Skip Navigation Links 
Go to NOAA's main page Go to the NWS main page National Weather Service Forecast Office

Houston/Galveston, Texas
background graphic

Local weather forecast by
"City, St" or zip code

   You are at NWS Houston/Galveston » Research Projects » VIL Density and Associated Hail Size Along the NW Gulf


Charles A. Roeseler, NWSO Houston, TX, and
Lance Wood, NWSO Lake Charles, LA


A large sample of hail events were studied across parts of the northwestern gulf coast to determine if already established vertically integrated liquid (VIL) densities from other parts of the country were effective in a warm marine environment such as the TX/LA gulf coast. Mathematically defined, VIL density=VIL/echo top, where VIL is in units of g/m3 and echo tops are measured in feet and converted to meters. VIL is a function of reflectivity and is defined for the National Weather Service (NWS) Doppler radar as:

VIL (kg/m2)= 3.44x10-6[(zi+zi+1)/2] 4/7h

where zi and zi+1 represent radar reflectivity at the bottom and top of a layer defined as h (depth in meters) (Federal Meteorological Handbook 11).


114 hail events that occurred in the Houston/Galveston (HGX) county warning area and 60 hail events from the Lake Charles (LCH) county warning area were analyzed. Dates and times of hail events were recovered from severe weather logs maintained at each office. Hail size varied from 0.75 in. to greater than 3.00 inches in diameter. Data was collected from 9/92 through 12/96 at HGX and from 1/95 through 5/97 at LCH. The hail cases studied were gathered from WSR-88D data at HGX, LCH and GRK (Fort Hood). Volume Coverage Pattern (VCP) 21 and 11 were employed during this study with a majority of the cases examined using VCP 21. VILs were selected from storms depicted on radar near the location and time as recorded on the severe weather log. Once a storm was identified, a maximum VIL was located and a matching echo top selected. Due to storms tilting with height, a 1:1 or pixel to pixel relationship was not always achieved. The selected echo top could be as much as two pixels away from the selected VIL value. Although choosing an echo top two pixels away from the VIL value may seem insignificant, the differences can be very large. Echo tops may vary as much as 10,000 feet in two pixels producing a pronounced affect on VIL densities, in most cases producing a lower VIL density than expected. See Amburn and Wolf (1996) for additional errors.

Data collection was further complicated during multi-cell events because multiple cores often existed, changing with each volume scan. In some situations, VIL and echo top values were rounded down to the lowest value in a range if similar values existed in a multi-cell state. Actual VIL values and echo tops were used during isolated activity or in situations when it was obvious that the maximum values indicated on radar corresponded to the investigated storm.

NWS severe weather warnings are verified when wind gusts exceeding 58 mph or 0.75 in. hail are reported. This has an effect on verifying hail size. Once either parameter is met, an active search for other severe weather parameters is lessened due to time and manpower constraints. In addition, human error in judging hail size, especially between 0.5 and 0.75 in. has an effect on NWS warning verification statistics and therefore on VIL densities which are calculated from these statistics. A small sample of 0.5 in. hail cases from HGX were examined and VIL densities were similar to 0.75 in. hail. The sample size was rather small and was excluded from the results of this paper. There is no guarantee that the hail size reported is representative of the size which occurred during a particular event.


Amburn and Wolf (1996) established that a relationship exists between VIL density and hail size. The goal of this study was to determine if similar results could be achieved in a coastal environment. For this study, hail size was categorized as severe (0.75 - 0.99 in.); large (1.00 - 1.99 in.) and very large (> 2.00 in.). The results indicate a relationship between VIL density and hail size, especially as hail size increases. A VIL density of 3.50 g/m3 correctly identified 125 out of 174 cases or 72% of all severe (0.75 in. or greater) hail events. This is less than either Troutman and Rose (1997) who reported 79% of all severe hail cases identified in the Nashville county warning area with a VIL density of 3.50 g/m3 and considerably less than Amburn and Wolf's findings (1996) in Oklahoma. Using a VIL density of 3.50 g/m3, they identified 90% of all severe hail events. A VIL density of 3.75 g/m3 correctly identified 72 out of 89 cases or 80% of all large (1.00 in. or greater) hail events and a VIL density of 4.25 g/m3 correctly identified 6 out of 6 or 100% of all very large (>2.00 in.) hail events (figure 1). 84% (147 out of 174) of all severe hail events would be identified if the VIL density was lowered to 3.28 g/m3; however a VIL density of this value would produce an unusually high number of false alarms if used strictly as a warning threshold. Therefore, it is suggested that a VIL density of 3.75 g/m3 and 4.25 g/m3 be used along the northwest gulf coast as a threshold for 1.00 in. hail and 2.00 hail, respectively (figure 2).

VIL density associated with low top convection, defined as 30,000 feet or less, producing 1.00 in. hail or greater produced an unusually high average VIL density level of 5.73 g/ m3. Low top convective VIL densities were consistently higher than other convective storms producing 1.00 in. hail. Seasonal variations in VIL density were examined, the only difference of note occurred with low top convection with a frequency maximum in the cool season of January through March.


The northwest gulf coast climate is much different than previous VIL density study areas such as the southern plains and the freezing level is at a much greater height, so large hail which develops in the upper levels of a thunderstorm has more time to melt as it descends. Mid and upper level winds are usually much weaker along the gulf coast and developing hail will fall through portions of the updraft. Warmer raindrops accelerating upward will melt the falling hail as the two phenomena collide near the edge of the updraft. It is important to remember that VIL density only indicates hail aloft (Amburn and Wolf, 1996), not what is or will be falling to the surface. Preliminary results indicate that VIL density does a good job of predicting hail size if the freezing level is low and a poor job if the freezing level is high. Forecasters should remain cognizant of the type of airmass in which thunderstorms are developing. Along the northwest gulf coast, VIL density should be used as a guide to forecast possible hail size development, not necessarily as the warning decision tool. VIL densities of 3.50 g/m3 would not aid in differentiating between 0.50 and 0.75 in. hail. As VIL levels increase to 3.75 g/m3, the threat for 0.75 in. hail or larger increase. Confidence levels in warning for large hail (greater than 1.00 in.) increase dramatically as VIL densities increase. Radar operators along the gulf coast must continue to examine other data such as storm structure, freezing level and storm history. Further research is required to address the effects of thermodynamic variables, changes in volume scanning strategy and elevated reflectivity cores on VIL density and hail size.

  • Amburn, S. and P. Wolf, 1996: VIL Density as a Hail Indicator. 18th conference on Severe Local Storms. San Francisco, CA, Amer. Meteor. Soc., 581-585.
  • Federal Meteorological Handbook 11, 1991: Doppler Radar Meteorological Observations, Part C, Office of the Federal Coordinator for Meteorological Services and Supporting Research, U.S. Department of Commerce/NOAA, Washington, D.C.
  • Troutman, T. W. and M. Rose, 1997: An Examination of VIL and Echo Top Associated with Large Hail in Middle Tennessee. NWS Southern Region Technical Attachment SR/SSD 97-15 5pp. Fort Worth, TX.
Weather Hazards | Weather Forecasts | Present Weather
Past Weather | Tropical Weather | Contact Us
Houston AMS | What's New?

National Weather Service
1353 FM 646 Suite 202
Dickinson, Texas 77539
Web Master's E-mail:
Page last modified: 2-Feb-2006 1400 GMT
Privacy Policy
About Us
Career Opportunities