Has NEXRAD Eliminated Thunderstorm Accidents?

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

This weekend's story about a trio of guys motoring around the Midwest in a Cessna 182 chasing tornadoes reminded me of what I don't see much of anymore: accidents involving penetration of thunderstorms. Twenty years ago, we used to see a couple of dozen of these every year. Now, I'd guess it's less than half that.

The reason should be obvious, even if I can't prove it. Datalink weather and ongoing improvements in weather displays in the cockpit have just about taken thunderstorms out of the accident equation. A 110-knot Cherokee can have a weather suite rivaling an Airbus, if not better.

Convective weather accidents often involved pilots flying places on instrument flight plans. They'd blunder into cells trying to run a line and rip the wings off the airplane or get hammered by hail, ice or turbulence. Stormscopes and Strike Finders helped; NEXRAD in the cockpit helped a lot more.

Flying into and around thunderstorms intentionally is nothing new and has been going on for years, although they never called it storm chasing. They called it weather research and the fact that it was done informs our knowledge of how thunderstorms and tornadoes form and move. These days, if YouTube and The Discovery Channel are to be believed, storm chasing has taken on a certain entertainment value. Not that I'm sniffing at that, mind you, for I'm way happier watching these guys run around the plains in camera-festooned SUVs than I am end-running a mesoscale line in an airplane.

I'm not sure what modern storm chasing is going to add to the knowledge base of avoiding cells in airplanes. As NEXRAD dissemination improves, the task is pretty simple: See the screen splotched with pulsating pepperonis? Don't fly there. If you've got onboard radar, too, so much the better, although it's good to understand attenuation. (In fact, I'm writing this from seat 25A in an airliner inbound to Wichita. It's skirting a dark area of cells off to the north.)

For those doing the chasing, thunderstorm hazards haven't changed much. And that applies to those of us who use NEXRAD or even just eyeballs to skirt a little too close. The late Dan Custis, who along with a few other pilots, flew a project to penetrate thunderstorms with an armored T-28, once told me that the experience led him to a surprising conclusion. And that was this: Except for one thing, if a pilot could keep the airplane upright in the clouds, a Cessna 172 could survive much of what might be encountered in a typical continental thunderstorm. What was the one thing? Hail. In this interview which appeared on AVweb in 2003, one of the T-28's pilot, Charlie Summers, said one problem the project consistently encountered was the engine pushrod tubes being hammered flat by hail impact. (The canopy itself had an armored outer cover.) Custis, by the way, wasn't suggesting anyone should penetrate cells in anything, but merely noting what his experience provided.

Hail can be just as deadly outside the cell, flying nearby in the bright sunshine, because hail shafts are often ejected from the top of the cell. Wayne Sand, an NCAR researcher at the time, once told me he was flying outside a cell with another pilot in a Twin Comanche. In the distance, they could see a large hailstone that seemed to be on a perfect trajectory to come through the windshield. It was and it did.

Back in my younger, dumber days, I skirted a cell too close and got into a vigorous column of grauple. If that stuff got another trip above the freezing level, it would have been grape-size hail and I might very well have been among those two dozen accident pilots common on those days. If being lucky is just as important as being good, I think I proved the point.

Comments (24)

How many flight hours does the average private pilot fly today compared to 20-30 years ago? How poor was all preflight weather information some 20-30 years ago?

Posted by: Mark Fraser | June 3, 2012 9:04 PM    Report this comment

@ Mark Fraser - agreed.

Posted by: john hogan | June 4, 2012 4:14 AM    Report this comment

Don't forget that approach radar has also improved in the depiction of storms as well, and ATC is supposed to inform pilots if they're flying into the bad stuff they see. Sometimes they don't and the results can be tragic - read the NTSB report on Scott Crossfield's fatal accident in a 210. The board placed a significant portion of the blame on the controller. It is a wise aviator that remembers that Nexrad is a strategic rather than a tactical tool.

Posted by: Jerry Plante | June 4, 2012 7:42 AM    Report this comment

I remember seeing the T-28 parked at Laramie years ago--with a badly cracked (albeit "armored") windscreen, and skin that looked a lot like an over-sized aluminum golf ball. Heavy rain sounds like a load of gravel being dumped on an airplane; I can't imagine what heavy hail sounds like--and the damage it could inflict could literally tear an airplane apart. The one thing that the T-28 had in spades, however, was enormous airframe strength.

Not to sound too morbid, but I think these folks in the 182 are flirting with becoming statistics. Strength-wise, a 182 is not built anything like a T-28, so that if the hail doesn't get them, the turbulence will.


