What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

ADS-B weather is.....meh.

rocketbob

rocketbob

Well Known Member
Two weeks ago I was flying with my college student son who is a newly minted private pilot. After having lunch we took off and were in the vicinity of some rain showers, some of which we flew thru and they were light. By the time the iPad weather got caught up it was showing yellow, red, and purple in where we were. And it was enough to make my son wisely question what we were doing. The 496 XM weather, disagreeing with the ADS-B radar, was not showing anything where we were. Looking out the window showed the same thing.

I read somewhere that ADS-B weather was 80% of XM's but 100% less in cost. I wish there were more options for XM weather receivers which would work with most EFB apps. I don't mind paying for weather that is accurate and with good radar resolution.

Good video showing what goes on behind the scenes and some good comparisions.

https://www.youtube.com/watch?v=AXPGRr5bKN4
 
Last edited:
I have always noted the same thing. The nexrad weather via satellite always seems more accurate and timely than ADSB weather.
G
 
Maybe help is on the way? From Sportys iPad pilot news: faster radar updates and this quote:
“The most important weather product for most pilots is radar, and that’s also new. The picture and the overall functionality is really the same, but it’s now being sourced from the Multi-Radar/Multi-Sensor radar (MRMS) product. The National Severe Storms Laboratory describes MRMS as “a system with automated algorithms that quickly and intelligently integrate data streams from multiple radars, surface and upper air observations, lightning detection systems, and satellite and forecast models.” The goal is a more accurate radar image. You can learn more at their website.”

https://ipadpilotnews.com/2019/06/a...aign=A19061C&utm_content=Fast+Five+-+June+8th
 
Last edited:
ADS-B precip depiction (map) weather is NEXRAD. Same feed the controllers and internet apps get.
Some displays tweak it a bit resolution wise but it all comes from the same central source which is a computer compilation of ground radar and satellite imagery that varies some in accuracy by location, terrain, and altitude. It defaults by design to over display rather than under display, hence the occasional ghost artifacts.
 
Remember both xm and adsb radar have latency - you see what was there, not what is there. Having great resolution really means little if the data is 15 minutes old. These products are both great for what they were designed for - long range strategic avoidance. They were never meant for close-in avoidance.
 
BobTurner said:
Remember both xm and adsb radar have latency - you see what was there, not what is there. Having great resolution really means little if the data is 15 minutes old. These products are both great for what they were designed for - long range strategic avoidance. They were never meant for close-in avoidance.
Click to expand...

I get that but I've always found XM to be much closer to reality.
 
GalinHdz said:
FWIW I don't use in flight weather for reality, I use it for trends. :cool:
Click to expand...
Exactly. As Garmin would say for the G3000, NEXRAD should be used for “early planning” and only on-board weather radar should be used for “tactical penetration”.

As someone involved in training, this has been a huge issue with the latest generation of pilots who came up on GPS and XM being in everything they flew - less situational awareness and over reliant on the XM picture.
 
In my recent case, there was no dangerous weather around. Just light to moderate rainshowers.

ADS-B was showing dangerous weather.

XM wasn't.
 
When I look at the ADSB WX, I also watch were the traffic ahead is going and at what altitude and what ATC is telling folks. Traffic is real time and helps to get a picture if there are build ups ahead.
 
I recently switched from many years of XM to FIS-B on a portable device. I liked XM a lot, particularly in the early years when it seemed almost magical. It wasn't perfect however; for example it commonly showed light precipitation where none existed. It also tended to be slow in initial acquisition, and occasionally stopped working entirely.

The video linked in the first post made a big deal about how XM could be animated to show storm movement, which I found puzzling because you can do the same with the NEXRAD transmitted via FIS-B (for example using Foreflight). If you're using a portable device its also easy to get the weather while still on the ground, via cell tower (either through your device directly or by tethering to your phone).

I'm curious about the update interval claims. My crude understanding is that the NEXRAD image is not actually a snapshot in time, but a composite of base reflectivity at multiple elevations that takes several minutes (up to 5?) to assemble. According to the current AIM (p. 4-5-20) the update interval for FIS-B is 5 minutes, but the transmission interval is 2.5 minutes. Is Sirius really presenting a more up to date picture? Or is it more complicated than that?
 
For a great demo of both, take a look at Matt Guthmiller’s last video. It shows the differences “In the Air” between the two, ADSB and XM weather. It sounds like an ad for XM, but it’s the best demo I’ve seen to date, IMO.
 
