Category Archives: Weather model discussion

Powerful hurricane to not enter Arizona even though the model shows this happening; horse story

Many of you probably were gasping for air after having seen the WRF-GFS model outputs from last evening’s 5 PM AST global data.

A large hurricane, really more the size of its typhoonic big brothers in the western north Pacific, and one that also dwarfs the late tropical remnant, “Newton” ,  that came through here a week or so ago, is shown to move along the SAME path as Newton into Arizona in about 13 days from now.

For those few of you who did NOT peruse the 00 GMT, CUT,  Z output, here are the fantastic fantasy hurricane depictions that this model, with all of its calculating power, shows entering AZ on the 26th.  Kind of fun to see even if it is bogus because it indicates that such a strong tropical cyclone could come through here one day.

Below, from IPS MeteoStar, these, maybe the best fake AZ hurricane depictions I have ever seen.  Note all the isobars, i.e., lines of equal pressure with this tropical cyclone in AZ, and then remember for all its rain, little Newton had virtually no signature on pressure maps! Hell, the pressure didn’t even fall at Nogales as Newt approached.  Pitiful.

But it wouldn’t be like that in this fantasy hurricane.  Tremendous pressure falls would occur as it entered AZ giving your microbarograph quite a workout as the pressure plummeted and then went up as the center passed by.

You do have a microbarograph don’t you?  If you don’t, think about it.

“Invalid” (haha) for 5 AM AST September 25th.
Invalid for 5 PM AST September 25th.

Next,  you’re curious, though,  about what steering pattern caused this hurricane, previously shown to stay far offshore and dissipate over some jellyfish and plastic particles way out in the Pacific in the models.

Let’s look, again from IPS MeteoStar at the steering situation at 500 millybars, or in around 20,000 feet or so:

Here the configuration. You're breathing a sigh of relief, maybe even chuckling: "That's not gonna happen." Ludicroous really, though withing the slightest realm of possibility, maybe one in a thousand. Like kicking a field goal that goes through the uprights after bouncing off an opposing player's helmut. I mean, it could happen, like a golf shot at Carmel that bounces off a stunted cypress and goes into the hole from 500 yards out, or.... OK, enough of that.
Here the configuration. You’re breathing a sigh of relief, maybe even chuckling: “That’s not gonna happen.” Ludicroous really, though withing the slightest realm of possibility, maybe one in a thousand. Like kicking a field goal that goes through the uprights after bouncing off an opposing player’s helmut. I mean, it could happen, like a golf shot at Carmel that bounces off a stunted cypress and goes into the hole from 500 yards out, or…. OK, enough of that.

2016091200_WST_GFS_500_HGT_WINDS_336 2016091200_WST_GFS_500_HGT_WINDS_348 2016091200_WST_GFS_500_HGT_WINDS_360

What you need to have any confidence is a big trough along or just offshore as we had with Newt, not a slight little itty bitty eddy aloft that has to be in exactly the right location at EXACTLY the right time.  I mean, its like a ball that goes for a home run after it bounces of the  center fielder’s head1

Hold your cash on the sand bags.

Finally, there’s really nothing from the spaghetti factory that supports this.  Boohoo.  What you need in spaghetti is strong support for a trough along the coast, not the below:

Valid at 5 PM AST September 25th.
Valid at 5 PM AST September 25th.

Yesterday’s clouds

Spectacular Altocumulus castellanus and floccus (no virga) passed overhead during the morning.  I hope you documented them with a few photos.

9:11 AM, on the trail looking at a superb example of Altocumulus floccus. Estimated height, 12,000 feet above ground level. No ice visible, so tops likely warmer than -10 C.
9:11 AM, on the trail looking at a superb example of Altocumulus floccus. Estimated height, 12,000 feet above ground level. No ice visible, so tops likely warmer than -10 C.  The bases of these clouds evaporated almost as soon as they formed, but the tops not so much, rose steadily after bottom disappeared.  Tallest ones were likely more than 1,000-2000 feet tall.
12:16 PM. By this time, which was good, smallish Cumulonimbus clouds recurred over the Catalina Mountains mostly east of Ms. Lemmon. Nice stages of ice development in the tops from newly risen, no sign of ice, to frizzy all ice remains, over and over again.
12:16 PM. By this time, which was good, smallish Cumulonimbus clouds recurred over the Catalina Mountains mostly east of Ms. Lemmon. Nice stages of ice development in the tops from newly risen, no sign of ice (right side here), to frizzy all ice remains (left side here), over and over again.

