Underwater grass beds in the Chesapeake expanded by nearly 10 percent during 2007, driven largely by the ongoing recovery of beds in the Susquehanna Flats and other low-salinity areas.

But scientists remain worried that grasses in most of the Bay continue to show little sign of recovery, and many areas have fewer beds than they did just a decade ago.

The annual Baywide aerial survey showed 64,912 acres of grass in the Chesapeake and its tidal tributaries last year, an increase of nearly 6,000 acres from 2006. Still, the underwater meadows cover only a third of the Bay Program's 185,000-acre goal.

And except for a spike in 2002, when 90,000 acres were observed during a lengthy drought that produced clear waters in much of the Bay, overall grass acreage has remained relatively constant for nearly two decades, ranging between 59,000 and 75,000 acres.

The lack of a Baywide improvement for submerged aquatic vegetation, or SAV, is at least partly driven by poor water clarity, which has shown a worsening trend in much of the Bay. "That is a very important item that we need to be focusing our attention on because it certainly plays a pivotal role with SAV in the Bay," said Bob Orth, a scientist with the Virginia Institute of Marine Science who oversees the annual Baywide survey.

The Bay Program's annual Health and Restoration Assessment Report showed that just 12 percent of the Bay met water clarity goals in 2007.

The annual aerial survey is a closely watched indicator of the Bay's health. Grass beds require clear water to absorb the sunlight they need to survive, so they are very susceptible to nutrient pollution, which spurs algae blooms, and sediment runoff, which clouds the water.

They also provide some of the most important habitat in the Chesapeake, providing shelter for juvenile crabs and small fish, and food for waterfowl.

Once, the Bay is thought to have supported hundreds of thousands of acres of underwater meadows but they began disappearing in the early to mid-1900s as pollution increased and water clarity worsened. In 1972, Hurricane Agnes delivered a devastating blow as torrential rains covered already weakened beds with layers of sediment while algae blooms filled the water.

Grasses bottomed out at an estimated 38,000 acres in 1984, and slowly increased until the 1990s. Since then, there has been no clear Baywide trend.

But the distribution of grass beds has changed.

"Over the past five to 10 years, we have witnessed large increases in many of the freshwater tributaries and segments of the Bay," said Lee Karrh, a scientist with the Maryland Department of Natural Resources and chair of the Bay Program's SAV Workgroup. "However, middle- and high-salinity areas, such as Eastern Bay and Tangier Sound, are well below their peaks."

The Susquehanna Flats, which had only a few scattered patches of grass a little more than a decade ago, has burst forth to become the largest, densest grass bed in the Chesapeake. Last year, grasses in that area surged to 11,726 acres, from 8,743 in 2006. That accounted for half of the entire Baywide increase.

"It's one of the few beds you can see from a satellite image. It just dominates that area," Orth said. "I look at satellite images all the time. I rarely see SAV beds. That bed shows up all the time. That's how big, and how dense, it is."

The bed is so large it is modifying the local environment, filtering water that flows by while producing billions of seeds, helping to fuel continued expansion.

Much of the rest of last year's increases came in low-salinity areas of the Potomac River, which gained about 2,000 acres over 2006.

Other low-salinity areas fared well also. The Bush, Gunpowder and Middle rivers in the Upper Bay showed sharp increases.

In Virginia, the upper tidal portions of the Pamunkey and Mattaponi showed increases as did the Chickahominy River and some tidal creeks along the James. In many of those areas, the return has been led by hydrilla, a nonnative species. But studies in the Potomac by U.S. Geological Survey scientists have shown that native plants, which had been wiped out by pollution, were able to flourish after hydrilla began growing in the river.

The recovery in low-salinity areas was not universal. Many plants in those areas have little tolerance for salinity increases. Scientists believe that last year's drought, which reduced freshwater flows into the Bay, raised salinities enough to cause diebacks in Maryland's Sassafras and Chester rivers, where beds had already been declining for several years.

Elsewhere in the Bay last year, grasses in mid-salinity areas generally declined. Scientists say poor water clarity, coupled with higher than normal salinities were likely contributing factors in the Magothy and Severn rivers as well as in Eastern Bay, where grass acreage fell from 565 acres in 2006 to 80 acres in 2007. Mid-salinity portions of Virginia waters were generally unchanged, although the Rappahannock showed a slight increase.

Meanwhile, most high salinity areas of the Bay, showed an increase in underwater grasses from 2006. That reflects a continued recovery of eelgrass-the main species found in high salinity water-which took a major hit in 2005 when warm temperatures were blamed for a major dieback. But Orth said the eelgrass beds remain fragile.

"We were pleased when we saw that the beds that died back are recovering. But I think the rate of recovery is being compromised by the poor water clarity that we now have," Orth said. "We are at such low levels, if we had another hot summer, it could knock back these plants to a level where there would really be a major concern that they would not recover."

Overall, eelgrass acreage had declined by about a quarter since the early 1990s, even before the 2005 die-off.

The annual Baywide grass estimate is derived from an analysis of more than 2,000 black-and-white aerial photographs taken between May and October.

For information about the aerial survey and survey results, go to www.vims.edu/bio/sav/.