Spatial gradients of temperature, accumulation and δ18O-ice in Greenland over a series of Dansgaard–Oeschger events
- 1Laboratoire des Sciences du Climat et de l'Environnement, UMR8212, CEA/CNRS/UVSQ, Gif sur Yvette, France
- 2Centre for Ice and Climate, Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark
- 3Climate and Environmental Physics, Physics Institute and Oeschger Centre for Climate Change Research, University of Bern, Bern, Switzerland
- 4Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA, USA
- 5British Antarctic Survey, Cambridge, UK
- 6UJF Grenoble 1/CNRS, Laboratoire de Glaciologie et Géophysique de l'Environnement (LGGE), UMR5183, Grenoble, 38041, France
Abstract. Air and water stable isotope measurements from four Greenland deep ice cores (GRIP, GISP2, NGRIP and NEEM) are investigated over a series of Dansgaard–Oeschger events (DO 8, 9 and 10), which are representative of glacial millennial scale variability. Combined with firn modeling, air isotope data allow us to quantify abrupt temperature increases for each drill site (1σ = 0.6 °C for NEEM, GRIP and GISP2, 1.5 °C for NGRIP). Our data show that the magnitude of stadial–interstadial temperature increase is up to 2 °C larger in central and North Greenland than in northwest Greenland: i.e., for DO 8, a magnitude of +8.8 °C is inferred, which is significantly smaller than the +11.1 °C inferred at GISP2. The same spatial pattern is seen for accumulation increases. This pattern is coherent with climate simulations in response to reduced sea-ice extent in the Nordic seas. The temporal water isotope (δ18O)–temperature relationship varies between 0.3 and 0.6 (±0.08) ‰ °C−1 and is systematically larger at NEEM, possibly due to limited changes in precipitation seasonality compared to GISP2, GRIP or NGRIP. The gas age−ice age difference of warming events represented in water and air isotopes can only be modeled when assuming a 26% (NGRIP) to 40% (GRIP) lower accumulation than that derived from a Dansgaard–Johnsen ice flow model.