Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Impact of microscreen pretreatment and biofilm photobioreactor design on efficiency of decentralized… Roberts, James 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_september_roberts_james.pdf [ 6.5MB ]
Metadata
JSON: 24-1.0380207.json
JSON-LD: 24-1.0380207-ld.json
RDF/XML (Pretty): 24-1.0380207-rdf.xml
RDF/JSON: 24-1.0380207-rdf.json
Turtle: 24-1.0380207-turtle.txt
N-Triples: 24-1.0380207-rdf-ntriples.txt
Original Record: 24-1.0380207-source.json
Full Text
24-1.0380207-fulltext.txt
Citation
24-1.0380207.ris

Full Text

  IMPACT OF MICROSCREEN PRETREATMENT AND BIOFILM PHOTOBIOREACTOR DESIGN ON EFFICIENCY OF DECENTRALIZED WASTEWATER TREATMENT by  James Roberts  B.A.Sc., The University of British Columbia, 2016  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Civil Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   July 2019  © James Roberts, 2019   ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis entitled:  Impact of Microscreen Pretreatment and Biofilm Photobioreactor Design on Efficiency of Decentralized Wastewater Treatment  submitted by James Roberts in partial fulfillment of the requirements for the degree of Master of Applied Science in Civil Engineering  Examining Committee: Ryan Ziels  Supervisor Pierre Bérubé Additional Examiner    iii Abstract  Biofilm photobioreactors rely on cooperation between algae and bacteria within a single biofilm to treat wastewater. Algae growth produces oxygen, which can subsequently be utilized by aerobic bacteria to degrade organic matter and produce carbon dioxide, which is then utilized as a carbon source by algae. Due to their relatively low maintenance and energy inputs, biofilm photobioreactors could be amenable for decentralized wastewater treatment. However, the impact of biofilm photobioreactor design on treatment efficiency has received little attention. Here, it was hypothesized that open (i.e. unsealed) versus closed (i.e. sealed) photobioreactors could promote different nitrogen removal pathways by altering redox conditions throughout a diel cycle. This study explored the effect of open versus closed photobioreactor configurations on nitrogen removal and the microbial community structure in two parallel photobioreactors treating microscreened decentralized wastewater. The reactors were intermittently lit in a 16hr-8hr light-dark cycle, and operated at an HRT of 2 days and SRT of 9 days. The influent feed regime and alkalinity addition were varied over three successive 30-day experimental phases.   Microscreening was an effective primary treatment step, removing 70 ± 6% (95% c.i.) of suspended solids and 39 ± 9% of COD. The photobioreactors removed 90 ± 6 % and 83 ± 3% of the remaining suspended solids and COD respectively, independent of operating conditions. Alternating oxic and anoxic conditions were observed in both reactors during the lit and unlit periods, respectively, resulting nitrification and denitrification. Optimal nitrogen removal conditions were observed under a sequencing batch feed regime with alkalinity addition. Under these conditions, TKN removal was significantly higher in the open reactor at 93 ± 5% compared to 78 ± 6% in the closed reactor due to higher rates of nitrification and N assimilation. TN removal   iv was similar at 77 ± 9% and 76 ± 8% in the open and closed systems, respectively. The dominant bacterial genus in the reactors was Tychonema, a cyanobacteria which comprised up to 87% of 16S rRNA gene amplicon reads. Overall, this study demonstrated that nitrogen removal pathways differ significantly in open and closed photobioreactors when operated at the same COD loading rate.    v Lay Summary  With climate variability and urban population growth threatening the capacity of fresh water supplies around the world, interest has grown for decentralized wastewater treatment to generate clean water that can be reused for landscape irrigation and other non-potable uses. However, treatment technologies that are implemented for centralized wastewater treatment are not easily scaled down, as per capita energy and maintenance demands dramatically increase. Thus, alternative technologies must be considered for small-scale decentralized wastewater treatment.  One such technology is the biofilm photobioreactor. Although there have been a number of studies exploring their use for wastewater treatment, the impact of reactor design on treatment efficiency has received little attention. Here, a bench-scale study was conducted that compared the treatment efficacy of open (unsealed) and closed (sealed) photobioreactors.  The results from the study can be used to provide evidence of feasibility for urban wastewater reuse and inform future research on sustainable decentralized treatment.    vi Preface This dissertation is original, unpublished, independent work by the author, James Roberts.   vii Table of Contents  Abstract ....................................................................................................................................iii Lay Summary ............................................................................................................................ v Preface ...................................................................................................................................... vi Table of Contents .................................................................................................................... vii List of Tables ............................................................................................................................. x List of Figures .......................................................................................................................... xi List of Equations ..................................................................................................................... xx Nomenclature ......................................................................................................................... xxi Acknowledgements ............................................................................................................... xxv Chapter 1: Introduction............................................................................................................ 1 Chapter 2: Literature Review................................................................................................... 3 2.1 Decentralized Wastewater Treatment and Reuse .......................................................... 3 2.2 Primary Treatment: Microscreening ............................................................................. 6 2.3 Biological Treatment: Biofilm Photobioreactor .......................................................... 10 Chapter 3: Thesis Objectives .................................................................................................. 15 Chapter 4: Materials and Methods ........................................................................................ 17 4.1 Raw Wastewater Collection ....................................................................................... 17 4.2 Raw and Microscreened Wastewater Characterization ............................................... 17 4.3 Biofilm Photobioreactors ........................................................................................... 18 4.3.1 Set-up ................................................................................................................ 18 4.3.2 Irradiance Calculation ........................................................................................ 20   viii 4.3.3 Inoculation ......................................................................................................... 21 4.3.4 Biomass Wasting ............................................................................................... 22 4.3.5 Influent Loading Rate ........................................................................................ 23 4.3.6 Phases of Operation ........................................................................................... 23 4.3.6.1 Influent Feed Regimes ................................................................................... 24 4.3.7 Diel Cycle Monitoring Experiments ................................................................... 25 4.3.8 Characterization of the microbial community with 16S rRNA gene amplicon sequencing ........................................................................................................................ 26 4.4 Sampling and Analytical Methods ............................................................................. 27 4.5 Calculations ............................................................................................................... 31 4.6 Significance Testing .................................................................................................. 34 Chapter 5: Results................................................................................................................... 36 5.1 Raw Wastewater Quality and Microscreen Performance ............................................ 36 5.2 Biofilm PBR Performance ......................................................................................... 37 5.2.1 Secondary Treatment Performance ..................................................................... 39 5.2.2 Nitrogen Removal Performance ......................................................................... 40 5.2.2.1 Phase A: Continuous Feed, No Alkalinity Addition ........................................ 41 5.2.2.2 Phase B: Continuous Feed, Alkalinity Addition.............................................. 47 5.2.2.3 Phase C: Batch Feed, Alkalinity Addition ...................................................... 52 5.2.2.4 Continuous and Batch Feed Regime Comparison ........................................... 56 5.2.3 Biomass Production ........................................................................................... 57 5.2.4 Microbial Community Composition of Microalgal Biofilm ................................ 58 Chapter 6: Discussion ............................................................................................................. 65   ix 6.1 Microscreen Pretreatment .......................................................................................... 65 6.2 Biofilm PBR Biological Treatment ............................................................................ 65 6.2.1 Comparative Performance of Open and Closed PBRs ........................................ 65 6.2.2 Other Factors Influencing PBR Performance ...................................................... 67 6.2.3 Microbial Dynamics........................................................................................... 74 Chapter 7: Conclusion ............................................................................................................ 78 7.1 Recommendations ..................................................................................................... 79 References ............................................................................................................................... 80 Appendix A: Microscreen Performance ................................................................................ 88 Appendix B: PBR Performance.............................................................................................. 92 Appendix C: PBR Biomass Characteristics ......................................................................... 110 Appendix D: 16s rRNA Gene Abundance Analysis Results ................................................ 114 Appendix E: Continuous Diel Monitoring Experiment Results .......................................... 120    x List of Tables Table 2.1 - Summary of North American treatment guidelines for wastewater reuse. .................. 5 Table 4.1 - PBR design parameters based on mass balances of carbon dioxide and oxygen ....... 21 Table 4.2 - HRT, effective HRT, and surface flow velocity used in existing literature ............... 23 Table 4.3 - Experimental conditions for the startup phase and three experimental phases observed in this study............................................................................................................................... 24 Table 4.4 – Influent (Inf), effluent (Eff), and solids (Sol) sampling and analysis regime in the open and closed PBRs over 27 days (3 SRT) of an experimental phase. .................................... 27 Table 4.5 - Summary of analysis techniques chosen for each measured analyte. ........................ 28 Table 4.6 – Summary of the statistical tests performed in this study. ......................................... 35 Table 5.1 - Influent and microscreened wastewater quality and consequent removal efficiency. 37 Table 5.2 - Influent and effluent wastewater quality for the PBRs through each experimental phase. ........................................................................................................................................ 38 Table 5.3 - PBR removal efficiency (%) of key parameters through the three experimental phases conducted in this study. ............................................................................................................. 39 Table 5.4 – Differences between the COD removal rate (g/m2/day) in the open and closed PBRs ................................................................................................................................................. 40 Table 5.5 – Summary of differences in reactor conditions and process rates between the open and closed PBRs. ............................................................................................................................. 41 Table 5.6 - Average characteristics of biomass harvested from the open and closed PBRs through three experimental phases. ........................................................................................................ 58 Table 6.1 - Comparison between experimental conditions and results observed in existing literature and this study. (O/C) Open/Closed. ............................................................................ 70   xi List of Figures Figure 2.1 – Example of drum filter microscreen design (Poseidon Resources, 2019).................. 6 Figure 2.2 – Example of rotary belt microscreen design (Salsnes Filter AS., 2017) ..................... 7 Figure 2.3 – Particle separation efficiency as a function of size (Ljunggren, 2006) ...................... 7 Figure 2.4 – Fractions of solids, COD, TN, and TP associated with particulate larger than 1000 µm, 100 µm, 10 µm, and 1 µm in raw blackwater (n = 16) (Todt et al., 2015). ............................ 9 Figure 2.5 - Symbiosis between microalgae and bacteria treating domestic wastewater (Boelee et al., 2014) ................................................................................................................................... 11 Figure 4.1 - Schematic diagram of the laboratory scale biofilm PBRs ....................................... 18 Figure 4.2 - Harvesting method used for collecting biomass from the two PBRs` ...................... 22 Figure 4.3 – Illustration of the frequency and volume of discharge/influent events under the continuous (hollow line) and batch (solid line) feed regimes. The discharge events sampled for daily analysis are denoted by an x. ............................................................................................ 25 Figure 5.1 - Effluent DO, pH, and alkalinity levels in the open and closed PBRs through the continuous diel monitoring experiment conducted at the end of Phase A. The dark period is denoted by the shaded area. ....................................................................................................... 43 Figure 5.2 - Concentrations of key parameters during the diel monitoring study at the end of Phase A with the PBRs operating under the “batch” feed condition. .......................................... 