Strong early December storms indicated for Arizona in yesterday’s model run, now 24 h old

It doesn’t get better than this if you like wind, rain, snow, piles of it, and newspaper stories about Southwest storms. From IPS MeteoStar, these renderings for December 6-8th from our BEST model, the WRF-GFS, crunched from yesterday morning’s global data taken at 5 AM AST:


Valid for Friday, December 6th at 5 PM AST. The colored regions where the model has calculated that precipitation will fall during the prior 12 h before this.  Precip indicated in just about all of Arizona

Valid at 5 AM, Saturday, December 7th. Twelve more hours of precip throughout most of Arizona!

Valid at 5 PM, Saturday, December 7th, precip center moves to just about over my house/Catalina and environs again as it has so many times in the past.

Valid for 5 AM, Sunday, December 8th. Rain area intensifies and grows in size over SE Arizona!  I was beside myself with joy when I saw this model output!


Now, after all this excitement, you may wonder why, with several model outputs since yesterday morning at 12 Z (5 AM AST) the writer has not updated these maps.  If you have followed this site for any time at all, you know the answer; the subsequent model outputs were not as good.  In fact, the last one, from yesterday evening’s 5 PM AST global data, had NO PRECIPITATION at all south of about Winslow.  Such an outrageous, even if possibly more accurate model output due to being based later global data, is NOT going to be displayed here!  I want you to be uplifted by thoughts of coming storms and rain, spring wildflowers in profusion, not droughty thoughts.

But we have a conundrum: Will it or won’t it rain here in Catallina in the first week in December?

And, of course, you know what happens here.  We examine that ball of yarn, those “Lorenz plots”, namely, the spaghetti plots from the NOAA spaghetti factory that puts them out, those plots you’ve come to love, to see if we can tease out the answer to our conundrum; which model run was the outlier, the one likely to be WRONG!

Let us first look at the projected contour map for 500 mb, about 18,000 feet above sea level, or about half way through the entire mass of the atmosphere, which is kind of scary when you think about how low that is:

Valid at 5 PM AST, December 7th.  We will compare the 552 height line with those "errorful" ones on the Lorenz-ball of yarn plots.

Valid at 5 PM AST, December 7th. We will compare the 552 height line with those “errorful1” ones on the Lorenz-ball of yarn plots below, and see if a lot of those same height lines show up SOUTH of Tucson, as it does here in this great prog.  Fingers crossed….  We need a LOT of those same contours south of us for this NOT to be an outlier, goofy model output.












Here it is:

Valid at 5 PM, December 7th, during height of predicted strong AZ storms shown above.

Valid at 5 PM, December 7th, during height of predicted strong AZ storms shown above.  The light blue lines are the errorful, key 552 contour lines.  Look, there’s only ONE south of Tucson, very unlike the stormy model output showed above!  Most of those bluish 552s are WAY to north, indicating that the stormy Tucson mod run was an outlier, not likely to happen as shown.

———————————-Gigantic explanatory footnote—————————————-
1“Errorful”–if you can imagine it, our government weather service puts little errors in the global measurements that were actually taken, and then runs their computer models over and over again with slightly different combinations of DELIBERATE errors.  They do this because we cannot measure the atmosphere accurately over the whole globe at the same instant.  There are always measurement errors, but we don’t know what they are.

The stronger a pattern is, the less it is changed in the future even with those tiny errors having been introduced at the beginning of the model run. The more fragile a pattern is, the more chaotic the future prediction of it gets. This means that where patterns are strong, the red and bluish lines are bunched together, and where the model is clueless, the lines are all over the place.

Naturally, in the first few days of a model run, pretty much all of the lines are bunched together, but after awhile, the fragile patterns start to emerge as the lines go crazy. Here in the western US, the plot above indicates, even though its 10 days out, that there is, without doubt, going to be a strong trough, lots of stormy action in the western US. The blue and red lines are RELATIVELY bunched together.

The thing that’s missing for us, is the AMPLITUDE of the pattern, that well-predicted trough with storms in the West does not appear like its going to reach down to Tucson, but rather produce rain and snow in the northern third of the State.  On the other hand, its not so far away that further slight changes won’t result in precip here, though not the behemoths indicated in the progs at the start of this blog, which I think was yesterday sometime.

In the cool half of the year, its essential for the core of the jet stream to be over or south of us for rain to get here, and that core is located between the red and blue lines above, so its not so far away.  Fingers crossed…

Finally, The End (of the footnote, and everything else)