Tropical paleoclimates at the Last Glacial Maximum: comparison of Paleoclimate Modeling Intercomparison Project (PMIP) simulations and paleodata

 Seventeen simulations of the Last Glacial Maximum (LGM) climate have been performed using atmospheric general circulation models (AGCM) in the framework of the Paleoclimate Modeling Intercomparison Project (PMIP). These simulations use the boundary conditions for CO₂, insolation and ice-sheets; surface temperatures (SSTs) are either (a) prescribed using CLIMAP data set (eight models) or (b) computed by coupling the AGCM with a slab ocean (nine models). The present-day (PD) tropical climate is correctly depicted by all the models, except the coarser resolution models, and the simulated geographical distribution of annual mean temperature is in good agreement with climatology. Tropical cooling at the LGM is less than at middle and high latitudes, but greatly exceeds the PD temperature variability. The LGM simulations with prescribed SSTs underestimate the observed temperature changes except over equatorial Africa where the models produce a temperature decrease consistent with the data. Our results confirm previous analyses showing that CLIMAP (1981) SSTs only produce a weak terrestrial cooling. When SSTs are computed, the models depict a cooling over the Pacific and Indian oceans in contrast with CLIMAP and most models produce cooler temperatures over land. Moreover four of the nine simulations, produce a cooling in good agreement with terrestrial data. Two of these model results over ocean are consistent with new SST reconstructions whereas two models simulate a homogeneous cooling. Finally, the LGM aridity inferred for most of the tropics from the data, is globally reproduced by the models with a strong underestimation for models using computed SSTs. Received: 9 September 1998 / Accepted: 18 March 1999     

Past and future polar amplification of climate change: climate model intercomparisons and ice-core constraints

Climate model simulations available from the PMIP1, PMIP2 and CMIP (IPCC-AR4) intercomparison projects for past and future climate change simulations are examined in terms of polar temperature changes in comparison to global temperature changes and with respect to pre-industrial reference simulations. For the mid-Holocene (MH, 6,000 years ago), the models are forced by changes in the Earth’s orbital parameters. The MH PMIP1 atmosphere-only simulations conducted with sea surface temperatures fixed to modern conditions show no MH consistent response for the poles, whereas the new PMIP2 coupled atmosphere–ocean climate models systematically simulate a significant MH warming both for Greenland (but smaller than ice-core based estimates) and Antarctica (consistent with the range of ice-core based range). In both PMIP1 and PMIP2, the MH annual mean changes in global temperature are negligible, consistent with the MH orbital forcing. The simulated last glacial maximum (LGM, 21,000 years ago) to pre-industrial change in global mean temperature ranges between 3 and 7°C in PMIP1 and PMIP2 model runs, similar to the range of temperature change expected from a quadrupling of atmospheric CO₂ concentrations in the CMIP simulations. Both LGM and future climate simulations are associated with a polar amplification of climate change. The range of glacial polar amplification in Greenland is strongly dependent on the ice sheet elevation changes prescribed to the climate models. All PMIP2 simulations systematically underestimate the reconstructed glacial–interglacial Greenland temperature change, while some of the simulations do capture the reconstructed glacial–interglacial Antarctic temperature change. Uncertainties in the prescribed central ice cap elevation cannot account for the temperature change underestimation by climate models. The variety of climate model sensitivities enables the exploration of the relative changes in polar temperature with respect to changes in global temperatures. Simulated changes of polar temperatures are strongly related to changes in simulated global temperatures for both future and LGM climates, confirming that ice-core-based reconstructions provide quantitative insights on global climate changes. An erratum to this article can be found at     

Mid-Holocene and last glacial maximum climate simulations with the IPSL model: part II: model-data comparisons

The climates of the mid-Holocene (MH, 6,000 years ago) and the Last Glacial Maximum (LGM, 21,000 years ago) have been extensively documented and as such, have become targets for the evaluation of climate models for climate contexts very different from the present. In Part 1 of the present work, we have studied the MH and LGM simulations performed with the last two versions of the IPSL model: IPSL_CM4, run for the PMIP2/CMIP3 (Coupled Model Intercomparion Project) projects and IPSL_CM5A, run for the most recent PMIP3/CMIP5 projets. We have shown that not only are these models different in their simulations of the PI climate, but also in their simulations of the climatic anomalies for the MH and LGM. In the Part 2 of this paper, we first examine whether palaeo-data can help discriminate between the model performances. This is indeed the case for the African monsoon for the MH or for North America south of the Laurentide ice sheet, the South Atlantic or the southern Indian ocean for the LGM. For the LGM, off-line vegetation modelling appears to offer good opportunities to distinguish climate model results because glacial vegetation proves to be very sensitive to even small differences in LGM climate. For other cases such as the LGM North Atlantic or the LGM equatorial Pacific, the large uncertainty on the SST reconstructions, prevents model discrimination. We have examined the use of other proxy-data for model evaluation, which has become possible with the inclusion of the biogeochemistry morel PISCES in the IPSL_CM5A model. We show a broad agreement of the LGM–PI export production changes with reconstructions. These changes are related to the mixed layer depth in most regions and to sea-ice variations in the high latitudes. We have also modelled foraminifer abundances with the FORAMCLIM model and shown that the changes in foraminifer abundance in the equatorial Pacific are mainly forced by changes in SSTs, hence confirming the SST-foraminifer abundance relationship. Yet, this is not the case in all regions in the North Atlantic, where food availability can have a strong impact of foraminifer abundances. Further work will be needed to exhaustively examine the role of factors other than climate in piloting changes in palaeo-indicators.     