Horse story

Have to depart from clouds and weather to tell this tale.  Yesterday I stopped here to let the mighty Zeus rest a little.  I let him graze “off leash” on some of the still-green nettle grass in a gravel parking area next to our cottage.  I then went to get a pail of water for him, the pail being on the north side of our house.  When he saw I was leaving,  he immediately followed me like a dog.  It was kind of cute.

But as we got to the gravel outside the north porch of our house, our two dogs, Banjo and Emma were going nuts at the sight of a horse outside the north windows.

Zeus got distracted by all of the commotion in the house and went onto the porch to look in one of the windows to see what was up, or maybe he saw his own reflection and thought it was another horse?  Here is the hilarious scene:

2:21 PM yesterday. Zeus looks in to see why the dogs are barking so much.
2:21 PM yesterday. Zeus looks in to see why the dogs are barking so much.


The End



1This actually happened in South Dakota,  at Mitchell’s Cadwell Park,  during a  baseball game I played in ’72.  I was catching in those days for Mitchell Commercial Bank.    Our center fielder,  a track star, ran to get a scorching line drive to medium depth center, and racing to his left,  reaching up to grab it, the ball instead bounced off his noggin and went some 40 or 50 feet over the fence!  He was OK.   We had no “concussion protocol” in those days.  Had a chance to bat against the legendary Canova, SD,  pitcher, Lee Goldammer in that game.  Whiffed on three pitches;  was maybe at bat for 30 seconds.

“Tweener” era begins today after pre-dawn sprinkles; one photo has birds in it

We’ll have to suffer through  a few days for the next storm, i. e., experience sunny weather with pleasant temperatures.  Its amazing that people all over America come to Tucson to experience sunny days with pleasant temperatures!

0.45 inches total in The Heights of Catalina in this latest round of rain, sounds of rain.    Actually, there was also some tiny graupel/soft hail in the rain yesterday, too.

Graupel indicates a lot of cloud droplet water overhead, and that ice crystals were colliding with them until they lost their identity and became little snowballs.  In regions where there are very few ice crystals,  graupel and the harder version,  hail often form.   Its likely that nearly all those rain drops that came down with the little baby graupel were melted graupels.

Graupels…..   Makes me think of that rock group, Led Graupelin, didn’t have the impact of Led Zepelin.  But I have LG’s one and only album entitled, “Compare to Led Zepelin.”   Was only $2.99, too!  Where’s my guitar?  I think I will play, “Stairwell to Heaven” now…

When graupel or hail occur,  there’s a pretty good electric charge up there in those Cumulonimbus clouds.  Its best to be indoors when its hailing until you know if a strike might occur (if there hasn’t been one already).  Besides, its not comfortable being out in hail.   And if you were listening to the rain, you heard a few blasts of thunder toward Lemmon around 2 PM that came out of one of the more enthusiastic Cumulonimbus clouds that went by.  Got 0.12 inches total to add to the night before’s rain of 0.33 inches.

Yesterday in clouds; a sojourn in clouds from morning to evening, in that order with no times noted











I know how much you like to see pictures of rain, so here’s one. You’re not like the “others” are you those people around you every day?


DSC_1569Then the piston of atmospheric subsidence slammed down to squash our Cumulus cloud tops to levels and temperatures where ice could not form….


Two picures in a row of a NWS-style rain gauge. Probably has never been done before. Has been getting a workout lately.  Everyone should have one.


The weather ahead

Kind of funny to see the Canadian GEM model internally plagiarize itself.  Compare last night’s panel at 500 millibars (below) with that same level’s panel  foretold for six days from now.

Yep, its the same thing over again in six days, though with less rain IMO:ann yesterday at 5 PM AST

ann 6 days from now
Valid on Tuesday, November 10th at 5 PM AST.  From the Canadians.

In another interesting model development, the best USA model, the WRF-GFS is having an internal CONFLICT of major proportions.  Check these progs out generated by data only six hours apart.  The first one, showing a big trough coming into Cal, was generated by global data taken at 5 PM AST last evening.  The panel below it was generated by the same model based on global data taken just six hours later, at 11 PM AST (so it the most recent WRF output available) and has a big ridge along the California coast.

Valid at 5 PM AST November 20th.

“Which one will the fountain choose?”, to quote old song lyrics1:

Valid at 11 PM AST November 20th.  From IPS MeteoStar.

Of course, spaghetti tells us which one is right, mostly.


The End


1Except that here we present only two “coins” not three.

Another Catalina rain day for May 2015

We received 0.08 inches here in “The Heights” for a third day with measurable rain in May already.  0.12 inches fell at the Bridge on Golder Ranch Dr. , while Saddlebrooke got up to a quarter of an inch (as estimated by CMP) in a tiny streak of clouds that erupted into shallow Cumulonimbus clouds, anvils and all yesterday afternoon between 4:30 and 5:30 PM. It was pretty much all over by 6:00 PM,  those shower clouds passing off toward Mammoth.

