Contents lists available at ScienceDirect Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat Coupling of a nitrate production model with HYDRUS to predict nitrate leaching J.-P. Matteau a,b, ⁎ , S.J. Gumiere a , J. Gallichand a , G. Létourneau a , L. Khiari a , M.-O. Gasser b , A. Michaud b a Dept. of Soils and Agri-Food Engineering, Université Laval, Québec, QC G1V 0A6, Canada b Institut de recherche et de développement en agroenvironnement (IRDA), Québec, QC G1P 3W8, Canada ARTICLE INFO Keywords: Nitrogen management Modeling Organic fertilizer Mineral fertilizer Water pollution ABSTRACT Losses of nitrogen fertilizer from agricultural watersheds are among the main causes of water quality de- gradation in Quebec. Under maize crop, leached nitrate can reach up to 185 kg N ha −1 . The synchronization of the nitrogen fertilization with crop uptake would be a way to optimize the use of nitrogen fertilizer and reduce nitrate leaching. Coupling an empirical nitrogen cycle model with a physical soil water dynamics model could simplify the optimization of nitrogen fertilizer use. The objective of this study was to couple the soil surface empirical nitrate production model adapted to the regional agro-climatic conditions of study area with HYDRUS to determine the most suitable combination in order to predict nitrate leaching into subsurface drains. In this study, we used estimation equations of nitrogen pools and transformations in nitrate. The equations, using historical feld data, were combined into 60 N-release patterns taking into account a dissolution function of nitrate, 4 transformation and dissolution functions for the ammonium, 5 organic nitrogen pools and 3 N organic nitrogen from the soil and manure release functions. These N release patterns were applied in a mineral fertilizer treatment and four organic fertilizer treatments for a total of 300 unique nitrate release patterns. The results demonstrate the potential of a method to evaluate the contributions of nitrogen based on at- mospheric data, crop rotation or soil particle sizes as 5% of the 300 combination gave a Nash–Sutclife efciency coefcient (NSE) above 0.70. The prediction method must be diferent according to the source of the fertili- zation, organic or mineral as only 1% of the combination gave NSE above 0.70 in more than one treatment. In addition, the study showed that the soil organic nitrogen contribution to nitrate leaching tends to decrease as the organic fertilizer applications rate increases as 75% of the combination giving NSE above 0.70 in high fertili- zation treatment used the lower organic N pool and only 17% used the lower organic N pool for the agronomic fertilization rate and mineral fertilizer. The simulations also showed that the nitrate-leached masses are closely related to the fertilizer supply nitrogen content. 1. Introduction Nitrogen used for agricultural fertilization is the main cause of an- thropogenic perturbations of the N cycle (Fowler et al., 2013). An ex- cessive input of nitrogen in waterways may result in accelerated eu- trophication of lakes and rivers (Dupas et al., 2015; Erisman et al., 2013). Along with pesticides, nitrate represents the main source of aquifer contamination in agricultural areas (Dupuy et al., 2012). Nitrate leachate can reach 185 kg ha −1 in groundwater under maize crop (Gollamudi, 2006). Despite these environmental risks, nitrogen fertili- zation remains the basis of modern agriculture; it is the limiting nu- trient for crop growth to the extent that nearly half of the human population depends on it for food supply (CRAAQ, 2010; Erisman et al., 2013; N’Dayegamiye et al., 2007). In 2009, in Canada, 75% of the mass of fertilizers used in agriculture was nitrogen (Agriculture and Agri-Food Canada, 2014). Nitrate leaching and the resulting surface and groundwater contamination are a concern, especially in maize cultivated areas because this crop re- quires high nitrogen fertilizer applications over extensive areas. One way to reduce nitrate leaching is to synchronize nitrogen fertilization with crop uptake. Synchronizing nitrogen availability with nitrate up- take can be performed indirectly by modeling nitrate leaching under the root zone or in drain tiles. Various process-oriented and physically- based solute transport models have been used to this end: Leaching https://doi.org/10.1016/j.agwat.2018.10.013 Received 16 February 2018; Received in revised form 9 October 2018; Accepted 11 October 2018 ⁎ Corresponding author at: Dept. of Soils and Agri-Food Engineering, Université Laval, Québec, QC G1V 0A6, Canada. E-mail address: jean-pascal.matteau.1@ulaval.ca (J.-P. Matteau). Agricultural Water Management 213 (2019) 616–626 0378-3774/ © 2018 Elsevier B.V. All rights reserved. T