The Transition from the Last Glacial Period in Inland and Near-Coastal Antarctica R
Total Page:16
File Type:pdf, Size:1020Kb
The transition from the Last Glacial Period in inland and near-coastal Antarctica R. Mulvaney, R. Röthlisberger, E. Wolff, S. Sommer, J. Schwander, M. Hutterli, J. Jouzel To cite this version: R. Mulvaney, R. Röthlisberger, E. Wolff, S. Sommer, J. Schwander, et al.. The transition fromthe Last Glacial Period in inland and near-coastal Antarctica. Geophysical Research Letters, American Geophysical Union, 2000, 27 (17), pp.2673-2676. 10.1029/1999GL011254. hal-03110176 HAL Id: hal-03110176 https://hal.archives-ouvertes.fr/hal-03110176 Submitted on 24 Jan 2021 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. GEOPHYSICAL RESEARCH LETTERS, VOL. 27, NO. 17, PAGES 2673-2676, SEPTEMBER 1, 2000 The transition from the last glacial period in inland and near-coastal Antarctica R. Mulvaney,1 R. R6thlisberger,2 E. W. Wolff,1 S. Sommer,2 J. Schwander,2 M. A. Hutterli 2'a and J. Jouzel4 Abstract. Recent studies suggestedthat, during the tran- Based on the timescale adopted, it was suggestedthat, dur- sition out of the last glacial period, one near-coastal site in ing the transition from the last glacial maximum to the Antarctica showeda responsesimilar to that of Greenland, Holocene, TD exhibits climate changes similar to those in and unlike that of central Antarctica. Here, we present a Greenland(Fig. 2) [Steiget at., 1998]. Specifically,TD ap- new high-resolutionrecord of calcium from Dome C, Antarc- peared to show a fast warming rather than a slow ramp in tica. Changesin flux of calcium, an indicator of dust input temperature, and a reversal whose timing appeared to match from other continents, should be synchronousacross the re- the YD rather than the ACR. gion and probably the continent. Using Ca to synchronise records,we find that the main warming at the near-coastal Synchronisation using the Ca record site of Taylor Dome was slower than suggestedpreviously, There is no doubt that climate can vary significantly over and similar to that of central Antarctica. Until there is fur- distances such as those between TD and Byrd or Vostok. ther evidence,it is still a reasonablehypothesis that Antarc- However, some geochemicalproxies should show a similarity tic climate behaved more or less as a single unit during the acrosslarge regions. Here we present a calcium record from transition. a new ice core drilled at Dome C, a central east Antarctic site ( 75ø06'S,123ø24'E, Fig. 1). The corefrom Dome C was Introduction analysed in the field to a depth of 580 m, using continuous flow analytical methods for a range of chemistry, including One of the significantclues for understanding the causes Ca and Na [RSthtisbergeret at., 2000]. The data are available and mechanismsof climate change during the last glacial at high resolution, but in this paper are used generally as I m cycle is the spatial pattern and temporal phasing of such averages. A timescale has been prepared based on a number changes. During the main transition from the last glacial of criteria, mainly comparisonof electricalhorizons [Wotff maximum to the Holocene, central Antarctic ice coresshow et at., 1999]and transitionswith coresfrom nearbyVostok, a different pattern to Greenland ice cores. In Greenland, and comparison of methane profiles with other well-dated the most prominent warming from the last glacial period records. However the arguments here do not rely on any is the rapid jump at about 14.8kyr (calendar) into the absolute dating. Because the Dome C ice considered here Allerod/Bollingwarm period [Johnsenet at., 1992]. This is all in the top 15% of the ice sheet, nearly 3000 m above was followedby a coolingduring the YoungerDryas (YD), bedrock, thinning corrections are small. The non-sea-salt and a rapid final warming at about 11.7kyr into the Pre Ca record from Dome C (calculatedwith referenceto Ns) Boreal. In cores from central Antarctica, most of the tem- is presented in Fig. 2. In the Holocene part of the record, perature increaseoccurs in a slow ramp from about 18.5 kyr, Ca is dominatedby sea-salt(based on calculationfrom Na), punctuated by a small coolingknown as the Antarctic Cold but during the transition and the last glacial period, Ca is Reversal(ACR) (e.g. [Loriuset at., 1979]). Synchronisation much elevated and originates from terrestrial dust. Ca at of coresusing methane records at Summit (Greenland), and TD (also shownas non-sea-saltCa in Fig. 2) also shows Byrd and Vostok (Antarctica) has shownthat the ACR be- elevatedconcentrations in the glacial period [Mayewskiet gan about 1000 yearsbefore the YD [Sowersand Bender., at., 1996; $teig et at., In Press]. 