|[Science] [BAS home] [Met home]||Antarctic Meteorology|
This is not really very general-public-y. If you don't recognise the terminology, you probably don't want the dataset...
Datasets 2, 3 and 4 can be differenced: here they are:
|4 - 3 (DGV - WMC)||2 - 3 (CG - WMC)||2 - 4 (CG - DGV)|
One can also compare the datasets along a traverse line, in this case from Casey to Vostok to Amundsen-Scott. The crosses are data points from DGV (dataset 5 in the list above) selected to be on the traverse line. The lines are datasets 2, 3 and 4 interpolated to the dataset 5 locations.
Looking at this, what do we see? DGV captures the high accumulation just near the coast best - although this is a difficult area, because there are low-accumulation places at the same latitudes. In general, the gridded datasets cannot be expected to resolve the coastal areas well, and don't. I had expected WMC and CG to look more similar than they do in the interior - since they are digitised from the same map! Polewards of 80 S WMC and CG agree, and DGV is somewhat higher: in terms of matching the observations it is hard to choose between them. Between 80 S and 75 S WMC, CG and DGV disagree; DGV has the edge nearer 75 S and CG nearer 80 S. Between 75 S and say 69 S all three agree well enough with the observations. North of 69 S, CG and WMC miss the accumulation peak the DGV captures.
These are from analyses (and below, regridded from "observations"). They are:
NMC (NCEP/NCAR) ERA ECMWF operational NMC-ERA
And here is ERA, annual mean:
The data used for this is available here.
Here is the temperature, regridded from Giovinetto and Bentley (exact ref...?), the difference from ERA, and the difference with expanded contour range.
The data used (in pp-format) is g_and_b.t.pp.
|Past last modified: 12/1/2004||
© Copyright Natural Environment Research Council - British Antarctic Survey 2002