Posted by: Cary Alburn | June 4, 2012 8:25 AM    Report this comment

I was a little unclear here. I was really referring to a rate. We used to sort accidents by type/cause and thunderstorms had its own category. Now they don't, because we don't see enough to merit mention.

Also, sketchy data suggests that a larger percentage of total hours flown are on IFR flight plans. This is because instrument ratings have grown substantially since the early 1990s and the fact that wealthier pilots who can afford new or high performance airplanes do more of the flying helps.

Despite all this, thunderstorm accidents appear to be down.

I'm not sure I agree that tactical versus strategic has a real influence here. At a certain point, they become one in the same. You can take off knowing there's a line 100 miles out and usd the NEXRAD to either pick a route around it from further out or just divert because you can't. To my mind, that's both strategic and tactical.

Pure tactical is picking a level 1 area to penetrate rather than a level 5. Some people do it with datalink and as long as the rain areas are broad, I don't see a problem with it. I've done it, too.

Like anything else, you just have to exercise judgment. Do it when you can, don't when you can't. Be we don't seem to be seeing many people doing the latter.

Posted by: Paul Bertorelli | June 4, 2012 8:32 AM    Report this comment

iPad, Foreflight, and Stratus has simplified and relegated flight planning to (almost!) the dummy-department. Careful ... I say that with much joy and relief.

Posted by: Phil Derosier | June 4, 2012 8:37 AM    Report this comment

I consider in flight Nexrad availability to be one of the most important GA flight safety advances in the last 50 years. It must have been 2003, maybe 2004, when I started using XM on a tablet lash-up, and seeing the big picture in the middle of a 3-5 hour flight suddenly allowed the sort of strategic/tactical choices that you mention, Paul. The ability to tell ATC exactly where I want to go on a revised clearance makes life easier for all of us. Yes, latency is a problem and there is still a role for sferics and on-board radar (I have and use both), but it's now more about pilot choices with good data, not luck and imprecise FSS oral descriptions, to stay away from the boomers and nasty stuff. Same with watching metars/pireps, it's all revolutionary. The folks who deliver this data to us in the air deserve a medal.

Posted by: Scott Dyer | June 4, 2012 10:14 AM    Report this comment

Back in the 80s and 90s, my annual Aviation Pilgrimage (to Oshkosh) resulted in about a 50% success rate of getting back to the west coast in a single day. The inevitable afternoon convective build ups over the Rockies resulted in a overnight stay in CO or KS about half the time.

Departing early in the morning, we had to make a best-guess where the buildups would be. We would then see how our guess worked out when we stopped for fuel in KS or CO and checked the national radar picture.

Now, with on-board Nexrad and satellite images, my success rate has been 100%. I'm able to see the developing situation from hundreds of miles away and make strategic planning decisions far enough away that there is little delay in my overall time enroute and still complete the trip under VFR.

Meanwhile, I hear pilots without Nexrad talking to FlightService and diverting to wait out the weather (just as I used to) when convective SIGMETs pop up on their route. Often those SIGMETs cover entire, if not multiple States.

Using Nexrad "tactically" to avoid embedded thunderstorms while IFR is more aggressive than I care to be.

Posted by: Kris Larson | June 4, 2012 3:00 PM    Report this comment

While I agree that NEXRAD is great, I am cautious about relying on it too heavily. In our King Air we are exposed to weather frequently. I have often found the position of cells to be displayed somewhat inaccurately, even if when recently updated. There have been several times when, if I had made a turn based on NEXRAD to avoid a storm, I would actually flown right into it. Therefore, I usually use the datalink radar for longer-range routing, and my onboard radar or onboard eyeballs to avoid them at closer range.

That being said, I still love NEXRAD, especially when combined with our onboard radar and strike finder.

Posted by: Joe Brand | June 4, 2012 4:55 PM    Report this comment

When I first used NEXRAD, I was sure thunderstorm encounter accidents were going to be a thing of the past for NEXRAD users. I was wrong, there are now pilots who use NEXRAD to cut even closer to convective activity. Any great technology can (and will be) defeated by human nature.

Posted by: Steven Hurst | June 5, 2012 7:16 AM    Report this comment

I haven't had the chance to experience in-cockpit NEXRAD on any of my flights, but have certainly had a few flights where I wish I had access to that data. It may not provide a precise "now" location of storm cells, but it would have helped in just knowing which way and how fast the storm cells are moving. Of course, it's really only when combined with a stormscope and/or on-board radar that NEXRAD really becomes a great tool.

