Physically based evaluation of climate models over the Iberian Peninsula

被引:4
作者
Sanchez de Cos, Carmen [1 ]
Sanchez-Laulhe, Jose M. [1 ]
Jimenez-Alonso, Carlos [1 ]
Sancho-Avila, Juan M. [1 ]
Rodriguez-Camino, Ernesto [2 ]
机构
[1] AEMET, Ctr Meteorol Malaga, Malaga 29010, Spain
[2] AEMET, Serv Cent, Madrid 28040, Spain
关键词
Climate models; Evaluation; Surface processes; PROJECTIONS; VARIABILITY; SIMULATIONS; UNCERTAINTY; EUROPE;
D O I
10.1007/s00382-012-1619-2
中图分类号
P4 [大气科学(气象学)];
学科分类号
0706 ; 070601 ;
摘要
A novel approach is proposed for evaluating regional climate models based on the comparison of empirical relationships among model outcome variables. The approach is actually a quantitative adaptation of the method for evaluating global climate models proposed by Betts (Bull Am Meteorol Soc 85:1673-1688, 2004). Three selected relationships among different magnitudes involved in water and energy land surface budgets are firstly established using daily re-analysis data. The selected relationships are obtained for an area encompassing two river basins in the southern Iberian Peninsula corresponding to 2 months, representative of dry and wet seasons. The same corresponding relations are also computed for each of the thirteen regional simulations of the ENSEMBLES project over the same area. The usage of a metric based on the Hellinger coefficient allows a quantitative estimation of how well models are performing in simulating the relations among surface magnitudes. Finally, a series of six rankings of the thirteen regional climate models participating in the ENSEMBLES project is obtained based on their ability to simulate such surface processes.
引用
收藏
页码:1969 / 1984
页数:16
相关论文
共 44 条
  • [1] [Anonymous], CLIMATE MODELS ASSES
  • [2] [Anonymous], 2007, PARAMETERIZATION SCH
  • [3] [Anonymous], ERA REPORT SERIES
  • [4] [Anonymous], 252 KNMI
  • [5] [Anonymous], 9 REGCLIM
  • [6] [Anonymous], 258 ECMWF
  • [7] [Anonymous], REPORTS METEOROLOGY
  • [8] [Anonymous], COSMO NEWSLETTER
  • [9] [Anonymous], 2006, 0617 DMI
  • [10] [Anonymous], MATH METHODS STAT