Weather: December 2007 Archives

First a Tropical Storm, Now an Ice Storm...

We in central Oklahoma just finished the bulk of an episodic, two-day ice storm that rendered nearly a half million residents without electricity, the largest figure in state history. Schools are closing for at least two days -- more because of power outages and direct blockages by fallen trees than from driving restraints imposed by ice. Around half an inch to an inch of solid accumulation is quite common, busting branches and sometimes taking down whole trees. I could hear the distinctive crack-n-crash sounds in the woods around my property for much of the day, and saw several large branches topple -- an experience common to anyone outdoors across several central OK counties.

For the Christmas-1987 ice storm, I was an OU student out of town for Christmas break, so my last experience with this magnitude of ice or greater was in Dallas, New Year's Day, 1979. That was worse, with more ice (1.5 to 2 inches) in a city of a million people, and 600,000 people without power. I vividly recall that Saturday night and Sunday morning, staying awake until after 3 a.m. to the noise of crashing limbs and the flashes of both lightning and power line blowouts. After we lost power, I remember listening in a transistor radio to Ray Ward, Dallas Power and Light spokesman, describing the mounting troubles throughout the night in various interviews. Another vivid memory from that event was all the yankee license plates (Ohio, Michigan, Illinois, Wisconsin, etc.) that I saw on vehicles that slid off Central Expressway and other area roads. [This was during a peak of the mass exodus of jobs to the Sun Belt from the Rust Belt.]

This Okie event definitely brought back some memories and points for comparison. A key difference is that the ground was warmer here, and the roads stayed largely ice-free. Unlike the 1987 event in Norman, or the 1979 Dallas storm, travel was inhibited in the latest ice storm only by fallen trees and by powerless signals. Nonetheless, while this ice storm isn't as severe in terms of absolute amount as that Dallas event, or separate but similarly thick ice accumulations several years ago in northwestern and southeastern Oklahoma, its location over relatively thickly populated areas has created a considerable mess.

Several rounds of mostly spotty, convective rains atop the subfreezing air caused this event. The heaviest thunderstorms, overnight last night, produced very short-lived and localized rain rates to near an inch an hour, with more common rates of .10-.25 inches over a larger area. Still, rain rates didn't add up to total accumulation, because there still was obvious runoff in the heaviest thunderstorm cores.

Here in Norman, we had a 36-hour period where no more than a couple of hours passed between seeing lightning or hearing thunder, all with below freezing surface temperatures, thanks in part to a series of perturbations in southwesterly flow aloft, and their destabilizing effects above an elevated plume of low level moisture. Last night, one of those lightning strikes hit in the woods next to my house, contributing to the first downed tree branch of many (either through direct strike, or through percussive effects).

This hasn't been Chamber of Commerce weather. We've had some power outages, but luckily, not the multi-day ones (as we did in Dallas in '79 or others in Oklahoma have currently). My neighborhood, while liberally festooned with downed branches and trees, still fared better than those of folks in the main part of Norman and north-northeastward across parts of Oklahoma and Lincoln Counties, where the heaviest storm totals were. Thousands upon thousands of trees and large limbs are down; and once the ice melts, it will look as if a hurricane selectively hammered only the vegetation.

Farther north, across much of KS, MO, southeast NE and IA, the ground is much colder; half as much rain will create much more travel problems and still take down some trees and limbs.

The sounds of chain saws -- including mine -- will permeate the air for a couple of weeks to come. Much as in parts of Dade County FL, after Hurricane Andrew, trees around these parts are going to look shorter, really haggard and imbalanced for a few years, until younger branches and trees grow enough to fill in the canopy.

I'm not going to complain too much. Those of us who choose to make our home in the Sooner State either do, or damn well should, accept that these things do happen here, and learn to cope. We get ice storms, deep droughts, 100+ degree heat waves, dust storms, tornadoes, tremendous flash floods, giant hail and derechos. All these combine to make life hard on trees in these parts, and in combination with poor soil, explain why we don't have many tall or fast-growing trees. This is living in Oklahoma (and, when applicable, adjacent north Texas and Kansas). We deal with it and forge forward. I make lemonade from lemons by doing both documentary and artistic photography, though my shift and sleep schedule won't allow much opportunity to get out and chronicle this event in images as thoroughly as other folks.

And for those of us who are meteorologists, we also learn as much as possible about these events in order to better predict them in the future. Fortunately, this one generally was well forecast; and the Norman forecast office deserves a good deal of credit for discussing ice potential a few days out in hazardous weather outlooks, then ramping up threat levels to an ice storm warning well before the first trees fell.

I've posted a small gallery of photography from the event, including the ones above, at this location online.

While it's still there, check out a Houston Chronicle article, within which former NHC director Neil Frank (for whom I do have a good deal of respect, BTW) questions NHC's decision to name several weak or debatably tropical 2007 storms.

