|
In analyzing 20km MAPS GRIB files I have noticed over the past months that sporadically the de-gribbed level 1 height is larger than the level 2 height. Usually this has occurred only once a month but yesterday it occured in 4 GRIB files (out of 30 analyzed) totalling 17 gridpoints. Today there were no such instances. The height difference between z1 and z2 is is not large in these cases, generally less than 1 meter. For me this is not a problem as I have a fix-up in my analysis program to handle it, but I wondered if this is of concern to MAPS modellers (or is this considered a tolerable "noise", which might result from the gribbing process for example).
|