The nitrogen will have to be cut by between 40 percent and 45 percent to reduce the dead zone to an average 2,000 square miles along the Louisiana coast, Scavia wrote.

The 30 percent reduction should result in a dead zone of between 2,560 square miles and 5,120 square miles, based on the model he used to reach his conclusions.

Scavia said, however, that the 2001 nutrient reduction plan still is a good first step at addressing a major environmental problem.

"We've set the initial target at 30 percent, and that's a pretty big job," Scavia said. "We need to start moving on the 30 percent reduction, and as we're doing that, continue to refine the models and analyses."

The reduction plan calls for largely voluntary efforts by farmers and others to reduce the use of nitrogen on crops and the release of nitrogen into the river. Those efforts would be helped if wetlands restoration programs targeted areas where they could remove nitrogen along the river and its tributaries.

The dead zone is a layer of water along the bottom of the Gulf that contains dangerously low levels of oxygen, a condition called hypoxia. Fish and shrimp avoid it, and their food sources, organisms that live on the bottom, are killed by it.

The zone forms when too much nitrogen is washed down the Mississippi and Atchafalaya rivers to the Gulf of Mexico. The nitrogen is released directly into the river and its tributaries from sewage treatment plants and septic tanks, or it's washed into the river after it's applied to crops as fertilizer or deposited on land from the air emissions of electric power plants. In the Gulf, the nutrients feed huge blooms of tiny algae, which die and decompose on the bottom, using up oxygen.

When the national nutrient reduction plan was authored in 2001, its goal of reducing the size of the dead zone to 2,000 square miles was based on existing measurements of nitrogen in the river and estimates of the size of the low-oxygen area in the years between 1980 and 1996. Research since then has indicated the low-oxygen area covered an average 6,400 square miles between 1997 and 2001. Last year, the dead zone was estimated to cover more than 8,000 square miles.

Since the plan was approved, Scavia, Louisiana Universities Marine Consortium scientist Nancy Rabalais and three scientists at Louisiana State University have developed a model to better measure the effects of different levels of nitrogen when combined with various forces affecting water in the Gulf of Mexico, such as weather and the westward movement of river water from the mouths of the Atchafalaya and Mississippi.

Historic measurements of nutrient load, current speed, the height and size of the low-oxygen area, and weather conditions were fed into the model, and estimates were made for the size of the dead zone for the years 1968 through 2002. The model's results closely matched actual measurements of the low-oxygen area made between 1985 and 2002, with only a few exceptions.

Using the model, the scientists said that before the mid-1970s, there was not enough nitrogen in the river water to create major dead zones along Louisiana's coastline. But as the use of fertilizer rapidly increased beginning in the mid-1970s, so did the size of the dead zone.

The model was then run to determine what would happen if nutrients were reduced by different amounts, between 10 percent and 60 percent, over the 15 years called for in the national plan. With a 40 percent reduction, the model determined that the dead zone would shrink to between 1,280 and 3,520 square miles, closer to the national program goals.