Openai/69235e49-f78c-8001-a70a-37b4e2caa0d3

This conversation was exported from ChatGPT

gpt-5-1

User: I’m looking for an automation to let me know where there’s new snow at Sunshine village or ski Louise.

I’m looking for an automation to let me know where there’s new snow at Sunshine village or ski Louise. There were some automated snow plots I could scrape but I think it’s gone. It was Mountain-weather.ca or something like that.

Assistant: Great—there is usable data for an alert/automation for fresh snow at Sunshine Village and likely Lake Louise — let me w…

Great—there is usable data for an alert/automation for fresh snow at Sunshine Village (and likely Lake Louise) — let me walk you through how you can set this up:

✅ Data sources found

• Sunshine Village: There’s a snow-plot station labelled “SUNQ1 – Sunshine Village Snow Site” on the Northwest River Forecast Center page, showing snow water equivalent (SWE), depth, etc. nwrfc.noaa.gov
• Alberta’s “Mountain Snow Course Data & Observation Reports” portal (via Alberta Environment and Parks / WaterPortal) shows they maintain similar data streams. Canada WaterPortal
• Lake Louise: There’s mention of an RSS feed for snow reports in the Lake Louise API docs (look at “SNOW-RSS.php”). api.skilouise.com

🛠 Proposed automation outline

Here’s how you could build an alert system for “new snow” (for example a threshold of +5 cm in 24 hours or similar) for either or both resorts:

1. Data extraction - For Sunshine: scrape the SUNQ1 station HTML. You’ll need to find the latest “snow depth” or “SWE change” values from the table. - For Lake Louise: hit the RSS endpoint (or any JSON/HTML endpoint if available) from the SkiLouise API to fetch the daily snowfall/24-hr change. - Store the previous reading each run (e.g., in a small database or plain file) so you can calculate delta (change) since last check.
2. Threshold logic - Define what “new snow” means to you (e.g., 5 cm or more depth increase, or some other SWE change threshold). - Compare the latest data to previous reading → if delta ≥ threshold, trigger alert.
3. Alert delivery - Choose your alert channel: email, push notification, Slack message, SMS, etc. - Include the resort name, time, depth change, maybe link to live webcam or conditions page.
4. Schedule - Set your script to run once or twice daily (e.g., early morning when new data comes in). - If you own a small server or use AWS Lambda / Google Cloud Functions + schedule, that's a clean solution.
5. Edge-cases / reliability - Ensure you handle missing data (e.g., early/late season when station not updating). - Might be good to fetch also the “resort‐reported overnight snowfall” from resort website as cross-check. For example SkiLouise says they report overnight, 24h, 48h values. Skilouise - Consider timezone / run time (Alberta MST/MDT) so you don’t miss the daily update window.

🔍 Sample implementation (pseudo)

# pseudocode
DATA_URL_SUN = "https://www.nwrfc.noaa.gov/snow/snowplot.cgi?SUNQ1"
THRESHOLD_CM = 5

prev_depth = load_prev("sun_depth.txt")
latest = scrape_depth_from_url(DATA_URL_SUN)   # parse html, find “Snow Depth (inches)”, convert to cm

if latest - prev_depth >= THRESHOLD_CM:
    send_alert(f"Sunshine Village: {latest-prev_depth:.1f} cm new snow; total depth {latest:.1f} cm")
save_prev("sun_depth.txt", latest)

And similarly for Lake Louise with its URL / RSS feed.

🤔 Some caveats

• The station data for Sunshine is in inches in that NOAA page. For example it shows “Snow Depth (inches)”. nwrfc.noaa.gov You’ll want to convert to cm if you prefer metric.
• The station may not exactly correspond to the same vertical terrain you ski on (though at Sunshine it looks like proper high-elevation station).
• Some resort reported “new snow” times might be slightly different than the raw station.
• Data availability: the NOAA page says “Please Note: Data provided by Alberta Environment and Parks” for Canadian stations. So the reliability is there but you must check whether updates are frequent. nwrfc.noaa.gov

If you like, I can look up the direct URLs for both resorts’ data (for example JSON endpoints or RSS feeds) and help you craft a full script (in Python or whatever you prefer) to run this automation. Would you like me to dig that up?