Posted by: Gary Baluha | June 5, 2012 8:56 AM    Report this comment

I've been using XM weather for several years (ever since the Garmin 396 was released), and I find it invaluable for actually getting to where you want to be reliably and for deciding when to "wait it out". I also find it extremely useful in negotiating re-routs with ATC while in the air. They seem to be much more responsive when you tell them that you are painting bad weather on your assigned route. I also have a strike-finder which generates a few phantom "strikes" now and then. However, if the Strike-finder is light up and the Nexrad shows red, then that's an area to avoid.

Personally, I would never try to spit cells with these technologies, but there are lots of times where a long line of weather features fairly innocuous areas and really bad stuff in the same line. The key is to get on the other side of that line with nothing more than a bit of turbulence. Without using these tools, I'd probably stay on the ground a whole lot more.

Posted by: Phil Ryder | June 5, 2012 4:51 PM    Report this comment

As a Garmin 696 user, I consider NEXRAD as mandatory equipment, not optional. Also, since it is portable there's an easy experiment you can perform from the safety of your easy chair at home to gain a better understanding of latency. In my local area one of the local TV stations feeds their own live radar image to a dedicated cable-TV channel. Turn on the TV radar channel when storms are in your neighborhood. Setup the portable NEXRAD equipment and satellite antenna next to a south facing window so it receives an adequate signal. Now compare the TV radar to the downloaded NEXRAD image and then compare the two to what is happening outside your window.

You will soon notice the major differences. I have seen bright red cell positioned over my house while both the TV radar shows the cell has moved away and the sun was shining outside. NEXRAD has also depicted the cell's magenta core as several miles away when it was actually overhead and doing its best to drown the neighbor's cat.

The lesson to learn is that this magic box is best used as a strategic planning tool and not for close-in tactical avoidance.

Posted by: Keith Gutierrez | June 6, 2012 6:47 AM    Report this comment

The technological advances in weather avoidance are simply mind blowing! It was 1955 and I was flying a Marine AD-5 (A1E) at night to Miami. The weather was awful to say the least. There were TRWs in all quadrants. My ADF needle was swinging all over the place. I made it to Miami radio and they assigned me 8000 feet in the stack. A massive thunder storm cell moved right over Miami radio. All heck broke loose. We had severe turbulence, rain, hail and more lightening than I ever want to see again. At one point Miami approach asked me about my position. I said something like: “I am between 9000 and 6000 feet on a yo yo. I am also inverted.” They cleared be out to sea until they got it all straightened out.

Posted by: Bruce Miller | June 6, 2012 12:12 PM    Report this comment

I have a Garmin GDL69A feeding XM Wx to my Garmin 530 and MX20. I have a Garmin 496 for backup.

I use the equipment in a tactical sense in our Monsoon season flying into central Arizona. I find the displays are updated almost on a minute-to-minute basis, depending on what you have selected for display, and the range displayed.

I'm under the impression (without any direct knowledge) that the fewer products you put on the display, and when a smaller range is displayed, updates are frequent. Both the 69A and the 496 allow you to put the 'age' of the data on screen. You can see if the info is 1 minute or 5 minutes or whatever old, and act accordingly.

I have a Stormscope feeding lightning to my Sandel HSI with updates at rapid intervals. which also helps (I don’t display the XM lightning on the 530 but the Stormscope also feeds the 530 and MX20.

Right now, the 496 in my office is showing the view of the west coast as 13 minutes old. I never see that ‘old’ of data in the plane.

I used to fly some heavy metal Lockheeds in the late ‘60s with then current radar. I never developed a confidence in radar that I have now.

Posted by: Edd Weninger | June 6, 2012 12:23 PM    Report this comment

I decided to actually take a quick data sweep to see if my knee-jerk estimate was correct. Normally, where we pick up thunderstorm accidents is when doing a Used Aircraft Guide on high performance airplanes--the Cessna 210s, the Malibus, the twins and so forth. The percentage of these airplanes involved in thunderstorm accidents seems to have declined.

Searching the NTSB records for thunderstorm-type accidents between 2006 and 2012, I found 22. For the period between 1990 and 2003, there were 123.

I didn't read them all, but most of the airplanes involved were high-performance models, including some airliners and turboprops. Not all were fatal, either. In some, thunderstorm activity was only incidentally involved.

If I get time and can find the data, I'll try to calculate a rate. But just skimming it, my impression seems correct. The number has trended down noticeably.

