The Multi-Radar Multi-Sensor (MRMS) system is an automated system that combines data from many radars with surface and upper-air observations, rain gauges, lightning detection, satellite data and numerical weather prediction models. It turns them into seamless, rapidly updating national mosaics and derived products used by the National Weather Service.
Where it comes from
The National Severe Storms Laboratory (NSSL) developed MRMS jointly with the Cooperative Institute for Mesoscale Meteorological Studies. MRMS was made operational at the National Centers for Environmental Prediction (NCEP) in 2014.
NCEP Central Operations now distributes the operational products. Radar inputs include Canadian radars, commercial radars and Terminal Doppler Weather Radars (TDWR), not only the NWS network.
Speed and resolution
MRMS works on a grid of about 1 km with a 2-minute update cycle, and its 3D reflectivity mosaic has 33 vertical levels. NOAA describes more than 100 products from the system.
Not every product refreshes every 2 minutes. Longer rainfall totals, for example, update hourly.
What it feeds
The product list includes reflectivity mosaics, Multi-Sensor rainfall estimates, the Maximum Estimated Size of Hail (MESH), Azimuthal Shear and Rotation Tracks, lightning density and lightning jump, echo tops and Vertically Integrated Liquid (VIL).
MRMS also estimates surface precipitation type and rate, and its radar-only rainfall totals are built by adding up that rate over time. MRMS rainfall also drives an NSSL flash flood modeling system called FLASH, which produces forecasts at 1 km and 5-minute resolution.
Where it can mislead
Combining radars does not remove what each radar cannot see. Far from a radar the beam widens and can average out rain features.
Products such as MESH and rotation tracks are radar-derived estimates. They point to where to look, and confirmation still comes from reports on the ground and damage surveys.