Tutorial workflow: messed up wind plot

[MetRonnie]

~/cylc-run/cylc-forecasting-workflow/run1/work/20210723T0300Z/consolidate_observations/wind.png

[MetRonnie]

Closing as this is only on cylc/cylc-doc#265 (wind data is broken anyway on master)

[MetRonnie]

Now that cylc/cylc-doc#265 merged, this is valid

[wxtim]

Do you still get this error? I don't.

[wxtim]

Do you still get this error? I don't.

And now.. I do

I'm pretty sure that one of the plots has got it's lat-long co-ords flipped, so that one of the plots is from
Lat -12 to +1 and Longitude 48 - 61 (A chunk of the Indian Ocean vaguely near the Seychelles), and the other from Lat 48-61 and Long -12 to +1 (A chunk of the North Atlantic with a few insignificant islands)...

As a result the overall plot is plotting a large chunk of Europe and the Western Asia.

I've managed to exclude the possibility that the problem is me messing up GCSE trig. Phew

When I changed the method of getting wind data I replaced a source which supplied (long, lat), with one which supplied the more conventional (lat, long).

If we didn't already have enough trouble with external dependencies I'd say lets use cartopy to add a map layer.
Failing that a simple dataset could add a nice UK outline.