Updated: Jul 19, 2019
If you haven't heard the acronym MRMS, you are not alone. There's no doubt that you will be hearing it more often in the coming months and years - of course, let everyone know where you heard about it first!
What is MRMS?
It actually consists of many radar-based products, but for all intents and purposes for this discussion, it's a radar mosaic. So, you ask, there are plenty of radar mosaics out there, why is this one so special? According to NOAA, "MRMS is 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." MRMS was made operational in 2014.
The goal of MRMS is to produce the best radar depiction (and other products not discussed here) to improve the decision-making capability for severe weather forecasts and warnings, hydrology, aviation, and numerical weather prediction. MRMS does this through a multi-radar and multi-sensor approach. Although, it's important to understand that not all MRMS radar products will use sensor data other than those from NEXRAD. The most popular products that use other sensor data are the reflectivities interpolated to constant temperature altitudes, the gauge-corrected QPEs, and the lightning probability just to name a few.
MRMS is now broadcast over FIS-B
What you probably didn't know is that as of April 23, 2019, the MRMS composite reflectivity radar mosaic is being broadcast to your FIS-B (ADS-B) receiver. More importantly, the substitution was completely seamless - your tax dollars hard at work. Vendors (certified and not) had no software changes to implement since the update was considered backward compatible. The FAA managed to have that fly in under the radar (pun intended) without any pomp and circumstance. Well, if you are not a student of airspace NOTAMs, you probably just missed it; shame on you. It's clear the FAA doesn't really understand how to do marketing. Their big reveal was to issue a bunch of NOTAMs for each ARTCC. Here's the NOTAM for the ZUA ARTCC as an example...
FDC 9/2633 (A0546/19) - AIRSPACE..SPECIAL NOTICE..THE FLT INFORMATION SERVICES BCST DATA SOURCE FOR ZUA ARTCC AIRSPACE IS CHANGING TO THE MULTI RADAR MULTI SENSOR (MRMS) WEATHER RADAR MOSAIC FOR FLT INFORMATION SERVICES BCST REGIONAL NEXRAD AND CONUS NEXRAD PRODUCTS. ALL FLT INFORMATION SERVICES BCST PRODUCT FORMATS REMAIN UNCHANGED AND EXISTING AVIONICS SHOULD REMAIN BACKWARD COMPATIBLE WITH THE NEW DATA SOURCE. THIS UPDATE WILL OCCUR BTN 1904231400 UTC AND 1904231600 UTC. ANY OBSERVED ISSUES SHOULD BE REPORTED VIA EMAIL TO ADSB@FAA.GOV. 23 APR 14:00 2019 UNTIL 23 MAY 14:00 2019. CREATED: 16 APR 16:43 2019.
What does this mean to you?
Likely the element that will have the biggest impact is the update rate of the radar mosaic. MRMS is built by the NWS every two minutes and FIS-B is broadcasting a brand new mosaic every two minutes. There are still inherent delays, but, in fact, it's likely the age of the MRMS radar mosaic you now see in the cockpit will be newer than the legacy radar product it used to broadcast. No, it's still not a real time depiction and should always be used as a strategic tool. This means it will more closely match what you see outside the cockpit which is great news.
The most critical element here is the date-time stamp. What is sent to your receiver is the actual NWS freeze time. That's the time or line in the sand where the NWS begins the process documented below to create the composite reflectivity mosaic.
Step 1: Combine most recent tilt from the Level 2 NEXRAD data feed as it arrives with tilts from existing and previous volume coverage pattern (VCP) mode to create an entire volume scan (referred to as a virtual volume).
Step 2: Perform quality control (QC) checks to remove all non-meteorological returns.
Step 3: Interpolate radar to 3D Cartesian coordinates using a spherical interpolation technique.
Step 4: Merge with other radars using an inverse (exponential) distance weighted mean to produce a 3D mosaic.
Step 5: Compute composite reflectivity by taking the column maximum reflectivity at each grid point.
The age old debate...
How old is the image relative to the actual weather? Of course, all of the processing above takes time and the mosaic considers a virtual volume which is typically a 4-6 minute period. Next, the results need to be distributed and broadcast. The results of the steps above are sent out over the NWS LDM feed where it is picked up by Harris and prepared to be transmitted via the ADS-B radio stations. Finally, your avionics vendor (e.g., Garmin) collects this data via your FIS-B datalink receiver (the mosaic is sent in continuous blocks) and it must be rendered on your portable or panel-mounted display. Notwithstanding the latency difference among vendors to render the data, the "age" of the refreshed mosaic will be 3-4 minutes old relative to the NWS freeze time. Then you will stare at that image for another two minutes until it's refreshed again.
In the end, there's no real way to determine the exact age of any particular pixel when it's initially rendered on your display. Best case scenario, that pixel is 4 minutes old relative to the weather outside the cockpit; worst case scenario it's 9 minutes old. This is primarily driven by the fact that the virtual volume from the NEXRAD sensors extends over a 4-6 minute window. On average you are looking at an image that's 5 or 6 minute old from reality and that ages for two minutes to 7 or 8 minutes old.
While there is always a chance that non-precipitation returns (clutter) find their way into MRMS, it's less likely due to the multi-radar algorithms used. Moreover, you still will have a regional radar depiction at the same resolution based on the 250 mile look-ahead factor and a national depiction that has been reduced in resolution by five times. MRMS has a bigger coverage than the legacy FIS-B radar depiction. For example, you'll further coverage off the coastal areas. Moreover, MRMS does include the Canadian Doppler weather radars data, but it's not being broadcast with FIS-B at this point in time due to delays in getting that component of the MRMS mosaic.
What's there to worry about?
After all, this is what you call progress, right? Yes, it is progress in the right direction. But given this was shipped out into the world without any formal and exhaustive testing, there could be issues not yet uncovered further hampered by the lack of promulgation from the FAA. Harris did perform some testing with legacy airborne equipment before the cut over and found no problems. But it wasn't an exhaustive test and it did not test all equipment. Keep an eye on that age of the data and note any weirdness that you may see. Report any unusual situations to your vendor so they can address any glitches that might be lurking in their software. Enjoy!
A big thanks to John Ferrara and Heather Reeves for their input. And if you want to learn more about aviation weather, consider purchasing the newest and most complete aviation weather book called Pilot Weather: From Solo to the Airlines.
Most pilots are weatherwise, but some are otherwise™
Weather Systems Engineer
CFI & former NWS meteorologist