Clim. Past, 10, 1253–1267, 2014 www.clim-past.net/10/1253/2014/ doi:10.5194/cp-10-1253-2014 © Author(s) 2014. CC Attribution 3.0 License. Siple Dome shallow ice cores: a study in coastal dome microclimatology T. R. Jones1, J. W. C. White1, and T. Popp2 1Institute of Arctic and Alpine Research, University of Colorado, Boulder, Colorado, USA 2Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark Correspondence to: T. R. Jones ([email protected]) Received: 19 April 2013 – Published in Clim. Past Discuss.: 24 May 2013 Revised: 10 May 2014 – Accepted: 12 May 2014 – Published: 26 June 2014 Abstract. Ice cores at Siple Dome, West Antarctica, re- determine the feasibility of long-term climate reconstruction ceive the majority of their precipitation from Pacific Ocean using coastal ice core sites in West Antarctica (WA). moisture sources. Pacific climate patterns, particularly the El Niño–Southern Oscillation (ENSO) and the Southern Annu- 1.1 Site description lar Mode (SAM), affect local temperature, atmospheric cir- culation, snow accumulation, and water isotope signals at Siple Dome is located in WA on the Siple Coast at approx- ◦ 0 ◦ 0 Siple Dome. We examine borehole temperatures, accumula- imately 81 40 S, 148 49 W (Fig. 1). The Siple Coast con- tion, and water isotopes from a number of shallow ice cores tains five major ice streams (A–E), which drain part of the recovered from a 60 km north–south transect of the dome. WAIS, and is immediately adjacent to the Ross Ice Shelf. The data reveal spatial gradients partly explained by oro- Siple Dome is located between ice streams C and D. Drill graphic uplift, as well as microclimate effects that are ex- sites include two cores on the north flank of the dome fac- pressed differently on the Pacific and inland flanks. Our anal- ing the Pacific Ocean (E and G), three cores from the summit yses suggest that while an ENSO and SAM signal are evi- of the dome (B, C and D), and two cores on the south flank dent at Siple Dome, differences in microclimate and possi- of the dome facing inland (H and F) (Fig. 1). The maximum − ble postdepositional movement of snow makes climate re- slope across the study area is relatively flat at 7.2 m km 1. construction problematic, a conclusion which should be con- Table 1 summarizes the details for each ice core location. sidered at other West Antarctic coastal dome locations. 1.2 Antarctic teleconnection The Siple Dome region is centrally located to study all of the effects connected to the El Niño–Southern Oscilla- 1 Introduction tion (ENSO), the Southern Annular Mode (SAM), and the Amundsen Sea low-pressure area (ASL). Previous studies The Siple Dome ice core was drilled as part of the West have shown that teleconnected ENSO climate signals are per- Antarctic Ice Sheet (WAIS) initiative managed by the Na- vasive across the Antarctic continent (Turner, 2004). In WA, tional Science Foundation (NSF) Office of Polar Programs ENSO has been shown to affect atmospheric circulation, lo- (OPP). Originally one of two deep WAIS initiative ice cores cal temperature, and snow accumulation (Delaygue et al., (the other being the inland and more elevated WAIS Divide), 2000; Bromwich et al., 2004; Guo et al., 2004; Schneider Siple Dome’s coastal location in the Pacific sector of Antarc- et al., 2004). A positive correlation exists between ENSO and tica was selected in part to study regional climate signals. In the SAM during times of strong teleconnection (Fogt and this study, we evaluate a series of shallow ice cores drilled at Bromwich, 2006), such that the SAM index expresses both sites across the dome to better characterize small-scale pat- high-latitude and tropical climate variability (Ding et al., terns in relationship with local climate variability as well as 2012). In turn, the SAM influences the depth of the ASL, and Published by Copernicus Publications on behalf of the European Geosciences Union. T. R. Jones et al.: Siple Dome Microclimatology 21 1254 T. R. Jones et al.: Siple Dome microclimatology o Table 1. The Siple Dome shallow ice core sites (B–H) and borehole 0 60 o temperature measurement sites (G, B, H, J) used in this study. E W −180 o Site Data type Latitude Longitude Length of Bottom (degrees) (degrees)14 ice record depth 60 o −190 (years) (m) 88 S Net Accumulation (cm yr WD E Ice 81◦18.140 S 148◦18.140 W 1908–1995 20.0 E 12 o ◦ 0 ◦ 0 SD o −200 0 S G Ice/temp 81 34.25 S 148 35.85 W 1917–1995 20.0 RI 8 D Ice 81◦38.730 S 148◦47.160 W 1901–1995 20.0 120 ◦ 0 ◦ 0 Amundsen −210 C Ice 81 39.30 S 