American experts who have studied the Soviet nuclear-weapons program say a picture is emerging of extremely intense radioactive contamination and a history of very high doses to production workers.

The experts, who visited the Soviet nuclear weapons production complex recently and scoured Soviet publications, say that the Soviets have recently confirmed that in the early years of their bomb program at Chelyabinsk, a complex about 900 miles east of Moscow, radioactive wastes were dumped into the Techa River until traces showed up in the Arctic Ocean, nearly 1,000 miles away.

So the bomb producers switched in September 1951 to pumping the wastes into Karachay Lake, a 100-acre body of water that has no outlet, until the accumulation was 120 million curies. That amount is about two and a half times greater than the release from the Chernobyl nuclear accident in 1986.

''This has got to be the most polluted spot on the planet,'' said Thomas B. Cochran, a physicist and senior staff scientist at the Natural Resources Defense Council, a private environmental group in Washington, and director of a research project that produced detailed publications on the American weapons complex.

Disagreement on Intensity

According to a new report by Dr. Cochran and Robert Standish Norris, also at the council, the dose rate on the lake shore near the outlet pipe is 600 roentgen per hour, a rate that would deliver a dose in an hour that would kill a person within weeks.

The Soviets, using lead-shielded bulldozers and dump trucks, are now seeking to stabilize the wastes by reducing the surface area and building separating dikes.

The work has been hampered, according to a paper by Boris V. Nekipelov, deputy minister of nuclear power and industry, and several other authors, by ''active aerosol carryover by wind.''

Other scientists who have visited the area recently confirmed that the lake was extremely contaminated, although they disagree about the intensity.

An 'Astronomical Dosage'

Frank L. Parker, an engineering professor at Vanderbilt University and chairman of the National Academy of Sciences board on radioactive waste management who was in Chelyabinsk in June, said that some American scientists on that tour then came away understanding that the dose rate was 600 roentgen per hour, while others believed it was 6. In either case, he said, that was an ''astronomical dosage.''

Even at the lower figure, a person on the scene for less than an hour would exceed the maximum permissible exposure for an American worker for a year, and the level is so high that cleanup would be extremely difficult. The limit now in use in this country and in the Soviet Union is 5 rem a year. A rem is a slightly different way of measuring radiation, referring to the dose absorbed in tissue rather than the intensity of the field, and is roughly equivalent to a roentgen.

Soviet documents describe a ''sanitary alienation zone'' in contaminated areas in which people are forbidden to live or travel. It was not known how large the zone is. Richard Wilson, a physicist at Harvard who has studied the Soviet nuclear complex, said, ''Provided you keep away from it, which they are, it is causing no harm to public health.''

He said that while the contamination was severe ''as compared with other industrial contamination of the Soviet Union, it is not outlandish.'' But he said , ''I would characterize it as greater than anything we've even contemplated in this country.''

Rise in Mortality Rate Seen

Washington is facing a bill in excess of $100 billion for cleanup of pollution from its bomb production. But the levels of contamination are far lower.

Recent Soviet disclosures also point to very high worker exposures. A researcher at Harvard, Alexander Shlyakhter, formerly of the Lenningrad Nuclear Physics Institute, translated a paper from the February issue of the quarterly Priroda, or Nature, that described doses of more than 100 rem to large numbers of workers.

The paper, also written by Mr. Nekipelov and others, says that workers who received more than 100 rem of radiation showed cancer mortality rates of 8.1 percent, compared with 4.3 percent for those whose exposures were less than 100 rem.

The figures, Dr. Shlyakhter pointed out, refer only to mortality thus far, since many of the people brought to the site in the late 1940's and early 1950's were young and are still living.

Compared with the current exposure standard of 5 rem per year, the 100 rem figure is extremely high, apparently having been accumulated in the period of a few months or years.

Dr. Shlyakhter said the paper also says the average dose for plutonium production workers in 1949, the peak year, was 94 rems, and in the chemical plant in 1951, the peak year there, 113 rems. He said the paper reports that 1.8 percent received more than 400 rems. If delivered in a few hours, a dose that size would kill a substantial number of those exposed.

The paper does not make clear the size of the worker population. ''It's still a secret military facility, and they declined to give any precise number of workers employed,'' said Dr. Shlyakhter. ''They give only proportions.'' But he estimated the number exposed to radiation at 6,000 to 6,500.

''The collective dose would seem to exceed those in Japan,'' he said, speaking of the survivors of the atomic-bomb blasts in 1945.

Similar to U.S. System

The basic Soviet system for making plutonium for bombs was similar to that used at the United States Governments' Hanford Reservation, near Richland, Wash. And like the production system in this country, large parts of the plants are contaminated and broken down.

In both countries, atoms of uranium were split in a nuclear reactor, releasing subatomic particles called neutrons. Some of those split other atoms, creating a chain reaction. But some were absorbed by uranium atoms, which were converted to plutonium. The fragments of uranium atoms were radioactive wastes, including cesium and strontium.

Then the fuel was removed from the reactor, and a chemical process was used to separate the plutonium, unused uranium, and wastes. In the United States, such so-called reprocessing wastes were concentrated and some liquids were poured into earthen basins in dry areas, in the unjustified belief that the soil would immobilize the radioactive material. The remainder was put in tanks, many of which have leaked and some of which pose an explosion hazard.

In the Soviet Union, much of the material was simply dumped into ''the open hydrographic network,'' as a Soviet document put it, meaning rivers and lakes.

Map: Chelyabinsk, U.S.S.R.