UC Berkeley

Anaerobic ammonium oxidation (anammox) is the basis for an innovative, biological treatment process that removes reactive nitrogen from wastewater. To date, over 100 full-scale anammox treatment processes have been installed at municipal and industrial wastewater treatment plants across the globe. Unfortunately, the bacteria responsible for anammox are easily inhibited and express low growth rates within the anammox treatment processes’ reactors. Often times, it can take up to six months to initiate a new anammox reactor or to restore the performance of an anammox reactor when an inhibition event occurs, which is unacceptably long for municipalities and industries who must adhere to strict nitrogen discharge limits. Moreover, these problems are compounded by a limited understanding of the complex microbial communities that comprise anammox reactors. The work presented in this dissertation seeks to fill this gap by investigating the temporal dynamics of anammox microbial communities during the start-up and continued operation of laboratory-scale anammox reactors, as well as the spatial dynamics of anammox microbial communities across a nitrogen-contaminated environment.

Chapter 2 begins with a review of previous literature that supports the idea of a core microbial community within anammox reactors. This review is combined with temporal-scale data from 440 days of continuous operation of a laboratory-scale anammox reactor to identify relationships between microbial community composition and associated reactor performance. Results suggest that anammox, denitrifying, and dissimilatory nitrate reducing to ammonium (DNRA) bacteria are omnipresent in the anammox reactor. Furthermore, results suggest that these three groups of nitrogen-cycling bacteria cooperate and maximize reactive nitrogen removal under desirable conditions, but that they compete and sabotage reactive nitrogen removal under undesirable conditions (primarily because they share nitrite as their electron acceptor). More research must be done to understand the conditions that support cooperation over competition among these three groups of nitrogen-cycling bacteria in an anammox reactor.

Chapter 3 builds off Chapter 2 and delves more deeply into the relationship between anammox, denitrifying, and DNRA bacteria in a laboratory-scale anammox reactor. The temporal dynamics of these three groups of nitrogen-cycling bacteria and associated reactor performance are investigated by manipulating the ratio of ammonium to nitrite in the anammox reactor’s feedstock. Results indicate that an ammonium to nitrite ratio of 1 to 1.32 in the reactor’s feedstock favors the enrichment of anammox bacteria, while lower ammonium to nitrite ratios (1 to 1.1 – 1 to 1.2) favor the enrichment of a more-diverse bacterial community that contains denitrifying and DNRA bacteria alongside anammox bacteria. Furthermore, results suggest that the more-diverse bacterial community has a greater capacity to remove reactive nitrogen from the feedstock (primarily because denitrifying and DNRA bacteria can transform nitrate, a product of anammox metabolism). Nevertheless, more research must still be done to understand the conditions that support cooperation among these three groups of nitrogen-cycling bacteria in an anammox reactor.

In Chapter 4, the capacity of two support media—polyvinyl alcohol-sodium alginate (PVA-SA) and clinoptilolite zeolite—to improve biomass retention (and hence, decrease startup and recovery times) within anammox reactors is investigated through a series of laboratory-scale anammox reactor experiments. Corresponding shifts in bacterial community structure in the presence of clinoptilolite zeolite are also investigated. Under the conditions provided in this study, results indicate that neither of the support media are capable of improving the performance of the laboratory-scale anammox reactors. Moreover, results indicate that the amendment of clinoptilolite zeolite to the laboratory-scale anammox reactors has no impact on the structure of the bacterial community within it. More research must be done (under different conditions) to definitively rule out the capacity of PVA-SA and clinoptilolite zeolite to improve biomass retention within anammox reactors.

While anammox bacteria have existed in anammox reactors for less than 20 years, they have existed for centuries (quite possibly millennia) in natural habitats. Thus, patterns found in the structure of anammox-containing microbial communities from natural habitats may be able to inform the structure and performance of microbial communities within anammox reactors. To this end, Chapter 5 investigates the abundance and distribution of anammox bacteria across nitrogen-contaminated natural habitats in New Zealand. The results of these investigations indicate that there are many similarities between the structure of anammox microbial communities in natural habitats and of anammox microbial communities in anammox reactors. Moving forward, more research must be done to transfer the patterns found within the structure of anammox-containing microbial communities in New Zealand’s natural habitats into anammox reactor conditions. Once equipped with these results, scientists and engineers can begin to enrich an anammox microbial community based on New Zealand’s unique natural habitats.

Ultimately, the results of the research described in this dissertation bolster the fundamental understanding of anammox microbial communities and their performance within anammox reactors. This understanding, in turn, will enable a more comprehensive control of the anammox treatment process and help facilitate its widespread adoption at municipal and industrial wastewater treatment plants across the globe.