Monday, June 8, 2009

VORTEX2 finally intercepts a tornado


After a long hiatus, we have a new post. Rachel Humphrey, a graduate student at the University of Colorado has written today's post.

I will add a pre-amble. The VORTEX2 fleet made tornado science history a couple days ago. We intercepted a tornadic supercell for nearly an hour, deploying mobile radars, sticknets, disdrometers, mobile mesonets, photogrammetry teams, and in situ tornado pods in an amazing integrated array. Integrated data were collected from about 20 minutes before the tornado formed, through its birth, and almost until it died. Never before has a tornado been studied in nearly as much detail.

Rachel's post:

Subject: It only took about five weeks...

...but today, V2 finally intercepted its first tornadic storm of its 2009 operations.

Our morning started off in Sterling, CO, and we left the hotel around 11:30ish, making our way north into Wyoming (in a similar location to where we chased yesterday, near the border of Wyoming and Nebraska) and opting to neglect the possibility of storm initiation further to the south. We targeted our first storm around 4:00pm local time, and headed towards the area of intercept in southeastern Wyoming. The environmental conditions were conducive to supercellular development (not amazingly great, but certainly better than what we had seen days prior), as there was sufficient shear and moisture present.

Once it became apparent that the targeted storm was more likely to become supercellular (and possibly tornadic) than the other cells in the region (there were two cells to the north of our storm, the topmost moving northeast, the second one down the line moving almost due east, and our storm moving east-southeast), the crew began to deploy their various instruments. The radars all got into position so there would be plenty of overlap for their radar lobes. The Probe vehicles began to get into position to begin their mesonet transects of the storm (as well as the eventual dropping of the PODs). The disdrometer teams began to scout out locations for good deployment sites of our own.

One of the main obstacles to overcome in our deployments was the lack of visibility due to the presence of large bluffs in the local topography. The radars had to find high terrain on which to scan, so as to not interrupt their beams. The Probe teams had to find terrain on which it was safe to deploy and where the PODs would be level and yet also exposed to the winds of whatever was coming their way. For the disdrometers, we had to make sure that we had good radar coverage as well as exposure to the elements. The bluffs were pretty to look at, but they definitely hampered all of our efforts to do that. However, we got it done - all of us - and once everyone put everything where it had to be, we were treated to the sight of a very large tonado that lasted (on the ground) over 20 minutes, snaking silently towards us (we were positioned near DOW 7 on US-85) while we watched in awe.

It's like a cliche scene from a movie: all the practice that we've been doing, all the dry runs, all the pseudo-deployments carried out in non-severe conditions all paid off today, as once the tornado formed, we were able to watch it from a safe distance while it came near (and, in some cases, crossed over) our instruments. Today's mission was a total success. Everyone's spirits were lifted (seriously, it was getting pretty snarky around these here parts), and all the data has been taken off the instruments already (everyone's eager to see what was obtained - after all, this is the biggest instrumented case so far this year!)

After the tornado roped out, we began to head down US-85 to connect up with I-80, with the intention of calling it a day and heading to Kearney, NE for the night (about 4 hours away from wehre we had just watched this amazing storm). However, while we were driving along, a cell to the north of us began to rapidly intensify, and showed signs of rapid rotation beneath its base. As a matter of fact, at times it looked like the entire mesocyclone had simply extended down to the ground! We (CU) headed north out of Chappell, hoping to rendezvous with DOW 6 to get some radar coverage as the core came over us, but that plan didn't exactly work out. For one thing, visibility was really strange beneath that storm, and it wasn't entirely sure that it would be safe for us to deploy in the core. So, we didn't. Though the storm itself was tornado warned, we never got confirmation that it actually produced a tornado, but we couldn't really be sure from where we were...We weren't in a !
position to deploy on it, and since it was becoming dark, the operations were ended for the day. However, on our ride back (we were on the second storm for over an hour, so our departure from the Chappell region was pretty delayed) along I-80, we at times were literally surrounded on three sides with storms which were either tornado- or severe-warned...It made for a very long, very slow, very intense, but very exciting ride back to Kearney tonight/this morning.

Speaking of which, I should wrap this up and get on with the pictures, since today is supposed to be active, as well - barring any wiping out of regional instability as the result of the propagation of an MCS through most of the state of Nebraska...

