of core-brGDGTs were produced in the acrotelm. Consequently, less than 12.8% of core-brGDGTs—derived from IPL brGDGTs—originated from the catotelm, suggesting that IPL brGDGTs in the catotelm made a limited contribution to the overall composition of core-brGDGTs. Thus, we suggest that peat core-brGDGTs are suitable for temperature estimation, while their signals are possibly time-smoothed due to peat accumulation in the acrotelm. Our results support the fundamental assumption for peat brGDGT-based paleotemperature reconstructions.
Effective fractionation factors (epsilon) relate the delta 2H of leaf waxes to ambient water and are essential for paleoclimate and paleotopography reconstructions using biomarker isotopes. The delta 2H of the water used by plants is often poorly constrained and may contribute to producing highly variable estimates of epsilon. Herein, we compile similar to 700 modern water delta 2H and delta 18O values from across the Himalaya and Tibetan plateau (TP) to calibrate a new orographic precipitation isotope model that produces precise estimates of long-term average primary precipitation isotopes. We couple data and precipitation isotope model predictions with similar to 200 leaf wax delta 2H values to calculate epsilon between leaf wax n-alkanes and meteoric water. Our results demonstrate that: (a) modern primary (non-evaporated) surface water isotopes across the TP and surrounding region can be effectively estimated with less error than existing spatial interpolation models, (b) the precipitation isotope field across this region is dominated by orographic lifting and condensation superimposed on synoptic precipitation patterns, and (c) the effective fractionation between hydrogen isotopes of leaf waxes in soils and meteoric water shows no correlation with elevation, temperature, relative humidity, or vapor pressure deficit and may be constrained by a single epsilon value of -120.2 parts per thousand, despite large variations in ecosystem and climate. This new precipitation isotope modeling approach enables reliable prediction of modern primary (non-evaporated) surface water isotopes (delta 2Hprecip and delta 18Oprecip) as well as reconstruction of ancient delta 2Hprecip using delta 2Hleaf wax.
Constraining how faunal communities vary over time and space in response to environmental change has long been a major goal of paleoecologists. Among fossilized organisms, few yield a better and greater detailed record than foraminifera to understand past marine conditions. The analytical frameworks used to build palaeoenvironmental reconstructions have progressed over the years, providing new perspectives within the field. One of these proposed frameworks, Ecological Trajectory Analysis (ETA), offers the possibility to compare geometric properties of trajectories to understand temporal changes within communities. We provide here an application of ETA to the fossil record by comparing benthic foraminiferal data sets from three previously studied cores from the French Atlantic margin. The three cores are each separated by approximately 350 km and cover the last deglaciation, thus allowing a comparison of community dynamics at spatial and temporal scales. A major perturbation in fossil benthic communities was identified in ETA metrics at the onset of Heinrich Stadial 1, with a twofold departing and recovering trend of communities. The three sites along the French Atlantic margin also show spatial differences during the deglaciation, testimony to a contrasting response of benthic foraminifera to ice sheet dynamics and fluctuating fluvial discharges. In this instance, the application of ETA allows to compare multiple fossil foraminiferal records. At a broader scale, the use of trajectory metrics within one unique multivariate space of community resemblance stands out as a powerful tool to compare many other fossil records within the field of paleoecology.
In the Northern Hemisphere, Eastern Pacific coastal upwelling systems host zones of high marine productivity and moderate the western United States' coastal climate by bringing cold, nutrient-rich waters to the surface. Reconstructions of regional sea-surface temperatures (SST) indicate that upwelling of cold waters began during the Pliocene. At ocean drilling program (ODP) Site 1014 in the Tanner Basin, southern California, this cooling is recorded by the unsaturation index, which finds cooling of similar to 11C, compared to only 3C in surrounding regions. Here, we used this record to test the suitability of reconstructing coastal California SSTs with oxygen and clumped-isotope paleothermometry. We measured carbonate-based isotope proxies (delta 18O and clumped isotopes) in bulk sediment and mixed-foraminifera fractions from ODP Site 1014, the core originally used for the measurements. Carbonate-based SST reconstructions in both bulk and foraminifera fractions show no cooling trend with time. Clumped-isotope temperatures, which do not rely on assumptions about seawater O, are on average 11C colder than alkenone-derived estimates, calculated as the difference between mean clumped-isotope (similar to 9C) and alkenone (similar to 20C) temperatures over 4,200 ka. To investigate this discrepancy, we evaluated three potential sources: (a) contamination; (b) disequilibrium isotope effects during carbonate precipitation; and (c) carbonate diagenesis during burial. In SEM images of foraminifera, we find visually apparent diagenetic alteration across all depths, including the 1000-year-old core top sample, where about two-thirds of specimens are severely altered. We propose that the discrepancy between the alkenone- and carbonate-stable-isotope-based temperatures dominantly reflects carbonate diagenesis occurring within millennia of deposition.
