Thursday, September 14, 2006

Another tropical storm of interest

I have been away on a trip to McCall, Idaho. Many wildfires burning up there with little ventilation, and skies and the air were filled with smoke on several days. Looks like things may be improving with the strong systems now affecting the northwest. I can report that all large trout in the streams around McCall easily avoided me.

I returned to find a bit of rain in the gauge, 0.04", from the unsettled weather of the last several days around Tucson.

There is another Tropical Storm, Lane, in the east Pacific and NHC forecasts it to become a hurricane and to take a track similar to John. This brings it up to the tip of Baja in about five days. However, all forecasts related to John were very poor and the GFS eventually did best with its track. The GFS is forecasting the storm northward, so we'll see how it does with this one.

Wednesday, September 06, 2006

Another dilemma re subtropics and westerlies

I have taken a quick look at some of the charts and model progs this morning.

First, the morning Tucson sounding has dried some and indicates that CAPE will likely be not very great today and that what there is should develop only over the mountains. So another quiet day is likely locally here in the Tucson metro area. The sounding (Fig. 1) shows a nice example of why the LI computed at 500 mb out here can be very deceptive!

Second, an infrared satellite image at 1230 UTC this morning (Fig. 2) shows a large area of storms and disturbed weather over and south of the southern end of the GoC. The question is what is this and will it affect our weather up here in Arizona? The only sounding available this morning near to this area is Mazatlan, which indicates a very moist and unstable air mass with low level-easterly winds. The nam and gfs models forecast a strenthening cyclone at 850 mb down in this region; however, as per usual this summer, the models have very different details in their forecasts.

I'll illustrate this with the 850 mb 42-h forecasts from the nam and gfs. The nam (Fig. 3 - left) indicates a low center SW of Baja with strong southerly winds intruding far north into northwestern Arizona, i.e., a significant surge of moisture into the lower Colorado River Basin. However, the gfs (Fig. 4 - right) predicts a cyclone over the southern GoC that is totally separated from the circulation fields further north over northern Sonora and the U.S. Southwest; i.e., no hint of a moisture surge into the U.S.

The weather in Arizona the nextcouple of days will, of course, depend upon which model's forecast is closest to what happens in the real atmosphere. Both models forecast several weak short-waves at 500 mb moving across Arziona and interacting either with increased moisture (nam with significant storms and rainfall) or residual moisture (gfs with isolated and mountain storms). I guess that I'd tend toward the gfs forecasts, given the two models' performance during Hurricane John.

Sunday, September 03, 2006

Status of Hurricane John and the back-door front

Pat Holbrook said at 2:12 pm on Thursday August 31st

Looking at satellite trends and station data, the lowest pressures are on the east side of the gulf of california...still think this one will be heading along the same direction as it is now and enter into the southern gulf and possibly up to Los Mochis area before landfall.

Pat called this one very well! Current visible loops seem to indicate that what's left of Hurricane John is making landfall now south of Guaymas. Obviously the GFDL model was far superior to the other models, and the consensus track has been consistently wrong. It is still possible that an orphan low-level circulation will jump Baja and drift out into the Pacific. But it appears that the main moisture and mid-level vorticity from John is heading across far southeast Arizona and into New Mexico.

CORRECTION TO ABOVE at 3:30 pm Labor Day - the circulation that appeared to go ashore in imagery yesterday wasn't the remnants of John. Imagery today indictes that John continues to be nearly stationary over the northern Gulf of California.

I've emphasized for several blog posts that when a TS comes directly into the Gulf the situation is much more complex than when a TS moves northwestward across the region southwest of Baja tip. This has certainly proved to be the case with John.

As for the backdoor front, the GFS model did very well (beginning last Tuesday or so) in forecasting this feature to broach the divide and come into southern Arizona. We were in the southeastern Arizona grasslands near Sonoita Friday afternoon and Saturday. By sunrise on Saturday (yesterday) the front had passed and we experienced a cool, suppressed day down there with brisk east winds. A careful look at the surface observations (and 24-h changes - an essential aspect of surface analysis in the Southwest during summer!) indicates that the front passed Tucson around 5 to 6 am yesterday morning; Casa Grande around 10 to 11 am; and Skyharbor in Phoenix between noon and 1 pm. High temperaturs yesterday east of Phoenix to Sells line were 5 to 10 F cooler than on Friday. The front appeared to stall from around Sells northward to west of Phoenix, providing a nice convergence zone that helped the development of the large MCS that occurred over south-central Arizona late yesterday afternoon and evening. This system produced a huge outflow of cool air that pushed southwest past Yuma and which came uphill across Tucson around 10:30 pm, with a shift to north-northwest winds and a jump in the dewpoint temperatures.

Storms today and tomorrow will occur where there's enough low-level heat, combined with moisture from John, to produce CAPE - mid-level temps are cooler over the north and west parts of the state, but deeper moisture is present over the south-central and southeast regions. Depending on exactly how remnants of John wander around, the models (which still have vastly different forecast scenarios) indicate that a dry-out may occur by Tuesday. Interestingly, the NAM spins another tropical system up southwest of Baja by 84 hours.

