Elucidating phosphorus removal dynamics in a denitrifying woodchip bioreactor

Gimhani N. Perera,Dorisel Torres Rojas,Aldrin Rivas,Greg Barkle,Brian Moorhead,Louis A. Schipper,Rupert Craggs,Adam Hartland
DOI: https://doi.org/10.1016/j.scitotenv.2024.170478
IF: 9.8
2024-02-08
The Science of The Total Environment
Abstract:Denitrifying woodchip bioreactors (DBRs) are an established nitrate mitigation technology, but uncertainty remains on their viability for phosphorus (P) removal due to inconsistent source-sink behaviour in field trials. We investigated whether iron (Fe) redox cycling could be the missing link needed to explain P dynamics in these systems. A pilot-scale DBR (Aotearoa New Zealand) was monitored for the first two drainage seasons (2017–2018), with supplemental in−field measurements of reduced solutes (Fe 2+ , HS − /H 2 S) and their conjugate oxidised species (Fe 3+ /SO 4 2− ) made in 2021 to constrain within-reactor redox gradients. Consistent with thermodynamics, the dissolution of Fe 3+ (s) to Fe 2+ ( aq ) within the DBR sequentially followed O 2 , NO 3 − and MnO 2 (s) reduction, but occurred before SO 4 2− reduction. Monitoring of inlet and outlet chemistry revealed tight coupling between Fe and P (inlet R 2 0.94, outlet R 2 0.85), but distinct dynamics between drainage seasons. In season one, outlet P exceeded inlet P (net P source), and coincided with elevated outlet Fe 2+ , but at ⁓50 % lower P concentrations relative to inlet Fe:P ratios. In season 2 the reactor became a net P sink, coinciding with declining outlet Fe 2+ concentrations (indicating exhaustion of Fe 3+ (s) hydroxides and associated P). In order to characterize P removal under varying source dynamics (i.e. inflows vs in-situ P releases), we used the inlet Fe vs P relationship to estimate P binding to colloidal Fe (hydr)oxide surfaces under oxic conditions, and the outlet Fe 2+ concentration to estimate in-situ P releases associated with Fe (hydr)oxide reduction. Inferred P-removal rates were highest early in season 1 ( k = 0.60 g P m 3 d −1 ; 75–100 % removal), declining significantly thereafter ( k = 0.01 ± 0.02 g P m 3 d −1 ; ca. 3–67 % removal). These calculations suggest that microbiological P removal in DBRs can occur at comparable magnitudes to nitrate removal by denitrification, depending mainly on P availability and hydraulic retention efficiency.
environmental sciences
What problem does this paper attempt to address?