Expedition Navigation Software 
1 - Antipodes
- Expert
Offline - Registered: 9/26/2015
- Posts: 10
Historical Grib data
Can historical grib data be downloaded in EXP and then used to analyse along a route?
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,965
Re: Historical Grib data
You can download reanalysis data from NOAA and ECMWF. This can be used in Exp.
See the links at
- kasszest
- Contributor
Offline
- From: London/Cowes
- Registered: 7/27/2019
- Posts: 7
Re: Historical Grib data
I am in the process of preparing for next year's OSTAR which will be starting 3 weeks earlier than previous editions, and so am running a lot of routings using reanalysis data. Getting 10m wind from ERA5 is no problem.
As the Gulf Stream is such a big factor I've also downloaded historical OSCAR ocean surface currents data, but it is in netcdf format, and doesn't appear to be the right kind of netcdf data for Expedition (or else I am doing something wrong). Does anyone have experience with using historical surface current data for routing? Is there a better source?
Thanks,
Kass
- jweisbaum
- Expert
Offline
- From: San Francisco, CA
- Registered: 12/21/2019
- Posts: 20
Re: Historical Grib data
Hey Nick, I'm also not able to load the NCEP Reanalysis NetCDF format successfully. Is this supported in Expedition? Thanks.
github.com/jweisbaum
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,965
Re: Historical Grib data
Only a few NetCDF formats are supported. They aren't nearly as well defined as Grib.
- jweisbaum
- Expert
Offline
- From: San Francisco, CA
- Registered: 12/21/2019
- Posts: 20
Re: Historical Grib data
Kasszest, I figured out how to do this.
You can use cdo and wgrib2 to convert NetCDF and change the message name to something meaningful to Expedition.
github.com/jweisbaum
- kasszest
- Contributor
Offline
- From: London/Cowes
- Registered: 7/27/2019
- Posts: 7
Re: Historical Grib data
Many thanks for that jweisbaum, I will try that. By 'message name' do you mean the Product Definition Section of the grib files? Any hints on where I can find a list of message names Expedition considers meaningful?
Thanks,
Kass
- jweisbaum
- Expert
Offline
- From: San Francisco, CA
- Registered: 12/21/2019
- Posts: 20
Re: Historical Grib data
The best way I have found is to do a side by side comparison between my troubled grib and a grib that I know works in Expedition (GFS, HRRR, whatever). I use CDO to convert to grib2 and then wgrib2 to edit, but there is a lot of overlap in these tools. I am not sure if it's the product definition section, but basically I compare the differences between the output of 'wgrib2 not_working.grb' and 'wgrib2 working.grb'. I'm still learning the details of the grib2 specs, and it also seems like different organizations have different standards.
There are many gotchas to this technique such as
1) Unit discrepancies
2) Projection discrepancies
3) Geodesic Model discrepancies
4) Wind altitude discrepancies (ie sig.0.995 vs surface vs 10m)
So make sure that you understand how Expedition is interpreting things. Also, if you find this technique helpful, consider sharing your script on GitHub. That way everyone can benefit and we can build a database of scripts for sailors to convert reanalysis into a useful format for Expedition.
github.com/jweisbaum
- jweisbaum
- Expert
Offline
- From: San Francisco, CA
- Registered: 12/21/2019
- Posts: 20
Re: Historical Grib data
Actually Kass,
I think if you look in your install directory, under Program Files(x86)> Expedition > Expedition4D > Config, you will see a bunch of xml files. Open those with an IDE or text editor to view what Expedition expects to be in each type of grib. You could experiment with creating your own as well, but make sure you make a backup just in case.
Jon
github.com/jweisbaum
- Nick
- Moderator
Offline
- Registered: 9/16/2015
- Posts: 2,965
Re: Historical Grib data
You can also ask for Grib options to be added. At present, Exp supports lat-lon (equirectangular & thinned), Lambert, Polar Stereographic, Gaussian and the fields you see in the settings interface.
Acceptable fields and levels are contained in internal look-up tables. Some of these are allocated priorities. For example, Exp can use surface winds as 10m winds for routing as some services mis-code them that way.