Evaluating four mathematical models for nitrous oxide production by autotrophic ammonia-oxidizing bacteria

Authors

  • Bing-Jie Ni,

    1. Advanced Water Management Centre, The University of Queensland, St. Lucia, Queensland 4072, Australia; telephone: 61-7-3365-4374; fax 61-7-3365-4726
    Search for more papers by this author
  • Zhiguo Yuan,

    Corresponding author
    1. Advanced Water Management Centre, The University of Queensland, St. Lucia, Queensland 4072, Australia; telephone: 61-7-3365-4374; fax 61-7-3365-4726
    • Advanced Water Management Centre, The University of Queensland, St. Lucia, Queensland 4072, Australia; telephone: 61-7-3365-4374; fax 61-7-3365-4726
    Search for more papers by this author
  • Kartik Chandran,

    1. Department of Earth and Environmental Engineering, Columbia University, New York, New York 10027
    Search for more papers by this author
  • Peter A. Vanrolleghem,

    1. modelEAU, Département de génie civil et de génie des eaux, Université Laval, 1065, Avenue de la Médecine, Québec, Canada QC G1V 0A6
    Search for more papers by this author
  • Sudhir Murthy

    1. DC Water and Sewer Authority, 5000 Overlook Ave., SW, Washington, District of Columbia 20032
    Search for more papers by this author

Abstract

There is increasing evidence showing that ammonia-oxidizing bacteria (AOB) are major contributors to N2O emissions from wastewater treatment plants (WWTPs). Although the fundamental metabolic pathways for N2O production by AOB are now coming to light, the mechanisms responsible for N2O production by AOB in WWTP are not fully understood. Mathematical modeling provides a means for testing hypotheses related to mechanisms and triggers for N2O emissions in WWTP, and can then also become a tool to support the development of mitigation strategies. This study examined the ability of four mathematical model structures to describe two distinct mechanisms of N2O production by AOB. The production mechanisms evaluated are (1) N2O as the final product of nitrifier denitrification with NOmath image as the terminal electron acceptor and (2) N2O as a byproduct of incomplete oxidation of hydroxylamine (NH2OH) to NOmath image. The four models were compared based on their ability to predict N2O dynamics observed in three mixed culture studies. Short-term batch experimental data were employed to examine model assumptions related to the effects of (1) NHmath image concentration variations, (2) dissolved oxygen (DO) variations, (3) NOmath image accumulations and (4) NH2OH as an externally provided substrate. The modeling results demonstrate that all these models can generally describe the NHmath image, NOmath image, and NOmath image data. However, none of these models were able to reproduce all measured N2O data. The results suggest that both the denitrification and NH2OH pathways may be involved in N2O production and could be kinetically linked by a competition for intracellular reducing equivalents. A unified model capturing both mechanisms and their potential interactions needs to be developed with consideration of physiological complexity. Biotechnol. Bioeng. 2013; 110: 153–163. © 2012 Wiley Periodicals, Inc.

Ancillary