Nitrogen loss from farmers’ fields is an economic loss to the farm and an environmental concern.
Runoff from a farm; Image created with AI (Grok)
Authors: Bradley Kennedy and Gurpal Toor
What is the issue?
Nitrogen loss from farmers’ fields is an economic loss to the farm and an environmental concern when it reaches the Chesapeake Bay, contributing to water quality degradation. Despite farmers implementing conservation practices for decades, there has been little change in the estimated nitrogen loads from agricultural land in the Chesapeake Bay Watershed. To understand why this is so and improve the science behind conservation practices, we need more comprehensive research on the specific characteristics and dynamics of nitrogen lost in agricultural runoff. For example, earlier research typically investigated only a few forms of nitrogen, only conducted sampling for part of the year, and/or relied on single composite samples to describe nitrogen loss over an entire storm event. Notably, most research has historically been conducted at the plot or watershed scale, rather than at the field scale (referred to here as small catchments, which range from five to 200 acres), the scale at which land is managed, and conservation practices are implemented.
What did we find and why does it matter?
Our ongoing data analysis is addressing these critical knowledge gaps by enhancing our understanding of when, why, and how nitrogen is lost from agricultural fields in Maryland, and in what forms. We are exploring the patterns of different nitrogen forms lost at various time scales, including within and across individual storm events, seasons, and years. Our analysis encompasses both nitrogen concentrations and total loads, as concentrations are useful for understanding the drivers of nitrogen loss, while loads are essential for assessing the overall amount of nitrogen loss and associated environmental risk. This information will be crucial in reducing nitrogen loss by helping to develop and refine conservation practices that protect water quality in Maryland's agricultural watersheds and those around the world.
We are analyzing over 4 years of data for our surface runoff site. Results from all three drainage types will be published in the coming months. Although the analysis is still ongoing, here are some of our preliminary findings for surface runoff:
- Dissolved organic nitrogen (DON) is a major fraction of dissolved nitrogen loss in surface runoff. Over the three years of our study, DON contributed most of the total dissolved N load, exceeding the combined contributions of nitrate and ammonium. This is an important finding because research often fails to measure DON or distinguish it from particulate organic N (PON), which has a different source and behaves differently in water.
- Our results indicate that most N is lost in just a handful of large runoff events. Further, we found that this effect was stronger for nitrate (91% lost in the 4 largest events) than for DON (60% lost in the 4 largest events). This suggests that conservation practices should be designed with large runoff volumes in mind.
- Our research also indicates that concentrations of both particulate and dissolved nitrogen are usually highest at the beginning of a runoff event. This means that when a conservation practice cannot treat all the runoff, it is more effective to treat the early part of it.
- Our data revealed strong seasonal differences in DON and nitrate losses. During the winter and spring, most dissolved nitrogen was lost in the form of DON, but in the summer, most dissolved nitrogen was lost as nitrate. During the fall, losses from both forms were similar. Understanding this type of seasonal variation is critical for designing effective conservation practices that target different forms of nitrogen at various times of the year.
Figure 1. Forms of dissolved nitrogen loss in different seasons.
What did we do?
We measured field-scale nitrogen loss on three working farms in Maryland with corn-soybean rotations. We monitored (1) surface runoff on the Western Shore of the Bay, (2) tile drainage on the mid-Eastern Shore, and (3) open ditch drainage on the lower Eastern Shore (see photos). Instruments at the catchment outlets measured rainfall and the amount of water leaving the field as surface or subsurface runoff. Automated samplers collected up to 24 runoff water samples during each storm event. Runoff samples were collected year-round to assemble the most complete data set possible. We visited the sites every week and brought samples back to the lab for analysis to determine the amount of nitrogen present in each form.
Because nutrient losses vary significantly from year to year depending on weather and other conditions, we needed several years of data for a thorough analysis. As of December 2025, we have nearly 4 years of data and have transitioned to ongoing data analysis. During the monitoring period, we collected the most samples from our ditch drainage outlet (691 samples in 70 events), followed by tile drainage (475 samples in 38 storm events) and surface runoff (210 samples in 27 storm events).
Credit: Gurpal Toor, University of Maryland
Publications completed for this work
Ji, W., Liu, Y., Wang, J., Toor, G. S., He, X., & Li, Z. (2024). Multiple isotopes reveal the impact of land use change on nitrate transport and transformation in deep loess deposits. Journal of Hydrology. https://doi.org/10.1016/j.jhydrol.2024.131056
Kennedy, B. C., & Toor, G. S. (2026). Field-scale nitrogen loss from agricultural land to water: Pathways, drivers, and management implications. Advances in Agronomy, 198. (In press).
Wang, S., Yu, X., Xia, Y., Gao, J., Chen, Z., Toor, G. S., & Zhou, J. (2024). Changes in nutrient surpluses and contents in soils of cereals and kiwifruit fields. Agronomy, 14, 2556. https://doi.org/10.3390/agronomy14112556
Yang, Y., Tfaily, M. M., Wilmoth, J. L., & Toor, G. S. (2022). Molecular characterization of dissolved organic nitrogen and phosphorus in agricultural runoff and surface waters. Water Research, 118533. https://doi.org/10.1016/j.watres.2022.118533
