Weather Modification Program

The program has been suspended due to funding issues.


A. Hail Suppression: We do not claim to be able to totally eliminate hail damage to crops and property in all storms. However, we are able to substantially reduce hail damage from what would have otherwise occurred during a growing season. Estimates of average damage reduction have ranged from 15% to 35%, or more. Total historic damage statistics have been used in each of the evaluations to arrive at their findings, however, we believe hail reduction now is likely to be on the order of 35% to 50% , or more, primarily due to the increased numbers of aircraft available, better storm targeting for seeding and increased seeding agent dispersal capability.

A 37:1 Cost-to-Benefit Ratio was calculated by the Kansas Water Office in 1994 for hail reduction to crops alone. That is, on average, for every dollar of program expenditure, $37 was returned to the sponsoring counties in the form of greater crop yields. The cost-to-benefit calculation did not include the greatly reduced crop hail insurance premiums that counties in Southwestern and Western Kansas have enjoyed since the program began in 1975. This reduction in crop insurance premiums over the years, by itself, is more than sufficient to pay yearly program costs…area farmers now have equal coverage for much less money than in 1975 (not including inflation).

B. Rain Augmentation: Although other weather modification programs similar to ours have shown statistically significant increases in rainfall of 9% – 10%, approximately, the WKWMP has not been able to show a statistically significant change in rainfall, either an increase, or decrease. We believe there is something fundamentally flawed in how the statistics are derived rather than what is happening in reality. Too much science has been done in this facet of weather modification for it not to be a workable technology given proper cloud conditions. On the WKWMP isolated clouds seeded have provided sufficient proof to us that rainfall increases in some clouds can increase to 1,000%, or more, cover hundreds of square miles and never pass over an official rain gauge. When averaged out against what the added moisture means to the entire target area of many thousands of square miles, the percent increase per square mile is very little, but what is observed is still unrecorded for official data analysis.

For several years we have tried to experiment with different seeding methods to increase rainfall. Many are promising. Extenuating factors caused problems in most of our more recent experiments. We will continue experimenting in the future. We believe there may be a good way, a better way and a best way in which to seed many cloud systems to increase rainfall. We need to be able to select the method of seeding which makes the most sense at the time and use the one which can be employed to obtain the best results with as little cost as possible.

It is possible to increase rainfall from some clouds. The clouds which best respond to such stimulation are ones in which rainfall would be produced naturally and become “parent” clouds…ones which would reproduce themselves through new, persistent, natural cloud growth attaching to themselves. Other clouds can be seeded which will produce lesser overall rainfall, but valuable nonetheless. There are a variety of clouds which are capable of responding to cloud seeding treatment and a variety of ways in which to treat them.

Not all clouds are capable of producing rainfall. We do not target candidate clouds for rainfall stimulation which do not appear to have a reasonable expectation of producing rainfall. Occasionally planes may be seen flying around some clouds from which rain doesn’t fall. However, such non-seeding flights are made only to ascertain what the cloud characteristics are in case seeding does become a possibility. If seeding is begun on marginal clouds and abruptly ended, the cloud has almost always collapsed for some natural reason before the seeding agent was able to rise into the sub-freezing domain within the cloud where a reaction may have been created. In such cases no effect was created and the collapsed cloud was due to natural circumstances. This often happens in Kansas.

