Raleigh vehicle preps, 'No Trip' outlier

What does Raleigh, North Carolina have to do with our storm chasing expedition out on the Plains? This week and next, I'll be in Raleigh to visit family and to start the annual pre-season re-configuration of the car. The past two seasons, Raleigh has been the site of the pre-chase expedition vehicle preparations. My grandmother's second home in the semi-rural southern outskirts of Raleigh has a large yard with plenty of space, making it ideal for spreading things out and doing any cleaning and installation work on the vehicle. Here in Charleston, such a task would involve parking on a street, with limited space and cars passing close by. In a previous post, I talked about some of the modifications that are planned to the car - so I won't repeat them here. I might get around to posting some photos of the vehicle work while I'm there.

No Trip?
I wanted to elaborate on that last little category on our departure date probability table - the somewhat obtrusive one that says 'No Trip'. There are a few reasons that might result in the total cancellation of our expedition. The first would be the most obvious - the weather's lack of cooperation. If no strong severe weather system occurs in the Plains during our standby period, we won't be heading west at all this season. Such completely barren seasons are rare, but are still possible. Last year was one such example, with no 'real' promising setups occuring during May and June. Last year, we elected to make the trip anyway, giving the benefit of the doubt to a weaker, short-lived pattern in late May. However, this year our standards will be high. We'll only make a trip, Lord willing, for a strong system that holds good promise for success. If 2007 is identical to 2006, there will be 'no trip' and we'll wait patiently for 2008.

The other reasons that could threaten our trip could be any logistical or financial contingencies that may arise. I've been known to press forward on a chase expedition when the bank accounts are low, making the investment in what I knew was good footage and pictures. Now that my stock video library is full of tornadoes, the financial justification for gathering more tornado imagery is less viable. Currently, I've got ample finances to make the trip, but a major car or house repair, delayed check from a client or other unforseen expense could tip the scales.

Despite all these possibilities, I always keep a strong optimism that the potential for an all-season no-go is very low. The chase expedition is a highlight of my year, and something I won't give up on too easily. As such, the 'no trip' probability will start out at 2% and remain that way throughout or standby period, barring any of the aforementioned problems.

On to the forecast outlook. Long-range models suggest a breakdown of the upcoming eastern trough and cold outbreak around mid-month, which might allow severe weather activity to ramp up once again out west after then. Mid-April is prior to our standby allocation and therefore still too early for us to plan a departure date, so our probability table for today will remain unchanged.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 2 April 21-30 10% May 1-10 22% May 11-20 22% May 21-31 22% June 1-15 22% No trip 2%

Raleigh vehicle work day #1: Tripod Compartment

Cooperative weather returned to Raleigh today, giving Matt (my brother) and I a full day to get started on the car (Why in Raleigh?). After removing literally every item from the car, including the accessory wiring, radios, and related chase gadgets, we gave it a good vacuuming and started the brainstorming session on how to improve the setup.

Cleaned and ready for work

When the rear seats are stowed, the back storage area platform overhangs about six inches up to the back of the second row seats. This leaves a roughly six-by-six cavity across the width of the interior behind the second row seats. We decided that this space will be ideal for the power console, and we'll be getting to that tomorrow.

The main project today, besides the initial cleaning of the car interior, was to construct one of the fruits of our brainstorm - a tripod compartment for the rear storage area. One of the biggest issues with past chase trips was space for fully-extended tripods. When you stop during a chase to set up your cameras, time is of the essence, and the task of extending tripod legs is often problematic. The tripods also need to be easily accessable, able to be deployed and re-stowed quickly. Our tripod compartment will solve these issues for us.

The tripod compartment is created by simply adding a 'second floor' to the rear storage area. For most SUVs and wagons, this can be built by nailing plywood to a set of 2x4 risers. On my car (Ford Freestyle), the rear storage/seating area features ledges that are conveniently suited to simple bridging, using a set of boards cut to fit:

Tripod compartment baseboards

We opted to use individual 1"x12" boards (with a small gap in between each) rather than a single sheet of plywood, which made installation easier. We purchased a 4'x6' piece of gray carpeting to serve as the surface of the 'second level'.

Installing the carpeting

We cut the carpet to fit the space, and used upholstery tacks to fasten it to the boards. We cut the carpet so that the ends could be wrapped around the front and the back of the boards. The carpet holds the boards together and keeps them from moving around. The carpet and gap in the boards allows the whole 'floor' to be folded up whenever it needs to be removed.

Matt and I, along with the dogs (Buffy and Beau) try out the finished tripod compartment.

The finished compartment has room for up to six extended tripods. The space on the second level has the same surface area as the original, which means we can use the same configuration for the rest of our gear as we have in the past.

Pelican case on the second level

The cables dangling from the bumper in the above photo are the security cables for the Pelican camera case. After we finished installing the tripod compartment, these cables were threaded through the frame of the car under the spare tire. I cut notches in the carpet to allow the ends of the cable to come up through the 'second floor', where they were threaded through the Pelican case and subsequently locked together:

Cutting notches for the security cable

Rain is in the forecast for tonight, so we're moving the chase vehicle setup operation to under the carport for tomorrow, when we'll start on rebuilding the electrical system console and wiring for the inverter and all 12-volt devices. Stay tuned for more photos!