1995; Btunier et at., 1997]. Although modelling studiessuggested that Australia may Recently, new records were published from an ice core be a significantsource of dustto Antarctica[Genthon, 1992], at Taylor Dome (TD, Fig. 1), a near-coastaldome site in it has now been shown from geochemical studies that dust East Antarctica[Steig et at., 1998]. Again, synchronisation in both Dome C and Vostok cores in the last glacial period was carried out using methane and oxygen isotopes in air. originatesmainly from Patagonia[Basile et at., 1997]. We therefore assume a Patagonian source, although our argu- ment would still be valid if Australia were the source. •British Antarctic Survey, Natural Environment Research Council, Cambridge, England, United Kingdom. The TD and Dome C Ca records, in common with all 2Climate and EnvironmentalPhysics, Physics Institute, Uni- other recordsof Ca or dust [Petit et at., 1990]from Antarc- versity of Bern, Bern, Switzerland. tica, show a major decreasebetween the last glacial period SNowat Departmentof Hydrologyand Water Resources,Uni- and the present-day. The magnitude of the decreaseis rather versity of Arizona, Tucson, Arizona. similar (factor of 30 in concentration)at TD and Dome C. 4Laboratoire des Sciences dd Climat et de l'Environnement However, the time period over which the decreaseoccurs is CEA-CNRS, Gif-sur-Yvette, France. quite different: the main decreaselasts around I kyr at TD, but about 3-4kyr at Dome C; again we emphasise that, Copyright2000 by the AmericanGeophysical Union. although the exact date of this decrease at Dome C is un- Papernumber 1999GL011254. certain, the time period covered by it is rather insensitive to 0094-8276/00/1999GL011254505.00 our assumptions. 2673 2674 MULVANEY ET AL.:TRANSITION FROM LAST GLACIAL IN ANTARCTICA concentrations)for the late glacial [$teig et al., In Press], but not sufficient to explain the change in Ca concentration observed. Thus, we find a clear result that the Ca concentration at Dome C reducesslowly over a period of several thousand years, during which TD apparently seesno change. TD con- centration then reducesrapidly to levels comparable to those at Dome C. Such a scenario is not credible, and we suggest that the timescale at TD is in error in the period previously dated as 14.5 to 17.5kyr, and the dating shouldbe extended through this section by at least 2 kyr. As further backing for our belief that Ca should show similar trends across the region, Ca at Dome Fuji on the opposite side of the conti- nent [Watanabeet al., 1999]shows a similardecrease lasting several thousand years at the transition. If the Ca record at TD is stretched to match more closely to Dome C, the effect is also to stretch the isotope (proxy temperature) record, such that the increasefrom the last glacial becomes a slow ramp that then looks like Figure 1. Location of ice core sitesreferred to in the text (B other Antarctic records, and unlike Greenland records. The - Byrd, DC- Dome C, DF - Dome F, TD- Taylor Dome, V- stretching of the TD record that we propose in this part of Vostok) in relation to the Patagonianand Australian deserts,the the core requires that the layer thickness at TD for this sec- possiblelong-range Ca sourceregions (Lambert azimuthal equal- tion is very low. The most likely explanation for this would area projection, scaletrue at centre). be greatly reduced accumulation rates. Although exception- ally low accumulationrates have been proposed[Morse et We now considerthe factors that can result in a change in al., 1998],we note that stretchingthe TD timescalefurther terrestrial dust depositing to the snow in Antarctica. These might be expectedto lead to higherconcentrations of •øBe are: (1) a changein the area, location, or aridity of the than were usedto calculatethe publishedtimescale ([$teig source,presumed to be in Patagonia;(2) a changein the et al.,1998] - •øBeis believed to havemaintained a constant uplift of dust from the source, due to changesin wind speed flux to the Antarctic ice sheet through the glacial stage, so or other conditionsat the source;(3) a changein efficiencyor any change in concentration observed in the ice implies a route of transport between the source and the ice core site; changein accumulationrate). We note that the radar in- (4) a changein the lossof dust during the commonoceanic ternal layers[Morse et al., 1998]show that this time period is thinner at the ice core location than at locations to the transport route; (5) a changein local depositionefficiency north and south. Because methane is rather invariant in in Antarctica. In relation to Patagonia (or indeedAustralia), TD and the period 17-30 kyr, the gasmatching (methane) provides Dome C are very close, and we must assume they have the little age controlin this part of the ice [Brooket al., 1999].