Forty feet above the soggy black mud of Florida's Everglades, Biologist Nicholas Aumen gestures out across an undulating sea of cattails.
"This is what we are trying to prevent in the Everglades," he says, motioning from the top of an observation tower toward the proliferating weeds clogging a portion of what once was part of the pristine "River of Grass."
The problem is that cattails and other aggressive plants are crowding out native saw grass, which is a cornerstone of the fragile ecosystem that extends from Lake Okeechobee south to Florida Bay. The cattails thrive on phosphorous-rich fertilizer runoff from nearby sugar farms.
Anywhere else in this wetland these cattails would be seen as an environmental catastrophe. But here, along the L-7 canal north of the Loxahatchee National Wildlife Refuge, Mr. Aumen and other scientists are encouraging the growth of cattails.
In effect the scientists are using this 3,800-acre patch of cattails and other fast growing plants to act as a huge natural filter to suck the fertilizer out of the runoff water, permitting cleaner water to be discharged into the Everglades.
The idea is to use cattails to help prevent the further spread of cattails. If water leaving the filter marsh is clean enough, saw grass should be able to hold its own and reestablish itself in areas contaminated with farm runoff carried deep into the Everglades by flood control canals.
Unless it's removed from the water, phosphorous threatens to change the face of Everglades forever.
"I'm extremely encouraged by its performance so far," says Richard Meeker, who runs the pilot project for the South Florida Water Management District, a state agency.
The experimental marsh has been operating since 1994. Initial targets were to reduce the phosphorous level in farm-runoff water from 120 parts per billion when it entered the filter marsh to less than 50 parts per billion when it left. Results have exceeded all expectations. In 1995, discharge water at the southern end of the marsh showed levels of 21 parts per billion. In 1996 it measured 24 parts per billion.
So far, scientists have found no observable negative effects on Everglades animals from exposure to the higher phosphorous levels in the filter marsh.
The science of filter marshes is so promising that construction has already begun on a second marsh, this one 2,280 acres just north of the Big Cypress Seminole Indian Reservation. Five other marshes are also in the design and planning stages. The total price tag - about $700 million.
Eventually the string of filter marshes will buffer the natural Everglades from some 700,000 acres of sugar fields and other farms.
But scientists say that the only way to truly restore the Everglades and make it safe for all the region's native species will be to reestablish that careful balance that existed for thousands of years when these wetlands were covered by saw grass.