Skip to main content
ARS Home » Midwest Area » West Lafayette, Indiana » National Soil Erosion Research Laboratory » Research » Publications at this Location » Publication #392019

Research Project: Managing Agricultural Systems to Improve Agronomic Productivity, Soil, and Water Quality

Location: National Soil Erosion Research Laboratory

Title: Source and transport controls on nutrient delivery to tile drains

item Williams, Mark
item Penn, Chad
item McAfee, Scott

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/29/2022
Publication Date: 7/8/2022
Citation: Williams, M.R., Penn, C.J., Mcafee, S.J. 2022. Source and transport controls on nutrient delivery to tile drains. Journal of Hydrology. 612(B).Article 127146.

Interpretive Summary: Nitrogen and phosphorus loss from agricultural tile-drained fields is controlled by the nutrient source and nutrient transport processes. In this study, we measured water, nutrient, and tracer movement in a tile-drained field in Indiana to understand how agricultural management practices and water flow pathways impacted nutrient delivery to tile drains. Results showed that management practices including fertilizer application rate, fertilizer placement, and fertilizer application timing directly influenced the amount of nutrient loss during the 30 day periods following fertilizer application. Since nitrate-nitrogen is highly mobile in the soil profile, a broad range of concentration was observed across different flow pathways and antecedent wetness conditions. Dissolved phosphorus concentration was greatest when the soil was dry and much of tile flow was comprised preferential flow, and lowest when the soil was wet and tile flow was comprised of groundwater. The results of this study provided new insights into the processes and management practices that influence nutrient loss to tile drains.

Technical Abstract: Apportioning sources, assessing flow pathways, and quantifying the interaction between source and transport factors are critical for decreasing nutrient loss from agricultural catchments. In this study, water, nutrient, and tracer (d18O, electrical conductivity) fluxes were measured over two years from a tile-drained field in Indiana, USA to quantify linkages among water flow pathways, management practices, and nutrient delivery to tile drains. Three-component hydrograph separation showed that tile discharge was, on average, sourced from groundwater (65±31%), shallow soil water (10-20-cm deep; 29±28%), and precipitation (6±8%). Daily nitrate-N (NO3-N) and dissolved reactive phosphorus (DRP) load ranged from 0.0-1.8 kg ha-1 and 0-101 g ha-1, respectively, while cumulative loads were 30.7 kg ha-1 and 1,040 g ha-1. Nutrient management practices (fertilizer rate, placement, timing) directly influenced the magnitude of incidental nutrient loss and proportion of load that was derived from a fertilizer source. It was estimated that 52% and 46% of cumulative NO3-N and DRP load, respectively, was incidental loss that occurred within 30 d of fertilizer application. Continuous re-distribution of NO3-N throughout the soil profile following fertilizer application resulted in a broad range of concentration across flow pathways and antecedent conditions. In contrast, dry conditions with tile water sourced from shallow soil water or precipitation resulted in greater, but more variable, DRP concentration compared to wetter conditions when tile water was largely comprised of groundwater. Findings suggest that groundwater table dynamics exerted a strong control over DRP concentration, as similar surface-tile hydrologic connectivity occurred during both dry and wet conditions but produced different water quality outcomes. Groundwater may therefore serve both as a chemical and hydrologic buffer of DRP concentration. Combined measurements of water, nutrient, and tracer fluxes revealed novel insights into processes controlling nutrient delivery to tile drains, with direct applicability for conservation practice implementation and improving process representation in models.