Visualisation of wind direction and speed

[jwagemann]

Seems simple but how can we better visualise wind direction and speed?

[dayo-coder]

I am interested

[khufkens]

I find streamlines (e.g. Windy / nullschool) super intuitive, especially when animated. Even static this kind of holds.

Here is a brief tutorial using hurricane Katrina wind data (ERA5 reanalysis) for an R workflow:
https://bluegreenlabs.org/post/map-building-5/

All credit goes to @eliocamp who coded the R ggplot geom - the package actually does a lot more and probably can be used to address other challenges as well.

[eliocamp]

metR has also geom_vector()

And thanks to contributions from twitter, geom_barb() will be coming soon.

[khufkens]

Forgot about that one! I'm actually adjusting this one to format old school wind arrows. Side project to render custom (arbitrary) pilot charts in R using ERA5.

Pilot charts are actually very dense representations of data which might serve as inspiration in this hackaton.

[dayo-coder]

Thanks for sharing

…

______________________________________________ POPOOLA Temidayo IsraelMeteorological Research & Training Institute  WMO Regional Training Centre (WMO-RTC)Nigerian Meteorological AgencyLagos. Nigeria. +234 (0)803-8845128 On Wednesday, June 1, 2022, 09:08:01 PM GMT+1, Koen Hufkens ***@***.***> wrote: I find streamlines (e.g. Windy / nullschool) super intuitive, especially when animated. Even static this kind of holds. Here is a brief tutorial using hurricane Katrina wind data (ERA5 reanalysis) for an R workflow: https://bluegreenlabs.org/post/map-building-5/ All credit goes to @eliocamp who coded the R ggplot geom - the package actually does a lot more and probably can be used to address other challenges as well. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

[milanavuckovic]

Hi,
It is an in person Hackathon and you can register for a few more hours at this link: https://events.ecmwf.int/event/305/registrations/189/

[iroy-tech]

I am interested.

[zakjan]

I'm sorry that I can't attend the hackathon in person. I built a library for animated streamlines deck.gl-particle (demo) that could be used for this task. I'm especially interested if it's possible to use the library as pydeck custom layer, but I'm not familiar with Jupyter. If there are any changes necessary in the bundling process, please raise an issue, I can fix it shortly.

[Candrasa]

Forgot about that one! I'm actually adjusting this one to format old school wind arrows. Side project to render custom (arbitrary) pilot charts in R using ERA5.

Pilot charts are actually very dense representations of data which might serve as inspiration in this hackaton.

I was curious about the windrose plot in this documentation and found another documentation that is quite similar https://windrose.readthedocs.io/en/latest/usage.html#subplots

[nguyenquangchien]

I was curious about the windrose plot in this documentation and found another documentation that is quite similar https://windrose.readthedocs.io/en/latest/usage.html#subplots

Great, you may try tweaking the windrose code, to produce the fishbone windroses and arrange them on a map.

But is the fishbone style more restrictive and cannot present as much information as the colourful style you showed above?