44 Figure 5.3 - Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase A. The dark period is denoted by the shaded area. ... 45 Figure 5.4 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase A at hour 17. .................................................................................................................................... 46   xii Figure 5.5 - Effluent DO, pH, and alkalinity levels through the “continuous” diel monitoring experiment conducted at the end of Phase B. The dark period is denoted by the shaded area. .... 48 Figure 5.6 - Concentrations of key parameters during the diel monitoring study at the end of Phase B with the PBRs operating under the “batch” feed condition. .......................................... 49 Figure 5.7 - Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase B. The dark period is denoted by the shaded area. ... 50 Figure 5.8 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase B at hour 17. .................................................................................................................................... 51 Figure 5.9 - Effluent DO, pH, and alkalinity levels through the “batch” diel monitoring experiment conducted at the end of Phase C. The dark period is denoted by the shaded area. .... 53 Figure 5.10 - Concentrations of key parameters during the diel monitoring study at the end of Phase C with the PBRs operating under the “batch” feed condition. .......................................... 54 Figure 5.11 – Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase C. The dark period is denoted by the shaded area. ... 55 Figure 5.12 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase C at hour 24. .................................................................................................................................... 56 Figure 5.13 - Relative abundance (%) of the 12 most abundant genera observed in triplicate at the end of the startup and three experimental phases, based on the results of the 16S rRNA gene amplicon sequencing. ................................................................................................................ 59 Figure 5.14 - Shannon Index of the relative abundance of genera present in the biomass samples taken at the end of the startup phase and each of the three experimental phases in the open (right bars) and closed (left bars) PBRs. .............................................................................................. 61   xiii Figure 5.15 - Principle coordinate analysis of the relative abundance of genera present at the end of the startup phase and each of the three experimental phases. X-axis (30 % explained variance), y-axis (12 % explained variance)............................................................................................... 62 Figure 5.16 - STAMP analysis comparing relative abundance of genera in the open and closed PBRs over the course of the entire 93-day experiment. Bars to the left and right of center indicate significantly greater abundance in the open and closed PBRs, respectively. The values on the far right indicate the p value for comparing the relative abundance of the genera between the two conditions. ................................................................................................................................ 63 Figure 6.1 – TKN (a) and TN (b) removal efficiency graphed as a function of the oxygen loading ratio for this and other major studies investigating wastewater treatment with open (•) and closed (■) PBRs. The first two weeks and second two weeks of Phase B were plotted separately (B1 and B2) due to elevated COD influent concentrations in the second half of the phase (Figure B.3). The results from Phase A were excluded from the graph because TKN removal was alkalinity limited (Figure B.1). ................................................................................................................. 73 Figure A.1 - COD removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ................................................................................................................................ 88 Figure A.2 - SS removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ................................................................................................................................ 89 Figure A.3 - TKN removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ................................................................................................................................ 89 Figure A.4 - TN removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ................................................................................................................................ 90   xiv Figure A.5 - TP removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ................................................................................................................................ 90 Figure A.6 - Turbidity removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ....................................................................................................................... 91 Figure B.1 - Alkalinity concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ........................................................................................................................... 92 Figure B.2 - BOD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 93 Figure B.3 - COD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 93 Figure B.4 - Filtered COD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ........................................................................................................................... 94 Figure B.5 - NH4-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 94 Figure B.6 - NO2-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 95   xv Figure B.7 – NO3-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 95 Figure B.8 – PO4-P concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 96 Figure B.9 - Suspended solids concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means......................................................................................................... 96 Figure B.10 - TKN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 97 Figure B.11 - Filtered TKN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means......................................................................................................... 97 Figure B.12 - TN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 98 Figure B.13 - TP concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 98   xvi Figure B.14 - TS concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means........................................................................................................................................ 99 Figure B.15 - Turbidity (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. 99 Figure B.16 - Transmittance at 254nm (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ......................................................................................................................... 100 Figure B.17 - DO concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means. ........ 100 Figure B.18 - PH (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means. ................................... 101 Figure B.19 - Temperature (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means. ................. 101 Figure B.20 - The COD removal rate (g/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 102 Figure B.21 - The denitrification rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 103   xvii Figure B.22 - The nitrification rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 104 Figure B.23 - The assimilation rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 105 Figure B.24 - The TKN removal rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 106 Figure B.25 - The TN removal rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 107 Figure B.26 - The TP removal rate (gP/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ........................................................................ 108 Figure B.27 - The specific alkalinity demand (gCaCO3/gN) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases B (left) and C (right). If the confidence   xviii interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. ...................................................................................... 109 Figure C.1 - Cartenoid content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ................................................ 110 Figure C.2 – Total chlorophyll content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ......................................... 111 Figure C.3 - Chlorophyll A content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ......................................... 111 Figure C.4 - Chlorophyll B content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ......................................... 112 Figure C.5 - Dry weight (g/m2/day) of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means. ..................................... 112 Figure C.6 - Dry to wet weight ratio of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means. ..................................... 113 Figure C.7 - Photosynthetic efficiency (%) of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means. .......................... 113 Figure D.1 - Relative abundance (%) of the 12 most common phyla observed in triplicate at the end of the startup and three experimental phases. .................................................................... 114 Figure D.2 - Principle coordinate analysis of the relative abundance of microbes at the end of the startup phase and each of the three experimental phases. Top-left graph shows the 1st (35% explained variance) and 2nd (12% explained variance) principle coordinates, top-right shows 1st and 3rd (10% explained variance), and bottom-left shows 1st and 4th (9% explained variance). 115   xix Figure D.3 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of the startup phase. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively. ............................................. 116 Figure D.4 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of the Phase A. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively. ............................................. 117 Figure D.5 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of Phase B. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively. ......................................................... 118 Figure D.6 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of Phase C. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively. ......................................................... 119 Figure E.1 - Concentrations of key parameters during the diel monitoring study at the end of Phase A with the reactors operating under the “continuous” feed condition. ............................ 120 Figure E.2 - Concentrations of key parameters during the diel monitoring study at the end of Phase B with the reactors operating under the “continuous” feed condition. ............................ 121   xx  List of Equations Equation 4.1 - Algal Growth Stoichiometric Equation ............................................................... 21 Equation 4.2 - Bacterial Growth Stoichiometric Equation ......................................................... 21 Equation 4.3 – NO3-N concentration (mg/L) ............................................................................. 31 Equation 4.4 - Organic nitrogen concentration (mg/L)............................................................... 31 Equation 4.5 - Total nitrogen concentration (mg/L) ................................................................... 31 Equation 4.6 - Removal efficiency (%) of a given parameter C ................................................. 31 Equation 4.7 - Production/consumption rate .............................................................................. 32 Equation 4.8 - Rate of assimilation of nitrogen into biomass ..................................................... 32 Equation 4.9 – PBR nitrification rate ......................................................................................... 32 Equation 4.10 – PBR denitrification rate ................................................................................... 33 Equation 4.11 - Specific alkalinity demand ............................................................................... 33 Equation 4.12 – Influent baseline alkalinity ............................................................................... 34 Equation 4.13 – Instantaneous production/consumption rate ..................................................... 34 Equation 6.1 – Oxygen loading ratio ......................................................................................... 71 Equation 6.2 – Oxygen production rate by photosynthesis ......................................................... 71     xxi Nomenclature %   Percent $   Dollars °C    Degrees Celsius  Alk   Alkalinity ANSI   American National Standards Institute AOA   Ammonia Oxidizing Archaea AOB   Ammonia Oxidizing Bacteria BOD   5 Day Biological Oxygen Demand C   Carbon CaCO3   Calcium Carbonate CBOD   Carbonaceous Biochemical Oxygen Demand CO2   Carbon Dioxide CFU   Colony Forming Unit COD   Chemical Oxygen Demand cm   Centimetre CT   Chlorine contact time multiplied by concentration DMF   Direct Membrane Filtration DNA   Deoxyribonucleic Acid DO   Dissolved Oxygen EGBC    Engineers of Geoscientists of British Columbia EPA   Environmental Protection Agency g   Gram   xxii GPD   Gallons Per Day H17   The 17th hour of the diel cycle H24   The 24th hour of the diel cycle hr   Hour HRT   Hydraulic Residence Time kg   Kilogram kWh   Kilowatt Hours L   Litre m   Metres M&E   Metcalf and Eddy MBR   Membrane Bioreactor mg   Milligram min   Minute mL   Millilitre mm   Millimetre MSW   Medium Strength Wastewater n   Number of samples N   Nitrogen N2   Nitrogen gas Nbio   Nitrogen content of reactor biofilm Neff   Nitrogen content of reactor effluent Ninf   Nitrogen content of reactor influent NH4   Ammonium-N   xxiii NO2   Nitrite-N NO3   Nitrate-N NOx   Sum of Nitrate-N and Nitrite-N NPS   Nominal Pore Size NS   Not Detected NSF   National Science Foundation NTU   Nephelometric Units O2   Oxygen OLR   Oxygen Loading Ratio PAR   Photoactive Radiation PBR   Photobioreactor PCA   Principle Component Analysis PE   Photosynthetic Efficiency pH   Power of Hydrogen PO4   Phosphate-P PSBR   Photosynthetic-Bioreactor PVC   Polyvinyl Chloride Redox   Oxidation-Reduction Reactions rpm   Revolutions per Minute rRNA   Ribosomal Ribonucleic Acid s   Second SI   Shannon Index SRT   Solids Retention Time   xxiv TC   Total Carbon TKN   Total Kjeldahl Nitrogen TOC   Total Organic Compound TP   Total Phosphorous TSS   Total Suspended Solids Turb   Turbidity UBC   University of British Columbia µm   Micrometre µmol   Micromol UV   Ultraviolet W   Watt WWTP  Wastewater Treatment Plant     xxv Acknowledgements I am greatly appreciative to the following people for their assistance,  For his supervision throughout my research project and thesis writing: Dr. Ryan Ziels, The University of British Columbia  For his time spent in revision of my thesis, Dr. Pierre Bérubé, The University of British Columbia  For their consistent help with laboratory analysis and sample collection: Otman Abida, The University of British Columbia Bryan Alexander, The University of British Columbia  For their assistance with the grants and parallel research that made this project possible: Sherry Zhao, Mitacs Canada Kimron Rink, ECO-TEK Angelique Pilon and Diana Lopez, UBC Sustainability Initiative  For their assistance with an endless stream of shop requests: Harald Schrempp, Doug Hudniuk, and Bill Leung, The University of British Columbia  For their assistance and companionship in the laboratory: Sylvia Woolley, Marcia Fromberg, Sarah Ries, Abhishek Dutta  And for listening to more wastewater ramblings than anyone should ever have to, my family and friends, they know who they are and they will spend a lifetime trying to forget. Chapter 1: Introduction Today, water management infrastructure around the world is more extensive and industrialized than in any other time in human history. This constitutes a triumph for urban sanitation as water has been used to efficiently collect predominantly soluble wastes and dispose of them outside of populated areas. However, this systematic centralization and disposal of urban by-products has led to a series of consequences that threaten the sustainability of the system as a whole. For one, urban water demand is increasing due to population growth, while the predictability of upstream water supplies is decreasing with climate change (Feo et al., 2014). This results in over-exploited ground and surface water supplies and consequent drying rivers and sinking cities (Ramakrishna and Babu, 1999). Meanwhile, the direction of nutrients and organics in domestic sewage, both treated and untreated, into natural water bodies not only can result in eutrophication and dead zones (Rutledge et al., 2011), but also wastes nutrients that could otherwise be recycled (Daigger, 2009). Finally, water infrastructure itself is under threat, as many countries do not have adequate capital to invest in new or aging infrastructure (American Society of Civil Engineers, 2013).  