WMOD00440

• Title: HIPLEX-1: Statistical Evaluation
• Date: 4/4/1984
• Weather Modification - Doc Type: Report

~PRIL 19S4

MIELKE ET AL.

515

TABLE 2. Primary and secondary response variables
~ and final test statistics.

I CIC2
2 CIC5
3 CCR5
4 PICS
5 MVDS
6 AWCS
7 TFPI
S TFE
9a TIPA
9b TIPR
lOa RERC
lOb AER

Primary response variables:

Cloud ice concentration, 2 min after treatment
Cloud ice concentration,S min after treatment
Concentration of crystals rimed, 5 min after
treatment
Precipitating ice number concentration, S min
after treatment
Mean volume diameter of precipitating ice
particles, S min after treatment
Average liquid water concentration, S min after
treatment
Time to first precipitating ice (particles with
diameters ;;;;, 0.6 mm in -::oncentrations > 0.1
L-')
Time to first SWR-75 radar echo (15 dBZ)
Time to initial precipitation at + 100C level,
aircraft measurement
Time to initial precipitation at + 100C level, SWR-
75 radar (15 dBZ)
Radar-estimated rainfall at + 100C level, using a
, constant Z-R relationship
Aircraft-estimated rainfall at + 100C level

FCC5

Secondary response variables:

2 M2EA
3 M3EA
4 D2EC
5 D3EC
6 MAXZ
7 TIPB

Fraction of crystals which are columnar,S min
after treatment
Maximum area of the 20-dBZ echo
Maximum area of the 30-dBZ echo
Time duration of the 20-dBZ echo
Time duration of'the 30-dBZ echo
Maximum radar reflectivity factor
Time to initial precipitation at cloud base (15-dBZ
echo on SWR-75 radar)

Final test statistics:

la PCPA Proportion of experimental units that precipitate,
based on TIPA
I b PCPR Proportion of experimental units that precipitate,
based 011 TIPR
2a A VRA Average volume of precipitation per experi!llental
unit, based on AER
2b A VRC Average volume of precipitation per experimental
unit, based on RERC

minated in early 1982, only 20 total cases were available
for analysis, with 7 cases of Class A-I clouds (4 seeded
and 3 nonseeded) and 13 cases of Gass B clouds (8
seeded and 5 nonseeded). (Table 3 presents the cloud
selection criteria for A-I, A-2 and B clouds in HIPLEX-
1.) Consequently, the, sample sizes were insufficient to
achieve the stated objectives of HIPLEX-1. In partic-
ular, a large P-value suggests that either 1) there exists
no obvious effects, or 2) the effects were sufficiently
small that the small sample sizes precluded any chance
of detecting the effect. Furthermore, a small P-value
in the present context suggests that either 1) a very
strong effect was detected, or 2) a bad break in the
randomization scheme occurred, i.e., a type-I statistical
error.

4. Empirical results

The HIPLEX-l primary and secondary response
variables are listed in Tables 4 and 5, respectively.
Table 6 and Figs. 1 and 2 present univariate MRPP
comparisons of nonseeded and seeded experimental
units for the HIPLEX-l primary response variables

TABLE 3. Cloud selection criteria for HIPLEX-1.

A Class A-l cloud criteria:

Average cloud liquid water concentration greater than 0.5 g
m-3 over approximately a I-km-Iong cloud region
determined by 10 s of flight at approximately 100 m s.'
(A WCO)"
2 Average ice-crystal concentrationb less than 1.0 L -, in the
I-km-Iong (10 s of flight) cloud region of maximum average
liquid water concentration (AI CO)
3 Maximum ice-crystal concentrationb less than 5.0 L-' for any
I-km-Iong (10 s of flight) cloud region (defined by FSSP
liquid water concentration greater than 0.01 g m.3) during
the test pass (MICO)
4 Vertical air velocity greater than -1.0 m s.' in the region
defined by item I, but if the vertical velocity is greater than
10.0 m s.' and the buoyancy is greater than I DC, reject the
candidate (VVLO and BOYO)
5 Length of the test penetration more than 2 and less than S km
as defined by an FSSP liquid water concentration greater
than 0.01 g m-3 (LPNO)
6 No radar echo detectable on the aircraft weather radar (RFLO)
7 Ooud-top temperature lower than -60C but higher than
-l20C (CTTO)
S Cloud-base temperature higher than OOC (CBTO)
9 Minimum separation between the current test cloud and
previous test clouds greater than 15 km to insure the
meteorological independence of the clouds (DBCO)

B Class A-2 cloud criteria:

I Items Al to A9 above
2 An average wind direction between the surface and SOO mb
from 250 to 0400 true (A WDO)
3 A 3D-mb-thick stable layer present with its base between 0 and
. -lOoC and its top temperature at least 1.50C higher than
the temperature extrapolated from the base of the layer to
the top using pseudoadiabatic ascent (DTCO)
4 A 100C dewpoint depression present somewhere within the
3D-mb layer of B3 above (DPDO)

C Class B cloud criteria:

I Items AI-A3, AS, A6, AS, and A9 abovec
2 Cloud-top temperature lower than -60C but higher than
-200C (CTTO)
3 Vertical air velocity greater than -1.0 m S.I in the region
defined by item A I, but no other vertical velocity or
buoyancy restrictions (VVLO and BOYO)

"Symbols in parentheses indicate the names assigned to the vari-
ables in the computer programs.
b Measured by the depolarization signal from the PMS 2D-C probe,
not the regular 2D-C image data. Although the depolarization signal
is believed to indicate only about 25% of the actual ice-crystal con-
centration, it was used because the image data include spurious
images that could not be rejected in real time.
c Oouds that failed to meet criteria A4 and A 7 for Class A were
then evaluated as candidates for Class B.