Wednesday, May 27, 2009

Backwards Storm Intercept

We intercepted the most interesting storm to date yesterday, in North Texas. For a while, a strong storm was approaching the Dallas/Fort Worth area. It was in 'bad' air, so we waited to see if it would cross Dallas and become a supercell when it emerged to the east. Instead it surprised us and split, and even more unusually, the right moving split died and the left moving split moved straight to the north. While 'left movers' are not VORTEX2's typical target type, we scrambled back to Slidell, Texas to get ahead of the storm. It had begun to rotate, and even had a feature that looked like a hook. A left moving storm like this is very difficult to intercept. We all had to put on our 'mirror glasses' since the storm structure was expected to be a mirror image of a normal supercell. It really taxed our adaptability, but the well practiced VORTEX2 scientists did great.

The storm made golfball sized hail and strong winds. One of the DOWs, the high-tech 6-beam Rapid-Scan DOW measured differences in winds across the rotation of 60 m/s (150 mph), which is quite intense. But the circulation was aloft and cut off from the ground by a cold outflow. There was not much chance of a tornado. The NSSL mesonets, the CSWR mesonets, and the TTU Sticknets all deployed in a box between the 6 X-band radars deployed close to the storm, so we collected very good data.

Eventually the storm morphed again, and a cyclonic, or normally rotating circulation developed. We collected radar data in that also.

It was our most intensively studied storm to date.

Today is a down day, so we are working on repairs, traveling north to our next target area, which may be in Nebraska or Iowa as soon as Friday, and doing our laundry.

Friday, May 22, 2009

Surprise supercell intercept

VORTEX2 has been conducting missions in Nebraska for the past few days. It has been pretty disappointing most days. The Plains continue to be pretty quiet. Strong jet stream winds, frequently necessary to provide a source of rotation for supercell thunderstorms, and tornadoes, are flowing only in the far northern tier of states, in North Dakota and South Dakota. Hot moist air, necessary to provide the energy for thunderstorms to form in the first place, has been confined to the far south. So, there have been few places where the conditions for rotating thunderstorms have overlapped.

One of those places was Alliance, Nebraska on Wednesday. VORTEX2 started in Ogallala, Nebraska, and went north to get under the edge of the jet stream. We spent much of the day in a nice town, Hemingford, Nebraska, waiting for storms to develop. We met a lot of the townspeople and even gave a tour to an elementary school class.

In mid afternoon storms began to percolate west of town. Then the chase began. We debated whether to drive north or southeast since there were storms in both areas. The choice was not very clear, and there were a variety of opinions. Both choices had merit. The southern storm was stronger, but the northern storm was in an environment more likely to cause it to become a supercell. It was my turn to be 'mission scientist' who serves as a tie breaker when the 'Steering Committee' of VORTEX2, can't reach a unanimous decision. So, I said, let's go southeast after the 'bird in the hand', the stronger storm. We stayed ahead of that storm for about an hour, and considered deploying the huge VORTEX2 armada around it. But, it never matured into a supercell, so was very unlikely to make a tornado. Would it morph into a supercell when it hit a cold front? Would it never change? Again, we did not know and it fell to me to break the tie by deciding to give up and head back west.

As we headed back to Alliance, Nebraska, most of the Steering Committee, including me, were not optmistic. We were discussing whether to wait for a while, or just to give up and let our crews get dinner at some reasonable hour (at real, non-fast-food, restaurants which are often closed when we roll into towns at 10pm). But, one of our group was more optimistic and recommended that we try a little longer. He sure was right. A weak storm near Alliance quickly got organized. It became a supercell. And supercells are what make tornadoes? We quickly deployed our radars, Sticknets were dropped, mobile mesonets drove back and forth. We finally had a mission, an intercept, on a viable storm, an interesting supercell.

The storm had a hook echo and rotation. One team observed a gustnado, which is a small, tornado-like spin up below a storm. VORTEX2 collected data on the storm for about an hour before it moved away from any paved roads. It never made a tornado. But, finally, we had intercepted a 'real' storm. We were very glad that we had driven all the way to northwest Nebraska, and that we had been tenacious and had not given up. It was also nice to see how the project leaders worked as a group, even when we had different opinions. As it turns out, neither the original choices of north or south were correct, a storm in the middle that all but one of us, including me, wanted to give up on, became the supercell, and we were there.

Today we're moved even more northward, to South Dakota. Chances are only slight for a supercell. But, like we saw on Wednesday, the chances are not zero. Even if chances are slight, VORTEX2 needs to be in the right place at the right time if another surprise happens this afternoon.

Tuesday, May 19, 2009

VORTEX2 Goes Way North

VORTEX2 has moved the massive fleet way north. Currently we are stationed in western Nebraska. We are poised to go more northwest in pursuit of severe weather. The problem for VORTEX2 (though not for the High Plains) is that the atmosphere has grown quiet. The jet stream is way up along the USA/Canada border. A storm in the Gulf of Mexico is preventing warm soupy air from flowing north into the Plains. The combination means hot dry conditions through tornado alley, and not many pickings for VORTEX2.