A coupled climate model simulation of the Last Glacial Maximum,Part 2: approach to equilibrium

The climate of the last glacial maximum (LGM) is simulated with a coupled climate model. The simulated climate undergoes a rapid adjustment during the first several decades after imposition of LGM boundary conditions, as described in Part 1, and then evolves toward equilibrium over 900 model years. The climate simulated by the coupled model at this period is compared with observationally-based LGM reconstructions and with LGM results obtained with an atmosphere-mixed layer (slab) ocean version of the model in order to investigate the role of ocean dynamics in the LGM climate. Global mean surface air temperature and sea surface temperature (SST) decrease by about 10 °C and 5.6 °C in the coupled model which includes ocean dynamics, compared to decreases of 6.3 and 3.8 °C in slab ocean case. The coupled model simulates a cooling of about 6.5 °C over the tropics, which is larger than that of the CLIMAP reconstruction (1.7 °C) and larger than that of the slab ocean simulation (3.3 °C), but which is in reasonable agreement with some recent proxy estimates. The ocean dynamics of the coupled model captures features found in the CLIMAP reconstructions such as a relative maximum of ocean cooling over the tropical Pacific associated with a mean La Niña-like response and lead to a more realistic SST pattern than in the slab model case. The reduction in global mean precipitation simulated in the coupled model is larger (15%) than that simulated with the slab ocean model (~10%) in conjunction with the enhanced cooling. Some regions, such as the USA and the Mediterranean region, experience increased precipitation in accord with proxy paleoclimate evidence. The overall much drier climate over the ocean leads to higher sea surface salinity (SSS) in most ocean basins except for the North Atlantic where SSS is considerably lower due to an increase in the supply of fresh water from the Mississippi and Amazon rivers and presumably a decrease in salt transport by the weakened North Atlantic overturning circulation. The North Atlantic overturning stream function weakens to less than half of the control run value. The overturning is limited to a shallower depth (less than 1000 m) and its outflow is confined to the Northern Hemisphere. In the Southern Ocean, convection is much stronger than in the control run leading to a stronger overturning stream function associated with enhanced Antarctic Bottom Water formation. As a result, Southern Ocean water masses fill the entire deep ocean. The Antarctic Circumpolar Current (ACC) transport through the Drake Passage increases by about 25%. The ACC transport, despite weaker zonal winds, is enhanced due to changes in bottom pressure torque. The weakening of the overturning circulation in the North Atlantic and the accompanying 30% decrease in the poleward ocean heat transport contrasts with the strengthening of the overturning circulation in the Southern Ocean and a 40% increase in heat transport. As a result, sea ice coverage and thickness are affected in opposite senses in the two hemispheres. The LGM climate simulated by the coupled model is in reasonable agreement with paleoclimate proxy evidence. The dynamical response of the ocean in the coupled model plays an important role in determining the simulated, and undoubtedly, the actual, LGM climate.     

High-resolution climate simulation of the last glacial maximum

The climate of the last glacial maximum (LGM) is simulated with a high-resolution atmospheric general circulation model, the NCAR CCM3 at spectral truncation of T170, corresponding to a grid cell size of roughly 75 km. The purpose of the study is to assess whether there are significant benefits from the higher resolution simulation compared to the lower resolution simulation associated with the role of topography. The LGM simulations were forced with modified CLIMAP sea ice distribution and sea surface temperatures (SST) reduced by 1°C, ice sheet topography, reduced CO₂, and 21,000 BP orbital parameters. The high-resolution model captures modern climate reasonably well, in particular the distribution of heavy precipitation in the tropical Pacific. For the ice age case, surface temperature simulated by the high-resolution model agrees better with those of proxy estimates than does the low-resolution model. Despite the fact that tropical SSTs were only 2.1°C less than the control run, there are many lowland tropical land areas 4–6°C colder than present. Comparison of T170 model results with the best constrained proxy temperature estimates (noble gas concentrations in groundwater) now yield no significant differences between model and observations. There are also significant upland temperature changes in the best resolved tropical mountain belt (the Andes). We provisionally attribute this result in part as resulting from decreased lateral mixing between ocean and land in a model with more model grid cells. A longstanding model-data discrepancy therefore appears to be resolved without invoking any unusual model physics. The response of the Asian summer monsoon can also be more clearly linked to local geography in the high-resolution model than in the low-resolution model; this distinction should enable more confident validation of climate proxy data with the high-resolution model. Elsewhere, an inferred salinity increase in the subtropical North Atlantic may have significant implications for ocean circulation changes during the LGM. A large part of the Amazon and Congo Basins are simulated to be substantially drier in the ice age—consistent with many (but not all) paleo data. These results suggest that there are considerable benefits derived from high-resolution model regarding regional climate responses, and that observationalists can now compare their results with models that resolve geography at a resolution comparable to that which the proxy data represent.     