If Neil was quoted in good context in that article -- never a given in the modern attack-media -- he seems to advocate turning back the clock and treating some of this year's hybrid or cold-to-warm core transitions as something other than tropical. If so, I don't agree. His statements are patently oversimplified and unfair to the forecasters now there. We as scientists need to do the best we can, under the best understanding currently available, and call a spade a spade. This means tropical if it meets all the criteria (as did, for example, Erin over Oklahoma, which was: warm core, 35+ kt sustained for several hours, structurally characteristic with an eye, eyewall and spiral band, effectively devoid of baroclinicity and ingesting low level temperature and moisture fields similar to the tropics). [Erin's final "best track" info still hasn't been fully determined yet.]

1979-era rules of thumb don't apply in the framework of today's vastly better sensor sampling. This includes QuikSCAT, coastal and inland fixed mesonets, mobile observing systems and deployed sensors in landfalling storms, fixed and mobile Doppler radars, and other tools that were either unavailable or much more primitive in that era. We're seeing named storms that wouldn't have been named way back when, you bet. But there are valid reasons.

This isn't because of some desire to inflate numbers for verification's sake or to support any "global warming" hysteria, as some of the reactionary but grotesquely ignorant BLOGgers have posted in response to that article. I'll respond for now to the BLOGs and postings in response to that article, not to Neil's quotes...

It's a lot harder to play verification games with TCs than with severe storms, given the intense scrutiny that every single Atlantic event gets in real time from storm aficionados of all flavors worldwide, as well as real meteorologists and meteorologist wanna-be's. I firmly believe that NHC is not playing verification games, has not, and won't be. I know those forecasters and they've got more integrity than that. By contrast, locally inflating a thunderstorm wind report to verify a warning, recording a dubious gust estimate that happens to occur in a warning, turning a "marble into a shooter" hailstone, or declaring an ongoing event as occurring one minute into the warning, is not so hard to do, and can go under the radar (pun intended) amidst the prevailing flurry of tens of thousands of other reports every year. Does that happen nowadays in verifying local storm warnings? I can't firmly vouch or prove either way. But I am quite confident that NHC is acting with the utmost integrity in classifying tropical systems. So...

As for those claiming NHC is playing "global warming" games, that's unadulterated crap. Real live NHC forecasters aren't dealing with climate change in their forecasting and verification duties -- just trying to make the best predictions possible and figure out what happened with each storm after the fact, using all available evidence. There is no grand conspiracy, period...no bogeyman under the bed, no chains rattling in the attic, no little green space-alien freaks planting chips in our heads, and no massive hidden agenda on naming tropical systems. Of course, irrational paranoids in the BLOGosphere never will be convinced otherwise!

Back to Neil's implication that marginal or hybrid systems could be left alone: Now isn't an era when some P.O.S., no-impact swirly in the middle of the subtropical Atlantic can be swept under the rug! NHC cannot just ignore marginal systems "way out there" or fail to name a storm that has strong evidence of qualifying characteristics.

Unlike in the 70s and 80s, satellite, radar, RECON and other data is available almost instantly, worldwide, to anyone with Internet connectivity. This audience runs the gamut from some really brilliant atmospheric scientists to beer-swilling, ill-educated conspiracy mongers drooling all over their keyboards in facile worship of Art Bell. From both extremes and everywhere between, accountability exists as never before, and the spotlight shines bright and hot. It's also true that not every customer will agree with the results, given the sheer size and diversity of the user base behind those spotlights. Clearly Neil and some of the BLOGgers aren't happy with it. So be it. Can't please everyone...

In the tropical arena, we are getting more marginal or atypical systems named than before for at least these three reasons, the first also being a legitimate reason we've got more severe local storm reports in the U.S. now as well:

1. They are being sampled much more thoroughly, both remotely and in situ, and

2. Scientific understanding of their existence, behavior and morphology has improved, contributing to better documentation of the processes and transitions and distinctions between tropical and extratropical. As meteorologists continue to gain understanding, and as technological and measuring capabilities improve, classification of tropical systems will and should evolve too!

3. Both policies and practices on naming subtropical systems have loosened since the 80s, and even since I was there in the early 90s. We also know the atmosphere doesn't give a flip about naming policy, and will keep cranking out these systems whether or not we decide to call them TS Bubba, STS Bubba or just "low." Therefore, the climatological record will include biases related to either written or unwritten rules changes on nomenclature. For a tropical or climate researcher, this sucks; but it is what it is. Adapt and adjust accordingly. Severe storms scientists are having to account for a much larger, crazier and more change-prone database.

These also are good reasons not to compare apples to oranges and try to make sweeping climatic conclusions from our flawed and inconsistently gathered records of Atlantic (or even worldwide) TCs. The same holds true even more so in the severe weather arena, where even relatively advanced U.S. records haven't been kept as long, and deep flaws in the data are well enunciated in the literature.

What is the good researcher to do? Any atmospheric scientist worth his/her degree(s) is absolutely obligated to either
1. Fully disclose and acknowledge the recording trends and biases in the cyclone data used for his/her work, or
2. Carefully and openly document, normalize and statistically detrend the data for all known biases, so as to minimize the impact on the storm climatology of the evolution in sampling, recording and naming of these storms.