In the past four years we have seen that the Target – Control approach to performing rainfall analysis is fundamentally flawed. The seasonal rainfall in NW Kansas is highly correlated to the rainfall in the WKWMP primary target area within Southwestern and Western Kansas. Many big storm days often occur in one region while perhaps not a whisper of storm is found in the other. Such storms occur at different times of the year when cloud moisture may be significantly different. If storms would pass through both regions every time, the statistics would be more believable.
Some people mistakenly think that by reducing hail in large, severe storms we also reduce a severe storm’s usable rainfall. We should be clear on one item. When precipitation comes in the form of crop- and property-damaging hail, it does no one any good and is not useful (or usable) precipitation. It is true that by reducing the intensity within the most intense part of the storm core (where hail is found within a severe storm) the rate of precipitation within that core is lowered, but this is exactly what has to be done in order to achieve some degree of successful hail suppression. The extremely high storm intensity readings by radar measure the high presence of hail within that core. However, in large severe storms the seeding agents become well-dispersed within the storm and tend to enhance the precipitation effects within much of the rest of the storm by spreading into the non-hail bearing parts of the cloud. Another effect of seeding a large, severe storm is that many of the ice crystals created move into the strongest part of the updraft and enter the high updraft/high intensity core where they are subsequently taken to the top of the cloud into its anvil. This provides another rainfall benefit not generally theorized in weather modification circles (except here). These ice crystals were previously thought to be totally lost in the anvil, wasted seeding agent.

However, it must be the case that they are somehow dispersed within the total anvil and sub-anvil areas containing a deep layer of ice crystals. When these ice crystals “settle out” from the rear part of the anvil of large storms they fall back to earth, warm up to become rain drops, scavenge other moisture from the cloud and become a continuous, light rain falling for a half-hour to an hour, or more, over hundreds, if not thousands of square miles. This “settling out” effect of the anvil’s ice crystals also creates the process causing the “rear inflow jet” in which the precipitation loading and evaporational effects from the rainfall moves airflow forward to the leading edge of the severe storm itself and provides convergence which helps sustain the storm’s persistence.


Probably the first rain stimulation proposal reasonably based upon science was by James P. Espy. In the April 5, 1839 issue of the National Gazette and Literary Register of Philadelphia, Espy proposed building large fires to generate updrafts. He reasoned that in a humid atmosphere cumulus clouds would eventually develop and produce rain. There are no records indicating the scheme led to any field trials, but in the 1880’s Congress did appropriate $10,000 to conduct some field experiments based on an old, widely-held idea that “it always rains after a battle”. Afterward, tests were performed with explosive charges carried aloft in balloons and optimistic reports followed. In the 1930s work done by Tor Bergeron and W. Findeisen led to the concept that clouds may contain both supercooled water and ice crystals. This led further to the concepts of “warm rain” and “cold rain.”

Modern scientific cloud modification had its serious beginnings in the late 1940s in the General Electric Laboratories at Schenectady, New York. There the scientists, Drs. Schaefer, Langmuir and Vonnegut, used dry ice and silver iodide as ice nucleating agents during in these early laboratory and field trial. The ice nucleating agents for cloud seeding have changed with time, Most seeding agents in use today to suppress hail continue to use formulations with silver iodide as one of their components. Presently, formulations used in Kansas are leading-edge materials within the industry. In 1972 the Kansas Legislature took a giant and progressive step forward when it enacted the Groundwater Management District Act. The act enabled interested groups to organize to implement area water conservation programs for themselves. Western Kansas Groundwater Management District #1 (WKGMD #1) thus became a legal entity of the State of Kansas.

As Groundwater District supporters began identifying program goals and specific objectives, an early objective was to organize, design and implement an operational weather modification program to seed convective clouds to increase rainfall and help alleviate the losses of sub-surface water in Western Kansas. The decision to implement such a program came after thoroughly reviewing results from the research program known as the Kansas Cumulus Project (KANCUP) and from state-sponsored seeding programs being conducted in North Dakota and South Dakota.

WKGMD #1 envisioned a perennially-supported program covering a large area in Western Kansas which would operate during the period in which crops were planted, grown and harvested. The program objectives would be to:

(1) increase area rainfall by seeding selected clouds in the absence of severe or potentially severe weather

(2) decrease the occurrence of crop-damaging hail by seeding potentially severe storms

(3) demonstrate the feasibility and effectiveness of projects of this type in the Western High Plains states

The weather modification program that was implemented in 1975 was first known as the Muddy Roads Program; later, it was changed to the present Western Kansas Weather Modification Program (WKWMP).