Simulations of the Eocene climate using state-of-the-art Earth system models provide a reference state for the future climate, as the Eocene was one of the warmest geological epoch with high atmospheric carbon dioxide (CO 2 ) concentration and global temperatures comparable to projections for the coming centuries. However, the paleogeographic configuration of the Eocene impact's distinct climate features. Here, we decompose the response of low-level monsoon dynamics over the Indian Ocean to the early Eocene hothouse using five climate model simulations from the Deep-time Model Intercomparison Project (DeepMIP). We see a circulation pattern resembling the paleo-monsoon across all models over the Indian Ocean. Surprisingly, we find low-level jet (LLJ) forming along the topographic barriers of the Eastern African Rift and the Deccan Plateau, which we refer to as the “Proto-LLJ.” Based on the analysis of the DeepMIP results, we find a reduction in the Proto-LLJ strength with elevated CO 2 . Under present-day conditions, the northward shift of monsoonal LLJ is attributable to the increased land-sea contrast under global warming. Even though land-ocean temperature contrasts increased during the Eocene hothouse, Proto-LLJ weakened due to tropical atmospheric stabilization. This stabilization reduced vertical temperature gradients, suppressed convection, and weakened atmospheric overturning, limiting the upward motion needed to drive strong monsoonal winds under CO 2 -induced warming.
Eccentricity cycles in deep-sea paleoclimate records suggest that astronomical forcing notably altered global temperatures and carbon cycle dynamics. Because changes in the distribution of insolation alone cannot explain the observed climate variability, climate-carbon cycle feedbacks must have amplified the response. However, the carbon sources and sinks operating on orbital timescales are poorly understood, especially in absence of dynamic ice sheets as during the early Cenozoic. Here, we use an Earth system model to explore the impact of astronomical forcing on the organic carbon cycle and its expression in key paleoceanographic variables, building on Vervoort et al. (2024, https://doi.org/10.1029/2023pa004826) who outlined the role of inorganic carbon cycle feedbacks. Results demonstrate that subtle changes in marine organic carbon burial, driven by nutrient (phosphate, P) availability, can produce 400-kyr cycles of negative delta 13C excursions during periods of elevated pCO2 and reduced CaCO3 preservation, consistent with typical orbital variations in Paleocene records. The magnitude and phasing of the response to eccentricity forcing are determined by the balance between P release (via temperature-dependent rock weathering) and P removal (via oxygen-dependent sedimentary P retention). Because these processes are strongly influenced by the distribution of landmasses and shelves, paleogeography exerts a first-order control on the expression of astronomical cycles. We do not reproduce the high amplitude 100-kyr "hyperthermal events" of the early Eocene, but our model identifies two potential mechanisms to amplify global warming on orbital timescales: reduced organic carbon burial as well as enhanced kerogen weathering could increase CO2 during eccentricity maxima under favorable conditions.
