Category Archives: Spaghetti plots

Coming to Arizona-Catalina on March 18-19th: wind and dust followed by RAIN!

Sounds more like something coming to your local googleplex movie compound…but I am pretty excited today to be able to report a great model prediction that the title alludes to, one that has a higher degree of reliability than we usually see for that far in advance.

Normally one would not fool with a forecast of rain here that far in advance, more than a week, with any rain at all!  How many times have we seen that predicted rain in the models that far and more in advance, evaporate as the predicted rain day gets closer?  Well, I have anyway.  Way too many times.

However, as it has been written, the reliability of a longer range forecast by the models, such as a week or two in advance,  is discernible in something we call “ensemble,” or more colloquially,  “spahgetti” plots.  (Skip to pictures if you don’t want to know about these…gets a little technical.)

Occasionally, there are patterns in the jet stream that have high predictability, and these patterns produce “ensemble” predictions that are pretty much the same for one to two weeks in advance, even though the initial conditions are perturbed-that is, deliberately changed on purpose (haha) to see how much difference there might be in those longer term predictions because of the little changes.

You might think of these “perturbations” as representing bad data, “bad balloon”, lack of data, etc.    These small changes are introduced at the very outset of the model run and the models are run completely over again out to one to two weeks or more, to see how differently they look as the predicted days days accumulate.  At first, a “bad balloon”, bad data point, won’t have much effect, and so model forecasts just a couple of days out usually change very little.

But, if the pattern looks pretty much the same after, say 10 days out AFTER these changes are put in, then such a prediction has high reliability, a strong signal; what will happen has a high confidence level.   When the pattern is changed drastically with these little changes, then there can be little confidence is what is predicted.

Now, why am I going into all this (maybe useless) detail?

The strength of today’s blog title is due to having one of those rare situations where a situation has such a strong “signal” in the data that the forecast of a very strong trough here eight days away, has is showing a LOT of reliability in those “spaghetti” plots (shown below).

Re-inforcing this prediction a bit, too, is that in our Catalina climo data, there was a suggestion of a higher chance of rain in the third week of March than in the second week.   There really could be something in the global circulation that “likes” to put a big trough in Arizona and the West in the third week of March.

Another factor is that troughs are more common in the interior of the West in the springtime than at any other time of the year.  In many locations in the West (e.g., Seattle, San Francisco, Grand Junction), because troughs are cold aloft, March is the coldest month of the year overhead!  So, we would expect these kinds of events, just based on our usual climatology.

OK, back to our Catalina rain prediction:

Finally, below, if you have made it this far,  is what we are REALLY interested in, the areas of rain being predicted for Arizona-Catalina on March 19th (see panel below from IPS Meteostar).

First the rain prediction in panel 1 and the configuration of the jet stream over this same domain in panel 2.  You can see a huge southward plunge of the jet stream along the West Coast toward Baja and then see that it curls to the northeast after passing overhead of Baja.

If you read this blog, you know it ain’t going to rain here in SE AZ without the maximum wind at 500 millibars being south of us, and so, when you see so much rain in all of AZ, as in panel 1, you could have already guessed what this jet stream configuration would look like!  You’re friends will be amazed.  And, voila, there it is, where the jet should be!  Unless you have a telescope handy, you’ll have to click on these to get a resolution that you can see what the HECK I am talking about.


Next, we go into the ethereal world of spaghetti plots, this last panel, from NOAA.  These lines, some representing perturbations in the models, are pretty darn compact over Arizona on March 19th, and that, in turn, means a pretty darn reliable forecast.

Some details on this assertion:  take a look at the red lines, indicating a contour height on a 500 millibar map of 5700 m.  Compare the spread of our red lines to those in the Atlantic, where in that domain, there can be little confidence in what is predicted eight days out.  That 5700 m line is pretty much near the edge of the jet stream here, and the 5520 m contour lines (turqoise lines) well within in it. So, we are nicely sandwiched by those contour lines, meaning there is high reliability that there is a jet here.   Also, the yellow lines and gray lines, indicating the times of different model runs, also converge over Arizona.  I am so happy!

Finally, if you can make out the green “climo” line, you will see that the long term climatology favors this “trough in the West” pattern at this time of year!  Its all good!

What else can be confidently predicted?  Cold in the West, but also likely record warmth in the East with this pattern on the 18th-19th.

It will be interesting to see if Mr. Cloud-maven person really knows what he is talking about in the week ahead.  Of course, nothing can be guaranteed, but it sure looks like a rain is coming, finally!  What’s really certain in this longer range prediction, is the dust and wind part, dammitall.

The End, at last!