ANNAPOLIS, Md. – Cleaning up the Chesapeake Bay is a lot like trying to sail into the wind. The heavy breeze keeps pushing back against efforts to make progress.
That’s illustrated by recent data from the state-federal Bay Program, which shows that roughly half of the efforts by states to reduce nitrogen pollution during the last 14 years have been offset by headwinds created by growth, climate change and the filling of the reservoir behind the Conowingo Dam.
The result: After nearly a decade and a half of work and record spending, states are less than a third of the way toward achieving the nitrogen reductions needed to meet the Bay’s clean water goals.
Further, the path forward is becoming more difficult, relying on pollution reductions from farms and development lands, where progress has been especially difficult.
New data from the Bay Program’s computer models reveal a cleanup effort that is further off track than generally recognized and without a clear trajectory toward achieving nutrient pollution goals set in 2010.
To be clear, there have been some improvements, the models show.
The latest data indicate an uptick in progress in 2023, suggesting that the recent influx of state and federal money to bolster cleanup efforts is having an impact.
“We are making progress,” said Lee McDonnell, who oversees science, analysis and implementation at the Bay Program. “But there are things that are making the job we [started] back in 2010 more challenging.
“Despite the increases we’ve had in population and agricultural production and impacts from climate change, we’re not going in reverse,” he said. “We’re making some gains, but not as quickly as we’d like to. But again, we’re fighting those headwinds.”
In 2010, the U.S. Environmental Protection Agency established a total maximum daily load, or TMDL, for the Bay region.
The TMDL specified the amount of nutrient reductions — nitrogen and phosphorus — needed in each state and major river to achieve Bay water quality objectives.
Nutrients spur algae blooms in the Chesapeake that cloud the water, killing underwater grass beds that provide critical habitat. When the algae die, they decompose in a process that removes oxygen from the water, making the water inhospitable or even deadly to aquatic life.
The nutrient reductions prescribed by the TMDL were intended to fix those problems.
But reaching those goals has proven more daunting than expected. And the EPA and Bay states have recognized they will fall short of their self-imposed 2025 deadline.
Adding up the offsets
Under the TMDL, states needed to reduce the amount of nitrogen reaching the Bay in a typical year by about 71 million pounds. Through 2023, they had taken actions to slash that by about 40 million pounds a year.
But several factors have offset much of that, according to Bay Program data.
- Climate change is increasing precipitation, which washes more nutrients off the land. That adds about 5 million pounds of nitrogen annually back into the equation.
- After many decades of trapping sediment and nutrients flowing down the Susquehanna River, the reservoir behind Conowingo Dam — about 10 miles upstream from the Bay — is essentially at capacity. As a result, nutrients and sediment are flowing downstream. It will require 6 million pounds of additional annual nitrogen reductions to offset the impact of the reservoir’s filling.
- Recently updated data, which reflects greater numbers of farm animals, increased fertilizer use on farms, impacts from developed lands and a variety of other changes has effectively erased more than 8 million pounds of estimated annual nitrogen reduction progress.
Taken as a whole, those headwinds offset nearly half of the nitrogen control actions taken by states since 2009.
And those trends are not expected to diminish. Agricultural intensification and development are projected to continue, and Bay Program computer modeling suggests climate change impacts on nutrients will accelerate.
Also problematic: Most nutrient reductions in the last 14 years came from wastewater treatment plant upgrades. Those are nearly completed. But state cleanup plans have suggested the amount of nutrients from those plants could eventually begin to increase because of expected population growth.
That means nearly all future nutrient reductions need to come from controlling runoff, or “nonpoint sources,” from farms — the largest source of nutrients — and developed lands.
All states face challenges in controlling those sources, but the issue is most pronounced in Pennsylvania, which has far more farms and developed lands than the other states.
Overall, the latest Bay Program figures show the amount of nitrogen from nonpoint sources has changed little since 2009, leaving the region with no clear trajectory as to when its nitrogen reduction goal would be met.
A report from the Bay scientific community last year highlighted the problem, warning that current actions and programs aimed at controlling runoff are unlikely to achieve their goals.
“It’s a nonpoint source game from here on out, in terms of making progress,” said Kurt Stephenson, a Virginia Tech professor of agricultural economics and one of the authors of the report.
“It’s hard to get enough implementation to move the needle in something that’s so pervasive,” Stephenson said. “That’s not even opening up the question of whether we are actually getting the reductions we think we are.”
Last year’s report raised questions about whether actions aimed at controlling runoff are as effective at reducing nutrients as assumed in computer modeling.
It said those questions were particularly significant for phosphorus, the other key nutrient affecting Bay water quality.
