Weather satellites have been measuring global temperatures since 1979, but have shown no climate warming — contrary to all expectations. According to computer models cited by the U.N.-sponsored science advisory body, the Intergovernmental Panel on Climate Change (IPCC), a substantial warming of nearly 1 degree Fahrenheit should have occurred during this 18-year period.

If the models cannot be supported by actual observations of the atmosphere, however, then we cannot and should not rely on their predictions of a future warming. Therefore, all the elaborate international machinery that's being constructed around the Global Climate Treaty is a complete waste of time and money. Policies are being developed to ration and tax carbon dioxide emissions from fuel burning, and thereby throttle energy use. Such policies would cause grave economic damage not only to industrialized nations, but also to developing countries that depend on trade and global prosperity.

In contrast to the satellite results, data collected from thermometers at weather stations around the world do show a warming — but only about half that predicted by the models. But these surface data do not cover the oceans, almost 70 percent of the Earth's surface, and large gaps exist elsewhere. Besides, such observations are subject to all kinds of local disturbances and interferences that can change over time, and thereby confuse measurement of temperature trends; for example, construction of a nearby building or even a growing tree will change wind patterns and affect the thermometer reading. But the biggest disturbing influence is the local warming from urban growth around the weather stations, which are usually located at airports.

This well-studied "urban heat island effect" produces a fictitious warming that is hard to eliminate from the surface temperature readings. Everyone knows this is so, but few will admit it. Instead, the attack has been focused on the satellite data. The fact they have reported a slight cooling since 1979 has been an embarrassing problem for global warming enthusiasts. (When corrected for the strong cooling effects of the two major volcanic eruptions in recent years, the satellite data suggest a slight warming trend — albeit very much smaller than that derived from surface observations or predicted by computer models). The divergence between satellite-derived temperature trends and surface thermometer trends has been growing larger year by year.

The campaign to discredit the satellite observations has been gathering momentum as it became more widely known that they demonstrate the inadequacy of the computer models. The objection at first was that the satellite record is too short; but after a decade or so, this excuse no longer works. The second approach has been to state that the surface thermometers and the satellite-borne instruments are both correct but that they measure different quantities. It is true the satellite instrument derives a temperature of the lower atmosphere rather than of the surface itself, but the two trends cannot be expected to diverge for very long. Both records cannot be right; at least one of them must be wrong.

The most recent attack, published in the March 13 issue of the international journal Nature accuses the satellite experimenters of technical errors in constructing the record that shows a cooling trend. The satellite experimenters will shortly reply with technical arguments that will be equally incomprehensible to the general public and even to non-specialist scientists.

The scientific debate is being conducted at a high professional level and will no doubt shed some light; but it hardly affects the fact the computer models are not being validated.

Personally, I believe the satellite data, although not perfect, are more nearly correct. The easiest way to demonstrate this is to appeal to a completely independent set of measurements for verification, instruments launched around the world by weather balloons show exactly the same temperature trends as the satellite instruments.

Further support for the satellite results comes from the actual demonstration of an urban heat island effect on surface thermometers. An article published in the July 1996 issue of the Bulletin of the American Meteorological Society graphically shows temperature readings from highly populated California counties exhibit a strong warming trend during this century while less populated countries show a zero trend. Similar results have now been obtained, comparing cities and rural stations in Australia and in South Africa. It is evident that cities have become warmer as population grows and releases more heat into the atmosphere.

So we're back to Square One. It is clear the computer models are not validated by observations and need more work before we can trust their predictions. But now a scientific attack on the satellite data has been published in a respected journal, we can — regrettably — expect to see efforts to discredit the satellite results and obfuscate the true situation.

And what about future warming? When all is said and done, we can conclude there will be some warming as a result of the growth in atmospheric greenhouse gases perhaps not more than 1 degree F by the year 2100, which would be barely measurable and completely inconsequential.

But long before then, a few more years of satellite data will make it obvious that global warming is a non-problem.