Much is still unknown about how some clouds grow, mature and dissipate with the passage of time, especially very severe storms. During the crop-growing period of the year rapidly growing convective clouds can quickly become severe producing hail which destroys both crops and property as well as producing highly destructive surface winds, including occasional tornadoes. The following is a brief explanation of how such convective clouds can form and the theory behind why hail suppression and rain stimulation is technologically feasible.

In order for a convective cloud to form air containing water vapor must first be cooled to condensation. Rising air can do this. Rising air cools by expansion as atmospheric pressure decreases causing it to cool adiabatically. Eventually, the air cools to a temperature at which the atmospheric water vapor in it condenses into water droplets. Condensation occurs first upon microscopic particles called cloud condensation nuclei CCN. CCN particles are relatively abundant in the world atmosphere and includes dust, smoke and salt particles. When a collection of these water droplets have grown to sufficient size, they are seen as clouds. In Western Kansas some of the mechanisms which causes air to lift and convective clouds to form are:

(1) surface heating – returning solar radiation to the atmosphere – warm air rising

(2) intruding cold and warm frontal systems – forcing air to lift over its advancing boundary

(3) relatively cold air in the upper atmosphere sinking into warmer air ahead of it causing warm, moist air to be displaced, or forced, upward

(4) upslope air flow which moves moisture from lower altitudes in Eastern Kansas into the higher altitude areas of Western Kansas and Eastern

(5) regions of horizontal convergence created by troughing at the earth’s surface, or aloft, forcing air to rise as it is squeezed together

(6) upper level low pressure systems (usually associated with low-level convergence, sinking cold air and/or upslope effects)

(7) convective scale interaction resulting from thunderstorm outflows digging under warm, humid air acting much like a mini-cold front forcing air
ahead of it to lift rapidly

(8) gravity waves—very small, internal perturbations traveling through the atmosphere, many of which are created randomly and not always easily
detectable in real-time

Other atmospheric particles are known as ice nuclei (IN), particles upon which, if found in condensed water droplets, enhance droplet freezing. Ice
crystals also may form directly from water vapor upon ice nuclei. Despite the atmospheric abundance of CCN, there is a relative scarcity of IN particles. It is this natural condition which we wish to address with cloud seeding.

Clouds can be made up of unfrozen water droplets, ice crystals or a combination of them. Within a convective cloud having a portion of it colder than freezing, some of the sub-freezing water droplets remain in a liquid state, and termed “supercooled”. Convective clouds often create a condition in which both unfrozen water droplets and ice crystals co-exist simultaneously. It is the supercooled cloud volume that is critical to the formation of rain and hail. Supercooled water can remain unfrozen down to as low as -40 C (-40 F) before spontaneously changing to ice. When spontaneous freezing occurs, it is termed homogeneous nucleation.

Supercooled water droplets containing ice nuclei freeze first. The speed at which supercooled water droplets convert into ice crystals increases as cloud temperature decreases, that is, as clouds grow in height above the freezing level. The process of vapor deposition starts to have a significant effect within clouds when ice crystals and supercooled water exist in the same medium. Surface pressures over ice crystals are lower than those over water droplets which creates a pressure gradient between them. This gradient causes liquid to flow from the droplets to the ice crystals, thereby growing at the expense of the droplets. Once ice crystals develop, they continue growing rapidly by using up surrounding water vapor and cloud water from nearby water droplets. Continuous unequal movements of water droplets and ice particles inside convective clouds ensure random collisions of ice and water droplets. The collisions promote the processes of coalescence, accretion and aggregation. Coalescence is a process in which the unfrozen water droplets collect other water droplets by impact, the freezing occurring after the impact. Accretion, or riming, occurs when droplets freeze upon impact with cloud ice particles. Aggregation is the process in which ice particles collect or attach to other ice particles. In advanced stages of cloud growth, ice particles will shatter, coalesce, grow larger and repetitively collide in a complex manner through the processes just mentioned. When the various sizes of ice particles eventually fall out of the cloud and drop below the freezing level, they begin melting. If melting is not complete, then hail, graupel or snow arrives at the ground as precipitation instead of rainfall.