No rain was reported at mountain sites, to give you an idea of how localized that was, localized practically to Basha’s Market parking lot,  Sutherland Heights’ Equestrian Trail Road,  and Saddlebrooke’s Acacia Drive, to exaggerate some.

The astounding aspect of a tiny line of showers that suddenly erupted over and a little downwind of Catalina was that it was EXACTLY predicted  in the University of Arizona’s 5 AM AST model run yesterday morning, one whose results are available by mid-morning!  So, there would have been a few hours notice of possible rain here in Catalina.

There is no rain predicted in that model run anywhere else except in extreme NW Arizona, just that tiny ribbon of rain right over us, and the U of AZ “Beowulf Cluster” weather calculator got it right.

However, unless you were in the right spot, you might not have even known that it rained, the shower streak was so narrow.

Below, the astoundingly accurate predictions for 3, 4, and 5 PM for that model run from yesterday morning.  No rain whatsoever is shown at 3 PM, as you will see.

Ann 4pm

Ann 5pmTo be “fair”, NO RAIN was predicted anywhere NEAR Catalina by that same model crunching the data from 5 PM AST the evening before our little rain event, leading CMP to be a little asleep at the wheel yesterday morning, no blog.

Some cloud shots before and as this predicted (or unpredicted, as the case may be) rain began to happen.  Of course, if you want to go to the movies and see this, go here, from the U of AZ:  Yesterday’s cloud movie

3:23 PM. Looking upwind. Nada going on, clouds to shallow for ice.
4:04 PM. Hmmmm. Clouds definitely fattening up upwind of us in a nice cloud street from upwind of and over Pusch Ridge to Catalina. Nice scene, anyway, even if nothing happens.
4:12 PM. Cumulus clouds are looking to gather down there at the corner of Pusch Ridge and Oracle Road, Huh. And they’re heading in this direction. Can ice really form in these guys today?
4:23 PM. Clouds over and downwind of Pusch Ridge continuing to gather while heading toward Catalina. Looking for ice now or virga, but don’t see any anywhere.
4:33 PM. Ice begins to show up in even modest clouds! And it started to show up everywhere in the form of virga. But then then the virga became thin shafts all the way to the ground. Cloud Maven Person is beside himself, but must go indoors for a social engagement!  Those of you who fancy yourselves as Cloud Maven Juniors, should have recorded this sighting of “first ice” in your cloud diaries for yesterday.
Also at 4:33 PM. Ice is now readily visible in that Cumulus mediocris massing upwind of Catalina and is not about over the south part of the Catalina CDP (“Census Designated Place” that might one day be absorbed by Oro Valley, rumour has it.)
4:41 PM. Just minutes before the first drops fall in Saddlebrooke, and CMP’s last photo of this incoming masterpiece of weather forecasting and little rain band; he can no longer comfortably jump up from dining room table in mid-conversation to say he has to pee again whilst actually taking a photo. You can only say you have to pee so many times in 10 minutes and still have your credibility intact. But, how can you say repeatedly, “I have to go look at some clouds?”, so I can’t hear the rest of your quite interesting story……”  Really came down for a couple of minutes several times there in Saddlebrooke between a quarter of five and 6 PM as one little raining cloud formed after another in this cloud stream.

Below,  sat view of this cloud streamer with radar, from IPS MeteoStar.  The image below is at the same time as the last photo above:

Satellite and radar imagery for 4:40 PM AST.
Satellite and radar imagery for 4:40 PM AST.  Note the many lines of clouds running almost due south to north into SE AZ from Mexico.

Here some more cloud stuff from the sounding launched at the U of AZ around 3:30 PM AST.

The TUS balloon sounding ("rawinsonde") for yesterday afternoon.  Looks like most tops were dabbling with the ice forming temperature of -10 C, but the sounding suggests that somewhat deeper tops could easily have arisen (and did!).  Interestingly, the model "knows" when ice forms, and it must have "known" that the ice-forming temperature was going to be surpassed in that little cloud line coming off Pusch Ridge.  Astounding, for the Nth time.
The TUS balloon sounding (“rawinsonde”) for yesterday afternoon. Looks like most tops were dabbling with the ice forming temperature of -10 C, but the sounding suggests that somewhat deeper tops could easily have arisen (and did!). Interestingly, the model “knows” when ice forms, and it must have “known” that the ice-forming temperature was going to be surpassed in that little cloud line coming off Pusch Ridge. Astounding, for the Nth time.  Bases were pretty cold, 0 C (32 F).

