This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: "Frank Colby" <address@hidden> >Organization: UMass Lowell >Keywords: 200008241436.e7OEaMN13073 >Dear Support, > >I'm trying to set up our ldm box to receive grib files for certain >models, and decode them into gempak files. I have used the pqact >configuration stuff that is on the unidata web site, but it's clear I'm >not getting the data for these actions, since I'm not creating the files >for some of the models. I am getting the eta 211 grid, usually out to >60 hours, and the gempak file seems complete. I can get some mrf data, >but I only get part of the US. I modified the pqact actions and now I >have the whole NH, but only from 84 hours to 240 hours. I'm happy to >puzzle out the actions, but I can't seem to find a list of what grib >files I should be looking for, so I don't know how to troubleshoot my >patterns. Here's what I'm requesting: > >REQUEST UNIDATA "(^[A-OQ-X])|(^[YZ].[^AHIJRU])" > >If I understand this, I should be getting anything that doesn't start >with P, Y or Z followed by anything, or anything that starts with Y or Z >and is followed by 1 character, followed by A, H, I, J,R or U. Or does >the last group in [] apply to the one character represented by the .? Frank, Actually, the set ^[YZ].[^AHIJRU] means, anything starting with Y or Z except products with any of [AHIJRU] as the third letter. With the ^ inside the square products, it means "not". > >Then the actions in pqact.conf are as follows: > >The 211 grid: > >HRS ^[YZ].[Q].*/mETA > >The 212 40 km grid: > >HRS ^[YZ].[R]... KWBE ([0-3][0-9])([0-2][0-9]).*/mETA > >which seems to not get anything to work on To get these grids for 212 ETA, you need to remove the R from the excluded set above. > >The E grid for the MRF > >HRS ^H.E[A-Z][0-9][0-9] KWB. (([0-3][0-9])([0-2][0-9]).. > >which is where I get the 84-240 hour forecasts only ( I thought the F >grid was for the SH -- is that right?) These MRF grids are 3.5 to 10 days. You are correct that E & F together give you northern and southern hemisphere. Nominally, if you want 0-72 hours on the global scale, then this period is covered by the AVN thinned grids ^H.[I-P]. The MRF is "extended" at a lower resolution from the AVN once per day. > >The 381km 201 grid is using > >HRS ^Y.A... KWBH > >which seems to get nothing, > >and the 190km grid 202 is > >HRS ^Y.I... KWBH > > Again, you are excluding the A & I grids in the request pattern above. >which also gets nothing. > >Can you point me in the right direction? > >Frank > > > One point to consider with eth ETA grids is to use the -m option with dcgrib to specify the maximum number of grids in the created file. The default is 1500 grids, but the ETA now exceeds that slightly in the 211 grids extending to 60 hours from 48. I suggest -m 2000 for 211 and -m 7500 for 212. Steve Chiswell