Climate: Long-Range Investigation, Mapping, And Prediction Research Articles

Constraints on the magnitude and patterns of ocean cooling at the Last Glacial Maximum

Observation-based reconstructions of sea surface temperature from relatively stable periods in the past, such as the Last Glacial Maximum, represent an important means of constraining climate sensitivity and evaluating model simulations1. The first quantitative global reconstruction of sea surface temperatures during the Last Glacial Maximum was developed by the Climate Long-Range Investigation, Mapping and Prediction (CLIMAP) project in the 1970s and 1980s (refs 2, 3). Since that time, several shortcomings of that earlier effort have become apparent4. Here we present an updated synthesis of sea surface temperatures during the Last Glacial Maximum, rigorously defined as the period between 23 and 19 thousand years before present, from the Multiproxy Approach for the Reconstruction of the Glacial Ocean Surface (MARGO) project5. We integrate microfossil and geochemical reconstructions of surface temperatures and include assessments of the reliability of individual records. Our reconstruction reveals the presence of large longitudinal gradients in sea surface temperature in all of the ocean basins, in contrast to the simulations of the Last Glacial Maximum climate available at present6, 7.

The Effect of a Large Freshwater Perturbation on the Glacial North Atlantic Ocean Using a Coupled General Circulation Model

Abstract The commonly held view of the conditions in the North Atlantic at the last glacial maximum, based on the interpretation of proxy records, is of large-scale cooling compared to today, limited deep convection, and extensive sea ice, all associated with a southward displaced and weakened overturning thermohaline circulation (THC) in the North Atlantic. Not all studies support that view; in particular, the “strength of the overturning circulation” is contentious and is a quantity that is difficult to determine even for the present day. Quasi-equilibrium simulations with coupled climate models forced by glacial boundary conditions have produced differing results, as have inferences made from proxy records. Most studies suggest the weaker circulation, some suggest little or no change, and a few suggest a stronger circulation. Here results are presented from a three-dimensional climate model, the Hadley Centre Coupled Model version 3 (HadCM3), of the coupled atmosphere–ocean–sea ice system suggesting, in a qualitative sense, that these diverging views could all have occurred at different times during the last glacial period, with different modes existing at different times. One mode might have been characterized by an active THC associated with moderate temperatures in the North Atlantic and a modest expanse of sea ice. The other mode, perhaps forced by large inputs of meltwater from the continental ice sheets into the northern North Atlantic, might have been characterized by a sluggish THC associated with very cold conditions around the North Atlantic and a large areal cover of sea ice. The authors’ model simulation of such a mode, forced by a large input of freshwater, bears several of the characteristics of the Climate: Long-range Investigation, Mapping, and Prediction (CLIMAP) Project’s reconstruction of glacial sea surface temperature and sea ice extent.

Evaluation of Last Glacial Maximum sea surface temperature reconstructions through their influence on South American climate

The consistency of land‐based and ocean‐based Last Glacial Maximum (LGM) proxy evidence on and near South America is evaluated by utilizing a regional climate model (RCM). The relatively high resolution of the RCM brings model output closer to the spatial scales of the geological reconstructions. This is important because proxy sites only a few tens of kilometers apart can yield markedly different interpretations of LGM climate. Along with differences in orbital parameters, CO2 concentrations, and vegetation, four versions of LGM sea surface temperatures (SSTs) (Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP), Schäfer‐Neth and Paul's (2003) core version, Paul and Schäfer‐Neth's (2003) line version, and coupled GCM output from Shin et al. (2003)), are used to generate four realizations of the LGM climate in the model. Mean annual surface temperature, mean temperature of the coldest month, and aridity differences from a present‐day simulation are compared with the geological reconstructions over South America to evaluate the consistency of the land and ocean‐based reconstructions. Model results suggest that the tropical Atlantic during the LGM was not significantly cooler than CLIMAP's original estimates. Simulations utilizing tropical Atlantic LGM SSTs cooler than CLIMAP estimates yield unrealistically warm LGM surface conditions over the equatorial Amazon basin. Reductions in evaporation and the atmosphere's moisture content over too‐cold tropical oceans lead to a reduction in cloud cover over the Amazon and increases in solar heating of the surface. All of the LGM SST simulations produce drier conditions over tropical South America; however, the extent of the aridity is associated with the degree of cooling in the surrounding oceans, especially the tropical Atlantic. The “line” version LGM SSTs of Paul and Schäfer‐Neth (2003) are recommended.

