Glacial – interglacial shifts in global and regional precipitation δ 18

Introduction Conclusions References


Conclusions References
Tables Figures

Introduction
Reconstructed isotope compositions of Pleistocene precipitation preserved in groundwaters, cave calcite, glacial ice, ground ice and lake sediments have been used to better understand past climate changes for more than a half century (e.g., Münnich, 1957;Thatcher et al., 1961;Münnich et al., 1967;Pearson and White, 1967;Tamers, 1967;Gat et al., 1969).Each type of isotopic proxy record is distinguished by its temporal resolution, preservation of one or both 18 O/ 16 O and 2 H/ 1 H ratios, and frequency of records on land surface.For example, groundwater records contain both 18 O/ 16 O and 2 H/ 1 H ratios with widespread global occurrence, but have a coarser temporal resolution than other paleoclimate proxies (Rozanski et al., 1985;Edmunds and Milne, 2001;Edmunds, 2009;Corcho Alvarado et al., 2010;Jiráková et al., 2011).Speleothem records, in contrast, have high temporal resolution but usually only report calcite 18 O/ 16 O ratios (without fluid inclusion 2 H/ 1 H data) and are less common than groundwater records (e.g., Harmon et al., 1978Harmon et al., , 1979)).Pleistocene glacier-and ground-ice records have high temporal resolution, can be analyzed for 18 O/ 16 O and 2 H/ 1 H ratios, but are rare on non-polar lands (Dansgaard et al., 1982;Thompson et al., Figures Back Close Full This study focuses primarily upon groundwater isotope records due to the relative density of groundwater records in the published literature in comparison to the more limited number of published isotopic records from speleothems and ice cores (compilations by Pedro et al., 2011;Stenni et al., 2011;Clark et al., 2012;Shah et al., 2013;Caley et al., 2014a).There exist roughly twice as many groundwater reconstructions of ice age precipitation δ 18 O (n = 61) as the combined total of speleothem and ice core precipitation δ 18 O records spanning the last ice age and late-Holocene time periods (n = 27; where δ 18  of paired precipitation-groundwater isotopic data demonstrated that modern annual amount-weighted precipitation and local, modern groundwater recharge isotope compositions follow systematic relationships with some bias toward winter and wetseason precipitation (Jasechko et al., 2014).Systematic rainfall-recharge relationships shown by Jasechko et al. (2014) support our primary assumption in this study that groundwater isotope compositions closely reflect meteoric water.Because groundwater records can only identify climate change occurring over thousands of years due to hydrodynamic dispersion during multi-millennial residence times (e.g., Davison and Airey, 1982;Stute and Deak, 1989), we limit the focus of this study to meteoric water isotope composition changes from the last ice age to the late-Holocene.The last ice age time period is defined as 19 500 to ∼ 50 000 years before present, defined using the end of the last glacial maximum as the more recent age limit (19 500 years before present; Clark et al., 2009), and the approximate maximum age of groundwater that can be identified by 14 C dating as an approximate upper age limit (i.e., groundwater ages more recent than ∼ 50 000 years).We adopt a definition of the late-Holocene as occurring within the last 5000 years following Thompson et al. (2006).Other work proposes the late-Holocene be defined as within the last 4200 years (Walker et al., 2012), which is consistent with the 5000 years before present definition (Thompson et al., 2006) within the practical uncertainty of 14 C-based groundwater ages (± ∼ 10 3 years).Further, although precipitation isotope Introduction

Conclusions References
Tables Figures

Back Close
Full compositions have varied over the late-Holocene, groundwater mixing integrates this variability, prohibiting paleoclimate interpretation at finer temporal resolutions.Pleistocene-to-Holocene changes in precipitation isotope compositions provide important insights into conditions and processes of the past.Perhaps the two bestconstrained global-in-scale differences between the last ice age and the late-Holocene are changes to oceanic and atmospheric temperatures (MARGO Members, 2009;Shakun and Carlson, 2010;Annan and Hargreaves, 2013), and changes to seawater δ 18 O (Emiliani, 1955;Dansgaard and Tauber, 1969;Schrag et al., 1996Schrag et al., , 2002)).
Atmospheric temperatures have increased by a global average of ∼ 4 • C since the last glacial maximum, with greatest warming at the poles and more modest warming at lower latitudes (Fig. S1 in the Supplement; e.g., Shakun and Carlson, 2010;Annan and Hargreaves, 2013).Seawater δ 18 O during the last glacial maximum was 1.0 ± 0.1 ‰ higher than the modern ocean, as constrained by paleo-ocean water samples collected from pore waters trapped within sea floor sediments (Schrag et al., 2002).