ENSO at 6ka and 21ka from ocean–atmosphere coupled model simulations

We analyze how the characteristics of El Niño-Southern Oscillation (ENSO) are changed in coupled ocean–atmosphere simulations of the mid-Holocene (MH) and the Last Glacial Maximum (LGM) performed as part of the Paleoclimate Modeling Intercomparison Project phase 2 (PMIP2). Comparison of the model results with present day observations show that most of the models reproduce the large scale features of the tropical Pacific like the SST gradient, the mean SST and the mean seasonal cycles. All models simulate the ENSO variability, although with different skill. Our analyses show that several relationships between El Niño amplitude and the mean state across the different control simulations are still valid for simulations of the MH and the LGM. Results for the MH show a consistent El Niño amplitude decrease. It can be related to the large scale atmospheric circulation changes. While the Northern Hemisphere receives more insolation during the summer time, the Asian summer monsoon system is strengthened which leads to the enhancement of the Walker circulation. Easterlies prevailing over the central eastern Pacific induce an equatorial upwelling that damps the El Niño development. Results are less conclusive for 21ka. Large scale dynamic competes with changes in local heat fluxes, so that model shows a wide range of responses, as it is the case in future climate projections.     

Influence of vegetation changes during the Last Glacial Maximum using the BMRC atmospheric general circulation model

 The influence of different vegetation distributions on the atmospheric circulation during the Last Glacial Maximum (LGM, 21 000 years before present) is investigated. The atmospheric general circulation model of the Bureau of Meteorology Research Center was run using a modern vegetation and in a second experiment with a vegetation reconstruction for the LGM. It is found that a change from conifer to desert and tundra causes an additional LGM cooling of 1–2 °C in Western Europe, up to −4 °C in North America and −6 °C in Siberia. An expansion of dryland vegetation causes an additional annual cooling of 1–2 °C for Australia and northern Africa. On the other hand, an increase of temperature (2 °C) is found in Alaska due to changes in circulation. In the equatorial region the LGM vegetation leads to an increased modelled temperature of 0.5–1.5 °C and decreased precipitation (30%) over land due to a reduction of the tropical rainforest, mainly in Indonesia, where the reduction of precipitation over land is associated with an increase of precipitation of 30% over the western Pacific. Received: 15 December 1999 / Accepted: 10 January 2001     

Impact of a realistic river routing in coupled ocean–atmosphere simulations of the Last Glacial Maximum climate

The presence of large ice sheets over North America and North Europe at the Last Glacial Maximum (LGM) strongly impacted Northern hemisphere river pathways. Despite the fact that such changes may significantly alter the freshwater input to the ocean, modified surface hydrology has never been accounted for in coupled ocean–atmosphere general circulation model simulations of the LGM climate. To reconstruct the LGM river routing, we use the ICE-5G LGM topography. Because of the uncertainties in the extent of the Fennoscandian ice sheet in the Eastern part of the Kara Sea, we consider two more realistic river routing scenarios. The first scenario is characterised by the presence of an ice dammed lake south of the Fennoscandian ice sheet, and corresponds to the ICE-5G topography. This lake is fed by the Ob and Yenisei rivers. In the second scenario, both these rivers flow directly into the Arctic Ocean, which is more consistent with the latest QUEEN ice sheet margin reconstructions. We study the impact of these changes on the LGM climate as simulated by the IPSL_CM4 model and focus on the overturning thermohaline circulation. A comparison with a classical LGM simulation performed using the same model and modern river basins as designed in the PMIP2 exercise leads to the following conclusions: (1) The discharge into the North Atlantic Ocean is increased by 2,000 m³/s between 38° and 54°N in both simulations that contain LGM river routing, compared to the classical LGM experiment. (2) The ice dammed lake is shown to have a weak impact, relative to the classical simulation, both in terms of climate and ocean circulation. (3) In contrast, the North Atlantic deep convection and meridional overturning are weaker than during the classical LGM run if the Ob and Yenisei rivers flow directly into the Arctic Ocean. The total discharge into the Arctic Ocean is increased by 31,000 m³/s, relative to the classical LGM simulation. Consequentially, northward ocean heat transport is weaker, and sea ice more extensive, in better agreement with existing proxy data.     

