10/31/26 Saturday 2:00 pm
I have been waiting and hoping that the river people update their forecasts. But it does not look like that is going to happen. I hope they have more updates tomorrow than 1. That would be irresponsible given the situation, especially Pagosa, forecast to hit 4 1/2 feet above “Major Flooding Stage” with 11,432 cfs (flood stage is 4476 cfs)
About 40 minutes after my preliminary update on the areas forecast to exceed prior flood levels, NWS GJT put out flood warnings for affected areas. Nice to see them changing their tune (again).
I have a couple of models to share: one I have used before, the other- not before current pre-storm prep.
First, the NBM: I have used it a lot over the last few weeks. It is a high-resolution model that smooths low-resolution multi-model solutions. The good about that is it takes multiple model runs from “model members”.– Got very close to the rabbit hole there. Why do they call it that? Do rabbits live in a hole?- See, I told you.
I will save all of that for a boring weather post; it’s a good idea to talk about the differences between the weather models.
Been awake since 1 am, not an excuse-anyway.
Here is the NBM
These totals are higher than yesterday’s, before last night’s/overnight’s precipitation. It does a good job with elevation (un)smoothing.
I have not mentioned this before, but look at the bottom right of the precip map. See that “Max 6.78″? You can match that with the colors; in this case, the three potential areas that the models predict that 6.78” of liquid equivalent precipitation could fall are La Platas, Wolf Creek, or the Southeast San Juans (south of Wolf Creek).
Now another new one in the lineup. It’s an experimental NOAA model ready to officially replace the NAM3k next year.
It is RRFS (Rapid Refresh Forecast System); NOAA calls it “next generation”. I can’t confirm its accuracy yet; I haven’t spent enough time with it, and I’m anxious to see how it verifies. Right now, it could be my favorite model, especially this winter.
It is a convection-allowing model suite built on an FV3 dynamical core… I did it again.
Here it is, bottom line: my first reaction was that the model handles elevation in our area the way it really is. You may have heard me say in the past, “this model run is impressive, but there is absolutely no way this could possibly happen”. This model is exactly the opposite.
Look up by Purgatory: It actually registers the lower elevation at Purgatory at 550 versus the peaks on either side. Same thing with Animas heading up to Silverton.
My biggest concerns are the bullseyes where we don’t need them-Above Vallecito, Wolf Creek, and the La Platas.
Very excited about turning the clocks back; that means I get model data an hour earlier! (nerd)