Conclusions References
Tables Figures

Dataset and methods
In order to examine spatial patterns of change to meteoric water δ 18 O values we compiled δ 18 O, δ 2 H, δ 13 C and 14 C data from 1713 groundwater samples collected from 61 aquifer systems reported in 75 publications (Fig. 1).δ 13 C, 3 H Introduction

Conclusions References
Tables Figures

Conclusions References
Tables Figures

Back Close
Full the versions submitted to the CMIP5 archive and participated in PMIP3.Notable exceptions include IsoGSM using different boundary conditions (Yoshimura et al., 2008), ECHAM5 not participating in CMIP5, and CAM3iso not participating in PMIP3.
The five models span a range of spatio-temporal resolutions and isotopic/atmospheric parameterizations described in detail in the above references.A selection of the intermodel similarities and differences are summarized in Table S1   18 O values are at a minimum (e.g., Bowen and Wilkinson, 2002).This broad spatial pattern is consistent with the nonlinear isotopic distillation of air masses undergoing progressive rainout (i.e., Rayleigh distillation).Because seawater δ 18 O values were ∼ 1 ‰ higher-than-modern during the last ice age (Schrag et al., 1996(Schrag et al., , 2002)), our finding that the majority of reconstructed ∆ 18 O ice age values are negative suggests that isotopic distillation of air masses was greater during the last ice age than under present climate.This finding is consistent with land surface temperature reconstructions that show larger glacial-to-modern changes to land temperatures at high latitude and continental settings (Fig. S1; Annan and Hargreaves, 2013).Introduction

Conclusions References
Tables Figures

Back Close
Full  Full as advised for PMIP2 (Braconnot et al., 2007), whereas the GISSE2 replaces Ice 5G Laurentide ice with that of Licciardi et al. (1999) and ECHAM5-wiso uses ice topography from PMIP3 (Braconnot et al., 2007(Braconnot et al., , 2012; PMIP3 follows ice sheet topography blended from multiple ice sheet reconstructions; e.g., Argus and Peltier, 2010;Toscano et al., 2011).Glacial topography is an important driver of simulated temperature, precipitation and atmospheric circulation at the Last Glacial Maximum (e.g., Justino et al., 2005;Pausata et al., 2011b;Ullman et al., 2014).Therefore it is likely that inter-model differences in paleo-ice sheet topographies impacts atmospheric circulation, high-latitude simulated precipitation δ 18 O at the Last Glacial Maximum, and thus simulated ∆ 18 O ice age values reported in this study (Fig. 3).
Differences in the specification of initial seawater δ 18 O may also lead to inter-model differences in simulated ∆ 18 O ice age values.Seawater δ 18 O is set to be globallyhomogenous in CAM3Iso, IsoGSM and LMDZ, and heterogeneous in ECHAM5-wiso (using modern gridded seawater δ 18 O heterogeneity of LeGrande and Schmidt, 2006) and GISSE2-R (coupled atmosphere-ocean model; seawater δ   and S3).Australian climate at the last ice age was more arid (Nanson et al., 1992), dustier (Chen et al., 1993) and cooler (Miller et al., 1997) than present day.Simulated ∆ 18 O ice age values across Australia are variable amongst the five models.
Reconstructed ∆ 18 O ice age values across Oceania have been attributed to temporal changes in the strength of monsoons and convective rains (Aggarwal et al., 2004;Partin et al., 2007;Williams et al., 2010) potentially impacted by ice-age-to-late-Holocene shifts in the position of the intertropical convergence zone (Lewis et al., 2010(Lewis et al., , 2011)).