The sizes and concentrations of all nuclei present in the atmosphere as well as their chemical and electrical properties all combine in important ways to determine how efficiently a cloud system can produce precipitation. Although there are massive amounts of water vapor in the atmosphere at any time, precipitation won’t occur if certain conditions required for the formation of precipitation are absent.

Two cloud types produce all precipitation: “warm clouds” and “cold clouds”. A “warm” cloud, is one in which its temperature is not below freezing and does not produce ice crystals in its cloud volume. The warm cloud is generally characterized by a relatively slow growth. Cloud water droplets eventually may grow to sufficient size and weight to fall from the cloud if given enough time. While falling, cloud droplets collect other cloud droplets by scavenging them along their downward paths. Although this type of cloud occasionally appears in Western Kansas, it doesn’t play a dominant role in producing precipitation here. However, large size warm-rain drops can be important embryo sources in the production of hail when they merge into sub-freezing clouds that are not of the warm-rain variety type and become carried aloft rapidly by updrafts where they freeze and grow into large hail.

Most important to Western Kansas is the “cold” cloud. Cold clouds have a portion of their volume which have grown into a temperature range below freezing. It is the interaction between supercooled water drops and ice crystals which initiates and promotes the process most responsible for producing
significant precipitation in Western Kansas.

The prevailing hypothesis under which the WKWM Program hail suppression portion operates is that hailstones grow to large sizes because there are too few ice crystals formed naturally in clouds during vigorous thunderstorm growth, thereby allowing relatively abundant supercooled cloud water to collect upon relatively few numbers of ice particles and other hail embryos. All too often those particles grow into hailstones too large to melt before reaching the earth’s surface. Current theory is that by vastly increasing ice crystal concentration within these ice crystal-deficient clouds, we are strongly increasing the competition for available cloud water thereby preventing hailstones from growing to a size large enough to damage crops and property. Property type, crop type, stage of crop growth and hail size are all important factors in determining damage severity.

Research has found that hail growth and movement within storms, especially very severe ones, can be very complex. However, most long-term hail suppression programs use similar seeding hypotheses and seed clouds in much the same way we do in Western Kansas. In 1994 the Kansas Water Office published the most recent evaluation of the WKWMP finding a 27% reduction of crop-hail damage statistically significant at the 5% level with a Benefit-to-Cost ratio of 37 : 1.

The hail suppression seeding agents used on the WKWM Program are either silver iodide-based or dry ice and are delivered directly into growing clouds by aircraft. The silver iodide seeding agents are vaporized in the updrafts found at cloud base, whereas, dry ice is dropped directly into growing cloud updrafts at temperature levels of -10C, approximately an altitude of around 20,000 feet in mid-summer.

Hygroscopic flares, first tested in 1997 and used for a short time in 1998, appear to be useful in attempting to stimulate rainfall. These flares are composed of sodium iodide, potassium iodide and lithium chloride. Unlike the silver iodide-based seeding agents, hygroscopic particles can enhance
rainfall in both warm and cold cloud temperature environments. There are still some lingering questions about the viability of hygroscopic seeding agents for use in hail suppression.

Over the years, the results from cloud physics research and other programs much like our own have been applied to the WKWM Program whenever possible. This helps ensure that the WKWMP retains in a reasonable state-of-the-art mode. We try to implement new ideas and test discoveries of new technological developments whenever possible, if it can be adapted to our specific conditions within our limited budget. Our innovations, however, generally have tended to be more in the area of enhancing our operational capability.