A probabilistic calibration of climate sensitivity and terrestrial carbon change in GENIE-1

In order to investigate Last Glacial Maximum and future climate, we “precalibrate” the intermediate complexity model GENIE-1 by applying a rejection sampling approach to deterministic emulations of the model. We develop ~1,000 parameter sets which reproduce the main features of modern climate, but not precise observations. This allows a wide range of large-scale feedback response strengths which generally encompass the range of GCM behaviour. We build a deterministic emulator of climate sensitivity and quantify the contributions of atmospheric (±0.93°C, 1σ) vegetation (±0.32°C), ocean (±0.24°C) and sea–ice (±0.14°C) parameterisations to the total uncertainty. We then perform an LGM-constrained Bayesian calibration, incorporating data-driven priors and formally accounting for structural error. We estimate climate sensitivity as likely (66% confidence) to lie in the range 2.6–4.4°C, with a peak probability at 3.6°C. We estimate LGM cooling likely to lie in the range 5.3–7.5°C, with a peak probability at 6.2°C. In addition to estimates of global temperature change, we apply our ensembles to derive LGM and 2xCO₂ probability distributions for land carbon storage, Atlantic overturning and sea–ice coverage. Notably, under 2xCO₂ we calculate a probability of 37% that equilibrium terrestrial carbon storage is reduced from modern values, so the land sink has become a net source of atmospheric CO₂.     

Coupled ocean-atmosphere surface variability and its climate impacts in the tropical Atlantic region

 This study examines time evolution and statistical relationships involving the two leading ocean-atmosphere coupled modes of variability in the tropical Atlantic and some climate anomalies over the tropical 120 °W–60 °W region using selected historical files (75-y near global SSTs and precipitation over land), more recent observed data (30-y SST and pseudo wind stress in the tropical Atlantic) and reanalyses from the US National Centers for Environmental Prediction (NCEP/NCAR) reanalysis System on the period 1968–1997: surface air temperature, sea level pressure, moist static energy content at 850 hPa, precipitable water and precipitation. The first coupled mode detected through singular value decomposition of the SST and pseudo wind-stress data over the tropical Atlantic (30 °N–20 °S) expresses a modulation in the thermal transequatorial gradient of SST anomalies conducted by one month leading wind-stress anomalies mainly in the tropical north Atlantic during northern winter and fall. It features a slight dipole structure in the meridional plane. Its time variability is dominated by a quasi-decadal signal well observed in the last 20–30 ys and, when projected over longer-term SST data, in the 1920s and 1930s but with shorter periods. The second coupled mode is more confined to the south-equatorial tropical Atlantic in the northern summer and explains considerably less wind-stress/SST cross-covariance. Its time series features an interannual variability dominated by shorter frequencies with increased variance in the 1960s and 1970s before 1977. Correlations between these modes and the ENSO-like Nino3 index lead to decreasing amplitude of thermal anomalies in the tropical Atlantic during warm episodes in the Pacific. This could explain the nonstationarity of meridional anomaly gradients on seasonal and interannual time scales. Overall the relationships between the oceanic component of the coupled modes and the climate anomaly patterns denote thermodynamical processes at the ocean/atmosphere interface that create anomaly gradients in the meridional plane in a way which tends to alter the north–south movement of the seasonal cycle. This appears to be consistent with the intrinsic non-dipole character of the tropical Atlantic surface variability at the interannual time step and over the recent period, but produces abnormal amplitude and/or delayed excursions of the intertropical convergence zone (ITCZ). Connections with continental rainfall are approached through three (NCEP/NCAR and observed) rainfall indexes over the Nordeste region in Brazil, and the Guinea and Sahel zones in West Africa. These indices appear to be significantly linked to the SST component of the coupled modes only when the two Atlantic modes+the ENSO-like Nino3 index are taken into account in the regressions. This suggests that thermal forcing of continental rainfall is particularly sensitive to the linear combinations of some basic SST patterns, in particular to those that create meridional thermal gradients. The first mode in the Atlantic is associated with transequatorial pressure, moist static energy and precipitable water anomaly patterns which can explain abnormal location of the ITCZ particularly in northern winter, and hence rainfall variations in Nordeste. The second mode is more associated with in-phase variations of the same variables near the southern edge of the ITCZ, particularly in the Gulf of Guinea during the northern spring and winter. It is primarily linked to the amplitude and annual phase of the ITCZ excursions and thus to rainfall variations in Guinea. Connections with Sahel rainfall are less clear due to the difficulty for the model to correctly capture interannual variability over that region but the second Atlantic mode and the ENSO-like Pacific variability are clearly involved in the Sahel climate interannual fluctuations: anomalous dry (wet) situations tend to occur when warmer (cooler) waters are present in the eastern Pacific and the gulf of Guinea in northern summer which contribute to create a northward (southward) transequatorial anomaly gradient in sea level pressure over West Africa. Received: 14 April 1998 / Accepted: 24 December 1998     

Last Glacial Maximum climate of the former Soviet Union and Mongolia reconstructed from pollen and plant macrofossil data