Forecast update
We are less than two weeks away from the start of our semi-standby date of April 20, and exactly three weeks away from our 'normal' chase standby date of May 1. The GFS is portraying a large trough moving across the US next week, with severe storms possible starting this weekend. After that, the GFS shows the strong trough over the eastern USA, suggesting a downtime in the Plains for at least a few days after April 20. Remember, for our chase trip, we look for a strong 'dip' (trough) in the jet stream to set up in the western USA - the classic large-scale pattern for severe storms and tornadoes. Compared to 2006, the main difference we're seeing this season so far is that the patterns have been 'progressive', that is, they are moving along rather than stagnating for weeks on end. Which is a good thing! That would suggest that we'll need to be on the alert for another western trough possibly moving in at the tail end of April. But for now, that's nothing more than speculation. As a result, our departure date probability will remain unchanged as of today.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 10 April 21-30 10% May 1-10 22% May 11-20 22% May 21-31 22% June 1-15 22% No trip 2%

Raleigh vehicle work day #2: Power/communications rack

Last year, we mounted all of our radios, routers and power system to a custom 'center console' made out of two stackable wire desk shelves. The 'console' was placed in the space between the second row seats.

Tom working with the 2006 'console' setup

This year, we're trying to make this setup even more compact as well as moving it more 'out of the way' in the space behind the second row seats. This has been more of a challenge than we first thought. After some lengthy brainstorming, we decided to use part of the old console shelf as a 'rack' grid, mounted vertically at the back of the second seat row. The space is tight, which presents a problem in not only fitting everything back there, but providing adequate ventilation.

Thankfully, Matt had a bunch of spare 12-volt PC case fans that work perfectly, even running straight off of the 12 volt car power. We started the build-out by working on the literal puzzle of fitting everything onto the wire shelf.

Individual components

After a couple of hours of trial-and-error, we finally finished piecing the 'puzzle' together.

Final rack grid assembly

Since this rack is going in behind the second row seats, we left out the CB radio and scanner (which will be installed up front in a location yet to be determined). The inverter will rest on the floor immediately behind the rack, cooled with the help of the four fans. We drilled holes into the rear storage platform and used wire ties to secure the rack:

Rack installed

The top-fastened rack can be swiveled up for maintenance or repair when needed.

Thursday, we'll finish up the wiring from the front of the car to the new rack. Not only is it getting late tonight, but there are storms moving in to the Raleigh metro area from the west - and there is a severe thunderstorm watch in effect. We'll take a break to try for some lightning shots and get back to the car in the morning.

No updates today to the forecast outlook for our trip, so once again the probability table will remain unchanged.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 11 April 21-30 10% May 1-10 22% May 11-20 22% May 21-31 22% June 1-15 22% No trip 2%

Raleigh vehicle work day #3: Wiring

Last night's storms in Raleigh produced plenty of tower lightning discharges while we watched, but the cloud bases were so low that the tops of the towers (and the lightning channels) were completely obscured. We gave up on the storms around 4AM.

After the day of rain, perfect weather returned on Thursday for our third day of chase vehicle prep work.

Raleigh spring weather

Today's task was to complete the wiring of the newly installed power/communications console. We installed the console so that it can be swiveled up to gain access to the back side.

Power console, swung up in service position

Wires needed to be run back to the rear of the car for the lightbar/warning strobes, and the main power connection to the front dash terminal needed to be redone. I started by installing a terminal block for the rear lights.

Rear warning light terminal

Wiring from the rear terminal to the power console was next.

Completed console wiring

After the console connections were finished, it was flipped back down into its normal position.

Finished console with seats upright

Next was the new connection to the main power panel under the dashboard, and the installation of a main power switch for the console. Since we'd moved the console back out of reach of the driver, a switch was needed so that the inverter and cooling fans could be shut off from up front. This switch was installed next to the existing switch panel for the warning lights.

Main power switch (on the left)

Buffy supervising the vehicle work

Although there are still a few car-related items to be completed before our chase expedition, we've finished all of the work that we needed Raleigh's warm weather and space to complete. The fine tuning and additional items can be taken care of in the coming weeks. All in all, this week's improvements to the car ran around $50 - not too bad.

Forecast update
Late April is looking less and less favorable for an active severe weather pattern in the Plains, so we're reducing the chances of a departure during that time.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 12 April 21-30 6% May 1-10 23% May 11-20 23% May 21-31 23% June 1-15 23% No trip 2%

May 1, chase standby on the horizon

After a long winter, we've finally reached one of the dates we've been waiting for. May 1, the start of our annual chase standby period, is now within the window of the 14-day GFS forecast model. That means we can start our forecasting excercises in an attempt to narrow down the departure date for our chase expedition.

As we've talked about last year, computer forecast models are very unreliable beyond 5 to 7 days, and virtually useless past 14 days. Models often flip-flop wildly in the long term (and even in the short term), giving us a 'roller coaster ride' from optimism to hesitation and back several times. Any forecast we derive from the models beyond day 7, therefore, must keep this fact in mind. While the 7- to 16-day models can be unreliable, they nevertheless can give us a rough idea of what the large-scale patterns might end up doing. This is particularly true if the models maintain a consistent forecast day after day. And since the long-range models are all we have to plan with, we try to work with whatever information we can get from them.

