The Bay Program is struggling to resolve a question that has festered for more than two decades: Just how accurate is its information about the tens of thousands of nutrient reduction efforts claimed to be taking place throughout the 64,000-square-mile watershed?

A growing chorus in the agricultural community strongly contends that farmers are not getting credit for all of the conservation strides they've made. Several recent reports seem to back up that claim.

At the same time, even as some of the best management practices, or BMPs, go uncounted, others contend that the benefits of many nutrient control efforts are overestimated. In some cases, practices are poorly installed or managed. In other cases, buffers and stream bank fences that vanished years ago — sometimes, along with the farms they were on - remain on the books, delivering phantom nutrient reductions to the Bay.

The problem of accurately knowing what practices are in place and how well they are working has been raised in reports going back at least a decade. But this spring, the National Research Council hammered home the issue in a tersely worded report. "The overall accounting of BMPs in the Bay watershed cannot be viewed as accurate," it said.

The urgency of addressing the problem has never been greater. Under the EPA's new Chesapeake Bay Total Maximum Daily Load, or pollution diet, states have to write detailed plans - known as watershed implementation plans, or WIPs - showing how they will meet their assigned nutrient reduction goals.

In addition, they have to establish two-year milestones that detail what actions they will take, and the amount of nutrient reductions those actions will achieve, in the coming 24 months. The milestones for 2012-13 are due at the end of 2011. If states fall short of goals, they can face sanctions from the EPA.

Apart from wastewater treatment plants, where nutrient discharges can be directly monitored, measuring progress toward meeting milestones and overall cleanup goals relies on reports about BMP implementation that states file with the EPA.

Ensuring the validity of that information, the National Research Council report said, is of "paramount importance" because that information is used to estimate the status of cleanup efforts.

Here's how the system works. Each year, states report to the EPA the number of practices that were reported to them from county conservation districts, other agencies and sometimes even conservation groups. That information is fed into a computer model that estimates the amount of nutrient reduction progress that should result from those actions.

The National Research Council review team said they could not conclude the magnitude of reporting errors, or the overall direction - whether cleanup efforts were being overestimated or underestimated.

In November, the state-federal Bay Program partnership is scheduled to complete a response to that report, the draft of which pledges ongoing efforts to resolve the problems, but offers little in the way of specifics.

Officials interviewed are adamant about fixing the problem. The EPA and the U.S. Department of Agriculture have signed an agreement committing them to developing a mechanism to improve the reporting, tracking and verification of BMPs by next July.

It's not just an agricultural issue. The new policy will cover other sectors, such as urban lands, where BMPs have not always been tracked. But many urban areas are covered by permits. And the vast majority of BMP implementation has taken place on agricultural lands, so that is where most of the focus is aimed.

But the issue of verifying what is actually on the ground is difficult, and expensive. Eventually, it may mean that trained technicians will need to visit a large percentage of the 84,000 farms in the watershed.

"Farmer-funded" practices

It's commonly accepted in the agricultural community that farmers are not getting credit for all of the conservation efforts they make.

What's reported to the EPA are primarily BMPs funded through various state or federal cost-share programs that help farmers build manure storage facilities, install buffers, plant cover crops or implement a host of other conservation measures that have been approved by the Bay Program for use in its models.

But many farmers take actions without collecting a cent from the government. A study by the U.S. Department of Agriculture's Natural Resources Conservation Service earlier this year indicated that the number of farmers using no-till practices that reduce erosion was far greater than previously reported.

Pilot studies by conservation districts in Pennsylvania's Lancaster and Bradford counties reached similar conclusions.

"We know we have a tremendous amount of BMPs that have never been accounted for," said Don McNutt, district administrator for the Lancaster County Conservation District. "The culture and the background here of the Plain Sect, and even the English here, is they would rather do it themselves."

A recent study in Maryland's Upper Chester River watershed estimated that a third of the 650 conservation practices on 125 farms in the area were implemented without assistance.

Officials in other counties and states have similar stories of farmers installing stream bank fences, switching to no-till management and even building manure storage lagoons without seeking assistance. As a result, local conservation districts often have no knowledge of those actions, and they go unreported to the EPA and uncounted toward nutrient reduction goals.

States have stepped up efforts to get better data; the Virginia General Assembly in 2010 required that the state start collecting such information. Elsewhere, grants have helped several pilot projects around the Bay watershed to better ground-truth BMPs. And the National Association of Conservation Districts is working with the states to develop a common procedure for documenting so-called "farmer-funded" or "voluntary" best management practices.

The most aggressive program may be in Maryland's Howard County, where more than a dozen technicians are being trained with an eye toward visiting each of the county's 335 farms.