 An improved concept of the best analogues method was used to reconstruct the Last Glacial Maximum (LGM) climate from a set of botanical records from the former Soviet Union and Mongolia. Terrestrial pollen and macrofossil taxa were grouped into broad classes – plant functional types (PFTs), defined by the ecological and climatic parameters used in the BIOME1 model. PFT scores were then calibrated in terms of modern climate using 1245 surface pollen spectra from Eurasia and North America. In contrast to individual taxa, which exhibit great variability and may not be present in the palaeoassemblages, even in suitable climates, PFTs are more characteristic of the vegetation types. The modified method thus allows climate reconstruction at time intervals with partial direct analogues of modern vegetation (e.g. the LGM). At 18 kBP, mean temperatures were 20–29 °C colder than today in winter and 5–11 °C colder in summer in European Russia and Ukraine. Sites from western Georgia show negative, but moderate temperature anomalies compared to today: 8–11 °C in January and 5–7 °C in July. LGM winters were 7–15 °C colder and summers were 1–7 °C colder in Siberia and Mongolia. Annual precipitation sums were 50–750 mm lower than today across northern Eurasia, suggesting a weakening of the Atlantic and Pacific influences. Reconstructed drought index shows much drier LGM conditions in northern and mid-latitude Russia, but similar to or slightly wetter than today around the Black Sea and in Mongolia, suggesting compensation of precipitation losses by lower-than-present evaporation. Received: 11 May 1998 / Accepted: 25 September 1998     

Tropical climates at the Last Glacial Maximum: a new synthesis of terrestrial palaeoclimate data. I. Vegetation, lake-levels and geochemistry

 Palaeodata in synthesis form are needed as benchmarks for the Palaeoclimate Modelling Intercomparison Project (PMIP). Advances since the last synthesis of terrestrial palaeodata from the last glacial maximum (LGM) call for a new evaluation, especially of data from the tropics. Here pollen, plant-macrofossil, lake-level, noble gas (from groundwater) and δ¹⁸O (from speleothems) data are compiled for 18±2 ka (¹⁴C), 32 °N–33 °S. The reliability of the data was evaluated using explicit criteria and some types of data were re-analysed using consistent methods in order to derive a set of mutually consistent palaeoclimate estimates of mean temperature of the coldest month (MTCO), mean annual temperature (MAT), plant available moisture (PAM) and runoff (P-E). Cold-month temperature (MAT) anomalies from plant data range from −1 to −2 K near sea level in Indonesia and the S Pacific, through −6 to −8 K at many high-elevation sites to −8 to −15 K in S China and the SE USA. MAT anomalies from groundwater or speleothems seem more uniform (−4 to −6 K), but the data are as yet sparse; a clear divergence between MAT and cold-month estimates from the same region is seen only in the SE USA, where cold-air advection is expected to have enhanced cooling in winter. Regression of all cold-month anomalies against site elevation yielded an estimated average cooling of −2.5 to −3 K at modern sea level, increasing to ≈−6 K by 3000 m. However, Neotropical sites showed larger than the average sea-level cooling (−5 to −6 K) and a non-significant elevation effect, whereas W and S Pacific sites showed much less sea-level cooling (−1 K) and a stronger elevation effect. These findings support the inference that tropical sea-surface temperatures (SSTs) were lower than the CLIMAP estimates, but they limit the plausible average tropical sea-surface cooling, and they support the existence of CLIMAP-like geographic patterns in SST anomalies. Trends of PAM and lake levels indicate wet LGM conditions in the W USA, and at the highest elevations, with generally dry conditions elsewhere. These results suggest a colder-than-present ocean surface producing a weaker hydrological cycle, more arid continents, and arguably steeper-than-present terrestrial lapse rates. Such linkages are supported by recent observations on freezing-level height and tropical SSTs; moreover, simulations of “greenhouse” and LGM climates point to several possible feedback processes by which low-level temperature anomalies might be amplified aloft. Received: 7 September 1998 / Accepted: 18 March 1999     

Impact of resolving the diurnal cycle in an ocean–atmosphere GCM. Part 2: A diurnally coupled CGCM