States needed to reduce the amount of phosphorus reaching the Bay each year by about 4 million pounds. According to computer model estimates, they have made more progress on phosphorus than nitrogen — even when the “headwinds” are factored in. But they still have to reduce phosphorus by about 1.3 million pounds per year.
There is significant uncertainty about those figures, though, because there is a much greater divergence between model estimates and water quality monitoring for phosphorus than with nitrogen.
Last year’s science report noted that monitoring shows “limited evidence of observable reductions in phosphorus concentrations.” Many areas of the Bay watershed are showing either no trends, or increasing trends, for phosphorus.
What to make of models
The Bay Program models are the key scorecard used to assess progress toward meeting cleanup goals.
They use a vast amount of data about land use, farms, discharges from wastewater treatment plants, impacts from air pollution and other factors to estimate the amount of water-fouling nutrients that reach the Bay.
The models also use state-generated information about pollution control actions taken each year, such as wastewater treatment plant upgrades, streamside buffer plantings and the use of cover crops, to calculate how much those actions would reduce nutrients in the Bay.
Water quality monitoring is not directly used to evaluate progress because there are natural year-to-year fluctuations in nutrient runoff. Years with more rainfall have more runoff into rivers, while dry years have less. Also, it often takes many years before runoff control actions impact water quality. Much of the nitrogen reaches streams through slow-moving groundwater, and things like forest buffers can take years to become fully effective.
Modeling offers a way to assess how actions reported by states would be expected to affect nutrients levels in the Bay when they are fully effective and under “average” weather conditions.
Increasingly, though, states, local governments and agricultural groups have questioned the models, saying they do not accurately reflect the impact of pollution control actions, often pointing to the lack of agreement between monitoring results and modeling data at specific locations.
States and agricultural groups have strongly questioned the accuracy of some of the newest data, especially figures showing significantly increased fertilizer use.
Debate over the new data lasted nearly two years before states last fall signed off on their use in future modeling, with the caveat that related nutrient increases would not have to be addressed until after 2025.
Trust but verify
Joe Wood, senior scientist with the Chesapeake Bay Foundation’s Virginia office, said the Bay Program needs a way to assess progress that ensures nutrient control actions are actually improving water quality rather than over-relying on model results.
Citing concerns raised in last year’s report from the scientific community, Wood said it should not be automatically assumed that all runoff control actions are as effective as thought.
Some are likely more effective than others, and some are likely more effective in certain places than in others, he noted — yet they all get the same nutrient reduction credit in models.
“Fundamentally, I think it’s really important that we don’t treat nonpoint sources as a problem that we have all the answers to and [assume] that we just need to get more practices on the ground,” Wood said.
Still, he said the models are helpful in assessing the relative level of effort states make each year as measured through data they report about the use of runoff controls.
Wood noted the uptick in actions in 2023 to control farm runoff, especially in Virginia and Pennsylvania, which he said were “a clear response from recent investments … It is important we acknowledge that, but we also have many issues with our current incentive programs that we must address, and it is going to take leadership and innovation to do so.”
Despite improvements in 2023, when the models incorporate the latest data, it shows that the region has a long way to go.
Under the TMDL, the region needs to reduce the amount of nitrogen reaching the Bay by about 71 million pounds a year, to 199 million pounds annually on average, to achieve Bay water quality objectives.
Factoring in the latest data, along with the impacts of climate change, growth and the Conowingo dam, the watershed as a whole still needed about 50 million pounds of annual nitrogen reductions at the end of 2023.
The District of Columbia and West Virginia have met their goals, but all of the other states are off track.
Under current cleanup plans, more than three fifths of future nitrogen reductions would need to come from Pennsylvania, which has more agriculture and developed lands in the Bay watershed than any other state.
The latest data, for 2023, does show that Pennsylvania had the greatest estimated nitrogen reductions from farmland based on actions taken that year. The reductions — about 1.8 million pounds — reflect greatly ramped-up spending.
But the most recent computer model figures, released in late May, also reveal some troubling signs:
- According to previous models, states had taken enough actions since 2009 to reduce the amount of nitrogen running off farms each year by nearly 12 million pounds. But when new data were incorporated, reflecting increased fertilizer use and animal populations, along with other updates, the region as a whole had netted a combined reduction of only about 2 million pounds from farms. Those figures do not reflect the impact of the Conowingo Dam, which would require an additional 6 million pounds of annual reductions, mostly from Pennsylvania farmland.
- Nitrogen runoff from developed lands has increased by about 1.5 million pounds a year since 2009, as more land is converted to roads, buildings and parking lots.
- Nitrogen from septic systems has increased by about 250,000 pounds annually.
- On the positive side, actual measured reductions from wastewater treatment plants account for the vast majority of nitrogen improvements, slashing the amount reaching the Bay each year by 29.6 million pounds.