Modeling the tropical climate and the impact of the western Pacific sea surface temperature at the Last Glacial Maximum

Changes in tropical atmospheric heat and circulation patterns, as well as the effects of sea surface temperature (SST) of the western Pacific during the Last Glacial Maximum (LGM), have been investigated using a climate model and the LGM SST compiled by Wang [1999]. The annual mean LGM SST was 2°–4°C lower than the present day in the tropical western Pacific west of 140°E. Simulations show that the largest decrease of atmospheric heat occurs in the tropics in a region stretching from the Bay of Bengal to the central Pacific Ocean, while there are no significant changes in atmospheric heat at higher latitudes. This indicates that the atmospheric heat in the tropics shows a stronger response to boundary changes between the LGM and the present day than it does at higher latitudes, since the tropical convection amplifies the response of the atmosphere in the former case. Because of this tropical heat change, the Walker circulation, the transverse monsoon, and the east Asian summer monsoon at the LGM were weaker than they are at present. The model with the Wang [1999] LGM SST simulated large changes in atmospheric circulation in the tropics and high northern latitudes compared to the Climate, Long‐Range Investigation, Mapping and Prediction (CLIMAP) [CLIMAP Project Members, 1981] project's reconstruction. The Wang [1999] LGM SST was also associated with a lower surface air temperature (SAT) at high northern latitudes and the Arctic and with a cooler temperature of the coldest month in Europe. The annual mean SAT decreased by 2°C in the Arctic, and the winter mean SAT decreased by 4°C in Europe.

Atmospheric Response to Modified CLIMAP Ocean Boundary Conditions during the Last Glacial Maximum*

Global climate simulations are conducted to examine the sensitivity of the Last Glacial Maximum (LGM) climate to prescribed sea surface temperatures (SSTs) that are modified from the Climate: Long-range Investigation, Mapping, and Prediction (CLIMAP) study. Based on the consensus from various LGM proxy data, the SSTs are cooled by 48C uniformly in the Tropics (308N‐308S) relative to CLIMAP, and the high-latitude sea ice extent is reduced. Compared to results from a simulation with CLIMAP SSTs, the modified LGM SSTs cause significant opposing changes in the hemispheric and regional-scale atmospheric circulation, which are most pronounced in the winter hemisphere. For instance, there is significant weakening of the midlatitude circulation and reduction of 500-hPa eddy kinetic energy and midlatitude precipitation resulting from the decreased meridional temperature gradient in the modified SST simulation. In contrast, reduced sea ice extent during the boreal winter causes increased regional baroclinicity and intensified atmospheric circulation in the western North Pacific and the North Atlantic. Cooled tropical SSTs also increase the land‐ocean temperature contrast, which strengthens the Asian summer monsoon circulation. Both LGM simulations produce enhanced low-level convergence and increased precipitation along the South Pacific convergence zone (SPCZ) relative to present day, despite the cooler LGM climate. The SPCZ orientation and intensity are closely linked to the distribution of South Pacific SSTs. Comparison of surface temperature estimates from land- and ocean-based proxy data with model output suggests that uniform cooling of the tropical SSTs and modification of the high-latitude sea ice extent may be sufficient to accurately simulate the first-order characteristics of the LGM climate.