Coupled ocean atmosphere general circulation models (GCM) are typically coupled once every 24 h, excluding the diurnal cycle from the upper ocean. Previous studies attempting to examine the role of the diurnal cycle of the upper ocean and particularly of diurnal SST variability have used models unable to resolve the processes of interest. In part 1 of this study a high vertical resolution ocean GCM configuration with modified physics was developed that could resolve the diurnal cycle in the upper ocean. In this study it is coupled every 3 h to atmospheric GCM to examine the sensitivity of the mean climate simulation and aspects of its variability to the inclusion of diurnal ocean-atmosphere coupling. The inclusion of the diurnal cycle leads to a tropics wide increase in mean sea surface temperature (SST), with the strongest signal being across the equatorial Pacific where the warming increases from 0.2°C in the central and western Pacific to over 0.3°C in the eastern equatorial Pacific. Much of this warming is shown to be a direct consequence of the rectification of daily mean SST by the diurnal variability of SST. The warming of the equatorial Pacific leads to a redistribution of precipitation from the Inter tropical convergence zone (ITCZ) toward the equator. In the western Pacific there is an increase in precipitation between Papa new guinea and 170°E of up to 1.2 mm/day, improving the simulation compared to climatology. Pacific sub tropical cells are increased in strength by about 10%, in line with results of part 1 of this study, due to the modification of the exchange of momentum between the equatorially divergent Ekman currents and the geostropic convergence at depth, effectively increasing the dynamical response of the tropical Pacific to zonal wind stresses. During the spring relaxation of the Pacific trade winds, a large diurnal cycle of SST increases the seasonal warming of the equatorial Pacific. When the trade winds then re-intensify, the increase in the dynamical response of the ocean leads to a stronger equatorial upwelling. These two processes both lead to stronger seasonal basin scale feedbacks in the coupled system, increasing the strength of the seasonal cycle of the tropical Pacific sector by around 10%. This means that the diurnal cycle in the upper ocean plays a part in the coupled feedbacks between ocean and atmosphere that maintain the basic state and the timing of the seasonal cycle of SST and trade winds in the tropical Pacific. The Madden–Julian Oscillation (MJO) is examined by use of a large scale MJO index, lag correlations and composites of events. The inclusion of the diurnal cycle leads to a reduction in overall MJO activity. Precipitation composites show that the MJO is stronger and more coherent when the diurnal cycle of coupling is resolved, with the propagation and different phases being far more distinct both locally and to larger lead times across the tropical Indo-Pacific. Part one of this study showed that that diurnal variability of SST is modulated by the MJO and therefore increases the intraseasonal SST response to the different phases of the MJO. Precipitation-based composites of SST variability confirm this increase in the coupled simulations. It is argued that including this has increased the thermodynamical coupling of the ocean and atmosphere on the timescale of the MJO (20–100 days), accounting for the improvement in the MJO strength and coherency seen in composites of precipitation and SST. These results show that the diurnal cycle of ocean–atmosphere interaction has profound impact on a range of up-scale variability in the tropical climate and as such, it is an important feature of the modelled climate system which is currently either neglected or poorly resolved in state of the art coupled models.     

Sensitivity of the tropical Pacific to a change of orbital forcing in two versions of a coupled GCM

 The changes of the variability of the tropical Pacific ocean forced by a shift of six months in the date of the perihelion are studied using a coupled tropical Pacific ocean/global atmosphere GCM. The sensitivity experiments are conducted with two versions of the atmospheric model, varied by two parametrization changes. The first one concerns the interpolation scheme between the atmosphere and ocean models grids near the coasts, the second one the advection of water vapor in the presence of downstream negative temperature gradients, as encountered in the vicinity of mountains. In the tropical Pacific region, the parametrization differences only have a significant direct effect near the coasts; but coupled feedbacks lead to a 1 °C warming of the equatorial cold tongue in the modified (version 2) model, and a widening of the western Pacific large-scale convergence area. The sensitivity of the seasonal cycle of equatorial SST is very different between the two experiments. In both cases, the response to the solar flux forcing is strongly modified by coupled interactions between the SST, wind stress response and ocean dynamics. In the first version, the main feedback is due to anomalous upwelling and leads to westward propagation of SST anomalies; whereas the version 2 model is dominated by an eastward-propagating thermocline mode. The main reason diagnosed for these different behaviors is the atmospheric response to SST anomalies. In the warmer climate simulated by the second version, the wind stress response in the western Pacific is enhanced, and the off-equatorial curl is reduced, both effects favoring eastward propagation through thermocline depth anomalies. The modifications of the simulated seasonal cycle in version 2 lead to a change in ENSO behavior. In the control climate, the interannual variability in the eastern Pacific is dominated by warm events, whereas cold events tend to be the more extreme ones with a shifted perihelion. Received: 14 December 1999 / Accepted: 24 May 2000     

Climatic changes in Eurasia and Africa at the last glacial maximum and mid-Holocene: reconstruction from pollen data using inverse vegetation modelling

