The setup for significant winter weather in the Southern Plains over the next two days is meteorologically pretty simple. A deep, frigid air mass at the surface has been driven south by high pressure centered in southwestern Canada. It lies underneath southwesterly flow on the east side of an expansive upper trough, which is providing both ample moisture and strong forcing for ascent (especially along embedded shortwaves).
Model guidance has converged on the idea of a transition from freezing rain to sleet to snow, moving southeastward with time as the warm nose around 700-800 mb cools. The heaviest precipitation appears likely to fall within a narrow band oriented roughly from southwest to northeast. The forecast challenges are
1. Nailing down precipitation types in space and time, and
2. Placing the band of heaviest precipitation
Precipitation types will depend almost completely on subtle variations in temperature in the lowest 3 km or so of the atmosphere. Even one day out, this is pretty tough to assess without relying on model forecast soundings. For Norman, consensus between the NAM and GFS seems to be some freezing drizzle and light sleet possible in the earliest stages of the event Thursday afternoon, changing to snow as the column wet-bulbs Thursday night. Moderate to heavy snow appears likely into Friday morning, ending midday. It appears the amount of sleet in central Oklahoma might be trending downward, with classic heavy snowfall soundings making more appearances there in both models late Thursday night/early Friday morning. Freezing rain looks to be a major, major issue for parts of southeastern Oklahoma into Arkansas, where the warm layer will be much more pronounced even after reaching saturation.
As far as the band of highest QPF, models have been wobbling back and forth across an area from I-44 to a Fort Smith-Dallas line. A lot will depend on the exact placement of the front (from the surface up to 700-800 mb) For what it's worth, the surface front ran a bit ahead of the forecast overnight. With that in mind and the major models coming into better agreement, Norman looks to catch at least the northwestern edge of some pretty hefty precip totals.
Norman forecast: A fine glaze of freezing drizzle on elevated surfaces during the day Thursday, with an inch or less of sleet accumulation through evening. Precip will rapidly turn to snow Thursday evening. Periods of heavy snow overnight into Friday morning; 2-5" total snowfall.
Wednesday, December 4, 2013
Tuesday, June 11, 2013
Ohio Valley forecast, Wednesday, June 12
Quick forecast discussion for likely severe weather event in Ohio Valley, 6/12/13...
Well-defined surface low pressure now in the lee of the Rockies is forecast to move toward the Midwest, with NAM and GFS agreeing on its center in NE MO/SE IA tomorrow. Even warmer, more moist air than is already in place will continue to be advected into the Ohio Valley over the next 12-24 hours; all guidance indicates this will result in CAPE > 3000 J/kg across much of the Ohio Valley/Upper Midwest/southern Great Lakes. Aloft, a 500-mb speed max with 40-60 kt, generally WNWly flow should overspread the area of interest, from MO through IL, IN, and OH into central WV. NAM forecasts a significant shortwave trough moving into the area midday.
While forecast CAPE is not on the order of the 5000-6000 J/kg available in the high-end Appalachian derecho of 6/29/2012, the amount of instability is still unusually high for the region. Flow from mid-levels on up will be stronger in this event than in last year's, and the increased shear is somewhat concerning in terms of supercells--whether discrete or embedded in a damaging QLCS. It appears the main source of lift will be near the western end of SPC's Day 2 moderate risk area, in IL and IN, in the afternoon hours. However, the pronounced cap from last year's derecho is absent on forecast soundings, so smaller shortwaves or convergence along any surface boundary may result in discrete convection farther east. Supercell composite forecasts from the SREF are alarming for OH and west WV, should discrete storms develop.
Forecast: Supercells initiate from central IL across IN, possibly into OH and/or far western WV, mid-afternoon. Some of these will pose a tornado threat before growing upscale into one or more lines. These squall lines will be capable of widespread significant wind damage as they turn southeastward from IL, IN, and OH through much of WV, and possibly SW PA. Any organized line of storms will likely continue to produce damaging winds late into the night. Severe hail will also be possible with the strongest storms.
The area most at risk in WV should be from Huntington to Wheeling eastward to the I-79 corridor.
The area most at risk in WV should be from Huntington to Wheeling eastward to the I-79 corridor.