Here’s a diagram of when ice forms in the type of clouds we mostly have in Arizona, “continental” ones with high droplet concentrations, and when ice should form in them.  As you can see, ice should form in them soon after the top temperature gets colder than 10 C WHEN the base temperature is about what it was yesterday.


From a survey of the onset of ice formation in continental clouds by Rangno and Hobbs (1995)1
From a survey of the onset of ice formation in continental clouds by Rangno and Hobbs (1995)1


“CMP” is not mentioning it at all, but yesterday was another kind of mucked up sky, not a Catalina postcard sky,  with lots of aerosols making the sky a whitish-blue, the lower aerosol stuff again from Mexico, but there was also a layer far above the cloud tops, likely a long-range transport event from thousands of miles away.

This higher haze layer still seems to be around if you look toward the horizons right now (5:59 AM).

We’ll be between two jet streams today, kind of a jet stream sandwich, and the stronger one is now approaching from the northwest with that mega upper low over Cal.  That means no rain today, subsidence rules, though we’ll have small, non-ice producing Cumulus, and likely some Altocumulus lenticulars, maybe a Cirrocumulus patch here and there.  Should be a pretty nice day for cloud photos, haze aside.

The best chance for rain is still after midnight tonight into mid-day tomorrow as the core of the stronger jet stream goes just about over us.  Still thinking a tenth of an inch will occur here, though mod run from the U of AZ at 5 PM completely dry.   A little snow likely on Ms. Mt. Lemmon, too!

The End, FINALLY!  Brain empty.


1From “A New Look at the Israeli Cloud Seeding Experiments.”




Canadian model wetting it up for SE AZ on the 10th; US mod just says “no” to AZ rain

Hot off the Canadian presses from last evening’s global data from 5 PM yesterday, this exciting depiction (from Enviro Can) for Friday afternoon, the 10th.  Note green and yellow regions in all of the SW, lower right panel!  Note big upper trough over Vegas, upper left panel.  Its all good, and, this being the 5th, its only a few days away!


Valid Friday afternoon at 5 PM AST. Note “tropical river” emanating from a tropical storm or depression off the tip of Baja (colored regions in lower left panel).  No reason to be depressed if this happens.

But let’s look deeper….deep into the heart of the US WRF-GFS model for the exact same moment in time, shown below:

Also valid at 5 PM AST on October 10th.  Perhaps the US model is sequestering rain for Arizona.... NONE, and in nil, is indicated from a model run generated from EXACTLY the same data as the Canadian one above and it shows no rain for much of anywhere in Arizona.  How bizarre is that?
Also valid at 5 PM AST on October 10th from IPS MeteoStar. Perhaps the US model is sequestering rain for Arizona…. NONE, as in nil, is indicated from a model run generated at the same time as the Canadian one above!   As you can see from the arrow, there is NO RAIN indicated for most of Arizona as a very strong upper level trough goes by. How bizarre is that?  Something’s goofy here and its not just me.

What caused this model bifurcation?

For one thing, the US mod sees no tropical depression off Baja, and our hurricane center is very skeptical that any tropical development will take place south of Baja, one that would move NW or N and send a moist plume up into a trough positioning itself over southern Cal on the 9th.    US mod has NO tropical moist stream coming out of the eastern Pac ahead of a trough over southern Cal at all, whether there is a low down there or not.  Boohoo.

So, any rain here in Catalina would be with the usual passage of the jet core and in the cold Pacific air behind the cold blast headed our way on the 10th-11th.  And that rain would be very light, likely less than a tenth of an inch.

The Canadian model, sadly,  has to be considered a little goofy right now with its copious SE AZ rains, and widespread rain/snow over the rest of AZ, too.  It would be good to inform your neighbors that you are thinking the same thing about the Canadian model, i.e. that its a bit looney as well in case they saw it and are touting a lot of rain on the way without really looking into things.  Embedded editorial note:  A lot of people vote like that, too, I think;  also read health claims in magazines without looking at the peer-reviewed literature, taling with their doctor, examine the results of double blind,  randomized trials, that kind of thing on which we base the whole edifice of medicine on.  No, they’d rather read on the side of a health pill food, “these claims have not been evaluated by the FDA” and then take the stuff anyway and enjoy the effects of a placebo, which, if you believe in it, can be pretty good.

No doubt the reports of low temperatures in Arizona with this coming big upper trough on the 10th-11th, and snow in the high mountains, will spur a flock of snowbirds to migrate south.  It’ll be windy and dusty, too, as it usually gets ahead of the passage of the cold front with these big troughs.  That will be interesting, as well as all the cloud forms that start to show up with it.

Note:  I’d show you some NOAA spaghetti so that we could get a little more on this model discrepancy, but its been sequestered from me, as you will see here.  In the meantime, will hope for Canadian vindication in the days ahead.