But, there are always chances. Today we are looking at slim chances in Nebraska. Tomorrow holds the possibility of operations in Nebraska, Wyoming or Colorado. The chances are slim, but we're out here anyway, waiting, so slim is better than none.

These types of lulls are par for the course in tornado research. Active periods alternate with periods of relative calm. During this time we practice on whatever storms we can find. With our uniquely numerous and diverse set of instruments we can learn valuable things about non-tornadic storms also.

Sunday, May 17, 2009

A few days of being stationary for VORTEX2

The skies over the High Plains have quieted for a while. After an excellent mission targeting a squall line in Oklahoma, VORTEX2 is now in a holding pattern waiting for the weather to become active again.

The squall line mission really tested out our equipment and tactics. We set up a line of 5 mobile radars (a first), both Sticknet lines in the dual-Doppler lobes (a first), and had 14 Mobile Mesonet vehicles driving back and forth across a pre-storm boundary and the storm itself. Disdrometers were deployed, and weather balloons were launched ahead of the storms and in the cold pool behind, we even dropped our array of tornado pods ahead of the line's gust front. Even though the line did not produce a tornado in our study area, the mission was very satisfying because we really succeeded in our choreography around the storms.

Now we have a few days with no missions. But, VORTEX2 is not at rest. CSWR is working hard on software and hardware modifications so that our radars work better. So are all the other groups. The Rapid-Scan DOW's generator is still stalling sometimes, so we're trying to shield it better from road splash. One of the DOWs went to Oklahoma City to get its hydraulics fixed. On Monday it will have yet another diagnosis of its drive shaft vibrations. We have a brand new International Truck that has had major problems, and we hope that International finally replaces our bent and vibrating shaft. And, one of the Tornado Pod's T/RH (temperature/relative-humidity) sensors got bent during travel yesterday.

We are gathering together the data we collected yesterday so that we can show it to and share it with other VORTEX2 scientists. Even though there was no tornado, we need to see how well the radars performed, how well the Sticks and Pods and Balloons, etc. did their jobs, whether they were directed to the best places at the best times, whether the data were recorded correctly and were calibrated.

Also, crews have to take care of personal tasks that are nearly impossible when we're nomadic, pulling into hotels late at night and leaving the next morning, things like laundry.

Thursday, May 14, 2009

No gold at the end of the rainbow yet


VORTEX2 conducted missions on Tuesday and Wednesday...but we didn't intercept any tornadoes. On both days we intercepted very interesting weather. Wednesday featured a line of storms which contained a couple supercells. One made a tornado, but VORTEX2 had chosen a different storm. We got a chance to test tornado pods on Tuesday and the Mobile Mesonets had a good mission on Wednesday. We deployed several of the radars on the storm, but not in the coordinated way we had written in our operations plan. The biggest problem was that the weather was very difficult with the target area moving in jumps from place to place (we call this discrete propagation. Also, the storms were moving in a very unfavorable direction compared to their shape, so radars had to run away from the hail cores very shortly after they parked.

Today we are 'down', which means we are working on various systems. We fixed a broken hydraulic foot on DOW7, fixed a broken generator on the Rapid-Scan DOW (yes the one we just purchased last week!), and worked on some computer problems on a couple mobile mesonets. With this many instruments (CSWR is fielding 21 separate weather stations), there are always things that need fixing.

Friday is our next mission day and we are hoping that we get some good data. And, we're hoping for an early night and some good rest this evening.

Wednesday, May 13, 2009

VORTEX2 hits the road


VORTEX2 had its first real mission today. We departed from Clinton, Oklahoma and rushed to Tulia, Texas. Then, as often happens when pursuing tornadoes, we waited and waited for a storm to start. Finally, around 5pm storms started to get stronger to our west towards New Mexico. We didn't have much optimism that they would make tornadoes, but the massive VORTEX2 fleet really needed practice. So, we headed away from town to try our luck attempting to intercept a real storm, with real wind, rain and maybe hail. Well, with 40 vehicles trying to coordinate for the first time, the result was predictable: a certain quantity of chaos. But, it was a great mission. It was very hectic in DOW7, with all 3 radios blaring with different teams' questions and reports. The mobile mesonets reported a road was closed because a whole line of power poles had snapped and fallen, probably from about 100 mph winds in one of the microbursts. The dust was beautiful. We deployed the DOWs and scanned through the storms, we had our pod teams actually deploy in front of the outflow, we tested communications with the several other teams. Some things worked, some things didn't.