Southeast Asia
Reconstructed Full coasts, but are more variable across western and northern China (Fig. 3).Chinese speleothem records show near-zero or positive reconstructed ∆ 18 O ice age values interpreted to reflect the reduced strength of the East Asian (Wang et al., 2001;Dykoski et al., 2005;Cosford et al., 2008) or Indian monsoons (Pausata et al., 2011a) Full

Africa
Reconstructed ∆ 18 O ice age values from Africa range from −2.9 to +0.4 ‰ (Fig. 1).14 of Europe and North America (Fig. 3), mechanisms driving this extra-tropical/tropical difference remain elusive and can be examined through future inter-model or modelreconstruction comparative studies.Northern African hydrological processes are influenced by multiple interlinked controls such as the strength of Atlantic meridional overturning circulation (Mulitza et al., 2008) and meridional shifts in the position of the intertropical convergence zone (Arbuszewski et al., 2013).Paleowater chemistry indicates that northern Africa was at least 2 • C cooler than today (Guendouz et al., 1998) and that westerly moisture transport was stronger than the present during the last ice age (Sultan et al., 1997;Abouelmagd et al., 2014).Paleowater isotope compositions of Northern Africa may have been impacted by higher-than-modern sea surface humidity as interpreted from lower-than-modern paleowater deuterium excess values (Rozanski, 1985).However, the deuterium excess of seawater during the last ice age may have been different from present day given that the Laurentide ice sheet had a deuterium excess value of ∼ 10 to ∼ 15 ‰ (e.g., Grasby and Chen, 2005;Ferguson et al., 2007).

Conclusions References
Tables Figures

Back Close
Full moisture travelling across the Congo rainforest (Levin et al., 2009).Lower-than-modern continental moisture recycling during the last ice age may partially explain negative reconstructed ∆ 18 O ice age values across some regions of inland tropical Africa (e.g., Risi et al., 2013).Although negative reconstructed ∆ 18 O ice age values in tropical Africa could be interpreted to reflect higher-than-modern upwind rainout during the last ice age (see Risi et al., 2008Risi et al., , 2010b;;Lee et al., 2009;Scholl et al., 2009;Lekshmy et al., 2014;Samuels-Crow et al., 2014), this explanation necessitates stronger-thanmodern convection during the last ice age, an explanation that would contradict the established cooler-than-modern land surface temperatures.Therefore, changes to atmospheric transport distances and vapor origins are more likely responsible for negative reconstructed ∆ 18 O ice age values across tropical Africa (Lewis et al., 2010).

Europe and the Mediterranean
Reconstructed in Turkey near to the Black Sea (−5.7 ‰) and potentially reflects a change to regional moisture source (Fleitmann et al., 2009;Arslan et al., 2013).Westerly air mass trajectories distal to the Fennoscandian ice sheet topography have not changed Introduction

Conclusions References
Tables Figures

Back Close
Full Mediterranean (Frumkin et al., 1999;Bar-Matthews et al., 2003;Ayalon et al., 2013) differ from negative reconstructed ∆ 18 O ice age values in nearby groundwater aquifers (e.g., Burg et al., 2013), advocating for further comparative research to ensure speleothem and groundwater isotope compositions each capture meteoric water δ 18 O unaltered by fractionating processes such as partial evaporation.Recent work suggests that speleothem δ 18 O data may be impacted by disequilibrium isotope effects (Asrat et al., 2008;Daëron et al., 2011;Kluge and Affek, 2012;Kluge et al., 2013) or by partial evaporation of drip waters resulting in 18 O-enrichment (e.g., Cuthbert et al., 2014a) and greater fractionation due to evaporative cooling (Cuthbert et al., 2014b), potentially explaining a portion of the difference between groundwater and speleothem reconstructed ∆ 18 O ice age values.
Changes to freeze-thaw conditions of the ground surface between last ice age and modern climates may have impacted the seasonality of the fraction of precipitation recharging aquifers and thus ∆ 18 O ice age (Darling, 2004(Darling, , 2011;;Jasechko et al., 2014).
Geomorphic evidence suggests permafrost covered portions of Hungary at the last glacial maximum, suppressing land temperatures by as much as 15 • C (Fábián et al., 2014).European pollen records indicate that northern Europe was tundra-like and that southern Europe was semi-arid at the last glacial maximum (Harrison and Prentice, 2003;Clark et al., 2012).The European glacial-to-modern transition from semi-arid deserts to temperate forests could have lowered ∆ 18 O ice age values as groundwater recharge ratios transitioned from more extreme winter-biased groundwater recharge ratios (e.g., semi-arid lands during last ice age) to less extreme but still winter-biased groundwater recharge ratios (e.g., forests during late-Holocene; Jasechko et al., 2014).

