Forecasting thoughts

I read Chuck Doswell's essay on project Vortex 2 and a few folks have commented who are in the know on some things. I found it to be quite interesting.

Chuck wrote:
Consensus wins by virtue of avoiding extremes. In the process, it is consistently mediocre. A really good forecaster can beat consensus consistently.

In my mind, consensus is essentially a human ensemble of forecasts. Take the average of all forecasts issued by the forecasters and you get consensus. I don't think consensus wins by avoiding extremes, actually it does well no matter how (and maybe even because) individual forecasters go extreme (in both directions sometimes). It covers the phase space. 

I am hard pressed to say that consensus is mediocre, though it is possible. Rather consensus tends to do quite well over the long term. (Wishing I had access to my old UAlbany forecast contest results). That being said, a good forecaster knows WHEN to beat a model that forecasters rely on and knows HOW to beat it. Playing the game you can easily make good bets or poor bets consistently, but to make big leaps in "skill" required risk taking. This comment applies mostly to the numbers (Max/ Min temperature, POP, precip amount).

Consensus on thunderstorm forecasting might be a different story, and supercells would be in another league entirely.

I too am a big fan of the dropsonde and I was very much disappointed that learjet dropsonde was not going to be used. (I tried to keep up with the mobile upsonde reports to see there progress. I really don't have much to say on if they succeeded in their mission or not, publications will have to tell that story.) I think what is most missed though will be a tremendous variety of environmental soundings. Not that the upsonde teams failed in any way. I think they have already published a paper on an MCS ...clearly not the goal of V2 alone.

But I think an opportunity was missed to really explore the environment around supercells which would have required a lot more forecasting precision (initiation location and time, relevant mechanisms teased out for sample flying patterns (across fronts, boundaries, moisture discontinuities, etc)).

The stated goals that were outlined in the essay:

• How, when, and why do tornadoes form? Why some are violent and long lasting while others are weak and short lived?
• What is the structure of tornadoes? How strong are the winds near the ground? How exactly do they do damage?
  
• How can we learn to forecast tornadoes better?

I know the mobile radars in their entirety were all about documenting the structure of tornados, getting close to ground wind speed measurements, and understanding the damage patterns. I think there is a clear linkage between the observation platforms and the goal. Even the first goal can be accomplished withe the variety of storm scale, tornado scale radars ... though I am pretty sure that dual doppler would be needed since we already have single doppler coverage.

I am less certain they can answer why a tornado was short lived (since we have no proof it should have lasted longer) or long lived (since we have no proof it should have been shorter lived). Thats where sample size becomes an issue (and where the project will prove its worth). Are 40 tornado's enough? Do you need to have 40 similar tornado's paired with 40 similar storms to draw some conclusions? Only time will tell.

The last goal sets the bar very high and is another reason why a dropsonde platform would have been so key. Researchers have opted to use the RUC model to circumvent this. It is a decent quality model that has proved its worth but it is still a model. It is always desirable to have observations since it is the small details that tend to matter most. And more soundings over a larger area is exactly what would help. Spatial and temporal coverage needs to improve, but when V2 is over and the mobile radars go home we are left with the "old" network to work with. It would be nice to have perspective and quantification on how good or uncertain some proximity soundings are, and if these proximity soundings tell us something more than tornadoes are possible. We still have no clue why some storms produce them, why some storms don't, and why some storms don't rotate when we think rotation is possible. They should be able to differentiate whether storm scale or environmental processes are dominating with the help of model sensitivity studies.

 Has anyone showed how poor or how good we already forecast tornado's? I believe there is one study showing the lead time of warnings but I am unaware of anything else that truly shows skill at tornado forecasting. I will have to check into that.

Nocturnal storms in the Tri Cities!

If only they could have occurred when I was awake. 1013 UTC thunderstorms developed over Southern WA. We have had altocumulus in the skies the last two days, morning and afternoon. Last night there was some well to the east and I figured perhaps the threat was gone and didnt even bother to pay attention to dessert weather. Surprise! at 315 I awoke to "plink!" "plink!" on the fireplace. A sign that it was raining big drops, but not pouring down. I peaked outside to verify and quickly went back to sleep.

Radar loops show this very nicely. I will have to make a movie of it since it is indeed rare (maybe 1 or 2 nocturnal events in the summertime). I saw very little on the surface chart, perhaps local flows were more dominant given the proximity of the hills around here. Area soundings showed 2C decreases at 500 hPa from 00 to 12 UTC, and a 3C increase at 700 hPa at Spokane. We happen to be on the cold side of the jet and it did not appear to be related to any jet streak circulation especially with anticyclonic curvature.

