When one looks at the Bay as a whole, the picture is often grim. Water clarity is worsening, underwater grass beds are declining, and oxygen-starved dead zones aren’t going away.

But when taking a broader look at the rivers that feed the Chesapeake, a new report points out, there are numerous, though scattered, signs of success throughout its watershed.

Nutrient discharge controls at wastewater treatment plants in some places have slashed the growth of toxic algae and helped beds of ecologically valuable underwater grasses to rebound. On farms where cows have been fenced out of streams, aquatic life has improved and pollution decreased. Air pollution controls have helped improve streams’ water quality in huge parts of the Bay watershed.

Those results were among the case studies presented in a recent state-federal Bay Program report that highlighted lessons learned from dozens of specific cleanup actions taken during the last three decades. The actions were taken in various parts of the Bay’s 64,000-square-mile watershed and all were closely monitored.

“We have learned that these local actions have results,” said Nick DiPasquale, director of the EPA Chesapeake Bay Program Office. “They have positive impacts on water quality.”

The recent report, “New Insights,” found clear instances where cleanup actions triggered local water quality improvements. But not all actions produced results, and in those cases the report sought to explore why. In some places, for instance, actions were overwhelmed by human population growth, intensification of agricultural activities or unique local conditions.

“We’ve got some challenges ahead,” said Bill Dennison, vice president of the University of Maryland Center for Environmental Science and one of the report’s coordinators. “But we can fix the Bay with these efforts. This process is something that can be measurable and real, and it is not going to be hopeless. In fact, it can be hopeful.”

Most of the instances where actions triggered measurable improvements in river health stemmed from upgrades to large wastewater treatment plants that were dominant sources of nutrients to local waters.

For instance, upgrades to the Back River plant in Baltimore resulted in nitrogen reductions that led to reduced algae growth within three years of the project’s completion.

Upgrades at the Blue Plains treatment plant in the District of Columbia slashed phosphorus and nitrogen discharges that translated into reduced downstream blooms of toxic cyanobacteria, along with a local recovery of underwater grasses.

Portions of the Patuxent River have seen improved water quality and the return of underwater grass beds that can be linked to improved upstream sewage treatment.

The report suggested more water quality improvements would take place as remaining plants are upgraded in coming years. According to Bay Program figures, 212 — or 45 percent — of the largest treatment plants in the watershed had been upgraded by the end of 2012.

“We can actually see those benefits when we turn on those sewage treatment plant upgrades,” said Don Boesch, president of the University of Maryland Center for Environmental Science. “In a short period of time — months to years — we can see the immediate benefits of that.”

Another success is the widespread reduction in nitrogen deposition on the Bay and its watershed from air pollution. During the last two decades, efforts to control nitrogen oxide emissions from cars, power plants and factories have reduced the amount of nitrogen landing on the watershed from those sources by about 30 percent.

The trend is most obvious in forested watersheds because there are few other sources of nitrogen, but the findings suggest that air pollution reductions likely have a positive impact on nitrogen concentrations in a broad portion of the watershed.

Likewise, the study highlighted several instances where agricultural actions had resulted in local improvements.

Studies by University of Maryland Wye Research and Education Center have shown that cover crops, planted quickly after fall crop harvest can absorb excess nitrogen, reducing the amount transported from the fields to aquifers by 40 percent over time.

Replacing poultry manure with chemical fertilizer — which is more easily controlled and applied — on a small farm on Green Run, a tributary of the upper Pocomoke River, resulted in a 30 percent decrease in nitrogen concentrations in the creek.

Reducing manure and fertilizer applications on a farm on Brush Run Creek in southcentral Pennsylvania cut the amount of phosphorus entering the waterway by 57 percent.

Many of the stream improvements in agricultural areas cited in the report tend to be in very small headwater streams; the Green Run watershed is about 1,200 acres, and Brush Run Creek is only about 400.

Those local effects often get diluted at larger scales, scientists said. “You can see real local water quality improvements, but usually when you move farther downstream you start to get a mix of several sources — agriculture, suburban, urban, septics — so it is hard to distinguish what is causing the improvement,” said Scott Phillips, Chesapeake Bay coordinator with the U.S. Geological Survey, and one of the report’s coordinators.

“The other piece is that you really need a lot of implementation to be able to detect that water quality improvement, and in many areas we might not have the needed degree of implementation,” Phillips added.

Indeed, the report cited a multi-year study on Bald Eagle Creek in York County, PA, in which the farm animal population was decreased by half, and the amount of nitrogen and phosphorus applied through manure and fertilizer was sharply reduced. But water quality improvements were never seen.

Researchers concluded the actions were offset by other activities on the farm, and by another landowner who didn’t participate in the study, changed the stream bank and channel and then allowed animals to graze near the water.

Another factor that makes detecting trends in agricultural watersheds difficult are “lag times.” Unlike wastewater treatment plant discharges, all of which go directly into the river, nutrients in agricultural systems are applied to the ground.

While some of those nutrients are flushed into waterways during storms, it takes years for other nutrients to make it to streams. Much of the nitrogen reaches streams through groundwater, a trip that can take years or decades. Similarly, phosphorus attached to sediment can be trapped in watersheds and stream beds for long periods before they make their way to the Bay.

“This tells us that we need to be a bit more patient with respect to seeing the outcomes,” Boesch said. ”But this should not be a blind patience…it has to be an informed patience where we are actually looking at things like changes in nitrogen concentration in groundwater. If we don’t see them, then maybe the practice isn’t working.”

Phillips agreed that more monitoring of groundwater was needed. Most existing monitoring is focused on surface waters.

The report offered little evidence of stormwater controls improving downstream water quality, but said that some new “green infrastructure” techniques such as rain gardens, which promote water infiltration into soils rather than discharges into streams, showed promise. Part of the problem with detecting an impact, officials said, is that because such techniques are relatively new, there is little monitoring data available, and often, the practices are not widely implemented.

“We haven’t seen the kind of real clear results from the stormwater sector that we’ve seen from the agriculture sector, from the wastewater sector from the atmospheric deposition sector,” said Rich Batiuk, associate director for science with the EPA’s Bay Program Office. “I think they are out there.”

But, the report acknowledged, not everything has produced positive results.

In the late 1980s, researchers established two miles of fences to keep farm animals from wandering into a stream in Lancaster County, PA, then watched for improvements in stream water quality.

They got a surprise. The data showed that the quality of streamwater leaving the fenced areas was actually worse than the quality of water when it entered.

Not to be deterred, they further examined the site and made a surprising discovery: Unusual geology under the site was actually drawing water away from the stream. Although the fences were reducing runoff, the stream failed to exhibit an improvement.

Wastewater treatment plants on the Choptank River had little impact on the river’s overall water quality because part of the improvement was offset by population growth, and because the nutrients from wastewater are relatively small compared with those from agriculture in the watershed.

Those lessons are important, too.

“It is important to see where things are not happening as we expected, and understand why that is the case,” Boesch said. “Is it because of lag time, or is it because we overestimated the effectiveness of what we were doing? All of those things are absolutely essential [to understand].

“To think that we have this blueprint that we are following, that it is perfect and that in the end it leads us to Valhalla is nonsense. We really need to watch what we are doing.”

The report is available here.