High numbers of ice nuclei can be produced from our liquid, silver iodide-based, seeding agent in a wing-tip generator by first vaporizing the solution. Wing generators are mounted to the tips of cloud base seeding planes and employ a combustion process in which a 2% silver iodide liquid seeding solution produces trillions of ice nuclei per gram of silver iodide consumed. The wing generators contain a built-in air tank within each generator which, when pressurized, forces the liquid seeding solution through an aperture and nebulizes it into a fine spray. The spray then flows into a combustion chamber where it is vaporized by burning. As the spray burns, very pure particles are formed and are exhausted out the tail-end of the generator into cloud base updrafts where the particles are carried aloft by natural action into the cloud’s supercooled region.

From 1987 through 1995 the liquid seeding agent was the same: Quantities of the oxidizers sodium perchlorate and ammonium perchlorate were added to a silver iodide-ammonium iodide-acetone-water solution resulting in a liquid solution containing 2% silver iodide by weight. However, in 1997 the
formulation was changed to contain amounts of silver iodide, sodium iodide, acetone and paradichlorobenzene (C6H4Cl2). Cloud chamber test results at CSU indicated the total number of ice crystals produced by the new solution at -10C were closely equivalent to the old formulation containing the perchlorates. Also, the particles initially act as hygroscopic condensation nuclei insuring that the formation of vast numbers of water droplets will contain ice nuclei particles. Any ice nuclei initially not trapped in the water droplets can be captured later by other droplets through random collisions within the cloud called “contact nucleation”. The entire process of hygroscopic condensation followed by freezing and contact nucleation, forms greater numbers of ice crystals at relatively warmer temperatures within a cloud than by simple contact nucleation.

The overriding reason to change to this formulation was due to its clean-burning properties. When this solution burns, little, if any, residue is produced leaving less generator maintenance needed over time—a critically important consideration during persistent, active operational periods. In the past two
seasons using this formulation, wing generator maintenance due to corroded parts has been nil compared to previous seasons while using the perchlorate additives.

In order to obtain the desired effect when seeding clouds, each cloud must be treated within a proper time interval, a “window of opportunity,” in order to produce the optimum ice crystal concentrations in clouds naturally deficient in them and to promote supercooled water droplet freezing within clouds. A cloud growing to maturity must be treated with enough time allowed so that the generated ice nuclei can be lifted by natural cloud updraft action into the appropriate temperature and moisture regime and kept there for a sufficient time to interact with the supercooled cloud water. If this opportunity window is missed when attempting rainfall stimulation, clouds may collapse prematurely resulting in wasted effort and resources. “Residence Time” in the supercooled cloud volume is critical to the success of both rain stimulation and hail reduction efforts.

The behavior of weakly and moderately growing cumuliform clouds can be altered through what’s called the “dynamic effect.” Under certain atmospheric conditions, clouds may be stimulated to grow larger and rain longer than would be the case if otherwise left unseeded. This is done by getting sufficient amount of seeding agent into the supercooled portion of a cloud to promote the rapid conversion from water droplets to ice crystals. When this water-to-ice conversion process occurs rapidly, the latent heat of fusion is released on a massive scale making the cloud slightly warmer and more buoyant. Updrafts will soon become invigorated drawing in greater amounts of water vapor into the cloud and supplying more moisture to the cloud for subsequent growth and “processing” into rainfall. This process enables a cloud to rain more and rain for a longer time than if left unseeded.

Although silver iodide produces greater numbers of ice nuclei than does dry ice, gram for gram, large numbers of ice nuclei can be produced more quickly by dropping comparatively larger amounts dry ice directly into the moisture-laden cloud updrafts found in the new-growth cloud towers. Relatively large amounts of dry ice are needed to produce an equivalent number of ice crystals from a given mass of silver iodide—roughly 1000 to 2000 grams of dry ice are needed to match one gram of silver iodide. Supercooled cloud droplets contacting dry ice which is falling through clouds, or those droplets brought into the wake of the falling dry ice, immediately change into ice crystals. Whereas, silver iodide-based seeding agents, while rising in the cloud, begin activating in droplets to form ice crystals at temperatures near -4C to -5C—roughly 2,000 to 2,500 feet above the freezing level.