Climate models and paleoclimate proxies have temperature variability that diverge from each other locally and at long timescales. It is unknown to what extent these divergences also apply to hydroclimate and whether long-term hydroclimate variability is fundamentally different than temperature variability. Here we evaluate the long-term variability of near surface air temperature (tas) and hydroclimate (Palmer Drought Severity Index [PDSI]) using a climate model (the Community Earth System Model-Last Millennium Ensemble [CESM-LME]) and a paleoclimate reconstruction based on this model (the Paleo Hydrodynamics Data Assimilation product [PHYDA]); this framework allows us to see how a model's long-term climate variability is affected by informing it with proxy data. Using power-scaling exponents, we find universally higher scaling values in PHYDA (except for global mean tas) compared to the CESM-LME model. Thus, PHYDA's global PDSI, local PDSI, and local tas are more dominated by low-frequency variability than CESM-LME's. Additionally, PDSI is spectrally flatter than tas in CESM-LME, whereas scaling values of tas and PDSI are comparable in PHYDA. These results indicate that the paleoclimate reconstruction process adds low-frequency variability that CESM-LME otherwise would not have. Based on a range of null reconstruction experiments, we attribute the origin of low-frequency variability in PHYDA to proxy information and not the mathematical properties of the data assimilation methodology. This implies that long-term variability in PHYDA is dependent on the selection of assimilated proxy data.
Branched glycerol dialkyl glycerol tetraethers (brGDGTs), bacterial membrane lipids, are widely used as temperature proxies. Although these proxies are effective for reconstructing past temperatures, brGDGT-based methods encounter limitation in extreme climate gradients and confounding factors such as salinity and aridity, conditions particularly prevalent in drylands. In Arid Central Asia (ACA), the commonly used linear brGDGT calibration exhibits significant biases, resulting in substantial errors in temperature reconstructions. This study compares two machine learning regression methods, Random Forest (RF) and Boosted Regression Trees (BRT), with traditional linear calibrations, using 761 surface samples from the ACA surface database. It also evaluates an unsupervised machine learning approach based on cluster and weighted combined calibrations. The analyses focuses on the robustness of BRT and RF methods, their effectiveness to reduce biases specific to drylands, and their performance in reconstructing Holocene climate, especially temperature and aridity. The results demonstrate that (a) machine learning methods reduce biases caused by confounding factors in drylands and improve temperature accuracy compared to linear models; (b) among the tested approaches, BRT outperforms RF in climate reconstructions; (c) machine learning enables independent and reliable predictions of both temperature and moisture; and (d) cluster-based calibrations provide additional improvements in specific archive contexts. This new machine leaning-based framework increases the reliability of brGDGT-derived climate reconstructions in drylands. Nonetheless, challenges persist due to past environmental fluctuations affecting the study context. Further expansion of surface data sets across a wider range of climates and archive types is essential to strengthen machine learning training and improve model precision.
Warm Slope Water (WSW), with a large Gulf Stream component, and cold Labrador Slope Water (LSW), with a strong Arctic contribution, fill deep shelf basins off southeastern Canada such as Emerald, on the Scotian Shelf, and Jordan in the Gulf of Maine. These slope waters leave their imprint in the stable isotope ratio of benthic foraminifera. We present a simple method for estimating the WSW/LSW blend using hydrographic data, the predicted equilibrium delta 18O in benthic foraminiferal calcite, and the assumption that the dominant control on delta 18O is temperature. Our results show that for most of the past 1800 years in deep Jordan Basin (deposition rate similar to 235 cm kyr-1; temporal resolution similar to 25 years), the slope waters were variable but generally consisted of a 60/40 mixture of LSW and WSW. About 800 years ago, LSW dominated (74% average contribution, 100% maximum contribution) Jordan and Emerald Basins for many decades, about the time that the climate was cooled by volcanic eruptions. During the Little Ice Age from similar to 500 to 300 and similar to 200 to similar to 150 years ago, the LSW contribution was about 84% and 78%, respectively. The most extreme event, a strong trend in delta 18O of -0.006 parts per thousand yr-1 (corresponding to similar to 3.2 degrees C increase in temperature), occurred during the Industrial Era warming beginning about 140 years ago. While hydrographic data indicate that the North Atlantic Oscillation can force short term slope water dynamics, it does not seem to be a primary driver of long-term variability seen in the core data.
The Younger Dryas chronozone (YD), between similar to 12,900 and 11,600 years ago, represents a climatically dynamic period in Earth's history that was defined by rapid millennial-scale climate changes. The cause of the climate change during the YD is typically attributed to substantial meltwater flux into the North Atlantic, which weakened the Atlantic meridional overturning circulation (AMOC). However, it is unclear if the climate signal from periods of AMOC shutdown were truly global in nature. Here, by utilizing the transient iTraCE simulation, a coupled climate model simulation of the last deglaciation, we test if the climate in Alaska was strongly influenced by an AMOC forcing. We find no significant impact of a North Atlantic meltwater forcing on Alaska climate during the YD. However, we do find millennial-scale climate changes concurrent with the YD. We argue that the millennial-scale climate change in Alaska during the YD was driven by changes in the tropical Pacific mean state, which could have been mistaken for an AMOC forcing.
