Monthly-Resolved Cave Proxy Evidence for Northward Gulf Stream Migration During the Little Ice Age

Monthly-Resolved Cave Proxy Evidence for Northward Gulf Stream Migration During the Little Ice Age | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Monthly-Resolved Cave Proxy Evidence for Northward Gulf Stream Migration During the Little Ice Age Edward Forman, James Baldini, Robert A. Jamieson, Franziska Lechleitner, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5973851/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Jul, 2025 Read the published version in Communications Earth & Environment → Version 1 posted You are reading this latest preprint version Abstract The Gulf Stream forms part of the upper-ocean limb of the Atlantic Meridional Overturning Circulation (AMOC), playing an essential role in redistributing heat northward and greatly influencing regional climates in the North Atlantic. Understanding Gulf Stream path and strength variability on longer timescales is vital to contextualise its present-day weakening and to fully appreciate its sensitivity to forcing. We present a 558-year long (1456–2013) proxy record of sea surface temperature from a Bermudan stalagmite using an indirect magnesium-temperature calibration based on a connection to wind speed. Our monthly-resolved terrestrial palaeo-oceanographic temperature reconstruction indicates that the Gulf Stream was likely positioned further south than today during the Little Ice Age. We suggest that a combination of reduced Gulf Stream transport, enhanced Labrador Current and Deep Western Boundary Current transport, and an extended negative North Atlantic Oscillation phase, caused the Gulf Stream to be at lower latitudes during the Little Ice Age, before migrating northward as the Little Ice Age abated. Earth and environmental sciences/Climate sciences/Palaeoclimate Earth and environmental sciences/Climate sciences/Palaeoceanography Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction The system of ocean currents collectively known as the Atlantic Meridional Overturning Circulation (AMOC) are a major driver of climate variability. The AMOC is a highly non-linear system and is thought to reach its critical threshold this century 1 , which would have serious ramifications for regional climates such as up to ~ 15°C temperature reductions in Northwest Europe 2 . Thus, understanding its response to climate forcing is critical. One key component of the AMOC is the Gulf Stream (GS), which transports warm water from the Gulf of Mexico northward towards Northwest Europe. Due to its presence, at least in part, winter surface air temperatures in western Europe are up to 10°C higher than the zonal mean at comparable latitudes 3 . GS path and AMOC strength are intrinsically linked. A weaker AMOC reduces the Deep Western Boundary Current (DWBC), weakens the northern recirculation gyre, and shifts the GS separation point northward 4 . Anthropogenic warming has weakened the AMOC to perhaps its weakest state over the past millennium, inducing a northward migration in the Gulf Stream North Wall 5 , 6 . Temperature records from the Gulf of Maine have traced the effects of GS positional shifting over the last 100 years 7 , with northward movement causing regional sea surface temperatures (SSTs) to rise. However, on longer timescales that are critical for understanding sensitivity to global temperature change, precise GS position remains elusive. The Little Ice Age (LIA) represents a key time interval for understanding how GS positioning and dynamics responds to climate change because it was characterised by extended Northern Hemisphere (NH) cold intervals between about 1300 and 1850 CE. Proxy evidence suggests the LIA featured a GS that was approximately 10% weaker and positioned further south, likely due to a stronger Labrador Current (LC) and enhanced recirculation gyre 8 – 10 . The lack of instrumental records from the LIA means that palaeoclimate reconstructions are crucial for inferring GS state and position. Stalagmites are increasingly providing important constraints on the interpretation of palaeoclimates, particularly due to their propensity to provide accurate and precise chronologies and potential for continuous, high-resolution records, in some cases at sub-monthly timescales. Strong environmental seasonality can result in visible or geochemical laminations, offering high-precision chronological control, complementing any radiometric (e.g., U-Th or 14 C) determinations (e.g., 11 , 12 ). Magnesium concentrations in stalagmites can exhibit a seasonal signal (e.g., 12 , 13 ) that can be employed both as a chronological tool (via seasonal cyclicity) and as a climate proxy (through Mg concentrations). Stalagmite Mg concentrations can record temperature based on the temperature dependency of Mg’s partition coefficient (D Mg ) in calcite (e.g., 12 , 14 ). However, sea spray contributions to rainfall may also affect dripwater especially in island karst systems and, consequently, stalagmite Mg concentrations 15 – 18 . Bermuda’s position within the recirculation gyres south of the GS makes it an ideal site to capture the effects of a southward GS shift, and the presence of extensive cave systems offers a rich speleothem archive for climate reconstruction. However, while other SST reconstructions for Bermuda exist (e.g., 19 – 21 ), a continuous annual SST record for Bermuda spanning more than the last three centuries has yet to be realised. Understanding the palaeoclimatic changes experienced in Bermuda on this temporal scale will provide new insights into how the LIA, with its clearly expressed NH climate anomaly, affected GS position and strength 8 , 22 – 24 . Here, we use Mg concentrations from a stalagmite from Bermuda to construct a monthly palaeotemperature record through sea spray contributions to dripwater, spanning 558-years (1456–2013 CE). We derived a Mg-SST calibration based on a Cochrane-Orcutt regression between our record and a previously developed coral Sr/Ca-SST record 20 . The resulting reconstruction yields SSTs covering most of the LIA and reveals GS migration patterns over the past 558 years at very high temporal resolution and chronological accuracy. Site Description Leamington Cave in north-eastern Bermuda (32° 20’ 31.64” N, 64° 42’ 30.93” W; entrance at 18 m a.s.l.) is a privately-owned former show cave connected to the ocean via flooded conduits and a large cave pool (Fig. 1 a). The cave contains many actively growing speleothems above the high tide line, with cave monitoring data revealing a same-day drip rate response to rainfall that suggests residence times capable of preserving a seasonal signal (Fig. 1 b). The Bermuda Islands consist of a carbonate platform above a volcanic seamount and are comprised almost entirely of aeolianites and palaeosols 25 . The Walsingham Formation, an aeolianite limestone unit (the oldest limestone unit, deposited approximately 1.1–0.8 Ma), hosts most of Bermuda’s caves (including Leamington Cave) and outcrops along the eastern border of the Harrington Sound and the western edge of Castle Harbour 26 , 27 . Bermuda has a humid subtropical climate (Cfa in Köppen-Geiger classification 28 ) and lies within the North Atlantic hurricane belt, with the hurricane season running from June to November each year. Meteorological data 29 indicate that mean sea temperatures are lowest in February (17.8°C) and peak in August (28.5°C) (Fig. 1 c). May historically experiences the lowest mean monthly rainfall (86.7 mm), whereas October has the highest (161.6 mm) 29 . The hurricane season accounts for a mean of 55% of annual rainfall. Except for September and October, when winds prevail from the east, winds typically come from SW and SSW, reaching a maximum in January (6.01 m/s) and a minimum in September (4.49 m/s) 29 . Results and Discussion Bermuda Temperature Trends Our SST reconstruction for Bermuda is derived from stalagmite BER-SWI-13 using an age model constructed from a Mg cycle count. The cycle count is modelled to a radiocarbon chronology in intervals where cyclicity is less pronounced and is better constrained and more precise than both the radiocarbon and U-Th chronologies (Fig. 2 ; full details of our chronology and age model construction are discussed in Materials and Methods). By virtue of the inverse relationship between wind speed and SST, our monthly-resolved Mg record serves as an indirect proxy for SST via sea spray contributions to dripwater. SST estimates were obtained through calibrating the stalagmite Mg record to a previously published coralline SST record (r = -0.51, p < < 0.001; 1782–1998; 20 ) (Fig. 3 , a and b). The resultant record spans 558 years (1456 to 2013 CE), making it one of the longest continuous high-resolution records for the region. The stalagmite-derived SST reconstruction exhibits notable agreement with both five previous coralline palaeotemperature records from offshore Bermuda and reanalysis data (Fig. 3 c and S8), illustrating its reconstructive power 19 – 21 , 30 . The record presented here cannot be confidently compared to observational SST data (starting in 1950) due to anthropogenic influences on speleothem deposition post-1967. However, because the coralline SST record used for our calibration is derived from a regression with instrumental data, its significant correlation with the stalagmite reconstruction reinforces the accuracy of our record 20 . The new stalagmite SST record (Fig. 3 ) suggests that prior to the 18th century, SSTs were relatively stable, fluctuating within a range of 1.4°C around a mean of 23.8°C (compared to the 1950–2018 average of 22.7°C; 29 ). A cooling trend lasting ~ 130 years began in ~ 1720 CE at a rate of approximately − 0.13°C per decade and ended with the onset of the post-1850 warming. The cooling period exhibits high frequency interannual oscillations with decadal-scale fluctuations of ~ 1.5°C (a range of 3.2°C around a mean of 22.9°C), compared to fluctuations of less than 0.5°C over the same timescale before ~ 1720 CE. The interannual variance also decreases after 1967 CE, but this is likely the product of anthropogenic impacts on the site’s hydrology and/or a shallower mixed layer depth 31 , 32 rather than changes in climate as the annual signal across that interval is also less pronounced, despite no change in the annual climate signal. Both the reconstruction presented here and the Goodkin et al. 20 coralline record indicate a warming of ~ 0.11°C per decade starting around 1850 CE. The records also both suggest a brief cooling interval starting in the late 19th century which lasted around two decades before warming resumed. The relatively cooler mean SSTs of the 18th and 19th century when compared to those pre-1720 CE are likely explained by wind speed changes artificially increasing estimates and/or aerosol forcing in the NH reducing SSTs 33 , 34 . Influences on Reconstructed SST Variability The SST record reconstructed here exhibits the effects of both changes in ocean currents and anthropogenic climate change. Enhanced radiative forcing from additional greenhouse gases and rising total solar irradiance at the end of the LIA are likely to be the cause of the increased temperatures observed after 1850 CE (Fig. 4 , a and g). Although, NH temperature anomalies remained largely low before the mid-1800s (Fig. 4 g), the unusually high SSTs reconstructed for Bermuda prior to 1720 CE suggest conditions influenced primarily by natural factors, without a significant anthropogenic impact. Neogloboquadrina pachyderma abundances in a sediment core from the northwest Atlantic 35 suggest a warming trend beginning in the 18th century (Fig. 4 a). Similarly, a Mg/Ca-derived temperature record for Chesapeake Bay 36 demonstrates moderate SSTs during the LIA before beginning to increase in the mid-1700s (Fig. 4 b). However, our reconstruction from Bermuda (Fig. 4 c) suggests regional ocean temperatures were high during the oldest part of the record (1456–1720 CE) until a cooling trend ensued around 1720 CE lasting approximately 130 years. The opposing temperature trends suggest a ‘thermal seesaw’ between the records, whereby an increase in SSTs at lower latitudes leads to a decrease at higher latitudes. This likely reflects southward GS migration during the LIA causing the observed rise in regional Bermudan SSTs compared to those seen at higher latitudes. We propose that a combination of weakened transport, a stronger LC and DWBC, and a sustained negative North Atlantic Oscillation (NAO) phase caused the GS separation point to be further south during the LIA, possibly reducing subtropical mode water potential vorticity. Keigwin and Pickart 37 suggested that northward slope water current migration during the LIA was responsible for an observed warming in several cores south of Newfoundland, but our SST record indicates a broader antiphase system suggesting the northward ocean current shift was more widespread. GS, DWBC and LC flow-speed reconstructions further support this proposed GS drift (Fig. 4 , d to e). Thornalley et al. 35 showed the DWBC was stronger towards the end of the LIA using two mean sortable silt ( \(\:\stackrel{-}{SS}\) ) records from sediment cores off Cape Hatteras (Fig. 4 d). A DWBC weakening from about 1750 CE also supports the notion of a northward shift in the GS separation point starting around the time. Rashid et al. 9 found the LC to be strongest during the LIA (over the last 1.45 ka) using \(\:\stackrel{-}{SS}\) data from a core off the southeastern Grand Banks (Fig. 4 d), which would also have caused the GS to separate from the coast further south 10 , 38 – 40 . Lund et al. 8 reconstructed Florida Current transport over the last millennium using sediment cores near the Dry Tortugas and Great Bahama Bank (Fig. 4 e). From this reconstruction they estimate that the GS was systematically weaker during the LIA by approximately 10%. Given a stronger subpolar gyre at the time this implies that the GS would have moved south and reduced northward heat transport 10 . Furthermore, studies have also found a link between the GS and the NAO, an atmospheric circulation phenomenon associated with changes in the surface westerlies across the North Atlantic 40 . The NAO index is defined by the normalised pressure difference between the Icelandic Low and the Azores High and modulates mid-latitude westerly strength as well as heat, freshwater, and momentum fluxes in the North Atlantic (e.g., 41 ). Studies have linked the NAO to changes in both North Atlantic Deep Water formation and Labrador Sea convection 41 . During a more negative NAO phase, decreased heat loss, higher precipitation, and weaker winds reduce convection in the Labrador Sea, corresponding to a weaker AMOC and in turn a more southerly GS path 42 . This is evident in current GS path migration (1993–2016), which may reflect an increasingly negative NAO (e.g., 40 , 43 ). Hence, the persistent negative NAO index values pre-1700 observed in the Luterbacher et al. 44 winter NAO reconstruction also corroborate a southward GS during the LIA (Fig. 4 f). The low Mg concentrations pre-1700 in Bermuda may be also attributable to weaker westerlies during an extended negative NAO phase. A weakening of the westerlies at this time could have reduced sea spray over the site and thus led to less Mg incorporation into the stalagmite. These weaker winds are evident in the relatively low implied seawater contribution from the Mg and Sr concentrations at the time (~ 4.97% compared to the mean ~ 6.10%; Figure S6). Considering that a transition in the 18th century can also be seen in sediment core-derived temperature and flow speed reconstructions (e.g., 35 , 36 ) there is likely a direct temperature effect in addition to any changes in wind speed. Thus, the GS likely did migrate at this time, but the potential for reduced wind speed over Bermuda to exaggerate the reconstructed SST signal complicates the quantification of the exact southward positional extent of the GS movement. Speleothem Mg concentrations are sensitive to the effects of prior calcite precipitation (PCP) and are often used as a proxy for rainfall amount (e.g., 45 , 46 ). However, areas with evenly distributed rainfall throughout the year usually do not exhibit a strong PCP effect 45 , 47 . Therefore, a change in cave processes (such as the degree of water-rock interactions or PCP) is unlikely to be the primary driver of Mg concentration variability as they are controlled by broader climatic factors, and only the period after 1967 exhibits the effects of PCP following anthropogenic impacts on the cave system (Figure S6). For this reason, although cave processes or the amount of fluid inclusions may reduce the accuracy of the temperature estimates, the trends exhibited here are likely not exclusive to the BER-SWI-13 record and reflect broader regional trends. Assuming our SST estimates reflect exclusively temperature change without amplification due to wind speed, the 1–2°C temperature change over 130 years suggests a moderately fast rate of GS migration when considering the