AI Gets Personal

 Regardless of how we feel about AI, it's definitely here to stay. While it feels intrusive sometimes and occasionally just plain silly, but also pretty scary, it is proving to be a huge assist in some ways. I have experimented using it in writing and sometimes it's helpful, but I really don't trust it and try to find sources outside of AI to confirm and prove facts. It is only as good as the information being fed to it by humans - who sometimes make mistakes.

Yesterday I received notice from Noah Newman (no, not the one on Y&R!) at Colorado Climate Center, the headquarters for Community Collaborative Rain, Hail and Snow (CoCoRaHS). I began recording precipitation for CoCoRaHS in June of 2009 - daily all those years with only a few breaks when I was away or like last winter when there were a few days I didn't feel safe going outside to measure with a bum knee. Noah let me know that the information I have been collecting was used to train AI on how to better predict snowfall and snow-to-liquid ratio. In recent studies it's been found that the measurements CoCoRaHS volunteers take and submit daily are more accurate and useful than what the sophisticated weather stations used by the National Weather Service provide. Here's a direct quote from his email:

 “For this study, researchers used data from nearly 1,000 observers in the Community Collaborative Rain, Hail and Snow Network (CoCoRaHS), a grassroots program of volunteers who measure precipitation in their yards and at other sites.” And  “…published online in January 2026 in Weather and Forecasting, … The authors found that their machine-learning method “outperforms existing methods” used by the National Weather Service.”

Being selected for part of this study was a pleasant surprise. If you go to the CoCoRaHS site, you will see how many daily reports there were for today, so it amazes me that I was chosen to participate. 

 For the last couple of years I've contemplated stopping reporting as winter approaches, but so far I'm still trudging out to shovel off the deck, braving wind, rain, and sleet to take measurements most mornings.I am thankful I'm able to do it - but each year it takes a bit more effort and sometimes I think to myself as snow is swirling around me and I'm pushing heavy snow out of the way, "You old fool, you're out of your mind!" Maybe. But then something like this study comes along and I realize that what I observe is useful in some way to make weather predictions more accurate and traveling safer. Out west, observers' reports help predict avalanches in the mountains. Sometimes "hands-on" is just better than technology, but with AI, perhaps the combination of both techniques will prove even more useful. So I'll probably hobble out there as long as I can and make my report - mostly daily, but sometimes I resort to "multi-day reports" and wait for a break in the weather before going out. The precipitation I'm measuring is relativelly the same, whether taken every day or batched into several! It isn't rocket science, but turns out it's useful.