The southern Tibetan Plateau (TP) is a crucial component of the “Asian Water Tower,” whose moisture dynamics influence both the terrestrial ecosystems and welfare of a large human population across Asia. However, the pattern of its moisture evolution and interactions with atmospheric circulation remain unclear, limiting our understanding of the current and future hydroclimatic changes. Here, we present geochemical, environmental magnetic, and color proxy records derived from a well-dated, high-resolution aeolian sedimentary record. Based on detailed evaluation of the environmental proxies and integrating data from other aeolian records, we investigated the spatial heterogeneity of moisture modes and their atmospheric responses. Our findings reveal two distinct Holocene moisture evolutionary modes in the southern TP, with their boundary delineated by the Shannan wide-valley, which roughly coincides with the southernmost northern boundary of the modern monsoon. In the western region, moisture was co-influenced by the mid-latitude Westerlies (MLW) and the Indian summer monsoon (ISM), characterized by higher moisture during the early Holocene due to strong ISM precipitation. This was followed by a sharp decrease in moisture and then a gradual increase from the middle Holocene onward, under the influence of an enhanced winter MLW. In contrast, the moisture mode in the eastern region was predominantly influenced by the ISM and shows a general decreasing trend. Using climate simulations, we project a future scenario where, as the winter MLW weakens and the ISM strengthens, moisture in the southern TP will experience a decreasing trend in the west and an increasing trend in the east.
Hydroxylated glycerol dialkyl glycerol tetraether (GDGT)-based proxies have recently been proposed for reconstructing sea surface temperatures (SSTs) in polar oceans, but their reliability has not yet been sufficiently validated. We investigated the spatial distribution of hydroxylated GDGTs in surface sediments from the western Arctic Ocean and the Bering Sea to evaluate the reliability of these proxies and apply them to a sediment core from the Chukchi inner shelf to reconstruct Holocene SST variability. Among the four tested proxies, the RI-OH' index shows the strongest correlation with annual mean SST, marking it the most reliable and well-calibrated SST proxy for this region. The RI-OH'-based SST reconstruction reveals two major cooling events at 9.1-8 ka and 4.6-3.7 ka. The earlier event coincides with intensified inflow of cold Pacific Winter Water into the Chukchi Sea, while the latter event is linked to expanded sea ice cover and advance of Alaskan glaciers, likely driven by regional shifts in atmospheric and oceanic circulation. A brief SST decline at 8.3 ka, concurrent with the well-known 8.2 ka event, is observed for the first time in the western Arctic Ocean. The latter cooling episode aligns with the 4.2 ka climate event across the Northern Hemisphere, suggesting that the Arctic climate dynamics contributed to the broader Holocene climatic transition.
Deoxygenation and expansion of the Oxygen Minimum Zones (OMZs) are among the most significant threats for the oceans posed by ongoing global warming. However, current knowledge of the underlying drivers is limited by short and sparse instrumental records, constraining the ability of models to project the future behavior of OMZs under global change. In this study, we used a multi-proxy approach to reconstruct the dynamics of the shallow OMZ in the eastern subtropical South Atlantic, aiming to identify the main drivers of its variability through time. Our results reveal a gradual intensification of the shallow OMZ off Angola from similar to 43 to similar to 25 kyr BP, followed by persistently low oxygen conditions thereafter, primarily driven by progressive warming of the subtropical South Atlantic. The warming-induced reduction in oxygen solubility and decreased ventilation due to enhanced upper ocean stratification dominated over the influence of regional upwelling and remote physical-biogeochemical processes in the Southern Ocean. Our findings support model projections and observational data that highlight the dominance of warming-related processes as key drivers of tropical OMZ expansion. These mechanisms are likely to play a central role in the future intensification of OMZs as ocean temperatures continue to rise under ongoing global climate change.