So, with that in mind, let's take a look at what the GFS model is suggesting for May 1. Here is the GFS 500mb forecast for the first day of May:

14-day GFS 500mb forecast

This image does not suggest a western trough in progress or approaching for the first few days of May. On this map, two small troughs are positioned across the north-central and northeastern USA, with a ridge (the opposite of a trough) out west. If the GFS is correct, May 1-3 - and possibly several days later - will probably not be the time for us to leave.

At this point I want to take a minute and explain what the significance of the 500mb map is. Our main focus with the extended models is the large-scale pattern on the 500 millibar forecast image. 500mb refers to a level in the atmosphere some distance above the surface layers, sometimes referred to as the 'mid-levels'. The models plot the pressure contours and wind direction at this altitude. Because the 500mb pattern is usually the 'indicator' of the overall large-scale pattern and jet stream path, we (and all chasers) use the 500mb forecast map to look for the pattern that is most conducive for tornadoes on the Plains: the west coast trough. The west coast trough will show up as a big 'dip' over the west coast/Rocky mountains on the '500mb' map.

Why is a western trough important for tornadoes? First of all, surface winds ahead of a trough tend to be southerly, resulting in a 'conveyor belt' of wind that pulls healthy moisture off of the Gulf of Mexico and transports it northward into the Great Plains. This provides fuel for storms and helps sharpen the 'dryline' boundary with the hot, dry air in the desert southwest. Second, low pressure systems develop along the periphery of the trough and move out over the Plains, providing a focus for storm development and low-level wind shear. Third, 'jet streaks' of high winds in the mid- and upper-levels of the atmosphere ride around the trough and move either close to or directly over the good moisture and wind shear, providing additional shear and assisting storm development. A slow-moving western trough can result in several days of prime tornado conditions on the Plains. In a typical spring season, a series of troughs will move through, one every week or so - with several good chase days in a row followed by a few 'down' days between events.

The 7-day GFS forecast shows a moderate-sized western trough for the April 22-24 timeframe:

7-day GFS 500mb forecast

So why are we not excited about the April 22-24 trough? The reason is because the unseasonably cold weather we've had in the last week or so has reached all the way to the Gulf of Mexico, meaning that this new trough will not have as much moisture in place as one will later in May. Dewpoints over the Gulf are reading in the 40s and 50s now, due to cold frontal passages pushing the moisture far to the south. We would like to see Gulf dewpoints in the mid 70s as far north as the Texas and Louisiana coastline, which would mean there is plenty of good moisture that can be transported north. The April 22-24 trough, while looking somewhat good for tornadoes, will not be 'all it can be' due to the lacking moisture.

After April 24, the GFS shows a deep trough settling in over the eastern USA, suggesting little chance for severe storms on the Plains through the end of the month:

12-day GFS 500mb forecast

So, we've all but eliminated the last half of April for a departure date, mainly due to the low surface dewpoints over the Gulf of Mexico and the lack of a good western troughing pattern suggested by the models. The first few days of May, based on the current GFS run, may also not be a good time to leave. Of course, this can all change with future outlooks. But for now, we'll adjust our probability table to reflect the data we've just looked at, giving more weight to the later days of May.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 17 April 21-30 3% May 1-10 20% May 11-20 25% May 21-31 25% June 1-15 25% No trip 2%

Going on alert for the weekend

Now that the new Tuesday evening run of the GFS is out, I'm finding it harder to ignore the upcoming troughs forecast to impact the Plains this weekend and next week. Here is the new 500mb GFS forecast for Friday morning. By the way, the colors and format of these GFS model images look a little different from the ones I posted yesterday, because I'm using the UCAR model outputs rather then the NCEP versions. I like the UCAR presentation better, but their images only go out to day 7. To get images for days 7-14, I have to use the NCEP maps. Now look at the decent trough out west, the upper-air pattern we chasers always hope for:

Friday's GFS 500mb forecast

It is hard for any chaser to look at that and say "I am definately not going to chase". Now here is the trough moving over the Plains on Saturday evening, as portrayed by the GFS:

Saturday's GFS 500mb forecast

If it was mid-May and we were looking at images like this, we'd probably be leaving here headed west within 24 hours. But, as I mentioned yesterday, there are two factors that make me hesitate in driving what will end up being a 3,000 mile plus round trip to chase this setup. First is the storm motion, which should be seasonably fast. While one can intercept and document a tornado with fast storm motions (see our September 22, 2006 chase), their forward speed makes an intercept more difficult for a variety of reasons. Second, the moisture forecasts are good, but not ideal:

Saturday's GFS dewpoint forecast

That last image shows widespread dewpoints in the 50s across the Plains, with a broad, peculiar axis of 60s nosing in. As a side note, notice the extreme drop-off of dewpoints parallel to the Texas/New Mexico border - this is the dryline, an important meteorological feature for chasing. I'll talk more in detail about the dryline later. The 50s dewpoints shown here are half-decent, but not optimal for tornadoes. Again, if that image showed a more prominent pool of 60s right up against the dryline, I'd be more excited.

