Integration of Passive Tracers in a Three-Dimensional Ice Sheet Model - Towards an Interdisciplinary Approach in Earth System Science - page 156

3 Discussion

The aim of this study was to asses the capability of the 3D ISM RIMBAY to
reproduce ice sheet chronologies. To this end model validations against an analytical
age-depth scale where carried out to test the accuracy of the implemented advective
transport schemes. Both schemes exhibit excellent accuracy down to 85 % depth of

the ice sheet. Below this depth high vertical resolutions are required to maintain a
chronology within 5 % of the analytical solution. Computational cost is the main
concern limiting the available number of vertical layers (n .

≈

100). Taking into
account the various unknowns in climate forcings when it comes to transient sim-
ulations of the Antarctic or Greenland ice sheet, a tracer age accuracy of 95 %
relative to the analytical solution throughout the ice sheet, in an idealized setup,
should be more than suf

cient. To compare simulated chronologies to ice core data,
transient simulations of ice sheet dynamics are necessary to reproduce realistic ice
fl flow patterns. Sensitivity studies of the Antarctic ice sheet are carried out in which

crucial flow parameters and boundary conditions were tuned to yield an equilibrium

ice sheet comparable to the present day con

fl

guration available from observations.
First results (not shown in this work) hint at a strong dependance of ice ages in the
interior on variable boundary conditions such as basal shelf melting, sea level and
accumulation effects. Following the validation of the tracer routine RIMTRACE, the
next steps include the transient simulation of ice core chronologies in glacial-
interglacial model runs. Comparing the results acquired in this fashion to ice core
data, will provide an important constraint to the ISM model results. Fitting the ISM
output to match the observed ice core chronologies will hopefully help in assessing
past ice sheet geometries. The results from test runs of the Eulerian and Lagrangian
advection methods, show that the Lagrangian scheme yields more accurate results
and includes the advantage of backtracking tracers to their point of origin. However
both schemes will be further improved. An additional implementation and com-
parison to the Semi-Lagrangian tracer scheme, as discussed in (Goelles et al. 2014 )
would complement the efforts to shed light on the polar climate archives. The last
step in the puzzle will be the implementation of the RIMBAY-TRACE ISM into the
coupled climate model COSMOS (Stepanek and Lohmann 2012 ) including a
complete isotope cycle which would provide the

first fully coupled model assess-
ment of the isotope and age distribution of the polar ice sheets.

References

Arthern RJ, Winebrenner DP, Vaughan DG (2006) Antarctic snow accumulation mapped using
polarization of 4.3-cm wavelength microwave emission. J Geophys Res Atmos 111:D06107
Beckmann A, Goosse H (2003) A parameterization of ice shelf-ocean interaction for climate
models. Ocean Model 5:157 - 170
Clarke GKC, Marshall SJ (2002) Isotopic balance of the Greenland ice sheet: modelled
concentrations of water isotopes from 30,000 BP to present. Quatern Sci Rev 21:419 - 430