Forcing of the deep ocean circulation in simulations of the Last Glacial Maximum

From the interpretation of different proxy data it is widely believed that the North Atlantic thermohaline circulation during the maximum of the last ice age ∼21,000 years ago was considerably weaker than today. Recent equilibrium simulations with a coupled ocean‐atmosphere‐sea ice model successfully simulated a reduction in North Atlantic Deep Water (NADW) formation consistent with reconstructions. Here we examine the influence of different air‐sea fluxes on simulated changes in the deep ocean circulation between the Last Glacial Maximum and present day. We find that changes in the oceanic surface freshwater fluxes are the dominant forcing mechanism for the reduced Atlantic overturning. Diminished export of freshwater out of the Atlantic drainage basin through the atmosphere decreases surface salinities in the North Atlantic, leading to less NADW formation in the colder climate. Changes in heat fluxes, which lead to increased sea surface densities in the North Atlantic and therefore to an enhanced overturning, are of secondary importance. Wind stress variations seem to play a negligible role. The degree to which the Atlantic freshwater export and hence the NADW formation are reduced depends on the formulation of the atmospheric hydrological cycle and on the strength of the overturning in the present‐day simulation. Simulated changes in sea surface properties for a large variety of overturning strengths are compared with different reconstruction data sets. The results depend strongly on the data set used. Sea surface temperature reconstructions from Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) and earlier salinity reconstructions based on planktonic foraminifera are most consistent with a significant reduction of the circulation, while recent reconstructions using dinocyst assemblages allow no unequivocal conclusion.

Tropical snowline depression at the Last Glacial Maximum: Comparison with proxy records using a single‐cell tropical climate model

The substantial lowering of tropical snowlines at the Last Glacial Maximum (LGM), circa 21 kya, is examined using a modified version of the single‐cell tropical climate model ofBetts and Ridgway[1989]. These authors concluded that it was difficult to reconcile the large depression of snowlines at the LGM with the small reduction in mean tropical sea‐surface temperature (SST) of the Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) reconstruction. Here, climatic implications of the snowline depression are compared, not with CLIMAP, but with temperature proxies (δ18O and Sr/Ca ratios in corals, noble gas concentrations in aquifers, alkenone and Mg/Ca ratios in deep‐sea sediments and recent faunal reconstructions) whose LGM values have been interpreted as implying lower surface temperatures in the tropics. It proves difficult, in the framework of this model, to reconcile the coldest of the indicated paleotemperatures with the observed snowline depression, which by itself is found to be consistent with an SST reduction of ∼3 K. A cooling of this magnitude corresponds most closely to recent warm‐pool LGM sea‐surface temperature estimates based on Mg/Ca paleothermometry. Discordance among the various proxy reconstructions may result, at least in part, from regional variations in surface temperature change at the LGM.

The status of the Indo-Pacific Warm Pool and adjacent land at the Last Glacial Maximum

Since the Climate: Long Range Investigation, Mapping and Prediction (CLIMAP) reconstructions for the Earth at the Last Glacial Maximum (LGM), there have been conflicting views on the extent of cooling of the oceans of tropical Australasia—here, referred to as the Indo-Pacific Warm Pool—in contrast with those temperatures registered on land. Based on sea-surface temperature (=SST) reconstructions for the Indo-Pacific Warm Pool, and on vegetation reconstruction for SE Asia as well as by considering the increase of land mass area engendered during low sea levels, we identify for the LGM a significant drop in precipitation in the Warm Pool region that would explain an increase in salinity while SST decreased by about 2 °C at the most. The latter would have caused a substantial decrease of large-scale atmospheric convection over the Indo-Pacific Warm Pool and suppressed deep atmospheric convection that would help maintain somewhat elevated SSTs. The drier atmosphere and diminished level of cloud cover would also have reduced nocturnal temperatures at elevation in the region and produced a steeper mean atmospheric lapse rate, forcing the tree line to drop and glaciers to be maintained down to much lower altitudes than today.

Foraminiferal faunal estimates of paleotemperature: Circumventing the No‐analog problem yields cool Ice Age tropics

The sensitivity of the tropics to climate change, particularly the amplitude of glacial‐to‐interglacial changes in sea surface temperature (SST), is one of the great controversies in paleoclimatology. Here we reassess faunal estimates of ice age SSTs, focusing on the problem of no‐analog planktonic foraminiferal assemblages in the equatorial oceans that confounds both classical transfer function and modern analog methods. A new calibration strategy developed here, which uses past variability of species to define robust faunal assemblages, solves the no‐analog problem and reveals ice age cooling of 5° to 6°C in the equatorial current systems of the Atlantic and eastern Pacific Oceans. Classical transfer functions underestimated temperature changes in some areas of the tropical oceans because core‐top assemblages misrepresented the ice age faunal assemblages. Our finding is consistent with some geochemical estimates and model predictions of greater ice age cooling in the tropics than was inferred by Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) [1981] and thus may help to resolve a long‐standing controversy. Our new foraminiferal transfer function suggests that such cooling was limited to the equatorial current systems, however, and supports CLIMAP's inference of stability of the subtropical gyre centers.

