Quantification of denitrification rate in shallow groundwater using the single-well, push-pull test technique

Denitrification has been identified as a significant nitrate attenuation process in groundwater systems. Hence, accurate quantification of denitrification rates is consequently important for the better understanding and assessment of nitrate contamination of groundwater systems. There are, however, few studies that have investigated quantification of shallow groundwater denitrification rates using different analytical approaches or assuming different kinetic reaction models. In this study, we assessed different analytical approaches (reactant versus product) and kinetic reaction (zero-order and first-order) models analysing observations from a single-well, push-pull tests to quantify denitrification rates in shallow groundwater at two sites in the Manawatū River catchment, Lower North Island of New Zealand. Shallow groundwater denitrification rates analysed using the measurements of denitrification reactant (nitrate reduction) and zero-order kinetic models were quantified at 0.42-1.07 mg N L^-1 h^-1 and 0.05-0.12 mg N L^-1 h^-1 at the Palmerston North (PNR) and Woodville (WDV) sites, respectively. However, using first-order kinetic models, the denitrification rates were quantified at 0.03-0.09 h^-1 and 0.002-0.012 h^-1 at the PNR and WDV sites, respectively. These denitrification rates based on the measurements of denitrification reactant (nitrate reduction) were quantified significantly higher (6 to 60 times) than the rates estimated using the measurements of denitrification product (nitrous oxide production). However, the denitrification rate quantified based on the nitrate reduction may provide representative value of denitrification characteristics of shallow groundwater systems. This is more so when lacking practical methods to quantify all nitrogen species (i.e., total N, organic N, nitrite, nitrate, ammoniacal N, nitrous oxide, nitric oxide, and nitrogen gas) in a push-pull test. While estimates of denitrification rates also differed depending on the kinetic model used, both a zero-order and a first-order model appear to be valid to analyse and estimate denitrification rate from push-pull tests. However, a discrepancy in estimates of denitrification rates using either reactant or product and using zero- or first-order kinetics models may have implications in assessment of nitrate transport and transformation in groundwater systems. This necessitates further research and analysis for appropriate measurements and representation of spatial and temporal variability in denitrification characteristics of the shallow groundwater system.