When we finally gave up on the storms we had a wonderful show, a beautiful sunset with storms and a rainbow and a spectacular lightning show at our hotel.

Tomorrow looks like a 'better' severe weather day, so we are trying to fix what we can overnight and prepare to start all over again in the morning.

Sunday, May 10, 2009

Glamour, Sciance, and the Start of VORTEX2


The glamorous part of our project is over. The real missions are about to begin. VORTEX2 cut the umbilical today and we became 'mobile'. I put this in quotation marks because we only deployed from Eastern to Western Norman, Oklahoma. But we are now all staying in the same hotel, working out of our mobile office not the comfort of the University, and doing our last minute preparations. Today we are holding a class, Tornado Pods 101, teaching our crews how to deploy the pods. We need them to know this like the back of their hands because they have 45 seconds to deploy each pod (we'll practice and practice until we get this time down to 45 seconds). Pod deployments happen in bad weather, to say the least. Our plan is that the 4 vehicles wait until the tornado is about 2 miles to their west, about 4 minutes away, before starting their deployment runs. Each team has to deploy 3 pods. 45 seconds for a deploy, 45 seconds to the next location, 45 seconds to deploy again, etc. Then they run away to the south of the tornado, wait until the tornado passes, then rush back in to collect their prizes, Pods with tornado data. After all that they rush back ahead of the tornado or tornadic storm and repeat the process.

Tomorrow we expect to go really mobile, and leave town.

Friday, May 8, 2009

Arrived in Norman for 'Media Day'

We finally arrived in Norman for Media Day. It was well orchestrated chaos. Even more than usual, there were print, photo, and TV reporters, documentary companies, seemingly every kind of press out to take pictures of the VORTEX2 fleet. I'll post some pictures later, but this was surely the largest gathering of severe weather science equipment ever. Dozens of science vehicles, a large crowd of media, students, scientists, etc. Even Miss Oklahoma, in costume and crown, was there. Who knew that science was so glamorous? I was assigned a satellite truck and gave 15 interviews between 5:30 am and 9:30 am, then gave one of the introduction talks at the media day gathering. Many VORTEX2 PI's, the Steering Committee, and other scientists were at the meeting and we received several interesting questions from the audience and through call in lines and e-mail. Then we all went to the parking lot to be interviewed some more.

Meanwhile real work had to continue. We were real happy that the broken generator on the rapid-scan DOW had been replaced...until we found out that the generator shop had installed the wrong generator. Now we have to get yet another generator overnight from Texas and beg the shop to work on in on Saturday. Nothing is easy it seems.

The DOW7 rushed out of media day in order to get its transmission replaced. Otherwise the team is working on software, weather instrument inatalls, and completing two Pods for CU.

Wednesday, May 6, 2009

CSWR team in Hays, Kansas


We've stopped in Hays, where we have rented a garage, so we can do some more work on a few of our systems. While we were driving we tested our radios, computers, and tracking systems in 'real world' conditions. Mostly things worked, but with this many radars, mobile mesonets, pods, computers, etc, there is always a bug list. One of the generators on the Rapid-Scan DOW is idling roughly, so we've sent it out ahead to Oklahoma City to get fixed. Maybe it will be fixed by the time we get there.

Our crew ate lunch at a real restaurant today. For a group that has been eating every lunch and every dinner in our Boulder hanger for the past 2 weeks, this was quite a treat.

This is Dr. Karen Kosiba and Paul Robinson working on some software installations. Karen is a recent PhD from Purdue. She is studying low level winds in tornadoes in VORTEX2. She has been on several DOW missions including 4 tornado seasons, a few hurricanes, and to a project called COPS in Germany. Paul is an analyst at CSWR and is working on radar and tornado pod analysis. He is going to have a very tough job this year, at the receiving end of the fire hose of all our data. Jon Lutz is inside the DOW7 making the antenna spin better. Jon is a radar engineer who has been working on DOWs since 1995, on the original DOW1. Jon has been on dozens of field programs with NCAR and other groups, all around the world and is one of the best all around people to have on any field project.

DOWs, Mobile Mesonets, Tornado Pods on the Road

The CSWR and CU teams have left Boulder, finally. Late yesterday we got the newest DOW, DOW7 working well enough that we could leave our lab. We made it to Limon, Colorado where we spent the night. Now we're heading to Hays, Kansas. We're spending the road time testing and shaking out our systems. As is to be expected we have a long bug list. This includes a broken air conditioner in DOW7. It is getting quite hot here in the back with all these computers and electronics. Spirits are way higher today; the crew really needed to get out of the shop and on the road.