They want to inventory everything that is happening on the farms. How many have fenced animals out of streams? How many have buffers, and how wide are they? They even want to know about the practices farmers employ that don't measure up to federal or state standards. For instance, do they have stream buffers that are less than the 35 feet required?

The $80,000 effort is overseen by Bob Ensor, district manager of the Howard Soil Conservation District and Dana York, a 30-year veteran of the USDA Natural Resources Conservation Service, who is now a consultant.

They envision training the technicians to meet with farmers individually and record what they see. A separate verifier will follow up to doublecheck the data. "There is no question, once it gets entered into our system as a verified practice, it is for real," Ensor said. "Anybody can go out there and find it."

They describe their effort as the "platinum" survey method, but their hope is that it will lead the way to less costly programs elsewhere.

Functional equivalent

When technicians hit the ground, they often find something that adds another layer of complexity to the tracking issues. They see a lot of things that look like best management practices but don't meet the rigorous guidance used by the USDA Natural Resource Conservation Service or the Bay Program.

For instance, stream banks might be fenced but the size of fence posts and number of wire strands may be different from state or federal requirements. Still, they keep cows out of streams. "That's a great example of something that we should be able to count," McNutt said.

That's given rise to a new term: functionally equivalent. Those are practices that don't meet the exact BMPs definitions that go into the Bay model, but would be expected to perform similarly.

It gets murkier with a related issue. There are many practices that fall short of the Bay Program or NRCS definitions for conservation practices, but which may have some benefit.

For instance, they require that farms have stream buffers at least 35 feet wide to get credit. Many farms have narrower stream buffers. Likewise, they may have grass waterways to carry runoff that are 8 feet wide instead of the required 10. Or they may be doing any number of other activities that fall short of the exacting specifications needed to receive cost-share funding, but presumably have some water quality benefit that has gone uncounted.

One of the major frustrations of farmers is getting blamed for Bay pollution, but not getting credit for their efforts, said Megan Dalton, district manager of the Shenandoah Valley Soil & Water Conservation District in Virginia. "That is the biggest issue that farmers have brought up," she said. "Why is it so hard to assign a credit? Everyone knows it is better than nothing."

Many of the pilot studies are taking note of such actions, in the hope that some nutrient reduction credit might be awarded in the future.

But the issue is difficult. The nutrient reductions set for various BMPs were established by teams of experts using the best available data. It might be difficult to set scientifically based nutrient reduction values for BMPs that do not meet federal standards because of a lack of research on such practices.

Phantom practices

While it's clear that some conservation practices have gone uncounted, the National Research Council report said that the data submitted to the EPA by states includes BMPs that don't exist, are double-counted or not functioning as expected.

Most BMPs have limited design lives, so their effectiveness can greatly diminish over time. Maryland and Virginia have programs to identify BMPs that have exceeded their design life and remove them from their records. Others stay on the books - or in the computer - generating nutrient reductions that may not be real.

For example, the Bay region has lost many dairy farms over the last two decades, some of which have been converted to other land uses. But the nutrient management plans, fences and stream buffers associated with those farms often stay on records.

"What is happening to these practices that were built a decade or longer ago?" asked Mark Dubin, an agricultural technical coordinator with the University of Maryland Extension who works with the Bay Program and also owns and operates a farm in Pennsylvania. "Maybe there are no cows there. It may be a golf course or housing development."

In many cases, no one knows. The level of field verification varies widely by state, the National Research Council report said. New York tries to inspect practices every year, while Maryland and Virginia have programs that spot check a percentage of BMPs annually. Others have little or no state-level verification of BMPs, relying instead on the reports of conservation districts and others.

As a result, in some instances, states or counties have sometimes submitted reports to the EPA claiming more BMPs in a given area than there was land to place them on.

Poorly functioning BMPs

In other cases, BMPs may not be functioning as expected. Bay Program BMP guideline assume that manure is not placed on cover crops planted in the fall to absorb the excess nutrients left in fields after the crops. But all states allow the practice, at least under certain circumstances, although it sharply reduces the nutrient reduction effectiveness of the cover crop, said Tom Simpson, executive director of Water Stewardship, an Annapolis-based nonprofit organization that provides independent conservation verification for farmers.

Simpson was also the chair of the Bay Program's Nutrient Subcommittee, which developed nutrient reduction assumptions for BMPs credited in the Bay Program models.

"It's better than putting manure down on bare soil," Simpson said, "and it's not like they shouldn't get any credit. But they surely shouldn't be getting this real high credit that is based on no manure application." In places - especially areas where there is excess manure - the practice can be common, he said.