There are two kickers for the dewpoint issue, however. One is that the GFS has been erring on the side of caution recently in the realm of dewpoints. That is, the actual dewpoints have been slightly higher than what the GFS has forecast. The other issue is that on the high Plains of western Kansas and Texas, the high elevation (from 3,000 to 4,000 feet) means that storms can still accomplish great feats with the lower moisture.

The next trough forecast to impact the Plains comes on Tuesday:

Tuesday's GFS 500mb forecast

This trough is smaller, but more intense - and moisture is forecast to be much better with this second system:

Tuesday's GFS dewpoint forecast

The main problem I see with trough #2 is that it may end up being too far south in less-than-ideal chase terrain, mainly south of the Dallas-Fort Worth area.

So what do we do? I'm convinced that we should watch these upcoming setups very closely. I'm not ready to say we're thinking of going - yet. There are some caveats, but it looks to possibly offer a few good chase days for the money. Our funding is limited, however, and any trip we make now will shorten the time we can stay out later in May and June. If we use a chunk of our funds now, we'll be in a bind if the Plains really gets rocking in mid to late May. We'll just have to wait and see what the model runs in the coming days show us. For the probability table, I'll give some weight back to our pre-standby time block of April 21-30 to reflect this weekend's prospects.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 18 April 21-30 10% May 1-10 22% May 11-20 22% May 21-31 22% June 1-15 22% No trip 2%

Sticking to the plan

It appears we'll be sitting the upcoming late April events out, regardless of how good the setups end up being. All of us have scheduling issues that will make a trip out west this early difficult or impossible. So, we'll officially axe our 'semi-standby' period from April 21-30 from this year's plans, as after these two upcoming troughs, the models are agreeing that the Plains will get quiet through the end of this month.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 18 April 21-30 0% May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No trip 2%

Going on alert - again!

I've all but decided to leave here on Sunday to chase a three-day setup that looks to be taking shape for Monday-Wednesday on the Plains. How's that for a change of plans? I knew better than to give April 21-30 a 0% percent on that last probability table. This trip, however, will not be our main 'chase expedition' but rather a solo one by me. If I end up chasing next week, it will mean I'll be doing two separate trips out west - this one, followed by our main trip in May.

The outlook for tornadoes has improved for Monday-Wednesday, with a strong western trough, good moisture, and a surface low pressure system all in a good spot for a decent three-day event. It's too early to nail down the finer details, but it appears the synoptics (IE, the large-scale features) will be very favorable. My premilinary target for Monday's chase is Amarillo, Texas, followed by Oklahoma City on Tuesday and central Arkansas on Wednesday. all of those will most likely change as the days get closer.

Now, on to May. A good-sized trough is showing up on the GFS for May 6:

May 6 GFS 500mb forecast

That's a good sign and something we'll be monitoring very closely for our main chase expedition. As I said before, models that far out are notorious for being flaky and flip-flopping, but what we can glean from this simply that we'll need to keep on watch.

Our probability table for this season is taking a beating already. I guess that's a good thing, since it means that things are getting active. Last year, our table was pretty consistent because we had such a long period of obvious downtime on the models (which verified). Now it's all over the place. Add to the fact that we're now looking at two separate trips, and I'm not even sure how to represent everything! For now, I'll show the main chase expedition with the red shading, and next week's prospects with blue and a separate percentage. That should cover it - for now.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 20 Red - main expedition; Blue - April 22-30 April 22-30 80% May 1-10 30% May 11-20 23% May 21-31 23% June 1-15 23% No trip 1%

April 23, 24, 25 chase trip a go

We just made Matt's flight reservation from Raleigh to Oklahoma City on Sunday evening after he leaves his work. This means we're pretty much committed to this trip. I may be leaving here as early as tomorrow afternoon (Saturday) to make it a more leisurely drive to get to OKC in time for the 10PM flight arrival.

The GFS is waffling again on early May, so we're back to equal percentages for our main chase expedition.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 20 Red - main expedition; Blue - April 22-30 April 22-30 99% No April trip 1% May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No main trip 2%

Last-minute preps, tornadoes already happening

As you can see, this blog is kicking in to high gear as things get going. I'll likely have 1 to 3 updates per day here during our trip, as time and internet access permit. The potential for this week seems to really be coming together. Tornadoes have already been reported in Nebraska on Friday, with more expected today in the Texas panhandle. The setups for Monday-Wednesday look very encouraging and we are optimistic. The SPC's Day 3 outlook for Monday is now out, with a nice hatched area down the dryline.

SPC outlook for Monday

Lord willing, t-minus 7 hours or so before I get on the road heading west. On a normal chase trip from West Virginia, I usually don't leave until the 12z model runs come out about noon or so and the SPC issues their new Day 2 outlook (for the following day) at 1:30PM. The new model runs and SPC outlook are the last new data for the next day's prospects I can get, and still make the target on time with a short night's sleep on the way. In other words, normally I'd be waiting until 1:30PM on Sunday to head west for Monday's chase. That way, I have every chance to back out if the chances for tornadoes decrease. But with a typical Plains chase target, in order to make it in time, it's a non-stop drive from WV that leaves me about 5 hours to sleep somewhere in central Missouri. Waiting for the 00z model runs (which come out in the late evening) would leave me no time to make it, even if I drove straight through with no sleep.