Monday, May 20, 2013
Moore musings
I'm home in West Virginia tonight, getting ready to move to Birmingham, AL, for a summer internship. As mostly weak, photogenic tornadoes touched down over previous days in the Southern Plains, it was a little frustrating to be missing the atmosphere at its best. Last night and especially now, it's sickening to be stuck here, unable to help at all. My prayers will continue to be with the people of Moore and Shawnee, as well as smaller communities hit.
First impressions:
1. The Moore mythology is back. After the May 3, 1999 F5 and the May 8, 2003 F4 carved partially overlapping paths of destruction through Moore, there was a strong sense that this unlucky city somehow attracted tornadoes. While there may be some unusual locations around mountains, large river valleys, etc., where geography plays a significant role in tornadogenesis, I can almost guarantee that's not the case in central Oklahoma. It's not pancake-flat, but it's close enough that I seriously doubt terrain influences the tracks of tornadoes. People have a wrongheaded view of what "random" means. A perfectly spaced distribution of tornadoes where nobody is ever hit twice would be anything but random. Random means last year's tornado strike has absolutely no bearing on your chance of being struck this year. A coin that lands tails-up ten straight times still has a 50/50 chance of landing tails-up on the next flip. The dramatic failure of the so-called Norman bubble on 5/10/10 and 4/13/12 had begun to detract from the Moore lore. However, recent town hall meetings conducted by Randy Peppler, Kim Klockow, and Rick Smith indicated the public still felt Moore was abnormally vulnerable to tornadoes.
There is no hope of containing the urban legend now. Admittedly, this is hard to explain away as a coincidence. Per SPC's tornado environment browser, the part of central Oklahoma surrounding OKC-Moore-Norman saw 21 violent tornadoes from 1962 to 2011. These tornadoes' damage swaths covered an area of 306.9 square miles. The radius of the region we're looking at is 75 miles, giving an area of about 17,662.5 square miles. So, even here in the tornado capital of the world, only about 1.74% of the land area was affected by F4-5 tornadoes over the course of 50 years. That roughly translates to any given point having a 0.49% chance of being hit in a 14-year period. There are not one, but two, points in the 21.9 square miles of Moore that were each hit twice from May 1999 to May 2013. In this span, four violent tornadoes have struck that sliver of real estate (a less infamous EF4 hit south Moore on 5/10/10). If the public perception that tornadoes tend to go through Moore had been somewhat weakened, that's no longer the case. And who can blame a non-meteorologist for feeling that way? We can only hope it doesn't create an illusion of security in surrounding areas.
2. Sometimes you just can't win. The most heartbreaking story I've heard in this, or any, tornado event is that of the children in the elementary school who sought shelter in designated areas and still did not survive. There were probably many other victims across Moore who did everything right and sought the best shelter available to them, but died anyway. Closets and bathrooms simply weren't enough.
The bottom line here is that high-end tornadoes in densely populated areas will inevitably take lives. Today, nature spotted us five points. First, this was a classic supercell in broad daylight. Most people in the path could see the beast approaching. Second, storm motion was reasonable--about 30 mph, compared to 50+ mph for some of the April 27, 2011 monsters in Dixie Alley. Third, it was not rush hour; 5/3/99 was. Fourth, the storm was tornado-warned 16 minutes before the funnel condensed. The proximity of the TDWR west of Norman provided an extraordinary view of the elongated hook echo at that time. Most of Moore had lead time in excess of 30 minutes. And fifth, this was Moore. Nowhere on earth is there a more storm-ready population. People there don't think their town is immune from tornadoes; if anything, they swear it's a twister magnet. Most residents have been through at least one big tornado and have some plan of action. Live severe weather coverage on OKC TV and radio stations is second to none. Most homes in Moore are solidly built, in contrast to the mobile homes the Shawnee tornado devoured less than 24 hours prior. Nature spotted us five points and still won big. Sometimes it just doesn't matter.
