Introducing Super GRIB !


Viewfax4This is a Super GRIB hot off the Saildocs press.  And it’s courtesy of  both the Tropical Analysis & Forecast Branch of NOAA, and Jim Corenman of SailMail Association, who forged the process in order for you to get it through your AirMail program.

It’s a new approach to looking at raw meteorological data, in that the data in the Super GRIB is cooked a bit.   It’s still a bit on the rare side, because at this stage it’s quite experimental!

Go ahead – click on the image for its full effect!

Unlike older GRIBS, which were purely driven by a computer model (in this case, the Global Forecast System model –  GFS),  the Super GRIB has a variety of controls so that professional meteorologists can tweak, combine,  and refine the output based on their experience with “local area effects”.

(In the image above,  I have set the options so that all you see is wind strength coded by  color (red is most intense). The shoreline is not visible in this view).  So what  kind of information is embedded in the SuperGRIB?

In the words of our friends at TAFB (part of the National Hurricane Center), “The gridded marine parameters to be included [are] surface wind direction and speed, wind gusts, significant wave heights and marine hazards. These [forecasts] are value added grids with forecaster input based on marine forecast expertise… The forecasters also use “smart tools” to take into account local marine effects and blend numerical model solutions as appropriate. This combination of tools and forecaster expertise allows gridded forecasts based on the best performing model(s), or an ensemble of model runs, in a given forecast scenario.”

Jim Corenman notifies me by email that some of the options are not yet functional.

There are also some interesting controls for the cruiser. Let’s take a look at them one by one.

Once you request the Super GRIB  (and more about how further down) and have opened it up using AirMail’s ViewFax software, click on the F11 key ( or use the View tab on the menu bar and select “GRIB Settings” option.  This will bring up the control panel. (Click on all images to get full size versions).

Super GRIB

Here you see the control panel, which is divided into two panes . On the top,  Surface Wind (SFC), and on the bottom, Waves.

Each panel has 5 layers (called “options’) and a slider which allows you to set the transparency of each “layer”. I don’t think that all 5 layers are operational just yet, but it will give you an idea of what will be possible in the future.

The layer shown to the left was produced by clicking on the check mark on in the upper left of the control,  the option for color gradient selected, and the transparency set to the right of the slider for full opacity.

In this image, the bottom pane has been checked, for wave information, but the transparency has  been set to render it invisible.   If you reverse the transparency, and make the wave option more visible and the wind all but invisible,  you get an entirely different perspective:

Super GRIBThis produces the band of blue signifying a portion of the Sea of Cortez where wave heights on this day were uniform.  (If you wander around the blue area with your mouse you’ll see an accompanying text box showing wave height in feet and wind direction in degrees True).

And if you put the transparency at midway, and select both wind and waves checkmark, you’ll get a superimposed image, which can be a bit tricky to resolve.

Again, click on any of the images to see full size.  Like, right now.

In this graphic to the left, I have set the transparency sliders midway for both wind and waves,  but for the wind I have chosen the traditional wind barbs, which may appeal to the older,  more traditional male.*

You’ll notice that one of the options is for “contours” but these contours are not (at least at the present time)  the sort of lines you see as isobars (gradient lines  of similar air pressure).  So don’t get confused. For some reason this current iteration does not include air pressure.

Still, the imagination goes wild at the thought of all these choices being offered, some by professional forecasters who may be tweaking the output, and the end user who like me* are eager to play around with a new feature.


* Not that there’s anything wrong with that.


How to request the Super GRIB:

If you are familiar with  requesting GRIBs using the GRIB requestrequest window in AirMail, you’ll know how to do the basic part. You’ll be outlining an area in blue on the map; you’ll click on the “request” button, make sure the “wind” option is checked and the “PRMSL” option is unchecked (below in a posting by Jim Corenman, he writes that in its present, experimental phase, data is limited to wind and wave information).

Once you make the selection and press “send”, you’ll find in your  outbox  an email ready to be posted.  Open it up and the code should look something like this:

send GFS:37N,15N,123W,099W|1,1|0,24,48|WIND

I change the GFS  to NDFD  and it should look like this:

send NDFD:37N,15N,123W,099W|1,1|0,24,48|WIND


The full instructions posted by Jim Corenman are below:

Gridded Marine Forecasts

NDFD: NDFD (National Digital Forecast Database) is a US National Weather Service (NWS) project to make forecaster-generated weather forecast data available in gridded format for the NWS areas of responsibility. These data files are prepared along with the forecast charts (radiofax charts) and text bulletins. The advantage of the grib format is a much more efficient file format (and smaller files) than chart images.

NDFD coverage areas in weird colors

NDFD is a developing project, and available data will be expanded. Currently wind and wave data is available for the Pacific from 20S to the equator and west to 120w, equator to 30n and west to 140w, 30n to 49n and west to 133w, and offshore waters around Alaska and Hawaii. The current Atlantic coverage is 03n to 31n from the Gulf eastwards to Africa, and from 31n to 50n from the coast east to 064W.

To request NDFD data use the code “NDFD”, specify the lat-lon area (respecting the available areas, above). The resolution is 0.12 deg (default is 1-deg if omitted), forecasts are available out to 7 days at 3 to 6-hour increments. Files are currently updated by 01:00z (for the 00z forecast), then 07:00, 13:00, 19:00 and 23:00z.


NDFD is still in “experimental” status which means that the data may not be available or correct (which is always true for computer data).

