Legacy Nutrient Management and Remediation in Diverse Agricultural Watersheds

Abstract

Excess nutrient loading from agricultural landscapes remains a primary driver of freshwater eutrophication in the United States, despite decades of conservation investment. In many watersheds, water quality impairments persist due to the accumulation of legacy nutrients—phosphorus (P) stored in soils and nitrogen (N) stored in groundwater—that continue to be mobilized long after management changes are implemented. This dissertation evaluates the magnitude, persistence, uncertainty, and mitigation potential of legacy nutrient pollution across watershed and edge-of-field scales, with a particular emphasis on phosphorus dynamics in surface waters and nitrogen removal from groundwater. The first chapter quantifies the spatial and temporal expression of legacy phosphorus export in a variable source area (VSA)–dominated agricultural watershed within the Lake Champlain Basin. Using the SWAT–VSA model calibrated for streamflow and total phosphorus (TP), this chapter examines long-term phosphorus export under current agricultural management. Results demonstrate that TP loads are highly persistent over decadal timescales, with a disproportionate contribution from hydrologically connected, high-risk landscape units. Soil phosphorus accumulation remains stable or increases in many areas, indicating limited natural drawdown of legacy P under baseline conditions and underscoring the challenge of achieving watershed-scale TP reductions without targeted intervention. The second chapter evaluates the effectiveness and uncertainty of widely implemented agricultural best management practices (BMPs) for mitigating legacy phosphorus at the watershed scale. A Monte Carlo ensemble modeling framework is used to propagate uncertainty in key P-related parameters and to assess both single and stacked BMP scenarios, including cover crops, forest buffers, reduced fertilizer application, no-till management, and crop phosphorus removal (P mining). Results show that most structural and management BMPs yield modest and highly uncertain TP reductions when applied uniformly across the watershed. In contrast, crop P removal consistently produces the largest and most predictable long-term reductions in watershed TP export by directly reducing soil P stocks. However, even under aggressive BMP implementation, the probability of meeting regulatory load reduction targets remains low, highlighting the dominant role of legacy P and the limitations of non-targeted BMP strategies. The third chapter shifts focus from surface-water phosphorus to groundwater-derived nitrate, evaluating the performance of a large-scale woodchip bioreactor treating emergent nitrate-rich groundwater in the Shenandoah Valley, Virginia. Using over two years of operational data, this chapter examines how hydraulic residence time (HRT), influent nitrate concentration, temperature, and flow variability influence nitrogen removal performance. A log-linear modeling framework is used to distinguish between normalized removal rates and total nitrogen removal. Results indicate that shorter HRTs maximize normalized nitrogen removal rates, while total nitrogen removal remains relatively stable across operating conditions due to compensating effects among flow, temperature, and loading. These findings emphasize the importance of designing bioreactors to accommodate hydrologic variability rather than targeting a single optimal residence time. Together, these three chapters provide an integrated assessment of legacy nutrient behavior and mitigation across scales, demonstrating that (1) legacy phosphorus strongly constrains watershed-scale water quality improvements, (2) BMP effectiveness is highly uncertain without explicit soil P drawdown strategies or targeted placement, and (3) engineered systems such as bioreactors can effectively mitigate legacy nitrogen when designed to operate under dynamic environmental conditions. The dissertation offers practical insights for watershed managers, conservation agencies, and policymakers seeking realistic, risk-informed strategies for long-term nutrient reduction.