relatively shallow nearby SST gradient in the modern North Atlantic (Fig. 4 and S9), implying that GS position is reasonably sensitive to forcing. Observational data tracking the current northward GS migration detected a shift in the GS north wall of ~ 1.5° in 52 years (1965–2017) between 50°W and 40°W 6 . Therefore, with the continuation of current trends in the GS and AMOC, the sensitivity of regional temperatures to GS path drift could have serious ramifications for future regional climates, ecosystems, and extreme weather events. Our study reconstructed regional SST using a stalagmite from Bermuda via sea spray contributions to dripwater. A Mg cycle count, modelled to a radiocarbon chronology, yielded a high-precision age model and a monthly-resolved Mg record spanning five centuries (1456–2013 CE). Palaeo-oceanographic temperature estimates were derived from an indirect magnesium-temperature calibration via a connection to wind speed, which modelled the Mg concentration record to a previously published coralline proxy SST record 20 . The resultant record indicates temperatures before the early 1700s were consistently high, contrary to more Northerly regional reconstructions and the NH average 35 , 36 , 48 . We propose this ‘thermal seesaw’ pattern suggests that the GS migrated northwards towards the end of the LIA following a prior period characterised by a more southerly GS path. Flow speed reconstructions for the GS, DWBC and LC support this interpretation 8 , 9 , 35 , providing evidence of weakening southward currents in the 18th century from strong conditions during the LIA. Further research should focus on quantifying the rates of positional change seen here, to gain a better understanding of GS sensitivity to additional forcing as well as its subsequent manifestations and effects. These results provide evidence of a moderately fast natural GS migration during the LIA. Previous research has proposed changes in ocean circulation towards the end of the LIA (e.g., 35 ), and our high-resolution SST reconstruction from Bermuda reinforces this concept. Methods Stalagmite BER-SWI-13 BER-SWI-13 is a 192 mm long stalagmite that grew in Leamington Cave, Bermuda (Fig. 1 , a to b). The stalagmite was selected for its favourable internal structure and active drip hydrology, which is responsive to rain events 49 and plots as a near-ideal seasonal flow 50 . Because cave monitoring data revealed a same-day drip rate response to rainfall events (Figure S1 ) 49 , the residence time of the seawater signal is sufficiently low to accurately preserve hydrology-induced seasonality at a monthly-resolution 50 . The cave pool’s connection to the ocean promotes daily cave ventilation by tidal flushing, suggesting the stalagmite never experienced condensation corrosion due to CO 2 buildup 49 . Additionally, monitoring data shows daily cave temperature is positively correlated to daily mean SST (r = 0.83, p < 0.01) (Figure S2) 49 . U-series and Radiocarbon Dating Reconnaissance CT scanning determined the position of stalagmite BER-SWI-13’s central growth axis and informed its sectioning. Cut sections were polished and then cleaned in an ultrasonic bath of deionised water (see Walczak 49 for full methods). Visual examination after sectioning showed no signs of growth hiatuses (Fig. 1 b). Sixteen ~ 400 mg carbonate powder samples (Table S1 ) were milled along the stalagmite growth axis. These were dated using multi-collector inductively coupled plasma mass spectrometric (MC-ICPMS) U-Th methods at the facilities of Bristol Isotope Group (BIG), University of Bristol, and University of St. Andrews Isotope Geochemistry (STAiG). U and Th separation and measurement protocols described in Hoffmann et al. 51 were followed, where CRM U112a and an inhouse Th standard (Teddii) was used for bracketing the samples and correcting for instrumental drift in mass bias, SEM yield. We adopt relevant decay constants from Cheng et al. 52 . U-series data was analysed using the IsoplotR software 53 . U-Th ages for material that is both young (i.e., < 1 ka) and has low 238 U/ 232 Th ratios are poorly constrained because there is no prior knowledge of the 230 Th/ 232 Th ratios at the time of calcite precipitation. Because of this, we attempted to assess the initial Th ratio using isochron methods for two of the milled samples (BER-SWI-UTh-13 and − 15) by performing replicate analysis (~ 100 mg sub-samples). However, there was insufficient 238 U/ 232 Th ratio variation in the two sets of replicate analysis to determine useful estimates of the initial 230 Th/ 232 Th component. Most researchers use an estimate of initial Th assuming it is equivalent to bulk Earth silicate composition ( 238 U/ 232 Th = 0.8 ± 0.4). However, there are many examples of elevated values of this ratio where either carbonate dust with high U/Th or hydrogenous elements in cave drip waters have probably influenced the isotopic composition: see review by Richards and Dorale 54 and recent papers by Wortham et al. 55 and Huang et al. 56 . Other speleothem samples from the North Atlantic and Caribbean region, including the Bahamas, Yucatan, and Puerto Rico, have been similarly affected 57 – 62 . In the absence of sufficiently robust estimates of the initial Th ratio via isochron methods, we considered the principle of stratigraphic constraint (see 63 ) and adopted an elevated value (( 238 U/ 232 Th) initial = 5 ± 2.5) that eliminates the major age inversion at the top of the sample (Figure S3) that is evident when using the bulk Earth initial Th ratios. Using this elevated value also shifts the basal age (BER-SWI-UTh-16) younger and would suggest a growth rate for the oldest material that is closer to that exhibited for the rest of the speleothem. One anomalous age remains for the sample (BER-SWI-UTh-5), which is likely to have been affected by post-depositional alteration. Additionally, twenty-five 8–9 mg carbonate powder samples were milled for 14 C analysis (Table S2) using a semi-automatic high-precision drill (Sherline 5400 Deluxe) at ETH Zurich. The 14 C samples were milled in increments of 5 mm for the top 50 mm, and 10 mm for the remaining 140 mm 49 . The top ~ 0.1 mm was discarded for each 14 C sample. Potential contamination during sub-sampling was minimised by cleaning all tools with methanol and drying with compressed air in between samples. Carbonate samples were prepared for accelerator mass spectrometry (AMS) analysis by graphitisation at the Laboratory for Ion Beam Physics (LIP) of ETH Zurich. The samples were processed on an automatic graphitisation system coupled to a carbonate-handling system (CHS-AGE, Ionplus, Switzerland) by acidification to CO 2 using 2 mL of 85% H 3 PO 4 and followed by graphitisation using excess H 2 over Fe catalyst. Radiocarbon contents were measured on a MICADAS accelerator mass spectrometer (Ionplus) following standard protocols for quality and blank assessment: oxalic acid II (NIST SRM 4990C) was used as the normalising standard, measured to a precision of better than 2‰. We used IAEA-C1 as a blank and IAEA-C2 and a modern coral standard as secondary standards. The procedural blank was established using a 14 C-free stalagmite (MAW-1, ~ 170 kyr old). Trace Element Analysis Trace element concentrations were measured using the prototype RESOlution M-50 excimer (193 nm) laser-ablation system with a two-volume laser-ablation cell coupled to an Agilent 7500ce/cs quadrupole ICPMS at Royal Holloway University, London 64 – 66 . Ablation tracks were pre-ablated to remove any superficial contamination and measured using a 140 by 10 µm rectangular laser slit across 50 mm sectioned blocks on the opposite half to the carbonate sampling. Sectioning was slightly off perpendicular to the growth direction so that overlaps in transects could compensate for any missing data between sections. A 15 Hz laser repetition rate and a laser fluence of ~ 4 J/cm 2 with a stage scan speed of 10 µm s - 1 of the LA cell were used during the continuous main track measurement 64 . Speleothem analyses were bracketed by analyses of NIST 612, NIST 610 for quantification using 43 Ca as internal standard 67 , and MACS3 standards for accuracy control (see 68 ). The resultant data were then reduced using the lolite software package, using NIST 610/612 standards for external standardisation 69 . BER-SWI-13 Chronology The U-Th chronology suggests an unusual range in speleothem growth rate for a humid subtropical cave environment (65 to 1636 µm yr - 1 ; where the 90th percentile of annual layers in a global review 11 is 404 µm yr - 1 ). This variability is most pronounced between depths from top of 26 and 35 mm where the growth rate implied from U-Th dating suddenly rises from 233 µm yr - 1 to 1636 µm yr - 1 for 5.5 years before returning to 196 µm yr - 1 , despite no hiatuses being observed. Additionally, the carbonate samples returned consistently low U concentrations of 83–136 ng/g, except for two samples that are 299 and 362 ng/g. The low U concentrations and unrealistically high variations in growth rate suggest that a chronology based solely on U-Th dating would yield inaccurate results. For this reason, we use the U-Th chronology only as confirmation when developing an alternative chronology based on radiocarbon dating and annual Mg cycle counting. First, we developed an ‘average’ chronological model using the radiocarbon-based technique presented in Lechleitner et al. 70 and refined in Fohlmeister and Lechleitner 71 . This modelling approach uses the radiocarbon data to construct a best fit linear growth rate model, taking into account atmospheric variability in 14 C and potential variability in dead carbon fraction. The chronology produced this way is anchored to a point of known age, which in our case is given by the clear increase in 14 C indicating the rise of the atmospheric bomb spike at 1955 CE (at a depth of 10.179 mm) (Fig. 2 b). The stalagmite collection year (2013 CE) provides a second chronological anchor. However, owing to possible localised anthropogenic influences on the stalagmite’s drip hydrology (e.g., land use change above the cave related to the development of the cave for tourism) starting ~ 1970 CE, we do not consider data post-1970 in our interpretations. The LA-ICPMS-derived Mg profiles exhibit well-developed cyclicity through much of the record. A comparison between the number of Mg cycles in the record and the U-Th and radiocarbon chronologies strongly suggests that the Mg cycles are annual. We conducted multiple Mg cycle counts with different observers to test their reproducibility, and constructed a chronology based on the annual Mg cycles (Fig. 2 ). The cyclicity is ambiguous for some sections of the record; thus, in the final cycle count these sections were modelled using the radiocarbon chronology mean growth rate (Figure S4). Mg concentrations in BER-SWI-13 are significantly inversely correlated (r = -0.51, p ≪ 0.0001, n = 217) with a previously developed coral Sr/Ca-derived proxy SST record from Bermuda 20 , which is contrary to that is expected based on the temperature-dependency of D Mg . Instead, Mg concentrations appear to be an indirect proxy for SST via sea spray contributions to rainfall. A clear inverse relationship exists between SST and wind speed (Figure S5). Colder temperatures are associated with higher wind speeds, which gives rise to higher sea spray contributions to rainfall and vadose dripwater. Seawater has a much higher Mg concentration than normal meteoric water and this influences the concentrations in cave dripwaters (and secondary calcite precipitates such as stalagmites). Additionally, because the Mg and Sr data do not plot along a PCP vector (Figure S6), PCP is unlikely to be a significant control on the record. The post-1970 CE data appear to show evidence of PCP supporting our decision to exclude these data from our interpretations due to alterations to the stalagmite’s drip hydrology. However, the Mg and Sr data do plot along a mixing curve between the concentrations of the bedrock and marine aerosols (Figure S6), supporting the perspective that sea spray (in turn controlled by SST) was the dominant control on Mg and Sr concentrations. By virtue of the inverse relationship between speleothem Mg and SST, the Mg cycle count was tuned intra-annually so that the Mg concentration minima were assigned a decimal date corresponding to the 15th of August (midway between the months with the lowest mean wind speed) and maxima (determined by the broad shape of the cycle to avoid individual weather events) were assigned a date corresponding to January 1st (start of month with the highest mean wind speed). The resultant high-precision final cycle count (1456–2013 CE) is consistent with the radiocarbon chronology, with a nearly identical mean growth rate (Fig. 2 a), as well as high intra-annual precision. The final cycle count yields markedly younger values than the U-Th data until depths from top greater than ~ 128 mm when the U-Th growth rate rapidly increases, whilst also having a more consistent annual growth rate (range in annual cycle length - final cycle count: 865.8 µm yr - 1 ; U-Th: 1571.8 µm yr - 1 ). The COnstructing Proxy Records from Age models (COPRA) algorithm 72 was used to develop an age-depth model with cycle count annual maxima and minima as inputs. The resultant COPRA model incorporated 2000 Monte Carlo simulations and piecewise cubic Hermite interpolating polynomial (PCHIP) interpolation to assign precise ages to every data point. The COPRA algorithm did not identify any hiatuses in the record. The complete dataset consists of 18,471 data points and creates a record of sub-monthly resolution with an average of 33.2 values per year. Mg-SST Calibration To derive a SST record, the speleothem Mg record was interpolated monthly and passed through a Savitzky-Golay filter to remove the influence of weather extrema and/or areas with anomalously high amounts of fluid inclusion before averaging into annual values. The correlation between the annual stalagmite Mg concentration record and the annual coralline SST record 20 over the ~ 200 years of overlap was then maximised by shifting the SST record within dating uncertainties using dynamic time warping (DTW) (see Figure S7). The DTW applied offset was negative (mean of -8.7 years) before 1864 CE, after which it changes to positive (mean of 6.0 years), likely due to the change in trend in both records around this time. Goodkin et al. 20 suggest that undercounting is most prevalent in their record before the mid-1800s when growth rates were extremely slow, supporting the negative DTW offset before 1864. An initial ordinary least squares (OLS) linear regression returned autocorrelated residuals and a non-statistically significant growth rate correlation term. Thus, the regression was corrected using a Cochrane-Orcutt estimation to include an AR(1) term and the growth rate term was excluded 73 . The final linear regression yielded the following correlation (Fig. 3 b): $$\:SST=25.8250\left(\pm\:0.2084\right)-0.0010(\pm\:0.0001)\bullet\:Mg$$ 1 $$\:(2\sigma\:,\:95\%\:conf.,\:{R}^{2}=0.69,\:\:p\ll\:0.0001,\:RMSE=0.5734^\circ\:C)$$ The modelled stalagmite SST record (Table S3) has less variance than the coralline record, likely because of aquifer mixing smoothing the signal (potentially the mixing of diffuse and concentrated flow paths). The signal-dampening effect that the strong correlation between cave temperature and SST would have via the temperature dependency of Mg fractionation may also be responsible for this reduced variance. Additionally, the regression’s predictive power was evaluated using split-period calibration/verification tests in both directions (see Figure S8). The regression passed these tests and thus yielded both a monthly and an annual SST record spanning 558 years, making it longer and higher resolution than many previous SST records from the region (Fig. 3 ). Declarations Data and materials availability All the data needed to evaluate the conclusions in the paper are available in the paper and/or the Supplementary Information. Author contributions statement E.C.G.F. and J.U.L.B. wrote the original draft. F.A.L., I.W.W. and S.R.S. collected the sample. D.C.N. and D.A.R. carried out the uranium-series dating. F.A.L., I.W.W. and C.M. constructed the radiocarbon chronology. R.A.J. and W.M. performed the trace element analysis. E.C.G.F. and J.U.L.B. conducted the layer counts, carried out the data analysis and interpreted the results. All authors contributed to the project, discussed manuscript ideas, and approved the final manuscript. Competing interests The authors declare no competing interests. Acknowledgements We thank the Bermuda Department of Conservation Services for a collecting permit for the speleothem. We thank A. Outerbridge for permission to enter Leamington Cave and G. Nolan for assistance with the cave monitoring equipment. We thank S.F.M Breitenbach for assisting with the age model development and use of the COPRA algorithm. We also thank P. Moffa-Sánchez for the insightful comments and suggestions regarding