As a result of these macro drivers, it is largely accepted that there will be a significant trend towards decentralized wastewater treatment and reuse in cities around the world (Metcalf and Eddy et al., 2013). The difficulty lies in the identification of treatment technologies that produce a treated wastewater quality suitable for water reuse while also maintaining low operation and maintenance costs without the benefit of economies of scale. This study explored two emerging technologies, microscreens and biofilm photobioreactors (PBRs), that together have been the potential to meet this demand, but have not been extensively investigated for decentralized   2 wastewater treatment to date.  Microscreens are gravity fed filters, with openings ranging from 50 to 500 microns, that have been used in centralized primary treatment to remove suspended solids with minimal energy and land requirements (Lema and Martinez, 2017). Microscreens have yet to be explored as a part of a decentralized treatment system. Biofilm PBRs on the other hand, are a secondary/tertiary treatment technology that use a single photosynthesizing biofilm to perform aeration, heterotrophic degradation, and biological nutrient removal functions all at once, reducing mechanical complexity and energy requirements. Although biofilm PBRs have been explored for decentralized wastewater treatment in a number of different studies to date (Boelee et al., 2011, 2014; Posadas et al., 2013; Zamalloa et al., 2013), information is lacking regarding the impact of PBR design and operating conditions on treatment efficiency.      3 Chapter 2: Literature Review 2.1 Decentralized Wastewater Treatment and Reuse Decentralized wastewater treatment and reuse is considered a solution to the current crisis of urban water management for a number of reasons: • Wastewater availability grows reliably with urban water consumption, ensuring it’s availability as a water source (USEPA, 2012a). • Treating wastewater closer to where people live more easily enables water and nutrient reuse in landscape irrigation and other non-potable uses while reducing disposal to bodies of water (Daigger, 2009). • Decentralized systems can be added incrementally as needed, quickly responding to demand, without having to predict and invest in future capacity (Capodaglio, 2017). • Spreading treatment systems out allows for the use of less compact treatment processes, potential reducing energy demand (Muga and Mihelcic, 2008).  However, there are also concerns regarding the operations and maintenance of decentralized wastewater treatment plants (WWTP’s). Traditionally, operation costs for multiple decentralized WWTP’s have been found to be higher than for a single centralized WWTP (Daigger, 2009). This is of concern because the operation of water and wastewater collection and treatment systems in developed countries is thought to be responsible for 3-5% of national electricity consumption, with 0.5kWh/m3 being used for water supply and treatment, and another 0.5kWh/m3 being used for wastewater collection and treatment (Kalogo et al., 2008). Although   4 the reuse of wastewater reduces the demand for fresh water and its associated electricity costs, the energy consumed for decentralized wastewater treatment should nonetheless be minimized.  Meanwhile, maintaining decentralized facilities to reliably produce the required effluent quality may be more difficult than for centralized facilities (Daigger, 2009). Wastewater treated for reuse in urban areas must meet strict guidelines in one of the following two categories (Table 2.1):  1. Unrestricted Urban/Agricultural Reuse: The use of reclaimed water for non-potable applications in municipal settings where public access is not restricted or to irrigate food crops intended for human consumption. These are two separate categories that often share same criteria. 2. Restricted Urban Reuse: The use of reclaimed water for non-potable applications in municipal settings where public access is controlled or restricted by physical or institutional barriers such as fencing, advisory signage, or temporal access restriction.   These concerns must be mitigated through the development of low energy and low maintenance wastewater treatment technologies capable of providing high quality effluent within an urban environment (Daigger, 2009).    5 Table 2.1 - Summary of North American treatment guidelines for wastewater reuse. Guideline Reference Uses pH CFU/ 100mL CBOD (mg/L) TSS (mg/L) Turb (NTU) UV Dose (mj/cm2) UV Trans (%) Cl Residual (mg/L) Med N (mg/L) Processes APEGBC Onsite Sewerage Guideline (Larocque et al., 2018) Disposal  <400       NO3<30 NH3<3.5  Secondary Filtration Disinfection  NSF/ANSI 40 (NSF, n.d.) Disposal     25  30       ISO16075-2 Very High Quality Treated Wastewater (ISO, 2015) Unrestricted Urban/ Agricultural Irrigation of Food Crops Eaten Raw   <10  5  5  2      Secondary, contact filtration or membrane filtration and disinfection ISO16075-2 High Quality Treated Wastewater (ISO, 2015) Restricted urban irrigation and agricultural irrigation of processed food crops  <200 10 10      Secondary, filtration and disinfection BC Reclaimed Water Guideline (Ministry of Agriculture, n.d.) Agriculture/ Urban Unrestricted  6-9 <1 <10  <2   0.5    BC Reclaimed Water Guideline (BC Ministry of Environment, 2013) Urban Restricted  6-9 <200 <45  <45     0.5    EPA  (USEPA, 2012b) Agriculture/ Urban Unrestricted  6-9  0 <10  <2   1   EPA (USEPA, 2012b) Urban Restricted 6-9 <200 <30 <30 <2   1   Range of State Guidelines (USEPA, 2012c) Agriculture/ Urban Unrestricted   2.2-20 5-60 5-60 0.5-3 80-100 >55 1-5 N<10 NH3<1-4  Range of State Guidelines (USEPA, 2012c) Urban Restricted  2.2-200 10-NS 5-NS 10-NS NS-100  NS-5 N<10 NH3<1-4 Secondary Disinfection   6 2.2 Primary Treatment: Microscreening Interest in microscreens as a primary treatment technology for municipal wastewater treatment has grown steadily over the past few decades as demand for compact, low cost treatment processes has grown. The main operating principle is simple; water flows by gravity through the filter, removing suspended particulate until the filter must be backwashed to remove and collect captured solids (Ljunggren, 2006). Microscreens are most commonly found in rotating drum (Figure 2.1) and rotary belt configurations (Figure 2.2).    Figure 2.1 – Example of drum filter microscreen design (Poseidon Resources, 2019)   7  Figure 2.2 – Example of rotary belt microscreen design (Salsnes Filter AS., 2017)  The filters are often operated such that the captured solids are allowed to form a cake on the surface of the filter capable of removing solids finer than the filter pore size itself (Ljunggren, 2006). Conversely, shear forces result in the breakup of some solids that are larger than the pore size, eventually allowing those particles to pass through the screen (Figure 2.3).    Figure 2.3 – Particle separation efficiency as a function of size (Ljunggren, 2006)   8 It has been shown that primary settling tanks in centralized treatment facilities can be replaced by microscreens using 1/10th the land (Franchi and Santoro, 2015). In centralized treatment, suspended solid removal efficiencies of 20-35% are possible with filter pore sizes of 200 µm (Särner, 1976)(Särner, 1978). Removal efficiency rose to 50% when a microscreen with an NPS of 20 µm was used (Eriksson, 1974). More recently, primary filtration of purely domestic wastewater showed SS and COD removal efficiencies of 50% and 30%, respectively using a 60 µm filter (Petterson, 2004).  Although the removal efficiencies demonstrated by microscreens at centralized treatment facilities are quite high, it is expected that removals would be even higher if microscreens were implemented in decentralized facilities. As wastewater passes through the sewer, suspended solids are subject to hydraulic shear and microbial hydrolysis, reducing particle sizes and solubilizing biodegradable contaminants (Levine et al., 1985; Roni et al., 2019). As a consequence, centralized treatment facilities may receive a higher concentration of smaller suspended solids and dissolved organics in comparison to decentralized treatment systems with shorter collection systems. This is important because the size distribution of particulate in wastewater has been shown to be a fundamental parameter in determining the performance efficiency of microscreens (Shea and Males, 1971).  Todt et al., (2015) characterized the concentration and size distribution of solids, COD, TN, and TP in separated blackwater and greywater streams from a dormitory before they entered the sewer system (Figure 2.4). This data gives an indication of removal efficiencies that could be achieved by microscreens with pore sizes within the size distribution explored.    9  Figure 2.4 – Fractions of solids, COD, TN, and TP associated with particulate larger than 1000 µm, 100 µm, 10 µm, and 1 µm in raw blackwater (n = 16) (Todt et al., 2015).  Ahn and Song, (2001) treated raw domestic wastewater from a resort complex using a submerged 0.1 µm ceramic membrane in a process referred to as direct membrane filtration (DMF). Although effluent equality was sufficient to meet the Korean Reuse Guideline, the system experienced a high fouling rate as pores clogged with dissolved organic material (Ahn et al., 2001). Chemically enhanced backflushing has been explored for the elimination of permanent fouling in DMF, but has not been demonstrated to be successful at sufficiently long operation times (Lateef et al., 2013). Although DMF has been demonstrated to achieve reuse quality effluent, fouling rates and consequent maintenance and monitoring requirements may limit the technology’s applicability in decentralized wastewater treatment. Conversely, microscreens in the range of 10-100 µm still have the capacity to remove significant concentrations of influent organic material but without pores small enough to clog with dissolved   10 organics. However, they have yet to be investigated for primary treatment of raw decentralized domestic wastewater.   2.3 Biological Treatment: Biofilm Photobioreactor A biological treatment technology that is emerging for decentralized wastewater treatment is the biofilm photobioreactor (PBR). Biofilm PBRs rely on the action of algae and bacteria, within a single biofilm, to remove organics and nutrients from wastewater that is recirculated across it. Algae growth produces oxygen under sunlit conditions, which can subsequently be utilized by aerobic bacteria to degrade organic matter and produce carbon dioxide, which is then utilized as a carbon source by algae (Figure 2.5) (Boelee et al., 2014). Nutrient removal is achieved through assimilation into algal and bacterial biomass but can also be supplemented by nitrification and denitrification under the appropriate redox conditions. Nitrification is the oxidation of ammonium to nitrate under oxic conditions, whereas denitrification is the reduction of nitrate to nitrogen gas under anoxic conditions.    11  Figure 2.5 - Symbiosis between microalgae and bacteria treating domestic wastewater (Boelee et al., 2014)  The only energy input for biofilm PBRs is for pumped recirculation, cycling water across the illuminated parallel plate on which the biofilm grows (Boelee et al., 2011). However, if a natural light source is not used for the biofilm PBRs, this would also add to energy requirements. Biofilm systems are known to be inherently robust, insensitive to fluctuations in influent concentrations and loading rate (Westerling, 2014). Due to their relatively low maintenance and energy inputs, biofilm PBRs could be amenable for decentralized wastewater treatment (Zamalloa et al., 2013).   Although biofilm PBRs have been explored for wastewater treatment in a limited number of different studies to date, information is lacking regarding the impact of PBR design and operating conditions on treatment efficiency. Using municipal wastewater, Zamalloa et al.,   12 (2013) showed that an open (i.e. unsealed) rooftop biofilm PBR exposed to sunlight on a 16hr-8hr diel cycle was an effective means of small-scale tertiary treatment to reach reuse-quality effluent. Posadas et al., (2013) examined the possibility of achieving both secondary and tertiary treatment within an open PBR. In this case, the comparative effectiveness of a 16hr-8hr light-dark cycled PBR and an unlit PBR was tested under varying HRT’s and recycle rates. The lit PBR exhibited significantly higher nitrogen and phosphorus removal rates under all conditions whereas a shift from a 10 to 3 day HRT significantly decreased phosphorus removal efficiency, and a shift from a 4.2 to 9 L/m2/min recycle rate did not significantly change treatment efficiency (Posadas et al., 2013). Boelee et al., (2014) studied the behaviour of a closed (i.e. sealed) biofilm PBR operated with continuous lighting and synthetic wastewater. It was determined that the PBR required alkalinity addition to build a symbiotic algal-bacterial biofilm but not to maintain it. However, the photoactive irradiance was only sufficient to support heterotrophic oxidation, inhibiting nitrification and its associated alkalinity demand. When nitrate was added to the influent, the PBR was shown to have a significant denitrifying capacity. Foladori et al., (2018) explored the impact of feed and mixing patterns on the performance of a continuous stirred microalgal-bacterial reactor operated under a 16hr-8hr light-dark diel cycle. It was observed that total nitrogen removal efficiency increased significantly when influent was fed to the reactor at the onset of the 8 hr dark period instead of the onset of the lit period, but only if mechanical mixing was also turned off during the dark period. Overall, there is very little consistency in operating conditions used in the existing literature. However, it can be summarized that three primary PBR design factors have been varied: the light-dark cycle, influent feed regime, and the open-closed condition, while operating parameters such as HRT, recycle rate, and alkalinity addition have been tuned.    