In order to improve the reliability of climate reconstruction, especially the climatologies outside the modern observed climate space, an improved inverse vegetation model using a recent version of BIOME4 has been designed to quantitatively reconstruct past climates, based on pollen biome scores from the BIOME6000 project. The method has been validated with surface pollen spectra from Eurasia and Africa, and applied to palaeoclimate reconstruction. At 6 cal ka BP (calendar years), the climate was generally wetter than today in southern Europe and northern Africa, especially in the summer. Winter temperatures were higher (1–5°C) than present in southern Scandinavia, northeastern Europe, and southern Africa, but cooler in southern Eurasia and in tropical Africa, especially in Mediterranean regions. Summer temperatures were generally higher than today in most of Eurasia and Africa, with a significant warming from ∼3 to 5°C over northwestern and southern Europe, southern Africa, and eastern Africa. In contrast, summers were 1–3°C cooler than present in the Mediterranean lowlands and in a band from the eastern Black Sea to Siberia. At 21 cal ka BP, a marked hydrological change can be seen in the tropical zone, where annual precipitation was ∼200–1,000 mm/year lower than today in equatorial East Africa compared to the present. A robust inverse relationship is shown between precipitation change and elevation in Africa. This relationship indicates that precipitation likely had an important role in controlling equilibrium-line altitudes (ELA) changes in the tropics during the LGM period. In Eurasia, hydrological decreases follow a longitudinal gradient from Europe to Siberia. Winter temperatures were ∼10–17°C lower than today in Eurasia with a more significant decrease in northern regions. In Africa, winter temperature was ∼10–15°C lower than present in the south, while it was only reduced by ∼0–3°C in the tropical zone. Comparison of palaeoclimate reconstructions using LGM and modern CO₂ concentrations reveals that the effect of CO₂ on pollen-based LGM reconstructions differs by vegetation type. Reconstructions for pollen sites in steppic vegetation in Europe show warmer winter temperatures under LGM CO₂ concentrations than under modern concentrations, and reconstructions for sites in xerophytic woods/scrub in tropical high altitude regions of Africa are wetter for LGM CO₂ concentrations than for modern concentrations, because our reconstructions account for decreased plant water use efficiency.     

CLIMAP SSTs re-revisited

Since the 1976 publication of the CLIMAP ice age sea surface temperature (SST) reconstruction showing a 1–2 ^∘C tropical cooling a substantial debate has arisen as to whether tropical SSTs may instead have been 4–5^∘ colder than present. Herein I review the arguments for large SST variations and question a number of key findings, particularly the validity of ice-age coral SST estimates and “down-projecting” tropical snowline changes to the surface. GCM results indicate that an intermediate solution requiring ∼2.5 ^∘C warm pool cooling is consistent with most quantitative low elevation surface land data and is small enough to allow the persistence of tropical biota in the ocean during glacial times. The proposal reduces estimated ice-age climate sensitivity (for a doubling of CO₂) from a “high-end” sensitivity of about 4.5 ^∘C (for a 5–6 ^∘C tropical cooling) to a “mid-range” sensitivity of about 3.0 ^∘C for a 2.5 ^∘C warm-pool decrease. Received: 28 July 1999 /Accepted: 12 August 1999     

Predictability of SST forced climate signals in two atmospheric general circulation models

 The predictability of atmospheric responses to global sea surface temperature (SST) anomalies is evaluated using ensemble simulations of two general circulation models (GCMs): the GENESIS version 1.5 (GEN) and the ECMWF cycle 36 (ECM). The integrations incorporate observed SST variations but start from different initial land and atmospheric states. Five GEN 1980–1992 and six ECM 1980–1988 realizations are compared with observations to distinguish predictable SST forced climate signals from internal variability. To facilitate the study, correlation analysis and significance evaluation techniques are developed on the basis of time series permutations. It is found that the annual mean global area with realistic signals is variable dependent and ranges from 3 to 20% in GEN and 6 to 28% in ECM. More than 95% of these signal areas occur between 35 °S–35 °N. Due to the existence of model biases, robust responses, which are independent of initial condition, are identified over broader areas. Both GCMs demonstrate that the sensitivity to initial conditions decreases and the predictability of SST forced responses increases, in order, from 850 hPa zonal wind, outgoing longwave radiation, 200 hPa zonal wind, sea-level pressure to 500 hPa height. The predictable signals are concentrated in the tropical and subtropical Pacific Ocean and are identified with typical El Ni?o/ Southern Oscillation phenomena that occur in response to SST and diabatic heating anomalies over the equatorial central Pacific. ECM is less sensitive to initial conditions and better predicts SST forced climate changes. This results from (1) a more realistic basic climatology, especially of the upper-level wind circulation, that produces more realistic interactions between the mean flow, stationary waves and tropical forcing; (2) a more vigorous hydrologic cycle that amplifies the tropical forcing signals, which can exceed internal variability and be more efficiently transported from the forcing region. Differences between the models and observations are identified. For GEN during El Ni?o, the convection does not carry energy to a sufficiently high altitude, while the spread of the tropospheric warming along the equator is slower and the anomaly magnitude smaller than observed. This impacts model ability to simulate realistic responses over Eurasia and the Indian Ocean. Similar biases exist in the ECM responses. In addition, the relationships between upper and lower tropospheric wind responses to SST forcing are not well reproduced by either model. The identification of these model biases leads to the conclusion that improvements in convective heat and momentum transport parametrizations and basic climate simulations could substantially increase predictive skill. Received: 25 April 1996 / Accepted: 9 December 1996     