the interpretation of our record. A European Research Council grant (ERC240167; J.U.L.B.) supported this work. Open access publication costs are supported by Durham University, Open Access Fund. References Ditlevsen, P. & Ditlevsen, S. Warning of a forthcoming collapse of the Atlantic meridional overturning circulation. Nature Communications 14, 1–12, DOI: https://doi.org/10.1038/s41467-023-39810-w (2023). van Westen, R. M., Kliphuis, M. & Dijkstra, H. A. Physics-based early warning signal shows that AMOC is on tipping course. Science Advances 10, eadk1189, DOI: https://doi.org/10.1126/sciadv.adk1189 (2024). Palter, J. B. The Role of the Gulf Stream in European Climate. Annual Review of Marine Science 7, 113–137, DOI:https://doi.org/10.1146/annurev-marine-010814-015656 (2015). Caesar, L., Rahmstorf, S., Robinson, A., Feulner, G. & Saba, V. Observed fingerprint of a weakening Atlantic Ocean overturning circulation. Nature 556, 191–196, DOI: https://doi.org/10.1038/s41586-018-0006-5 (2018). Caesar, L., McCarthy, G. D., Thornalley, D. J. R., Cahill, N. & Rahmstorf, S. Current Atlantic Meridional Overturning Circulation weakest in last millennium. Nature Geoscience 14, 118–120, DOI: https://doi.org/10.1038/s41561-021-00699-z (2021). Seidov, D., Mishonov, A., Reagan, J. & Parsons, R. Resilience of the Gulf Stream path on decadal and longer timescales. Scientific Reports 9, 11549, DOI: https://doi.org/10.1038/s41598-019-48011-9 (2019). Whitney, N. M. et al. Rapid 20th century warming reverses 900-year cooling in the Gulf of Maine. Communications Earth & Environment 3, 179, DOI: https://doi.org/10.1038/s43247-022-00504-8 (2022). Lund, D. C., Lynch-Stieglitz, J. & Curry, W. B. Gulf Stream density structure and transport during the past millennium. Nature 444, 601–604, DOI: https://doi.org/10.1038/nature05277 (2006). Rashid, H., Zhang, Z. W., Piper, D. J. W., Patro, R. & Xu, Y. P. Impact of Medieval Climate Anomaly and Little Ice Age on the Labrador Current flow speed and the AMOC reconstructed by the sediment dynamics and biomarker proxies. Palaeogeography Palaeoclimatology Palaeoecology 620, 111558, DOI: https://doi.org/10.1016/j.palaeo.2023.111558 (2023). Thibodeau, B. et al. Last Century Warming Over the Canadian Atlantic Shelves Linked to Weak Atlantic Meridional Overturning Circulation. Geophysical Research Letters 45, 12376–12385, DOI: https://doi.org/10.1029/2018gl080083 (2018). Baker, A. et al. The Properties of Annually Laminated Stalagmites-A Global Synthesis. Reviews of Geophysics 59, e2020RG000722, DOI: https://doi.org/10.1029/2020RG000722 (2021). Carlson, P. E., Miller, N. R., Banner, J. L., Breecker, D. O. & Casteel, R. C. The potential of near-entrance stalagmites as high-resolution terrestrial paleoclimate proxies: Application of isotope and trace-element geochemistry to seasonally-resolved chronology. Geochimica Et Cosmochimica Acta 235, 55–75, DOI: https://doi.org/10.1016/j.gca.2018.04.036 (2018). Huang, H. M. et al. Seasonal variations in Sr, Mg and P in modern speleothems (Grotta di Ernesto, Italy). Chemical Geology 175, 429–448, DOI: https://doi.org/10.1016/S0009-2541(00)00337-5 (2001). Drysdale, R. et al. Magnesium in subaqueous speleothems as a potential palaeotemperature proxy. Nature Communications 11, 5027, DOI: https://doi.org/10.1038/s41467-020-18083-7 (2020). Baldini, L. M. et al. Regional temperature, atmospheric circulation, and sea-ice variability within the Younger Dryas Event constrained using a speleothem from northern Iberia. Earth and Planetary Science Letters 419, 101–110, DOI: https://doi.org/10.1016/j.epsl.2015.03.015 (2015). Fairchild, I. J. et al. Controls on trace element (Sr-Mg) compositions of carbonate cave waters: implications for speleothem climatic records. Chemical Geology 166, 255–269, DOI: https://doi.org/10.1016/S0009-2541(99)00216-8 (2000). Sinclair, D. J. et al. Magnesium and strontium systematics in tropical speleothems from the Western Pacific. Chemical Geology 294, 1–17, DOI: https://doi.org/10.1016/j.chemgeo.2011.10.008 (2012). Tremaine, D. M. et al. A two-year automated dripwater chemistry study in a remote cave in the tropical south Pacific: Using Cl – as a conservative tracer for seasalt contribution of major cations. Geochimica et Cosmochimica Acta 184, 289–310, DOI: https://doi.org/10.1016/j.gca.2016.03.029 (2016). Draschba, S., Pätzold, J. & Wefer, G. North Atlantic climate variability since AD 1350 recorded in δ 18 O and skeletal density of Bermuda corals. International Journal of Earth Sciences 88, 733–741, DOI: https://doi.org/10.1007/s005310050301 (2000). Goodkin, N. F., Hughen, K. A., Curry, W. B., Doney, S. C. & Ostermann, D. R. Sea surface temperature and salinity variability at Bermuda during the end of the Little Ice Age. Paleoceanography 23, PA3203, DOI: https://doi.org/10.1029/2007pa001532 (2008). Kuhnert, H., Pätzold, J., Schnetger, B. & Wefer, G. Sea-surface temperature variability in the 16th century at Bermuda inferred from coral records. Palaeogeography Palaeoclimatology Palaeoecology 179, 159–171, DOI: https://doi.org/10.1016/S0031-0182(01)00410-2 (2002). Broecker, W. S. Was a change in thermohaline circulation responsible for the Little Ice Age? Proceedings of the National Academy of Sciences of the United States of America 97, 1339–1342, DOI: https://doi.org/10.1073/pnas.97.4.1339 (2000). Lapointe, F. & Bradley, R. S. Little Ice Age abruptly triggered by intrusion of Atlantic waters into the Nordic Seas. Science Advances 7, eabi8230, DOI: https://doi.org/10.1126/sciadv.abi8230 (2021). Schleussner, C. F., Divine, D. V., Donges, J. F., Miettinen, A. & Donner, R. V. Indications for a North Atlantic ocean circulation regime shift at the onset of the Little Ice Age. Climate Dynamics 45, 3623–3633, DOI: https://doi.org/10.1007/s00382-015-2561-x (2015). Vacher, H., Hearty, P., Rowe, M., Curran, H. & White, B. Stratigraphy of Bermuda: nomenclature, concepts, and status of multiple systems of classification. Special Papers-Geological Society of America, 271–294 (1995). Hearty, P. J. & Vacher, H. L. Quaternary Stratigraphy of Bermuda - a High-Resolution Pre-Sangamonian Rock Record. Quaternary Science Reviews 13, 685–697, DOI: https://doi.org/10.1016/0277-3791(94)90099-X (1994). Iliffe, T. M. & Calderón-Gutiérrez, F. Bermuda's Walsingham Caves: A Global Hotspot for Anchialine Stygobionts. Diversity 13, 352, DOI: https://doi.org/10.3390/d13080352 (2021). Peel, M. C., Finlayson, B. L. & McMahon, T. A. Updated world map of the Koppen-Geiger climate classification. Hydrology and Earth System Sciences 11, 1633–1644, DOI: https://doi.org/10.5194/hess-11-1633-2007 (2007). Bermuda Weather Service. Climate Data , available at: https://www.weather.bm/climate.asp (2018). Rayner, N. A. et al. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research-Atmospheres 108, 4407, DOI: https://doi.org/10.1029/2002jd002670 (2003). Michaels, A. F. & Knap, A. H. Overview of the US JGOFS Bermuda Atlantic Time-series Study and the Hydrostation S program. Deep Sea Research Part II: Topical Studies in Oceanography 43, 157–198, DOI: https://doi.org/10.1016/0967-0645(96)00004-5 (1996). Steinberg, D. K. et al. Overview of the US JGOFS Bermuda Atlantic Time-series Study (BATS): a decade-scale look at ocean biology and biogeochemistry. Deep Sea Research Part II: Topical Studies in Oceanography 48, 1405–1447, DOI: https://doi.org/10.1016/S0967-0645(00)00148-X (2001). Booth, B. B., Dunstone, N. J., Halloran, P. R., Andrews, T. & Bellouin, N. Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability. Nature 484, 228–232, DOI: https://doi.org/10.1038/nature10946 (2012). Wild, M. Global dimming and brightening: A review. Journal of Geophysical Research: Atmospheres 114, DOI: https://doi.org/10.1029/2008JD011470 (2009). Thornalley, D. J. R. et al. Anomalously weak Labrador Sea convection and Atlantic overturning during the past 150 years. Nature 556, 227–230, DOI: https://doi.org/10.1038/s41586-018-0007-4 (2018). Cronin, T. M. et al. The Medieval Climate Anomaly and Little Ice Age in Chesapeake Bay and the North Atlantic Ocean. Palaeogeography Palaeoclimatology Palaeoecology 297, 299–310, DOI: https://doi.org/10.1016/j.palaeo.2010.08.009 (2010). Keigwin, L. D. & Pickart, R. S. Slope water current over the Laurentian Fan on interannual to millennial time scales. Science 286, 520–523, DOI: https://doi.org/10.1126/science.286.5439.520 (1999). Joyce, T. M., Kwon, Y. O., Seo, H. & Ummenhofer, C. C. Meridional Gulf Stream Shifts Can Influence Wintertime Variability in the North Atlantic Storm Track and Greenland Blocking. Geophysical Research Letters 46, 1702–1708, DOI: https://doi.org/10.1029/2018gl081087 (2019). Sanchez-Franks, A., Hameed, S. & Wilson, R. E. The Icelandic Low as a Predictor of the Gulf Stream North Wall Position. Journal of Physical Oceanography 46, 817–826, DOI: https://doi.org/10.1175/Jpo-D-14-0244.1 (2016). Taylor, A. H. & Stepens, J. A. The North Atlantic oscillation and the latitude of the Gulf Stream. Tellus Series a-Dynamic Meteorology and Oceanography 50, 134–142, DOI: https://doi.org/10.1034/j.1600-0870.1998.00010.x (1998). Zhang, W. Z., Chai, F., Xue, H. J. & Oey, L. Y. Remote sensing linear trends of the Gulf Stream from 1993 to 2016. Ocean Dynamics 70, 701–712, DOI: https://doi.org/10.1007/s10236-020-01356-6 (2020). Medhaug, I., Langehaug, H. R., Eldevik, T., Furevik, T. & Bentsen, M. Mechanisms for decadal scale variability in a simulated Atlantic meridional overturning circulation. Climate Dynamics 39, 77–93, DOI: https://doi.org/10.1007/s00382-011-1124-z (2012). Chaudhuri, A. H., Gangopadhyay, A. & Bisagni, J. J. Response of the Gulf Stream transport to characteristic high and low phases of the North Atlantic Oscillation. Ocean Modelling 39, 220–232, DOI: https://doi.org/10.1016/j.ocemod.2011.04.005 (2011). Luterbacher, J. et al. Extending North Atlantic Oscillation reconstructions back to 1500. Atmospheric Science Letters 2, 114–124, DOI: https://doi.org/10.1006/asle.2001.0044 (2001). Jia, W. et al. Highly resolved δ 13 C and trace element ratios of precisely dated stalagmite from northwestern China: Hydroclimate reconstruction during the last two millennia. Quaternary Science Reviews 291, 107473, DOI: https://doi.org/10.1016/j.quascirev.2022.107473 (2022). Wassenburg, J. A. et al. Calcite Mg and Sr partition coefficients in cave environments: Implications for interpreting prior calcite precipitation in speleothems. Geochimica Et Cosmochimica Acta 269, 581–596, DOI: https://doi.org/10.1016/j.gca.2019.11.011 (2020). Denniston, R. F. et al. A Last Glacial Maximum through middle Holocene stalagmite record of coastal Western Australia climate. Quaternary Science Reviews 77, 101–112, DOI: https://doi.org/10.1016/j.quascirev.2013.07.002 (2013). Mann, M. E. et al. Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proceedings of the National Academy of Sciences of the United States of America 105, 13252–13257, DOI: https://doi.org/10.1073/pnas.0805721105 (2008). Walczak, I. Holocene climate variability revealed using geochemistry and Computed Tomography scanning of stalagmites from the North Atlantic Basin. Doctoral thesis thesis, Durham University, Durham, United Kingdom, (2016). Baldini, J. U. L. et al. Detecting and quantifying palaeoseasonality in stalagmites using geochemical and modelling approaches. Quaternary Science Reviews 254, 106784, DOI: https://doi.org/10.1016/j.quascirev.2020.106784 (2021). Hoffmann, D. L. et al. Procedures for accurate U and Th isotope measurements by high precision MC-ICPMS. International Journal of Mass Spectrometry 264, 97–109, DOI: https://doi.org/10.1016/j.ijms.2007.03.020 (2007). Cheng, H. et al. Improvements in 230 Th dating, 230 Th and 234 U half-life values, and U–Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth and Planetary Science Letters 371, 82–91, DOI: https://doi.org/10.1016/j.epsl.2013.04.006 (2013). Vermeesch, P. IsoplotR: A free and open toolbox for geochronology. Geoscience Frontiers 9, 1479–1493, DOI: https://doi.org/10.1016/j.gsf.2018.04.001 (2018). Richards, D. A. & Dorale, J. A. Uranium-series chronology and environmental applications of speleothems. Reviews in Mineralogy and Geochemistry 52, 407–460, DOI: https://doi.org/10.2113/0520407 (2003). Wortham, B. E., Banner, J. L., James, E. W., Edwards, R. L. & Loewy, S. Application of cave monitoring to constrain the value and source of detrital 230Th/232Th in speleothem calcite: Implications for U-series geochronology of speleothems. Palaeogeography, Palaeoclimatology, Palaeoecology 596, 110978, DOI: https://doi.org/10.1016/j.palaeo.2022.110978 (2022). Huang, S. et al. An integrated study of constraining the initial 230 Th of a stalagmite and its implications. Quaternary Geochronology 80, 101497, DOI: https://doi.org/10.1016/j.quageo.2024.101497 (2024). Moseley, G. E. et al. Early–middle Holocene relative sea-level oscillation events recorded in a submerged speleothem from the Yucatán Peninsula, Mexico. The Holocene 25, 1511–1521, DOI: https://doi.org/10.1177/0959683615585832 (2015). Beck, J. W. et al. Extremely large variations of atmospheric 14 C concentration during the last glacial period. Science 292, 2453–2458, DOI: https://doi.org/10.1126/science.1056649 (2001). Hoffmann, D. L. et al. Towards radiocarbon calibration beyond 28 ka using speleothems from the Bahamas. Earth and Planetary Science Letters 289, 1–10, DOI: https://doi.org/10.1016/j.epsl.2009.10.004 (2010). Fensterer, C. et al. Cuban stalagmite suggests relationship between Caribbean precipitation and the Atlantic Multidecadal Oscillation during the past 1.3 ka. The Holocene 22, 1405–1412, DOI: https://doi.org/10.1177/0959683612449759 (2012). Rivera-Collazo, I. et al. Human adaptation strategies to abrupt climate change in Puerto Rico ca. 3.5 ka. The Holocene 25, 627–640, DOI: https://doi.org/10.1177/0959683614565951 (2015). Warken, S. F. et al. Persistent link between Caribbean precipitation and Atlantic Ocean circulation during the Last Glacial revealed by a speleothem record from Puerto Rico. Paleoceanography and Paleoclimatology 35, e2020PA003944, DOI: https://doi.org/10.1029/2020PA003944 (2020). Hellstrom, J. U–Th dating of speleothems with high initial 230 Th using stratigraphical constraint. Quaternary Geochronology 1, 289–295, DOI:https://doi.org/10.1016/j.quageo.2007.01.004 (2006). Jamieson, R. A. Trace element geochemistry of Belizean and Bermudan stalagmites: new tools, proxies and applications. Doctoral thesis thesis, Durham University, Durham, United Kingdom, (2017). Evans, D., Erez, J., Oron, S. & Müller, W. Mg/Ca-temperature and seawater-test chemistry relationships in the shallow-dwelling large benthic foraminifera Operculina ammonoides. Geochimica et Cosmochimica Acta 148, 325–342, DOI: https://doi.org/10.1016/j.gca.2014.09.039 (2015). Müller, W., Shelley, M., Miller, P. & Broude, S. Initial performance metrics of a new custom-designed ArF excimer LA-ICPMS system coupled to a two-volume laser-ablation cell. Journal of Analytical Atomic Spectrometry 24, 209–214, DOI: https://doi.org/10.1039/b805995k (2009). Longerich, H., Jackson, S. & Gunther, D. Laser ablation inductively coupled plasma mass spectrometric transient signal accquistion and analyte concentration calculation. Journal of Analytical Atomic Spectrometry 11, 899–904 (1996). Stoll, H. M., Müller, W. & Prieto, M. I-STAL, a model for interpretation of Mg/Ca, Sr/Ca and Ba/Ca variations in speleothems and its forward and inverse application on seasonal to millennial scales. Geochemistry, Geophysics, Geosystems 13, DOI: https://doi.org/10.1029/2012GC004183 (2012). Paton, C., Hellstrom, J., Paul, B., Woodhead, J. & Hergt, J. Iolite: Freeware for the visualisation and processing of mass spectrometric data. Journal of Analytical Atomic Spectrometry 26, 2508–2518, DOI: https://doi.org/10.1039/c1ja10172b (2011). Lechleitner, F. A. et al. A novel approach for construction of radiocarbon-based chronologies for speleothems. Quaternary Geochronology 35, 54–66, DOI: https://doi.org/10.1016/j.quageo.2016.05.006 (2016). Fohlmeister, J. & Lechleitner, F. A. STAlagmite dating by radiocarbon (star): A software tool for reliable and fast age depth modelling. Quaternary Geochronology 51, 120–129, DOI: https://doi.org/10.1016/j.quageo.2019.02.008 (2019). Breitenbach, S. F. M. et al. COnstructing Proxy Records from Age models (COPRA). Climate of the Past 8, 1765–1779, DOI: https://doi.org/10.5194/cp-8-1765-2012 (2012). Cochrane, D. & Orcutt, G. H. Application of Least Squares Regression to Relationships Containing Autocorrelated Error Terms. Journal of the American Statistical Association 44, 32–61, DOI: https://doi.org/10.2307/2280349 (1949). Copernicus Marine Service. ODYSSEA Global Sea Surface Temperature Gridded Level 4 Daily Multi-Sensor Observations, available at: https://doi.org/10.48670/mds-00321 (2024). Rouse, J. R. P. The Provenance and Geochemical Alteration fo Bermudan Eolianites. Doctoral thesis thesis, Queen’s University, Ontario, Canada, (2016). Huang, Y. M. & Fairchild, I. J. Partitioning of Sr 2+ and Mg 2+ into calcite under karst-analogue experimental conditions. Geochemica Et Cosmochimica Acta 65, 47–62, DOI: https://doi.org/10.1016/S0016-7037(00)00513-5 (2001). Chester, R. Marine Geochemistry. John Wiley & Sons (2009). Tierney, J. E. et al. Tropical sea surface temperatures for the past four centuries reconstructed from coral archives. Paleoceanography and Paleoclimatology 30, 226–252, DOI: https://doi.org/10.1002/2014PA002717 (2015). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformation.docx Supplementary Information TableS1.xlsx Table S1 TableS2.xlsx Table S2 TableS3.xlsx Table S3 Cite Share Download PDF Status: Published Journal Publication published 14 Jul, 2025 Read the published version in Communications Earth & Environment → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5973851","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":417099596,"identity":"1f6a6160-07bf-4f32-966c-e562eb4b2df0","order_by":0,"name":"Edward Forman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYJCCD2CSvQFIGFgQpYNxBpjiOQDSIkGKFokEMElYPT/78YsNH/dskzeXfH51w48CCQb+9u4EvFoke3IKG2c8u224c3ZO2c0eoMMkzpzdgFeLwQ2e9Mc8B24zbridk3aDB6jFQCIXvxb7GzyJzUAt9htunkm7+YcYLQYS7AdBWhI33GA/dpsoWyTO5DA2zjhwO3nDmRy22zIGEjwE/cLffvxhw4cDt203HD/+7OabPzZy/O29+LUAo9AAhcFDQDkIsD9AZ4yCUTAKRsEoQAUAksZPgmvcVmMAAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0000-7350-1920","institution":"Durham University","correspondingAuthor":true,"prefix":"","firstName":"Edward","middleName":"","lastName":"Forman","suffix":""},{"id":417099597,"identity":"ac76dc1c-0c34-4399-902d-e52e60164045","order_by":1,"name":"James Baldini","email":"","orcid":"","institution":"Durham University","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"","lastName":"Baldini","suffix":""},{"id":417099598,"identity":"ef141d4e-6d12-4997-9d3a-8da0d9e0902d","order_by":2,"name":"Robert A. Jamieson","email":"","orcid":"https://orcid.org/0000-0001-6385-3930","institution":"University of Leeds","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"A.","lastName":"Jamieson","suffix":""},{"id":417099599,"identity":"43d0a088-96ee-46d0-ae53-6bc5136d5165","order_by":3,"name":"Franziska Lechleitner","email":"","orcid":"https://orcid.org/0000-0002-1697-0429","institution":"University of Bern","correspondingAuthor":false,"prefix":"","firstName":"Franziska","middleName":"","lastName":"Lechleitner","suffix":""},{"id":417099600,"identity":"1f4f3d97-b4ac-40b3-8332-ddd3de20e37b","order_by":4,"name":"Izabela Walczak","email":"","orcid":"","institution":"Durham University","correspondingAuthor":false,"prefix":"","firstName":"Izabela","middleName":"","lastName":"Walczak","suffix":""},{"id":417099601,"identity":"94d179b9-83ed-4faa-a3ab-432e2df3f777","order_by":5,"name":"Dan Nita","email":"","orcid":"","institution":"Babes-Bolyai University","correspondingAuthor":false,"prefix":"","firstName":"Dan","middleName":"","lastName":"Nita","suffix":""},{"id":417099602,"identity":"1768f0a0-313b-4838-a3ed-14fa9673ada8","order_by":6,"name":"S. R. Smith","email":"","orcid":"https://orcid.org/0000-0002-3078-0015","institution":"Bermuda Aquarium, Museum and Zoo","correspondingAuthor":false,"prefix":"","firstName":"S.","middleName":"R.","lastName":"Smith","suffix":""},{"id":417099603,"identity":"2942752f-90b4-428d-85b3-6ccc1d2414e5","order_by":7,"name":"David Richards","email":"","orcid":"https://orcid.org/0000-0001-8389-8079","institution":"University of Bristol","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Richards","suffix":""},{"id":417099604,"identity":"ad28ed35-0959-45fb-a69f-dd73c0dcfdea","order_by":8,"name":"Lisa Baldini","email":"","orcid":"","institution":"Teesside University","correspondingAuthor":false,"prefix":"","firstName":"Lisa","middleName":"","lastName":"Baldini","suffix":""},{"id":417099605,"identity":"b66eb7d8-edfb-4244-9cf1-c7524c7687d6","order_by":9,"name":"Cameron McIntyre","email":"","orcid":"https://orcid.org/0000-0001-8517-9836","institution":"SUERC","correspondingAuthor":false,"prefix":"","firstName":"Cameron","middleName":"","lastName":"McIntyre","suffix":""},{"id":417099606,"identity":"3630c024-dc85-4cfe-a44e-def36fbf1640","order_by":10,"name":"Wolfgang Muller","email":"","orcid":"https://orcid.org/0000-0001-5368-8011","institution":"Goethe University Frankfurt","correspondingAuthor":false,"prefix":"","firstName":"Wolfgang","middleName":"","lastName":"Muller","suffix":""}],"badges":[],"createdAt":"2025-02-06 13:15:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5973851/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5973851/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s43247-025-02446-3","type":"published","date":"2025-07-14T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":79252951,"identity":"3187aa0c-8aa7-4c37-aafa-1537bb99b019","added_by":"auto","created_at":"2025-03-26 08:27:10","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":430184,"visible":true,"origin":"","legend":"\u003cp\u003eThe study site and collected sample. \u003cstrong\u003ea\u003c/strong\u003e Leamington Cave survey map by members of the Bermuda Cave Diving Association. Purple circle denotes the location of the stalagmite. \u003cstrong\u003eb\u003c/strong\u003e Scan of stalagmite BER-SWI-13 with positions of dating powders (\u003csup\u003e14\u003c/sup\u003eC, red; U-Th, blue) and isotope milling track (grey) annotated (latter not used in this study)\u003csup\u003e 49\u003c/sup\u003e. Laser ablation transects were performed on the facing half which is not shown\u003csup\u003e64\u003c/sup\u003e. \u003cstrong\u003ec\u003c/strong\u003e Bermuda climatology. Sea surface temperature (red; 1950-2018), precipitation (blue; 1949-2018) and wind speed (green; 1975-1995) averaged monthly from daily resolution\u003csup\u003e29\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/a6474051e51f6636b15f719f.png"},{"id":79252953,"identity":"2cb43519-b3e1-4a37-afec-32d6a863d861","added_by":"auto","created_at":"2025-03-26 08:27:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":190624,"visible":true,"origin":"","legend":"\u003cp\u003eAge model for stalagmite BER-SWI-13. \u003cstrong\u003ea\u003c/strong\u003e Development of the BER-SWI-13 chronology, comparing the preliminary cycle count (green) and the radiocarbon (black) to the final cycle count (blue) developed from both. Grey shaded area signifies the 2σ confidence interval of the radiocarbon chronology. Black areas (left-hand side) mark depths modelled to the radiocarbon mean. \u003cstrong\u003eb\u003c/strong\u003e Radiocarbon data recorded in the stalagmite (purple) compared to the U-Th dates (cyan). The black arrow indicates the trajectory of the radiocarbon data if only influenced by the Suess Effect. Note that the radiocarbon ages here have not been corrected for dead carbon fraction and that the U-Th dates use (\u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e232\u003c/sup\u003eTh)\u003csub\u003einitial\u003c/sub\u003e = 5 ± 2.5. \u003cstrong\u003ec\u003c/strong\u003e Mean growth rates based on radiocarbon (black, 0.345 mm yr\u003csup\u003e-1\u003c/sup\u003e), preliminary (green, 0.373 mm yr\u003csup\u003e-1\u003c/sup\u003e), and final (blue, 0.360 mm yr\u003csup\u003e-1\u003c/sup\u003e) chronologies.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/13e2c0609b22b4f268a9e306.png"},{"id":79253418,"identity":"9d1e702c-c55f-4776-bdf2-49e7447ccbe2","added_by":"auto","created_at":"2025-03-26 08:35:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":386740,"visible":true,"origin":"","legend":"\u003cp\u003eThe stalagmite-derived SST reconstruction. \u003cstrong\u003ea\u003c/strong\u003e Cochrane-Orcutt regression model output (1456-2013 CE; 558 years) at both a monthly (light blue) and annual (dark blue) resolution plotted with a coralline SST record (1782-1998 CE; 217 years) for Bermuda\u003csup\u003e20\u003c/sup\u003e. \u003cstrong\u003eb\u003c/strong\u003e post-DTW coralline SSTs regressed against stalagmite Mg concentrations. \u003cstrong\u003ec\u003c/strong\u003e Comparison of the stalagmite Mg concentration-based record to five coralline palaeo-ocean temperature records for Bermuda\u003csup\u003e19-21\u003c/sup\u003e. Note that the Kuhnert et al.\u003csup\u003e 21\u003c/sup\u003e record was averaged to annual spacing from monthly to reduce noise and that axes have been inverted to match the direction of the SST axis.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/7a963ecceb7bec70c99d05d2.png"},{"id":79253421,"identity":"8532522d-7946-461b-a1bf-12ab6211693e","added_by":"auto","created_at":"2025-03-26 08:35:11","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":418028,"visible":true,"origin":"","legend":"\u003cp\u003eComparison between BER-SWI-13 SST and other reconstructions from the North Atlantic. (top) Map of North Atlantic regional SSTs\u003csup\u003e74\u003c/sup\u003e. Coloured circles on the map denote the sites of the records in the panel below. (bottom) Palaeoclimatic records for the study period. \u003cstrong\u003ea\u003c/strong\u003e Percentage abundance of the polar species N. pachyderma (sinistral) in a marine sediment core (OCE326-MC13) from the Northwest Atlantic\u003csup\u003e35\u003c/sup\u003e. \u003cstrong\u003eb\u003c/strong\u003e Chesapeake Bay reconstructed SST data with a 15-point moving average\u003csup\u003e36\u003c/sup\u003e. \u003cstrong\u003ec\u003c/strong\u003e The annual SST model for Bermuda with a 15-point moving average. \u003cstrong\u003ed\u003c/strong\u003e Mean sortable silt (\u003cem\u003eSS\u003c/em\u003e\u003csup\u003e\u003cem\u003e--\u003c/em\u003e\u003c/sup\u003e) grain size data from three cores (red, 56JPC; orange, 48JPC; grey, MO217)\u003csup\u003e 9,35\u003c/sup\u003e. \u003cstrong\u003ee\u003c/strong\u003e Gulf Stream transport estimates in the Florida Straits\u003csup\u003e8\u003c/sup\u003e. \u003cstrong\u003ef\u003c/strong\u003e Normalized winter North Atlantic Oscillation index data with 15-point moving average\u003csup\u003e44\u003c/sup\u003e. \u003cstrong\u003eg\u003c/strong\u003e Northern Hemisphere temperature anomaly reconstruction with 95% confidence interval for proxy-derived estimates\u003csup\u003e48\u003c/sup\u003e. Grey area denotes the temporal extent of the LIA.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/6e94c2ec72ebf04fb69c1250.png"},{"id":86742534,"identity":"ebe330a7-c3ed-4a50-af64-7b9334f0b04e","added_by":"auto","created_at":"2025-07-15 07:11:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2175636,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/5051a676-b767-4e70-8c7e-c90d9d3398fd.pdf"},{"id":79252955,"identity":"8291f92d-6a9c-4d48-b7f7-bc7be8dd892c","added_by":"auto","created_at":"2025-03-26 08:27:10","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1457066,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/d34bea7364482794e3cae7ee.docx"},{"id":79252949,"identity":"9bca40a0-ed9f-46dd-b919-b1f545842e2c","added_by":"auto","created_at":"2025-03-26 08:27:10","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":17530,"visible":true,"origin":"","legend":"\u003cp\u003eTable S1\u003c/p\u003e","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/5ce078e508880008fe01179f.xlsx"},{"id":79253419,"identity":"f21ece5c-b8b6-4b09-b5e1-90dc04bc743f","added_by":"auto","created_at":"2025-03-26 08:35:10","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":12162,"visible":true,"origin":"","legend":"\u003cp\u003eTable S2\u003c/p\u003e","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/bd8bef5c873502a0035fa3a9.xlsx"},{"id":79252957,"identity":"f8f16f1f-03cf-4155-994b-3f582c13d1a0","added_by":"auto","created_at":"2025-03-26 08:27:10","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":546914,"visible":true,"origin":"","legend":"Table S3","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5973851/v1/f836b0547fcd3cc25874047a.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Monthly-Resolved Cave Proxy Evidence for Northward Gulf Stream Migration During the Little Ice Age","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe system of ocean currents collectively known as the Atlantic Meridional Overturning Circulation (AMOC) are a major driver of climate variability. The AMOC is a highly non-linear system and is thought to reach its critical threshold this century\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e, which would have serious ramifications for regional climates such as up to ~\u0026thinsp;15\u0026deg;C temperature reductions in Northwest Europe\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Thus, understanding its response to climate forcing is critical. One key component of the AMOC is the Gulf Stream (GS), which transports warm water from the Gulf of Mexico northward towards Northwest Europe. Due to its presence, at least in part, winter surface air temperatures in western Europe are up to 10\u0026deg;C higher than the zonal mean at comparable latitudes\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. GS path and AMOC strength are intrinsically linked. A weaker AMOC reduces the Deep Western Boundary Current (DWBC), weakens the northern recirculation gyre, and shifts the GS separation point northward\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Anthropogenic warming has weakened the AMOC to perhaps its weakest state over the past millennium, inducing a northward migration in the Gulf Stream North Wall\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Temperature records from the Gulf of Maine have traced the effects of GS positional shifting over the last 100 years\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, with northward movement causing regional sea surface temperatures (SSTs) to rise. However, on longer timescales that are critical for understanding sensitivity to global temperature change, precise GS position remains elusive.\u003c/p\u003e \u003cp\u003eThe Little Ice Age (LIA) represents a key time interval for understanding how GS positioning and dynamics responds to climate change because it was characterised by extended Northern Hemisphere (NH) cold intervals between about 1300 and 1850 CE. Proxy evidence suggests the LIA featured a GS that was approximately 10% weaker and positioned further south, likely due to a stronger Labrador Current (LC) and enhanced recirculation gyre\u003csup\u003e\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. The lack of instrumental records from the LIA means that palaeoclimate reconstructions are crucial for inferring GS state and position.\u003c/p\u003e \u003cp\u003eStalagmites are increasingly providing important constraints on the interpretation of palaeoclimates, particularly due to their propensity to provide accurate and precise chronologies and potential for continuous, high-resolution records, in some cases at sub-monthly timescales. Strong environmental seasonality can result in visible or geochemical laminations, offering high-precision chronological control, complementing any radiometric (e.g., U-Th or \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC) determinations (e.g., \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e). Magnesium concentrations in stalagmites can exhibit a seasonal signal (e.g., \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e) that can be employed both as a chronological tool (via seasonal cyclicity) and as a climate proxy (through Mg concentrations). Stalagmite Mg concentrations can record temperature based on the temperature dependency of Mg\u0026rsquo;s partition coefficient (D\u003csub\u003eMg\u003c/sub\u003e) in calcite (e.g., \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e). However, sea spray contributions to rainfall may also affect dripwater especially in island karst systems and, consequently, stalagmite Mg concentrations\u003csup\u003e\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBermuda\u0026rsquo;s position within the recirculation gyres south of the GS makes it an ideal site to capture the effects of a southward GS shift, and the presence of extensive cave systems offers a rich speleothem archive for climate reconstruction. However, while other SST reconstructions for Bermuda exist (e.g.,\u003csup\u003e\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e), a continuous annual SST record for Bermuda spanning more than the last three centuries has yet to be realised. Understanding the palaeoclimatic changes experienced in Bermuda on this temporal scale will provide new insights into how the LIA, with its clearly expressed NH climate anomaly, affected GS position and strength\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan additionalcitationids=\"CR23\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHere, we use Mg concentrations from a stalagmite from Bermuda to construct a monthly palaeotemperature record through sea spray contributions to dripwater, spanning 558-years (1456\u0026ndash;2013 CE). We derived a Mg-SST calibration based on a Cochrane-Orcutt regression between our record and a previously developed coral Sr/Ca-SST record\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. The resulting reconstruction yields SSTs covering most of the LIA and reveals GS migration patterns over the past 558 years at very high temporal resolution and chronological accuracy.