13  The light-dark cycle, influent feed regime, and the open-closed condition are of particular interest because they directly impact ammonia and nitrate removal by manipulating the redox conditions within a PBR to affect nitrification and denitrification processes. Light intensity and diel cycle can impact oxygen production via photosynthesis, determining how dissolved oxygen (DO) concentrations vary over a 24 hr period. Oxic and anoxic conditions within the biofilm of a PBR can thereby be either produced continuously or cycled with a diel lighting pattern. Similarly, varying the open-closed condition of the PBR determines whether atmospheric aeration is present to supplement aeration via photosynthesis. Finally, the influent feed regime can affect redox conditions of the PBR through batch vs. continuous loading of COD. For instance, batch feeding promotes high initial COD concentrations that could result in transient anoxia due to excess respiration of oxygen by heterotrophic bacteria. With continuous feeding, the initial COD would be lower, and oxic conditions could persist if sufficient aeration from algal photosynthesis is supplied. Although it is also important to note that different redox conditions may also occur within the biofilm itself, due to limited oxygen diffusion flux into deep biofilms (Zamalloa et al., 2013).   In addition to redox conditions, nitrogen removal pathways are significantly impacted by members of the microbial community contained within a PBR biofilm. Nitrification in particular can only be carried out by select bacterial groups, referred to as ammonia-oxidizing bacteria (AOB) and nitrite-oxidizing bacteria (NOB) (Seviour and Nielsen, 2010). Additionally, certain lineages of archaea can oxidize ammonium, and are referred to as Ammonium Oxidizing Archaea (AOA) (Könneke et al., 2005). Krohn-Molt et al., (2013) carried out the first   14 metagenomic study of biofilm PBRs in general, exploring community composition and dynamics in biofilms PBRs used for biofuel production. The results suggested that the metabolic and catabolic potential of the microbes living within the biofilm was highly diverse (Krohn-Molt et al., 2013). Metagenomic analyses have been performed on biomass collected from suspended growth PBRs treating wastewater (Carney et al., 2014; Krustok et al., 2015). However, no study has investigated microbial community structure in biofilm PBRs treating wastewater.  Despite the dependence of redox conditions and consequent biological processes on PBR operation and design, there has yet to be a side by side comparison of nitrogen removal pathways in open versus closed biofilm PBRs within the same study. Furthermore, the microbial communities in PBR biofilms treating wastewater have not been characterized as of yet.     15 Chapter 3: Thesis Objectives The motivation of this work was to gain new understanding of the combined performance of microscreens and biofilm PBRs for treating decentralized wastewater, and in the case of biofilm PBRs, investigate the impact of reactor design and influent feed regime on treatment efficiency and microbial community structure. Two biofilm PBRs were set up and operated in parallel for three consecutive 30-day experimental phases, using microscreened decentralized wastewater as feedstock.   The research was conducted to fulfill the following objectives: 1. To determine the effectiveness of microscreening as a primary treatment step in decentralized wastewater treatment. Raw wastewater was collected from the sewer at UBC campus and characterized before and after microscreening for critical wastewater parameters. The goal of this analysis was to provide insight into whether the removal efficiencies and compact nature of the microscreen are sufficient to justify its associated primary solids management requirement.  2. To determine the influence of open versus closed biofilm PBR configurations on redox conditions, treatment efficiency, and microbial community structure. To do so, the top of one PBR was sealed, while the second PBR was left open to the atmosphere. The reactors were intermittently lit in a 16hr-8hr light-dark cycle to encourage alternating oxic-anoxic periods, creating optimal conditions for both nitrification and denitrification to occur. The PBRs were continuously fed, without supplemental alkalinity in Phase A, and with supplemental alkalinity in Phase B. Phase B was necessitated based on an alkalinity   16 limitation observed in Phase A. The goal from the analysis of these results was to provide clarity on how open and closed PBRs differ in order to better inform future designs.  3. To determine the influence of influent feed regime on redox conditions and treatment efficiency. In Phase C, all conditions were kept the same as in Phase B, except the PBRs were fed once per day, at the outset of the 8-hour dark period, in an effort increase carbon availability for denitrification. The overarching goal from this phase was to optimize PBR performance and improve effluent quality.   17 Chapter 4: Materials and Methods 4.1 Raw Wastewater Collection Domestic wastewater was collected once per week in ~15 L samples for a period of 150 days (between January 2018 and June 2018) from the sewers flowing from the Acadia Park and Orchard Commons residences at UBC Vancouver Campus between the hours of 8 and 9 AM for immediate analysis. There was no significant difference in wastewater quality between the two residences, including for COD (Figure A.1) and SS (Figure A.2), so their results are summarized as one sample source.   4.2 Raw and Microscreened Wastewater Characterization The bench scale microscreening unit used in this study was a Salsnes filter with a 54 µm nominal pore size (Trojan Technologies, London, CA). The first step of sample processing involved filtering an entire 15 L sample through a 1 mm sieve to remove large solids for separate quantification. The filtrate was then gently stirred with a magnetic stirrer to keep the remaining settleable solids in suspension. Three 250 mL subsamples were immediately collected from the stirred tank to characterize the 1mm-sieved wastewater, while three more 250 mL subsamples were microscreened at 54 µm before analysis to characterize the performance of the microscreen. The sieved solids (>1 mm) were blended and diluted before undergoing separate characterization. The mass of each analyte present in the sieved solid dilution was averaged over the 15 L sample volume and added to analyte concentrations in the 1mm-sieved wastewater for an overall raw wastewater concentration.  The parameters characterized include: total, suspended, and fixed solids, turbidity, transmittance, alkalinity, pH, COD, TKN, TP, NH4-N,   18 NOx-N, NO2-N, PO4-P as well as soluble fractions of COD, TKN, and TP. Soluble fractions were separated using a 1.2 µm glass fibre filter. The boundary between soluble and particulate matter has been defined as at a particle diameter less than 2µm (Standard Methods for the Examination of Water and Wastewater, 2003), although glass fibre filters of 1.2 µm are widely used for high solids wastewaters (Todt et al., 2015).  4.3 Biofilm Photobioreactors 4.3.1 Set-up Two identical PBRs were constructed from ½” acrylic based on the reactor geometry used by Posadas et al. (2013). The PBRs were each comprised of a distribution chamber and settling chamber, each 15 cm wide, 10 cm long, and 10 cm deep, bridged by a slanted cultivation surface (15 cm wide and 30 cm long) with a 0.5% slope to maintain uniform flow (Figure 4.1). The height of cultivation surface was 8.5 cm at the influent side and 8.35 cm at the effluent side.  Figure 4.1 - Schematic diagram of the laboratory scale biofilm PBRs   19 The influent and recirculated effluent were pumped into the distribution chamber and distributed evenly across the cultivation surface at a depth of approximately 5 mm. The cultivation surface consisted of acrylic plastic with 60 grit sandpaper fixed to the surface to encourage biofilm attachment. A 250 µm microscreen was added to the top of the settling chamber to capture any sloughed solids. The microscreen support took up 0.35 L of the collection tank volume such that the overall working volume was 2.5 L. The influent and recycle inlets were located 1 cm from the bottom of the distribution tank whereas the effluent and recycle outlet were located higher, 7 cm from the bottom of the collection tank, to allow for any settleable solids to fall out of suspension before discharge. NPT barbed male pipe adapters (1/8 inch), size 16 Masterflex norprene tubing, and 600 rpm Masterflex L/S pumps were used for the influent, effluent, and recycle lines (Cole Parmer, Vernon Hills, USA). The tubing size (L/S 16) was selected to reduce the risk of clogging in the system. A second effluent port was subsequently added in the base of the distribution chamber to facilitate batch operation in Phase C.  The two PBRs were located side by side underneath a full spectrum 300 W LED grow light operated on a 16hr-8hr light-dark regime (Super Bright LEDs, Earth City, USA). The PBRs and light were contained within a reflective housing such that the irradiance was even across the cultivation surface. The light was located approximately 1 metre above the cultivation surface, distanced to provide an irradiance of 150 µmol/m2/s as measured with an irradiance meter (QSL-100, Biospherical Instruments, San Diego, US).   The distribution tank, cultivation surface, and settling tanks for the two PBRs were fitted with individual lids, each composed of a ¼” acrylic cap with a ½” acrylic inset. The caps were   20 screwed into the frame of the PBR, and the insets were sunk inside, to eliminate gas headspace between water level and the lid. The lids covering the distribution and settling tanks were opaque, and were installed on both PBRs to prevent the growth of suspended algae in those areas. The lid covering the cultivation surface was transparent, and was installed only on the closed reactor. A 1/8” rubber gasket was also fitted for top of the closed PBR, between the lids and the frame, to prevent gas transfer with the surrounding atmosphere. An 11 mm hole was drilled in the centre of the lid for each distribution tank, and plugged with a pierceable TPE stopper, to facilitate in-situ liquid sampling. Adjacently, two 1”  holes were drilled and fitted with cable glands to mount in-situ DO and pH probes.  Influent was comprised of microscreen effluent mixed with tap water, proportioned to approximate the desired influent COD concentration of 250 mg/L. This influent was stored in bulk at 4°C in a walk-in fridge and replaced with freshly sampled and screened wastewater on a weekly basis. Every second day, influent was manually transferred from the walk-in fridge to a 6 L influent tank in a small fridge (4°C) and located adjacent to the PBRs. From there, the influent was continuously mixed gently using a magnetic stirrer, and automatically fed to the PBRs. The influent pump, effluent pump, and light were controlled with a cyclical timed controller built at UBC. The PBR temperature and the pH were left uncontrolled, but were monitored.   4.3.2 Irradiance Calculation The minimum light requirement for complete assimilation of easily degradable COD for the given flow rate was calculated stoichiometrically according to the algal (Equation 4.1) and bacterial (Equation 4.2) growth equations laid out by Boelee et al., (2014).   21  Equation 4.1 - Algal Growth Stoichiometric Equation CO2 + 0.12NH4+ + 0.01H2PO4− + 0.69H2O → 1.19O2 + 0.11H+ + 1 C1H1.78N0.12O0.36P0.01 Equation 4.2 - Bacterial Growth Stoichiometric Equation CH3COO− + 0.88O2 + 0.22NH4+ + 0.019H2PO4− + 0.8H+ → 0.91CO2 + 1.6H2O + 1.1C1H1.4N0.2O0.4P0.017  It was calculated that 110 mg/L of O2 would be required by the bacteria to consume the target influent COD concentration, 250 mg/L, and that the corresponding CO2 production would be 156 mg/L. For the algae to produce 110 mg/L of O2, it requires 127 mg/L of CO2, leaving a residual concentration of 29 mg/L CO2, which can be consumed by the algae for further nutrient removal. The lighting required was calculated based on a CO2 consumption of 156 mg/L through algae growth at a quantum yield of 0.03 mol O2/mol PAR photons (400-700nm) (Boelee et al., 2011). This resulted in an applied Photon Flux Density of 100 µmol/m2/s. However, the PBRs were only lit for 16 hours a day so the lighting was set at 150 µmol/m2/s.   Table 4.1 - PBR design parameters based on mass balances of carbon dioxide and oxygen PBR Design Parameter Value Target COD Consumption (mg/L) 250 Quantum Yield (mol O2/mol PAR photons) 0.03 Illumination Period (hrs/day) 16 Photon Flux Density (µmol/m2/s) 150  4.3.3 Inoculation The PBRs were inoculated in accordance with the procedure used by Boelee et al., 2011. One week prior to commencement of the startup phase, biomass, harvested from the activated sludge   22 tank in the onsite wastewater treatment facility at UBC’s Centre for Interactive Research on Sustainability, was rubbed on the cultivation surface. Two hours prior to startup, the PBRs were filled with 50% wastewater and 50% trickle filter effluent collected from Langley wastewater treatment plant in Metro Vancouver. The objective was to inoculate the system with algae and both heterotrophic and nitrifying bacteria acclimatized to municipal wastewater.   4.3.4 Biomass Wasting A portion of the solids on the cultivation surface were harvested every second day such that the solids retention time (SRT) was 9 days, a figure based on the 12-day biofilm lifecycle observed by Boelee et al., 2011. To ensure that exactly 2/9ths of the cultivation surface could be sampled at a time, grooves were machined into the PBR walls at ten regular intervals along the length of the cultivation surface such that dividers could be inserted to control the area of biofilm harvested (Figure 4.2).   Figure 4.2 - Harvesting method used for collecting biomass from the two PBRs`    23 4.3.5 Influent Loading Rate Often biofilm PBRs have some storage volume such that the wastewater only spends a fraction of its HRT in contact with the biofilm. However, this volume varies significantly from one study to the next, resulting in HRT’s ranging from a few hours to over a week (Table 4.1). For this reason, the influent loading rate was selected based on “effective” HRT, or time spent in contact with the biofilm, which is comparatively stable in the literature (Table 4.1). The effective HRT’s used in this study were 4.9 and 4.2 hours in the continuous and batch experiments, respectively. The minor decrease in effective HRT, from continuous to batch operation was not intentional, but a consequence of the adaptation of the PBR design to suit the batch feed regime. The internal recycle rate was set at 135 mL/min to provide a similar surface velocity, 0.9 mm/s, as was used in comparable studies (Table 4.1).   Table 4.2 - HRT, effective HRT, and surface flow velocity used in existing literature Reference HRT Effective HRT Surface Flow Velocity   (hrs) (hrs) (mm/s) (Zamalloa et al., 2013) 24 1.8 2 (Posadas et al., 2013) Phase III 250 20 14 (Posadas et al., 2013) Phase IV 125 10 14 (Posadas et al., 2013) Phase V 74 6 14 (Boelee et al., 2014) 4.5 4.5 0.6 This Study 46 - 54 4.2 - 4.9 0.9  4.3.6 Phases of Operation A 2-month start up period was conducted in which it was established that operation was stable and similar in the two reactors before Reactor 2 was sealed. Three separate experiments, A-C, were then run to determine the comparative performance of open and closed biofilm PBRs while   24 varying two operational conditions: alkalinity addition and the influent feed regime (Table 4.2). Originally, only two phases were planned, one operating under a continuous feed regime and the other under a batch regime. However, a third phase was necessitated when an alkalinity limitation was observed during continuous operation in Phase A. The continuous feed regime was then repeated for Phase B, except with the addition of 100 mg/L CaCO3 of supplemental alkalinity to the influent. In Phase C, supplemental alkalinity was continued, but a batch influent feed regime was applied. Each experimental phase was then run for a period of at least 27 days each (three SRT’s) to provide sufficient time for the biofilm to stabilize.  Table 4.3 - Experimental conditions for the startup phase and three experimental phases observed in this study. Phase Days Open/Closed CaCO3  Added Influent Regime Influent Rate Influent Period HRT  Effective HRT    Reactor 1 Reactor 2 (mg/L)  (mL/day) (hrs) (hrs) (hrs) Startup  Open Open - “Continuous” 1100 2.4 54 4.9 A 0-30 Open Closed - “Continuous” 1100 2.4 54 4.9 B 36-64 Open Closed 100 “Continuous” 1100 2.4 54 4.9 C 65-93 Open Closed 100 “Batch” 1300 24 46 4.2  4.3.6.1 Influent Feed Regimes The feed regime was adjusted between a “continuous” condition in Phase A and B, and a “batch”  condition in Phase C. True continuous operation was not possible due to the small influent rate required, coupled with the size of influent tubing needed to avoid clogging. As a result, influent was fed in semi-continuous discrete doses under both conditions, but at different frequencies and volumes. During an influent event, effluent was first pumped from the PBRs before an equivalent volume of influent was added. In the continuous feed regime, the PBRs were fed once   25 every 2.4 hours with 110 mL of fresh influent, and effluent was sampled for analysis at hour 17 (Figure 4.3). Under the batch feed regime, the PBRs were fed once every 24 hours with 1300 mL of fresh influent, and effluent was sampled at hour 24 (Figure 4.3). In the results and discussion, these effluent readings are referred to as the H17 and H24 concentrations, respectively.         1300 Dark Period Light Period x                                                                                                                 110         x                    0   8  17    24  Figure 4.3 – Illustration of the frequency and volume of discharge/influent events under the continuous (hollow line) and batch (solid line) feed regimes. The discharge events sampled for daily analysis are denoted by an x.  4.3.7 Diel Cycle Monitoring Experiments Two diel monitoring experiments were performed at the end of each experimental phase to observe the behaviour of the PBRs over the course of a 24-hour cycle, as well as ascertain instantaneous reaction rates. In Phases A and B, when the PBRs were operating under the continuous feed regime, in-situ samples were taken every 25 minutes in order to observe behavioural trends within both the 16hr-8hr light cycle and the 2.4 hour influent feeding cycle. In all three of the phases, 24-hour batch tests were also conducted in which the PBRs were sampled Influent Volume (mL) Hour   26 every 3 hours in-situ, allowing for the estimation of hourly reaction rates. Samples were analysed for COD, pH, DO and filtered for analysis of dissolved COD, NH4-N, NOx-N, NO2-N, TKN, TP, PO4-P. Alkalinity was also measured but only at the beginning and end of the light/dark periods due to the large volume required for titration.  4.3.8 Characterization of the microbial community with 16S rRNA gene amplicon sequencing 16S rRNA gene amplicon sequencing was conducted to explore the impact of PBR operating conditions on the microbial community structure. A sample of biomass was collected after each experimental phase and stored at -20°C for batch DNA analysis at the end of the experiment. DNA extraction, amplification, and sequencing were performed in triplicate by Microbiome Insights (Vancouver, CA). DNA extraction was carried out according to the method taken from MOBIO Laboratories Inc using the Mobio Power Mag soil DNA isolation bead plate (Qiagen, Hilden, DE) and KingFisher Flex magnetic particle processor (Thermo Fisher Scientific, Waltham, US). PCR amplification and sequencing was done according to the method taken from Kozich et al., (2013) using the Miseq Gene Sequencer with 250-bp Paired-End Kit (Illumina, San Diego, US). 16S 515F/806R amplicon primers were used to target the V4 region of the 16S rRNA gene. Bioinformatics analysis was conducted using R statistical computing software (v. 3.5.1). Reads were processed with quality-filtering and denoising using the DADA2 package (v. 1.9.2)  (Callahan et al., 2016). The read numbers per sample were then normalized using the DESeq2 package (v. 1.20.0) (Love et al., 2014) and classified with the RDP classifier algorithm (Wang et al., 2007) against the SILVA rRNA database (v. 132) (Quast et al., 2013). Differential abundance analysis was conducted to detect genera that significantly differed in the open and   27 closed PBRs using STAMP (v. 2.1.3) (Parks et al., 2014). The Shannon Index of the relative abundance at the genus rank was calculated using the vegan package (v. 2.5.3) (Oksanen et al., 2018). A variance stabilizing transform was then applied, again using the DESeq2 package (Love et al., 2014), followed by principle coordinate analysis of the relative abundance of genera using the mixOmics package (v. 6.6.0) (Rohart et al., 2017).   4.4 Sampling and Analytical Methods Liquid effluent samples were taken two to three times per week from the PBRs and analysed according to Standard Methods for the same parameters monitored in the microscreen tests (Table 4.3). Two-ninths of the cultivation surface were scraped every second day and analysed for wet weight, dry weight, chlorophyll and carotenoid content, %C, %N, and microbial composition. All analyses were carried out in triplicate. Methods are provided for all analyses below (Table 4.4).   Table 4.4 – Influent (Inf), effluent (Eff), and solids (Sol) sampling and analysis regime in the open and closed PBRs over 27 days (3 SRT) of an experimental phase.   1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Effluent pH x x x x x x x x x x x x x x x x x x x x x x x x x x x DO x x x x x x x x x x x x x x x x x x x x x x x x x x x All i x   x    x   x    x   x    x   x   Influent All i x   x    x   x    x   x    x   x   Solids Wet Weight x  x  x  x  x  x  x  x  x  x  x  x  x  x Dry Weight x    x    x    x    x    x    x   Pigmentsii  x    x    x    x    x    x    x   DNA                           x EA                           x                                                i All of the same liquid parameters monitored in the microscreen tests. ii Pigments includee chlorophyll A, chlorophyll B, and cartenoids.   28 Table 4.5 - Summary of analysis techniques chosen for each measured analyte. Analyte Method Instrument Preparation  Container Liquid Analysis     Alkalinity Standard Method 2320 B (Titration) TetraSipTM (Mantech, Guelph, ON) Immediate Analysis 200mL Glass Beaker Ammonia  (NH3-N) Standard Method 4500-NH3 H (Flow Injection) Flow Injection Analyser (QuickChemâ 8000 Series, Lachat Instruments, Milwaukee, US) 1.2 µm filtration on nitrocellulose, pH<2 with H2SO4, refrigerated at 4°C PS Culture Tube Biochemical Oxygen Demand  (BOD) Standard Method 5210 B (5-Day BOD Test) Dissolved Oxygen Meter (DOH-SD1, Omega Engineering, Norwalk, US) Immediate Analysis 300mL BOD Bottles with Pennyhead Stoppers Chemical Oxygen Demand (COD) Standard Method 5220 D (Closed Reflux Colormetric) Spectrophotometer (DR2800, Hach Co., Loveland, US) COD Reactor Block  (accu-TESTTM , Bioscience, Inc., Allentown, PA) Immediate Analysis; for soluble COD: 1.2 µm filtration on nitrocellulose 10 mL glass vial with PP cap Dissolved Carbon  Standard Method 5310 B (Total Carbon) TOC Analyzer (TOC-LCPH/CPN, Shimadzu Co., Kyoto, Japan) Immediate Analysis, 1.2 µm filtration on nitrocellulose 20mL Glass Vial with PTFE Lined Cap Dissolved Oxygen  (DO) Standard Method 4500-O G (Membrane Electrode) Dissolved Oxygen Meter (DOH-SD1, Omega Engineering, Norwalk, US) Immediate Analysis 200mL Glass Beaker Dissolved Solids Standard Method 2540 C (Total Dissolved Solids Dried at 180°C) Forced Air Oven (1350 FM, VWR Scientific, Radnor, US) Immediate Analysis, 1.2 µm filtration on nitrocellulose 50mL Porcelain Evaporating Dish Fixed Solids Standard Method 2540 E (Fixed and Volatile Solids Ignited at 550°C) 30400 Furnace (Type 30400 Thermolyne, Thermo Fisher Scientific, Waltham, US) Immediate Analysis, 1.2µm filtration on nitrocellulose for FSS 50mL Porcelain Evaporating Dish, 42mL Aluminum Weigh Dish Microscopy Standard Method 10200 D Microscope As per method 25x75mm Glass Slides   29 (Axioplan 2, Carl Zeiss Microscopy, Jena, DE) Nitrite  (NO2-N) Standard Method 4500-NO2- H (Cadmium Reduction Flow Injection) Flow Injection Analyser (QuickChemâ 8000 Series, Lachat Instruments, Milwaukee, US) 1.2 µm filtration on nitrocellulose, pH<2 with H2SO4, refrigerated at 4°C PS Culture Tube Nitrogen Oxides  (NOx-N) Standard Method 4500-NO3- I (Cadmium Reduction Flow Injection) Flow Injection Analyser (QuickChemâ 8000 Series, Lachat Instruments, Milwaukee, US) 1.2 µm filtration on nitrocellulose, pH<2 with H2SO4, refrigerated at 4°C PS Culture Tube Orthophosphate  (PO4-P) Standard Method 4500-P G (Flow Injection) Flow Injection Analyser (QuickChemâ 8000 Series, Lachat Instruments, Milwaukee, US) 1.2 µm filtration on nitrocellulose, pH<2 with H2SO4, refrigerated at 4°C PS Culture Tube pH Standard Method 4500-H+ B (Electrometric) Mantech TitraSipTM System Immediate Analysis 200 mL Glass Beaker Total Kjeldhal Nitrogen  (TKN) Standard Method 4500-Norg D (Block Digestion and Flow Injection) Flow Injection Analyser (QuickChemâ 8000 Series, Lachat Instruments, Milwaukee, US) 1.2 µm filtration on nitrocellulose, pH<2 with H2SO4, refrigerated at 4°C PS Culture Tube Total Solids (TS) Standard Method 2540 B (Total Solids Dried at 180°C) Forced Air Oven (1350 FM, VWR Scientific, Radnor, US) Immediate Analysis 50 mL Porcelain Evaporating Dish Total Suspended Solids  (TSS) Standard Method 2540 D (Total Suspended Solids Dried at 103-105°C) Forced Air Oven (1350 FM, VWR Scientific, Radnor, US) Immediate Analysis, 1.2 µm filtration on nitrocellulose 42 mL Aluminum Weigh Dish Turbidity Standard Method 2130 B (Nephelometric) Turbidimeter (2100P, Hach Co., Loveland, US) Immediate Analysis 10 mL Turbidity Vial Solids Analysis     Cartenoids Adapted from Leu and Hsu, 2005. (Lysis and Spectrophotometry) Spectrophotometer (DR2800, Hach Co., Loveland, US) Refrigerated at 4°C; collected every second day and analysed once per week 15 mL PP Centrifuge Tube   30 Chlorophyll Adapted from Leu and Hsu, 2005. (Lysis and Spectrophotometry) Spectrophotometer (DR2800, Hach Co., Loveland, US) Refrigerated at 4°C; collected every second day and analysed once per week 15 mL PP Centrifuge Tube Dry Weight Standard Method 2540 B (Total Solids Dried at 180°C) Forced Air Oven (1350 FM, VWR Scientific, Radnor, US) Refrigerated at 4°C; collected every second day and analysed once per week 50 mL Porcelain Evaporating Dish Elemental Analysis Method taken from Verardo et al., 1990 Elemental Analyser (Vario MICRO Cube, Elementar Americas, Langenselbold, DE) Centrifuged at 3000 rpm for 10 minutes, decanted, and frozen at -20°C 15 mL PP Centrifuge Tube 8x5 mm Tin Capsules Photosynthetic Efficiency Method taken from Posadas et al., 2013 Quantum Scalar Irradiance Meter (QSL-100, Biospherical Instruments, San Diego, US) n/a n/a   31 4.5 Calculations Influent and effluent from the two PBRs were monitored for identical parameters as in the characterization of the raw wastewater to allow for mass balance analyses. Those calculations are outlined below.  A nitrogen mass balance was performed to calculate the nitrate (NO3-N) (Equation 4.3), organic nitrogen (Norg) (Equation 4.4), and total nitrogen (TN) (Equation 4.5) in the liquid phase:   Equation 4.3 – NO3-N concentration (mg/L) !"#$ = !"#& − !"#( Equation 4.4 - Organic nitrogen concentration (mg/L) !)*+ = !,-" − !"./  Equation 4.5 - Total nitrogen concentration (mg/L) !," = !,-" + !"#&  where: Nx is the concentration of the given nitrogenous species, mgN/L.  From influent and effluent concentrations, daily removal efficiencies (Equation 4.6) and production/consumption rates (Equation 4.7) were calculated:  Equation 4.6 - Removal efficiency (%) of a given parameter C  1234567	9::;<;2=<> = ?1 − ABCCADECF × 100    32 Equation 4.7 - Production/consumption rate  1I = (ABCC − ADEC)1000 × LM  where: where Cinf and Ceff represent any parameter concentration in the influent and effluent (mg/L), respectively, Q is the daily influent rate (L/day), A is the cultivation surface area (m2), and RC is the production/consumption rate (g/m2/day).  The rate of assimilation of nitrogen into biomass was calculated based on the dry biomass production rate (g TS/day) and the nitrogen mass fraction in the dried biomass (g N/g TS) (Equation 4.8):  Equation 4.8 - Rate of assimilation of nitrogen into biomass 1ND)OPQQ," = 1ND)OPQQ × !%  where: RBiomass is the average dry weight of biomass produced per day in a given week (g TS/m2/day) and N% is the fraction of nitrogen present in the biomass sample taken for elemental analysis at the end of each phase (g N/g TS).  Nitrification (Equation 4.9) and denitrification (Equation 4.10) process rates were then calculated based on the consumption of TKN and TN in the PBRs:  Equation 4.9 – PBR nitrification rate !;TU;:;<6T;4=	16T2 = 1,-" − 1ND)OPQQ,"   33 Equation 4.10 – PBR denitrification rate  V2=;TU;:;<6T;4=	16T2 = 1," − 1ND)OPQQ,"  where: RTKN is the total consumption of TKN in the PBR (gN/m2/day), RTN is the total consumption of nitrogen in the PBR (gN/m2/day), and RBiomass,N is the biomass assimilated nitrogen (gN/m2/day).  Similar equations were proposed by Zamalloa et al., (2013) for biofilm PBRs where ammonia evaporation was also considered. Ammonia evaporation was considered to be insignificant in this study, as the pH never approached 10, the recommended pH for ammonia stripping (USEPA, 2000). Furthermore, as aeration was provided passively as opposed to mechanically, no bubbles were present to encourage the stripping process. The low temperature and limited airflow around the enclosed PBRs also reduced potential for evaporation.  Specific alkalinity demand was also calculated to determine the net result of the alkalinity consuming (nitrification, ammonia assimilation, and heterotrophic assimilation) and producing (ammonification and denitrification) wastewater treatment processes (Equation 4.11). This was taken to be the difference between the initial or “baseline” alkalinity concentration (Equation 4.12) and the effluent alkalinity concentration, divided by influent TN:  Equation 4.11 - Specific alkalinity demand  WX2<;:;<	M7Y67;=;T>	V236=Z = MNPQB − M[CC\!]EC    34 Equation 4.12 – Influent baseline alkalinity  MNPQB = M,^ + M_``B`	  where: ABase is equal to the influent baseline alkalinity (mg/L CaCO3), AEff is the alkalinity of the effluent (mg/L CaCO3), ATW is the alkalinity of tap water prior (mg/L CaCO3), AAdded is the alkalinity added for treatment (mg/L CaCO3) and TNInf is the influent total nitrogen concentration (mg/L CaCO3).  Hourly production/consumption rates in the batch diel monitoring experiment were calculated every three hours for NO3-N and COD (Equation 4.13):  Equation 4.13 – Instantaneous production/consumption rate  1I = (Aa − Aabc)1000 × LM  where: RC is the instantaneous production/consumption rate (g/m2/day), Ct is the concentration of an analyte at time t, Ct-3 is the concentration of the analyte three hours earlier (mg/L), Q is the daily influent rate (L/day), and A is the cultivation surface area (m2).  4.6 Significance Testing The data in this paper is summarized by the use of a mean and a 95% confidence interval. Statistical significance was tested when a parameter was compared between the open and closed PBRs, when a parameter was compared between consecutive phases, and when a parameter was compared with a reference. The associated statistical tests are presented below (Table 4.5).   35 Normality was confirmed using a Shapiro-Wilk normality test. Supplemental graphs of the data produced in this study are provided in Appendices A, B, C, D, and E to reduce dependence on p-values and allow greater insight into effect sizes.  