WMOD00404

• Title: Thailand Applied Atmospheric Research Program - Final Report - Volume II
• Date: 3/1/1994
• Weather Modification - Doc Type: Report

enhanced the rainfall from individual cells by over 100 percent, thereby replicating many of
the Florida results. In addition, rain increases of over 50 percent are indicated for t.he
experimental unit (i.e., the small mesoscale convective cluster) that covers nearly 2,0100
square kilometers (Rosenfeld and Woodley, 1989; 1993).

Finally, an operational seeding program in West Texas has provided evidence of areal rainfaIl
increases through dynamic seeding. This evidence comes from an assessment of five years
of warm season cloud seeding over a watershed serving San Angelo, Texas (Woodley and
Solak, 1989). Target-control regressions that had been derived from historical rainfall records
were used to estimate the effect of dynamic seeding in this program. The analysis suggests
more rainfall in each of the five years. .The probability of this increase happening by chance
is about 3 percent. An overall effect of seeding of about +17 percent for the target for all
years of operation is indicated. Sensitivity testing supports the interpretation that seeding
was responsible for a sizeable portion of this apparent rainfall increase.

In summary, the scientific evidence from cloud seeding research programs in Florida and
Texas that have employed dynamic seeding techniques indicates that rainfall can be
increased from convective clouds by over 100 percent on the scale of individual cells, by 25
to 50 percent on the scale of groups of convective clouds, and by 10 to 25 percent over targets
up to 13,000 square kilometers in size. As the target size increases, the fraction of convective
clouds within that target that are seedable and can be effectively seeded decreases. The
strength of the evidence for enhanced rainfall decreases as the scale ofthe rainfall increases.
The evidence is strongest for individual cells where the seeding signal is largest, and weakest
for large target areas where the seeding signal is weakest. Much remains to be known as to
how and why dynamic seeding works to increase areal rainfall, and this question is the basis
for continuing research in the United States.

2.4 Summary of Thai Cold Cloud Seeding Experiments

2.4.1 Introduction

At the outset of the Thai cold cloud studies, all participants agreed that it would be a mistake
to begin a full-scale area seeding experiment in Thailand based on what was known at the
time about the microphysical and dynamic characteristics of Thai Clouds. Such experiments
were felt to be too consumptive of time and money to warrant beginning with such an effort.
On the other hand, the participants recognized that years of basic meteorological studies are
not necessary before the potential of seeding can be evaluated. Too much is known from
seeding experiments in other climatologically similar areas to require beginning "at first
principles" in Thailand.

~

The components of the Florida (FACE) and Texas [SWCP (Southwest Cooperative Project)]
studies that focused on the convective cell proved successful with relatively small samples,
and they gave credibility to the overall effort. This result is not surprising when one
considers that the building blocks of convective cloud systems are the individual convective
elements, which consist initially of individual moist updrafts and later as downdrafts filled
with precipitation. These convective elements are often seen visually as towering cumulus
clouds and on radar as reflectivity cores or cells. Most of the water vapor feeding the cloud
is ingested via its updrafts. This water vapor, which is condensed in the updraft, may be
converted into precipitation particles, or it may be lost by mixing and evaporation into the
ambient air.

6