SUJET :

Hi,

To try and get to the bottom of this kindly load a file that gives you the problematic cloud display, take several screen shots at 1 step intervals that include the entire application including the color bar on the right.
Please post the images here and also post the same grib file (you will need to zip it first).

Thanks for your patience,
David

Hello. You can download the files from here.

wetransfer.com/downloads/c0e9529448780e9...0181121164835/4564dd

You don't permit files greater than 2Mb.

Can I ask you something else please. I tried to download icon grib files directly from here
opendata.dwd.de/weather/nwp/icon-eu/grib/
The format should be grib2 (the file extension is .grib2) but I am not able to open it. Neither Xygrib can open it nor wgrib2.
Do you know something about it? Do you download and use these files?

Thanks
Gian

Hi,

Thanks for the files. Please leave the grib file online for a while as I will ask someone else to take a look.
I found the grib file to be perfect and I do not have the same cloud issues on every second time step. This is related to your particular computer for some reason.
I kindly ask you to do one more check. Please create a meteo table on the center of Cyprus as this shows cloudy at 0800 1000 and 1200 UTC but clear at 0900 and at 1100. Does this occur in the meteo table too or are the values cloudy (near 100%) for all hours 0800 to 1200 on Wed the 21st?

As for ICON-EU gribs - they do open in XyGrib for total cloud coverage use 'clct'. They even open while still having the bz2 compression. Not all the parameters are recognized by XyGrib and as there is only a single parameter in each ICON-EU file XyGrib will not open a file that does not have at least one recognized parameter.

The ICON global gribs will not open as they are on a special grid projection and require some processing before being usable in most grib readers.
David

Hi,
In a previous message I said that they are consistent , I meant Weather Data Board, meteotable and graphics, all of them are showing the hole in the data. Today I opened the same data that I sent to you and the hole is in different hours. In the image you can see the meteo table that you asked. If I closed and reopened Xygrib the clouds seem to be ok but If I loaded the grib file from file or download a new one there is a big probability to cause again a hole in the clouds (only). If you think there is some other test or debug I can do for you I am at your disposal .

Now I would like to ask you something about the grib data copyright. I the home page I can read "Weather forecast data sets created or derived by OpenGribs.org are licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License." but this license say also "Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation."
So I would like to ask you if it is permited to use GFS, Icon-EU or Arpege-EU grib data download with XyGrib to produce some weather graphics for a TV weather forecast using another software (that I am working on).

And also (if you are so gently) do you know how to make a script to download grib data directly from Meteofrance?

Thanks
Gian

Hi,
Regarding the cloud issue: I have opened an issue for this in our github area and I suggest waiting to see if the problem occurs again with other users or if we are able to replicate it.

Regarding licenses and terms of use: The data provided is in many ways a derivative work of the original data. This includes changes in projection, normalization of accumulated values, code table changes to make it more compatible with grib readers and then custom packaging according to requests. This give us the right to make some claims on our derivative work.

This said, we are not at all concerned with secondary derivative usage as you suggest. Creating graphics or helping to prepare a forecast and the likes, even if these are commercial activities. In fact the aim of our activity is to provide free an open use of good meteo resource to a broad spectrum of users and uses.

What we are trying to prevent is the commercialization of our compiled data itself in the forms of a subscription or paid for feature in other client software packages.

I thank you for bringing up the issue as it appears that we need to make our terms of usage and the related license a bit more clear.

David