Dry ice is dispensed from a container auguring dry ice into an opening in the aircraft floor which lets the dry ice fall directly into the clouds. The container carries about 200 lbs of pelletized dry ice and is released at a rate of 5 lbs per minute.

Dispensing ejectable silver iodide flares at cloud top is not done on the WKWMP primarily because it is a much more expensive form of seeding agent than is dry ice, despite dry ice subliming while in storage between operational periods.

The cloud systems listed below, and variations of them, are most responsible for producing rain and hail in Western Kansas:

(1) the air-mass storm complex

(2) multiple celled storms

(3) the squall line

Air-mass storms often become complex after starting out as an isolated cloud system with a well-organized cloud base and its new growth updraft area usually visible somewhere around its base. Multiple cloud turrets often develop around the initial ‘parent’ storm and subsequent storm movements can become erratic depending upon several factors such as its severity, terrain effects, dynamic factors within the storm, cloud height, variability of wind speed and direction with height plus the blocking of steering winds caused by large upwind cloud systems.

Air-mass storm complexes often transition into a large, multi-celled systems. Fig. 2 shows a “classic” storm with new growth on its upwind (left) side. Updrafts upwind of a storm’s direction of movement are termed ‘trailing’ or ‘backside,’ whereas, the front side is on its downwind, or leading side. Most often, updrafts pertinent to the hail process are found along its trailing edge below cloud base at some distance behind the precipitation.

Air-mass storms can transition into a line of storms containing multiple cells showing characteristics more similar to those of a small squall line. During the gradual development of these multi-cellular lines, cloud base updrafts frequently shift around although they are still found near some of the individual
cell elements comprising it making proper seeding treatment quite difficult at times. The updraft locations important to the hail process on these cloud systems are along a line on its front side, running from a few miles to many miles in length. Other times, the best seeding area may be around only one end of the line. Multi-celled lines may also appear as a remnant of a weakened squall line or as part of a line of storms associated with fronts, surface troughs and thunderstorm outflows.

Under some conditions multi-celled storms may become very large, developing several new growth areas simultaneously with distinct “cores” growing embedded in and around the periphery of the cloud boundary while the cloud system is in transition from being a relatively small severe storm into a large supercell. One characteristic of a supercell is that a some point it exhibits a “right-turning” motion relative to the direction of the mean steering wind. Supercells can be quite dangerous as they are capable of ejecting hail long distances in different directions, occasionally throwing it into the flight paths of seeding aircraft both at cloud base and cloud top. Supercells produce the most destructive tornadoes, however, not all supercells produce tornadoes; some estimates have indicated about one-quarter of all supercells may be associated with tornadic development.

The cloud system known as a squall line is an organized line of cumulonimbus clouds many miles in length. Important updraft areas are found along its advancing cloud edges. Updrafts important to the precipitation and hail processes are seldom found along the trailing edges of these lines except at its
end, or at significant breaks within the line. Squall lines can be extensive, crossing a few counties within a state or crossing more than one state; frequently squall lines are associated with surface troughing ahead of frontal passages. Updrafts can easily exceed 2,000 feet per minute and produce “scud” clouds visible nearly to the ground. Ahead of the squall line, updraft areas are usually smooth.

