Maximum Estimated Size of Hail (MESH) is a radar-derived estimate of the largest hailstone diameter a storm is likely producing, calculated from reflectivity measured in the sub-freezing part of the storm column.
How it is calculated
MESH is derived from the Severe Hail Index (SHI). SHI integrates reflectivity through the vertical column, weighting most heavily the reflectivity that lies between the environmental freezing level and the minus-20 degree Celsius level. That layer is where large hail is grown, so reflectivity there is treated as a stronger hail signal than reflectivity closer to the ground.
The MRMS implementation pulls its temperature profile from the operational mesoscale model, which lets the freezing level and minus-20 heights vary in space rather than being set to a single value across the domain. MESH then applies an empirical fit that maps SHI to a maximum expected stone diameter, historically expressed in inches or millimeters.
How forecasters use it
Warning meteorologists watch MESH climb in real time as a storm intensifies. Rising MESH values inside a persistent core support severe thunderstorm warnings for hail, and threshold crossings feed the hail tag on the warning product. The MESH Tracks product accumulates the maximum MESH along a storm's path, producing a swath that is used after the event to guide damage surveys and to reconstruct where the largest stones likely fell.
MESH is also a tuning input to broader severe weather guidance. Verification studies against reported hail sizes are used to update the empirical fit, and MESH swaths are one of the standard radar-derived proxies compared against local storm reports and hail pad networks.
Where it misleads
MESH is a radar-based estimate, not a ground observation. Storms embedded in strong deep-layer shear tilt away from vertical, which spreads the hail growth zone across horizontal distance and can leave a MESH underestimate over the ground-truth fall location. Left-moving supercells and storms with giant Bounded Weak Echo Regions can show a MESH hole under the updraft even as they produce very large stones on the flanks. Low-density, dry stones return less reflectivity for their size and read low.
Bright band contamination from the melting layer and small biases in the mesoscale model temperature profile propagate into MESH as errors that vary by season and location. Treat any single MESH value as a guide that says the storm is capable of stones near that size, not a measurement of what actually fell.