Open-meteo with marine forecast. Hoping for plugin

[allancph]

Hi programmers

This wx service is free (no api key) for non-commercial and has very nice nice features
up to 16 days forecast
marine forecast :)
etc
And very nice european coverage :)
https://open-meteo.com/en/docs

Unfortunately I'm not a programmer, but hoping somebody will like the idea.

The owner is interested in the plug feeding some real time wx data the other way
His email here
Patrick Zippenfenig [email protected]

Best regards Allan Lavridsen

[patrick-zippenfenig]

Creator of Open-Meteo here. Let me know if you need any additional information to integrate weather forecasts!

[panaaj]

1. The SignalK data model indexes objects with a unique id. A uuid can be assigned by the plugin for the location so there is no issue if an id is not returned.

2. Speed in m/sec, angles in radians, height/distance in meters, pressure in Pascals, duration in msec.

[patrick-zippenfenig]

1. In this case I would recommend you calculate a UUID inside the plugin based on the requested coordinates
2. The API only returns angles in degrees, pressure in hPa and duration in seconds. Although it is feasible the extend the API with this functionally, I do not see much benefit. Can you convert data inside the plugin? The API returns information about the unit and you can rely on this field to convert all fields with hPa to pressure in Pascal. An extensive list can be found in the FlatBuffers encoding scheme.

[panaaj]

@patrick-zippenfenig Question regarding grid-cells, are cell sizes uniform or do they vary depending on the model applied, geographic location, etc?
I ask because I am looking to retrieve data from a cache when requests are made for a location within the "boundary" of a cell already in the cache.

[patrick-zippenfenig]

Yes, they vary from model to model. Local high resolution models also use different grid projects which makes it very hard to estimate if two locations would use the same grid cell. Additionally, the system may select a different grid-cell, although two coordinates are less than 2 km apart from each other due to elevation correction. It should be safe to assume that two coordinates within 1 km use the same grid-cell, but at this point it is questionable if the cache adds any value.

[panaaj]

Thanks for the details.