Problems with 20090317_gdas0p5-20090329_gdas0p5

[croc]

Dear hysplit user forum,
I am having problems running hysplit using meteo files from
20090317_gdas0p5 to 20090329_gdas0p5, obtaining NaN values in
tdump files.
My guess is that the problem is related to the missing/filled in fields in
meteo files as stated in readme_gdas0p5_missings.txt:

2009 March 17 - March 29: original surface/flux fields failed; filled-in
with gdas1 data interpolated to half-degree; SH2M missing

Is there any way we can use those gdas0p5 files or we just should use
the gdas1 ones?
Regards,