Modeling hourly subsurface drainage using steady-state and transient methods

Computer models have been frequently used to simulate the hydrologic and environmental processes in subsurface-drained cropland. The widely-tested steady-state Hooghoudt (ssH) equation, implemented in the Root Zone Water Quality Model (RZWQM2, version 2.94.00), serves in simulating subsurface drainage. However, transient methods such as the integrated Hooghoudt (inH) and van Schilfgaarde (vanS) equations have seldom been implemented in models. In the present study, RZWQM2's hydrologic component was modified to initiate the soil water redistribution process when rainfall occurred. The three drainage equations (ssH, inH and vanS) were tested in each of two versions of RZWQM2 (original and modified). Field data from Iowa (2007-2008) and Ontario (2009-2010) were used to evaluate different model version × equation combinations' simulation accuracy at both daily and hourly scales, evaluated using the percent of bias (PBIAS), Nash-Sutcliffe efficiency coefficient (NSE), and the Index of Agreement (IoA). On a daily scale and across equations, for the Iowa data the original model (PBIAS⩽14.96,NSE⩾0.40,⩾0.69) was outperformed by the modified model (PBIAS⩽6.48,NSE⩾0.70,IoA⩾0.76). Similarly, for the Ontario data, the original model (PBIAS⩽8.87,NSE⩾0.19,IoA⩾0.65) was outperformed by the modified model (PBIAS⩽3.59,NSE⩾0.31,IoA⩾0.67). However, based on a parity of PBIAS, NSE and IoA values, hourly scale tile drainage computed using the modified model equipped with transient equations did not improve model performance compared with the original ssH equation.