WMOD00439

• Title: HIPLEX-1: Experimental Design and Response Variables
• Date: 4/4/1984
• Weather Modification - Doc Type: Report

APRIL 1984

SMITH ET AL.

499

an open-ended process because the number of possible
scenarios is boundless.

3. The experimental unit

The designated experimental unit for HIPLEX-l
was the small semi-isolated cumulus congestus cloud
(Fig. 1). It was the only type of cloud that was thought
to be well-enough understood to permit the design of
a physically meaningful randomized seeding experi-
ment. Clouds were sought that had diameters from
-2 to 8 km and depths from 2.5 to 4.5 km, and that
were visually distinct and separable from neighboring
cumulus congestus clouds. The desired cloud base
temperature was about +50C, with cloud top tem-
perature preferably no lower than -12oC. Other de-
sired conditions included cloud liquid water concen-
trations of -1 g m - 3, low concentrations of ice particles
around the -lOoC level, the absence of any radar echo,
and a cloud lifetime of about one-half hour after treat-
ment. These general conditions were, on the basis of
the exploratory studies and trial seeding experiments
carried out prior to the beginning of the randomized
seeding experiment, expected to lead to a sample of
clouds that would be amenable to seeding according
to the HIPLEX-l physical hypothesis (Section 5). Table
I presents detailed objective criteria that were adopted
to quantify these general conditions for cloud selection
in HIPLEX-l. These quantities were measured by the
project aircraft, as described below.
The test-case clouds were classified as Class A-I, A-

2, or B. Weak to moderate updrafts were required for
Class A-I clouds, the primary category for which the
experimental hypothesis was developed. Class A-I
clouds were considered unlikely to precipitate naturally
because of warm tops (-6 to -12oC) at the time of
selection and a presumed absence of ice multiplication
processes. Class A-2 clouds were clouds in the same
temperature range but in which an ice multiplication
process was considered likely to occur. (The criteria
for Class A-2 clouds were based on observations during
the exploratory studies, but no test-case clouds were
classified as Class A-2 during the randomized seeding
experiment. Evidence from the HIPLEX-I physical
studies suggests that the stated criteria for Class A-2
clouds were too narrow to separate out all of the cases
with active ice multiplication processes.) Clouds with
stronger updrafts or top temperatures between -12
and - 200C at the time of selection were placed in the
third category, Class B. Class B clouds were considered
likely to precipitate naturally, with or without ice mul-
tiplication, due to their relatively cold tops. The dif-
ferences in behavior between seeded and nonseeded
clouds were expected to be most readily detected in
Class A-I clouds and hardest to detect in Class B clouds.
Class B clouds were selected for experimentation only
when no suitable Class A clouds were available.
The initial selection of candidate clouds was made
primarily on the basis of visual appearance from the
project aircraft. Aircraft passing through and over the
top of the candidate clouds (Fig. 2) then made the
various measurements suggested in Table 1. In this

,',r
.r.

FIG. 1. Photograph of HIPLEX-I Test Case 9 (22 July 1979) looking south from the seeding aircraft
at the -lOoC level -17 min after the treatment time.