Vertically Integrated Liquid (VIL) is a radar-derived estimate of the mass of liquid water plus ice in a column above a point, calculated by integrating reflectivity through the depth of a storm and reported in kilograms per square meter.
What it is measuring
The VIL calculation traces back to Greene and Clark's 1972 work relating radar reflectivity to raindrop size distributions and liquid water content. At each grid point, the algorithm integrates a function of reflectivity from the ground to the top of the storm and reports the total in kilograms per square meter. To keep large hail from producing unphysical liquid mass estimates, reflectivity values above 56 dBZ are capped at 56 dBZ before the integration.
The result is best understood as a proxy for how much precipitation the storm is currently holding aloft, ice included, rather than a literal measurement of liquid water. Two storms with the same VIL can have very different vertical distributions of that mass, which matters when the reader wants to know how tall or how deep the water load actually is.
How forecasters use it
Warning forecasters watch VIL for two patterns. A rapidly rising VIL in a growing storm suggests the updraft is lofting and holding more precipitation, and a sudden VIL collapse often precedes a downburst or a hail dump as the load falls out. Absolute VIL thresholds for severe hail vary strongly with air mass and season, which is why the concept of a VIL-of-the-day has been used, comparing today's storms against a baseline for the environment they are in.
VIL density normalizes VIL by the storm's echo top height, which partly removes the tall-versus-short ambiguity and improves the correspondence to hail size. A high VIL density in a warm-season environment is a stronger hail signal than raw VIL, though it too is a guide, not a threshold switch.
Important limits
VIL was calibrated on the assumption that reflectivity comes from a rain-like distribution of drops, so the algorithm systematically misreads ice and hail even with the 56 dBZ cap. VIL is underestimated for fast-moving and highly tilted storms because the column integration assumes precipitation is stacked over one grid point. Colder air masses tend to run lower baseline VIL values, warmer air masses run higher, and the same storm strength reads differently in each.
Bright band contamination from the melting layer can inflate VIL in stratiform regions where no severe weather is occurring. Read VIL trends rather than single values, and always pair VIL with dual-polarization variables and reflectivity structure before assigning a threat.