3. Meteorologically, this came out of left field. When the emotional side of my brain wore itself out and the geeky side kicked in, I started to wonder why this freakish storm did what it did. If this singular tornado had never existed, or had hit rural areas, you could scarcely justify using the word "outbreak" at all. There were 16 preliminary reports of tornadoes, scattered from central Texas to northeastern Missouri, and including a couple landspouts on the High Plains far from the main action. This was not a May 3rd or a May 24th, a day with obscene parameters across the board. In fact, some chasers thought the previous day's storms would be the best this system had to offer. Here's a special sounding taken at Norman less than three hours before the Newcastle-Moore tornado began doing damage:
Yeah, thermodynamically, this is a loaded gun in the process of going off. But those winds...SRH around 150 is not what you associate with violent tornadoes. You would expect a large curved or even sickle-shaped hodograph in the lower levels, not that anemic veering. It's absolutely unbelievable that one of the most catastrophic tornadoes in recorded history came in an environment where 900-mb winds were SSW at 20 kt. I know 5000 J/kg SBCAPE will make up for a lot. It may effectively alter wind profiles in the vicinity of the storm. Yet if that's all there is to it, why didn't we see an outbreak? Every other storm in the region, despite the instability, behaved in a way appropriate for this wind profile. This storm exploded from nothing, immediately generated an exceptionally powerful tornado over a 17-mile track (mediocre for an EF4-5), and did not produce another significant tornado. Something critical happened on the mesoscale level near Newcastle. I want to know what it was.
Updated: Here's spotty data from the Purcell wind profiler leading up to the event. It indicates no real improvement in environmental low-level wind fields through 2:00 p.m. local time, less than an hour before the tornado touched down.
4. I love Oklahoma. From the people who stepped out of their cellars and only said, "The Lord giveth and the Lord taketh away," to the thousands upon thousands now focusing their efforts on the survivors, there is no better place on the planet. I'm not saying those traits are exclusively Oklahoman. They're just on display in the Sooner State time and again when bad things happen to its good people. You might call my second home flat, backward, redneck, overweight, and a little rebellious, but I'd do just about anything to be there tonight.
First impressions:
1. The Moore mythology is back. After the May 3, 1999 F5 and the May 8, 2003 F4 carved partially overlapping paths of destruction through Moore, there was a strong sense that this unlucky city somehow attracted tornadoes. While there may be some unusual locations around mountains, large river valleys, etc., where geography plays a significant role in tornadogenesis, I can almost guarantee that's not the case in central Oklahoma. It's not pancake-flat, but it's close enough that I seriously doubt terrain influences the tracks of tornadoes. People have a wrongheaded view of what "random" means. A perfectly spaced distribution of tornadoes where nobody is ever hit twice would be anything but random. Random means last year's tornado strike has absolutely no bearing on your chance of being struck this year. A coin that lands tails-up ten straight times still has a 50/50 chance of landing tails-up on the next flip. The dramatic failure of the so-called Norman bubble on 5/10/10 and 4/13/12 had begun to detract from the Moore lore. However, recent town hall meetings conducted by Randy Peppler, Kim Klockow, and Rick Smith indicated the public still felt Moore was abnormally vulnerable to tornadoes.
There is no hope of containing the urban legend now. Admittedly, this is hard to explain away as a coincidence. Per SPC's tornado environment browser, the part of central Oklahoma surrounding OKC-Moore-Norman saw 21 violent tornadoes from 1962 to 2011. These tornadoes' damage swaths covered an area of 306.9 square miles. The radius of the region we're looking at is 75 miles, giving an area of about 17,662.5 square miles. So, even here in the tornado capital of the world, only about 1.74% of the land area was affected by F4-5 tornadoes over the course of 50 years. That roughly translates to any given point having a 0.49% chance of being hit in a 14-year period. There are not one, but two, points in the 21.9 square miles of Moore that were each hit twice from May 1999 to May 2013. In this span, four violent tornadoes have struck that sliver of real estate (a less infamous EF4 hit south Moore on 5/10/10). If the public perception that tornadoes tend to go through Moore had been somewhat weakened, that's no longer the case. And who can blame a non-meteorologist for feeling that way? We can only hope it doesn't create an illusion of security in surrounding areas.
2. Sometimes you just can't win. The most heartbreaking story I've heard in this, or any, tornado event is that of the children in the elementary school who sought shelter in designated areas and still did not survive. There were probably many other victims across Moore who did everything right and sought the best shelter available to them, but died anyway. Closets and bathrooms simply weren't enough.