This time, with Matt's flight reservation made and paid for, I'm committed to making the chase at this point. The downside to that is that if the outlook for tornadoes diminishes at the last minute, we can't back out. The upside, though, is that I don't have to wait until the last possible minute to leave here. That translates to a much easier drive, and time for full night's sleeps before the chase days begin. I can enjoy the drive, under the speed limit, stopping for limited sightseeing and photography here and there. No pressure to get somewhere 1000 miles away by the next morning. Consequently, I'm planning on heading out today as early as possible, and aiming for somewhere from St. Louis to Joplin for the first night's stay. After church, it's another leisurely drive to Oklahoma City, where I won't need to be until 10PM to pick Matt up from Will Rogers airport. More than plenty of time barring any problems. There is a small risk for severe storms in southern Missouri and northeast Oklahoma on Sunday afternoon, which I may have time to check out. After I pick up Matt in OKC, we'll drive a couple more hours west on I-40 (Elk City or thereabouts) to be in position to make any target in the Texas panhandle, western Kansas or southeast Colorado on Monday.

Forecast and target
The latest NAM/WRF model output shifts Monday's risk slightly to the east, centered on the Texas/Oklahoma border. However, our preliminary rough target will remain Amarillo, Texas. If the risk shifts back west or north (as is still possible this far out), shooting for Amarillo will leave us some time to adjust accordingly. There is some concern now, based on the latest NAM output, that the surface low and better upper dynamics might remain too far west of the moisture and instability. However, this would simply mean tornadoes happening more toward dark rather than in the afternoon. Tornadoes harder to see and photograph, but tornadoes nonetheless.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 21 Red - main expedition; Blue - April 22-30 April 22-30 99.9% No April trip 0.1% May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No main trip 2%

DAY 1: Charleston to St. Louis

I'm sitting in my car in southeast St. Louis typing this, so my journalistic eloquence (if I had any to begin with) is going to suffer. After a few more hours of sleep this morning, I finally was on the road westbound before noon. Perfect travel weather with crystal clear skies. I was surprised to see Kentucky's, Indiana's and Illinois' foliage wasn't any farther along than ours. I passed Lexington around 2:30PM. (My digital camera decided to stop working today, so all of the shots on this post are still frames from the Sony FX1.)

Lexington water tank mural along I-64

I stopped for a ginger ale at a gas station in Shelbyville, KY. The clerk asked me if I was going to 'Thunder'. Slightly puzzled, I said 'No, Oklahoma!'. I soon found out what 'Thunder' was. Outside Louisville, traffic backed up enough for me to be able to see planes flying in formation up ahead in the distance. As I passed downtown, it was apparent a huge event was taking place, as every inch of waterfront was packed with people.

Huge crowds in Louisville

I have never seen a festival of this scale anywhere else. Not only was downtown and the waterfront overtaken with people, but the residential neighborhoods outside of the city were packed as well, with hundreds of people sitting in lawn chairs in their yards, most with big party tents and food grilling. This event, I found out, is called "Thunder over Louisville" and is a day-long airshow, literally right over downtown, featuring military aircraft and a massive fireworks display afterwards. I stopped for an hour in one of the residential areas (getting downtown was out of the question) and shot some video. The sights and sounds were really spectacular, and I wish I had time to stay longer.

I stopped again west of Evansville, Indiana to shoot some of the landscape.

Rural Indiana road

I love seeing this sign

As the sun sank lower, I made one more stop near Belle Rive, Illinois to watch the sun set on the southern Illinois prairie (just as flat as western Kansas).

The car's shadow at sunset

Southern Illinois sunset

I reached St. Louis just after 9PM, central time. The arch is another great sight for an eastern US chaser - meaning the Plains aren't that much farther away.

St. Louis - a welcome sight

I'm heading down I-44 for a couple more hours before finding a place to stop for the night. From there, I've got all day to make the remaining 7 hours to Oklahoma City.

Our probability table will go on hiatus while we're traveling, since we won't have time to do much long-range forecasting. Forecasting in the short term while driving all day is enough to keep us busy.

DAY 2: St. Louis to Oklahoma City

Wow, Monday and Tuesday's tornado outlooks look great. More on that later! After a short night's stay near Springfield, I was back on I-44 westbound by 7AM. Most of the area along I-44 in southwestern Missouri was still showing the heavy damage from the winter's ice storms. Every tree had at least minor damage, and in some spots the destruction was almost indistinguishable from a tornado damage path. It looked as if a giant lawn mower had lopped off the tops of all the trees everywhere.

Springfield, MO ice storm damage

Damaged treetops in Springfield

I crossed the Oklahoma state line just before 9AM.

