Estimation of flow and transport parameters for woodchip-based bioreactors: I. laboratory-scale bioreactor

In subsurface bioreactors used for tile drainage systems, carbon sources are used to facilitate denitrification. The objectives of this study were to investigate the validity of first- and zero-order reactions for a laboratory-scale bioreactor, and to estimate flow and transport parameters for a laboratory-scale bioreactor. The laboratory-scale bioreactor used in this study consisted of a polyvinyl chloride (PVC) pipe (0.25 m in diameter and 6.1 m in length) filled with woodchips, with a drainage control structure attached to each end. Creek water and deionised water with NO3-N concentrations ranging from 8 to 33.7 mg l(-1) was passed through the bioreactor at several flow rates. For the tests with creek water, complete (100%) nitrate reduction and approximately 10-40% nitrate reduction were observed at high retention times and at low retention times, respectively. The model CXTFIT2, developed by Toride et al. (1999, Research report No. 137, U.S. Salinity Lab., ARS, USDA, Riverside, California), and an analytical solution introduced by van Genuchten and Alves (1982, Technical Bulletin No. 1661, U.S. Salinity Lab., USDA, Riverside, California) were used to investigate the assumptions that the reactions obeyed first- and zero-order kinetics. The results revealed that the assumption of first-order reaction kinetics resulted in closer agreement between the model results and the data. However, the estimated reaction terms varied over a factor of 5. These variations were not improved with the single parameter estimation. From these laboratory-scale bioreactor studies, it was concluded that the assumptions of first-order decay for nitrate transport are justified for each single run, and that woodchip-based bioreactors may be suitable for significant nutrient reduction from agricultural fields, artificially drained by tile drain systems. (C) 2009 IAgrE. Published by Elsevier Ltd. All rights reserved.