The mineral dust aerosol cycle during the Last Glacial Maximum

The Canadian Centre for Climate Modelling and Analysis general circulation model (GCM) is used to investigate the factors influencing the mineral dust distribution at the time of the last glacial maximum (LGM). Simulated deposition rates are used in conjunction with existing paleoclimate data to infer likely characteristics of the mineral dust conditions at that time and identify contributions to the pattern of enhanced “dustiness” of the LGM relative to the present, indicated by ice core and loess data. Maximal consistency between the GCM, the Climate: Long Range Investigation, Mapping and Prediction (CLIMAP) data set, and the ice core data requires approximately a 15‐ to 30‐fold increase in dust production globally during the last glacial maximum relative to the preindustrial Holocene. In particular, regional 10‐ to 50‐fold and 20‐fold LGM increases are indicated in eastern Asia and South America, respectively. These implied source enhancements are quite insensitive to the details of the dust model. Analysis of the contributing factors suggests an increase in fine‐particle availability in the source regions.

Project takes a new look at past sea surface temperatures

Advancements in knowledge about sea surface temperature (SST) at key points in Earth's past are on the horizon through a new research project called TEMPUS that is funded by the European Commission's Environment and Climate Programme. The first priority of the project is to reconstruct the last ice age, which serves as a stringent test for appraising the validity of climate models. SST data for this effort are being obtained from the Climate/Long‐Range Investigation, Mapping, and Prediction (CLIMAP) project 18 kyr reconstruction [CLIMAP, 1976, 1981]. CLIMAP represented a turning point in the study of climate change and helped to fuel explosive growth in climate research based on the analysis of sediment cores. Maps produced 20 years ago from CLIMAP data are still at the core of much groundbreaking research.

Late Quaternary variations of sea surface salinity and temperature in the western tropical Atlantic: Evidence from δ18O of Globigerinoides sacculifer

In order to reconstruct hydrographic changes during glacial‐interglacial cycles for a transequatorial transect we analyzed oxygen isotopes of Globigerinoides sacculifer (without sac‐like chamber) and abundances of Globorotalia truncatulinoides (dextral) from FS Meteor cores GeoB 2204‐2 (Brazilian continental slope) and GeoB 1523‐1 (Ceara Rise). Δδ18O values of G. sacculifer between the two cores were calculated. Modern Δδ18O (G. sacculifer) is ∼0.2‰ between the two core positions, reflecting differences in sea surface salinity (SSS). Higher SSS at GeoB 1523‐1 (Ceara Rise) is the result of increased precipitation in the region of the Intertropical Convergence Zone. During glacials the δ18O records from the two cores converge to the same absolute value, resulting in Δδ18O values of around 0‰. Maximum abundances of G. truncatulinoides (dex) correlate with minimum Δδ18O, suggesting a possible increase of SSS at GeoB 1523‐1 during stages 2, 3, 4, and 6, which is related to a glacial weakening of the tropical Hadley Cell [Gates, 1976]. Variations in tropical sea surface temperatures are assumed to be low [Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP), 1981].