The Colombian Andes has a hydroclimate with multiple proposed drivers of interannual, sub-decadal and multidecadal variability, including the meridional movement of the Inter Tropical Convergence Zone, El Ni & ntilde;o Southern Oscillation (ENSO), and Pacific Decadal/Atlantic Multidecadal Oscillation. Globally, the late Holocene (3-4 k yrs B.P.) is of interest as multiple proxy studies suggest modern multidecadal rainfall variability developed during this period. Limited high-resolution proxy studies assess changes in rainfall during the late Holocene from the Tropical Andes. We present a record of stable isotopes (delta 18O and delta 13C) from two speleothems, C2A-1 and C8A, that grew during the late Holocene in the Eastern Cordillera of the Colombian Andes from Cueva de la F & aacute;brica. Hendy Tests on both stalagmites were consistent with isotopic equilibrium at the time of calcite deposition, suggesting delta 18O values reflect climate variability. Frequency analyses revealed statistically significant oscillations at the sub-decadal and multidecadal scale in the faster-growing portion of C2A-1. We isolated the sub-decadal delta 18O variability through a band-pass filter, calculated the amount of rainfall for C2A-1, and compared the distribution against modern (1979-2016 C.E.) rainfall. The unimodal distribution of band-pass filtered speleothem delta 18O data suggests a response to an internal mode of climate variability, such as ENSO. A regional speleothem proxy comparison over 3-4 k yrs B.P. suggest a bimodal expression of rainfall at the multidecadal timescale with drier central Amazonia and wetter coastal and high-Andes mean rainfall states. This complex regional picture of multidecadal hydroclimate is likely attributed to internal climate forcings between the Pacific and Atlantic Oceans.
Two episodes of expansion and strengthening of subsurface oxygen minimum zones (OMZ) associated with seafloor hypoxia occurred in the North Pacific during the Bolling-Allerod (B/A) and early Holocene (EH) warm periods. The mechanisms driving these hypoxic events at high latitude, however, remain unclear. Here, we integrated multiple long-chain alkenone temperature records (U-37(k ')) from the eastern and western margins of the North Pacific to infer the dynamics of the subpolar gyre (SPG). It was found that the mean temperature difference between NE and NW Pacific effectively reflects the variation in intensity of the SPG since the Last Glacial Maximum. During the B/A and EH warm periods, the northward shift of the westerlies led to strengthening of the SPG, and consequently, an increase in upwelling. The enhanced upwelling of nutrient-rich waters highly likely resulted in increased biological productivity, and thus led to expansion of the OMZ underneath the SPG during the B/A and EH periods.
Biological productivity, shaped by climate and environmental factors, is critical to climate feedback mechanisms and the global carbon cycle. This study investigates the measurement of six phytosterols (β‐sitosterol, stigmasterol, campesterol, dinosterol, epi‐brassicasterol, and 24‐methylene cholesterol) from core MD01‐2414 in the central Okhotsk Sea, covering the past 1.5 million years (Ma). These sterols serve as proxies for terrestrial and marine productivity in the central Okhotsk Sea and northeast Siberia. Sterol concentrations reflect global glacial/interglacial cycles between 1.2 and 0.6 Ma, with higher and lower values during interglacial and glacials intervals, respectively. X‐ray fluorescence (Ba/Ti) and total organic carbon/total nitrogen (TOC/TN) ratios indicate shifts in marine and terrestrial sources, confirming biological productivity as a key driver of sterol deposition. Sterol fluxes, combined with sea surface temperature records from the northwest Pacific and sea ice proxies from the Bering Sea, reveal an extreme cold interglacial (Marine Isotope Stage, MIS 23) and prolonged glacial conditions during MIS 22, which stressed both terrestrial and marine ecosystems. Increased sea ice expansion during this period likely fostered North Pacific Intermediate Water formation, reducing upwelling and CO 2 exchange between bottom waters and the atmosphere. Integration of sterol data with regional records, including pollen, temperature, and sedimentary facies from Lake El'gygytgyn, highlights a warming event at the onset of MIS 32, peaking in late MIS 32. This warming precedes the “super‐interglacial” MIS 31 and coincides with maxima in boreal and austral summer insolation, underscoring its significance in regional climate evolution.