\u003c/p\u003e\n\u003ch3\u003eSite Description\u003c/h3\u003e\n\u003cp\u003eLeamington Cave in north-eastern Bermuda (32\u0026deg; 20\u0026rsquo; 31.64\u0026rdquo; N, 64\u0026deg; 42\u0026rsquo; 30.93\u0026rdquo; W; entrance at 18 m a.s.l.) is a privately-owned former show cave connected to the ocean via flooded conduits and a large cave pool (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). The cave contains many actively growing speleothems above the high tide line, with cave monitoring data revealing a same-day drip rate response to rainfall that suggests residence times capable of preserving a seasonal signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). The Bermuda Islands consist of a carbonate platform above a volcanic seamount and are comprised almost entirely of aeolianites and palaeosols\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. The Walsingham Formation, an aeolianite limestone unit (the oldest limestone unit, deposited approximately 1.1\u0026ndash;0.8 Ma), hosts most of Bermuda\u0026rsquo;s caves (including Leamington Cave) and outcrops along the eastern border of the Harrington Sound and the western edge of Castle Harbour\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBermuda has a humid subtropical climate (Cfa in K\u0026ouml;ppen-Geiger classification\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e) and lies within the North Atlantic hurricane belt, with the hurricane season running from June to November each year. Meteorological data\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e indicate that mean sea temperatures are lowest in February (17.8\u0026deg;C) and peak in August (28.5\u0026deg;C) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). May historically experiences the lowest mean monthly rainfall (86.7 mm), whereas October has the highest (161.6 mm) \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. The hurricane season accounts for a mean of 55% of annual rainfall. Except for September and October, when winds prevail from the east, winds typically come from SW and SSW, reaching a maximum in January (6.01 m/s) and a minimum in September (4.49 m/s) \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Results and Discussion","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eBermuda Temperature Trends\u003c/h2\u003e \u003cp\u003eOur SST reconstruction for Bermuda is derived from stalagmite BER-SWI-13 using an age model constructed from a Mg cycle count. The cycle count is modelled to a radiocarbon chronology in intervals where cyclicity is less pronounced and is better constrained and more precise than both the radiocarbon and U-Th chronologies (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; full details of our chronology and age model construction are discussed in Materials and Methods). By virtue of the inverse relationship between wind speed and SST, our monthly-resolved Mg record serves as an indirect proxy for SST via sea spray contributions to dripwater. SST estimates were obtained through calibrating the stalagmite Mg record to a previously published coralline SST record (r = -0.51, p\u0026thinsp;\u0026lt;\u0026thinsp;\u0026lt;\u0026thinsp;0.001; 1782\u0026ndash;1998; \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, a and b). The resultant record spans 558 years (1456 to 2013 CE), making it one of the longest continuous high-resolution records for the region. The stalagmite-derived SST reconstruction exhibits notable agreement with both five previous coralline palaeotemperature records from offshore Bermuda and reanalysis data (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec and S8), illustrating its reconstructive power\u003csup\u003e\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. The record presented here cannot be confidently compared to observational SST data (starting in 1950) due to anthropogenic influences on speleothem deposition post-1967. However, because the coralline SST record used for our calibration is derived from a regression with instrumental data, its significant correlation with the stalagmite reconstruction reinforces the accuracy of our record\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe new stalagmite SST record (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) suggests that prior to the 18th century, SSTs were relatively stable, fluctuating within a range of 1.4\u0026deg;C around a mean of 23.8\u0026deg;C (compared to the 1950\u0026ndash;2018 average of 22.7\u0026deg;C; \u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e). A cooling trend lasting\u0026thinsp;~\u0026thinsp;130 years began in ~\u0026thinsp;1720 CE at a rate of approximately \u0026minus;\u0026thinsp;0.13\u0026deg;C per decade and ended with the onset of the post-1850 warming. The cooling period exhibits high frequency interannual oscillations with decadal-scale fluctuations of ~\u0026thinsp;1.5\u0026deg;C (a range of 3.2\u0026deg;C around a mean of 22.9\u0026deg;C), compared to fluctuations of less than 0.5\u0026deg;C over the same timescale before ~\u0026thinsp;1720 CE. The interannual variance also decreases after 1967 CE, but this is likely the product of anthropogenic impacts on the site\u0026rsquo;s hydrology and/or a shallower mixed layer depth\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e rather than changes in climate as the annual signal across that interval is also less pronounced, despite no change in the annual climate signal. Both the reconstruction presented here and the Goodkin et al. \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e coralline record indicate a warming of ~\u0026thinsp;0.11\u0026deg;C per decade starting around 1850 CE. The records also both suggest a brief cooling interval starting in the late 19th century which lasted around two decades before warming resumed. The relatively cooler mean SSTs of the 18th and 19th century when compared to those pre-1720 CE are likely explained by wind speed changes artificially increasing estimates and/or aerosol forcing in the NH reducing SSTs\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eInfluences on Reconstructed SST Variability\u003c/h3\u003e\n\u003cp\u003eThe SST record reconstructed here exhibits the effects of both changes in ocean currents and anthropogenic climate change. Enhanced radiative forcing from additional greenhouse gases and rising total solar irradiance at the end of the LIA are likely to be the cause of the increased temperatures observed after 1850 CE (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, a and g). Although, NH temperature anomalies remained largely low before the mid-1800s (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eg), the unusually high SSTs reconstructed for Bermuda prior to 1720 CE suggest conditions influenced primarily by natural factors, without a significant anthropogenic impact.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eNeogloboquadrina pachyderma\u003c/em\u003e abundances in a sediment core from the northwest Atlantic\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e suggest a warming trend beginning in the 18th century (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). Similarly, a Mg/Ca-derived temperature record for Chesapeake Bay\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e demonstrates moderate SSTs during the LIA before beginning to increase in the mid-1700s (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). However, our reconstruction from Bermuda (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec) suggests regional ocean temperatures were high during the oldest part of the record (1456\u0026ndash;1720 CE) until a cooling trend ensued around 1720 CE lasting approximately 130 years. The opposing temperature trends suggest a \u0026lsquo;thermal seesaw\u0026rsquo; between the records, whereby an increase in SSTs at lower latitudes leads to a decrease at higher latitudes. This likely reflects southward GS migration during the LIA causing the observed rise in regional Bermudan SSTs compared to those seen at higher latitudes. We propose that a combination of weakened transport, a stronger LC and DWBC, and a sustained negative North Atlantic Oscillation (NAO) phase caused the GS separation point to be further south during the LIA, possibly reducing subtropical mode water potential vorticity. Keigwin and Pickart\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e suggested that northward slope water current migration during the LIA was responsible for an observed warming in several cores south of Newfoundland, but our SST record indicates a broader antiphase system suggesting the northward ocean current shift was more widespread.\u003c/p\u003e \u003cp\u003eGS, DWBC and LC flow-speed reconstructions further support this proposed GS drift (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, d to e). Thornalley et al. \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e showed the DWBC was stronger towards the end of the LIA using two mean sortable silt (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{SS}\\)\u003c/span\u003e\u003c/span\u003e) records from sediment cores off Cape Hatteras (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). A DWBC weakening from about 1750 CE also supports the notion of a northward shift in the GS separation point starting around the time. Rashid et al. \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e found the LC to be strongest during the LIA (over the last 1.45 ka) using \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\stackrel{-}{SS}\\)\u003c/span\u003e\u003c/span\u003e data from a core off the southeastern Grand Banks (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed), which would also have caused the GS to separate from the coast further south\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Lund et al. \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e reconstructed Florida Current transport over the last millennium using sediment cores near the Dry Tortugas and Great Bahama Bank (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ee). From this reconstruction they estimate that the GS was systematically weaker during the LIA by approximately 10%. Given a stronger subpolar gyre at the time this implies that the GS would have moved south and reduced northward heat transport\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFurthermore, studies have also found a link between the GS and the NAO, an atmospheric circulation phenomenon associated with changes in the surface westerlies across the North Atlantic\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. The NAO index is defined by the normalised pressure difference between the Icelandic Low and the Azores High and modulates mid-latitude westerly strength as well as heat, freshwater, and momentum fluxes in the North Atlantic (e.g., \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e). Studies have linked the NAO to changes in both North Atlantic Deep Water formation and Labrador Sea convection\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. During a more negative NAO phase, decreased heat loss, higher precipitation, and weaker winds reduce convection in the Labrador Sea, corresponding to a weaker AMOC and in turn a more southerly GS path\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. This is evident in current GS path migration (1993\u0026ndash;2016), which may reflect an increasingly negative NAO (e.g., \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e). Hence, the persistent negative NAO index values pre-1700 observed in the Luterbacher et al. \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e winter NAO reconstruction also corroborate a southward GS during the LIA (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef).\u003c/p\u003e \u003cp\u003eThe low Mg concentrations pre-1700 in Bermuda may be also attributable to weaker westerlies during an extended negative NAO phase. A weakening of the westerlies at this time could have reduced sea spray over the site and thus led to less Mg incorporation into the stalagmite. These weaker winds are evident in the relatively low implied seawater contribution from the Mg and Sr concentrations at the time (~\u0026thinsp;4.97% compared to the mean\u0026thinsp;~\u0026thinsp;6.10%; Figure S6). Considering that a transition in the 18th century can also be seen in sediment core-derived temperature and flow speed reconstructions (e.g., \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e) there is likely a direct temperature effect in addition to any changes in wind speed. Thus, the GS likely did migrate at this time, but the potential for reduced wind speed over Bermuda to exaggerate the reconstructed SST signal complicates the quantification of the exact southward positional extent of the GS movement.\u003c/p\u003e \u003cp\u003eSpeleothem Mg concentrations are sensitive to the effects of prior calcite precipitation (PCP) and are often used as a proxy for rainfall amount (e.g., \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e). However, areas with evenly distributed rainfall throughout the year usually do not exhibit a strong PCP effect\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. Therefore, a change in cave processes (such as the degree of water-rock interactions or PCP) is unlikely to be the primary driver of Mg concentration variability as they are controlled by broader climatic factors, and only the period after 1967 exhibits the effects of PCP following anthropogenic impacts on the cave system (Figure S6). For this reason, although cave processes or the amount of fluid inclusions may reduce the accuracy of the temperature estimates, the trends exhibited here are likely not exclusive to the BER-SWI-13 record and reflect broader regional trends.\u003c/p\u003e \u003cp\u003eAssuming our SST estimates reflect exclusively temperature change without amplification due to wind speed, the 1\u0026ndash;2\u0026deg;C temperature change over 130 years suggests a moderately fast rate of GS migration when considering the relatively shallow nearby SST gradient in the modern North Atlantic (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and S9), implying that GS position is reasonably sensitive to forcing. Observational data tracking the current northward GS migration detected a shift in the GS north wall of ~\u0026thinsp;1.5\u0026deg; in 52 years (1965\u0026ndash;2017) between 50\u0026deg;W and 40\u0026deg;W\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Therefore, with the continuation of current trends in the GS and AMOC, the sensitivity of regional temperatures to GS path drift could have serious ramifications for future regional climates, ecosystems, and extreme weather events.\u003c/p\u003e \u003cp\u003eOur study reconstructed regional SST using a stalagmite from Bermuda via sea spray contributions to dripwater. A Mg cycle count, modelled to a radiocarbon chronology, yielded a high-precision age model and a monthly-resolved Mg record spanning five centuries (1456\u0026ndash;2013 CE). Palaeo-oceanographic temperature estimates were derived from an indirect magnesium-temperature calibration via a connection to wind speed, which modelled the Mg concentration record to a previously published coralline proxy SST record\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. The resultant record indicates temperatures before the early 1700s were consistently high, contrary to more Northerly regional reconstructions and the NH average\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. We propose this \u0026lsquo;thermal seesaw\u0026rsquo; pattern suggests that the GS migrated northwards towards the end of the LIA following a prior period characterised by a more southerly GS path. Flow speed reconstructions for the GS, DWBC and LC support this interpretation\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, providing evidence of weakening southward currents in the 18th century from strong conditions during the LIA. Further research should focus on quantifying the rates of positional change seen here, to gain a better understanding of GS sensitivity to additional forcing as well as its subsequent manifestations and effects. These results provide evidence of a moderately fast natural GS migration during the LIA. Previous research has proposed changes in ocean circulation towards the end of the LIA (e.g., \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e), and our high-resolution SST reconstruction from Bermuda reinforces this concept.