Effect of Pipe Pressurization on the Discharge of a Tile Drainage System

Abstract

purpose, subsurface drainage may aff ect the catchment’s hydrologic response (Robinson, 1990; Skaggs et al., 1994). Studies on its downstream impact, especially on fl ooding, involve many fl ow processes taking place at diff erent scales (Robinson and Rycroft, 1999). At the fi eld scale, subsurface drainage draws the water table down, which conse-quently increases the storage capacity of the soil and signifi cantly decreases surface runoff (Augeard et al., 2005; Kao et al., 1998; Konyha et al., 1992; Robinson and Rycroft, 1999; Skaggs et al., 1994). At the catchment scale, the transfer of drained water from the fi eld to the outlet, through buried pipe collector networks, is subject to hydraulic resistance occurring (i) through the soil profi le (Dierickx, 1999; Stuyt et al., 2005) and (ii) in the buried pipe col-lector networks themselves (Dolezal et al., 2001; Oberlin, 1981).First, resistance of water fl ow in the soil profi le depends on the soil’s hydraulic properties and the drainage system’s design parameters (e.g., Lesaff re and Zimmer, 1988; Wiskow and van der Ploeg, 2003); both are used to describe the drainage discharge in relation to the infl ating or defl ating processes of the water table (Bouarfa and Zimmer, 2000) and in relation to the partial ponding conditions caused by the high water table (Salem and Skaggs, 1998). Around and along the drain pipe, the convergence of streamlines toward the inlet perforations of the drain induces radial and entrance resistance as well as a corresponding head loss (Dierickx, 1999; Kohler et al., 2001; Stuyt et al., 2005). For high drainage rates, these head losses can lead to the water table level rising above the drain and then throughout the soil profi le. Head losses due to streamline convergence toward drain inlets have been taken into account using an equivalent drain diam-eter smaller than the actual one (Dierickx, 1999). Th eoretical formulas for drainage rates and simulation models, such as DRAINMOD (Skaggs, 1980), SIDRA (Bouarfa and Zimmer, 2000), and HYDRUS-2D (Simunek et al., 1999), propose meth-ods to compute drainage rates during high fl ows, taking into account head losses attributable to soil and drain resistance.Th e drainage models assume free pressure conditions in the drain; however, water transfer in the buried pipe network can be constrained by pipe pressurization, which may signifi cantly modify both drainage discharge and water table behavior. Th e pressurization may result from undersize pipes (e.g., Cavelaars et al., 1994; El-Gammal et al., 1995) or from submersion of the