The bottom line here is that high-end tornadoes in densely populated areas will inevitably take lives. Today, nature spotted us five points. First, this was a classic supercell in broad daylight. Most people in the path could see the beast approaching. Second, storm motion was reasonable--about 30 mph, compared to 50+ mph for some of the April 27, 2011 monsters in Dixie Alley. Third, it was not rush hour; 5/3/99 was. Fourth, the storm was tornado-warned 16 minutes before the funnel condensed. The proximity of the TDWR west of Norman provided an extraordinary view of the elongated hook echo at that time. Most of Moore had lead time in excess of 30 minutes. And fifth, this was Moore. Nowhere on earth is there a more storm-ready population. People there don't think their town is immune from tornadoes; if anything, they swear it's a twister magnet. Most residents have been through at least one big tornado and have some plan of action. Live severe weather coverage on OKC TV and radio stations is second to none. Most homes in Moore are solidly built, in contrast to the mobile homes the Shawnee tornado devoured less than 24 hours prior. Nature spotted us five points and still won big. Sometimes it just doesn't matter.
3. Meteorologically, this came out of left field. When the emotional side of my brain wore itself out and the geeky side kicked in, I started to wonder why this freakish storm did what it did. If this singular tornado had never existed, or had hit rural areas, you could scarcely justify using the word "outbreak" at all. There were 16 preliminary reports of tornadoes, scattered from central Texas to northeastern Missouri, and including a couple landspouts on the High Plains far from the main action. This was not a May 3rd or a May 24th, a day with obscene parameters across the board. In fact, some chasers thought the previous day's storms would be the best this system had to offer. Here's a special sounding taken at Norman less than three hours before the Newcastle-Moore tornado began doing damage:
Yeah, thermodynamically, this is a loaded gun in the process of going off. But those winds...SRH around 150 is not what you associate with violent tornadoes. You would expect a large curved or even sickle-shaped hodograph in the lower levels, not that anemic veering. It's absolutely unbelievable that one of the most catastrophic tornadoes in recorded history came in an environment where 900-mb winds were SSW at 20 kt. I know 5000 J/kg SBCAPE will make up for a lot. It may effectively alter wind profiles in the vicinity of the storm. Yet if that's all there is to it, why didn't we see an outbreak? Every other storm in the region, despite the instability, behaved in a way appropriate for this wind profile. This storm exploded from nothing, immediately generated an exceptionally powerful tornado over a 17-mile track (mediocre for an EF4-5), and did not produce another significant tornado. Something critical happened on the mesoscale level near Newcastle. I want to know what it was.
Updated: Here's spotty data from the Purcell wind profiler leading up to the event. It indicates no real improvement in environmental low-level wind fields through 2:00 p.m. local time, less than an hour before the tornado touched down.
4. I love Oklahoma. From the people who stepped out of their cellars and only said, "The Lord giveth and the Lord taketh away," to the thousands upon thousands now focusing their efforts on the survivors, there is no better place on the planet. I'm not saying those traits are exclusively Oklahoman. They're just on display in the Sooner State time and again when bad things happen to its good people. You might call my second home flat, backward, redneck, overweight, and a little rebellious, but I'd do just about anything to be there tonight.
Wednesday, April 17, 2013
Thoughts on Wednesday
I don't know if I'll chase today. Probably will. It looks like the front placement may split the difference between the NAM and GFS, and that leads to an interesting thought. Just wanted to mention something that's floating around in the realm of possibilities.
- Mid-afternoon: Convection initiates on the dryline not too far south of the triple point in NW TX.
- Strongly backed low-level flow carries the developing updraft almost due north (this is why I think nothing initiated by the front will survive).
- Late afternoon: The storm strengthens and turns to the northeast in the ideal environment just south of the front, in SW OK.
- With a continued right turn, the supercell's motion approaches the SW-NE boundary orientation.
- Early evening: Motion becomes parallel to the boundary at exactly the time the storm reaches it.
- Through late evening: LLJ arrives in force and the rogue boundary-riding supercell drops a string of significant tornadoes across OK.
- Front begins advancing southeastward again, finally undercutting the storm.
As a low-probability (storm has to fire in precisely the right place), high-impact (could produce tors along a 100+ mile path) event, is that far-fetched?
More likely, we'll see a couple nice supercells around Snyder-Wichita Falls-Walters-Lawton. These would easily be capable of producing significant tornadoes. Convection along the front stands no chance of becoming established before moving northward into colder air, so this may limit severe potential in much of Oklahoma. RAP's orientation of the front in NE OK may, however, allow a few storms to get into the warm sector in NE OK/SE KS/SW MO. This would mean a second tornado hotspot up there, where hodographs are just obscene. For now, I think the only major tornado threat to central OK is as described above, with a storm from the dryline rooting on, or even turning eastward from, the frontal boundary. No matter what happens, it's time for me to get sleep now.
