Methodology:

The isotopic response to idealised changes in AIS elevation are simulated using the isotope-enabled coupled ocean--atmosphere--sea-ice General Circulation Model, HadCM3 (Tindall et al., 2009). Two control simulations were used: a pre-industrial (PI) simulation, and a 128 ka simulation centred on the last interglacial (LIG) Antarctic isotope maximum including a modern day AIS configuration (Holloway et al., 2016). Then a suite of eight idealised AIS elevation change simulations were performed using orbital and greenhouse-gas forcing at 128 ka. Each experiment scaled the AIS and relates the change to elevation at the EPICA Dome C (EDC) ice core site following:

beta = ZEDC / (ZEDC + DeltaZ)

where ZEDC is the EDC ice core site elevation in the modern day AIS configuration, DeltaZ is the prescribed elevation change which extends to +/- 1000 m, and beta is the scaling coefficient. Elevations across the Antarctic continent are then increased or decreased proportional to beta;

ZA' = ZA / beta

where ZA is the two-dimensional array of modern AIS elevations and ZA' is a new array of altered AIS elevations. This approach maintains the modern shape of the AIS, thus reducing the influence of changing ice sheet configuration on circulation and climate and isolating the effect of elevation changes alone. We perform experiments with DeltaZ equal to (+/-) 100, 200, 500 and 1000 m. Each of the above elevation change scenarios is integrated for a total of 500-years to ensure that surface and mid-depth climate fields are sufficiently spun-up with the imposed elevation changes. The last 50 years of each simulation are analysed.

LIG Antarctic isotope maximum of between +2-4 per mille above PI in d18O are recorded in East Antarctic ice cores. We evaluate our elevation scenarios against LIG d18O maxima from five published ice core records from East Antarctica (Masson Delmotte et al., 2011): Vostok, Dome Fuji, EPICA Dome C, EPICA Dronning Maud Land and Talos Dome Ice Core. The records are processed following the approach outlined in Holloway et al. (2017): The ice core isotope records are synchronised to the EDC3 age scale (Parrenin et al., 2007) and interpolated onto a common 100 year time grid. Any residual temporal misalignment between the ice cores is minimised by applying a 1500 yr low-pass filter to each record before taking the LIG peak (Sime et al., 2009). Fractional isotopic content is expressed for oxygen-18 as:

d18O = 1000 x (H218O / H216O) / RVSMOW-1

in per mille, where RVSMOW is the ratio of H218O to H216O for Vienna standard mean ocean water.

Data collection:

Model code developed to analyse the outputs can be found on Github: https://github.com/aitens/Antarctic-Ice-Sheet-elevation-impacts-on-water-isotope-records-during-the-Last-Interglacial. To analyse the outputs using the model code, one needs python 2.7 (or 3), with modules likes matplotlib, numpy, iris.

The model was run using 128k orbital parameters for 100 years. Data consist of surface outputs (1 level) with global coverage (longitude from 0 to 360 degrees and latitude from -90 to 90 degrees).

When analysing the outputs, a focus was made on location of ice cores covering the last interglacial:

Vostok (E106.8, S78.5), Dome F (E39.7, S77.3), EDC (E123.4, S75.1), EDML (E0.07, S75.0), Taylor Dome (E158.7, S77.8), TALDICE (E159.2, S72.8), WAIS Divide (E247.9, S79.5), Hercules Dome (E255, S86), Skytrain (E280.3, S78).

Data quality:

No quality control procedures out of the described methodology were processed.