The End.

Hoping Canadians win model forecast duel

This from Canada last night for April 8th at 5 PM AST:5 PM APril 8 00_054_G1_north@america@zoomout_I_4PAN_CLASSIC@012_144  appropriate descriptor for AZ; “juicy.”  This was such a great find early this morning!

Note deep upper low at 500 mb in eastern Cal (upper left panel) and gigantic surface low centered in Four Corners area (upper right panel). Would almost surely mean rain here in Catalina 24 h after these panels (the last one in the forecast series).  If these maps verify this would be another billion-trillion dollar value storm in drought relief for the Southwest AND Plains States over the days following this map.

However, you will be as moribund as I was after looking at the USA WRF-GFS model output for the SAME moment in time, 5 PM AST April 8th below.  I had hoped both models would show the same thing, which would build confidence on what’s going to happen on the 8th-9th, but they are vastly different!  Take a look at THIS upper level pattern: no low in eastern Cal, just a strong jet pouring down from the Pac NW with cool air.  No moisture of consequence here in Catlandia with a pattern like the one below.

Also valid at 5 PM AST April 8th, for the 500 millibar pressure level.
Also valid at 5 PM AST April 8th, for the 500 millibar pressure level.

So who you gonna call when these kinds of things happen?  Spaghetti!

Well, if you know anything at all about “spaghetti” you will only get more moribund, maybe start crying when you see it for the same time as these models.  The Enviro Can mod is clearly an outlier “solution” even if though are dealing with different models.  Inserting some “chaos” in form of “bad balloons” (bad data) at the start of the USA model run does not reveal a pattern with a low in eastern Cal in any of the “ensemble” model runs,  but only results in a strong signal (line bunching) for a jet to whoosh down the interior of the West Coast that then loops back toward the northeast over and east of us–not good for precip here, just a rush of cool-cold air.

However, it WILL still be a pattern that’s great for Texas and the Plains States in general, so let us not be selfish in our dryness and begrudge others who get rain, but rejoice with those droughty others who will get so much relief beginning around the 8th as shown here–and that relief lasts for about three days, too.  Just hope there aren’t too many tornadoes in Texas and eastward…

As an aside, it might be worth the drive to central Texas to see some of those Big Boys out there, get some perspective on Nature’s power.

Valid for 5 PM AST, April 8th.
Valid for 5 PM AST, April 8th.


Yesterday’s clouds, those small Cumulus, almost beyond the curvature of the earth to the north-northwest

Only the truly great cloud maven juniors of our time would have observed and logged in their weather diaries those tiny Cumulus clouds (humilis and fractus) that appeared momentarily off to the north-northwest horizon, barely visible, from Catalinaland around 5:12 PM. Since I had foretold some distant Cumulus to the north yesterday, I damn well was looking the whole afternoon, straining eyeballs, and was starting to feel sad until I saw this one cloud, and then I was SO happy, euphoric really. I think this one, and a couple of others needing a microscope to see, were there for about 15 minutes is all. Here is that photo-documentation of a small, distant Cumulus humilis. I’ve added some writing on this photo to help you find it, but you will have to blow it up.

Also, its OK to log things you missed in retrospect into your weather diary; it helps make it more complete.           SONY DSCBTW, there were also a few little patches of Cirrus.

The End.



Rejected! (that 06 Z mod run from last night with no rain here for 15 days)

I had to laugh when I saw the very latest WRF-GFS model run, the run from last night’s 11 PM AST (06 Zulu) that had not ONE green pixel of rain over Catalina in the next 15 days!  You can see that really bad run here, from IPS MeteoStar.  You’ll get quite a guffaw out of it, as I did.  Quite bogus, really, or, as Wallace Shawn’s character repeats several times in the movie classic, “Princess Bride”, “Its inconceivable.”  (Of course, he was wrong, but we’ll ignore that.)

Completely dry cannot happen here over this 15-day period, period.

I was really afraid you had looked at this model run right off the bat this morning and had gotten down because of it, so I thought I’d better take it on right from the get go to get you going; so you wouldn’t be sad at work thinking about it all day.

What would be the reasons for firmly rejecting this latest model run, as a defender might a potential layup in a basketball game, this model run send up, the result of billions of calculations on our best, fastest computers?  They’d better be pretty good ones, we know.

1.  Its based on data at 06 Z when there are not so many observations, lots of gaps in the data, filled in by what the prior model run thought would be in those gaps (possible “fantasy” weather in those).

2.  Wildly different from the model run just 6 h earlier (00 Z) 5 PM AST, one that was based on oodles of global data.

3.  Is inconsistent with a steady depiction of several rains here over the next two weeks in prior model runs. The big storms during the last week of December into early January, written about here for some days now, are only close calls for us in this latest 06 Z run.

4.  It doesn’t agree with the superior Environment Canada model run, either, on the first of these coming rains.   For example, the Enviro Can mod has rain here late on Christmas Eve into Christmas Day, as it has steadily of late, while the 06 Z run has zip.

5.  Principal reason designate last night’s 11 PM (06 Z) run as a “laugher”, an “outlier”, completely bogus piece of computer trash?

The NOAA-NCEP (National Centers for Environmental Prediction) “Ensembles of Spaghetti“, which we often discuss here.  Let’s look at one of these, one valid for 5 PM AST, December 28th.  Looks like rain here to me.

Here’s where both nascent CMJs (cloud-maven juniors) can really shine because they know about “spaghetti” and when to reject “outlier” model runs.  And they can also impress others with their black, really great-looking,   “I ‘heart’ spaghetti” tee-shirts with the colorful example on the front.  Shows your in with the “in” weather crowd, always a great thing.

Note the positions of all the red lines (570 decameter heights of the 500 millybar surface) in the above example.  They bulge down toward the equator magnificently in the eastern Pacific and to the south of the SW.  These red lines virtually guarantee a big trough in the lower (southern) portions of the jet stream on this day and the days around it.

While ordinary folk, and maybe some of the weather practitioners on TEEVEE making huge amounts of money might be chagrined, taken aback by that latest, 06 Z model run and back off big claims and dreams of strong storms ahead, the CMJ will stand tall against this model run and continue to proclaim significant rains in the last week of December into January, ones that will juice up our water year precip totals significantly.

Well, at least I will.

One final note….  The exact day of the major storms in late December are unknown; those will wobble around until the smaller elements/waves in the global picture come more into focus over the coming days.  Significant rains might occur on the 28th-29th of December and/or the 30th and 31st, with still another possibility in early January (a bit more dicey, of course.)

In the meantime, before Christmas, a glorious Arizona winter storm break.

The End.



Rain and when today

For those in a hurry:  no rain before 5 PM AST, rain beginning between 5 PM and 8:42 PM; most of rain after midnight into mid-day.   Since people forget, I’ve added a musical reminder there: “After midnight, we’re gonna let it all hang down.”  Could be that incoming front talkin’  to us.

It will also be a fantastic day for clouds, probably some nice lenticulars today downwind of the Catalinas and elsewhere as the wind strengthens above us.  Keep your camera handy.

For retirees, or others who have a LOT of time….

The long story line about how to forecast when

Its ALWAYS fun and a challenge to try to foretell when the rain will start to fall by the clock.  Did some radio work like that in Seattle, i.e., “rain beginning between 11 AM and 2 PM”, “no rain before noon”, and also for the Washington Husky baseball and softball teams, when to take the tarp off, resume play, when put the tarp on, etc.

Those radio forecasts were made early in the morning on weekdays (KZAM-FM, 1982) and weekends (the latter during Weekend Edition on KUOW-FM, an NPR affiliate, 1987-1992).  It was thought that such a forecast, given early in the day,  would be the most useful one for a listener planning his/her day if it could be done well enough. This would be ESPECIALLY true on weekends when the listener could do whatever with his/her time rather than be in a building all day at work.  But, could you do it well enough using the surface obs, weather maps, satellite imagery, and model forecasts generated the evening before, such as they were in those days.  It was deemed a forecasting experiment in those days.  It wasn’t being done elsewhere in SEA.

BTW, there was no weather radar in Seattle during those days if you can imagine such a thing.

However, I will also add that radar is really not that useful in SEA anyway since forecasts for the next 8-12 h is a window of time in which radar can’t be used, for example if the incoming rain is still off the coast of Washington!  Reading the sky was much more important in determining how close the rain was, integrating that interpretation with the satellite imagery.

Or those days when showers would form that aren’t present in the early morning when you’re giving your forecast.  The passage of a front in Seattle, BTW, is usually followed by a brief clearing followed by passing showers, unlike say, back East where once a band of rain goes by, that’s it for the rest of the day.

So, here we are in Catalina facing a great storm, our best of the winter.  How close can you come to getting the start time here in Catalina?  That’s the game.  Today, the models are SO MUCH better, you have to look at them carefully, and know whether they run a bit fast in bringing in rain, and if there are any significant errors built in.

OK, here goes the first look, based on the passage of the core of the jet stream at the 500 millibar level, around 18,000 feet.  The criteria that rain starts when its overhead and after it passes was developed during the early 1970s whilst the author, a forecaster, was working for a HUGE randomized cloud seeding experiment in Durango, Colorado.  It was found that almost no precip fell in Durango before the 500 mb jet passed (95% of the wintertime precip there fell only after this happened).  That study was later expanded to the ENTIRE US and it was found that wintertime (November through April) precip in the interior of the Southwest also followed that Durango criteria very closely; it was almost a black and white predictor.

So, let’s look at when the jet passes over us here (using IPS MeteoStar’s great forecast renderings again) and see what time it goes by (of course, in this era, the models also “know this” relationship; not so well in the early 1970s).

Valid for 5 PM AST today! Note the coloring. It shows where the velocity of the wind is higher and lower.  Yellow, browns and reds are highest.

Valid for 11 PM AST tonight. Should be raining real good by now; jet has passed overhead and is to the south and east of us.

Like Seattle in some ways, this early onset of rain is from clouds for the most part, are ones that haven’t formed yet!  They’ll be forming and filling in to the southwest of us and here, deepening as this big trough works inland toward us.

Next, if you were to measure the movement of the frontal rain band now in southern California over the past 12 h or so with a ruler, or piece of paper, you would find that at its present rate of movement that frontal band would be here around 10 AM tomorrow.  So, you would have rain starting in the evening, peak rain in the morning.  Done.

Now after this simple exercise, let’s see what the BEST model has in mind for the rain start, that from the U of A, processed by their intimidating “Beowulf cluster” (you can call it up here) based on data from 11 PM AST last night.  This should be a very accurate forecast, and we hope resembles what was said above; no rain before 5 PM, but raining soon after that.  (BTW, I have not looked yet; part of the “game” today;  can you do as well as the model with a simple approach?

OK, here is the coverage of rain expected by 5 PM AST in the U of A model:  not much, but its upwind.

Coverage of rain by 5 PM AST. Not much.
Rain coverage by 9 PM tonight. Should have rained some to be redundant.

In sum, that simple technique was not so bad.

The main frontal band is still to the west at this point, 9 PM AST,  and it doesn’t pass over until after midnight. Then the rain lingers into the mid-morning to early afternoon.  Amounts still look VERY substantial. This model projects around a half an inch or more here in Catalina by tomorrow night.  The range of values, given the various uncertainties in models and clouds and weather, bottom amount 0.25 inches (only 10% chance of less), top amount, an inch (only 10% chance of more), these percentages generated from this keyboard, BTW.

Here is the Beowulf total storm rain ending at midnight tomorrow night:


What’s after this?

Mods still showing spate of storms over the next couple of weeks, this one not a one-hit wonder.  And there’s support for these storms in our venerable ensembles of spaghetti.  Main brunt of storms: California.  Expect to read about them.

But are our rains too late for our spring flowers?  Dunno.

The End, at last.

Looking for rain in all the wrong places…

Like in our best USA! models.

Pretty upset this early AM to find that the US’s Weather Forecasting and Research-Global Forecast System (WRF-GFS) model run, a model costing millions of dollars BTW, ingested last night’s 5 PM AST global data, BUT then threw up an identical twin that matched the Canadian Enviro Can model output that came out 24 h earlier!  It was unbelievable to see this, humiliating really, something akin to a reverse nose job.

Recall that the USA! model had rain here and a big cold trough right over Catalina on the evening of December 9th into Monday morning the 10th.  The Canadian model had that SAME trough over Cornhusky Stadium, Lincoln, Nebraska!

The Canadian model was right.

Here’s are the two forecast maps made within 24 h and each for for the SAME DAY AND TIME by our own WRF-GOOFUS model:  on the left,  the rainful run from the previous day that made me so happy (until I had some “spaghetti” and saw it was likely a bogus output).  The panel on the right is the sickening output from last night, both rendered by IPS.

Valid for Monday, December 10th at 5 AM AST. Sweet!

From last night, also valid at 5 AM AST, Monday, December 10th. Horrible, unbelievable amount of change between the two.  Makes you feel sad for weathermen and weatherwomen that have to deal with these things.









I really wanted Enviro Can to eat some crow with their forecast of MY trough over Nebraska.  But no!  “Bow down to Canada”, as heard here if you substitute in your mind the word, “Canada” for “Washington.”  Hey, its got the lyrics at this site and so it should be pretty easy for you to sing along with it.

BTW, the Canadians (Enviro Can) don’t feel they have to show “spaghetti” plots to reveal how bad their numerical forecasts might be because they are always so right (in the 144 h time frame available from Enviro Can).  “Don’t need no spaghetti.”

Can we say the same?

Doesn’t seem like it.  We need “spaghetti” so we can see how bad our model forecasts might be.  Calling Obama now…. not “happy with crappy”, to quote some overseas manufacturer’s creed, here.  OK, our models aren’t exactly “crappy” but they aren’t as good as they should be.