Tuesday, May 5, 2009


I am writing from Boulder, unfortunately. We tried until 12:30am last night, but couldn’t get our last radar working. The problem is that the antenna will not spin. We got all packed up, but then had to have our crew go back to homes and hotels. This morning we’re trying again and we hope to leave by afternoon.
The picture is of a radar that we did get working. It is the Rapid-Scan DOW. Normal radars send out one beam, kind of like a flashlight beam. This is great for most uses, but not for studying tornadoes. The problem is that even the very fast-spinning DOWs, take a full minute to collect a 3D volume of data. Essentially we just get a picture of a tornado once per minute. But, tornadoes change very very quickly, sometimes evolving a lot during that minute between our ‘pictures’, and we miss how those changes happen.

The Rapid-Scan DOW has a pretty clever design that allows it to send out 6 beams simultaneously, but at different inclinations. So, we can ‘rake’ the sky with these beams, and in one sweep of the antenna we have a 3D image in the lowest few thousand feet of a tornado…every 5-7 seconds. Now we can see those extremely rapid changes.
The first time we took the Rapid-Scan out to see tornadoes, it was able to collect 3D images. We collected data in 6 tornadoes in 2005. But, it was not able to see enough, was not able to see what radar meteorologists call ‘clear air’ echoes just outside the tornado. We have that fixed, and have upgraded other components, and we’re ready to go. We did some final tests by placing a microwave ‘beacon’ on the roof of my house, and seeing how the 6 different radar beams saw it.

But, back to today. We are itching to go. We’re all waiting for the final DOW radar, DOW7, to have its problems fixed and tested. Maybe my next post will be from the road

Monday, May 4, 2009


We are in our final hours of frantic preparation before leaving Boulder. We’re behind schedule, so everyone is very anxious to leave. We have lots of our equipment out in the parking lot and we’re packing boxes of spare radar parts, medical kits, cables, computers, etc.

Most of our crew are here. So, we have a dozen or so people working on our 9 vehicles. An IMAX crew showed up yesterday too. They’re filming an IMAX documentary about VORTEX2. They shoot through a whole roll of very expensive film in about 3 minutes.

We’ve had a couple very very long days here since my last post. Some of the hardest tasks are pretty mundane, but critical. Here we’re trying to pull radio cables through a hose that coils up a 56 foot mast on one of the DOWs. It took hour and many attempts. It was a hard, messy greasy job.

The vehicle behind the hose pulling team is one of our mobile mesonet / tornado pod deployers. It has a 13 foot mast with two anemometers (in case one is smashed by hail), and a thermometer and relative humidity instrument inside an S-shaped shield which keeps dry air flowing over them. Yes, we've hit branches, wires, and fast-food drivethru overhangs with them; it is pretty ugly. But, getting the instruments as high as possible above the ground and into the wind is important.

Our newest DOW will be tested in just a few hours. If it works, we can go, if it doesn’t, it is another long night of work in our hangar.

Well, that’s all for now. Next time I write, I hope it is from the road, after some successful practice deployments.

Friday, May 1, 2009


The VORTEX2 project is about to begin. We have been working on asking which questions about tornadoes are most important, designing VORTEX2, planning, writing proposals, gathering community input, more planning, etc. for several years. With substantial funding from the National Science Foundation and the National Oceanic and Atmospheric Administration, several universities, government labs, and private companies are coming together to begin the largest, most ambitious tornado project ever. It all starts in several days.

I run the Center for Severe Weather Research in Boulder Colorado, and perhaps the largest individual group within VORTEX2. We will be fielding 3 radars, 4 mobile mesonet vehicles which deploy a total of 12 tornado pods, and support vehicles, operated by a staff of about 24, ranging in background from senior engineers and meteorologists, recent PhD's, to current graduate and undergraduate students.

We hope to leave Boulder very very soon so that we can get in a week of practice before
the main VORTEX2 operation starts. Preparations are ongoing at a frenetic pace. We're my dedicated, but a bit burnt out, crew is here from 0800 to 2300 every single day. (Is it a weekend? Not in the CSWR DOW hangar.)

I'll be introducing our vehicles, instruments and crew in upcoming posts. For now, just one
picture of our two new DOW radars. One is raised up next to the other since we had to drain all the diesel fuel from our newest truck so it could pass its initial registration weigh in (yep, it is legal to drain the fuel). We had to come in at 25,999 lbs or less, we squeaked in at 25,940.
Whew.