Effect of vegetation on an ice‐age climate model simulation

A growing number of studies suggest that vegetation changes can significantly influence regional climate variations. Herein we utilize a climate model (GENESIS) with a land surface vegetation package to evaluate the potential role of the very large vegetation changes that occurred during the last glacial maximum (LGM). In particular, we focus on the potential response to a significant reduction in the area of tropical rainforest. Simulations employed a global vegetation reconstruction for the LGM and Climate/Long‐Range Investigation, Mapping and Prediction (CLIMAP) sea surface temperature (SST) estimates. Results indicate that expansion of dryland vegetation causes a 15–30% additional LGM cooling for Australia (0.4°C) and Africa (0.9°C), respectively. Turnover from conifer to tundra also causes cooling of 2°–4°C or more in western Europe and Siberia. However, for the largest rainforest area (Amazon Basin), inclusion of realistic vegetation increased modeled temperatures 2°–4°C and decreased precipitation by 10–35%. These latter results are similar to those obtained with sensitivity experiments of the effects of future Amazon deforestation. Initial assessment of the potential effect of decreased stomatal resistance due to lower ice age CO2levels indicates little significant response to this effect. Comparison of model‐predicted low‐elevation LGM temperature changes with estimates from proxy data indicate that inclusion of realistic vegetation estimates for the LGM results in slightly more than 50% agreement between models and data for low‐elevation sites in low‐mid latitudes. Data at variance with model predictions would appear to be explainable by considering additional changes in vegetation, ice age dust, or a 1°–2°C cooling below CLIMAP values. This conclusion is at variance with a 3°–4°C tropical cooling suggested by some studies for explaining estimated land temperature changes during the LGM. In some western European sites model temperatures are colder than proxy data by 2°–8°C. This model‐data discrepancy may be explained by less sea ice in the subpolar North Atlantic than stipulated by CLIMAP, a conclusion consistent with new marine data from that region.

Spatial and temporal oceanographic variability of the eastern equatorial Pacific during the Late Pleistocene: Evidence from radiolaria microfossils

Eight 150,000 year long records of sea surface temperatures combined with two additional records spanning 400,000 years constrain the spatial and temporal patterns of oceanographic change in the eastern equatorial Pacific and possible mechanisms of variability in the region. Empirical orthogonal function analysis shows two important modes of variability, one associated with the eastern boundary current and another associated with the North Equatorial Countercurrent. The two long time series located in the equatorial divergence and within the Peru Current have very different patterns of change. The spectrum for the time series from the Peru Current is dominated by orbital periods of 100, 41, and 23 kyr and is similar in variance distribution and phase to records from the Southern Ocean. In contrast, the equatorial divergence site has spectral concentrations at the orbital frequencies and also concentration of variance at the nonorbital 31,000 year period. The phase and amplitude spectra of these two sites support the importance of changes in eastern boundary advection and also document a nonlinear response of the equatorial Pacific to orbital changes. Finally, these data provide a new evaluation of the temperature change in the eastern equatorial Pacific during the last glacial maximum. Cooling in the Peru Current region is predicted to be about 4°C, and cooling in the equatorial divergence is estimated to be 3° to 5°C. The estimated cooling of the region is of the order of 2°C greater than the cooling predicted by Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP).

Radiolarian‐based transfer functions for estimating mean surface ocean temperatures and seasonal range

New radiolarian‐based transfer functions to estimate sea surface temperature (SST) and seasonal range are presented. The transfer fonctions are based on the approach originated by Imbrie and Kipp [1971]. The transfer functions differ from previous studies in the following three important ways: (1) extensions to Q‐mode factor analysis provide an objective method to cull species in the very diverse radiolarian population; (2) a log transform of the relative abundance data is used to normalize the species percent data; and, (3) rather than writing equations for specific seasons, which are not independent data sets, statistically independent equations are developed to predict mean annual sea surface temperatures as well as seasonal temperature range. One hundred and seventy surface sediment samples from the Pacific Ocean are used to generate the SST and season temperature range transfer functions. All samples were counted using a standardized radiolarian taxonomy. Forty one radiolarian species were used in the final regression equation. Q‐mode factor analysis of this data set identified seven assemblages. These assemblages, the tropical, transitional, Antarctic, Bering Sea, western Pacific, central gyre, and eastern boundary current, are named for the oceanographic regions where these assemblages are important. The seven assemblages are used in a regression analysis to predict SST and seasonal temperature range. The standard error of estimates for both mean SST and seasonal temperature range is 1.6°C. Comparison between radiolarian‐based SSTs and SST estimates from alkenone Uk37 in a 20,000 year long record from the northeast Pacific shows excellent agreement. Comparison of mean SST estimates for the last glacial maximum (LGM) based on radiolarian and foraminifera in 10 eastern equatorial Pacific also show excellent concordance. These new LGM estimates suggest that the original Climate: Long‐Range Investigation Mapping, and Prediction (CLIMAP) reconstruction for this region underestimated surface ocean cooling.