Convective Scale Interaction is a term given to the process in which a collapsing storm produces precipitation and downdrafts and promotes subsequent new cloud growth. The downdraft air, also called a gust front or outflow boundary, fans out below its cloud base undercutting relatively warmer and often more moist air. If moisture is sufficient in the air being lifted above the gust front, it can rise into an unstable atmosphere, growing rapidly into another new severe storm which quickly reaches maturity. Then, the storm collapses, producing its own downdraft, thereby repeating the whole sequence again and can do so over and over, again many times. Single outflow boundaries have been known to be strong enough to travel 100-200 miles, or more, from its parent storm. Satellite views of clouds forming along these moving gust fronts often show them aligning into a semi-circular, fan-shape orientation which are called “arc-clouds.” Some of these clouds, themselves can develop into large, severe convective storm systems. Single storms, multiple storms and supercells all have been identified as forming along these gust fronts. Previous research in the southeastern part of the U.S.A. has estimated 60%-75% of the storms existing in late afternoon on a typical storm day were caused by scale interaction. Two, or more, colliding gust fronts frequently create extremely severe storms, although the severe storms are often short-lived. Severe aircraft turbulence is frequently found in gust front air between the parent storm and the leading edge of the gust front. Ahead of the gust front the air is generally smooth. When gust fronts drop out of high-based clouds, micro-burst activity is prevalent which has been known to flatten buildings, crops and cause aircraft accidents while landing and taking-off.

Convective Scale interaction as described here frequently occurs on the WKWMP and when identified on radar or reported by pilots, its occurrence and direction of movement is monitored carefully for subsequent new storm growth to develop above it. Also, it has been observed that severe new storm growth often develops in weak, old non-hail bearing precipitation areas, which are undercut by gust fronts. Satellite imagery can also give advance warning about subsequent new storm development potential which can’t be seen immediately on radar or visually by pilots.

Under some conditions rainfall augmentation over large areas have been produced by seeding atop the leading edge of a gust front as air is lifted over it which causes only weak cumuliform clouds to form. Updrafts found above gust fronts have wide variability—from 100 – 200 feet per minute to a more normal 1000 – 1500 feet per minute, or more. If this particular condition occurs at night with little threat of hail developing from new storm growth and weak updrafts are prevalent, rainfall stimulation seeding can be highly productive over large areas. In such cases the cloud’s microphysical characteristics are attempted to be altered by seeding. It’s likely under these conditions the dynamic effect is markedly changed, whereas, the “static” seeding effect is being achieved. It is likely that hygroscopic seeding above a gust front may also produce good results, however, it has not been tested on this program yet under the circumstances given.

There is another form of cloud system which has important seeding potential, on occasion, for producing precipitation in Kansas: the multiple-celled-convective system. This starts as a cluster of small, weak air-mass clouds developing within a relatively small area—typically 10 – 30 miles in diameter. If one, or more, clouds can grow sufficiently to merge with another, the resulting merged cloud tends to continue growing, thereby promoting further cloud mergers which further increases both cloud volume and intensity. Such cloud systems are capable of eventually producing precipitation over large areas and persisting much longer than they normally would, otherwise. Updrafts initially found within such a cluster of cells are often embedded and difficult to locate, however, once such a system grows to a certain size, updrafts generally organize better and the cloud system becomes easier to continue seeding. Under natural conditions, many times these are the regions that become the ‘first echo’ development seen on radar. First echoes have a high correlation to being the day’s first severe storm.

Most of the important research on the dynamics of the multiple-celled-convective system was done years earlier in West-Central Texas. By the same token, earlier radar studies from 1972 to 1974 of Northwest Kansas clouds found this area was a fertile region for these smaller cell clusters to develop. Comparisons of data seem to suggest Western Kansas may be even better than in Texas where the most extensive research has been conducted. In the early years scant attention was paid to those weak-appearing multiple-celled cloud clusters until one of them had grown much larger. If some of those systems had been treated early, instead of waiting to see whether they grew and became much larger, more frequent success for rain stimulation and/or hail suppression probably would have occurred. Today, however, we anticipate the development of these smaller cloud clusters and realize they may have the potential to become ‘targets of opportunity’ to produce rainfall over large areas as well as regions of preferred hail development. When observed, the attempt must be made to begin seeding them as early as possible in hopes of catching the “window of opportunity” for dynamic seeding effect discussed earlier. Spectacular results appear to have occurred on some past occasions from seeding such cloud systems.