Posted by: Paul Bertorelli | June 6, 2012 6:32 PM    Report this comment

So what is new about researching severe storms and possible tornadoes. I did it 33 years ago.If they get their 182 into and inflow that is going into the storm faster then they can descend they better be prepared for instant IFR and have a way out. Been there,

Posted by: Lester Zinser | June 7, 2012 1:27 PM    Report this comment

Ironic that this article gets written and shortly afterwards N950KA (PC12) breaks up in flight @FL260 over Florida killing a family of six. Thunderstorm related? Time will tell but wouldn't be a shock.

Reminds me when a sports announcer says "this guy hasn't missed an extra point in two seasons". Doink.


Posted by: Mark Hangen | June 8, 2012 1:18 PM    Report this comment

Will be interesting to the cause for PC accident.

Posted by: Lester Zinser | June 8, 2012 3:51 PM    Report this comment

we fly aircraft (Senecas, C340s, C90s, Cheyennes) around thunderstorm systems for a living (cloud seeding). none of our airplanes is equipped with xm or other weather currently, but all except the Senecas have onboard radar. Nexrad radar is great for getting the "big picture" but DON'T USE IT TO TRY TO NAVIGATE THROUGH THUNDERSTORMS. Radar is HISTORY - radar only depicts precip-sized water - rain. The radar systems (Nexrad) perform a "volume scan" procedure to gather the data - one rotation at 0 degrees, then at 1 degree, etc - this takes a few minutes. Then once it's gathered the current scan data, it sends it to the display. Then, there's a time delay for XM or whatever service to upload it to your receiver. The upshot is that what you are seeing for "radar" displays is maybe 15-20 minutes ago! Meanwhile, growing cumulus cells have updraftsof 1000-5000 ft per minute - do that math and see how the Nexrad is old news once you see it in your airplane. Our pilots are talking to a dedicated project ground radar (which does it's own volume scan, so still HISTORY) to get the big picture, but we're not trying to fly THROUGH a line of storms to get to Grandma's house, only flying up to the cells and then going wherever the cells go.

Posted by: Hans Ahlness | June 9, 2012 1:34 AM    Report this comment

Sadly, the Pilatus that crashed on Friday, June 8th in Florida probably had live radar, stormscope and Nexrad. The pilot appears to have flown into severe weather. The NTSB will figure out what happened. But, I continue to see people trusting devices like Nexrad, that have considerable delay in updating their picture, too close to severe weather. I think for too many people technology provides a false sense of security. Thunderstorms will kill you and they need a wide berth. Fred Ahles

Posted by: fred ahles | June 10, 2012 6:03 AM    Report this comment

Pilots wanting to learn more about thunderstorm avoidance are in luck. AOPA is running a great program next week. See below:

Storm Week just around the corner Watching a thunderstorm mature from your front porch can be beautiful. Flying anywhere in the vicinity of one can be ghastly. Combat convective weather with the Air Safety Institute and Storm Week—June 11 through 15. Each day the Air Safety Institute will launch a thunderstorm-themed product, including an Ask ATC video, a quiz, and a live webcast. Mark your calendar now for the live webcast on June 14. Renowned experts explore convective weather, ATC’s role, and when to say “no” to a flight. Tap into the Air Safety Institute’s Storm Week and apply what you learn!

Posted by: fred ahles | June 10, 2012 6:16 AM    Report this comment

In my opinion, deciding when *not* to go is the easy part: if you don't like anything about what you see, you just don't go. Which means sometimes canceling a flight due to normal summer airmass thunderstorms. The real problem is deciding when *to* go. Everyone talks about when to make a no-go decision, but little is mentioned about how to make a "go" decision when the weather isn't entirely cooperative. Which really equates to not talking about how to apply risk and resource management to situations where the go/no-go decision isn't so clear cut, and where proper use would allow one to make a "go" decision in certain circumstances.

Posted by: Gary Baluha | June 11, 2012 10:31 AM    Report this comment

The NTSB has just put out a finding about the weaknesses of doing a flight while using Nexrad data. Some of that info can be over 20 minutes old when it shows up on the display. From someone who flew on demand cargo and now flies fractional nothing beats having a functional on board weather radar. The next best tool is ATC but their radar can have some serious limitations. For the weekend flyer it is best to avoid flying in thunderstorm coditions all together.

Posted by: matthew wagner | June 21, 2012 12:00 PM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?


Enter your information below to begin your FREE registration