Sunday, July 29, 2007

Spectacular wall cloud, heavy rain, and wet downburst move across north Tucson (07-28-07)

Heavy storms with flooding rains and damaging downbursts hit Tucson again early yesterday afternoon (Saturday 07-28-07).

On the north-central side of town, where we live, a storm cell moved southward from the west end of the Catalina Mountains. This cell had a spectacular wall cloud, see:
(NOTE: file size = 91.3 mb)

and Photos 1, 2, and 3. I took the first two from the house looking north at around 12:50 pm MST and tried to get the entire wall cloud/mesocyclone in the view. Photo 3 shows the east side of the wall cloud from the UofA's "Arizona Webcam." (This cam looks due north; the vantage point is a window in the Gould-Simpson building.)

A massive wall of rain developed to the rear and to the east side of the wall cloud. An associated wet downburst took down limbs and entire trees in a region along Ft. Lowell stretching 2 to 4 miles east-southeast from our house. I was out at sunrise taking some photos of the damage. Two large mesquite trees at Ft. Lowell and Dodge fell, taking out the power lines at intersection - police were still directing traffic through that intersection at 5 pm last evening. Photos 4, 5, and 6 show photos of downed trees in the area mentioned above.

Rains of 2 to 3 inches were measured over much of the Tucson metro region causing serious flooding, high water rescues, and the usual problems caused by stupid motorists. The news story at:

has a nice slide show focused on the flooding and water damages, plus a helicopter rescue.

Thursday, July 26, 2007

Heavy rains and damanging winds hit Tucson metro area

Heavy and severe thunderstorms hit Tucson on Monday and Tuesday, the 23rd and 24th of July 2007.

Have returned after 3 days in Boise, Idaho. Saw the indications of serious urban flooding that occurred last couple of days as we drove from airport to house. Quite a few limbs and trees down here and there all the way from airport to our neighborhood. Neighbors seemed to think that winds were worse with the Tuesday storm. Monday storm led to 4 severe wind reports and Tuesday's storm produced 3 severe reports at SPC - including a gust to 63 mphat DM AFB.

When I walked this morning, I noted many limbs and a couple of trees down - but of course, I don't know if the damage occurred Monday or Tuesday. Workers were sawing up large sycamore limbs that were downed at Campbell and River (my guess would be on Tuesday).

Newspaper stories indicate that there was fairly widespread damage acrossparts of the metro area on both days - but the severity seems to have beengreatest on Tuesday. Stories for anyone interested are at:


The second story has a nice slide show of storm events from both Monday and Tuesday. A good amount of rain accumulated in our gauge while we were gone - 1.94" Guess I should leave more often, or perhaps it's the combined effect of Mullen and Maddox both out of town? I took a quick look at the soundings from Sunday morning (the 22nd) through this morning (the 26th). Given that the atmosphere has been very moist and that winds have been very light through most of period, I find it distressing that the evening soundings continue to exhibit a drybias, see:

With the moist atmosphere, there have been numerous sounding problems due to wetting of the Sippican thermistors. I note bad layers in the following: 22/12Z; 23/12Z; 24/00Z; 24/12Z; and 25/12Z; with 25/00Z suspect. It's hard to have a very high degree of confidence in these "new and improved" upper-air data!

Today's morning sounding continues to show light and variable winds in much of troposphere, except that upper-levels have some flow and also indicate substantial difluence over SE AZ - so large-scale upward motion may help things along a bit today.

Monday, July 23, 2007

Off to Idaho . . .

I will be away until Thursday - we are flying to Boise ID in
a couple of hours. Will have camera in hand, since several weak
short waves, and moisture from the SW, are moving up the
back of the 500 mb anticyclone toward ID, i.e., a "Ring of Fire"

Coverage of the Mexican supercell is at end of the third post down.

Sunday, July 22, 2007

More severe storms on 21 July 2007

Prepared at 8 PM MST Sunday July 22, 2007

Although Friday the 21st of July was a "down" day in the local Tucson metro region, intense storms returned on Saturday July 21st late in the evening (good Friday call by Pat Holbrook!). Severe storms with heavy rains and downbursts moved in from the E to NE, producing very heavy rains of more than an inch and downburst winds over much of the metro area. Except of course here at the house, where we had mostly winds and lightning and thunder followed by very light rains after midnight.

A measured gust of 56 mph occured at DM AFB, and TUS reported a gust in the 40's. Most of the downburst damage occured on the east and south sides of the metro area. I drove down Grant and Kolb this morning and observed that there were a number of trees down in a large Target parking lot at Kolb and Grant, with sctattered trees and limbs down at least to south of Kolb and 22nd - about four miles- DM is about another 2 miles to the south, so I suspect that there was more wind damage south of where I drove.

Newspaper reports indicate that power poles were downed in the south metro area just north of the Tucson airport and that power lines were downed in the area around Alvernon and I-10, a few miles ENE of the airport.

I was able to get a photo of this storm as it was moving into the metro area. I took this shot looking about due east from the front of my house at about 7:15 pm MST.

Storm reports for 19 July 2007

Prepared 7:00 pm MST 22 July 2007

As things turned out, the 19th was a very active storm day in SE AZ - refer to previous post.

There are many severe events documented under storm reports at the SPC page.

Here at the house, the event was mostly wind and dust followed by some nighttime trailing rain.
There was a very bad dust storm here right before dark. The Catalina Mountains were totally obscured when the gust front blew in - I don't think that I've observed that before. At atmo the roof-top anemometer registered gusts to ~ 75 mph. Unfortunately, there was not enough light for me to get photos.

In Phoenix, Skyharbor airport reported gusts to 55 mph. There was a fatality due to power lines downed by the outflow - BUT, although I have seen numerous newspaper accounts of this death that occurred after the storms moved through, I don't find any severe storm reports at SPC.

However, the most amazing storm report for the 19th was copied to me by NWS forecaster Jeff Davis. A very severe, westward moving supercell ocurred in Mexico, just south of the border.

It hit the town of Cananea (see map), and produced, at a minimum, a spectacular hailstorm. The Tucson NWS Doppler radar documented an impressive mesocyclone with this storm. Since the storm was moving toward the WSW, the mesocyclone was located on the right flank of the storm, i.e., the north flank of the storm. The four-level reflectivity product and the four-level
Doppler velocities indicate the intense nature of this storm a bit after 00 UTC (5 pm MST).

Thursday, July 19, 2007

Big severe storm day today in AZ?

Prepared at 8:00 am MST, 19 July 2007

The summer thunderstorm season in southern Arizona has been very slow and sparse to date - except in the far south borderlands. The last few days have been characterized by little CAPE at low elevations, with storms mostly clinging to the mountains in the Tucson area. The low-level moisture levels have been adequate to support some storms, and it has generally come from weak pushes of higher dewpoint air into southern AZ, some nice outflows from storm complexes to the east and south. these have combined with abundant recyling to keep the IPW values in the 30 to 40 mm range. Yesterday, with stronger steering-level winds finally, a storm cell moved off the western end of the Catalina Mountains and produced a wet downburst in Oro Valley. Lots of thunder and wind and dust here, and then a brief burst of rain that left 0.05" in the gauge. We still await a serious push of deep sub-tropical moisture into the state.

But, lots of interesting things are finally happening - a large, Sonoran MCS decayed over the central GoC this am and it appears to have started what will finally be a serious push of very moist subtropical air into southern AZ Doppler radar VAD winds at Yuma show nice SSE winds up to 4,000 ft. The upper-level inverted trough over northern Mexico appears will affect southern AZ late pm - early night today with nice difluence aloft. By tomorrow evening it will be right overhead, with the difluence weakened and shifted to NW AZ. During this period there is upper-level ridging over a subtropical circulation that is located south of the mouth of GoC.

This system is predicted to strengthen considerably by the NAM and may becomea TS? Regardless, it is perfectly positioned to intensify the surge of moisture into AZ for the next 60 hours!

The NAM suggests a not very favorable upper-level shear profile this afternoon with easterlies of nearly equal strength from 500 to 200 mb, but everything else looks like it's coming together, and the last two days the model predicted winds have been erratic over this corner of state.

So which day will be the big day? I'm going to go with:

Likely big severe tstm day today, and then two days of good, heavy rain storms.

Sunday, July 15, 2007

Slow start to summer thunderstorm season around Tucson

Prepared at 8:30 am MST 15 July 2007

Yet another day of no storms yesterday in the lower elevations of southeast/south-central AZ. Dry BL air again advected in from the NW, but not as severely as the last couple of days.Quite a slow start to the storm season around here - although Art Douglas has been reporting on the abundant rains to the south along the borderlands.

Misc. points of interest this morning.

Precipitable water and dewpoints are down relative to 24 hours ago.

It's hard to find possible CAPE in the morning sounding, which was about 3 mm on the dry side vs GPS. But there are mid-level buildups and light showers drifting by, so there is a bit of elevated CAPE around.

Winds aloft are very light indicating heavy rain potential, if storms do develop.

Surface pressures have fallen about 3 mb, relative to 24 hours ago over the southern half of state.

The NWS POPs for today are the lowest (10%) of the last several days. However, one of the showers passed over our house between 6 and 7 am MST (I know the timing for certain since it caught me in middle of a morning walk), and actually produced 0.02" of rain in the gauge. Thus, today is the 4th day in July that there's been measurable precip here at the house (grand total to date for July is only 0.26").

The ESE-WNW cloud band that has been south of us the last couple of days is now directly overhead. It appears to be co-located with a weak 300 mb trough/deformation zone. Heavy rains have been associated with this feature to our south yesterday and day before.

The NAM 500 mb analysis at 1200 UTC this morning says that the trough over central TX curls back north over northern MX as a weak inverted trough that extends north into the cloud band mentioned above. I can't see any indication of this feature, but it's again something that the model says is in the great data void to our south. The models do very interesting things with the Texas trough during the coming five days!

Certainly hope that there's some persistence associated with the cloud band and that the heavy storms shift northward with it today.

Monday, July 09, 2007

More on sounding analysis post-NWS RRS

On Saturday morning Jon wrote:

"The emphasis on the AFD and SPC forecasts, however, was whether coverage of the *severe* potential would be greater on Fri compared to Thu. Sure, it was more convectively active on Friday over Thursday. But, it is tough to say, at this point, on the coverage of severe thunderstorms based on no mention of such in media, print or LSRs from Friday. Have you seen any bonified severe reports from Friday?"

Actually, while there was lots of severe storm verbiage in the FD discussion the bottom line, as I interpreted it, had to do with general coverage of storms and POPs, which were left at previous levels. On Friday storms formed early; rained and anviled out by late afternoon; and there were no severe storm events, even though coverage of storms was greater. This often happens in the Tucson area, but it was not clear to me from the morning sounding that Friday was going to be an early storm day. Which leads to the issue of importance, as Jon stated:

"I agree--sounding assessment is tricky with the known problems. Thank goodness for PWAT sounders across the region."

I think that "tricky" may be an understatement. Consider this morning for example: The TWC sounding indicates IPW of 41mm with CAPE of well over 1000. These are quite high values for here and may (?) indicate much more storm potential today vs yesterday. The dilemma is that it appears that IPW is 5 to 6 mm too high wrt GPS IPW. The question is how does one know how to modify the "wet" Sippican data this morning? Where exactly is the phantom moisture that needs to be subjectively removed? How will this wet sounding affect the morning model runs?

Last evening the situation was just the opposite. The high resolution WRF model forecast indicated moderate CAPE at TWC at 0000 UTC, but topped by a layer of considerable CIN. However, the TWC sounding for that time indicated no CAPE whatsoever - but comparison of the IPWs indicated that last evening's Sippican data were about 7 mm too dry. Again the dilemma is: how do we know what the structure of the real atmosphere was?

I guess that I must admit that I am puzzled as to why there is so little reaction by any or all who are in atmospheric research and/or weather forecasting who use upper-air data or derived products. As for me, I am very angry and frustrated by the damage that's been done by the RRS program to the quality of the US upper-air sounding data!

Friday, July 06, 2007

Sounding Analyses When Data Are From Unreliable RRS Sippican Sondes

Prepared 4:00 pm MST on July 6, 2007

I think that the Tucson NWS upper-air soundings from yesterday afternoon and from this morning illustrate several problems that forecasters, and other users, will have with the data from the unreliable Sippican sondes being used operationally by the NWS RRS program.

First, let's examine the 5 pm MST (0000 UTC) Tucson sounding from yesterday - See Fig. 1. This sounding was taken on a very hot afternoon, about two hours before severe thunderstorms moved into the Tucson metro area. Things of importance:

  • The well-mixed adiabatic boundary layer (BL) reaches to 600 mb in the temperature profile. A layer of steep lapse rate is just above, reaching from 600 to 450 mb, i.e., an excellent thermal situation for high-based convection. Winds in the thermal BL are light and variable, meaning that advective changes were minimal, until the storms moved into Tucson.

  • The moisture profile from the 1st data point above the surface to 700 mb appears to be too dry in the extreme. There is no physically realistic BL captured in the Td profile. There should either be a BL well-mixed in moisture or, if shear at 600 mb was mixing dry air downward into the BL, decreasing mixing ratio with height would be expected.

  • Computed CAPE for this severe thunderstorm proximity sounding was zero.

The analysis and forecasting dilemma is obvious. Since the Sippican sondes tend to measure Td (RH) too low, we have no idea what the actual BL moisture structure was
prior to the thunderstorms and macrobursts.

Second, let's look at this morning's sounding - See Fig. 2. Things of importance:

  • The moisture profile bears no resemblance to that observed the evening before. Now we must ask whether we should accept this profile as accurate.

  • Note that if this profile becomes well-mixed (in T and q) during daytime heating to 550 mb, there would be at most a tiny sliver of CAPE. There is advection from the SE to consider however.

  • The GPS IPW at the time of the sounding was about 6 mm greater than that of the sounding. This calls into question, again, the accuracy of the Td profile.

Given the unreliable character of the Sippican moisture measurements, it is not clear how the forecaster, or other users, can actually diagnose what is the real potential for convective storms. The morning forecast discussion from the Tucson NWS Office, based to some degree on this sounding, concluded that:


Interestingly, here at the house, there have been thunderstorms and occasional showers since about 2 pm. Updraft bases appear to be at about 600 mb and an intense downburst appears to be in progress a couple of miles to the east. I suspect that the morning sounding was dry and that there has also been some additional increase due to advection. This combines to make this afternoon much more convectively active than yesterday.

Damaging macroburst(s) hit Tucson metro area last evening

Prepared at 10:30 am MST Friday July 6, 2007

This post will document the thunderstorm and macroburst(s) that moved across Tucson last evening between about 7 pm and 8:30 pm MST (0200 to 0330 UTC). I will follow this with a second post examining the RRS Sippican soundings from the Tucson NWS site. The second post will examine how difficult it is to diagnose the storm potential and local environment with unreliable sounding data.

There is one severe storm report posted for Tucson at the SPC storm reports page this morning:

This area of town is just to the west-northwest of our house. On a walk this morning, I observed and photographed trees and moderate limbs down, and everyone I encountered talked about the vicious wind storm last evening. When I drove out to have some car repairs done I observed numerous moderate limbs and trees down; several intersections had damage to the light poles and police were directing traffic. I only observed one building that had roof damage (at the intersection of Campbell and Grant). This was in north central Tucson and not covered in above report. I am told that the atmo roof-top anemometer measured a gust to 80 mph; this is near the downtown and just to the northeast of the NWS Office. The newspaper reports four different power outage areas scattered over a large part of the metro area from 0200 UTC until 0330 UTC. Obviously this was a significant severe thunderstorm event!

The following photos document the storm and a bit of the damage:

Photo 1 - taken from a campus web cam at 7 pm (0200 UTC,) looking north toward the Catalina Mountains, shows the very high-based updraft over the mountains. There is a slight bit of blowing dust visible, but this was coming from a different cell off to the east.

Photo 2 - shows the updraft from our house, at 0230 UTC, over the west end of the Catalina Mountains. Note that a precipitation shaft has developed and is reaching the ground. At this time the storm was characterized by frequent cloud-to ground and in-cloud lightning.

Photo 3 - shows the precipitation moving off the west end of the mountains toward the south. This photo was taken 3 minutes after Fig. 2.

Photo 4 - shows a large salt cedar limb down. Figures 4, 5 and 6 were taken within two blocks of our house to the northwest, the east, and the south.

Photo 5 - shows large gum tree limbs down in front of a house.

Photo 6 - shows a small paloverde tree that was uprooted.

When the macroburst struck at our house the period of strong gusts lasted about 20 minutes. I estimate that we had maximum gusts of 40 to 60 mph during this period. The strong winds were initially from the north but, after about 15 minutes, switched to the south. I saw some trees down from south winds, but most were downed by the initial north winds. We had a trace of rainfall accompanied by blowing mud.

All in all, a quite a spectacular start to the summer thunderstorm season in the Tucson area!

Thursday, July 05, 2007

A very complicated weather situation

Discussion prepared 1530 UTC 5 July 2007

Last evening we were able to see, from Tucson, widespread anvil and even some convective towers in the distance from north to east to southeast. There were some impressive storms over, and near, far east and southeast Arizona and these produced outflows that have affected all of southeastern Arizona during the night - increasing dewpoints and precipitable water, while keeping low temps very high (Tucson's "low" this morning was 87F). Douglas reported a severe thunderstorm wind gust as several outflows passed there. These storms were associated with a weak 500 mb shortwave through that has come over the top of the western US ridge and moved south-southwestward, to the west of the amazingly persistent cyclone that has been anchored over the southern Plains for several weeks. This cyclone, which again this morning seems to have taken on warm-core tropical characteristics, is forecast by the models as (and appears to be) breaking apart with pieces going several directions. The part that moves into northern Mexico does not really swing across southern Arizona with middle-level winds forecast to remain northeasterly though 48 to 72 hours. See the following 500 mb analysis and forecasts from the NAM:

The morning upper-air soundings at Tucson and Phoenix have extremely deep residual boundary layers form yesterday. Tucson's old BL extends to 450 mb this morning! So, even if low-level moisture continues to increase, there will only be small CAPE available at low-elevations for very high-based convection this afternoon. This is a perfect setup for severe downbursts and strong outflows from more intense mountain storms that will try to move into the deserts. The Phoenix sounding is similar.

So, a key question for tomorrow appears to be: Will low-level moisture continue to increase slowly due to evaporation and recycling OR will deeper GoC moisture push in below 850 mb, giving the precipitable water a real kick from south of the border? Yuma appears to be experiencing at least a moderate southesterly push of moisture this morning, similar to the past several days. The KYUX Doppler winds - see the VAD display available at -

indicates that fairly strong SSE winds extend up to about 3000 feet at this time.

My feeling at this time is that the piece of the Texas cyclone that pushes off southwestward across the southern GoC will be strong enough to trigger a significant low-level surge of Gulf Moisture northward into Arizona after midnight tonight. Thus, I anticipate that the summer thunderstorm season is off and running, and that tomorrow will bring increased CAPE and more extensive storm activity across lower elevations of south-central and southeast Arizona.

Monday, July 02, 2007

Additional Comments on the NWS RRS program

My webpage comments on the NWS RRS program have been posted on the Roger Pielke Climate Science weblog

There have been several comments and questions posted at this site as follow:

Question: How many other sites have changed sondes and rendered breaks in seemingly good time series?

The site changes have been documented according to NWS procedures. The problem is that the announcements are not easily seen by many in the data user and research communities. I hope that NWS systems operations people will find ways to improve the dissemination of information about changes in observing systems to all external users.

The new sondes, as per Fig. 1 map shown at, are now being used at a significant percentage of the NWS upper-air sites. Thus, because of the unreliable temperature and humidty sensors, more noise is being introduced into the long-term time series for, eventually, the entire US upper-air network.

I mistakenly did not indicate on Fig.1 that KOUN is now using the Sippican sondes. As of today, KSLE began using the new sonde.

Question: The purpose of the move was to? What do radiosondes measure and how is it used by weather forecasters?

The purpose of the Tucson move was apparently to save money, since continuing the observations at the airport required a service contract. That contract was terminated. The move will eventually save the NWS money, even though the new sondes cost twice as much as the ones being discontinued.

Simply put, radiosondes measure temperature, humidty and winds from the surface into the stratosphere. These data feed into the numerical prediction models used by all weather forecasters. Unreliable observations will affect the performance of these models. See comment # 5 at the Pielke site also.

Question: Who is responsible for this program?

Those responsible for the RRS program and the operational implementation of unreliable sondes are apparently the high-level managers of the NWS Office of Operational Systems (OOS). The Director is John McNulty

The OOS branch responsible for upper-air operations is the Field Systems Operations Center. The Director of this branch is John Van Kuren