The dew point has been higher than usual, especially with the missed rain of the other day. It rained south of here 2 days ago, but yesterday was dry and the dew points dipped down into the mid-50s only, then came back up to 60. Of course our weather has been marked decidedly by the regular passage of midlatitude cyclones, which I must say has had a profound effect on Seattle, but not apparently us.


Day 2 update: more TS obs ! 1230-4am, and again this am from 9-11am!

They seem to initiate off the the hills to the south of the basin. Looks similar to the cap rock area.

This afternoon there is a TS playground out by Yakima which is awesome from my westward facing balcony. Just a wall of cumulonimbus:

 

I watched the failed attempts at CI along the boundary as it moved in, then behind the boundary attached to the weakening storms. Some interesting cloud features for around these parts.

multiscale convection initiation 29 July 2010

Tis the season for CI close to the mountains. I have been saving the water vapor image for 0645 UTC so I can look at the monsoon season over AZ and MCSs over the midwest for the season in its entirety. The other evening I saw that the WV imagery depicted some weak mountain generated convection in Northern CO and Southern WY. This is a late season hot spot for convection (proof required, so this is anecdotal at the moment). The story unfolded at the surface as simply southeasterly flow with dew points in the upper 60's. At 850 hPa the situation could be described  as a moisture pool along a warm front.

In any case, it didnt look dynamic until 300 hPa, where an upper level anticyclonic shear zone was situated along the NE SD border, with an anticyclonically curved jet streak approaching. So the situation unfolded as short-lived convection in a pretty humid pouch failed ... drifted eastward and by 0845 UTC CI was underway in this pouch. It was not a big event but did drop copious amounts of rain in the FSD area in eastern SD over into Storm Lake in Iowa.

This appears to be an established corridor for storms this evening. Some could argue it is intimately tied to the developing monsoon moisture stream. Tonights storms initiated over the Rapid City area (mountain generated). They have currently split with a slow moving portion of the high cloud tops over NE and drifting slowly SE, while further to the east in SD the storms are just slightly north of last nights storms.

Storm reports are absent in this corridor with the exception of tonight, so this type of event would only stand out as being different in terms of the precip. The NSSL model run handled this quite nicely though the second diurnal cycle appears to be extreme ... this case could be examined quite well.

Hopefully these types of events don't get ignored since the storm report numbers were low. How was convection initiated? What were the dynamical controls, and how much did the upslope, moist flow at the surface contribute, and what role did pre-existing convection play and how did it manifest itself (humidification or boundary development)?

Predictability

Tough forecast day for SPC in light of the disappearance of the spring forecast experiment model suite. I know I felt the loss of NCARs model when I was making my forecast for Minnesota and Iowa.

The story starts with obscenely high values of dew points across the midwest ... 80+ dew points in Iowa and above 75 just about everywhere else. A potent little low was forecast to move through the area around the Twin Cities with a moderate risk of tornadoes given the extreme instability and large vertical shear.

What was more obvious today was that the hodograph curvature would be decreasing through the day as the cold front advanced eastward. A remnant MCS was forecast to depart the area and another was forecast to erupt in its wake. The NSSL model had a hard time with that component of the forecast, instead developing the convection along the cold front and slightly ahead.

The hard part was utilizing this model run to determine what role the remnant MCS would play in terms of any leftover boundaries, the role of the associated cloud field to the west, and the cold front even further to the west. As it turned out, the departing MCS gave way to another MCS accompanied by damaging winds and tornadoes. The cold front lit up in sections appearing to be linear but isolated enough to warrant tornado warnings before collapsing. About 5 hours later as the front passed the Twin Cities it lit up again.

All in all it was a hard forecast because the model simulation was not quite capturing the episodic nature of the convection. This is one forecast where the model was working against the forecasters despite the somewhat strong forcing for ascent. The details were absolutely lacking in the model forecast and this case will be a prime candidate for further research into why this happened.

This radar image sums it up nicely with the western most convection marking the cold front and accompanied by TOR warnings, Then the cloud field marked by the 25-30 dBz echo, then the isolated convection which exploded into the MCS over the Twin Cities, and the remnant MCS is off to the east.

Updated: MCS

Here is the update of the MCSs that occurred over the last 5 weeks (May 28 - July 4) at 0645 UTC:
If I counted correctly, something like 26 systems (of various sizes) over that span,
not including multiple MCS events, or the fact that multiple systems could be identified on radar under one cloud shield.

