MESH
Hail and wind · Cloudbreak team · Updated October 29, 2026 · 10 sources
Definition
MESH, the Maximum Estimated Size of Hail, is a radar estimate of the largest hailstone a storm is producing, based on how strong its echoes are in the parts of the storm cold enough to grow hail. It is a gridded product of NOAA's Multi-Radar Multi-Sensor (MRMS) system, computed in millimeters and shown in inches on NWS displays.1,2
How it works
Radar cannot see a hailstone's size directly. MESH infers it: strong echoes high in a storm, where the air is below freezing, mean the storm is growing large hail. The estimate is for hail aloft, and the hail that reaches the ground can be larger, smaller or melted.
MESH is built on the Severe Hail Index (SHI), published by Witt and others in 1998. SHI converts radar reflectivity into hail kinetic energy and adds it up through the depth of the storm. Echoes get partial weight from 40 dBZ and full weight at 50 dBZ; layers get partial weight above the 0°C (32°F) level and full weight at -20°C (-4°F).3,2,4
In millimeters, MESH equals 2.54 times the square root of SHI. Witt and others fit that curve to the 75th percentile of 147 hail reports, so about 75% of observed hail should fall below the estimate. All the reports came from Oklahoma and Florida, and none was smaller than 0.75 inch.5,1,2
NOAA's National Severe Storms Laboratory (NSSL) runs it in MRMS, which merges many radars into a 3D grid and takes the 0°C and -20°C heights from numerical model analyses. MESH comes out on a 0.01° grid every 2 minutes. Merging radars makes up for the cone of silence above each radar, beam broadening far from it and terrain blockage.3,1,6
Hail swath maps show the highest MESH at each grid point over a period. MRMS keeps these maximums over 30, 60, 120, 240 and 360 minutes and 24 hours (1440 minutes), and NSSL says they help with warnings and with checking afterward where hail fell.7,4
Example
In 2013, IBHS field teams measured hailstones across two storm swaths and compared them with NSSL's multi-radar MESH:8
| May 19, 2013, Wichita, Kansas | May 20, 2013, near Antioch, Oklahoma | |
|---|---|---|
| Radar | WSR-88D a few miles away; beam 300 to 400 ft above the road | WSR-88D about 40 miles away; 0.5° beam 2,700 ft up |
| Largest measured hail | 1.26 inches (112 stones at 9 sites) | 1.5 to 2 inches over a 2 to 3 mile stretch |
| MESH error | Too high by a mean of 0.24 inch | Too low by a mean of 0.31 inch |
In the Wichita case, MESH and the radar's hail classification both showed hail west of one site, but teams found none to measure. The report gives three possible causes: stones too small to measure, melting, or radar error.8
Known limits
MESH is an estimate of hail aloft, not a measurement on the ground, and it can be off in either direction. The main known sources of error:
- MESH tends to underestimate hail in storms tilted by strong wind shear, in left-moving supercells, in supercells with a giant bounded weak echo region (BWER), and in storms with low-density, dry hailstones.1
- It relies on the model's temperature profile, and it does not remove bright band contamination.1
- Tested against 5,954 hail reports from 30 storm days in 2013 and later, the original equation underestimated small hail and overestimated large hail relative to the 75th percentile of reports.5
- The MESH value that best separates severe (1 inch or larger) from smaller hail is 1.14 inches (29 mm) in two studies and 1.19 inches (30 mm) in a third. Two studies found no MESH value with skill for 2-inch hail.2
- From 2012 to 2019, MESH showed about 20 more severe hail hours a year than Storm Data across much of the High Plains. The SPC authors judge that the true rate lies between the two, and say MESH adds value at night and where few people live.2
Where you'll see it
- NWS forecast offices, aviation weather units and the Storm Prediction Center, through NWS systems and the public MRMS Operational Product Viewer.1,6
- The public MRMS data feed from NOAA's National Centers for Environmental Prediction (NCEP), as GRIB2 files such as MRMS_MESH_Max_1440min, the 24-hour hail swath.9,6,7
- MYRORSS, a NOAA reanalysis hosted on AWS, with radar hail size estimates on a 0.01° grid from April 1998 through December 2011.10
- Hail swath maps of accumulated MESH, which NSSL says help verify where hail fell, and field studies that compare those swaths with stones on the ground.4,8
Questions
- What does MESH mean on a hail map?
- Maximum Estimated Size of Hail: the radar-estimated largest hailstone at each grid point. Swath maps show the highest value over a period from 30 minutes to 24 hours.1,7
- How accurate is MESH hail size?
- It is designed so that about 75% of observed hail falls below it. In two 2013 field cases, it was too high by a mean of 0.24 inch in one and too low by 0.31 inch in the other.2,8
Cloudbreak
Cloudbreak's proprietary predictive storm engine, built with AI and backed by meteorologists, tracks every storm against every building, before it hits and after it passes.
Sources
Written by the Cloudbreak team. Numbers, dates and rules cite the published sources below.
- 1. NOAA NWS Warning Decision Training Division: Maximum Estimated Size of Hail (MESH)
- 2. American Meteorological Society, via NOAA Storm Prediction Center: An Hourly Climatology of Operational MRMS MESH-Diagnosed Severe and Significant Hail with Comparisons to Storm Data Hail Reports (Wendt and Jirak, Weather and Forecasting, 2021)
- 3. NOAA NWS Warning Decision Training Division: Severe Hail Index (SHI)
- 4. NOAA National Severe Storms Laboratory: Severe Weather 101: Hail Detection
- 5. American Meteorological Society, via PubMed Central: Severe Hail Fall and Hailstorm Detection Using Remote Sensing Observations (Murillo and Homeyer, Journal of Applied Meteorology and Climatology, May 2019)
- 6. NOAA National Severe Storms Laboratory: NSSL Projects: Multi-Radar/Multi-Sensor System (MRMS)
- 7. NOAA National Severe Storms Laboratory: Operational MRMS GRIB2 Tables (MRMS Version 12.2)
- 8. IBHS: 2013 IBHS Characteristics of Severe Hail Field Research Summary (Giammanco and Brown, February 25, 2014)
- 9. NOAA National Centers for Environmental Prediction: MRMS Data: MESH_Max_1440min
- 10. NOAA Open Data Dissemination, Registry of Open Data on AWS: NOAA Multi-Year Reanalysis of Remotely Sensed Storms (MYRORSS)