Long-term monitoring data from agricultural watersheds are needed to determine if efforts to reduce nutrient transport from crop and pasture land have been effective. Goodwater Creek Experimental Watershed (GCEW), located in northeastern Missouri, is a high-runoff-potential watershed dominated by claypan soils. The objectives of this study were to: (i) summarize dissolved NH-N, NO-N, and PO-P flow-weighted concentrations (FWC), daily loads, and yields (unit area loads) in GCEW from 1992 to 2010; (ii) assess time trends and relationships between precipitation, land use, and fertilizer inputs and nutrient transport; and (iii) provide context to the GCEW data by comparisons with other Corn Belt watersheds. Significant declines in annual and quarterly FWCs and yields occurred for all three nutrient species during the study, and the decreases were most evident for NO-N. Substantial decreases in first- and fourth-quarter NO-N FWCs and daily loads and modest decreases in first-quarter PO-P daily loads were observed. Declines in NO-N and PO-P transport were attributed to decreased winter wheat ( L.) and increased corn ( L.) production that shifted fertilizer application from fall to spring as well as to improved management, such as increased use of incorporation. Regression models and correlation analyses indicated that precipitation, land use, and fertilizer inputs were critical factors controlling transport. Within the Mississippi River Basin, NO-N yields in GCEW were much lower than in tile-drained areas, but PO-P yields were among the highest in the basin. Overall, results demonstrated that reductions in fall-applied fertilizer and improved fertilizer management reduced N and P transport in GCEW.

The U.S. Department of Agriculture‐Agricultural Research Service's (ARS) Experimental Watershed Network grew from Dust Bowl era efforts of the Soil Conservation Service in the mid‐1930s with the establishment of small experimental watersheds. In the 1950s, five watershed research centers with intensively instrumented watersheds at the scale of 100 to 700 km² were established. Primary network research objectives were to quantify on‐site and downstream effects of conservation practices and develop rainfall‐runoff relationships for design of water conservation structures. With passage of the Clean Water Act in 1972, research objectives have evolved to add a variety of observations relevant to the water quality issues. Many of the watersheds within the network have served, and continue to serve, as core validation sites for satellite sensors. As a result of the network's long history and intensive monitoring, coupled with mission‐driven research, a deep knowledge base of watershed processes has been developed. This has led to the extensive development and validation of numerous watershed models that are in widespread use today. The visionary investments in building and maintaining this network and associated scientific investigations for more than half a century have not only resulted in numerous high‐impact research accomplishments but also a wide array of accomplishments that directly benefit society. The ARS Experimental Watersheds formed the core of the Conservation Effects Assessment Project (CEAP) as well as the recently established Long‐Term Agroecosystem Research (LTAR) network. LTAR will expand the mission of the ARS Watersheds Network to include agricultural intensification, maintaining or improving ecosystem services while enhancing rural prosperity.

The Agricultural Policy Environmental eXtender (APEX) model is capable of estimating edge‐of‐field water, nutrient, and sediment transport and is used to assess the environmental impacts of management practices. The current practice is to fully calibrate the model for each site simulation, a task that requires resources and data not always available. The objective of this study was to compare model performance for flow, sediment, and phosphorus transport under two parameterization schemes: a best professional judgment (BPJ) parameterization based on readily available data and a fully calibrated parameterization based on site‐specific soil, weather, event flow, and water quality data. The analysis was conducted using 12 datasets at four locations representing poorly drained soils and row‐crop production under different tillage systems. Model performance was based on the Nash–Sutcliffe efficiency (NSE), the coefficient of determination (r²) and the regression slope between simulated and measured annualized loads across all site years. Although the BPJ model performance for flow was acceptable (NSE = 0.7) at the annual time step, calibration improved it (NSE = 0.9). Acceptable simulation of sediment and total phosphorus transport (NSE = 0.5 and 0.9, respectively) was obtained only after full calibration at each site. Given the unacceptable performance of the BPJ approach, uncalibrated use of APEX for planning or management purposes may be misleading. Model calibration with water quality data prior to using APEX for simulating sediment and total phosphorus loss is essential.Core Ideas: Uncalibrated, APEX produced unacceptable site‐specific sediment and TP estimates. Acceptable runoff estimates do not translate to acceptable water quality estimates. Distributions of successfully calibrated performance indicator values were normal. Calibration of the APEX model with water quality data remains an essential step.

Potential impacts of climate change on the hydrological components of the Goodwater Creek Experimental Watershed were assessed using climate datasets from the Coupled Model Intercomparison Project Phase 5 and Soil and Water Assessment Tool (SWAT). Historical and future ensembles of downscaled precipitation and temperature, and modeled water yield, surface runoff, and evapotranspiration, were compared. Ensemble SWAT results indicate increased springtime precipitation, water yield, surface runoff and a shift in evapotranspiration peak one month earlier in the future. To evaluate the performance of model spatial resolution, gridded surface runoff estimated by Lund–Potsdam–Jena managed Land (LPJmL) and Jena Diversity-Dynamic Global Vegetation model (JeDi-DGVM) were compared to SWAT. Long-term comparison shows a 6–8% higher average annual runoff prediction for LPJmL, and a 5–30% lower prediction for JeDi-DGVM, compared to SWAT. Although annual runoff showed little change for LPJmL, monthly runoff projection under-predicted peak runoff and over-predicted low runoff for LPJmL compared to SWAT. The reasons for these differences include differences in spatial resolution of model inputs and mathematical representation of the physical processes. Results indicate benefits of impact assessments at local scales with heterogeneous sets of parameters to adequately represent extreme conditions that are muted in global gridded model studies by spatial averaging over large study domains.