South America
Reconstructed ∆ 18 O ice age values across South America range from −6.2 to 0.3 ‰ (Fig. 1).The highest-magnitude, negative reconstructed ∆ 18 O ice age values (Andean ice cores; Thompson et al., 1995Thompson et al., , 1998) ) are found in similar locations to the lowest present Introduction

Conclusions References
Tables Figures

Back Close
Full day precipitation δ 18 O values across South America (Bowen and Wilkinson, 2002).
Here the importance of upstream convection upon modern Andean precipitation δ 18 O has been highlighted at inter-annual (e.g., Hoffmann et al., 2003;Vuille and Werner, 2005), seasonal (e.g., Vimeux et al., 2005;Samuels-Crow et al., 2014) and daily time scales (e.g., Vimeux et al., 2011).It is therefore possible that upstream convection controls past changes to Andean precipitation isotope compositions recorded in ice cores.Further, upwind changes to continental moisture recycling driven by shifts in plant transpiration fluxes may have altered continental gradients in precipitation δ 18 O.Glacial-interglacial changes to the density of Amazonian vegetation are supported by oceanic pollen records (Haberle and Maslin, 1999) (Salati et al., 1974).Eastern Brazil was 5 • C cooler than today during the last ice age (Stute et al., 1995b).Four of the five general circulation models simulate positive ∆ 18 O ice age values across eastern Brazil (Fig. 3), highlighting differences between simulated and reconstructed ∆ 18 O ice age values in parts of the tropics.The negative reconstructed ∆ 18 O ice age value in eastern Brazil has been previously interpreted to reflect higher-than-modern precipitation during the Pleistocene (Salati et al., 1974).Lewis et al. (2010) show that localized precipitation fluxes govern precipitation δ Full

North America
Reconstructions of ∆ 18 O ice age from North American proxy records range from −5.5 to +1.0 ‰.Canadian records of subglacial recharge from the Laurentide ice sheet (e.g., Grasby and Chen, 2005;Ferguson et al., 2007) are not included in this synthesis because of possible transport along paleo-glacial flow paths in the Laurentide ice sheet.
Reconstructed ∆ 18 O ice age values along the USA east coast show the highest, positive values in Georgia (latitude: 32  Hertig et al., 2002).Decreasing ∆ 18 O ice age values with increasing latitude along the USA east coast may be explained in part by the isotopic distillation of air masses advected northward from the subtropics under cooler-than-modern final atmospheric condensation temperatures.The chemistry of Maryland groundwaters has been interpreted to show that the region was more arid and 9-12 • C cooler during the last ice age relative to modern climate conditions (Purdy et al., 1996;Aeschbach-Hertig et al., 2002;Plummer et al., 2012).This glacial-to-modern temperature change is larger than most other temperature proxy records at similar latitudes (Annan and Hargreaves, 2013).Impacts of higher-than-modern ice age seawater δ 18 O upon terrestrial precipitation ∆ 18 O ice age may have been offset by lower sea levels that increased atmospheric transport distances during the last ice age (Clark et al., 1997;Aeschbach-Hertig et al., 2002;Tharammal et al., 2012) Full greater upstream air mass distillation during the last ice age than under modern climate.Pollen, vadose zone and groundwater records show that Pleistocene southwestern USA was ∼ 4 • C cooler, had greater groundwater recharge fluxes, and had more widespread forests than present day (Stute et al., 1992(Stute et al., , 1995a;;Scanlon et al., 2003;Williams, 2003) Full  Geosci., 3, 114-117, 2010.

CPD Introduction Conclusions References
Tables Figures
Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | 18 O ice age values are unevenly distributed among western Europe (n = 10), eastern Europe and the Middle-East (n = 12), Africa (n = 18), southeastern Asia (n = 6), Australia, Oceania and the Malay Archipelago (n = 3), South America (n = 2), temperate and subtropical North America (n = 8) and the High Arctic (n = 2, ground ice records).Half the compiled Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |

18 O
is calculated by the ocean model).Simulated precipitation δ 18 O values either show little change (±0.1 ‰) or show increases of up to 1.5 ‰ when modern spatial heterogeneous of surface ocean δ 18 O values are included (LeGrande and Schmidt, 2006).The incorporation of heterogeneous seawater δ 18 O into model simulations can impact simulated ∆ 18 O ice age values in cases where simulated moisture sources or simulated over-ocean meteorology change between the two climate states.The models also show deficiencies in simulating reconstructed ∆ 18 O ice age values in the tropics.This finding could, in part, be related to the high sensitivity of precipitation δ 18 O to convective parameterizations (Lee et al., 2009; Field et al., 2014), although future research is required to test this.Another reason may lie on the fact that the reconstructed ∆ 18 O ice age integrates the hydroclimatological signal over multimillennial time scales, whereas the simulated ∆ 18 O ice age explicitly simulate the last glacial maximum and pre-industrial/present-day climate conditions.The stronger extratropical agreement between the sign of simulated and reconstructed ∆ 18 O ice age Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Arslan, S., Yazicigil, H., Stute, M., and Schlosser, P.: Environmental isotopes and noble gases in the deep aquifer system of Kazan Trona Ore Field, Ankara, central Turkey and links to paleoclimate, Quaternary Res., 79, 292-303, 2013.Asmerom, Y., Polyak, V. J., and Burns, S. J.: Variable winter moisture in the southwestern United States linked to rapid glacial climate shifts, Nat.

Figure 1 .Figure 2 .
Figure 1.Meteoric water δ 18 O change from the latter half of the last ice age (19 500 to ∼ 50 000 years ago) to the late-Holocene (within past ∼ 5000 years).The low temporal resolution of groundwater records means that δ 18 O variations within each time period are smoothed and likely represent unequal temporal weighting.References for reconstructed meteoric water δ 18 O changes for ice cores, groundwater and cave calcite are presented in the Supplement.
The objective of this study is to analyze spatial patterns of reconstructed precipitation δ18O changes since the last ice age from published groundwater, ground ice, glacial ice and cave calcite records, and to compare these observations with output from five state-of-the-art isotope-enabled general circulation model simulations of last glacial maximum and more recent climate conditions.Synthesizing paleowater δ 18 O 18O changes from the last ice age to the late-Holocene are described herein as reconstructed ∆ 18 O ice age , where reconstructed ∆ 18 O ice age = groundwater records are located in the tropics or subtropics (that is, within 35 • of the equator) and half are located in the extra-tropics.Speleothem and ice core isotope proxy records were also compiled.Lacustrine sediment δ18 O records are not considered in this study because these records may preserve meteoric waters impacted by evaporative isotope effects.Speleothem and ice core reconstructed ∆ 18 O ice age values were calculated by subtracting average δ18O values for each of the two time intervals defined for the groundwater records: the late-Holocene (< 5000 years before present) and last ice age (19 500-50 000 years before present).This step effectively lowered the temporal resolution of speleothem and ice core precipitation isotope records to be consistent with the temporal resolution of the groundwater records.A correction factor was applied to speleothem δ 18 O values to account for different H 2 O-CaCO 3 isotopic fractionation factors during the last ice age and the late-Holocene imparted by the different land surface temperatures during each time period (details presented within the Supplement).Simulated ∆ 18 O ice age values were compiled from five isotope-enabled general circulation models (simulated ∆ 18 O ice age = δ 18 O last glacial maximum − δ 18 O pre-industrial ):