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStalagmite BER-SWI-13\u003c/h2\u003e \u003cp\u003eBER-SWI-13 is a 192 mm long stalagmite that grew in Leamington Cave, Bermuda (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, a to b). The stalagmite was selected for its favourable internal structure and active drip hydrology, which is responsive to rain events\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e and plots as a near-ideal seasonal flow\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Because cave monitoring data revealed a same-day drip rate response to rainfall events (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e, the residence time of the seawater signal is sufficiently low to accurately preserve hydrology-induced seasonality at a monthly-resolution\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. The cave pool\u0026rsquo;s connection to the ocean promotes daily cave ventilation by tidal flushing, suggesting the stalagmite never experienced condensation corrosion due to CO\u003csub\u003e2\u003c/sub\u003e buildup\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Additionally, monitoring data shows daily cave temperature is positively correlated to daily mean SST (r\u0026thinsp;=\u0026thinsp;0.83, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01) (Figure S2) \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eU-series and Radiocarbon Dating\u003c/h2\u003e \u003cp\u003eReconnaissance CT scanning determined the position of stalagmite BER-SWI-13\u0026rsquo;s central growth axis and informed its sectioning. Cut sections were polished and then cleaned in an ultrasonic bath of deionised water (see Walczak\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e for full methods). Visual examination after sectioning showed no signs of growth hiatuses (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Sixteen\u0026thinsp;~\u0026thinsp;400 mg carbonate powder samples (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) were milled along the stalagmite growth axis. These were dated using multi-collector inductively coupled plasma mass spectrometric (MC-ICPMS) U-Th methods at the facilities of Bristol Isotope Group (BIG), University of Bristol, and University of St. Andrews Isotope Geochemistry (STAiG). U and Th separation and measurement protocols described in Hoffmann et al. \u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e were followed, where CRM U112a and an inhouse Th standard (Teddii) was used for bracketing the samples and correcting for instrumental drift in mass bias, SEM yield. We adopt relevant decay constants from Cheng et al. \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. U-series data was analysed using the IsoplotR software\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eU-Th ages for material that is both young (i.e., \u0026lt; 1 ka) and has low \u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e232\u003c/sup\u003eTh ratios are poorly constrained because there is no prior knowledge of the \u003csup\u003e230\u003c/sup\u003eTh/\u003csup\u003e232\u003c/sup\u003eTh ratios at the time of calcite precipitation. Because of this, we attempted to assess the initial Th ratio using isochron methods for two of the milled samples (BER-SWI-UTh-13 and \u0026minus;\u0026thinsp;15) by performing replicate analysis (~\u0026thinsp;100 mg sub-samples). However, there was insufficient \u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e232\u003c/sup\u003eTh ratio variation in the two sets of replicate analysis to determine useful estimates of the initial \u003csup\u003e230\u003c/sup\u003eTh/\u003csup\u003e232\u003c/sup\u003eTh component.\u003c/p\u003e \u003cp\u003eMost researchers use an estimate of initial Th assuming it is equivalent to bulk Earth silicate composition (\u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e232\u003c/sup\u003eTh = 0.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4). However, there are many examples of elevated values of this ratio where either carbonate dust with high U/Th or hydrogenous elements in cave drip waters have probably influenced the isotopic composition: see review by Richards and Dorale\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e and recent papers by Wortham et al. \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e and Huang et al. \u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Other speleothem samples from the North Atlantic and Caribbean region, including the Bahamas, Yucatan, and Puerto Rico, have been similarly affected\u003csup\u003e\u003cspan additionalcitationids=\"CR58 CR59 CR60 CR61\" citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn the absence of sufficiently robust estimates of the initial Th ratio via isochron methods, we considered the principle of stratigraphic constraint (see \u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e) and adopted an elevated value ((\u003csup\u003e238\u003c/sup\u003eU/\u003csup\u003e232\u003c/sup\u003eTh)\u003csub\u003einitial\u003c/sub\u003e = 5 \u0026plusmn; 2.5) that eliminates the major age inversion at the top of the sample (Figure S3) that is evident when using the bulk Earth initial Th ratios. Using this elevated value also shifts the basal age (BER-SWI-UTh-16) younger and would suggest a growth rate for the oldest material that is closer to that exhibited for the rest of the speleothem. One anomalous age remains for the sample (BER-SWI-UTh-5), which is likely to have been affected by post-depositional alteration.\u003c/p\u003e \u003cp\u003eAdditionally, twenty-five 8\u0026ndash;9 mg carbonate powder samples were milled for \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC analysis (Table S2) using a semi-automatic high-precision drill (Sherline 5400 Deluxe) at ETH Zurich. The \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC samples were milled in increments of 5 mm for the top 50 mm, and 10 mm for the remaining 140 mm\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. The top\u0026thinsp;~\u0026thinsp;0.1 mm was discarded for each \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC sample. Potential contamination during sub-sampling was minimised by cleaning all tools with methanol and drying with compressed air in between samples. Carbonate samples were prepared for accelerator mass spectrometry (AMS) analysis by graphitisation at the Laboratory for Ion Beam Physics (LIP) of ETH Zurich. The samples were processed on an automatic graphitisation system coupled to a carbonate-handling system (CHS-AGE, Ionplus, Switzerland) by acidification to CO\u003csub\u003e2\u003c/sub\u003e using 2 mL of 85% H\u003csub\u003e3\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e and followed by graphitisation using excess H\u003csub\u003e2\u003c/sub\u003e over Fe catalyst. Radiocarbon contents were measured on a MICADAS accelerator mass spectrometer (Ionplus) following standard protocols for quality and blank assessment: oxalic acid II (NIST SRM 4990C) was used as the normalising standard, measured to a precision of better than 2\u0026permil;. We used IAEA-C1 as a blank and IAEA-C2 and a modern coral standard as secondary standards. The procedural blank was established using a \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC-free stalagmite (MAW-1, ~\u0026thinsp;170 kyr old).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eTrace Element Analysis\u003c/h3\u003e\n\u003cp\u003eTrace element concentrations were measured using the prototype RESOlution M-50 excimer (193 nm) laser-ablation system with a two-volume laser-ablation cell coupled to an Agilent 7500ce/cs quadrupole ICPMS at Royal Holloway University, London\u003csup\u003e\u003cspan additionalcitationids=\"CR65\" citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAblation tracks were pre-ablated to remove any superficial contamination and measured using a 140 by 10 \u0026micro;m rectangular laser slit across 50 mm sectioned blocks on the opposite half to the carbonate sampling. Sectioning was slightly off perpendicular to the growth direction so that overlaps in transects could compensate for any missing data between sections. A 15 Hz laser repetition rate and a laser fluence of ~\u0026thinsp;4 J/cm\u003csup\u003e2\u003c/sup\u003e with a stage scan speed of 10 \u0026micro;m s\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e of the LA cell were used during the continuous main track measurement\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. Speleothem analyses were bracketed by analyses of NIST 612, NIST 610 for quantification using \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003eCa as internal standard\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, and MACS3 standards for accuracy control (see \u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eThe resultant data were then reduced using the lolite software package, using NIST 610/612 standards for external standardisation\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eBER-SWI-13 Chronology\u003c/h3\u003e\n\u003cp\u003eThe U-Th chronology suggests an unusual range in speleothem growth rate for a humid subtropical cave environment (65 to 1636 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e; where the 90th percentile of annual layers in a global review\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e is 404 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e). This variability is most pronounced between depths from top of 26 and 35 mm where the growth rate implied from U-Th dating suddenly rises from 233 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e to 1636 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e for 5.5 years before returning to 196 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e, despite no hiatuses being observed. Additionally, the carbonate samples returned consistently low U concentrations of 83\u0026ndash;136 ng/g, except for two samples that are 299 and 362 ng/g. The low U concentrations and unrealistically high variations in growth rate suggest that a chronology based solely on U-Th dating would yield inaccurate results. For this reason, we use the U-Th chronology only as confirmation when developing an alternative chronology based on radiocarbon dating and annual Mg cycle counting. First, we developed an \u0026lsquo;average\u0026rsquo; chronological model using the radiocarbon-based technique presented in Lechleitner et al. \u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e and refined in Fohlmeister and Lechleitner\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. This modelling approach uses the radiocarbon data to construct a best fit linear growth rate model, taking into account atmospheric variability in \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC and potential variability in dead carbon fraction. The chronology produced this way is anchored to a point of known age, which in our case is given by the clear increase in \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003eC indicating the rise of the atmospheric bomb spike at 1955 CE (at a depth of 10.179 mm) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The stalagmite collection year (2013 CE) provides a second chronological anchor. However, owing to possible localised anthropogenic influences on the stalagmite\u0026rsquo;s drip hydrology (e.g., land use change above the cave related to the development of the cave for tourism) starting\u0026thinsp;~\u0026thinsp;1970 CE, we do not consider data post-1970 in our interpretations.\u003c/p\u003e \u003cp\u003eThe LA-ICPMS-derived Mg profiles exhibit well-developed cyclicity through much of the record. A comparison between the number of Mg cycles in the record and the U-Th and radiocarbon chronologies strongly suggests that the Mg cycles are annual. We conducted multiple Mg cycle counts with different observers to test their reproducibility, and constructed a chronology based on the annual Mg cycles (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The cyclicity is ambiguous for some sections of the record; thus, in the final cycle count these sections were modelled using the radiocarbon chronology mean growth rate (Figure S4).\u003c/p\u003e \u003cp\u003eMg concentrations in BER-SWI-13 are significantly inversely correlated (r = -0.51, p ≪ 0.0001, n\u0026thinsp;=\u0026thinsp;217) with a previously developed coral Sr/Ca-derived proxy SST record from Bermuda\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, which is contrary to that is expected based on the temperature-dependency of D\u003csub\u003eMg\u003c/sub\u003e. Instead, Mg concentrations appear to be an indirect proxy for SST via sea spray contributions to rainfall. A clear inverse relationship exists between SST and wind speed (Figure S5). Colder temperatures are associated with higher wind speeds, which gives rise to higher sea spray contributions to rainfall and vadose dripwater. Seawater has a much higher Mg concentration than normal meteoric water and this influences the concentrations in cave dripwaters (and secondary calcite precipitates such as stalagmites). Additionally, because the Mg and Sr data do not plot along a PCP vector (Figure S6), PCP is unlikely to be a significant control on the record. The post-1970 CE data appear to show evidence of PCP supporting our decision to exclude these data from our interpretations due to alterations to the stalagmite\u0026rsquo;s drip hydrology. However, the Mg and Sr data do plot along a mixing curve between the concentrations of the bedrock and marine aerosols (Figure S6), supporting the perspective that sea spray (in turn controlled by SST) was the dominant control on Mg and Sr concentrations.\u003c/p\u003e \u003cp\u003eBy virtue of the inverse relationship between speleothem Mg and SST, the Mg cycle count was tuned intra-annually so that the Mg concentration minima were assigned a decimal date corresponding to the 15th of August (midway between the months with the lowest mean wind speed) and maxima (determined by the broad shape of the cycle to avoid individual weather events) were assigned a date corresponding to January 1st (start of month with the highest mean wind speed). The resultant high-precision final cycle count (1456\u0026ndash;2013 CE) is consistent with the radiocarbon chronology, with a nearly identical mean growth rate (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea), as well as high intra-annual precision. The final cycle count yields markedly younger values than the U-Th data until depths from top greater than ~\u0026thinsp;128 mm when the U-Th growth rate rapidly increases, whilst also having a more consistent annual growth rate (range in annual cycle length - final cycle count: 865.8 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e; U-Th: 1571.8 \u0026micro;m yr\u003csup\u003e-\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e). The COnstructing Proxy Records from Age models (COPRA) algorithm\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e was used to develop an age-depth model with cycle count annual maxima and minima as inputs. The resultant COPRA model incorporated 2000 Monte Carlo simulations and piecewise cubic Hermite interpolating polynomial (PCHIP) interpolation to assign precise ages to every data point. The COPRA algorithm did not identify any hiatuses in the record. The complete dataset consists of 18,471 data points and creates a record of sub-monthly resolution with an average of 33.2 values per year.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eMg-SST Calibration\u003c/h2\u003e \u003cp\u003eTo derive a SST record, the speleothem Mg record was interpolated monthly and passed through a Savitzky-Golay filter to remove the influence of weather extrema and/or areas with anomalously high amounts of fluid inclusion before averaging into annual values. The correlation between the annual stalagmite Mg concentration record and the annual coralline SST record\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e over the ~\u0026thinsp;200 years of overlap was then maximised by shifting the SST record within dating uncertainties using dynamic time warping (DTW) (see Figure S7). The DTW applied offset was negative (mean of -8.7 years) before 1864 CE, after which it changes to positive (mean of 6.0 years), likely due to the change in trend in both records around this time. Goodkin et al. \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e suggest that undercounting is most prevalent in their record before the mid-1800s when growth rates were extremely slow, supporting the negative DTW offset before 1864. An initial ordinary least squares (OLS) linear regression returned autocorrelated residuals and a non-statistically significant growth rate correlation term. Thus, the regression was corrected using a Cochrane-Orcutt estimation to include an AR(1) term and the growth rate term was excluded\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e. The final linear regression yielded the following correlation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb):\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:SST=25.8250\\left(\\pm\\:0.2084\\right)-0.0010(\\pm\\:0.0001)\\bullet\\:Mg$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:(2\\sigma\\:,\\:95\\%\\:conf.,\\:{R}^{2}=0.69,\\:\\:p\\ll\\:0.0001,\\:RMSE=0.5734^\\circ\\:C)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eThe modelled stalagmite SST record (Table S3) has less variance than the coralline record, likely because of aquifer mixing smoothing the signal (potentially the mixing of diffuse and concentrated flow paths). The signal-dampening effect that the strong correlation between cave temperature and SST would have via the temperature dependency of Mg fractionation may also be responsible for this reduced variance. Additionally, the regression\u0026rsquo;s predictive power was evaluated using split-period calibration/verification tests in both directions (see Figure S8). The regression passed these tests and thus yielded both a monthly and an annual SST record spanning 558 years, making it longer and higher resolution than many previous SST records from the region (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eData and materials availability\u003c/h2\u003e \u003cp\u003eAll the data needed to evaluate the conclusions in the paper are available in the paper and/or the Supplementary Information.