Nexrad Radar Can Lie

I was headed south from my home airport in Muskegon, Michigan, on what was a typical summer day in what has been a very untypical spring and early summer. It was very warm, haze cut the visibility to around 7 or 8 miles, and there were a few very high scattered cirrus clouds. The forecast called for those conditions to exist all the way along my route to Atlanta.

Not long after takeoff I was surprised to see an east-west line of weather appear on the XM Weather Nexrad radar display. The echoes where stretched from over Grissom Air Force base to east of Fort Wayne, Indiana. The returns were solid level one and two green, with sizable areas of yellow, and even a few spots of red.

The Garmin G600 display in my Baron shows six levels of the possible 19 that are displayed on the returns from a Nexrad radar site. Each level is actually a measure of the radio frequency energy that is bounced back to the radar antenna. Big rain drops and wet hail are the best energy reflectors in the atmosphere and are also indicative of a strong thunderstorm. So, the stronger the echo, the more likely there is a thunderstorm with severe turbulence and maybe hail.

The new Garmin GTN750, a follow-on system for the Garmin 530. Top photo: An example of anomalous propagation (AP). Courtesy Lyndon State College

It’s pretty easy to tell the light green of level one from the darker green of level two. But the two shades of yellow that indicate levels three and four are very close in shading. And the two reds for level five and six are really hard to tell apart unless there is a huge area of each level showing to compare.

However, when a Nexrad return contains at least some areas of level three or four yellow, and some specks of level five or six red, it deserves attention. The precipitation indicated may not be coming from a convective cloud, but any precip reaching level three or four usually has some noteworthy turbulence.

I also have an Avidyne satellite receiver that operates on the Sirius network. The weather information supplied over the Avidyne/Sirius satellite comes from WSI, the weather information that is in the flight planning room at most FBOs. The data on XM is supplied by Barron Weather Services.

The line of weather continued to be displayed by XM as new radar mosaic pictures arrived about every five minutes, but there was not a single speck of weather showing on the Avidyne/Sirius radar picture from WSI. I know it’s weird to have both services in the same airplane, but that’s a long story.

I had to choose to believe the XM or the Avidyne/Sirius radar picture. If I believed that the significant weather showing on XM was real, a deviation plan was necessary. But if I believed the Avidyne/Sirius picture there was nothing to avoid.

Given the total lack of a chance of precip in the forecast, and the actual weather conditions with no clouds in site and a stable air mass, it sure looked like Avidyne/WSI had it right. And nobody on the frequency was talking about any weather ahead. I droned on.

After five cycles of Nexrad images showing essentially the same line of weather on XM, the next mosaic arrived and the line had broken up. Five minutes later, the radar image on XM was totally clear.

AP can be caused by anything from flocks of migrating birds to strong temperature changes in the atmosphere. Courtesy: Unisys

I had obviously seen a prime example of anomalous propagation (AP) on XM. For all kinds of reasons a weather radar can “see” precipitation or other targets that simply aren’t there and that’s called AP. Some AP really isn’t all that anomalous because it is caused by a terrain feature, or an object sticking up in the radar antenna path. But other AP can be caused by anything from flocks of migrating birds to strong temperature changes in the atmosphere.

The radar image we see in the cockpit from XM or Avidyne/Sirius is a mosaic of returns from individual Doppler radars on the ground. In most areas of the U.S. the coverage of the radars overlap so the same parcel of air is scanned by perhaps three, or even four, radars. To make a mosaic that covers the whole country the private weather services such as XM/Barron and WSI use a computer algorithm that selects the return from what they believe is the optimum individual radar site. The algorithm attempts to account for known AP spots and weeds them out. The computer also attempts to use other “logic” tests to determine if an echo is real or AP before including the return in the mosaic.

The Avidyne FlightMax EX500.

Obviously, something didn’t work quite right at Barron Weather Services on this day, but it did at WSI. Who knows why? WSI has a human meteorologist scan each mosaic before transmission and perhaps the human spotted the AP over Indiana. But my point is that we sill must use satellite weather radar as a broad strategic weather avoidance tool, not a close in tactical guide. Given the delays – at least 10 minutes – necessary to gather the returns from an individual radar, create the mosaic, and transmit it, the weather picture we see in the cockpit may have changed from the reality we are flying in. And creation of the mosaic itself can occasionally present a false picture of the real weather.

In this case the AP posed no threat to safety because it showed something that wasn’t there. It cried wolf when none was present. If I had flown around the fake wolf it would have added a little time to the trip but no extra risk. But a Nexrad mosaic can also occasionally fail to show a rapidly building storm for a cycle or two and that can lead you down an unsafe path.

Nexrad in the cockpit is great. It has changed the way we fly more than any development in a long time and adds greatly to safety potential. But occasionally I get a reminder that Nexrad isn’t perfect and we still need to consider the whole weather picture.

This entry was posted in Airmanship, Safety, Technology. Bookmark the permalink.

9 Responses to Nexrad Radar Can Lie

  1. Lawrence Sciortino says:

    Good commentary – it’s worthwhile to review that no one indication can be trusted entirely when it comes to mother nature. imo, the second hand radar imagery with its inherent time delays, such as you have in your Baron, is a lot better than nothing, but is no substitute for real time on board radar, especially in approach and departure realms. The development and position of cells can change too quickly for close quarter work.

    • Tex Hull says:

      This comment pretty much nails it. I would add two thoughts. First, Nexrad beats on board radar hands down at long range, e.g. at 200 nm. Second, lightning detection adds valuable information.

    • Mauda Palmer says:

      Are you Larry Sciortino from Eastern Airlines? If so, please contact me at maudsiek@

  2. Pete Zaitcev says:

    Funny coincidence. I logged to DUATS today and it warned about a SEVERE THUNDERSTORM in the area of the airport (KAEG). Went to the airport – not a cloud in sight.

  3. Mac says:

    Even when the radar returns are real, and not AP, radar images can be difficult to analyze. But more often than not, radar fibs by showing something that is not there, rather than not showing precip that is there.

    Mac Mc

  4. noticias says:

    advertising and *********** with Adwords. Anyway I?m adding this RSS to my email and could glance out for much extra of your respective fascinating content. Make sure you update this once more very soon..

  5. wsi says:

    Hmm it seems like your site ate my first comment (it was extremely long) so I guess I’ll just sum it up what I wrote and say, I’m thoroughly enjoying your blog. I too am an aspiring blog blogger but I’m still new to everything. Do you have any recommendations for newbie blog writers? I’d genuinely appreciate it.

  6. It’s truly a great and useful piece of info. I am happy that you simply shared this useful info with us. Please keep us up to date like this. Thank you for sharing.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>