For more info about NDFD see:


Computer Forecast Models:

The primary source for Saildocs computer-model data is NOAA’s “GFS” global model, requested by using the code “GFS:” (or “Grib:”) in a grib-request. Data from other models is also available. Here are the details for the available models:

GFS: The most-often used model is NOAA/NCEP’s global “GFS” (Global FOrecast System) model. Data is generally available on a 0.5 x 0.5-deg grid every 3 hours out to 192 hours, and then on a 2.5-deg grid every 12 hours out to 384 hours (16 days). Note, however, that forecast accuracy becomes increasingly speculative after 4-5 days.

The request-code is “GFS:” as used in the examples in the “gribinfo” document. The code “grib:” can also be used, the difference is that for “grib:” the forecast-times are always adjusted to provide forecast-times relative to 00z. Available parameters are PRMSL,WIND,HGT500,TEMP. Data is almost always available by HH+05:00 hours (“HH” being the synoptic-time), so by 05:00z for the 00z forecast cycle.

NAVGEM: This is a US Navy model (formerly known as NOGAPS), with data is available on a 1-deg grid with forecasts every every 3 hours to 24, then 6 hours to 96, then 12 hours to 144, and updated every 6 hours by HH+06:00. Available parameters are PRMSL,WIND,HGT (the default is PRMSL,WIND). The general view is that GFS has better accuracy on average, but Navgem is an independent model and is valuable as a “second opinion”.

COAMPS: This is a meso-scale (i.e. more detailed) regional model provided by the US Navy. Data is available for three regions: eastern Pacific, west Atlantic, and Carib/Central America. The syntax is the same as for a basic grib, except the request-code “coamps:” is used in place of “gfs:” or “grib:”. Forecast data is available on a 0.2 x 0.2-deg grid at 6-hour intervals out to 48 hours (72 for west-Atlantic). Available (and default) parameters are PRMSL,WIND.

The areas for the currently available coamps areas are:

NE Pac:  34n,64n,172w,118w
So Cal:  22n,43n,128w,109w  (wind only)
Eq. Am:  12s,32n,122w,058w  (wind only)
W Atl:   20n,55n,093w,055w
Euro:    29n,66n,015w,040e

The same “coamps:” code is used for all regions, and Saildocs chooses the appropriate region based on the requested lat-lon area. Requests which fall partially outside the selected region will be trimmed at the edge of the region, and a zero-length file will result if the request is completely outside any of the regions. It is not possible to span multiple regions because data from different files does not match at the region edges, apparently a boundary-condition issue with the model.

WW3: This is NOAA’s WW3 wave model, data is available on a 1 x 1.25 grid (lat x lon), forecasts every 3 hours out to 180, updated every 6 hours usually by HH+05:00 (5 hours after the synoptic time, or 05:00 for the 00z forecast). All of the WW3 data is available, the default parameter is “HTSGW” (height of sig waves) or “WAVES”, other WW3 parameters are defined here:

Also, the “WAVES” parameter can be added to a GFS request and Saildocs will combine the files, adding the WW3 sig-wave data layer to the same grib file (on the 1×1.25 grid if coarser than the user-specified grid).

Regional WW3 models are also available, updated every 6 hours and available around HH+05:00, as follows:

ww3akw: 45n,75n,160e,123w  0.3 x 0.5 deg grid, to 180 hrs @ 3 hr increment
ww3enp: 05n,60n,170w,070w, 0.3 x 0.25 grid, to 180 hrs @ 3 hr inc
ww3wna: 00n,50n,098w,030w, 0.3 x 0.25 grid, to 180 hrs @ 3 hr inc
ww3med: 29n,66n,015w,045E, 0.2 x 0.2 grid,  to 96 hrs @ 6 hr inc

The first three are also available as “ww3merged” to allow spanning Central America.

If the generic “WW3” code is used (or “Waves” added to a GFS request) then then Saildocs will choose the best data file for each request, i.e. whichever returns the most data points.

send ndfd:22N,17N,148W,114W|0.5,0.5|0,24,48|PRMSL,WIND

This is an ocean current forecast model from NOAA’s RTOFS system, based on the HYCOM model (see for more details). It is actually two different models and three overlapping areas:
The smallest-area gulf-stream dataset (code RTOFS-GS) covers 25N-48N, 083W-052W. That file has higher resolution with a data grid of 0.05 x 0.06 deg (lat x lon) and forecast times of 0,24,48,72,96,144 hours updated daily at around 15:00z.
The larger-area Atlantic file (RTOFS-ATL) goes out to 144 hours at a 6-hour increment, 0.26-deg resolution, and covers 26S to 75N, North America to Africa/Europe.  The global RTOFS model (RTOFS-GLOBAL) has a 0.08 deg resolution (data grid) and goes out to 192 hours at 24-hour increments.

The RTOFS code can be used for all regions, and Saildocs selects whichever file best covers your request (i.e. returns the most data points). The specific models can also be requested with the noted code. Available parameters are CUR (or CURRENT or UOGRD,VOGRD), WTMP (or WATER_TEMP), DSL_M (sea-level deviation), and SALTY (salinity). The default (if parameters are omitted) is CURRENT.

OSCAR: OSCAR (Ocean Surface Current Analysis- Realtime) is an ocean current analysis based on satellite observations (scatterometer and altitude) over a 5-day period, and updated every 5 days. Resolution is 0.33 deg, global data is provided but the focus is on the tropical Pacific. The code is “OSCAR”, the forecast-time field is not used, and the default (and only) parameter is surface current (CURRENT or UOGRD,VOGRD).
For more info see when internet is available.


Thanks for using Saildocs and Good Sailing!

(updated 2014-05-13)


Please let me know how it works (or doesn’t work) for you by filling in the comment form below.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s