\u003c/p\u003e \u003c/div\u003e\n\u003ch2\u003eAuthor contributions statement\u003c/h2\u003e \u003cp\u003eE.C.G.F. and J.U.L.B. wrote the original draft. F.A.L., I.W.W. and S.R.S. collected the sample. D.C.N. and D.A.R. carried out the uranium-series dating. F.A.L., I.W.W. and C.M. constructed the radiocarbon chronology. R.A.J. and W.M. performed the trace element analysis. E.C.G.F. and J.U.L.B. conducted the layer counts, carried out the data analysis and interpreted the results. All authors contributed to the project, discussed manuscript ideas, and approved the final manuscript.\u003c/p\u003e \n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eWe thank the Bermuda Department of Conservation Services for a collecting permit for the speleothem. We thank A. Outerbridge for permission to enter Leamington Cave and G. Nolan for assistance with the cave monitoring equipment. We thank S.F.M Breitenbach for assisting with the age model development and use of the COPRA algorithm. We also thank P. Moffa-S\u0026aacute;nchez for the insightful comments and suggestions regarding the interpretation of our record. A European Research Council grant (ERC240167; J.U.L.B.) supported this work. Open access publication costs are supported by Durham University, Open Access Fund.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDitlevsen, P. \u0026amp; Ditlevsen, S. Warning of a forthcoming collapse of the Atlantic meridional overturning circulation. Nature Communications 14, 1\u0026ndash;12, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-023-39810-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-39810-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Westen, R. M., Kliphuis, M. \u0026amp; Dijkstra, H. A. Physics-based early warning signal shows that AMOC is on tipping course. Science Advances 10, eadk1189, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.adk1189\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.adk1189\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePalter, J. B. The Role of the Gulf Stream in European Climate. Annual Review of Marine Science 7, 113\u0026ndash;137, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eDOI:https://doi.org/10.1146/annurev-marine-010814-015656\u003c/span\u003e\u003cspan address=\"DOI:10.1146/annurev-marine-010814-015656\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaesar, L., Rahmstorf, S., Robinson, A., Feulner, G. \u0026amp; Saba, V. Observed fingerprint of a weakening Atlantic Ocean overturning circulation. Nature 556, 191\u0026ndash;196, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-018-0006-5\u003c/span\u003e\u003cspan address=\"10.1038/s41586-018-0006-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaesar, L., McCarthy, G. D., Thornalley, D. J. R., Cahill, N. \u0026amp; Rahmstorf, S. Current Atlantic Meridional Overturning Circulation weakest in last millennium. Nature Geoscience 14, 118\u0026ndash;120, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41561-021-00699-z\u003c/span\u003e\u003cspan address=\"10.1038/s41561-021-00699-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeidov, D., Mishonov, A., Reagan, J. \u0026amp; Parsons, R. Resilience of the Gulf Stream path on decadal and longer timescales. Scientific Reports 9, 11549, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-019-48011-9\u003c/span\u003e\u003cspan address=\"10.1038/s41598-019-48011-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitney, N. M. \u003cem\u003eet al.\u003c/em\u003e Rapid 20th century warming reverses 900-year cooling in the Gulf of Maine. Communications Earth \u0026amp; Environment 3, 179, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s43247-022-00504-8\u003c/span\u003e\u003cspan address=\"10.1038/s43247-022-00504-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLund, D. C., Lynch-Stieglitz, J. \u0026amp; Curry, W. B. Gulf Stream density structure and transport during the past millennium. Nature 444, 601\u0026ndash;604, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature05277\u003c/span\u003e\u003cspan address=\"10.1038/nature05277\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRashid, H., Zhang, Z. W., Piper, D. J. W., Patro, R. \u0026amp; Xu, Y. P. Impact of Medieval Climate Anomaly and Little Ice Age on the Labrador Current flow speed and the AMOC reconstructed by the sediment dynamics and biomarker proxies. Palaeogeography Palaeoclimatology Palaeoecology 620, 111558, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.palaeo.2023.111558\u003c/span\u003e\u003cspan address=\"10.1016/j.palaeo.2023.111558\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThibodeau, B. \u003cem\u003eet al.\u003c/em\u003e Last Century Warming Over the Canadian Atlantic Shelves Linked to Weak Atlantic Meridional Overturning Circulation. Geophysical Research Letters 45, 12376\u0026ndash;12385, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018gl080083\u003c/span\u003e\u003cspan address=\"10.1029/2018gl080083\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaker, A. \u003cem\u003eet al.\u003c/em\u003e The Properties of Annually Laminated Stalagmites-A Global Synthesis. Reviews of Geophysics 59, e2020RG000722, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2020RG000722\u003c/span\u003e\u003cspan address=\"10.1029/2020RG000722\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarlson, P. E., Miller, N. R., Banner, J. L., Breecker, D. O. \u0026amp; Casteel, R. C. The potential of near-entrance stalagmites as high-resolution terrestrial paleoclimate proxies: Application of isotope and trace-element geochemistry to seasonally-resolved chronology. Geochimica Et Cosmochimica Acta 235, 55\u0026ndash;75, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gca.2018.04.036\u003c/span\u003e\u003cspan address=\"10.1016/j.gca.2018.04.036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, H. M. \u003cem\u003eet al.\u003c/em\u003e Seasonal variations in Sr, Mg and P in modern speleothems (Grotta di Ernesto, Italy). Chemical Geology 175, 429\u0026ndash;448, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0009-2541(00)00337-5\u003c/span\u003e\u003cspan address=\"10.1016/S0009-2541(00)00337-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDrysdale, R. \u003cem\u003eet al.\u003c/em\u003e Magnesium in subaqueous speleothems as a potential palaeotemperature proxy. Nature Communications 11, 5027, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-020-18083-7\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-18083-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaldini, L. M. \u003cem\u003eet al.\u003c/em\u003e Regional temperature, atmospheric circulation, and sea-ice variability within the Younger Dryas Event constrained using a speleothem from northern Iberia. Earth and Planetary Science Letters 419, 101\u0026ndash;110, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.epsl.2015.03.015\u003c/span\u003e\u003cspan address=\"10.1016/j.epsl.2015.03.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFairchild, I. J. \u003cem\u003eet al.\u003c/em\u003e Controls on trace element (Sr-Mg) compositions of carbonate cave waters: implications for speleothem climatic records. Chemical Geology 166, 255\u0026ndash;269, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0009-2541(99)00216-8\u003c/span\u003e\u003cspan address=\"10.1016/S0009-2541(99)00216-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSinclair, D. J. \u003cem\u003eet al.\u003c/em\u003e Magnesium and strontium systematics in tropical speleothems from the Western Pacific. Chemical Geology 294, 1\u0026ndash;17, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.chemgeo.2011.10.008\u003c/span\u003e\u003cspan address=\"10.1016/j.chemgeo.2011.10.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTremaine, D. M. \u003cem\u003eet al.\u003c/em\u003e A two-year automated dripwater chemistry study in a remote cave in the tropical south Pacific: Using Cl\u0026thinsp;\u0026ndash;\u0026thinsp;as a conservative tracer for seasalt contribution of major cations. Geochimica et Cosmochimica Acta 184, 289\u0026ndash;310, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gca.2016.03.029\u003c/span\u003e\u003cspan address=\"10.1016/j.gca.2016.03.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDraschba, S., P\u0026auml;tzold, J. \u0026amp; Wefer, G. North Atlantic climate variability since AD 1350 recorded in δ\u003csup\u003e18\u003c/sup\u003eO and skeletal density of Bermuda corals. International Journal of Earth Sciences 88, 733\u0026ndash;741, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s005310050301\u003c/span\u003e\u003cspan address=\"10.1007/s005310050301\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodkin, N. F., Hughen, K. A., Curry, W. B., Doney, S. C. \u0026amp; Ostermann, D. R. Sea surface temperature and salinity variability at Bermuda during the end of the Little Ice Age. \u003cem\u003ePaleoceanography\u003c/em\u003e 23, PA3203, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2007pa001532\u003c/span\u003e\u003cspan address=\"10.1029/2007pa001532\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuhnert, H., P\u0026auml;tzold, J., Schnetger, B. \u0026amp; Wefer, G. Sea-surface temperature variability in the 16th century at Bermuda inferred from coral records. Palaeogeography Palaeoclimatology Palaeoecology 179, 159\u0026ndash;171, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0031-0182(01)00410-2\u003c/span\u003e\u003cspan address=\"10.1016/S0031-0182(01)00410-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBroecker, W. S. Was a change in thermohaline circulation responsible for the Little Ice Age? Proceedings of the National Academy of Sciences of the United States of America 97, 1339\u0026ndash;1342, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.97.4.1339\u003c/span\u003e\u003cspan address=\"10.1073/pnas.97.4.1339\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLapointe, F. \u0026amp; Bradley, R. S. Little Ice Age abruptly triggered by intrusion of Atlantic waters into the Nordic Seas. Science Advances 7, eabi8230, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.abi8230\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.abi8230\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchleussner, C. F., Divine, D. V., Donges, J. F., Miettinen, A. \u0026amp; Donner, R. V. Indications for a North Atlantic ocean circulation regime shift at the onset of the Little Ice Age. Climate Dynamics 45, 3623\u0026ndash;3633, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-015-2561-x\u003c/span\u003e\u003cspan address=\"10.1007/s00382-015-2561-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVacher, H., Hearty, P., Rowe, M., Curran, H. \u0026amp; White, B. Stratigraphy of Bermuda: nomenclature, concepts, and status of multiple systems of classification. Special Papers-Geological Society of America, 271\u0026ndash;294 (1995).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHearty, P. J. \u0026amp; Vacher, H. L. Quaternary Stratigraphy of Bermuda - a High-Resolution Pre-Sangamonian Rock Record. Quaternary Science Reviews 13, 685\u0026ndash;697, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0277-3791(94)90099-X\u003c/span\u003e\u003cspan address=\"10.1016/0277-3791(94)90099-X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIliffe, T. M. \u0026amp; Calder\u0026oacute;n-Guti\u0026eacute;rrez, F. Bermuda's Walsingham Caves: A Global Hotspot for Anchialine Stygobionts. Diversity 13, 352, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/d13080352\u003c/span\u003e\u003cspan address=\"10.3390/d13080352\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeel, M. C., Finlayson, B. L. \u0026amp; McMahon, T. A. Updated world map of the Koppen-Geiger climate classification. Hydrology and Earth System Sciences 11, 1633\u0026ndash;1644, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/hess-11-1633-2007\u003c/span\u003e\u003cspan address=\"10.5194/hess-11-1633-2007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBermuda Weather Service. \u003cem\u003eClimate Data\u003c/em\u003e, available at:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.weather.bm/climate.asp\u003c/span\u003e\u003cspan address=\"https://www.weather.bm/climate.asp\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRayner, N. A. \u003cem\u003eet al.\u003c/em\u003e Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research-Atmospheres 108, 4407, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2002jd002670\u003c/span\u003e\u003cspan address=\"10.1029/2002jd002670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMichaels, A. F. \u0026amp; Knap, A. H. Overview of the US JGOFS Bermuda Atlantic Time-series Study and the Hydrostation S program. Deep Sea Research Part II: Topical Studies in Oceanography 43, 157\u0026ndash;198, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0967-0645(96)00004-5\u003c/span\u003e\u003cspan address=\"10.1016/0967-0645(96)00004-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteinberg, D. K. \u003cem\u003eet al.\u003c/em\u003e Overview of the US JGOFS Bermuda Atlantic Time-series Study (BATS): a decade-scale look at ocean biology and biogeochemistry. Deep Sea Research Part II: Topical Studies in Oceanography 48, 1405\u0026ndash;1447, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0967-0645(00)00148-X\u003c/span\u003e\u003cspan address=\"10.1016/S0967-0645(00)00148-X\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBooth, B. B., Dunstone, N. J., Halloran, P. R., Andrews, T. \u0026amp; Bellouin, N. Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability. Nature 484, 228\u0026ndash;232, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature10946\u003c/span\u003e\u003cspan address=\"10.1038/nature10946\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWild, M. Global dimming and brightening: A review. Journal of Geophysical Research: Atmospheres 114, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2008JD011470\u003c/span\u003e\u003cspan address=\"10.1029/2008JD011470\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThornalley, D. J. R. \u003cem\u003eet al.\u003c/em\u003e Anomalously weak Labrador Sea convection and Atlantic overturning during the past 150 years. Nature 556, 227\u0026ndash;230, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-018-0007-4\u003c/span\u003e\u003cspan address=\"10.1038/s41586-018-0007-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCronin, T. M. \u003cem\u003eet al.\u003c/em\u003e The Medieval Climate Anomaly and Little Ice Age in Chesapeake Bay and the North Atlantic Ocean. Palaeogeography Palaeoclimatology Palaeoecology 297, 299\u0026ndash;310, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.palaeo.2010.08.009\u003c/span\u003e\u003cspan address=\"10.1016/j.palaeo.2010.08.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeigwin, L. D. \u0026amp; Pickart, R. S. Slope water current over the Laurentian Fan on interannual to millennial time scales. Science 286, 520\u0026ndash;523, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.286.5439.520\u003c/span\u003e\u003cspan address=\"10.1126/science.286.5439.520\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJoyce, T. M., Kwon, Y. O., Seo, H. \u0026amp; Ummenhofer, C. C. Meridional Gulf Stream Shifts Can Influence Wintertime Variability in the North Atlantic Storm Track and Greenland Blocking. Geophysical Research Letters 46, 1702\u0026ndash;1708, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018gl081087\u003c/span\u003e\u003cspan address=\"10.1029/2018gl081087\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanchez-Franks, A., Hameed, S. \u0026amp; Wilson, R. E. The Icelandic Low as a Predictor of the Gulf Stream North Wall Position. Journal of Physical Oceanography 46, 817\u0026ndash;826, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/Jpo-D-14-0244.1\u003c/span\u003e\u003cspan address=\"10.1175/Jpo-D-14-0244.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor, A. H. \u0026amp; Stepens, J. A. The North Atlantic oscillation and the latitude of the Gulf Stream. Tellus Series a-Dynamic Meteorology and Oceanography 50, 134\u0026ndash;142, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1034/j.1600-0870.1998.00010.x\u003c/span\u003e\u003cspan address=\"10.1034/j.1600-0870.1998.00010.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1998).