148 47.66 W10 1836–1995 30.0 o D (per mil) ◦ 0 ◦ 0 Ross δ Sea 72 S B Ice/temp 81 39.53 S 148 48.72 W 1657–1995 54.0 Sea ◦ 0 ◦ 0 −1 120 H Ice/temp 81 44.37 S 148 58.61 W ice eq) 1898–1995 20.0 −220 ◦ 0 ◦ 0 o o F Ice 81 54.51 S 149 20.22 W8 1872–1995 20.0 W South Pacific 64 S ◦ 0 ◦ 0 JWater Temp Isotopes 81 55.50 S 149 22.58 W Sector −230 Water Isotopes 1 Sigma o Net Accumulation 56 S o W E G D C B H F 180 Shallow Ice Core 600 tive SAM index years are more likely during strong La Niña Inland 500 years, while Bromwich et al. (1993) found a connection be- 90 30km Flank 120 60 Elevation (m asl) 400 tween katabatic surges over inland WA and across the Ross IceE Shelf duringG ElD C NiñoB H events. F 20 F Shallow Ice Core 150 30 −22 1.3 E est. Accumulation at Siple Dome 10 −23 H −24 SatelliteJ microwave F est. observations G at Siple Dome first sug- −25 B 180 Summit D C B 0 gested a north to south decreaseH in accumulation rates 20 m Borehole Temp (C) −26 (Zwally400 450 and Gloersen500 550, 1977600 ). Subsequent650 studies by Bromwich (1988Elevation) concluded(m a.s.l.) that orographic processes con- G 210 330 trolled the distribution of snow across Siple Dome from Figure 3.stormsTop) Shallow originating ice core mean in theδD (solidRoss black and line, Amundsen un- seas. Radar E certainty barsstratigraphy attached), δD was standard later deviation used (black to confirm dashed line), a northward migra- 240 300 and mean net accumulation (solid green line, standard error bars at- Pacific 270 tached) ontion Siple ofDome the from flow 1920-1995. divide Middle) (Nereson The elevation et al., pro-1998), and modeling Flank file of shallowof the ice cores radar at layerSiple Dome. spacing Bottom) at millennialShallow 20 meter resolution showed a borehole temperatures70 % decrease recorded in at accumulation Siple Dome. Dark from blue north squares to south across the Figure 1. Top panel: map of Antarctica centered on the Ross Seadenote temperature measurements. Blue lines designate tempera- ture trends.dome Dashed (Nereson black lines designateet al., 2000 adiabats). A(0.82 recentoC/100m). accumulation study region. The red dot is Siple Dome (SD), the blue dot is the WAISNo boreholederived temperature from was the taken analysis in the vicinity of beta of Core radioactivity E (30km in shallow firn DivideFigure (WD), 1. Top) and Map the of green Antarctica dot centered is Roosevelt on the Ross Island Sea (RI). region. WD and The red dot is Siple Dome (SD), the blue dot is WAIS Divide (WD), north) or Corecores F (30km showed south). that These accumulation temperature values is (lightgreatest blue 30 km N of Siple RI areandrecently the green dotdrilled is Roosevelt ice cores Island in (RI). close WD proximity and RI are to recently Siple Dome.squares) areDome estimated and using decreases adiabats and by temperature half at 30 trends. km S, based on net accu- Bottomdrilled panel: ice cores local in closemap proximity of Siple Dometo Siple including Dome. Bottom) grid Local locations of mulation averages from 1955 to 1997 (Hamilton, 2002). shallowmap of ice Siple cores Dome B–H. including The Pacific grid locations flank ofcores shallow are ice to thecores north of B-H. The Pacific Flank cores are to the North of the Dome Summit, the dome summit, and the inland flank cores are to the south of the 1.4 Water isotopes domeand summit. the Inland Flank cores are to the South of the Dome Summit. To study the effects of ENSO, SAM, and the ASL at Siple Dome, we utilize water isotopes as a climate tracer. The iso- both the location and depth of the ASL affects the climatol- topic composition of water molecules is expressed in delta ogy of the Ross Sea region and the interior of West Antarctica notation (δ) relative to VSMOW (Vienna Standard Mean by influencingwww.jn.net storm tracks (Turner et al., 2012). Ocean Water) using the following equation:J. Name Research by Hosking et al. (2013) shows that ∼ 25 % of variance expressed in the SAM index is related to ENSO δsample = Rsample/RVSMOW − 1 × 1000, variations. However, the exact location, phasing, and climate forcing of the ASL in relation to ENSO has been the topic where R is the isotopic ratio 18O/16O or D/H in the sample of numerous studies. Turner et al. (2012) showed that the or VSMOW. The δD and δ18O isotopic composition of pre- ASL is significantly deeper during La Niña as compared cipitation is related to the temperature at which condensation to El Niño, but the zonal location of the ASL is not sta- occurs in a cloud (Dansgaard, 1964).