Biological Systems Engineering, Department of


Document Type


Date of this Version



Water 2017, 9, 517; doi:10.3390/w9070517.


Messer, Burchell, & Bírgnd in MDPI Water (2017) 9. Copyright © 2017, the authors. Licensee MDPI, Basel, Switzerland. Open access, Creative Commons Attribution license 4.0.


The objective of the study was to determine the kinetic model that best fit observed nitrate removal rates at the mesocosm scale in order to determine ideal loading rates for two future wetland restorations slated to receive pulse flow agricultural drainage water. Four nitrate removal models were investigated: zero order, first order decay, efficiency loss, and Monod. Wetland mesocosms were constructed using the primary soil type (in triplicate) at each of the future wetland restoration sites. Eighteen mesocosm experiments were conducted over two years across seasons. Simulated drainage water was loaded into wetlands as batches, with target nitrate-N levels typically observed in agricultural drainage water (between 2.5 and 10 mg L-1). Nitrate-N removal observed during the experiments provided the basis for calibration and validation of the models. When the predictive strength of each of the four models was assessed, results indicated that the efficiency loss and first order decay models provided the strongest agreement between predicted and measured NO3-N removal rates, and the fit between the two models were comparable. Since the predictive power of these two models were similar, the less complicated first order decay model appeared to be the best choice in predicting appropriate loading rates for the future full-scale wetland restorations.