Comparative evaluation of phosphorus losses from subsurface and naturally drained agricultural fields in the Pike River watershed of Quebec, Canada M. Eastman a , A. Gollamudi a , N. Sta ¨ mpfli a , C.A. Madramootoo a , A. Sarangi b, * a Brace Centre for Water Resources Management, McGill University, Montreal, Canada b Water Technology Centre, IARI, New Delhi, India 1. Introduction Excessive phosphorus (P) loading into freshwater bodies has deteriorated water quality in many regions of the world. Intensive upstream agricultural land use activities tend to drastically increase P loading in downstream water bodies. Eutrophication has been identified as one of the leading threats to lake water quality worldwide (Harper, 1992). Negative effects of eutrophica- tion include: dissolved oxygen depletion, increases in suspended solids, decreased light penetration, and reduction in aquatic flora and fauna species (Migliaccio et al., 2007). Algal mats get deposited at the surface, limiting the recreational use of the lake and facilitate growth of toxin producing cyanobacteria, which pollute the water bodies. These pollutants complicate the water treatment process, often rendering the lake water unsuitable for human consumption. In response to elevated P concentrations in water bodies, various management strategies are being practiced around the world. The use of nutrient management plans and beneficial management practices are being promoted in North America. In addition, many regions use a modified version of the P-Index, which is used as a field scale tool to determine the risk of P loss from agricultural lands. There are many versions of the P-Index, developed specifically for a certain region; however, they all tend to be similar. Site factors such as soil characteristics, slope, soil test phosphorus (STP) concentration, percent P saturation (P sat ), fertilization practices, tillage practices, cropping patterns and distance of field to the nearest watercourse are typically included in the model (Sharpley et al., 2001). Tools such as the P-Index have been made available along with the development of hydrological and water quality models at the field, watershed and basin scales. Models such as ANSWERS (Beasley et al., 1980), CREAMS (Knisel, 1980), EPIC (Williams et al., 1984), GLEAMS (Leonard et al., 1987), AGNPS (Young et al., 1987), and SWAT (Arnold et al., 1998; Gollamudi et al., 2007) have all been used to simulate hydrology and P transport at different scales. Traditionally, it was believed that P loss from agricultural landscapes occurred primarily during surface runoff events and very little was lost through subsurface drainage. However, research investigations have revealed that the subsurface drainage systems in agricultural fields also discharge significant phosphorus quantities under a wide range of soil characteristics and Agricultural Water Management 97 (2010) 596–604 ARTICLE INFO Article history: Received 4 April 2009 Accepted 24 November 2009 Available online 22 December 2009 Keywords: Non-point source pollution Nutrient transport Surface runoff Subsurface drainage Water quality monitoring ABSTRACT Phosphorus (P) is the limiting nutrient responsible for the development of algal blooms in freshwater bodies, adversely impacting the water quality of downstream lakes and rivers. Since agriculture is a major non-point source of P in southern Quebec, this study was carried out to investigate P transport under subsurface and naturally drained agricultural fields with two common soil types (clay loam and sandy loam). Monitoring stations were installed at four sites (A, B, C and D) in the Pike River watershed of southern Quebec. Sites A–B had subsurface drainage whereas sites C–D were naturally drained. In addition, sites A–C had clay loam soils whereas sites B–D had sandy loam soils. Analysis of data acquired over two hydrologic years (2004–2006) revealed that site A discharged 1.8 times more water than site B, 4 times more than site C and 3 times more than site D. The presence of subsurface drainage in sandy loam soils had a significant beneficial effect in minimizing surface runoff and total phosphorus (TP) losses from the field, but the contrary was observed in clay loam soils. This was attributed to the finding that P speciation as particulate phosphorus (PP) and dissolved phosphorus (DP) remained relatively independent of the hydrologic transport pathway, and was a strong function of soil texture. While 80% of TP occurred as PP at both clay loam sites, only 20% occurred as PP at both sandy loam sites. Moreover, P transport pathways in artificially drained soils were greatly influenced by the prevailing preferential and macropore flow conditions. ß 2009 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +91 9811400885. E-mail addresses: ads_wtc@rediffmail.com, arjamadutta.sarangi@mail.mcgill.ca (A. Sarangi). Contents lists available at ScienceDirect Agricultural Water Management journal homepage: www.elsevier.com/locate/agwat 0378-3774/$ – see front matter ß 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.agwat.2009.11.010