Crossing into Oklahoma

Now take a look at that last image (and the Missouri tree damage photos, for that matter). Notice that most of the trees have not started budding and leafing yet. Spring 'greenage' seemed to drop off to near nothing along the Missouri/Oklahoma line. In fact, of everywhere I saw in daylight the past two days, Charleston back home had more green than anywhere in Kentucky, Indiana, Illinois and Missouri. That fact surprised me, but not as much as what I saw as I drove further into Oklahoma.

Here is the giant McDonald's near Vinita, OK that spans the Interstate. This is supposed to be the biggest McDonald's in the world.

Vinita McDonald's on I-44

Now look closely in the distance in that last image. You see that on the horizon? Green! Not just spotty green, but fully leafed trees. In a short distance near Vinita, the trees suddenly went from barren to near-fully leafed, and it continues that way from that point westward. This is puzzling, as there is no dramatic change in elevation, climate or tree type between Joplin and Tulsa. The only explanation I can figure is that this region has been in drought for some time, and recent heavy rains may have been localized enough in Oklahoma that the drought has been alleviated there - with a sharp cutoff of rainfall totals in northeast Oklahoma.

Here you can see the state of the trees in Claremore as a train rolls through town.

Union Pacific train in Claremore

This train rolled right by where I ended up going to worship services, at First Baptist Church of Claremore. After church, it was on to Tulsa.

Tulsa skyline in the distance

Getting ready to head for OKC

I drove through a marginal cumulus field between Tulsa and Stroud, but opted to ignore the slight threat for storms, as they would likely take me too far away from our lodging point tonight. I arrived in OKC around 3:30PM and am currently at a hotel on Meridian west of town.

Finally, Oklahoma City

Matt's plane is scheduled to arrive at 10:08PM. I'm only a few minutes from the airport, so I've got some time to relax and prepare for the next two days, which, from a forecasting standpoint, look to be banner tornado setups.

DAY 3: Target in the Texas panhandle

We are currently in Canadian, Texas checking data, after leaving Oklahoma City early this morning. Supercells and tornadoes are expected over this area later in the afternoon. Today should be a good chase day providing we position ourselves correctly. Tomorrow also looks like a huge outbreak scenario over Oklahoma and Kansas. Stay tuned for updates!

DAY 3: Three tornadoes in southwest Kansas

Blessed with success! We caught three tornadoes today near Protection, Kansas. A full report and log will follow soon, but for now here are a few photos:

DAY 4: Challenging day in central Kansas

We're currently in Wichita, Kansas for the night, after starting the day in Pratt and covering ground from Great Bend, Hays, Russell, Ellsworth to McPherson. Tuesday's setup originally promised a major outbreak of tornadoes, but very early on, the day's prospects diminished considerably as a large line of storms fired from Texas to Nebraska. Nonetheless, one storm managed to go tornadic on the dryline near Hutchinson, Kansas. Where were we? Shooting a bright rainbow and a massive 'whales mouth' cloud formation 20 miles away. We had given up on the day 30 minutes earlier, after a long, hard day of driving and waiting in vain for storms to fire and then for them to produce. A loose USB connection had just caused my WxWorx data stream to quit, and I didn't bother fixing it because I felt the day was over. So, we turned our attention to the other sights the storms had to offer.

Rainbow and pump jacks

'Whale's mouth' behind gust front

I saw the tornado images that just about everyone else captured today. While I would have loved to have bagged the same thing, I can't say I'm hurting too much after what we were blessed with yesterday. My mistake today was pretty clear-cut. I let the mental exhaustion of a long, unproductive chase day influence my decision making - rather than sticking to time-tested chasing tenets (like not giving up it out until the day is over). A lot of chasing is sticking to the realities of what is happening rather than what you're feeling. I knew that the day shouldn't be over until the last tail-end storm dies, but the trends of the day got the best of me. That, and the lack of radar to alert me to what was happening to the south, cost us the tornado(es). Nonetheless, I am happy with the images that we captured in leu of the tornado event.

I'll write more about Tuesday's (and Monday's) chase later on when I've got more time.

Tomorrow, we've got one more Great Plains chase day in southeast Kansas/northeast Oklahoma that will conveniently take us a little further east toward home. After that, it looks like we'll have a possible lightning chase from Indiana through Kentucky on Thursday. We'll arrive in Raleigh late Thursday or Friday, when yet another storm photography chance will be had from the same system we will have been following across the country. After that, I'm heading back to West Virginia, at which point I'll return to long-range forecasting mode for our main chase expedition in May.

More tornado images from Monday
Matt has gone through some of his digital photos (Canon 5D) of the Protection, Kansas tornadoes, and has posted a few of them on his web site at raleighskyline.com.

Protection, Kansas tornado #2, by Matt Robinson

Protection, Kansas tornado #3, by Matt Robinson

Protection, Kansas supercell at dusk, by Matt Robinson

DAY 5: Low-topped supercell in southwest Missouri

I'm typing this from the car next to Busch Stadium in downtown St. Louis while Matt shoots some skyline photos. After leaving Wichita, we played around with some cold-core storms in southeast Kansas and northeast Oklahoma today, covering ground from Sedan, KS to Bartlesville, OK. The activity was not showing much promise after a few hours, so we decided to begin the trek back east toward home.