Too, I have to deal with Canadian relatives that will be gloating today, I am sure.   Maybe this spectacular example of “model divergence”,  as we would call it, Canadian vs. US, is the talk of Canada today, and that’s what makes today’s wrf-goofus output sting so much.

I really want to call President Obama on this and tell him about it; I know he would add it to his list of things that need to be fixed in our country.  Even if you have only a tinge of jingoism, you HAVE to be upset that the Canadians in their big little country, have a better weather forecasting model than we do!  I think I am going to have to lie down for awhile…calm down.

So, what is ahead in our weather?

Of course, we have to look at the Canadian model first to get the most reliable one to see if they have anything for us…  (hahahahah, sort of). I always do look at that one first, but I don’t brag about it.  The summary of last night’s Enviro Can run, out to 144 h:  they got nothin’ for us, just some cooler air over time.  Cirrus clouds will be floating by from time to time as they do on most days.  Did you know that Cirrus is a precipitating cloud?  Yep, little ice crystals are always settling out leaving those pretty trails.  Mt. Everest would know this…

Yesterday’s Cirrus clouds, sunrise to sunset

Feel another song coming on…. key lyric, “I don’t remember getting older…”

Hope you had some good log entries describing the varieties and species of Cirrus…  If you did, you’ll be getting closer to getting that Cloud Maven Junior Tee.

7:01 AM. Sunrise Cirrus.
5:32 PM. Sunset Cirrus, maybe with a contrail in there, dammitall.



Future weather has a lot of AZ rain, but its more uncertain using US models

Here’s is the latest model run from our USA WRF-GFS (aka, “goofus”, as the Europeans might call it, looking down their noses at our inferior weather predicting model compared with their “ECMWF” model as described (here) in the November 9th issue of Science.

It was an upsetting read, BTW. Seems the Euros use bigger, faster computers than we do, ones that they were able to afford by charging a lot of money to see the results.  Very bad.

In case you want the meat of that Science article:  “From the BEGINNING (this writer’s emphasis) ECMWF has been the world champ in medium range forecasting. Today ECMWF forecasts remain useful into the next week, out to 8.5 days.  That leaves the rest of the forecasting world, inculding the U. S. National Weather Service with its less powerful computer, in the dust by a day or more.”

What have our guys (includes women) been doing all these years?  (Just kidding, maybe.)

OK, onward with what we have to work with…..

This WRF-GFS run is just from last nights 11 PM global data crunch, the VERY latest as of this writing.  I picked it out from earlier runs to show because this run latest has a lot of rain in Arizona.   Namely, it was a subjective call to display a few snapshots from it.  Displaying the results of this run has nothing to do with scientific objectivity.  Enjoy; it might not be real rain that falls to the ground, only real in the model’s calculations.  Still, its great to see and think about.

Instead of showing the full size of these model outputs as I normally would do, I thought I would size them in proportion to their credibility based on the Science article.  We can’t see the better ECMWF-British model results unless we pay a lot of money, so this will have to do.  Unless you click on these below, you’ll have to use a microscope…

Valid for November 29th, 11 PM AST, only 264 hours away!
Valid for 11 AM AST, November 30th, 12 h later
Valid for 11 PM AST, November 30th–off and on rains now for TWENTY-FOUR hours!
Valid for 11 AM AST, December 1st. Still raining around here.
Valid for 11 PM AST, December 1st. Rain still falling in the 12 h ending at this time.












So, once again, our late November-early December storm has returned to the model fold. Its been coming and going.   For example, the 5 PM AST global model run had NO RAIN in AZ, so I didn’t want to show those results.

 But just ahead….this

In the nearer future…..  Seems the Environment Canada computer model, built around the SUPERIOR ECMWF model, has rain here in about 48 h from now resulting from  a tiny, weak low that ejects from the deep tropics right over us. Cool, though the air itself would be warmer and more moist than we usually see at this time of year in a rain situation (higher dewpoints).  Must regard this as a serious rain threat now.  Here’s a snapshot of that rain day from Enviro Can (see lower right panel for 12 h rain totals and areas covered–would fallen overnight tomorrow night into Wednesday morning.  The whole better than the US model runs is here.

Yesterday’s clouds

Another fabulous early winter day in Arizona.  Out of state license plates picking up in number.  Can’t blame ’em.   Here’s a sample of yesterday’s skies and another great sunset:

2:06 PM. Cumulus humilis and fractus (shred clouds).
5:23 PM. Small Cumulus and distant Cirrus add highlights to an Arizona sunset.
5:25 PM, looking south.