Preliminary target: Grandfield, OK
- Mid-afternoon: Convection initiates on the dryline not too far south of the triple point in NW TX.
- Strongly backed low-level flow carries the developing updraft almost due north (this is why I think nothing initiated by the front will survive).
- Late afternoon: The storm strengthens and turns to the northeast in the ideal environment just south of the front, in SW OK.
- With a continued right turn, the supercell's motion approaches the SW-NE boundary orientation.
- Early evening: Motion becomes parallel to the boundary at exactly the time the storm reaches it.
- Through late evening: LLJ arrives in force and the rogue boundary-riding supercell drops a string of significant tornadoes across OK.
- Front begins advancing southeastward again, finally undercutting the storm.
As a low-probability (storm has to fire in precisely the right place), high-impact (could produce tors along a 100+ mile path) event, is that far-fetched?
More likely, we'll see a couple nice supercells around Snyder-Wichita Falls-Walters-Lawton. These would easily be capable of producing significant tornadoes. Convection along the front stands no chance of becoming established before moving northward into colder air, so this may limit severe potential in much of Oklahoma. RAP's orientation of the front in NE OK may, however, allow a few storms to get into the warm sector in NE OK/SE KS/SW MO. This would mean a second tornado hotspot up there, where hodographs are just obscene. For now, I think the only major tornado threat to central OK is as described above, with a storm from the dryline rooting on, or even turning eastward from, the frontal boundary. No matter what happens, it's time for me to get sleep now.
Preliminary target: Grandfield, OK
Tuesday, April 16, 2013
The supercell drought ends
Storms? What are those? It's been so long since the Southern Plains has seen convection worth a drive that I almost forgot what it was like. Today was a marginal day, with SPC extending, then removing, then extending again a slight risk area from E OK down through N TX. I got into the first few problems of math homework early in the afternoon, took care of a couple other things, and noticed short-range models continuing to insist on a few storms along the stalled front. With 2000-3000 J/kg CAPE and 40+ kt shear, these would probably be supercells. I started rolling south on I-35. While the unbroken pattern of cloud streets in E and SE OK suggested strong capping persisted in that area, well-defined Cu lined the frontal boundary in N TX. A mesoscale discussion came out from the SPC, followed in short order by a tornado watch box. I initially pegged Madill, OK, as a target, but got lost in Ardmore after taking the wrong exit. Might've been for the best, as I returned to I-35 instead of pressing east to Madill. A quick stop on the south side of Ardmore combined gas, water, and weather radio battery resupply. About that time, KFWS radar finally came back into service after being down for a while. Seeing a cluster of storms approaching Bowie, TX, I was pretty sure I could intercept in Texas.
Maneuvering around Gainesville on I-35 at 4:30 was an adventure in itself, with what appeared to be exit ramps on the map never materializing. I finally found a frontage road and stayed on it into Valley View, where I headed west on two-lane back roads with the nowcasting guidance of a friend back in Norman. I caught a glimpse of the lead cell, dominant at the time, in a rural area between Decatur and Gainesville.
This marginally severe supercell became less impressive as it tracked just north of my location. I edged northwestward--road options forced a northwesterly or southwesterly path--to position for another supercell at the back of the cluster as it became dominant. Many miles in the distance, I noticed an interesting feature under its updraft. It looked to be scud, but persisted for a while with noticeable motion. I stopped on a gravel spot along the shoulder, but was immediately met by a local resident coming home. He was pretty touchy when I tried to talk to him. It turned out that one of the gravel roads that rolled over the hill from that pull-off spot was his driveway. I was a couple feet off the roadway, well outside his fence. There was no house anywhere to be seen. But rural Texans own guns, and if they say you're violating their privacy on the shoulder of a public road, you don't stand there in your crimson OU hat and argue the point. I apologized and rolled on, catching a couple quick shots of the feature of interest. Other chasers claim it as a funnel, but nobody seems to have been close enough to verify that.
I came into Forestburg, TX, and stopped for a couple more structure shots across a football field in disrepair.
This supercell was peaking as I left Forestburg; it was just about to dump 1"+ hail on Sunset, TX, to the southwest. I wanted to get to Alvord, TX, just to the southeast of Sunset. I missed the turn for Alvord, and found myself headed for Sunset. I turned around, went back, and got on the right road. Flying solo screws you up sometimes. As I headed southwestward toward Alvord, the supercell's structure became otherworldly.