The nutrient reduction credit given for riparian buffers allows a minimum 35-foot-buffer, but the Bay Program assumes that on average, most buffers are wider. Yet in heavy production areas, buffers wider than 35 feet are rare, Simpson said.

In some instances, changes in farm operations mean a well-designed BMP no longer performs as planned. If a 100-cow dairy farm builds a manure storage facility but then adds 50 cows, that facility is no longer able to store manure as effectively, or as long, as planned. As a result, the farmer may have to apply manure to fields when it's not needed.

"If we are going to count all of these voluntary BMPs, we certainly need to determine whether those things are already giving credit to are functioning as defined by the Bay Program," Simpson said.

Nutrient management plans, which provide direction for the application and management of manure and fertilizer, are the most widely used BMP in the Bay watershed. This spring's USDA report said that while the vast majority of farmers in the Bay watershed had nutrient management plans, many did not fully implement them. Those using manure, which typically results in more nutrient runoff than chemical fertilizer, tended to have the lowest level of plan compliance.

Officials can check records, but say it is hard to verify that manure and fertilizer are being applied in accordance with plans.

Virginia has a program that calls for inspecting at least 5 percent of cost-shared BMPs each year. Those inspections show more than 91 percent compliance, said Stephanie Martin, acting head of district programs for the Virginia Department of Conservation and Recreation. But, she said, verifying nutrient management is problematic. "We can't chase the spread wagon," she said.

Acts of God

Acts of God can also be a problem. Severe flooding in the wake of Tropical Storm Lee washed BMPs from the land, but not necessarily from the record books. Several officials in New York and northern Pennsylvania reported the loss of buffers, stream bank fencing, stream restoration and other BMPs. In some cases, even the land where those BMPs were located is gone.

No states have programs to account for BMPs lost in extreme events. The management practices already on record will contribute to modeled nutrient reduction progress, even though some no longer exist. They could stay in the model for years.

Severe weather can also degrade practices. For example, they can erode paths through grass waterways or streamside buffers which shunt water to streams, bypassing the buffers. "It doesn't matter if you have a 500-yard buffer if you have three gullies eroded in it which funnel all the water straight through," said Michael Helfrich, the Lower Susquehanna RiverKeeper.


EPA officials have said they are open to crediting "farmer-funded" and "functionally equivalent" BMPs, and even to establishing nutrient reduction values for lesser efforts, such as narrower stream buffers. But officials say adding previously uncounted efforts is only part of the equation. They say it needs to go hand-in-hand with new verification procedures and removing ineffective or practices that no longer exist from the books.

"You can't be loading a lot of practices into the model from the non-cost share side if you are not also making sure you have cleaned up any historic data that shouldn't be in there anymore," said Kelly Shenk, agricultural policy coordinator with the EPA's Chesapeake Bay Program Office.

EPA officials note that claimed nutrient reductions ultimately have to be reflected in actual water quality improvements in the Chesapeake.

Tracking farmer-funded BMPs presents a variety of challenges for state and county conservation programs already short of staff and funding.

"Ultimately, I think we are talking literally about going county by county, farm by farm. And there are only certain times of the year we can get out there to do this. When there are crops in the field you can't go out there. During hunting season you certainly can't go out there," said John Roderick, administrator for resource conservation with the Maryland Department of Agriculture.

Another challenge is verifying that practices, once counted, stay in place. If a farmer uses his own money to implement a BMP, he's also free to remove it whenever he wishes — unlike those who participate in a government cost-share program. In addition, cooperation is voluntary; farmers not participating in cost-share programs don't have to discuss what they do on the ground, unless they are subject to other regulations.

There is a price to be paid for accuracy. Time and money spent collecting BMP data doesn't implement new pollution control practices. "This is something the staff is not wild about," McNutt acknowledged. "They would rather just go out and put more BMPs in."

But there may be ways to improve tracking and control costs in the future. "Technology is coming along behind us," said York, who is helping to design the Howard County project. She envisions a day, not far in the future, where much of the job of verification, from ensuring the widths of buffers to ensuring the planting of cover corps, can be monitored by satellite.

But that will not completely replace the need for people on the ground, which has its own benefits. With the stepped-up verification efforts, field technicians will get closer interaction with farmers. For BMPs that don't quite meet standards, it may give them a chance to help farmers make the needed improvement.

"It's a great opportunity to sit with the farmer and say, based on what you see, some programs can assist them," Roderick said. And in some cases, he said, "we will be able to tell the farmer when we walk away that 'you are not part of the problem. You have met your obligation.'"