Sea surface temperature changes in the southern California borderlands during the last glacial‐Interglacial cycle

A variety of evidence suggests that average sea surface temperatures (SSTs) during the last glacial maximum in the California Borderlands region were significantly colder than during the Holocene. Planktonic foraminiferal δ18O evidence and average SST estimates derived by the modern analog technique indicate that temperatures were 6°–10°C cooler during the last glacial relative to the present. The glacial plankton assemblage is dominated by the planktonic foraminifer Neogloboquadrina pachyderma (sinistral coiling) and the coccolith Coccolithus pelagicus, both of which are currently restricted to subpolar regions of the North Pacific. The glacial‐interglacial average SST change determined in this study is considerably larger than the 2°C change estimated by Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) [1981]. We propose that a strengthened California Current flow was associated with the advance of subpolar surface waters into the Borderlands region during the last glacial.

Reconstructing the stable isotope geochemistry and paleotemperatures of the equatorial Atlantic during the last 150,000 years: Results from individual foraminifera

This study represents an attempt to extract paleoclimatic data from the deep‐sea record by analyzing foraminiferal shells individually. Using the oxygen (δ18O) and carbon (δ13C) isotopic composition of individual Orbulina universa and Neogloboquadrina dutertrei, we present an approach to reconstruct the δ18O of seawater (δ18Ow), the δ13C of ΣCO2, and seasonal maximum sea surface temperatures (SST) in the western and eastern equatorial Atlantic. We examine the glacial and interglacial extremes of the last 150,000 years (isotope stages 1, 2, 5e, and 6). Comparison of recent water column hydrography with reconstructions from core top assemblages shows that O. universa and N. dutertrei δ18O and δ13C values accurately record hydrographic conditions in the mixed layer and upper thermocline at both sites. By analyzing shells individually, we can evaluate the effect of bioturbation on the range of δ18O and δ13C values in each interval and take it into consideration in our data interpretations. Downcore results show that N. dutertrei δ18O values in the western equatorial Atlantic reflect glacial to interglacial changes in δ18Ow due to continental ice formation (Δδ18O=1.30 ‰). We use changes in N. dutertrei δ18O values between core intervals to estimate the ice‐volume effect in paleotemperature calculations for the mixed layer. To validate the use of O. universa for mixed layer reconstructions, we have added individual Globigerinoides sacculifer data for stages 1 and 2 at both sites. Paleotemperature reconstructions from O. universa δ18O values indicate that maximum seasonal mixed layer temperatures in the equatorial Atlantic decreased by at most 2.6°C between isotope stages 1 and 2 and by no more than 3.4°C between stages 1 and 6. Individual shell data from G. sacculifer yield similar results indicating that maximum O. universa δ18O values reflect the mixed layer environment. In agreement with Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) [1981] SST reconstructions for stage 2, these data indicate little change between glacial and interglacial paleotemperatures in the equatorial Atlantic.An electronic supplement of this material may be obtainedon a diskette or Anonymous FTP from KOSMOS.AGU.ORG.(LOGIN to AGU's FTP account using ANONYMOUS as theusername and GUEST as the password. Go to the rightdirectory by typing CD APEND. Type LS to see what files areavailable. Type GET and the name of the file to get it.Finally, type EXIT to leave the system.) (Paper 95PA03773,Reconstructing the stable isotope geochemistry andpaleotemperatures of the equatorial Atlantic during the last150,000 years: Results from individual foraminifera,Katharina Billups and Howard J. Spero). Diskette may beordered from American Geophysical union, 2000 FloridaAvenue, N. W., Washington, DC 20009; $15.00. Paymentmust accompany order.

