search

inbo / crow

Online tool to visualize birds detected by weather radars
https://inbo.github.io/crow/
MIT License
15 stars 7 forks source link

FAQ / about? #119

Open peterdesmet opened 3 years ago

peterdesmet commented 3 years ago

A colleague looked at the application for the first time and found it difficult to interpret the results. E.g. for Essen he wondered if there are lots of birds flying at 4km:

Screenshot 2021-02-11 at 20 38 35

I think this is an artefact in the data, as the MTR is 0. I guess we won't be able to explain all these things in the short text above the chart. Would it be worth to explain some of these things in a FAQ, about or "how to interpret these charts", with e.g. annotated screenshots? cf. https://birdcast.info/news/a-primer-for-using-weather-surveillance-radar-to-study-bird-migration/

mrtnrey commented 3 years ago

The German radars are indeed quite noisy. We could consider to limit the density plot to the bins that have a valid sd_vvpin the vpts-data. The sd_vvp is the final check in the vol2bird algorithm whether the bin is bird-reflectivity or something else. This should be quite easily implemented&tested. We have then also a better consistency with the MTR-charts. The drawback is that most plots will be quite empty...

The explanation at the birdcast website is really great. Some basic explanation in CROW should be integrated indeed. Not sure where & how (I like the fact that CROW is a onepager for the moment).

niconoe commented 3 years ago

I also like the explanations at birdcast, really understandable, even for me :)

Technically speaking, we could indeed add a second page (that'll stay a SPA, so it would stay a single HTML page, but the content would be replaced by JS code to give a "multipage" look and feel).

Another option would be some sort of popup/modal/overlay, but I'm not sure how nice it would be with long content. We're using Bootstrap and BootstrapVue for the UI, so any bootstrap-ready solution (Bootstrap modals, for example) should be pretty easy to try.

In terms of content, I think it would be also good to explain the strengths/weaknesses/use cases of the different colour scales.

mrtnrey commented 3 years ago

@hvangasteren how is this managed in FlySafe, these noisy Germans? (see above)

hvangasteren commented 3 years ago

Which noisy data? J

peterdesmet commented 3 years ago

@hvangasteren see the yellow band at 4000m in the image at https://github.com/inbo/crow/issues/119#issue-806688998

peterdesmet commented 3 years ago

Or, as BirdTAM:

Screenshot 2021-02-12 at 16 36 28
hvangasteren commented 3 years ago

You're using the data from baltrad, which is filtered. We're using the raw data downloaded from the German services. If you want to know more about this, please forward to Hidde. He knows the differences exactly.Best HansVerzonden vanaf mijn Galaxy -------- Oorspronkelijk bericht --------Van: Peter Desmet notifications@github.com Datum: 12-02-2021 16:37 (GMT+01:00) Aan: inbo/crow crow@noreply.github.com Cc: Hans van Gasteren hvangasteren@xs4all.nl, Mention mention@noreply.github.com Onderwerp: Re: [inbo/crow] FAQ / about? (#119) Or, as BirdTAM:

—You are receiving this because you were mentioned.Reply to this email directly, view it on GitHub, or unsubscribe.

mrtnrey commented 3 years ago

No Hans: we're receiving the raw data from DWD too at KMI ("TH" data). We don't use baltrad.