Friday, September 01, 2006

Summer rainfall

I have been keeping daily rainfall records here at the house starting July 1, 1999. Since the summer season (June, July, and August) has ended, I thought I'd provide a summary for 2006 and compare to recent years.

My rain gauge is at approximately 32.27 N and 110.93 W, with elevation of 2369 ft MSL.

Rainfall here in 2006 was: June = 0.50" July = 4.88" August = 1.09" Total = 6.47"

There was measurable rainfall on 27 of 92 days (that infamous 30% POP we see so often in the summer forecasts).

Heavy storms on just 4 days accounted for 65% of the summer total. Summer rainfall, rank- ordered, for last eight years at the house:

1999 - 9.41" (no records for June)
2006 - 6.47"
2000 - 6.30"
2005 - 5.43"
2003 - 4.69"
2002 - 4.15"
2001 - 2.70"
2004 - 2.44"

So, a nice wet summer here - not at all close to wettest, but much better than driest year.

How complex can things get?

I have been tied up with other work this morning and have just now gotten around to looking at this morning's weather charts and model progs out through 48 hours.

This is one of the most complex and difficult situations that I can remember seeing here in the Southwest.

I have taken a very careful look at the NAM and GFS models out through 48-h; I have little confidence in any of the models or their forecasts beyond 48-h. There are strong and competing features in battle with each other during the next 48-h, as Hurricane John moves north at the same time as a very large upper-air blocking pattern sets up over the north-central U.S. While these things happen, a strong Canadian cold front moves rapidly southward down the front range of the Rockies, bringing much rain and moisture in its wake. This front back-doors into the Southwest and brings low-level winds around to the east over the eastern half of Arizona. Finally, low-level moisture surges very rapidly northward after after 24 to 30 hours. The only NWS offices discussing in their FDs the role of the cold front in determining the Holiday weekend weather are ELP and ABQ, but it seems fairly clear that the front is going to affect the evolution of events over a large part of Arizona also.

Brief summary of the models progs at 12, 30, and 48 hours follows:

12 - hours: Both models indicate light northerly winds at 850 and 700 mb. At 500 mb both models have light westerly flow over southern Arizona, with a weak anticylone south of the border. Upper-level winds are westerly, are strong and appear confluent over southern Arizona.

30 - hours: Winds at 850 mb and 700 mb continue northerly over much of southern Arizona. The cold front has pushed into southern New Mexico and is further west in NAM with stronger easterly winds. The NAM has northerly winds south to about Hermosilla; whereas, the GFS indicates the expected Gulf Surge has moved into the lower Colorado River Basin. Both models indicate moisture advection at 700 mb into southeastern Arizona from the east. At 700 to 500 mb there is a pronounced inverted trough swinging around Hurrican John and approaching the New Mexico bootheel. Increasing heights at 700 mb over west Texas act to increase the gradient and southeasterly winds to the northeast of John over much of northwestern Mexico. Upper-level winds continue strong from west but may be slightly difluent over southern Arizona. The NAM is moving John up the east side of Baja and the GFS is moving John up along the west coast of Baja.

48 - hours: At this time the two models have diverged markedly in their forecasts. The NAM has begun moving John westward, probably because of the large block to north and northeast. Moisture has surged rapidly northward as far as southern Nevada. The backdoor front has produced strong easterly winds over southeastern Arizona, but signficant moist advection continues at 700 mb from the cool side of the front. There is a weak 500 mb trough and vort max over southeastern Arizona, but 500 mb winds are strong northerly over most of the state. Upper-level flow appears divergent.

In contrast, the GFS continues moving John northward; the surge has moved far north and into the southern Great Basin. Easterly winds have increased over southeastern Arizona in lower-levels. GFS also has weak trough and vort max at 500 mb over southeastern Arizona. Upper-level flow is weaker and difluent over southeast Arizona.

So what does all this mean? It appears that through 48 hours the most signficant storms will occur Saturday afternoon and night.

Southern New Mexico appears to have most certain chance of widespread significant storms and rains. Southeastern Arizona may have a good storm event, but timing of moisture increases from John and encroachment of front from the east will be critical. If the front comes through too soon, the best activity zone will be shifted to the west and northwest. There is a good liklihood that the influx of low-level moisture will push far enough north that northwestern Arizona and southern Nevada (and nearby California) will experience signficant storms also.

As I said at the start of this ramble, things are really complex and getting the forecast right for tomorrow afternoon and night is a huge challenge! In the longer-term, the GFS continues John northward into the lower Colorado basin, carving out a weak trough to the west of the block. The magnitude of the rain event that the GFS forecasts after 48 hours is HUGE in the lower Colorado River Basin. The NAM takes John to the west and just forecasts a nice pattern for strong storms over the Southwest for several days.