Comparing simulated glacial climate and paleodata: A reexamination

Glacial sea surface temperatures (SSTs) simulated by an atmosphere‐mixed layer ocean model are compared with those reconstructed by the Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) Project using planktonic microfossils. Two methods of comparison are employed. The first is global and uses the subjectively analyzed (i.e., hand contoured, then digitized) data set published by CLIMAP. The second is restricted to only those discrete locations where the CLIMAP sediment cores were taken. Both methods indicate that many aspects of the reconstructed glacial SST changes are simulated reasonably well by the model, although there are areas of disagreement. The extent of the disagreement appears smaller when the SSTs are sampled at discrete locations, because the largest discrepancies occur in the subtropical Pacific where data are sparse. When examined separately for each ocean, the magnitude of the disagreement in low latitudes roughly corresponds to the magnitude of the uncertainties in SST estimation using planktonic microfossils, being largest in the Pacific and smallest in the Atlantic. Because the largest discrepancies occur where uncertainties in estimation are large, no clear determination of whether the climate model exaggerates or underestimates low‐latitude climate sensitivity appears possible. Nonetheless, sampling at discrete locations may be the best procedure for evaluating climate model performance, because errors associated with extending analyses to data‐void areas can be avoided and uncertainties associated with inadequate spatial sampling made more evident.

SIMMAX: A modern analog technique to deduce Atlantic sea surface temperatures from planktonic foraminifera in deep‐sea sediments

We present a data set of 738 planktonic foraminiferal species counts from sediment surface samples of the eastern North Atlantic and the South Atlantic between 87°N and 40°S, 35°E and 60°W including published Climate: Long‐Range Investigation, Mapping, and Prediction (CLIMAP) data. These species counts are linked to Levitus's [1982] modern water temperature data for the four caloric seasons, four depth ranges (0, 30, 50, and 75 m), and the combined means of those depth ranges. The relation between planktonic foraminiferal assemblages and sea surface temperature (SST) data is estimated using the newly developed SIMMAX technique, which is an acronym for a modern analog technique (MAT) with a similarity index, based on (1) the scalar product of the normalized faunal percentages and (2) a weighting procedure of the modern analog's SSTs according to the inverse geographical distances of the most similar samples. Compared to the classical CLIMAP transfer technique and conventional MAT techniques, SIMMAX provides a more confident reconstruction of paleo‐SSTs (correlation coefficient is 0.994 for the caloric winter and 0.993 for caloric summer). The standard deviation of the residuals is 0.90°C for caloric winter and 0.96°C for caloric summer at 0‐m water depth. The SST estimates reach optimum stability (standard deviation of the residuals is 0.88°C) at the average 0– to 75‐m water depth. Our extensive database provides SST estimates over a range of −1.4 to 27.2°C for caloric winter and 0.4 to 28.6°C for caloric summer, allowing SST estimates which are especially valuable for the high‐latitude Atlantic during glacial times.An electronic supplement of this material may be obtained on adiskette or Anonymous FTP from KOSMOS.AGU.ORG. (LOGIN toAGU's FTP account using ANONYMOUS as the username and GUESTas the password. Go to the right directory by typing CD APPEND. TypeLS to see what files are available. Type GET and the name of the file toget it. Finally type EXIT to leave the system.) (Paper 95PA01743,SIMMAX: A modern analog technique to deduce Atlantic sea surfacetemperatures from planktonic foraminifera in deep‐sea sediments, UwePflaumann, Josette Duprat, Claude Pujol, and Laurent D. Labeyrie).Diskette may be ordered from American Geophysical Union, 2000Florida Avenue, N.W., Washington, DC 20009; Payment mustaccompany order.

Cooling of Tropical Brazil (5°C) During the Last Glacial Maximum

A 30,000-year paleotemperature record derived from noble gases dissolved in carbon-14-dated ground water indicates that the climate in lowland Brazil (Piaui Province, 7 degrees S, 41.5 degrees W; altitude, 400 meters) was 5.4 degrees +/- 0.6 degrees C cooler during the last glacial maximum than today. This result suggests a rather uniform cooling of the Americas between 40 degrees S and 40 degrees N. A 5.4 degrees C cooling of tropical South America is consistent with pollen records, snow line reconstructions, and strontium/calcium ratios and delta(18)O coral records but is inconsistent with the sea-surface temperature reconstruction of CLIMAP (Climate: Long-Range Investigation, Mapping and Prediction). On the basis of these results, it appears that the tropical Americas are characterized by a temperature sensitivity comparable to that found in higher latitudes.