I pulled off the road at an oil or natural gas facility after passing numerous chasers. They had the right idea. Even with my wide-angle camera, which is usually a pain because the subject of the photo isn't big enough, I couldn't capture the full structure. Still, being alone under this mothership-style LP storm, watching the striated front side of the thing stream northward into the updraft, was incredible.
Maneuvering around Gainesville on I-35 at 4:30 was an adventure in itself, with what appeared to be exit ramps on the map never materializing. I finally found a frontage road and stayed on it into Valley View, where I headed west on two-lane back roads with the nowcasting guidance of a friend back in Norman. I caught a glimpse of the lead cell, dominant at the time, in a rural area between Decatur and Gainesville.
This marginally severe supercell became less impressive as it tracked just north of my location. I edged northwestward--road options forced a northwesterly or southwesterly path--to position for another supercell at the back of the cluster as it became dominant. Many miles in the distance, I noticed an interesting feature under its updraft. It looked to be scud, but persisted for a while with noticeable motion. I stopped on a gravel spot along the shoulder, but was immediately met by a local resident coming home. He was pretty touchy when I tried to talk to him. It turned out that one of the gravel roads that rolled over the hill from that pull-off spot was his driveway. I was a couple feet off the roadway, well outside his fence. There was no house anywhere to be seen. But rural Texans own guns, and if they say you're violating their privacy on the shoulder of a public road, you don't stand there in your crimson OU hat and argue the point. I apologized and rolled on, catching a couple quick shots of the feature of interest. Other chasers claim it as a funnel, but nobody seems to have been close enough to verify that.
I pulled off the road at an oil or natural gas facility after passing numerous chasers. They had the right idea. Even with my wide-angle camera, which is usually a pain because the subject of the photo isn't big enough, I couldn't capture the full structure. Still, being alone under this mothership-style LP storm, watching the striated front side of the thing stream northward into the updraft, was incredible.
And that's really all that's worth seeing. The storm went through some contortions as it decayed, dying out with the decline of surface heating. Is it worth driving to rural northern Texas to experience a few minutes of epic storm structure? Of course!
On a side note, as I finished writing this, central Oklahoma felt a preliminary magnitude 4.3 earthquake. Here's hoping this apartment complex makes it through the night.
Monday, April 1, 2013
Spring!
It may finally be springtime on the plains of Oklahoma. I chased a dying hailstorm near El Reno, OK, on 3/29, then showed up late to the Medicine Park-Walters hailer the following day with a couple friends from SoM. Not a lot to show from either day. Norman got run over by a beast of a severe thunderstorm about 4 a.m. on Easter morning, 3/31. Our apartment complex got plenty of hail, quarters and smaller, while golf balls fell farther south. Video:
The GFS and ECMWF are in agreement that an upper trough will dig into the Southwest early next week. As is often the case, the Euro slows things down a bit. The latest GFS (00z 4/2) is the fifth consecutive run to indicate a significant severe weather event in W OK and/or NW TX on Monday, April 8. It has the highest potential of any run yet, with 3000 J/kg CAPE, 50-60 kt shear, and the enormous sickle-shaped hodographs of a violent tornado outbreak. I don't like to trust models at this range, but these last several runs are hard to ignore. Based on the consistency of the GFS, even with some doubts about exact timing and location, I wouldn't be surprised to see a Day 7 area highlighted in the SPC's next Day 4-8 outlook.
Even less certain is the setup for the following day, Tuesday, April 10. A couple older GFS runs hinted at a major event in AR. The most recent has 3000-3500 J/kg CAPE across most of E/C TX, with tantalizing wind profiles up around the DFW metro. The Euro solution might lend itself even more to multiple days of severe weather in the Southern Plains if the system stalls. I think it's safe to say Monday and Tuesday are worth watching...and I'll be wrapping up homework by this weekend. I wasn't in the TX Panhandle for today's surprise tornado, in part because I was doing Matlab at 3:30 in the morning and couldn't consider making a 3+ hour one-way drive. That isn't going to happen again!
The GFS and ECMWF are in agreement that an upper trough will dig into the Southwest early next week. As is often the case, the Euro slows things down a bit. The latest GFS (00z 4/2) is the fifth consecutive run to indicate a significant severe weather event in W OK and/or NW TX on Monday, April 8. It has the highest potential of any run yet, with 3000 J/kg CAPE, 50-60 kt shear, and the enormous sickle-shaped hodographs of a violent tornado outbreak. I don't like to trust models at this range, but these last several runs are hard to ignore. Based on the consistency of the GFS, even with some doubts about exact timing and location, I wouldn't be surprised to see a Day 7 area highlighted in the SPC's next Day 4-8 outlook.