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, W. Z., Chai, F., Xue, H. J. \u0026amp; Oey, L. Y. Remote sensing linear trends of the Gulf Stream from 1993 to 2016. Ocean Dynamics 70, 701\u0026ndash;712, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10236-020-01356-6\u003c/span\u003e\u003cspan address=\"10.1007/s10236-020-01356-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMedhaug, I., Langehaug, H. R., Eldevik, T., Furevik, T. \u0026amp; Bentsen, M. Mechanisms for decadal scale variability in a simulated Atlantic meridional overturning circulation. Climate Dynamics 39, 77\u0026ndash;93, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-011-1124-z\u003c/span\u003e\u003cspan address=\"10.1007/s00382-011-1124-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChaudhuri, A. H., Gangopadhyay, A. \u0026amp; Bisagni, J. J. Response of the Gulf Stream transport to characteristic high and low phases of the North Atlantic Oscillation. Ocean Modelling 39, 220\u0026ndash;232, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ocemod.2011.04.005\u003c/span\u003e\u003cspan address=\"10.1016/j.ocemod.2011.04.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuterbacher, J. \u003cem\u003eet al.\u003c/em\u003e Extending North Atlantic Oscillation reconstructions back to 1500. Atmospheric Science Letters 2, 114\u0026ndash;124, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1006/asle.2001.0044\u003c/span\u003e\u003cspan address=\"10.1006/asle.2001.0044\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia, W. \u003cem\u003eet al.\u003c/em\u003e Highly resolved δ\u003csup\u003e13\u003c/sup\u003eC and trace element ratios of precisely dated stalagmite from northwestern China: Hydroclimate reconstruction during the last two millennia. Quaternary Science Reviews 291, 107473, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2022.107473\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2022.107473\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWassenburg, J. A. \u003cem\u003eet al.\u003c/em\u003e Calcite Mg and Sr partition coefficients in cave environments: Implications for interpreting prior calcite precipitation in speleothems. Geochimica Et Cosmochimica Acta 269, 581\u0026ndash;596, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gca.2019.11.011\u003c/span\u003e\u003cspan address=\"10.1016/j.gca.2019.11.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDenniston, R. F. \u003cem\u003eet al.\u003c/em\u003e A Last Glacial Maximum through middle Holocene stalagmite record of coastal Western Australia climate. Quaternary Science Reviews 77, 101\u0026ndash;112, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2013.07.002\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2013.07.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMann, M. E. \u003cem\u003eet al.\u003c/em\u003e Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proceedings of the National Academy of Sciences of the United States of America 105, 13252\u0026ndash;13257, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0805721105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0805721105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalczak, I. \u003cem\u003eHolocene climate variability revealed using geochemistry and Computed Tomography scanning of stalagmites from the North Atlantic Basin.\u003c/em\u003e Doctoral thesis thesis, Durham University, Durham, United Kingdom, (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaldini, J. U. L. \u003cem\u003eet al.\u003c/em\u003e Detecting and quantifying palaeoseasonality in stalagmites using geochemical and modelling approaches. Quaternary Science Reviews 254, 106784, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2020.106784\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2020.106784\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann, D. L. \u003cem\u003eet al.\u003c/em\u003e Procedures for accurate U and Th isotope measurements by high precision MC-ICPMS. International Journal of Mass Spectrometry 264, 97\u0026ndash;109, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ijms.2007.03.020\u003c/span\u003e\u003cspan address=\"10.1016/j.ijms.2007.03.020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng, H. \u003cem\u003eet al.\u003c/em\u003e Improvements in \u003csup\u003e230\u003c/sup\u003eTh dating, \u003csup\u003e230\u003c/sup\u003eTh and \u003csup\u003e234\u003c/sup\u003eU half-life values, and U\u0026ndash;Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth and Planetary Science Letters 371, 82\u0026ndash;91, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.epsl.2013.04.006\u003c/span\u003e\u003cspan address=\"10.1016/j.epsl.2013.04.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVermeesch, P. IsoplotR: A free and open toolbox for geochronology. Geoscience Frontiers 9, 1479\u0026ndash;1493, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gsf.2018.04.001\u003c/span\u003e\u003cspan address=\"10.1016/j.gsf.2018.04.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichards, D. A. \u0026amp; Dorale, J. A. Uranium-series chronology and environmental applications of speleothems. Reviews in Mineralogy and Geochemistry 52, 407\u0026ndash;460, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2113/0520407\u003c/span\u003e\u003cspan address=\"10.2113/0520407\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWortham, B. E., Banner, J. L., James, E. W., Edwards, R. L. \u0026amp; Loewy, S. Application of cave monitoring to constrain the value and source of detrital 230Th/232Th in speleothem calcite: Implications for U-series geochronology of speleothems. Palaeogeography, Palaeoclimatology, Palaeoecology 596, 110978, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.palaeo.2022.110978\u003c/span\u003e\u003cspan address=\"10.1016/j.palaeo.2022.110978\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, S. \u003cem\u003eet al.\u003c/em\u003e An integrated study of constraining the initial \u003csup\u003e230\u003c/sup\u003eTh of a stalagmite and its implications. Quaternary Geochronology 80, 101497, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quageo.2024.101497\u003c/span\u003e\u003cspan address=\"10.1016/j.quageo.2024.101497\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoseley, G. E. \u003cem\u003eet al.\u003c/em\u003e Early\u0026ndash;middle Holocene relative sea-level oscillation events recorded in a submerged speleothem from the Yucat\u0026aacute;n Peninsula, Mexico. The Holocene 25, 1511\u0026ndash;1521, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0959683615585832\u003c/span\u003e\u003cspan address=\"10.1177/0959683615585832\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeck, J. W. \u003cem\u003eet al.\u003c/em\u003e Extremely large variations of atmospheric \u003csup\u003e14\u003c/sup\u003eC concentration during the last glacial period. Science 292, 2453\u0026ndash;2458, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1056649\u003c/span\u003e\u003cspan address=\"10.1126/science.1056649\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann, D. L. \u003cem\u003eet al.\u003c/em\u003e Towards radiocarbon calibration beyond 28 ka using speleothems from the Bahamas. Earth and Planetary Science Letters 289, 1\u0026ndash;10, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.epsl.2009.10.004\u003c/span\u003e\u003cspan address=\"10.1016/j.epsl.2009.10.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFensterer, C. \u003cem\u003eet al.\u003c/em\u003e Cuban stalagmite suggests relationship between Caribbean precipitation and the Atlantic Multidecadal Oscillation during the past 1.3 ka. The Holocene 22, 1405\u0026ndash;1412, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0959683612449759\u003c/span\u003e\u003cspan address=\"10.1177/0959683612449759\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRivera-Collazo, I. \u003cem\u003eet al.\u003c/em\u003e Human adaptation strategies to abrupt climate change in Puerto Rico ca. 3.5 ka. The Holocene 25, 627\u0026ndash;640, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0959683614565951\u003c/span\u003e\u003cspan address=\"10.1177/0959683614565951\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarken, S. F. \u003cem\u003eet al.\u003c/em\u003e Persistent link between Caribbean precipitation and Atlantic Ocean circulation during the Last Glacial revealed by a speleothem record from Puerto Rico. Paleoceanography and Paleoclimatology 35, e2020PA003944, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2020PA003944\u003c/span\u003e\u003cspan address=\"10.1029/2020PA003944\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHellstrom, J. U\u0026ndash;Th dating of speleothems with high initial \u003csup\u003e230\u003c/sup\u003eTh using stratigraphical constraint. Quaternary Geochronology 1, 289\u0026ndash;295, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eDOI:https://doi.org/10.1016/j.quageo.2007.01.004\u003c/span\u003e\u003cspan address=\"DOI:10.1016/j.quageo.2007.01.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJamieson, R. A. \u003cem\u003eTrace element geochemistry of Belizean and Bermudan stalagmites: new tools, proxies and applications.\u003c/em\u003e Doctoral thesis thesis, Durham University, Durham, United Kingdom, (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEvans, D., Erez, J., Oron, S. \u0026amp; M\u0026uuml;ller, W. Mg/Ca-temperature and seawater-test chemistry relationships in the shallow-dwelling large benthic foraminifera Operculina ammonoides. Geochimica et Cosmochimica Acta 148, 325\u0026ndash;342, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gca.2014.09.039\u003c/span\u003e\u003cspan address=\"10.1016/j.gca.2014.09.039\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026uuml;ller, W., Shelley, M., Miller, P. \u0026amp; Broude, S. Initial performance metrics of a new custom-designed ArF excimer LA-ICPMS system coupled to a two-volume laser-ablation cell. Journal of Analytical Atomic Spectrometry 24, 209\u0026ndash;214, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1039/b805995k\u003c/span\u003e\u003cspan address=\"10.1039/b805995k\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLongerich, H., Jackson, S. \u0026amp; Gunther, D. Laser ablation inductively coupled plasma mass spectrometric transient signal accquistion and analyte concentration calculation. Journal of Analytical Atomic Spectrometry 11, 899\u0026ndash;904 (1996).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStoll, H. M., M\u0026uuml;ller, W. \u0026amp; Prieto, M. I-STAL, a model for interpretation of Mg/Ca, Sr/Ca and Ba/Ca variations in speleothems and its forward and inverse application on seasonal to millennial scales. Geochemistry, Geophysics, Geosystems 13, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2012GC004183\u003c/span\u003e\u003cspan address=\"10.1029/2012GC004183\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaton, C., Hellstrom, J., Paul, B., Woodhead, J. \u0026amp; Hergt, J. Iolite: Freeware for the visualisation and processing of mass spectrometric data. Journal of Analytical Atomic Spectrometry 26, 2508\u0026ndash;2518, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1039/c1ja10172b\u003c/span\u003e\u003cspan address=\"10.1039/c1ja10172b\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLechleitner, F. A. \u003cem\u003eet al.\u003c/em\u003e A novel approach for construction of radiocarbon-based chronologies for speleothems. Quaternary Geochronology 35, 54\u0026ndash;66, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quageo.2016.05.006\u003c/span\u003e\u003cspan address=\"10.1016/j.quageo.2016.05.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFohlmeister, J. \u0026amp; Lechleitner, F. A. STAlagmite dating by radiocarbon (star): A software tool for reliable and fast age depth modelling. Quaternary Geochronology 51, 120\u0026ndash;129, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quageo.2019.02.008\u003c/span\u003e\u003cspan address=\"10.1016/j.quageo.2019.02.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBreitenbach, S. F. M. \u003cem\u003eet al.\u003c/em\u003e COnstructing Proxy Records from Age models (COPRA). Climate of the Past 8, 1765\u0026ndash;1779, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/cp-8-1765-2012\u003c/span\u003e\u003cspan address=\"10.5194/cp-8-1765-2012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCochrane, D. \u0026amp; Orcutt, G. H. Application of Least Squares Regression to Relationships Containing Autocorrelated Error Terms. Journal of the American Statistical Association 44, 32\u0026ndash;61, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2280349\u003c/span\u003e\u003cspan address=\"10.2307/2280349\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1949).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCopernicus Marine Service. ODYSSEA Global Sea Surface Temperature Gridded Level 4 Daily Multi-Sensor Observations, available at:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48670/mds-00321\u003c/span\u003e\u003cspan address=\"10.48670/mds-00321\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRouse, J. R. P. \u003cem\u003eThe Provenance and Geochemical Alteration fo Bermudan Eolianites.\u003c/em\u003e Doctoral thesis thesis, Queen\u0026rsquo;s University, Ontario, Canada, (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, Y. M. \u0026amp; Fairchild, I. J. Partitioning of Sr\u003csup\u003e2+\u003c/sup\u003e and Mg\u003csup\u003e2+\u003c/sup\u003e into calcite under karst-analogue experimental conditions. Geochemica Et Cosmochimica Acta 65, 47\u0026ndash;62, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0016-7037(00)00513-5\u003c/span\u003e\u003cspan address=\"10.1016/S0016-7037(00)00513-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChester, R. Marine Geochemistry. \u003cem\u003eJohn Wiley \u0026amp; Sons\u003c/em\u003e (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTierney, J. E. \u003cem\u003eet al.\u003c/em\u003e Tropical sea surface temperatures for the past four centuries reconstructed from coral archives. Paleoceanography and Paleoclimatology 30, 226\u0026ndash;252, DOI:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/2014PA002717\u003c/span\u003e\u003cspan address=\"10.1002/2014PA002717\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5973851/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5973851/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Gulf Stream forms part of the upper-ocean limb of the Atlantic Meridional Overturning Circulation (AMOC), playing an essential role in redistributing heat northward and greatly influencing regional climates in the North Atlantic. Understanding Gulf Stream path and strength variability on longer timescales is vital to contextualise its present-day weakening and to fully appreciate its sensitivity to forcing. We present a 558-year long (1456\u0026ndash;2013) proxy record of sea surface temperature from a Bermudan stalagmite using an indirect magnesium-temperature calibration based on a connection to wind speed. Our monthly-resolved terrestrial palaeo-oceanographic temperature reconstruction indicates that the Gulf Stream was likely positioned further south than today during the Little Ice Age. We suggest that a combination of reduced Gulf Stream transport, enhanced Labrador Current and Deep Western Boundary Current transport, and an extended negative North Atlantic Oscillation phase, caused the Gulf Stream to be at lower latitudes during the Little Ice Age, before migrating northward as the Little Ice Age abated.\u003c/p\u003e","manuscriptTitle":"Monthly-Resolved Cave Proxy Evidence for Northward Gulf Stream Migration During the Little Ice Age","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-26 08:27:06","doi":"10.21203/rs.3.rs-5973851/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-earth-and-environment","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commsenv","sideBox":"Learn more about [Communications Earth and Environment](https://www.nature.com/commsenv/)","snPcode":"","submissionUrl":"","title":"Communications Earth \u0026 Environment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d6876783-dbbb-4e01-b6b6-9e0f7bb83e40","owner":[],"postedDate":"March 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":44469721,"name":"Earth and environmental sciences/Climate sciences/Palaeoclimate"},{"id":44469722,"name":"Earth and environmental sciences/Climate sciences/Palaeoceanography"}],"tags":[],"updatedAt":"2025-07-15T07:11:08+00:00","versionOfRecord":{"articleIdentity":"rs-5973851","link":"https://doi.org/10.1038/s43247-025-02446-3","journal":{"identity":"communications-earth-and-environment","isVorOnly":false,"title":"Communications Earth \u0026 Environment"},"publishedOn":"2025-07-14 04:00:00","publishedOnDateReadable":"July 14th, 2025"},"versionCreatedAt":"2025-03-26 08:27:06","video":"","vorDoi":"10.1038/s43247-025-02446-3","vorDoiUrl":"https://doi.org/10.1038/s43247-025-02446-3","workflowStages":[]},"version":"v1","identity":"rs-5973851","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5973851","identity":"rs-5973851","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}