
Atmospheric contributions to the surface mass balance of Greenland in the HadAM3 atmospheric model
Author(s) -
Murphy Bradley F.,
Marsiat Isabelle,
Valdes Paul
Publication year - 2002
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2001jd000389
Subject(s) - greenland ice sheet , climatology , orography , environmental science , ice sheet , snow , atmospheric sciences , glacier mass balance , climate model , cryosphere , geology , precipitation , sea ice , climate change , glacier , meteorology , geography , oceanography , paleontology , geomorphology
Atmospheric climate models now present powerful tools for studying atmospheric parameters relevant to the mass balance of the Greenland ice sheet, a significant uncertainty in global sea level. This study evaluates the simulations of the UK Hadley Centre atmospheric model (HadAM3) against observational data and reanalyses. HadAM3 reproduces the main precipitation patterns, but a positive bias exists in autumn over southern Greenland. The modeled North Atlantic storm track is well placed, but it and the polar vortex and hemispheric‐scale wave activity are weaker than observed, which may impact on moisture flux over Greenland. Surface temperatures are 2°–3°C too warm in summer and up to 5°C too cold in winter, probably owing to the snow albedo varying too little throughout the year, and excess summer downward short‐wave radiation flux and winter long‐wave loss, related to insufficient cloud cover. The summer warm bias and the too coarsely represented ice sheet area lead to Greenland‐mean melting being overestimated, and a negative surface mass balance is simulated. A higher‐resolution ice sheet mask improves these quantities significantly. Increasing resolution is expected to remove most errors when orography and the ice sheet, particularly around the coastal margins where gradients are large, are better defined, as orographic dynamical deficiencies are minimized. A high‐resolution version of the model is seen to improve the simulation of precipitation. Minor ice and snow physics improvements are needed, but the large‐scale simulation produces good‐quality inputs with no major bias for higher resolution limited area or downscaling models.