Table 4.6 – Summary of the statistical tests performed in this study. Comparison Statistical Test Alpha Between PBRs Paired Sample T-Test 0.05 Between Phases Independent Two Sample T-Test Unequal Variance 0.05 With a Reference One Sample One Sided T-test 0.05   Chapter 5: Results 5.1 Raw Wastewater Quality and Microscreen Performance The raw wastewater characteristics were quite variable in the 17 samples analysed (Table 5.1)(Appendix A). This was presumably due to proximity of sample collection to the source, as at that location, the sewer flow is still better described as a small number of discrete discharges, not yet averaged by scale (Harder, 2012). The mean values were largely similar to those of medium to high strength domestic wastewater (Metcalf and Eddy et al., 2013). For instance, total solids in the raw wastewater was 716 ± 352 mg/L compared with 720 mg/L for medium strength domestic wastewater (MSW) (Metcalf and Eddy et al., 2013). However, the average COD/TN ratio of the sampled wastewater was 8.50 ± 2.97, which was significantly lower than the ratio for MSW of 10.75 (p < 0.001) (Metcalf and Eddy et al., 2013). The relatively low COD/TN ratios likely had implications for nitrification rates in the PBRs, which have been shown to improve at lower COD/TN ratios (Sharma and Gupta, 2004).  The fraction of total solids present as suspended solids was relatively high, 0.44 ± 0.09, in comparison to the value of 0.29 for MSW (Metcalf and Eddy et al., 2013). The high suspended fraction translated into particularly high suspended solids and total solids removal efficiencies of the microscreen, which averaged 70 ± 6% and 38 ± 9%, respectively. A microscreen in centralized treatment would expect significantly lower suspended solids removal efficiencies of 50% (p < 0.001) (Salsnes Filter AS., 2017). Assuming a TSS/TS ratio of 0.29 (Metcalf and Eddy et al., 2013), this translates to a TS removal efficiency of just 15%. A primary clarifier in centralized treatment would also expect significantly lower TSS and TS removal efficiencies of 40% (p < 0.001) and 9% (p < 0.001), respectively (Metcalf and Eddy et al., 2013).     37 Table 5.1 - Influent and microscreened wastewater quality and consequent removal efficiency. Parameter n Raw wastewater Microscreened Effluent  Microscreen Removal Efficiency (%) TSS (mg/L) 17 403 ± 331 82 ± 26 70 ± 5.7 TS (mg/L) 17 716 ± 352 380 ± 82 38 ± 8.8 Sus. Fraction 17 0.44 ± 0.09 0.21 ± 0.04 n/a Fixed Fraction 17 0.27 ± 0.06 0.35 ± 0.07 n/a Turbidity (NTU) 17 138 ± 83 113 ± 52 24 ± 6.9 Transmittance (%) 14 8.2 ± 6.5 9.4 ± 4.2 n/a COD (mg/L) 17 916 ± 499 470 ± 111 39 ± 9.1 COD Soluble (mg/L) 17 267 ± 78 294 ± 68 -1.3 ± 6.1 TKN (mg/L) 17 110 ± 37 98 ± 29 13 ± 10 TKN Soluble (mg/L) 17 30 ± 7.2 30 ± 6.1 -15 ± 21 NH4 (mg/L) 17 24 ± 6.8 26 ± 4.6 n/a NO3 (mg/L) 17 -0.09 ± 0.26 0.10 ± 0.21 n/a NO2 (mg/L) 17 0.29 ± 0.16 0.25 ± 0.13 n/a TN (mg/L) 17 111 ± 37 98 ± 29 13 ± 10 Total Phos. (mg/L) 14 7.9 ± 4.7 4.8 ± 1.8 30 ± 18 PO4 (mg/L) 17 5.2 ± 1.1 5.4 ± 1.6 n/a Alkalinity (mg/L) 11  116 ± 45 n/a  5.2 Biofilm PBR Performance Overall performance of the PBRs in this study relative to reuse guidelines (Table 2.1) can be effectively separated into secondary treatment performance and nitrogen removal. The target parameters associated with secondary wastewater treatment are effluent suspended solids, BOD, and turbidity whereas the target parameters for nitrogen removal are effluent NH4-N and NO3-N. The PBR effluent in this study met secondary treatment guidelines for unrestricted urban reuse independent of operating conditions whereas nitrogen removal varied. Effluent quality (Table 5.2) and removal efficiency (Table 5.3) are reported below. Graphs of the raw influent and effluent data, and process rates are provided in Appendix B.   38 Table 5.2 - Influent and effluent wastewater quality for the PBRs through each experimental phase. Parameter A n = 12 i B n = 10 i C n = 8 i Influent Influent Influent Open Effluent Closed Effluent Open Effluent Closed Effluent Open Effluent Closed Effluent TSS (mg/L) 51 ± 11 97 ± 61 73 ± 16 7.3 ± 3.5 5.3 ± 2.5 7.1  ± 2.6 7.3  ± 3.3 6.9 ± 10 9.0 ± 10 TS (mg/L) 186 ± 33 483 ± 150 405 ± 78 177 ± 30 153 ± 34 308 ± 37 244 ± 58 231 ± 29 211 ± 25 Turbidity (NTU) 41 ± 7.1 133 ± 122 76 ± 42 1.9 ± 0.5 1.6 ± 0.3 2.2 ± 0.4 2.1 ± 0.5 1.8 ± 0.3 2.8 ± 1.7 Transmittance (% 254nm) 23 ± 9.7 15 ± 12 5.2 ± 3.0 60 ± 3.5 66 ± 2.5 51 ± 3.4 60 ± 5.1 57 ± 6.0 56 ± 7.0 COD (mg/L) 220 ± 61 401 ± 198 329 ± 45 63 ± 8.0 50 ± 9.5 68 ± 15 57 ± 22 62 ± 26 68 ± 22 COD Soluble (mg/L) 132 ± 74 204 ± 98 204 ± 54 46 ± 8.9 35 ± 7.8 60 ± 10 38 ± 11 70 ± 37 53 ± 14 BOD (mg/L) 66 ± 17 153 ± 96 144 ± 34 3.4 ± 0.9 1.6 ± 0.5 5.3 ± 2.4 5.1 ± 3.8 4.0 ± 0.7 6.3 ± 3.2 TKN (mg/L) 62 ± 13 64 ± 24 94 ± 30 20 ± 2.4 17 ± 4.1 10 ± 11 13 ± 17 13 ± 17 23 ± 16 TKN Soluble (mg/L) 40 ± 12 30 ± 14 39 ± 9.9 16  ± 2.0 11 ± 2.8 5 ± 8.2 9 ± 14 10 ± 15 21 ± 15 NH4-N (mg/L) 38 ± 13 26 ± 11 32 ± 7.4 18 ± 2.2 12 ± 3.1 5 ± 6.6 7 ± 10 9 ± 11 18 ± 12 NO3-N (mg/L) -0.2  ± 0.3 0.1 ± 0.4 0.0  ± 0.4 20 ± 1.0 13 ± 4.3 10 ± 5.1 8 ± 4.5 10 ± 5.7 0.6 ± 1.3 NO2-N (mg/L) 0.4 ± 0.4 0.1 ± 0.1 0.0 ± 0.0 0.4 ± 0.2 2.1 ± 4.0 0.1 ± 0.3 0.0 ± 0.0 0.2 ± 0.3 2.0 ± 2.3 TN (mg/L) 62 ± 13 64 ± 24 94 ± 30 40 ± 2.9 32 ± 5.4 20 ± 11 21 ± 15 21 ± 13 23 ± 14 TP (mg/L) 2.4 ± 0.7 2.9 ± 1.1 3.5 ± 1.6 1.0 ± 0.3 0.8 ± 0.4 1.3 ± 0.8 2.1 ± 1.6 1.3 ± 1.0 1.9 ± 1.2 PO4-P (mg/L) 4.5 ± 1.5 4.4 ± 2.4 7.5 ± 3.6 2.6 ± 0.3 2.6 ± 0.4 1.8 ± 1.4 2.0 ± 1.4 3.7 ± 1.6 3.6 ± 1.2 Alkalinity (mg/L) 188 ± 135   280 ± 79 263 ± 53 -4.7 ± 17  -3.4 ± 14  157 ± 196 193 ± 238 114 ± 66 191± 76 COD/TN  3.7 ± 1.2 7.5 ± 5.4 4.2 ± 0.9 1.7 ± 0.3 1.6 ± 0.3 9.4 ± 9.7 9.1 ± 10.2 3.8 ± 1.6 4.4 ± 2.4                                                i The actual n value for each parameter varies slightly due to missing data and can be found in Appendix B with the raw data.   39 Table 5.3 - PBR removal efficiency (%) of key parameters through the three experimental phases conducted in this study. Parameter A n = 12 i B n = 10 i C n = 8 i Open Closed Open Closed Open Closed TSS 85 ± 8 90 ± 6 88 ± 7 86 ± 9 87 ± 20 85 ± 21 Turbidity 95 ± 1 96 ± 1 94 ± 3 94 ± 3 97 ± 2 96 ± 1 COD 71 ± 4 77 ± 4 75 ± 8 81 ± 7 83 ± 3 81 ± 2 BOD 95 ± 1.1 98 ± 1 95 ± 2 96 ± 2 97 ± 1 95 ± 2 TKN 69 ± 4 76 ± 4 88 ± 9 85 ± 15 93 ± 5 78 ± 6 TN 37 ± 8 52 ± 4 66 ± 11 67 ± 13 77 ± 9 76 ± 8 TP 53 ± 17 67 ± 13 56 ± 17 46 ± 24 59 ± 22 37 ± 26  5.2.1 Secondary Treatment Performance While influent COD concentrations fluctuated around the target value of 250 mg/L over the three experiments, secondary treatment performance was stable and satisfied all treatment targets for unrestricted urban (non-potable) reuse except for turbidity(Figure B.3). Effluent suspended solids were consistently below the 10 mg/L target at 7.8 ± 2.3 and 7.2 ± 2.2 mg/L in the open and closed PBRs, respectively (Figure B.9). Effluent BOD levels were likewise below the 10 mg/L target at 4.3 ± 0.9 and 3.7 ± 1.4 mg/L (Figure B.2). Effluent turbidity levels however did not fall below the 2 NTU target at  2.1 ± 0.3 and 2.0 ± 0.3 NTU in the open and closed PBRs, respectively (Figure B.15). COD removal rates for the three experimental phases are reported below and were statistically similar between the open and closed PBRs (Table 5.4)(Figure B.20).                                                   i The actual n value for each parameter varies slightly due to missing data and can be found in Appendix B with the raw data.   40 Table 5.4 – Differences between the COD removal rate (g/m2/day) in the open and closed PBRs   A n = 13i B n = 10i C n = 7i  Open Closed Open Closed Open Closed COD Removal 3.84 ± 0.36 4.17 ± 0.35 7.66 ± 4.57 7.93 ± 4.49 7.54 ± 1.58 7.28 ± 1.31 (g/m2/day) p  = 0.06 p = 0.06 p = 0.14  5.2.2 Nitrogen Removal Performance Differences in DO, pH, and even temperature were observed between the open and closed PBRs in the three experimental phases and these differences produced significant differences in nitrogen removal mechanisms (Table 3.2). The results of each of the three experimental phases are individually discussed below alongside the results from the 24-hour diel studies, to provide insights into the link between PBR redox conditions and measured nitrification/denitrification rates.                                                    i The actual n value for each parameter varies slightly due to missing data and can be found in Appendix B with the raw data.   41 Table 5.5 – Summary of differences in reactor conditions and process rates between the open and closed PBRs.   A n = 11i B n =10i C n = 7i  Open Closed Open Closed Open Closed DO 6.9 ± 0.2 7.2 ± 0.8 8.9 ± 1.3 6.2 ± 0.8 6.4 ± 0.7 2.8 ± 0.5 (mg/L) p = 0.22 p < 0.001 p < 0.001 pH 5.0 ± 0.4 5.4 ± 0.4 8.5 ± 0.3 7.3 ± 0.3 7.8 ± 0.5 7.3 ± 0.1  p < 0.001 p < 0.001 p = 0.70 Temperature  21.4 ± 0.2 22.7 ± 0.2 22.8 ± 0.5 24.0 ± 0.6 24.2 ± 0.2 25.5 ± 0.3 (°C) p < 0.001 p = 0.001 p = 0.02 Nitrification 0.90 ± 0.19 0.96 ± 0.11 0.90 ± 0.21 0.90 ± 0.19 1.61 ± 0.53 1.41 ± 0.51 (gN/m2/day) p = 0.57 p = 0.99 p < 0.001 Denitrification 0.42 ± 0.17 0.58 ± 0.11 0.57 ± 0.26 0.62 ± 0.19 1.30 ± 0.56 1.37 ± 0.49 (gN/m2/day) p = 0.04 p = 0.36 p = 0.41 Assimilation 0.21 ± 0.04 0.26 ± 0.02 0.32 ± 0.02 0.23 ± 0.02 0.33 ± 0.03 0.23 ± 0.01 (gN/m2/day) p = 0.04 p < 0.001 p < 0.001 TKN Removal 1.12 ± 0.22 1.25 ± 0.14 1.21 ± 0.22 1.13 ± 0.19 1.95 ± 0.51 1.65 ± 0.51 (gN/m2/day) p = 0.05 p = 0.23 p = 0.001 TN Removal 0.64 ± 0.20 0.87 ± 0.12 0.88 ± 0.30 0.85 ± 0.22 1.64 ± 0.53 1.60 ± 0.48 (gN/m2/day) p < 0.001 p = 0.62 p = 0.64  5.2.2.1 Phase A: Continuous Feed, No Alkalinity Addition In Phase A, the open and closed PBRs were operated with continuous feeding during 16 hr-8 hr light-dark diel cycles to compare their treatment performance. However, it was observed that the influent wastewater conditions in Phase A caused an alkalinity limitation in both PBRs. The average H17 pH was 5.0 ± 0.4 and 5.4 ± 0.4 in the open and closed PBRs, respectively (Figure B.18), prompting the regular measurement of alkalinity through the remainder of the study. The                                                i The actual n value for each parameter varies slightly due to missing data and can be found in Appendix B with the raw data.   42 two H17 alkalinity measurements taken at the end of the Phase A averaged  -4.7 ± 17 mg/L CaCO3 and -3.4 ± 14 mg/L CaCO3 in the open and closed PBRs, respectively (Figure B.1). Negative alkalinity occurs when the pH dips below 4.5 and is an effective measure of PBR acidification. The low effluent alkalinity was likely caused by the low alkalinity in the tap water of Vancouver (12 ± 3.4 mg/L as CaCO3 in seven samples taken at UBC over the course of this study), and the presence of nitrification in both reactors (Figure B.22), which consumes alkalinity. The continuous diel test showed the open PBR operated at negative alkalinity levels throughout its 24-hour cycle whereas DO levels did not dip below 2 mg/L (Figure 5.1). The pH spiked every 2.4 hours as natural alkalinity was added with each new influent dose and was subsequently consumed via nitrification (Figure 5.1).   The H17 DO concentration in the open and closed PBRs was high and similar, at 6.9 ± 0.2 mg/L and 7.2 ± 0.8 mg/L, respectively (Figure B.17)( p = 0.22). During the dark period of the continuous diel test, the closed PBR DO levels dropped to 0 mg/L and a corresponding generation of alkalinity through denitrification was observed (Figure 5.1). This influx of alkalinity was likely responsible for increased pH stability in the dark period in the closed PBR (Figure 5.1). However, the alkalinity was quickly consumed again under lit conditions, and pH instability resumed (Figure 5.1).   43  Figure 5.1 - Effluent DO, pH, and alkalinity levels in the open and closed PBRs through the continuous diel monitoring experiment conducted at the end of Phase A. The dark period is denoted by the shaded area.  In the batch diel study, biomass net nitrate production (from nitrification) in the open PBR decreased to near zero during the dark period, yet net consumption (from denitrification) was not observed (Figure 5.2), indicating limited denitrifier activity relative to nitrifiers. Conversely, the closed PBR achieved an instantaneous net nitrate removal of 1.5 gNO3-N/m2/day during unlit conditions (Figure 5.3), attributable to the fact that the DO concentrations dropped to 0 mg/L (Figure 5.2). TN concentrations declined faster than NO3 concentrations in both the open and closed reactors during the unlit period (Figure 5.2), which could be attributed to simultaneous Open Closed●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●024602040600 5 10 15 20HourDO & pH●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●024602040600 5 10 15 20HourAlkalinityParameter ●ALK  ( mg/L ) DO  ( mg/L ) pH  ( NA )Experiment Hour  44 nitrification/denitrification. Biodegradable COD removal was complete in both PBRs by the end of the dark phase (Figure 5.2).  Figure 5.2 - Concentrations of key parameters during the diel monitoring study at the end of Phase A with the PBRs operating under the “batch” feed condition.  Open Closed● ● ● ● ● ● ● ● ●010203040500 5 10 15● ● ● ● ● ● ● ● ● ●010203040500 5 10 15Parameter●TN  ( mg/L )TKN.F  ( mg/L )NH3  ( mg/L )NO2  ( mg/L )NO3  ( mg/L )●●● ● ● ● ●● ● ● ●● ●01020300 5 10 15●●● ● ● ● ● ● ●● ●● ●01020300 5 10 15Parameter●DO  ( mg/L )Temp  ( *C )pH  (  )●●● ● ●●●●●●0501001500 5 10 15●●●●● ●●●● ●0501001500 5 10 15Parameter● COD  ( mg/L )ALK  ( mg/L )510150 5 10 15510150 5 10 15ParameterTP.F  ( mg/L )Experiment Hour-N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L )   45  Figure 5.3 - Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase A. The dark period is denoted by the shaded area.  Despite the differences in alkalinity and DO levels, overall nitrification rates were statistically similar (p = 0.57) at 0.90 ± 0.19 and 0.96 ± 0.11 g N/m2/day in the open and closed PBRs, respectively (Figure B.22), and nitrification represented 81 ± 6% and 77 ± 3% of TKN removal, respectively (Figure 5.4). The average daily nitrogen removal rate via denitrification was significantly higher in the closed PBR at 0.58 ± 0.11 gN/m2/day compared to 0.42 ± 0.17 gN/m2/day in the open PBR (p = 0.04) (Figure B.21) and denitrification represented 65 ± 9% and 65 ± 6% of TN removal, respectively (Figure 5.4). Biomass assimilation rates were also significantly higher in the closed PBR at 0.26 ± 0.02 gN/m2/day compared to 0.21 ± 0.04 gN/m2/day in the open PBR (p = 0.04)(Figure B.23). Overall daily TKN removal was similar in the two PBRs at 1.12 ± 0.22 and 1.25 ± 0.14 gN/m2/day in the open and closed PBRs, respectively (p = 0.05)(Figure 8.24). However, the closed PBR did achieve significantly higher TN daily removal rates at 0.87 ± 0.12 gN/m2/day compared to 0.64 ± 0.20 gN/m2/day in the open ●●●●●●●●−10123570 5 10 15HourdNO3 (g/m2/day)Parameter●OpenCloseddNO3-N (gN/m2/day)   46 PBR (p < 0.001)(Figure 8.25). This was due to the combined effect of higher assimilation and denitrification rates.   Effluent NO3-N concentrations averaged 20 ± 1.0 and 13 ± 4.3 mg/L in the open and closed PBRs, respectively (Figure B.7). Effluent NH3-N concentrations averaged 18 ± 2.2 and 12 ± 3.1 mg/L, respectively (Figure B.5).     Figure 5.4 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase A at hour 17.  0.000.250.500.751.000 10 20 30DayOpen0.000.250.500.751.000 10 20 30DayClosedkeyNorgNH4BiofilmNO2NO3N2Fraction of Influent TN  47 5.2.2.2 Phase B: Continuous Feed, Alkalinity Addition In Phase B, the PBRs were operated as in Phase A but alkalinity was added to the feed in order to remove the alkalinity limitation. The addition of alkalinity at 100 mg/L CaCO3 was sufficient  and the pH stabilized in both PBRs (Figure 5.5). The H17 pH through Phase B in the open and closed PBRs was 8.5 ± 0.3 and 7.3 ± 0.3, respectively (Figure B.18). The H17 alkalinity concentration was 157 ± 196 mg/L and 193 ± 238 mg/L, respectively (Figure B.1), indicating that even with the 100 mg/L of supplemental alkalinity removed, the overall wastewater treatment process, from tap water to effluent, had a net positive effect on alkalinity. Specific alkalinity demand was -0.75 ± 4.5 gCaCO3/gN and -1.1 ± 4.3 gCaCO3/gN in the open and closed PBRs, respectively, indicating a net production of alkalinity (Figure B.27).   