Results and discussion 3.1 Reconstructed ∆ 18 O ice age values
(Supplement).For clarity, data-based ∆ 18 O ice age values from groundwater, speleothems, ground ice and ice cores are referred to herein as reconstructed ∆ 18 O ice age , whereas simulated precipitation isotope compositions from general circulation models are referred to as simulated ∆ 18 O ice age .We acknowledge that the general circulation models explicitly the equator and within 400 km of the nearest coastline (e.g., Bangladesh ∆ 18 O ice age of +1.6 ‰, less than 300 km from the coast; Figs. 1 and 2; Fig. S2).In comparison, negative reconstructed ∆ 18 O ice age values are found in both coastal regions and farther inland.Negative reconstructed ∆ 18 O ice age values of the greatest magnitude are located analyze the last glacial maximum and the pre-industrial climate conditions (i.e., simulated ∆ 18 O ice age = δ18O last glacial maximum − δ 18 O pre-industrial ), whereas proxy record reconstructions of ∆ 18 O ice age integrate hydroclimatology over multi-millennial time scales that are different from the model simulations (i.e., reconstructed ∆ 18 O ice age = δ 18 O ice age − δ 18 O late-Holocene ). 3 Reconstructed groundwater (n = 61), speleothem (n = 15) and ice core (n = 12) ∆ 18 O ice age values are presented in Fig. 1 (references presented in the Supplement).Reconstructed ∆ 18 O ice age values range from −7.1 ‰ (i.e., δ 18 O ice age < δ 18 O late-Holocene ) to +1.8 ‰ (i.e., δ 18 O ice age > δ 18 O late-Holocene ).Three-quarters of compiled records have negative reconstructed ∆ 18 O ice age values and one-quarter of compiled records have positive reconstructed ∆ 18 O ice age values.More than 80 % of reconstructed ∆ 18 O ice age values of greater than zero are located in the within 35 • of Introduction Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | 18 O ice age values (Antarctic and Greenland ∆ 18 O ice age values range from −3.6 to −7.1 ‰ ; Fig. 2).Reconstructed ∆ 18 O ice age values synthesized in this study generally show that tropical ∆ 18 O ice age values are closer to 0 ‰ (i.e., no change) than high latitude, continental regions that generally have high magnitude, negative reconstructed ∆ 18 O ice age values.High magnitude, negative reconstructed ∆ 18 O ice age values are most common where present day precipitation δ

3.2 Simulated ∆ 18 O ice age values
Simulated precipitation ∆ 18 O ice age values from five general circulation models are presented in Fig. 3.At least four of the five models agree on the sign of simulated ∆ 18 O ice age values for 68.8 % of grid cells covering Earth's surface (68.7 % of overocean areas and 68.9 % of over land areas; multi-model calculation completed using 3 of 4 models as a threshold at high-latitudes where IsoGSM data was not available).Simulated ∆ 18 O ice age values are consistently negative over the North Atlantic Ocean and the Fennoscandian and Laurentide ice sheets and consistently positive over most of the tropical oceans, whereas low agreement is found over tropical land surfaces.The negative simulated ∆ 18 O ice age values over the Northern Hemisphere ice sheets and North Atlantic are likely driven by the difference in topography and sea ice cover, respectively, between the last ice age and pre-industrial climate.The glacial-interglacial change in ice sheet topography and sea ice cover impacted surface temperatures.Surface temperatures were more than ∼ 20 • C cooler over most of present-day Canada during the last ice age and this temperature shift is likely to impact simulated ∆ 18 O ice age values (Fig. S1).A comparison of simulated ∆ 18 O ice age values over tropical Africa, South America and Oceania shows inter-model disagreement (Fig. 3).Different tropical simulated ∆ 18 O ice age values amongst the models reflect the different isotopic parameterizations, inter-model spread in simulated precipitation fluxes, glacial-interglacial shifts in seawater δ 18 O (inter-model seawater δ 18 O ice age minus seawater δ 18 O pre-industrial ranges from +0.7 to +1.1 ‰) and seawater δ 18 O heterogeneity used in each model.Inter-model spread in simulated ∆ 18 O ice age values in some regions highlights the importance of this global synthesis of proxy record reconstructed ∆ 18 O ice age values as a constraint for climate model simulated ∆ 18 O ice age values.Another potential source for the model disagreement is introduced by the different ice-sheet topography used in each model.CAM3Iso, IsoGSM and LMDZ4 have used Ice 5G (Peltier, 1994) Introduction

