Scientists ask, does rainmaking work?

Press, 17 November 1981, Page 20

Scientists ask, does rainmaking work?

From “The Economist,” London

In America, would-be weather modifiers coin perhaps S4M-S6M a year from farmers, water utilities and so on by seeding clouds with, e.g., silver iodide to promote rain. Yet even today, more than three decades after rainmakers first went commercial, there are doubts about their ability to deliver.

One problem is how to measure success. Typically a company will claim it can increase precipitation in a, given area in a given season by 10-20 per cent over the. longterm average. Those are figures well within the range of variation that occurs naturally. So what do they prove? Suppose rainfall does increase by 10 per cent. Hats off to the commercial rainmaker? Or would-that increase have occurred anyway? Suppose rainfall does not increase. Thumbs down to the rainmaker? Or would rainfall in your area have decreased 10 per cent but for the company’s intervention?

One way of trying to answer such questions is to set up a properly controlled experiment. Broadly, there are two approaches: O Randomised seeding. You could choose a target area, record all the days judged suitable for rainmaking but actually seed clouds only on some of them. To avoid bias, the choice of which days to seed should be made on the flip of a coin or the like. Then you

could see whether rainfall was greater on seeded days with favourable weather conditions than on unseeded days with similar weather conditions. • Control areas. Alternatively, you could choose a target area for seeding operations and one or more control (unseeded) areas. The control areas would have to be ones in which precipitation is normally very similar to that in the target area — and yet unaffected by operations over the target area. Then you could see whether rainfall differed significantly in the two areas.

Of course, commercial rainmakers are not in the business of conducting such thorough scientific experiments. But. over the years, a number of government and other teams have run experiments based on one or the other (or a combination) of these approaches. At first blush, it is hard to make much sense of their results. Some experiments did seem to suggest that cloud seeding produced more rain. Some were patently inconclusive. Three experiments in the United States disconcertingly seemed to show that cloud seeding significantly reduced precipitation. Analyse the experiments more deeply and a more coherent picture emerges. The key conclusion sounds deceptively simple: cloud seeding can enhance rainfall — but only under certain conditions. Many of the early experiments were carried out “blindly."

Their designers did not know enough about the dynamics of clouds or the weather systems harbouring them to get their tactics right. They selected the wrong clouds. Or seeded good candidates. too early. Or pumped in too little — or too much — seeding material. Or ignored wind conditions that literally blew their efforts away. That sort 6f mistake accounted for many of the disappointing results. Not that all the experiments claiming positive results stand up to scrutiny. Some claims rested on too little data; some were biased by just a few days of exceptionally heavy rain and so on. The one experiment that everybody agrees was impeccably run — and threw up convincing results — was conducted in Israel. It showed average increases in rainfall due to seeding of about 15 per cent. So the key questions are two. Under what specific conditions does cloud seeding make sense? And do such conditions occur frequently and widely? The answers depend on the physics of rainfall, which is only partially understood. A cloud is more a dynamic process than a thing. It begins to develop when an up-draught of warm, .moist air rises, expands and cools, condensing, moisture in droplets around dust and other impurities in the atmosphere. Initially, these droplets are very small: typically only a few microns (thousandths of a millimetre) across. To fall to earth as rain, they must grow at least one million times larger and do so before the cloud dies away —

often within an hour. Condensation alone is unlikely to yield droplets of more than 40-60 microns on average.

The key to further growth is turbulence and the tendency of a cloud to climb upwards into colder reaches' of the atmosphere as condensation within it releases latent heat and increases its buoyancy. In these turbulent conditions, cloud droplets can grow by colliding and coalescing with one another. But the crucial step is the formation of ice crystals which can sweep up lots of cold liquid droplets and splinter (through collisions) to generate still more icy scavengers. These can then combine into large particles. In practice, for ice crystals to form, there need to be “nuclei” (e.g. particles of clay or silicate) around which they can grow. In a cloud, spontaneous freezing of pure water

droplets will begin only at hardjo-attain temperatures of about -40 deg. C; given nuclei, ice crystals will begin to form at higher temperatures. Efficient nuclei are not always to hand. That is the rationale behind seeding clouds: the materials used to seed them are chosen for their ability to act as ice nuclei. The Israelis found they got their best results when seeding clouds whose tops were at temperatures in the range of - 15deg. C to -21 deg. C. On days when cloudtop temperatures were either above -luoeg. U or below -21 deg. C, the effects of seeding were negligible. But it would be wrong to assume there is a simple recipe for seeding clouds. The ranges of size of the droplets in particular clouds, the concentrations of natural ice nuclei and the dynamics of air currents in them are also relevant

— and not well understood. The behaviour of individual clouds can be as erratic and eccentric as that of the weather itself.

Cloud expert Sir John Mason, head of Britain's Meteorological Office, argues more work needs to be done to find out precisely why the Israeli clouds seem more responsive to seeding than rather similar clouds in other parts of the world. Indeed, the met. office has now begun talks with the Israelis about the possibility of joint experiments. And, in the wings, is the rain in Spain. The World Meteorological Organisation is mulling over preliminary results of three years’ work in Spain undertaken in preparation for a possible large-scale, international experiment in rainmaking. The decision whether to go ahead should be made in June next year.