The H17 DO concentration was significantly higher in the open PBR at 8.9 ± 1.3 mg/L compared to 6.2 ± 0.8 mg/L in the closed PBR (p < 0.001)(Figure B.17). DO concentrations reached zero during the dark period of the continuous diel test in both PBRs (Figure 5.5).     48  Figure 5.5 - Effluent DO, pH, and alkalinity levels through the “continuous” diel monitoring experiment conducted at the end of Phase B. The dark period is denoted by the shaded area.  The DO in both PBRs dropped to 0 mg/L in the dark phase of the batch diel test as well (Figure 5.6), and as such, net denitrification rates were observed (Figure 5.7). Unlike in Phase A, TN concentrations through the dark phase did not reduce faster than nitrate concentrations, indicating simultaneous nitrification/denitrification was likely not taking place (Figure 5.6). This is likely due to elevated COD concentrations in the influent, resulting in an initial COD concentration of roughly 225 mg/L after dilution from feeding (Figure 5.6), which is 100 mg/L Open Closed●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●● ●●01020304001002003004000 5 10 15 20HourDO & pH●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●01020304001002003004000 5 10 15 20HourAlkalinityParameter ●ALK  ( mg/L ) DO  ( mg/L ) pH  ( NA )Experiment Hour  49 higher than in Phase A (Figure 5.2). COD removal was not completed in the dark phase in the closed PBR as it was in Phase A (Figure 5.6).  Figure 5.6 - Concentrations of key parameters during the diel monitoring study at the end of Phase B with the PBRs operating under the “batch” feed condition.  Open Closed●● ● ● ● ● ● ● ●02040600 5 10 15 20●● ● ● ● ● ● ● ● ●02040600 5 10 15 20Parameter●TN  ( mg/L )TKN.F  ( mg/L )NH3  ( mg/L )NO2  ( mg/L )NO3  ( mg/L )●● ● ● ● ● ● ● ● ●01020300 5 10 15 20●● ● ● ● ● ● ● ● ●01020300 5 10 15 20Parameter●DO  ( mg/L )Temp  ( *C )pH  (  )●●●●● ● ●●●● ●1002003000 5 10 15 20●●●● ●●●● ●● ●1002003000 5 10 15 20Parameter● COD  ( mg/L )ALK  ( mg/L )24680 5 10 15 2024680 5 10 15 20ParameterTP.F  ( mg/L )Experiment Hour-N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L )   50  Figure 5.7 - Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase B. The dark period is denoted by the shaded area.  Even with the significant difference in H17 DO concentrations, overall nitrification rates were statistically similar at 0.90 ± 0.21 and 0.90 ± 0.19 gN/m2/day in the open and closed PBRs, respectively (p = 0.99)(Figure B.22), and nitrification represented 72 ± 4% and 77 ± 5% of TKN removal (Figure 5.8). Denitrification rates were likewise statistically similar at 0.57 ± 0.26 and 0.62 ± 0.19 gN/m2/day in the open and closed PBRs, respectively (p = 0.36)(Figure B.21) and denitrification represented 61 ± 6% and 72 ± 4% of TN removal, respectively (Figure 5.8). Biomass assimilation rates were significantly higher in the open PBR, at 0.32 ± 0.02 gN/m2/day compared to 0.23 ± 0.02 gN/m2/day in the closed PBR (p < 0.001)(Figure B.23). The net result was no significant difference in the effluent TKN (p = 0.23)(Figure B.24) or TN removal (p = 0.62)(Figure B.25) between the two PBRs. Effluent NO3-N concentrations averaged 9.8 ± 5.1 and 7.7 ± 4.5 mg/L in the open and closed PBRs, respectively (Figure B.7). Effluent NH3-N concentrations averaged 4.6 ± 6.6 and 6.8 ± 10 mg/L, respectively (Figure B.5).   ●●●● ●● ● ●−2025112241710 5 10 15 20HourdNO3 (g/m2/day)Parameter●OpenCloseddNO3-N (gN/m2/day)   51  It is worth noting however that the influent used for the second half of Phase B was far more concentrated, with a COD of 569 ± 237 mg/L, compared to just 150 ± 84 mg/L in the first half of Phase B (Figure B.3). When only looking at the first two weeks of Phase B, nitrification rates (p = 0.02)(Figure B.22) and denitrification (p = 0.004)(Figure B.23) rates were significantly higher in the closed PBR and effluent ammonia levels were 0.32 ± 0.84 mg/L and 0.57 ± 1.2 mg/L (Figure B.1).    Figure 5.8 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase B at hour 17.  0.000.250.500.751.0050 60DayOpen0.000.250.500.751.0050 60DayClosedkeyNorgNH4BiofilmNO2NO3N2Fraction of Influent TN  52 5.2.2.3 Phase C: Batch Feed, Alkalinity Addition In Phase C, the PBRs were batch fed at the outset of the unlit period in order to test whether denitrification could be promoted using the alternative feed regime. The pH remained stable during Phase C at 7.8 ± 0.5 and 7.3 ± 0.1 in the open and closed PBRs, respectively (Figure B.18). The H17 alkalinity was 114 ± 66 and 191± 76 mg/L (Figure B.1), translating to specific alkalinity demands of 0.58 ± 0.40 gCaCO3/gN and -0.44 ± 0.45 gCaCO3/gN in the open and closed PBRs, respectively (Figure B.27). The H17 DO concentration was significantly lower in the closed PBR at 2.8 ± 0.5 mg/L, while the open PBR remained relatively oxygen-rich at 6.4 ± 0.7 mg/L (p < 0.001) (Figure B.17). DO concentrations reached zero in both PBRs during the dark period of the batch diel study though the DO level in the closed PBR remained near zero through lit periods as well (Figure 5.9).     53  Figure 5.9 - Effluent DO, pH, and alkalinity levels through the “batch” diel monitoring experiment conducted at the end of Phase C. The dark period is denoted by the shaded area.  Net denitrification was observed in both PBRs in the batch diel monitoring experiment (Figure 5.11) and almost all COD removal took place before the end of the unlit period, despite similar initial COD concentrations to Phase B (Figure 5.10). Simultaneous nitrification and denitrification was observed in both PBRs during the unlit period as nitrate levels were near zero and photosynthesis had ceased, yet TN levels steadily declined (Figure 5.10).  Open Closed●●●●● ●●●●● ●01020304001002003004000 5 10 15 20HourDO & pH●●●● ● ● ● ● ●● ●01020304001002003004000 5 10 15 20HourAlkalinityParameter ●ALK  ( mg/L ) DO  ( mg/L ) pH  ( NA )Experiment Hour  54  Figure 5.10 - Concentrations of key parameters during the diel monitoring study at the end of Phase C with the PBRs operating under the “batch” feed condition.   Open Closed● ● ● ● ● ● ● ●03060900 5 10 15 20● ● ● ● ● ● ● ●03060900 5 10 15 20Parameter●TN  ( mg/L )TKN.F  ( mg/L )NH3  ( mg/L )NO2  ( mg/L )NO3  ( mg/L )● ●● ● ● ● ● ● ● ● ●01020300 5 10 15 20●●● ● ● ● ● ● ● ● ●01020300 5 10 15 20Parameter●DO  ( mg/L )Temp  ( *C )pH  (  )●● ●● ● ● ●●●1002003004000 5 10 15 20●●● ● ● ● ●● ●1002003004000 5 10 15 20Parameter● COD  ( mg/L )ALK  ( mg/L )68100 5 10 15 2068100 5 10 15 20ParameterTP.F  ( mg/L )Experiment Hour-N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L ) -N ( mg/L )   55  Figure 5.11 – Observed rate of change in nitrate-N concentration (gN/m2/day) in the batch diel monitoring experiment at the end of Phase C. The dark period is denoted by the shaded area.  Due to higher DO levels in the open PBR, the nitrification rate was significantly higher in the open PBR at 1.51 ± 0.53 gN/m2/day compared to 1.41 ± 0.51 gN/m2/day in the closed PBR (p = 0.009)(Figure 8.22), and nitrification represented 81 ± 5% and 84 ± 4% of TKN removal (Figure 5.12). Denitrification rates were statistically similar at 1.30 ± 0.56 and 1.37 ± 0.49 gN/m2/day (p = 0.41)(Figure 8.21), and denitrification represented 77 ± 7% and 83 ± 4% of TN removal, respectively (Figure 5.12). Assimilation rates remained similar to Phase B, with the open PBR producing significantly more biomass at 0.33 ± 0.03 gN/m2/day compared to 0.23 ± 0.01 gN/m2/day in the closed PBR (p < 0.001) (Figure B.23). TN removal efficiency was similar in both PBRs at 77 ± 9 % and 76 ± 8 %, respectively (p = 0.64)(Figure B.25), although the TKN removal was significantly higher in the open PBR at 93 ± 5 % versus 78 ± 6 % (p = 0.001)(Figure B.24).   ●●● ● ● ●●−1.1−0.8−0.5−0.20.00.20.50.80 5 10 15 20HourdNO3 (g/m2/day)Parameter●OpenCloseddNO3-N (gN/m2/day)   56 Effluent NO3-N concentrations averaged 9.8 ± 5.7 and 0.63 ± 1.3 mg/L in the open and closed PBRs, respectively (Figure B.7). Effluent NH3-N concentrations averaged 9.1 ± 11 and 18 ± 12 mg/L, respectively (Figure B.5).     Figure 5.12 - Fractional speciation of nitrogen leaving the open and closed PBRs in Phase C at hour 24.  5.2.2.4 Continuous and Batch Feed Regime Comparison The use of real wastewater with fluctuating characteristics made it difficult to compare the performance of the PBRs between experimental phases and to comment on optimal operating conditions. However, significantly higher denitrification rates were observed under the batch 0.000.250.500.751.0075 80 85 90 95DayOpen0.000.250.500.751.0075 80 85 90 95DayClosedkeyNorgNH4BiofilmNO2NO3N2Fraction of Influent TN  57 feed regime in Phase C than under the continuous feed regime in Phase B in both the open (p = 0.019) and closed (p = 0.009) PBRs (Table 5.5). However, the overall TN removal efficiency did not significantly differ in the open (p = 0.094) or closed (p = 0.21) PBRs (Table 5.3). This may have been caused by the higher influent TN concentration in Phase C than in Phase B, at 94 ± 30 mgN/L and 64 ± 24 mgN/L, respectively (Figure B.12).  5.2.3 Biomass Production The biofilm effectively self-adhered to the sandpaper surface at the flow velocity (0.9 mm/s) used in this study. Some sloughing was observed, but was mostly a result of the harvesting regime due to the short solids retention time used in the study (Boelee et al., 2011). Without alkalinity addition in Phase A, biomass production in the closed PBR was significantly higher, although that pattern reversed with the addition of alkalinity in Phases B and C (Figure B.23). Maximum production was observed in the open PBR in Phase C at 7.3 ± 2.4 g/m2/day (Figure C.5) and a photosynthetic efficiency of 5.8 ± 1.9 % (Figure C.7).  Biomass production averaged 4.5 ± 1.1 and 3.7 ± 0.6 g/m2/day in the open and closed PBRs, respectively, over the course of the three phases (Figure C.5). Chlorophyll density was 530 ± 129 and 428 ± 61 mg/m2 in the open and closed PBRs, respectively (Figure C.2). Photosynthetic efficiency was 3.5 ± 0.8 % and 2.9 ± 0.5%, higher than the 1.5% observed by Posadas et al., 2013 (Figure C.7). Nitrogen content of the biomass fluctuated slightly, averaging 6.3 ± 3.5% and 6.9 ± 4.7% in the open and closed PBRs, respectively, through four samples (Table 6.11). Carbon content of the biomass fluctuated as well, averaging 34 ± 22% and 35 ± 34% in the open and closed PBRs, respectively, through four samples (Table 5.6).    58  Table 5.6 - Average characteristics of biomass harvested from the open and closed PBRs through three experimental phases.  Ai Bi Ci Open Closed Open Closed Open Closed Dry Weight (2/day) 2.5 ± 1.2 3.6 ± 1.6 4.5 ± 1.0 4.1 ± 1.0 7.3 ± 2.4 3.8 ± 0.8 Photosynthetic Efficiency (%) 2.0 ± 0.9 2.8 ± 1.3 3.5 ± 0.8 3.2 ± 0.8 5.8 ± 1.9 3.0 ± 0.6 Dry/Wet Ratio 0.06 ± 0.02 0.06 ± 0.01 0.06 ± 0.01 0.07 ± 0.01 0.09 ± 0.01 0.09 ± 0.01 Chlorophyl A (mg/m2)   429 ± 81 346 ± 58 350 ± 173 351 ± 130 Chlorophyl B (mg/m2)   169 ± 109 91 ± 18 88 ± 64 65 ± 41 Cartenoids (mg/m2)   206 ± 48 147 ± 32 180 ± 76 153 ± 62 %N 7.4 9.0 6.7 5.4 4.7 6.3 %C 42.3 51.0 35.7 26.8 24.4 27.4  5.2.4 Microbial Community Composition of Microalgal Biofilm A sample of biomass was collected after each experimental phase for DNA extraction and 16S rRNA gene amplicon sequencing to explore the impact of PBR operating conditions on the microbial community structure. After quality-filtering and denoising, the number of sequences retained per sample averaged 35,900 ± 4,800.   Throughout all experimental phases, bacterial 16S rRNA genes were overwhelmingly attributed to the Proteobacteria and Cyanobacteria phyla, which together comprised over 85% of reads in all samples (Figure D.1). The dominant cyanobacterial genus in the PBRs was Tychonema, which comprised up to 87% of 16S rRNA reads (Figure 5.13).                                                i The n value for each parameter varies slightly due to missing data and can be found in Appendix C with the raw data. g/m   59   Figure 5.13 - Relative abundance (%) of the 12 most abundant genera observed in triplicate at the end of the startup and three experimental phases, based on the results of the 16S rRNA gene amplicon sequencing.  The PBRs were operating under identical conditions during the startup phase, and were dominated by the same three genera at the start of Phase A (day 0) despite some difference in their relative abundances (Figure 5.13). STAMP taxonomic analysis showed that the greatest number of statistically significant differences between genera in the two PBRs were present at the end of Phase A (Figure D.4), coinciding with the observation of an alkalinity limitation (Figure 8.7) and consequent low pH (Figure B.18), particularly in the open PBR.   0 28 64 93ClosedOpen1 2 3 1 2 3 1 2 3 1 2 302550751000255075100RepRelative Abundance (%)GenusSM1A02HolophagaSimplicispiraNovosphingobiumRudaeaChryseobacteriumRhodanobacterAeromonasCandidatus_ParacaedibacterJanthinobacteriumYersiniaTychonema_CCAP_1459−11BDay  60  A wide variety of genera were present in the open and closed PBRs over the 93-day experimental period. One metric for assessing microbial community composition is the Shannon Index (SI). Normality of the SI could not be confirmed as data was only available in triplicate. However, t-tests were conducted to supplement a visual comparison of community diversity between experimental phases. The SI of the microbial community in the open PBR increased significantly between the day 0 and day 28 (p = 0.008), decreased significantly between day 28 and day 64 (p < 0.001), and did not significantly change between day 64 and day 93 (p = 0.11) (Figure 5.14).  On the other hand, the SI did not significantly change between any experimental phases in the closed PBR (p = 0.43, p = 0.47, p = 0.43, respectively)(Figure 5.14).    61  Figure 5.14 - Shannon Index of the relative abundance of genera present in the biomass samples taken at the end of the startup phase and each of the three experimental phases in the open (right bars) and closed (left bars) PBRs.  Principle component analysis (PCA) showed that both PBRs were closely clustered at the end of Phase B and C, phases in which alkalinity was added (Figure 5.15). This indicates that the alkalinity limitation may have been the primary source of variability in the microbial community for both the first, (x-axis, 30% explained variance), and second (y-axis, 12% explained variance) principle components.  ●●●●●●●●●●●●●●●●●●●●●●●●1.01.52.02.53.00 28 64 93DayShannon Diversity IndexGroup●●ClosedOpen  62  Figure 5.15 - Principle coordinate analysis of the relative abundance of genera present at the end of the startup phase and each of the three experimental phases. X-axis (30 % explained variance), y-axis (12 % explained variance).  Differential abundance analysis with STAMP on all of the samples taken from the open and closed PBRs revealed that the most biologically relevant and statistically significant difference to persist over the 93-day experimental period was the increased abundance of the genus Rhodoferax in the closed PBR (Figure 5.16), measured in relative abundances of up to 0.3% and 2.5% in the open and closed reactors respectively. Rhodoferax are known denitrifiers and mixotrophs in full scale activated sludge reactors (McIlroy et al., 2016). Other genera that were present in statistically different abundances between the two PBRs included elevated levels of ●●●●●●PCA sample plot−60 −30 0 30 60−40040xyGroup●●ClosedOpenDay● 0286493  63 Flavihumibacter and Dinghuibacter in the closed PBR and elevated levels of Polaromonas and Defluviimonas in the open PBR (Figure 5.16).   Figure 5.16 - STAMP analysis comparing relative abundance of genera in the open and closed PBRs over the course of the entire 93-day experiment. Bars to the left and right of center indicate significantly greater abundance in the open and closed PBRs, respectively. The values on the far right indicate the p value for comparing the relative abundance of the genera between the two conditions.   