Scientists wait to see if volcano brings early winter and argue over whether CO2 buildup will warm up the Earth

A National Academy of Sciences committee has reaffirmed that carbon dioxide emitted by burning fossil fuels could warm the Earth. But a committee report notes, some scientists who doubted that conclusion ''remain unconvinced.''

Thus the first installment of a congressionally ordered reassessment of the so-called carbon dioxide problem has done little to resolve the subject's uncertainties.

Besides the degree of warming, if any, these uncertainties include possible climate changes from a warming and whether or not these would be harmful to humanity.

Some scientists have speculated the Antarctic ice cap might melt and flood coastal cities. Some studies indicate rain belts would shift so that many presently productive farmlands would become marginal for agriculture. Other experts, however, have suggested that more carbon dioxide in the air could make crops grow more abundantly and use water more efficiently. The net effect, they say, could be an increase in world food supply.

Carbon dioxide (CO2) is an odorless, colorless gas emitted when wood and fossil fuels burn. It also is the main source of carbon for photosynthetic plants, which use carbon as a basic building material. In the atmosphere, CO2 intercepts heat radiating upward and re-radiates much of it back toward Earth's surface.

If there were reason to expect a carbon dioxide warming and to expect it would have bad effects, energy planners might have to restrict future use of coal. Much of the extra carbon dioxide would come from that fuel. Thus Congress has asked the President's Office of Science and Technology Policy, together with the National Academy of Sciences, for an overview of the issue. The new NAS report, looking only at the question of a possible warming, is a partial response to this request.

The NAS committee used a 1979 NAS assessment as the basis for its study. The committee's main finding is that ''the present study has not found any new results that necessitate substantial revision of the conclusions of the (1979) report.'' The main conclusion of the earlier report was that a doubling of atmospheric CO2 - expected within the next 50 to 100 years - would most probably bring a global warming of 1.5 to 4.5 degrees C. (2.7 to 8.1 degrees F.)

Many scientists interested in the CO2 problem concur with this. But there are dissenters.

Reginald E. Newell of the Massachusetts Institute of Technology and T. G. Dopplick of Scott Air Force Base in Illinois have studied the response of the tropical ocean to a CO2 doubling. They suggest that the surface temperature rise would be slight.

One of the most insistent critics is Sherwood B. Idso of the US Department of Agriculture Water Conservation Laboratory in Phoenix, Ariz. He has tried to estimate the atmosphere's response to increased heat radiation by studying natural events. These include changes in dust levels and in atmospheric moisture over the Phoenix area and seasonal changes in solar radiation over the US. Dr. Idso has reported that he believes the warming due to a CO2 doubling would be only about 0.25 degrees C. ( 0.45 F.) - a negligible amount. He finds the same result from his own analysis of the study of Newell and Dopplick.

The NAS committee judged both studies to be ''flawed and incomplete.'' It criticized them for not being global in extent and for neglecting key energy flows and the overall energy balance of the relevant processes. Yet in the report's preface, committee chairman Joseph Smagorinsky admitted that such criticism had failed to make the dissenters change their minds.

Neither Newell nor Dopplick was available for comment. However, Idso confirmed he still considers the projected CO2 warming to be a ''nonproblem.''

For his part, he criticizes the computer studies on which the NAS committee, and many other scientists in this field, rely. These are elaborate mathematical simulations of the atmosphere. They include many factors. But some key processes are left out or poorly represented, Idso maintains. He notes, for example, that the influence of clouds is not understood. He says he believes that when the computer models can include all the key factors and represent them adequately, they also will show negligible CO2 induced warming.

Although the NAS panel expresses confidence in the computer models, it does acknowledge their shortcomings. The report says that the models ''have some predictive reliability'' for CO2 studies. Yet it adds, ''present models are not sufficiently realistic to provide reliable predictions in the detail desired for assessment of most impacts.''

Chairman Smagorinsky summed it up by saying: ''Ultimately, of course, nature will reveal to us all the truth.''