四种再分析辐射资料在东南极高原适用性评价

利用东南极Panda-1站2011年2月至2012年1月期间的辐射观测资料,检验了四种再分析资料在该地区的适用性。结果表明:对各辐射分量ERA interim在Panda-1地区的适用性都明显好于其他三种再分析资料,这主要归因于其四维变分(4D-VAR)数据同化系统的应用、新的云预报方程和改进的参数化方案以及同化了更多的卫星资料雷达等非常规探测资料。对于向下短波辐射,NCEP-1与实测值之间偏差最大(18.7 W·m-2),可能原因是模式对大气透明度的高估和对云量的低估。对反射率模拟的偏差直接导致了各模式对净短波辐射模拟偏差。NCEP-1与JCDAS都低估了Panda-1地区的地表反射率,模式中,地表吸收了更多的向下短波辐射,最终导致对净短波辐射模拟偏高。对向下长波辐射,四种再分析资料都存在不同程度的低估,冬季偏差大于夏季,其中NCEP-1与NCEP-2偏差最大(分别为-62.6 W·m-2和-37.3 W·m-2)。四种再分析资料均不能很好地反映Panda-1地区净辐射的年变化情况,一般而言,夏季偏差小,冬季偏差大。虽然再分析资料存在明显的缺陷和不足,在广袤的东南极高原地区,观测站点稀少,实测资料无法满足需要,再分析资料仍不失为研究东南极地区气候的一种有效工具。

EVALUATION OF FOUR REANALYSIS RADIATION DATASETS FROM THE EAST ANTARCTIC PLATEAU

We assessed the correspondence of reanalysis radiation from NCEP-1, NCEP-2, ERA interim, and JCDAS with observational data from the Panda-1 station on the east Antarctic plateau for February 2011 to January 2012. Results indicate that the applicability of ERA interim radiation data at Panda-1 was significantly better than for the other three datasets. This could be attributed mainly to the application of a four-dimensional variational data assimilation system: a new cloud prediction equation, an improved Parameterization scheme, and greater assimilation of satellite and radar data. For downward shortwave radiation, the maximum bias (18.7 W·m^-2) was found between NCEP-1 and the observations. One possible cause could be model overestimation of atmospheric transparency and underestimation of cloud coverage. Net shortwave radiation bias is caused mainly by the bias of albedo. NCEP-1 and JCDAS underestimated the surface albedo in the Panda-1 area, which meant greater ground absorption of downward shortwave radiation and overestimation of net shortwave radiation. All reanalysis datasets overestimated downward longwave radiation to some degree and the bias in winter was larger than in summer; the largest bias was found in NCEP-1 and NCEP-2 (?62.6 W·m^-2 and -37.3 W·m^-2, respectively). None of the reanalysis datasets could reflect the annual variation of net radiation, i.e., smaller bias in summer, larger bias in winter. Even though the reanalysis datasets have obvious shortcomings, the lack of observation stations on the Antarctica plateau means that observational data cannot satisfy research requirements and thus, reanalysis datasets remain effective tools in Antarctic climate research.