Modeling the Tropical Pacific Ocean Using a Regional Coupled Climate Model

A high-resolution tropical Pacific general circulation model (GCM) coupled to a global atmospheric GCM is described in this paper. The atmosphere component is the 5°×4°global general circulation model of the Institute of Atmospheric Physics (IAP) with 9 levels in the vertical direction. The ocean component with a horizontal resolution of 0.5°, is based on a low-resolution model (2°×1°in longitude-latitude).Simulations of the ocean component are first compared with its previous version. Results show that the enhanced ocean horizontal resolution allows an improved ocean state to be simulated; this involves (1) an apparent decrease in errors in the tropical Pacific cold tongue region, which exists in many ocean models,(2) more realistic large-scale flows, and (3) an improved ability to simulate the interannual variability and a reduced root mean square error (RMSE) in a long time integration. In coupling these component models, a monthly "linear-regression" method is employed to correct the model's exchanged flux between the sea and the atmosphere. A 100-year integration conducted with the coupled GCM (CGCM) shows the effectiveness of such a method in reducing climate drift. Results from years 70 to 100 are described.The model produces a reasonably realistic annual cycle of equatorial SST. The large SSTA is confined to the eastern equatorial Pacific with little propagation. Irregular warm and cold events alternate with a broad spectrum of periods between 24 and 50 months, which is very realistic. But the simulated variability is weaker than the observed and is also asymmetric in the sense of the amplitude of the warm and cold events.     

High-resolution simulations of the last glacial maximum climate over Europe: a solution to discrepancies with continental palaeoclimatic reconstructions?

The analyses of low-resolution models simulations of the last glacial maximum (LGM, 21 kyr BP) climate have revealed a large discrepancy between all the models and pollen-based palaeoclimatic reconstructions. In general, the models are too warm relative to the observations, especially in winter, where the difference is of the order of 10°C over western Europe. One of the causes of this discrepancy may be related to the low spatial resolution of these models. To assess the impact of using high-resolution models on simulated climate sensitivity, we use three approaches to obtain high-resolution climate simulations over Europe: first an atmospheric general circulation model (AGCM) with a stretched grid over Europe, second a homogeneous T106 AGCM (high resolution everywhere on the globe) and last a limited area model (LAM) nested in a low-resolution AGCM. With all three methods, we have performed simulations of the European climate for present and LGM conditions, according to the experimental design recommended by the Palaeoclimate Modeling Intercomparison Project (PMIP). Model results have been compared with updated pollen-based palaeoclimatic indicators for temperature and precipitation that were initially developed in PMIP. For each model, a low-resolution global run was also performed. As expected, the low-resolution simulations underestimate the large cooling indicated by pollen data, especially in winter, despite revised slightly warmer reconstructions of the temperatures of the coldest month, and show results in the range of those obtained in PMIP with similar models. The two high-resolution AGCMs do not improve the temperature field and cannot account for the discrepancy between model results and data, especially in winter. However, they are able to reproduce trends in precipitation more closely than their low-resolution counterparts do, but the simulated climates are still not as arid as depicted by the data. Conversely, the LAM temperature results compare well with climate reconstructions in winter but the simulated hydrological cycle is not consistent with the data. Finally, these results are discussed in regard of other possible causes for discrepancies between models and palaeoclimatic reconstructions for the LGM European climate.     

Mid-Holocene tropical Pacific climate state,annual cycle,and ENSO in PMIP2 and PMIP3

Using the Paleoclimate Modeling Inter-comparison Project Phase 2 and 3 (PMIP2 and PMIP3), we investigated the tropical Pacific climate state, annual cycle, and El Niño-Southern Oscillation (ENSO) during the mid-Holocene period (6,000 years before present; 6 ka run). When the 6 ka run was compared to the control run (0 ka run), the reduced sea surface temperature (SST) and the reduced precipitation due to the basin-wide cooling, and the intensified cross-equatorial surface winds due to the hemispheric discrepancy of the surface cooling over the tropical Pacific were commonly observed in both the PMIP2 and PMIP3, but changes were more dominant in the PMIP3. The annual cycle of SST was weaker over the equatorial eastern Pacific, because of the orbital forcing change and the deepening mixed layer, while it was stronger over the equatorial western pacific in both the PMIP2 and PMIP3. The stronger annual cycle of the equatorial western Pacific SST was accompanied by the intensified annual cycle of the zonal surface wind, which dominated in the PMIP3 in particular. The ENSO activity in the 6 ka run was significantly suppressed in the PMIP2, but marginally reduced in the PMIP3. In general, the weakened air-sea coupling associated with basin-wide cooling, reduced precipitation, and a hemispheric contrast in the climate state led to the suppression of ENSO activity, and the weakening of the annual cycle over the tropical eastern Pacific might lead to the intensification of ENSO through the frequency entrainment. Therefore, the two opposite effects are slightly compensated for by each other, which results in a small reduction in the ENSO activity during the 6 ka in the PMIP3. On the whole, in PMIP2/PMIP3, the variability of canonical (or conventional) El Niño tends to be reduced during 6 ka, while that of CP/Modoki El Niño tends to be intensified.