Repository navigation
Wildfire - #1036
Wildfire#1036
Conversation
Adds examples/discovery/wildfire_smoke.py, a sphinx-gallery example that tracks Canadian wildfire smoke over the US Midwest and Great Lakes using act.discovery.get_airnow_bounded_obs as the sole data source. AirNow also re-serves Canadian provincial monitors, so one API path covers both countries. The example plots entirely with ACT displays: GeographicPlotDisplay for a regional map of the latest hourly surface PM2.5 at every station, and TimeSeriesDisplay for 24-hour PM2.5 series at one site each in Duluth, Toronto, and Chicago on a shared y-axis. The token is read from the AIRNOW_API environment variable and the date window is pinned for reproducibility. Co-authored-by: Claude <claude@anthropic.com>
Add wildfire smoke AirNow discovery example
Place GeographicPlotDisplay (map) and TimeSeriesDisplay (three city series) on a single matplotlib figure instead of two separate figures, and shade the EPA PM2.5 AQI category bands behind each time series. Co-authored-by: Claude <claude@anthropic.com>
Combine wildfire smoke example into one four-panel figure
|
@scollis I have moved a renamed the example so sphinx knows how to build the example. Fixed the PEP8 as well. |
|
Thanks Zach. I will check this out
…On Thu, Jul 23, 2026 at 9:53 AM Zach Sherman ***@***.***> wrote:
*zssherman* left a comment (ARM-DOE/ACT#1036)
<#1036 (comment)>
@scollis <https://github.com/scollis> I have moved a renamed the example
so sphinx knows how to build the example. Fixed the PEP8 as well.
—
Reply to this email directly, view it on GitHub
<#1036?email_source=notifications&email_token=AAGJQB2N4C3A4JOIXE5SMRD5GIRG3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKMBVHE4DSMJZGEZKM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-5059891912>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAGJQB7LGJTQ4PQGXT3XK235GIRG3AVCNFSNUABFKJSXA33TNF2G64TZHMYTONJQGI3TMMRUHNEXG43VMU5TIOBZGQ2TQMRWGA3KC5QC>
.
Triage notifications, keep track of coding agent tasks and review pull
requests on the go with GitHub Mobile for iOS
<https://github.com/notifications/mobile/ios/AAGJQB6HDQBFD727L5NTKKD5GIRG3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKMBVHE4DSMJZGEZKM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJKTGN5XXIZLSL5UW64Y>
and Android
<https://github.com/notifications/mobile/android/AAGJQB6SFDR73NOG532BUVD5GIRG3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKMBVHE4DSMJZGEZKM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLTGN5XXIZLSL5QW4ZDSN5UWI>.
Download it today!
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
@AdamTheisen I was able to get the tests working etc, and this PR is ready for a final review. Figured to ping you as well for a review as this is AI assisted and for the first AI coauthored PR figured a second review would be good. |
|
Thanks @zssherman ! There are definitely some quirks that we might want to clean up. As of right now it looks to just be "faking" ACT plots |
|
Yep. We (Claude and myself) struggled with having a map and a set of non geo axes. |
|
@zssherman what do you think on this one? |
|
I saw this on my feed and I figured I'd mention if it helps, you can use WxData to retrieve data from the AirNow API as WxData has an AirNow API Python interface. Current Observations: https://edrewitz.github.io/WxData/air%20now%20api%20observations Historical Observations: https://edrewitz.github.io/WxData/air%20now%20api%20historical%20observations I hope you find this helpful! |
4 participants
I need to work out why the linting test fails.
AIRNow example from ongoing Canadian fires. Wanted a set up four panel plot with a regional plot and three time series.
Claude tried to follow examples in the sphinx gallery. May have messed up.
This should be treated as super low priority (fun). Happy to make any requested changes.