Alan Carroll said:
I'm curious about the update interval claims. My crude understanding is that the NEXRAD image is not actually a snapshot in time, but a composite of base reflectivity at multiple elevations that takes several minutes (up to 5?) to assemble. According to the current AIM (p. 4-5-20) the update interval for FIS-B is 5 minutes, but the transmission interval is 2.5 minutes. Is Sirius really presenting a more up to date picture? Or is it more complicated than that?
Click to expand...
I think you have it correct. The time stamp on our displays shows the last uplink (or down link, for xm); the actual age of the data is unknown, but longer. Is 2.5 minute uplink times better than 5? Probably yes, but only marginally so. The Xm data looks nicer than ADSB. But given the latency, the user must always be careful to make it fuzzy in interpreting it - to account for the unknown changes since the last data was taken.
 
I fly with both. XM seems a little more accurate based on the view from the canopy. I agree that it sometimes shows areas of light rain when there is no rain. Perhaps it's more sensitive than ADSB and picking up mist from higher altitudes. John
 
rocketbob said:
In my recent case, there was no dangerous weather around. Just light to moderate rainshowers.

ADS-B was showing dangerous weather.

XM wasn't.
Click to expand...

The ADS-b data coming from the ground just gives the dB levels of the returns. The display device takes that data against it's internal programs and conventions to display it. Each device decides for itself what color to paint for a 45 db return, for example. There is variation among the device manufacturers. Data from the same receiver in my 6 shows a much scarier picture on my GRT display than it does on my ipad/foreflight. Also, a 4 color algorithm usually looks scarier than the same data displayed in a 5 color scheme, due to the large ranges for each color in the 4 color scheme. FF has a slider for displaying a 4 color rendition. If you turn that on, the picture gets a lot scarier. A lot of the green turns yellow and some yellow turns red.

Larry
 
Last edited:
lr172 said:
The ADS-b data coming from the ground just gives the dB levels of the returns. The display device takes that data against it's internal programs and conventions to display it. Each device decides for itself what color to paint for a 45 db return, for example. There is variation among the device manufacturers. Data from the same receiver in my 6 shows a much scarier picture on my GRT display than it does on my ipad/foreflight. Also, a 4 color algorithm usually looks scarier than the same data displayed in a 5 color scheme, due to the large ranges for each color in the 4 color scheme. FF has a slider for displaying a 4 color rendition. If you turn that on, the picture gets a lot scarier. A lot of the green turns yellow and some yellow turns red.

Larry
Click to expand...

I just checked the Stratux source code. The nexrad block intensities are arrays of integers so I suspect the RGB value is just the value of what's in the FIS transmission. These numbers can be offset by the software vendor to alter the shading.
 
Maybe Fis-B is showing the future weather :). I flew to Bowie, Texas 0F2 on Sunday morning. Sky clear. Ten miles out from Bowie the ADS-B weather on my Aera 660 showed red and yellow covering all of Bowie. The horizon showed blue. Five miles out, same thing. On top of Bowie airport, no cloud, no rain, nothing. FIS-B yellow and red. I was actually happy to discover this, since I am planning on flying to Oshkosh this year and in the past used XM which was quite accurate other than slow updates. I'll be leery of what FIS-B shows (or doesn't show) on this trip without cross-checking METARS, visuals and listening in on frequency.

BTW, my comment about showing future weather: 6 hours later, we had deep black sky, lowering clouds and 80 mph winds at my house in North Dallas. Lost my shade tree in the back yard in the first few seconds of the wind, rain and hail. Scary looking stuff. I wish I had the FIS-B readout for this. My Radar Pro app showed purple as it passed over my location. No weather alert from NOAA either.

Chris
 
rocketbob said:
I just checked the Stratux source code. The nexrad block intensities are arrays of integers so I suspect the RGB value is just the value of what's in the FIS transmission. These numbers can be offset by the software vendor to alter the shading.
Click to expand...

I had thought they were sending db levels, buy maybe they are sending identification of one of the 18+ blocks in the nexrad standard (5 db range for each color block). The likely problem is that the Nexrad convention for shading has 3 levels of green, yellow and red. Most displays only have a few colors to work with, so the developers are likely making arbitrary decisions on how to fit 9 nexrad blocks/colors (the basic green, yellow red that we see most typically) into 4 or 5 colors/shadings available on their UI.

I definately have seen different apps do different things with the same basic data. My computer representation of Nexrad will show me 3 levels of yellow, FF shows me two levels or yellow, with FIS-b data and GRT shows me one level of yellow.

Larry
 
Last edited:
I flew home from Evansville last summer under a solid overcast.. the Indianapolis area was painting yellow with lots of red but I could clearly make out the downtown area from 60 miles away. We pressed on and flew through areas shown in red on my GRT efis and barely saw a drop of rain and no turbulence. I attributed it to the composite data presentation, but the color coding issues described here would also be a factor in the experience.

Bottom line- trust your eyeballs. Anything in your cockpit can (and will) lie to you.
 
You must log in or register to reply here.
Back
Top