Even less certain is the setup for the following day, Tuesday, April 10. A couple older GFS runs hinted at a major event in AR. The most recent has 3000-3500 J/kg CAPE across most of E/C TX, with tantalizing wind profiles up around the DFW metro. The Euro solution might lend itself even more to multiple days of severe weather in the Southern Plains if the system stalls. I think it's safe to say Monday and Tuesday are worth watching...and I'll be wrapping up homework by this weekend. I wasn't in the TX Panhandle for today's surprise tornado, in part because I was doing Matlab at 3:30 in the morning and couldn't consider making a 3+ hour one-way drive. That isn't going to happen again!
Tuesday, March 5, 2013
Saturday 3/9 forecast
Social media's already starting to light up with talk of Saturday as a possible chase day. Greg Forbes is leading the charge with a "TORCON" of 3 in W OK, I think. The setup is still fuzzy. The Euro has an upper low centered over CO by Saturday evening, while the GFS keeps it farther southwest and almost cuts it off a little bit. Both models forecast a modest surface cyclone centered around the N TX Panhandle, the OK Panhandle, and SW KS. The Euro leans toward a broader, weaker low. Even the GFS only manages to advect mid/upper 60s surface temps and upper 50s dewpoints into the area of interest; I have to imagine the Euro solution would result in even less favorable low-level conditions. The GFS solution has <750 CAPE along the dryline except for a little bullseye in KS. Forecast hodographs are unconvincing as well, with less than impressive low-level shear and a weird kink around 500 mb. Unless model guidance changes a lot, I don't expect more than maybe a short drive west for lightning after dark. Even shelf cloud shots along a squall line would be tough, with a lot of cloud junk around.
Tuesday, February 26, 2013
2013 so far
The only real weather events here this winter have been a severe weather bust, an underperforming snowstorm, and a blizzard that missed OKC-Norman. The pattern gets a lot less interesting now, if you believe long-range models. This might be all until the first real spring outbreak, so here:
Wednesday, February 20, 2013
Tough forecast in OK
It's 12:30 Thursday morning. With another winter storm brushing the area, it's a hard call for forecasters along I-40 as far as rain/sleet/freezing rain. Snow looks likely in far northern OK, and rain is certain to the south. It's a classic problem that comes out at least once a winter in central Oklahoma, and one that requires a look through multiple layers of the troposphere. RAP forecast soundings show a large "warm nose" from ~700 mb to ~925 mb, with temperatures diving well below freezing around 950 mb, then rising back to within a few degrees of the freezing mark at the surface. NAM had been leaning this way since Wednesday morning in the OKC area. RAP suggests moderate ZR from OKC or Shawnee eastward along the I-40 corridor into Arkansas, extending northward through the northeastern part of the state--Tulsa certainly included.
Currently, Mesonet observations indicate lowest dewpoints in eastern Oklahoma--generally upper 20s. Get a little precip falling into that, and surface temps hovering at freezing will drop sufficiently for a significant icing event. Based on those observations, as well as numerical guidance, I might as well throw out a forecast:
OKC: Some freezing rain early morning with little accumulation. Any icing will be very short-lived as the surface warms.
Norman: Rain.
Locally significant freezing rain event in eastern/far northeastern OK, including:
Vinita
Miami
Jay
Tahlequah
Sallisaw
Stigler
Eufaula
Checotah
Okemah
All I've got time for. Back to math.
Currently, Mesonet observations indicate lowest dewpoints in eastern Oklahoma--generally upper 20s. Get a little precip falling into that, and surface temps hovering at freezing will drop sufficiently for a significant icing event. Based on those observations, as well as numerical guidance, I might as well throw out a forecast:
OKC: Some freezing rain early morning with little accumulation. Any icing will be very short-lived as the surface warms.
Norman: Rain.
Locally significant freezing rain event in eastern/far northeastern OK, including:
Vinita
Miami
Jay
Tahlequah
Sallisaw
Stigler
Eufaula
Checotah
Okemah
All I've got time for. Back to math.
Subscribe to:
Posts (Atom)