The precipitation anomaly for the last 30 days still indicate 8" swaths from NE through IA. Lets recall the radical flooding in OK City on 14-15 June (which also has their 30 day anomaly above 5"). KS also participated in the events but it was much less a focus for extreme  precipitation totals.

60 day anomalies were significantly less revealing the reduced May rainfall in Ia, NE and northern KS.

Compared to 2008 and 2009, June 2010 was literally the summation of the last two Junes. In 2008, 8" anomalies were common in IA and northern MO and a small part of eastern KS. In 2009, 8" anomalies were common in western NE and a sliver in KS, NE and MO (1 small corridor). 2006 and 2007 were very different from these last 3 years, but 2005 was similar to 2009.

Iowa is currently in flood mode with 35 of 99 counties under some kind flood watch or warning. Interestingly, the Iowa Environmental Mesonet reported that DSM had 22 days of measurable rain and more is on the way this week. up until the 22nd the IEM reported the highest mixing ratio average for the state since 1902!

The complete scientist

http://blog.agu.org/sciencecommunication/2010/06/29/complete-scientists/

A classic argument between being just a scientist and being an applied scientist. It is worthy of considering, since science is not just done for pure fun and amusement by the curious. We are, afterall, interested in using science for the betterment of society. Knowledge gains are practical too, since they provide a foundation from which applied science may be constructed.

I took issue with little in this blog post, but I have my 2 cents:
1. The requirement to go out and be social! I like this point but it is far from practical for your average university professor or average young scientist. These are people busy building not just their knowledge base and communication skills, but also their reputation in the scientific community. They are busy expanding their career and acquiring future funding. There is seldom time for such outreach activities. This is not to say they should not be doing outreach, but rather that their time is better spent in the construction and development phase of their career.  Personally I would much rather the elder scientists participate in outreach activities and from what I see, this happens fairly regularly.
 

2. watches and warnings are not intuitive. Oh yeah:
person 1: watch out, their might be a something bad around the corner.
person 2: warning, anvil is about to fall on your head from that coyote!

Seems pretty intuitive to me. watch = prepare, warning=action. See the difference?

The confusion lies in the perception of risk that is built into those words from the culmination of life experiences. I was watching "After the catch" about a fisherman from LA who survived Katrina. He did not evacuate and he almost payed dearly. He thought: I survived before and this one wont be so bad. the gist of his story is his wife floated away, to be found alive 6 hours later, and most of his family held onto trees until they got in a boat.

This type of story is common. The forecasting community recognizes these types of social behavior but there isn't too much to do yet. I do know that an NWS office issued tornado warnings when the eyewall of a certain storm was passing through LA. They wanted the gravity of that situation understood in plain words: all hell is about to break loose if you can hear this message (and in particular, if you havent already realized the gravity of the situation). Kudos to you NWS. 

Breaking the myth of how poor weather forecasts used to be and educating the public on how good they are is difficult to do. I was in an airport a while back and a random conversation ensued: "As a meteorologist I get to keep my job precisely because the forecast was wrong". May seem strange, but bad forecasts imply there are tools we need to improve and situations we need to learn from. The publics perception is that we are always wrong (not precise). But that is the entire point we need to convey ... being precise is not what makes a good forecast. Knowing the evolution is much better that getting all the numbers correct*.

*getting all the numbers correct serves a different purpose: Farmers want to know how the heat will affect crops, or how much rain to expect. People in general want to know what the temp will be and how humid and how sunny. But being precise isnt what is promised. A range is always better since local variations will always be present.

Interesting Iowa weather

 Radar imagery from the morning of the 27th showed what could be either a gravity wave, bore, or cold outflow. The two waves seen as finelines of about 25 dBz are over AMW and approaching DSM.

The time series from the roof of the ISU Agronomy building:
http://mesonet.agron.iastate.edu/plotting/iemcc/?year=2010&month=6&day=27

The time series shows a small drop (1.2F in 5 minutes) in temperature and dew point, a brief recovery, then a second drop (3F in 7 minutes). The third drop is actually the outflow boundary from the storms to the west, not the north. So it is plausible that the bore or gravity wave arrived first followed closely by the outflow boundary. 

The temp and pressure relationship was non-existent for the 1st "wave" (dp = 0), but strong for the 2nd (dp = 0.5 hPa). the wind speeds in the 2nd wave peaked after the min pressure occurred during the t fall. The last wind speed increase occurred in the wake of the 2nd wave but prior to the rapid, small pressure oscillations. I wonder if the pressure sensor is affected at particular wind directions.