After passing through a developing cell's precip core on I-44 at Springfield, I looked back to see this, and promptly exited the highway:

Rain-free base at Springfield, MO

We followed this storm north for the next two hours. At first, we drove right up under the area where the wall cloud was trying to condense. As we approached, cool west winds suddenly turned to southeasterly inflow. We watched for a while longer, when a large RFD slot developed and the lowering grew larger. Rising motion was modest but persistent. The base took on a rounded appearance with a horseshoe shape.

After a few minutes, the small wall cloud's rotation and rising motion increased, and condensation began reaching downward further. The huge RFD cut was very impressive. The storm was tiny with meager inflow at the surface, but nonetheless, at this point the overall structure was getting me excited.

A few times, a pointy tail cloud condensed diagonally toward the ground, making the lowering a classic 'fakenado'. So much so I wondered how long it would take for someone to report it as a tornado/funnel cloud.

Here is the wall cloud at its lowest point. Persistent and clear rotation was noted here. If this had gone on for another 30 seconds, I would have called it in.

We saw only one other chase vehicle on this cell. A nice catch for a day that initially promised little.

We're headed to Evansville, Indiana for tonight's stay. A marginal but widespread severe weather risk will be present for the continuance of our eastward trip tomorrow through Kentucky and yes, West Virginia. We'll likely stay at my house in Charleston tomorrow night, and make the final trip to Raleigh to get Matt back home.

DAY 6: Lexington, Kentucky chase

We observed storms around the Lexington, Kentucky area on Thursday, a convenient target on our way home from our first Plains chase trip. We watched a cell develop just northwest of town that quickly condensed a nicely-shaped wall cloud complete with inflow tail. The storm was racing to the NE, and with few road options to intercept, we let it go in favor of more development moving toward us from the SW. We watched the original storm move north with rock-hard convection punching through pileus caps. The structure down low looked great and I wish we could have stayed with it - but 40mph NE storm motions and Kentucky roads do not mix. This storm soon began hooking in on radar with long-lived rotation markers on both WxWorx and GRLevel3, and was tornado warned for at least 2 hours as it moved into Ohio.

We hung around Lexington to wait for the action from the south to move in, which quickly lined out before arriving. We stayed ahead of the storms east along I-64 all the way to the WV state line. Despite intermittent tornado and severe warnings for this line, we saw nothing visually to get excited about. A shelf cloud and a few nice CGs were the only items of note.

Near Morehead, Matt got a decent lightning strike on video that produced a bright orange glow on the mountainside where it hit. I'll post that and a few other shots sometime soon.

We just spent the night at my house in Charleston, and we'll be leaving here in a few minutes for Raleigh so Matt can return home.

Forecast update
The 2007 season is in stark contrast to 2006. Last year, we were stuck in an extended dead severe weather pattern, watching for any glimmer of hope on the models. This year, the start of May is looking so active that it looks like our problem will be picking the best of a number of good chase setups. We've already bagged three tornadoes, and our normal standby period hasn't even arrived! Since we're still technically in 'travel mode', I haven't had time to look at the models in detail. What I do see is a major trough setting up out to the west, which could mean an active severe weather pattern sometime during the first ten days of May. While this is encouraging, it's a little too early to go back out after we've just arrived home. I'd prefer to wait a little longer if neccessary, but we'll keep a close eye on the upcoming system nonetheless. All that said, I'll stick with equal probabilities to all weeks in our upcoming standby period.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 27 May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No trip 2%

Plains Trip #1 summary, April 23 log

I'm currently in Raleigh for a few days before heading back to Charleston. In the meantime, here are the stats for trip #1:

States: 8
Chase days: 4
Tornadoes: 4 or 5
Miles: 3,100
Fuel cost: $384.96
Motel cost $200.06
Plane ticket: $280.00
Oil Change: $20.00
Total cost: $885.02

The total cost came in much lower than I expected. My original estimates of the fuel costs were way too high, and we were able to find satisfactory hotels for around $50 a night. Not counting the plane ticket for Matt, the whole 6-day trip came in at about $600, not bad for a trip of that length.

Chase reports for trip #1
The completed log for the April 23 tornadoes, along with photos and video, can be viewed at the April 23 report page. Reviewing video confirms we had three tornadoes on the ground at the same time! I'm still working on the logs and photos for the rest of the chases.

Forecast update

GFS 500mb forecast for May 5

A nice-looking trough - just as nice as this last one - is showing up on the GFS for the May 3 to 6 period. After that, a large trough appears to set up on the west coast for an extended time. Currently the later trough appears too far west to bring significant severe weather to the Plains, but as we're talking many days in advance, this can easily change.

This is looking like one of those seasons where there will be so many good setups that we will have no choice but to miss a few of them. Ideally, I'd love to just take the month off and stay in Kansas or Oklahoma for the entire month. But, there's no way we can afford that this year. It is going to be difficult for us to regroup and go back out west for another week so soon after getting back from our last trip. We may have to sit this upcoming series of events out unless it looks so huge that it is a 'must-be-there' situation. Although we've still got the entire month of May to go, sometimes the good part of the season happens early, followed by nothing else the rest of the month (like in 2003).

