cdo&geocat垂直插值气压层比较

一大碗年糕

之前一直用的geocat的interp_hybrid_to_pressure来插值模式输出数据，晚上试了下cdo的插值。两者总体还是一致的geocat：interp_hybrid_to_pressure

1. ds = xr.open_dataset("var_tmp.nc")
   
2. ds_interp = gc.interp_hybrid_to_pressure(ds['Z3'], ds['PS'],ds['hyam'],ds['hybm'],ds['P0'])

复制代码

cdo：ml2pl

1. # Extract variable
   
2. ncks -O -v ${VAR},ilev $filename var_tmp.nc
   
3. # Add layer interface 'ilev' as bounds of vertical coordinate 'lev’
   
4. ncatted -a bounds,lev,c,c,"ilev" var_tmp.nc
   
5. # Interpolate from hybrid sigma-pressure to pressure levels
   
6. cdo ml2pl,20000.,25000.,30000.,50000.,70000.,85000.,92500.,100000. var_tmp.nc var_ml2pl.nc

复制代码

结果：向上其他气压层缺测区域基本一致，但是在1000hPa上cdo的缺测面积较少。大小差异几乎可以忽略不计