Regional reconstructed and simulated ∆ 18 O ice age values 3.3.1 Australia and Oceania
values (i.e., positive or negative) relative to the tropics is most likely linked to the substantial changes to extra-tropical ice-sheet topography and sea-ice cover between the two climate states in northern North America and Europe.In this case the extreme temperature anomaly between last glacial and pre-industrial climate largely overwhelms a potential bias induced by smearing reconstructed δ 18 O late-Holocene values and reconstructed δ 18 O ice age values over multiple millennia.
18O ice age values from Australia and Oceania fall between −1 and 1 ‰ (Figs. 1 18 reconstructed ∆ 18 O ice age values from Africa are negative.Near-zero reconstructed ∆ 18 O ice age values are generally found near to coasts (e.g., Senegal, ∆ 18 O ice age of +0.3 ‰ ; Madioune et al., 2014), whereas higher magnitude, negative reconstructed ∆ 18 O ice age values in Africa are found farther inland (e.g., Niger, ∆ 18 O ice age of −3.0 ‰: ∼ 800 kilometers from the Atlantic coast).General circulation models show poorer agreement with reconstructed ∆ 18 O ice age values over tropical Africa compared to . Negative reconstructed ∆ 18 O ice age values in parts of South America may have been driven in part by lower-than-modern continental moisture recycling during the last ice age.The reconstructed groundwater ∆ 18 O ice age value located in eastern Brazil is −2.7 ‰ Salati et al. (1974)Modern precipitation δ 18 O values are lowest in eastern Brazil when precipitation rates are at a maximum; extending Lewis et al.'s interpretation linking local precipitation amount to precipitation δ 18 O suggests the negative reconstructed ∆ 18 O ice age value found in eastern Brazil may record wetter-than-modern conditions at the last ice age as proposed bySalati et al. (1974).Introduction • N; reconstructed ∆ 18 O ice age of +1.0 ‰; Clark . Reconstructions of ∆ 18 O ice age values in the central and southwestern USA have the highest magnitude, negative reconstructed ∆ 18 O ice age values of temperate North America, ranging from −0.7 to −3.4 ‰.Central and southwestern USA reconstructed ∆ 18 O ice age values contrast the positive reconstructed ∆ 18 O ice age values found along the eastern USA coast at similar latitudes.Consistently negative ∆ 18 O ice age values in central and southwest USA suggest that advected moisture to the region underwent Introduction

Supplement related to this article is available online at doi:10.5194/cpd-11-831-2015-supplement. Introduction
Wagner et al. (2010)14)d ∆18O ice age values found in the southwest USA have been ascribed to lower-than-modern summer precipitation (New Mexico,Phillips et al., 1986), latitudinal shifts in the positions of the polar jet stream and the intertropical convergence zone(New Mexico, Asmerom  et al., 2010)and changes to over-ocean humidity, temperature or moisture sources(Idaho, Schlegel et al., 2009).USA reconstructed ∆18O ice age values could also be influenced by changes to groundwater recharge ratio seasonality as land surface conditions changed(Jasechko et al., 2014).Wagner et al. (2010)interpret decreases to southwestern precipitationδ 18 O to reflect cooler and more-humid conditions.Extending this interpretation to negative reconstructed ∆ 18 O ice age across the southwestern USA values supports earlier conclusions that the region was cooler and more humid than today during the last ice age, possibly linked to changes in air mass trajectories and moisture sources (Asmerom et al., 2010; Wagner et al., 2010).Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | subtropical coasts.75 % of reconstructed ∆ 18 O ice age values are negative, with the highest magnitude differences between δ 18 O ice age and δ 18 O late-Holocene observed at high latitudes and far from coasts.This spatial pattern suggests stronger isotopic distillation of advected air masses during the last ice age than under present climate were able to override higher-than-modern glacial seawater δ 18 O values at most locations.Future paleo-precipitation proxy record δ 18 O research can use this new global map of ∆ 18 O ice age records to target and prioritize developing new records in certain regions and to compare ∆ 18 O ice age shifts from different proxy records.In the near term, a global compilation of large lake sediment isotope records that accounts for paleo-evaporative isotope effects could enhance spatial coverage of interglacial-glacial δ 18 O shifts.General circulation models agree on the sign and magnitude of terrestrial precipitation ∆ 18 O ice age values better in the extra-tropics than in the tropics.Differences in simulated precipitation isotope composition changes amongst the models might be linked to different parameterizations of seawater δ 18 O, glacial topography and convective rainfall, however, this hypothesis requires further testing.Future model research should focus on quantifying the relative roles of inter-model spread in the simulated climate vs. the isotopic response to climate change on resulting simulated precipitation δ 18 O.This would provide guidelines to interpret model-data isotopic differences and to identify what aspects of the ice age climate and hydrology models have difficulties in capturing.
4 ConclusionsCompiled groundwater, speleothem, ice core and ground ice records of δ 18 O changes between the last ice age and the late-Holocene range from −7.1 (i.e., δ 18 O ice age < δ 18 O late-Holocene ) to +1.8 ‰ (i.e., δ 18 O ice age > δ 18 O late-Holocene ).Aquifers with positive reconstructed ∆ 18 O ice age values (25 % of records) are most common along the Introduction