For these reasons, I'm going to remain with equal departure date probabilities for all days in our standby period, until the upcoming system can be looked at in greater detail as the events get closer.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 28 May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No trip 2%

Forecast update for April 29

The past few models runs have been trending keeping the upcoming western trough to the northern teir of US states. This suggests severe weather possible in the northern Plains during the May 5-8 period.

GFS 500mb forecast for May 5

The dryline looks to be active during this time in the southern Plains, but if the trough goes as far north as forecast, upper support will be lacking for the dryline down south. After May 8 or so, the models suggest a more zonal pattern (no real troughs or ridges across the USA). Zonal flow is better than ridging, but not that great for severe storms.

As a result of this information, we'll reduce the chances for a May 1-10 departure.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 29 May 1-10 17% May 11-20 27% May 21-31 27% June 1-15 27% No trip 2%

May 5-10 looking good

And here comes May!

The western trough showing up for later this weekend is looking better with each model run. Southwest Kansas looks to be in the sweet spot for Sunday per the latest GFS forecast as of 00z today. (00z is 00:00 UTC, or Coordinated Universal Time).

Take a look at these three model output images for Sunday afternoon. Again, we're talking six days out, so this may change a little. But you can't ask for a better-looking picture here for chasers. First of all, the front end of a strong 500mb trough is moving over the Plains, providing excellent mid and upper-level wind fields for supercells.

GFS 500mb forecast for Sunday, May 6

Second, abundant moisture, pulled north from the Gulf of Mexico, should be in place providing ample fuel for storms and helping the dryline become well-defined. Look at the widespread 65F dewpoints and large areas of 70F dewpoints over the Plains, as well as the sharp dryline in western OK, KS and the Texas panhandle.

GFS dewpoint forecast for Sunday, May 6

Last but not least, a nice surface low pressure should be in place across the high Plains, providing low-level shear and backed surface winds, all juxtaposed with the mid-and upper-level winds and good low-level moisture. A classic tornado setup!

GFS surface forecast for Sunday, May 6

In chasing, there's always a challenge. Despite this upcoming system being within our long-planned for May standby period, some of us still have work/scheduling issues that may interfere with our ability to make this series of chase days. Tornadoes look like a definite lock for Sunday through Tuesday, maybe for more days afterward.

As a result, I may look at going on yet another 'pre-expedition' chase trip like the one we just returned from, meaning that we'll end up having three trips out west rather than two. This is similar to 2004, when we had multiple active patterns that necessitated three round trips to the Plains to catch all the action. For now, I'll give May 1-10 back the probabilites I removed earlier, to reflect the possibility that all of us will end up being able to make this upcoming event.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of April 30 May 1-10 24.5% May 11-20 24.5% May 21-31 24.5% June 1-15 24.5% No trip 2%

"It's here!" and possible May 4-8 chase trip

May 1 is a date that has the same significance to chasers as December 25 does to kids, the start of arguably the greatest of all months (cue Andy Williams' "It's The Most Wonderful Time Of The Year"). Now that it's here, our 'official' standby period begins as the classic chase season kicks in to high gear.

So now that May is here, what's the plan? The models indicate a nice, slow-moving trough bringing setups for severe storms and tornadoes to the Great Plains from Friday through Tuesday. The potential show starts in the High Plains of western Kansas on Friday and Saturday, then moves east on Sunday-Tuesday as the trough advances. It appears like there is potential for four or five solid chase days in a row with this system, if the models' forecasts come to pass.

Now, the bad news. Unfortunately, our crew is not going to be able to make it to this one. Between the four of us, each of us have work and/or family obligations that will prevent us from making a trip this week. However, while Matt, Tom and Bill are tied down this weekend, my obligation is a meeting with a client scheduled for Thursday morning. This means I can still make it out west for Friday's setup, Lord willing, if I left immediately afterwards. This would be a solo trip for me.

Since May 1-10 looks to be an active period in which I may make a second trip out west, we'll break the probability table into two segments again - one for the upcoming week, the other for our main expedition.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date:

2007 Chase Expedition - Departure Date Probability as of May 1 Red - main expedition; Blue - May 4-10 trip May 4-10 65% No May 4-10 trip 35% May 11-20 33% May 21-31 33% June 1-15 33% No trip 1%

Trip #2 a go, for now

SPC outlook for Friday

Tonight's 00z model runs are now out, and both the GFS and NAM/WRF are still showing precipitation in central and northern Kansas on both Friday and Saturday evenings. SPC's new Day 2 outlook (above) reflects the favorable setup. Precip showing up in the forecasted high-instability, moderately-capped, high-shear environment means that supercells are likely. Since I have the luxury this time of waiting until the last possible minute (1:30PM EDT) for the SPC's afternoon Day 2 outlook and the morning's 12z model runs, I'll hold off on making an absolute decision about leaving until early afternoon. Nonetheless, at this point I'm about 90% sure I'm heading west after my meeting at work.

Western Kansas is a long haul from here, so I might go ahead and leave immediately after my meeting just to get a jump on the drive. I need to make at least Salina, Kansas, which is about 16 hours away. That gives me some time for rest without too much driving in the morning.

Based on the current outlook, this probability table charts the chance of our trip starting on a particular date: