Monday, September 26, 2016


This discussion will cover the potential for severe weather tomorrow.  An anomalous retrograding cut-off low has moved to a position over the Baja spur this morning. Guaymas is reporting a -12@ 500mb and the low was analyzed at 5760 m. There is a moderate amount of moisture to the south and east of the low in the range of 25-35 mm. Plus, there is a TS (Roslyn) at 119W, 17N moving slowly to the NE which may add additional moisture over the next few days.

By tomorrow, the low begins to eject to the north with quite cool mid level temperatures over southern Arizona of around -12 to -15C.

Some low level moisture is present over southern Arizona with 850mb dewpoints in the 7-10 C range, which is fairly high for this time of year.
CAPE isn’t all that impressive, but the wind profile certainly is with a Great Plains supercell type of directional shear. Speed shear isn’t as good as was being forecast the other day but is still fair. The PBL becomes deeply mixed by early afternoon thus strong to severe storms are likely from around 1-4pm in and around the Tucson area. There is a risk of large hail. Storms are likely to exhibit rotation with the possibility of funnel clouds or maybe a brief tornado. The 12Z WRF GFS and WRF NAM model runs are quite similar thus there is some confidence in this forecast. The RR is missing at 12Z. I’ll check to see if there is any data available and make a run if that is the case.850_32.gif

Deep convection initiates around 20Z over the higher terrain of central Pima and Santa Cruz counties. By mid-afternoon, a broken line of strong to severe storms moves to the NW and impacts Pinal, eastern Pima, and western Cochise counties. Again, both models are quite similar with location and timing. The WRFGFS, below, is about an hour faster than the WRFNAM and has strong storms in the Tucson area around 3-4pm.

Thunderstorm winds are expected to be strong with isolated areas of severe winds.

The broken line of storms continues to move to the north and northeast into the early evening.  At this point, the model runs diverge with the WRFGFS (below) moving the dissipating storms to the east of Phoenix and continuing strong storms over Cochise county plus another area of storms moving off of the White Mountains.

The WRFNAM is a few hours slower and has storms dissipating over Arizona by 8-9pm.

The storms move fairly quickly so rainfall isn’t extreme.  However, some locations could see over an inch or more.  Mode consistency has been remarkable as all the runs from that last 18 hours have nearly identical solutions.  Saying that, there are some issues that have to be monitored.  All model runs allow full heating tomorrow and as past history has shown, the model does have issues with insufficient clouds so if we wake up to clouds tomorrow, there may be much less activity than predicted.  

Tuesday, September 13, 2016


Previous Day
A very active day for much of eastern Arizona as there were widespread thunderstorms, with some severe winds and large hail. Precipitation was heavy, but only for short periods as storm motion was quite rapid.

Both 6z runs were good as well as the 12z WRFNAM. The model runs were correct on predicting repeated development of storms throughout the afternoon as it rained 3 different times at my house. The 12Z WRFNAM had a bit too much precipitation, but location and timing were good. The 12Z WRFGFS had too much activity and was too far west (in and around Phoenix), as usual. It didn’t always use to be like this as a few summers ago, before the GFS upgrade, the WRFGFS (subjectively) performed as well or better than the WRFNAM.

Enough moisture has leaked into southern Arizona ahead of the west coast trough to support storms again today. Mid-level morning clouds are again present over central Arizona along with an area of high clouds over SW Arizona associated with a strong jet passing over the northern Baja mountains. The 6Z runs were mainly devoid of clouds while the 12Z NAM had some, but not enough. The RRx has been missing from ESRL the past few days.

A fairly strong area of low pressure is present at 850mb over southern NV is helping to advect moisture into the eastern part of the state.

There were no major IPW initialization errors by any of the 6 or 12Z runs. Again, clouds are an issue over central Arizona. However, this shouldn’t impact the forecast much as low-level moisture is insufficient. The model runs should perform well across eastern Arizona.

Day 1
Moisture continues to advect into southern Arizona ahead of a strong west coast trough. The model runs from yesterday had much drier air across the state for today, but the wet air was never far away in Mexico and it doesn’t take much to advect it into the state. Goes to show you how quickly the forecast can change. By late morning, there is significant low-level moisture over southeastern Arizona as dewpoint temperatures are above 10C.

The model runs disagree where the most unstable air is located. The WRFNAM has it over far SE Arizona while the WRFGFS (below) is most unstable over Santa Cruz/eastern Pima. A close look at the 15Z Suominet data indicates that the WRFGFS seems to have the best handle on the moisture as the WRFNAM is too wet over Cochise county. However, it is quite wet due south of Cochise county at Moctezuma (PW is 31mm). In fact, all of northern Sonora has IPW values in the low to mid 30mm range. WRFGFS below.

Are we going to have a repeat of yesterday? Early afternoon CAPE around Tucson is around 900 J/Kg and the PBL is mixed deeply. What is different from yesterday is that the directional shear is somewhat less by early afternoon. Speed shear in the WRFGFS (below) is a bit better with mid-level winds about 5-15 knots stronger, but the WRFNAM winds are similar to yesterday.  So, it’s looking favorable again for fast moving and organized convection from Tucson and to the east, Hail is again a possibility.

The 2 12Z runs are nearly identical in developing deep convection during the early afternoon N/S of Tucson as well as over NW Arizona where there was also sufficient CAPE.

Big storms continue to develop and move to the NE along this line throughout the afternoon.  Quite similar to yesterday but offset to the east by about 50-100 miles.

In some ways, today looks even better for severe weather as there is more upper support as a strong jet associated with the trough moves towards the state. Upper difluence and divergence are present over much of the state during the afternoon.

Wet air continues over the eastern part of the state into the evening.

As they did last night, storms continue into the evening hours, east of Tucson.

Note that storms redevelop over the same locations throughout the day into the evening.  Some areas could get quite a bit of rain, like the mountains around Tucson.

Day 2
Some moisture continues to hang around eastern Arizona but should be insufficient to support much deep convection.

Monday, September 12, 2016


12Z WRFGFS Update
The WRFGFS initialized OK with only minimal IP errors. Like the other runs, it did not have enough clouds and at 17Z, scattered to broken mid level clouds continue over central Arizona. IPW was initialized well.

I wanted to mention the 12Z WRFGFS as it's much more active around Phoenix this afternoon as forecast CAPE is much higher than the WRFNAM.

Strong storms develop over much of northern Maricopa county during the mid to late afternoon. My two big concerns with this forecast are the clouds that continue to be present that may restrict heating and the over-prediction of deep convection that has been typical of the WRFGFS this summer. So, I find the below unlikely and the coverage/intensity should be less and perhaps somewhat later in the day.

I took a quick look at predicted vs observed IPW at 16:45Z and the WRFGFS is definitely too wet west of Phoenix and even the WRFNAM is 2-3mm too wet over much of SW and southcentral Arizona. I've been unable to initialize off of the RR this morning and may go try again in order to see what it says. At this point, the WRFNAM is likely to be too early with deep convection and perhaps too widespread too.

Previous Day
Yesterday’s WRF runs produced deep convection mainly over far SE Arizona which was close to observed. Precipitation developed northeast of Phoenix early this morning which was not forecast by any of the model runs.

An elevated unstable layer is present on the morning sounding with over 1000 J/Kg of CAPE.  This instability has already triggered deep convection to the NE of Phoenix, which was not handled well by the overnight model runs. The 12z NAM is also missing this convection and the associated cloud cover.

The NAM had a moderate wet bias in Sonora this morning probably due to the bad sounding data there. Other than that, only minor errors. I’m not waiting for the 12z WRFGFS as I need to get this discussion finished quickly due to the early start of the activity, so I’ll also use the WRFGFS from 6Z. Its IPW was initialized well. Overall, the model initializations are fair and confidence is medium.  

Day 1
A weak shortwave trough is passing through Arizona today with an area of cooler mid level air (below -10C) that can be seen over eastern Arizona by late morning.

Low-level moisture has increased significantly over the eastern ½ of the state with late morning 850mb dewpoint temperatures in the 11 to 13C range.  

CAPE is forecast to be over 1500 J/Kg in eastern Pima county with higher amounts off to the west. With CAPE this high and cool air aloft, hail is a possibility with storms today.

The Tucson forecast skew-t indicates over 1300 J/Kg of CAPE and a well-mixed PBL by around mid-day. This should lead to early afternoon deep convection. The wind profile has some directional shear thus some organization is possible, and with CAPE this high, some very strong to severe storms are possible.

It looks like eastern Pima/western Cochise become very active by early afternoon (12z WRFNAM below). The 6Z WRFGFS is similar except for an hour or two later.  In fact, all the model runs over the past 24 hours have been very similar thus confidence is high.

Storms move off to the NE quite fast as mid level flow is around 20-30 knots.  Additional storms form on the edges of the outflows from the initial activity.

Besides hail, isolated very strong winds are a possibility with the storms.

By late afternoon, the PBL is very deeply mixed at Phoenix, but CAPE is somewhat lacking with about 500 J/Kg (CAPE is not calculated correctly on the skew t below). If storms can form, they will have a high risk of very strong winds.

The WRFNAM does have isolated storms around Pinal and eastern Maricopa late this afternoon and early evening. Also, note the redevelopment of some activity over eastern Pima county.  Both the 6z WRFGFS and 12Z WRFNAM (below) are similar.

Day 2
Drier air returns with the only moisture of significance over SE Arizona.

Only an isolated storm or two during the late afternoon.

Tuesday, September 6, 2016

20160906 #2

This is an update to the earlier discussion, written around 2Z.

Previous Forecast
So far, the 12Z WRFGFS and 12/18Z WRFNAM are close to the actual position of Newton.  Satellite imagery shows the center of Newton over the Gulf of California, just to the north of Loreto while the runs have it about 50 miles or so to the southeast. The various WRFRR runs are too slow and too far to the southwest. A few areas of deep convection have redeveloped, but no eye is visible in the satellite imagery. The 21Z NHC advisory was ‘generous’, estimating the peak sustained winds at 65 knots. This is close to the model runs but maybe slightly on the low side, thus the runs may be slightly too strong. Model IR clouds from the WRFGFS/NAM are very similar to the observed satellite clouds.

Only the 0Z NAM will be used for this discussion. The initial location of hurricane Newton looks to be accurate while the wind speeds may be a bit high as surface winds are as high as 55 knots and 700mb are as high as 80 knots. Typically, there are no soundings at 0Z from Mexico.  However, one brave individual sent up a balloon at Guaymas. Fortunately, the surface winds in Guaymas were not strong yet, but at 925 mb winds were already 35 knots. I’m sure that this data point is going to be of great help to the model runs at 0Z. The 700mb trough continues to be present over southern California but looks quite weak. Maybe with weaker flow Newton will take a slightly more west path? We shall see what the WRFNAM says below.

(barely) Hurricane Newton remains over the GofC for about 8 hours before crossing Isla Tiburon and coming onshore. This is a major difference from the 12Z WRFNAM and GFS runs as they already had it inland about 75 miles to the northwest. It is also very similar to the location of the 15Z WRFRR. The only difference is that the WRFRR is quite a bit stronger.  

This westward shift in the track means that the circulation will 1) remain over the warm and low friction water longer and 2) avoid the high friction, higher terrain in eastern Sonora. So, a stronger storm that remains intact longer. By 18Z, it’s still not across the border and in a nearly identical position as the 15Z WRFRR. The main difference is the 0Z WRFNAM is stronger!

During the late evening into the early hours, light rain develops over much of the SE part of the state.

Heavy rain begins over central Pima county by late morning as the center of circulation approaches. This is quite a shift from earlier and I’d like to see more 0Z runs before fully buying into it. Check for yourself! Certainly, this increases the risk for very strong winds and very heavy rain in the Tucson area.

After moving quite quickly, Newton slows down over far NW Sonora and finally crosses the border into central Pima county during mid afternoon (must be backed up at the Port of Entry at Lukeville due to Labor day)

As depicted by the WRFRR, the center passes just to the west of Tucson with very strong winds for much of eastern Pima county during the late afternoon with gusts perhaps as high as 50mph.

The only positive with this run is that the most intense precipitation remains to the west of Tucson.  However, this band of 6+ inches could certainly be another 25-50 miles to the east.

Strongest winds in the Tucson area are around late afternoon with some areas with sustained winds as high as 40 knots with higher gusts maybe around 50 knots.

It is currently running and a quick look at the initialization shows it did well with the placement of Newton. Just to throw more uncertainty into the mix, it rapidly moves Newton across the GofC and comes onshore around 9Z about 75 miles to the SE of the WRFNAM.

This track causes Newton to  spin down over the higher terrain of central and eastern Sonora.  So, in spite of such a short time to the event, the models continue to have a wide spread in their solutions with no clear favorites.

Heavy precipitation begins over far southern Arizona around 12-14Z.

The heaviest rain remains just south of the border. So, the message continues to be the same as this morning which is some areas of southern Arizona will get a lot of rain: up to 6 inches.  There is a possibility that areas from central Pima county all the way to western Cochise county could get tropical storm force winds. When and where? We’ll have to “wait and see”.