Simulation and evaluation of 2-m temperature over Antarctica in polar regional climate model

• Published in: Science China Earth Sciences
• Main Authors: Xin, YuFei, Bian, LinGen, Rinke, Annette, Dethloff, Klaus
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: Springer 2014
• Subjects: Alfred Wegener Institute; Antarc*; Antarctic; Antarctica
• Online Access: https://epic.awi.de/id/eprint/37088/; https://epic.awi.de/id/eprint/37088/1/xin2013.pdf; http://link.springer.com/article/10.1007%2Fs11430-013-4709-z; https://hdl.handle.net/10013/epic.44809; https://hdl.handle.net/10013/epic.44809.d001

The European Centre for Medium-Range Weather Forecasts Reanalysis ERA40, National Centers for Environmental Prediction (NCEP) 20th-century reanalysis, and three station observations along an Antarctic traverse from Zhongshan to Dome-A stations are used to assess 2-m temperature simulation skill of a regional climate model. This model (HIRHAM) is from the Alfred Wegener Institute for Polar and Marine Research in Germany. Results show: (1) The simulated multiyear averaged 2-m temperature field pattern is close to that of ERA40 and NCEP; (2) the cold bias relative to ERA40 over all of Antarctic regions is 1.8°C, and that to NCEP reaches 5.1°C; (3) bias of HIRHAM relative to ERA40 has seasonal variation, with a cold bias mainly in the summer, as much as 3.4°C. There is a small inland warm bias in autumn of 0.3°C. Further analysis reveals that the reason for the cold bias of 2-m temperature is that physical conditions of the near-surface boundary layer simulated by HIRHAM are different from observations: (1) During the summer, observations show that near-surface atmospheric stability conditions have both inversions and non-inversions, which is due to the existence of both positive and negative sensible heat fluxes, but HIRHAM almost always simulates a situation of inversion and negative sensible heat flux; (2) during autumn and winter, observed near-surface stability is almost always that of inversions, consistent with HIRHAM simulations. This partially explains the small bias during autumn and winter.