WMOD00316

• Title: Comments on "Further Exploratory Evaluations of Grossversuch IV Using Hailpad Data: Analysis of Hail Patterns and Stratification by Storm Type for Seeding Effect"
• Date: 6/6/1991
• Weather Modification - Doc Type: Report

Reprinted from JOURNAL OF ApPLIED METEOROLOGY, Vol. 30. No.6, June 1991
American Meteorological Socif'ty

Comments on "Further Exploratory Evaluations of Grossversuch IV Using Hailpad Data:
Analysis of Hail Patterns and Stratification by Storm Type for Seeding Efl'ect"

A. S. DENNIS

United States Bureau of Reclamation, Denver, Colorado

PAUL L. SMITH AND JAMES R. MILLER, JR.
Institute of Atmospheric Sciences, South Dakota School of Mines and Technology, Rapid City, South Dakota
14 September 1990

The article by J.-F. Mezeix (1990) provides a num-
ber of detailed analyses of hail pad data collected during
Grossversuch IV and interesting suggestions for future
hail suppression experiments. His suggestion that future
experiments take into account the possibility of dy-
namic effects of seeding appears to be particularly
valuable. However, two points, which are made more
than once in the article, and one omission puzzle us.
In his conclusion, the author states that "to our
knowledge, none of the experiments conducted to date
have dealt directly with a possible effect on the hailed
area. In other words, the area factor has never been
controlled." As operational cloud seeders have been
attempting for years to reduce the areas affected by
damaging hail as well as hail intensity, it is reasonable
that an area of hailfall be used as a response variable
in hail suppression experiments. In fact, area of hail fall
is listed as a secondary response variable (SG) in the
paper "Main Results of Grossversuch IV" of which
Dr. Mezeix is a coauthor (Federer et al. 1986). Perhaps
it was because of the close correlation between SG and
global hailpad kinetic energy EG, which was a primary
response variable in Grossversuch IV, that the designers
did not think it necessary to include SG in the list of
primary response variables. An evaluation of a ran-
domized crossover experiment in the Greek National
Hail Suppression Program (Flueck et al. 1986) in-
cluded as response variables the numbers and per-
centages of hail pads hit; these are essentially measures
of the hailed areas. Changes in the area hailed by in-
dividual storms would also be reflected in seasonal
summaries of insured losses, which have been used to
evaluate operational hail suppression projects in the
north-central United States (e.g., Dennis et al. 1980;
Miller and Fuhs 1987; Smith et al. 1987) and France
(Dessens 1986).

Corresponding author address: Dr. Arnett S. Dennis, Water Aug-
mentation Group, Mail code 0-3720, Bureau of Reclamation, P. O.
Box 25007, Denver, CO 80225-0007.

The second point concerns the accumulation zone
hypothesis, which we understand to require the pres-
ence of supercooled rather than frozen raindrops. Only
if such supercooled drops are present can the seeding
as practiced in Grossversuch IV produce additional
hailstone embryos quickly enough to have significant
impact on the hail development. In discussing a fa-
vorable but inconclusive statistical result for type I
storms (single evolution cells), Mezeix ( 1990) states
that "as a result of seeding in the -50 to --lOoC zone,
there could be rapid formation and additional frozen
drop embryos in a high water content zone (accu-
mulation zone) followed by sufficient embryo com-
petition for melting to give a reduced number of hail-
stones during the hailfall." Data collected during
Grossversuch IV by the South Dakota School of Mines
and Technology T -28 aircraft in 1982 and 1983 showed
that at the -80C level the high-reflectivity zones in the
( nonseeded) hailstorms studied contained frozen pre-
cipitation particles rather than supercooled raindrops
(Waldvogel et al. 1987). It is true that the T-28 pen-
etrated only one "type I" storm, on 21 June 1983,
which evidently produced no hail at the ground. Wald-
vogel et al. did not discuss that storm in detail, largely
because the radar-identified big drop zone (rBDZ)
signature was no longer present during the aircraft
penetrations. However, the observations show that even
on the first penetration shortly after conditions had
dropped below the rBDZ criteria, only ice particles
much like those reported in the other storms investi-
gated were present in the high reflectivity-updraft zone.
Recently reported polarimetric radar observations
(Holler 1990) from a region not far away from the
Grossversuch IV site also provide no indication that
the rBDZ of a storm contains any accumulation of
large supercooled raindrops.
It is difficult to see how seeding of such zones with
artificial ice nuclei could lead to formation of additional
hail embryos capable of competing against the natural
t:mbryos already present. We suggest that any future