64  Among genera with no significant different in abundance between the two reactors, a number have been identified as known AOB’s and denitrifiers. Of the known genera of AOB’s, only Nitrosomonas was observed and at very low abundances (<0.1%). Thauera terpenica is a known denitrifying species (Foss and Harder, 1998) and was detected, particularly in Phase B and C, at abundances of up to 0.2% of reads. Zoogloea ramigera, a common bacteria in activated sludge treatment associated with floc formation (Rossello-Mora et al., 1995), was also identified at low abundances (<0.1%). Members of the Zoogloea genera are also suspected denitrifiers as they are often present in WWTP’s with denitrification (Rossello-Mora et al., 1995), and individual species have been isolated and shown to perform denitrification (Lv et al., 2017). Other genera known for denitrifying species were also detected, including Pseudomonas, Paracoccus, and Flavobacterium (Seviour and Nielsen, 2010). No known NOB’s were detected.      65 Chapter 6: Discussion 6.1 Microscreen Pretreatment The microscreen pretreatment in this study, even in operation without a cake, exhibited significantly higher TSS and TS removal rates (70 ± 6% and 38 ± 9%, respectively) than what could be expected from a primary clarifier or microscreen in operation at a centralized wastewater treatment facility. In decentralized wastewater, solids may not be subjected to the hydraulic shear and microbial hydrolysis that takes place in a sewer during conveyance to a centralized treatment facility (Roni et al., 2019). The elevated solids removal rates of microscreens applied to decentralized wastewater in this study may thus be attributed to a higher fraction of total solids present in the suspended (particulate) form than what could be expected for influent entering a centralized treatment facility. Additional sampling of particle size distributions throughout a sewer collection network would be needed to verify this conclusion. Nonetheless, the removal rates for TS (38 ± 8.8%), COD (39 ± 9.1%), TP (30 ± 18%), and TN (13 ± 10%) found in this study were largely in agreement with the size distributions presented in a study by Todt et al., (2015), in which the proportions of nutrients present at varying size fractions in dormitory wastewater were investigated (Figure 2.4).  6.2 Biofilm PBR Biological Treatment 6.2.1 Comparative Performance of Open and Closed PBRs Nitrification and denitrification proved to be the dominant TKN and TN removal pathways, respectively, in both the open and closed PBRs in all three experimental phases. When the alkalinity limitation was removed in Phase B, it was expected that lower DO conditions in the closed PBR relative to the open PBR would have an inhibitory effect on nitrification, resulting in   66 lower TKN removal rates, but, conversely, encourage denitrification and overall TN removal. However, observed nitrification and denitrifications rates were both similar between the open and closed reactors in Phase B. H17 DO in the open and closed PBRs (8.9 ± 1.3 and 6.2 ± 0.8 mg/L, respectively) remained above the reported inhibitory DO levels for nitrifying biofilms of 4 mg/L (Nogueira et al., 1998) while the continuous diel test at the end of Phase B showed DO levels in both PBRs falling below the accepted threshold for effective denitrification, 0.5 mg/L (Figure 5.5)(Metcalf and Eddy et al., 2013). With that said, denitrification rates in the closed PBR were actually higher than in the open PBR until the last two effluent measurements of Phase B, where denitrification performance in the open PBR improved (Figure B.21). This was likely a result of elevated COD concentrations in the influent in the latter half of Phase B (Figure B.3) which caused a reduction in the DO levels in the open PBR (Figure B.17), thus promoting anoxic conditions in its biofilm and a rapid denitrification of accumulated nitrate (Figure B.7). It can be concluded that while DO levels tended to be lower in the closed PBR, differences in the nitrogen removal pathways between the open and closed PBRs were dependent on COD loading.  In Phase C, it was hypothesized that batch feeding at the start of the unlit period would promote denitrification in both PBRs by encouraging a period of high carbon, low DO conditions. The denitrification rates observed under batch feed conditions in Phase C (1.30 ± 0.56 and 1.37 ± 0.49 gN/m2/day in the open and closed PBRs, respectively) were significantly higher than those observed in under continuous feed conditions in Phase B (0.57 ± 0.26 and 0.62 ± 0.19 gN/m2/day in the open and closed PBRs, respectively). These results are consistent with previous research that showed batch feeding under anoxic conditions improved denitrification rates in suspended microalgal-bacterial systems (Foladori et al., 2018).   67 6.2.2 Other Factors Influencing PBR Performance  Overall, varying the open/closed condition significantly affected redox conditions within the PBRs, which impacted nitrogen removal efficiencies. However, this and other studies have shown that under the right conditions, both open and closed PBRs have the capacity to achieve effective nitrification and denitrification. Four critical parameters in the determination of these conditions are influent feed regime, lighting regime, COD loading rate, and alkalinity concentration.  Although simultaneous nitrification and denitrification has been observed in PBRs under oxic conditions (Foladori et al., 2018) and under anoxic conditions in this study (Figure 5.2), temporally separated oxic and anoxic periods fed intermittently have been shown to improve total nitrogen removal efficiency in complete mix PBRs (Foladori et al., 2018). The biofilm PBRs in this study were shown to behave similarly when operated in Phase C with the same intermittent lighting and feed regime as Foladori et al., (2018) and achieved significantly higher denitrification rates than in Phase B, where the PBRs were operated with a continuous feed regime (Table 5.5). However, lighting and feed regimes alone are not enough to ensure oxic and anoxic conditions, COD loading and photosynthetic oxygen production must be balanced as well. When COD loading is too high and irradiance is too low, oxic conditions would not be created and nitrification would be inhibited. When COD loading is too low or irradiance too high, anoxic conditions would not be created and denitrification would be inhibited. In addition, sufficient alkalinity must be present as a precursor for nitrification to take place. Both of these processes are necessary in order to meet restrictive urban reuse guidelines requiring roughly 95% ammonia removal and 85% TN removal (Table 2.1).   68  Operation of the closed PBR in Phase A was similar to the experiment by Boelee et al., (2014) with continuous feed, no alkalinity addition, and a similar effective HRT (Table 6.1).  However, the experiment presented here showed higher TN and TKN removal efficiencies than in Boelee et al., (2014) (Table 6.1). The elevated TKN removal observed in this study relative to those reported by Boelee et al. (2014) may be related to lower COD loading (5.4 gCOD/m2/day here compared to 56 gCOD/m2/day in Boelee et al. (2014)) (Table 6.1). DO levels were around 0 mg/L for the whole study in Boelee et al. (2014), likely limiting nitrification, whereas the DO concentration during the lit period here was 7.2 ± 0.8 mg/L (Figure B.17). This study also used real wastewater with alkalinity (188 ± 135 mg/L CaCO3 in Phase A (Figure B.1)) whereas HCO3- was only added to the synthetic influent for the first six days of the experiment by Boelee et al., (2014). Nitrification by nitrifying biofilms has been shown to be inhibited at CaCO3 concentrations less than 45 mg/L (Biesterfeld et al., 2003). Elevated denitrification rates and TN removal efficiency in this study were likely a result of the alternating oxic and anoxic periods produced by the 16 hr-8 hr light-dark cycle and moderate COD loading in this study. In contrast, continuous lighting was used in Boelee et al., (2014) and even so, oxic conditions were never produced.  pH instability was not observed in Boelee et al., (2014) as pH control was exercised using recycle line pH monitoring and automatic adjustment of influent acetate and acetic acid concentrations.   The operation of the open PBR in Phase A was similar to the experiment performed by Posadas et al., (2013) with continuous feed, no alkalinity addition, a similar effective HRT, and also in this case, a 16hr-8hr light-dark cycle and real wastewater (Table 6.1). However, the open PBR in   69 this study did not achieve 100% TKN removal whereas the PBR in Posadas et al. (2013) did, likely through the use of an influent with even lower COD loading of roughly 1.1 gCOD/m2/day compared to 5.4 gCOD/m2/day here (Table 6.1). Posadas et al., (2013) also experienced pH instability, as a result of the high rate of nitrification rate and a limited denitrification rate; however alkalinity was not a limiting factor, as TKN was completely removed (Table 6.1).   Table 6.1 - Comparison between experimental conditions and results observed in existing literature and this study. (O/C) Open/Closed.  Reference O/C Light Eff. HRT HRT Influent Pretreat Influent  Removal Efficiency RCOD  RO2   On-Off PAR hrs hrs  COD mg/L TN mg/L PO4 mg/L COD (%) TN (%) TKN (%) PO4 (%) g/m2/day g/m2/day (Zamalloa et al., 2013) O 16-8 213 1.8  24 Settling, Adsorption 61 ±27 36 ±9 1.7 ±0.9 viii 39 ±18 69 ±14 91 ±20 94 ±8 8.4 ±4 12 ±11 (Posadas et al., 2013) Phase III O 16-8 88 6.0 74 Screening,Settling 181 ±69 ix 91 ±14 7 ±3 90 ±3 ii 72 ±13 100 85 ±9 1.1 ±0.4 4.9 (Posadas et al., 2013) Phase IV O 16-8 88 10 125     86 ±3 ii 59 ±11 100 57 ±17 2.2 ±0.8  (Posadas et al., 2013) Phase V O 16-8 88 20 250     86 ±6 ii 54 ±8 100 36 ±22 3.6 ±1.4  (Boelee et al., 2014) C 24 340 4.5 4.5 Synthetic 350 50 10 77 ±8 40 ±6 40 ±6 25 ±3.5 56 28 This Study Phase A O 16-8 150 4.9 54 Screening 220 ±61 62 ±13 4.5 ±1.5 71 ±4 37  ±8 69 ±4 45 ±10 5.4 ±1.5 8.3 C        77 ±4 52  ±4 76 ±4 46 ±7   This Study Phase B O 16-8 150 4.9 54 Screening 401 ±198 64 ±24 4.4 ±2.4 75 ±8 66  ±11 88 ±9 54 ±14 9.8 ±4.8 8.3 C        81 ±7 67  ±13 85 ±15 46 ±14   This Study Phase C O 16-8 150 4.2 46 Screening 329 ±45 94 ±30 7.5 ±3.6 83 ±3 77  ±9 93 ±5 44 ±24 9.5 ±1.3 8.3 C        81 ±2 76 ±8 78 ±6 43 ±25                                                  viii TP data used in lieu of unavailable PO4 data ix TOC data used in lieu of unavailable COD data  A potentially useful metric in describing the optimal COD loading rate to biofilm PBRs could be the oxygen loading ratio (OLR) (Equation 6.1), the ratio of oxygen demand (RCOD) to oxygen production by algae (RO2) (Equation 6.2):  Equation 6.1 – Oxygen loading ratio  !"# = #%&'#&(  Equation 6.2 – Oxygen production rate by photosynthesis  #&( = )*&( ∙ , ∙ - ∙ .&(  where: OLR is the oxygen loading ratio (gCOD/gO2), RCOD is the COD loading rate (g/m2/day), RO2 is the oxygen production rate by photosynthesis (g/m2/day), QYO2 is the quantum yield of oxygen evolution, T is the length of the daily irradiation period (s), I is the photon flux density (mol photons/m2/s), and MO2 is the molar mass of oxygen (g/mol). . This parameter was calculated for all biofilm PBRs used to treat domestic wastewater reported thus far. In studies in which the quantum yield was not directly measured, it was estimated to be 0.03 mol O2/mol photons as was measured in Boelee et al., 2011. When the oxygen loading ratio was graphed against TKN and TN removal efficiency, it illustrated that the performance of the open and closed PBRs follow similar patterns, though separated by a linear translation (horizontal shift) along the x-axis (Figure 6.1). The size of this translation is potentially proportional to the atmospheric aeration rate in the open reactor such that if it were added as a term in the denominator of the oxygen loading ratio, the trendlines would overlap. With that said the atmospheric aeration rate is dependent on the DO concentration and therefore varies with   72 OLR and diel fluctuations, making it difficult to estimate. TKN removal efficiency is inversely correlated to the oxygen loading ratio, as lower COD loading rates result in more oxic conditions and better nitrification (Figure 6.1). TN removal efficiency on the other hand, appears to follow a more complex pattern (Figure 6.1).  As TKN removal declines at high COD loading rates, so does TN removal because denitrification becomes limited by the lower amount of NOx produced via nitrification (Figure 6.1). At lower COD loading rates, DO levels would increase and denitrification rates would become limited by the presence of oxygen (Figure 6.1). However, at even lower loading rates, the slower process of assimilation into biomass, may begin to have a more significant effect on TN removal. In general, the TN removal was inversely correlated to the oxygen loading ratio in the closed system (Figure 6.1), indicating that limitations in TKN removal at high relative COD loadings are the major bottleneck for TN removal in the closed PBRs. However, in open PBRs, the oxygen loading ratio had minimal impact on TN removal over the range of conditions examined in this study and in previous literature (Figure 6.1). The higher TKN removal rates allowed at higher oxygen loading rates in the open reactors may thus help to promote TN removal under such conditions by supplying adequate NOx for denitrification. The above analysis assumes a relatively standard COD/TN ratio found in domestic wastewater.     73  Figure 6.1 – TKN (a) and TN (b) removal efficiency graphed as a function of the oxygen loading ratio for this and other major studies investigating wastewater treatment with open (•) and closed (■) PBRs. The first two weeks and second two weeks of Phase B were plotted separately (B1 and B2) due to elevated COD influent concentrations in the second half of the phase (Figure B.3). The results from Phase A were excluded from the graph because TKN removal was alkalinity limited (Figure B.1).  The above approach can be used as a rule of thumb design tool for selecting COD loading and irradiation rates for open and closed PBRs with a particular effluent TKN target. For instance, to a) b)   74 reach 95% TKN removal, the oxygen loading ratio in open and closed PBRs should be roughly 0.4 gCOD/gO2 and 0.8 gCOD/gO2, respectively (Figure 6.1).   6.2.3 Microbial Dynamics The microbial communities in the two PBRs varied over the course of the 93-day experiment. PCA showed no variability on the two principle components once alkalinity was added in Phase B, indicating that the lack of alkalinity addition in the start-up phase and Phase A was the primary source of variability in the microbial community throughout the experiment (Figure 5.15). Analysis of the Shannon Index (SI), a measure of community composition, showed that the SI varied significantly between consecutive phases in the open PBR, Start-up to Phase A and Phase A to Phase B, when alkalinity was not added in both of the consecutive phases (Figure 5.14). The SI in the closed PBR on the other hand did not vary significantly between operational phases (Figure 5.14). This may be explained by the Anna Karenina effect, in which dysbiotic or environmentally stressed individuals (i.e. PBRs) vary more in microbial community composition in healthy individuals (i.e. PBRs) (Zaneveld et al., 2017). The variability of the SI in the open PBR without alkalinity addition, indicated that it was in a stochastic, unstable state, potentially due to environmental stress. The SI in the closed PBR on the other hand, was statistically similar from one experiment to the next, even in the absence of alkalinity addition. This is likely due to the fact that the closed PBR demonstrated a more consistent capacity for denitrification, and thereby self-sufficient alkalinity production. If operation under alkalinity-limited conditions is necessary, it may be the case that higher COD loading or closed operation could be used to limit alkalinity demand and promote a more stable treatment process.    75 The 4 most common phyla observed in the PBRs in this study: Cyanobacteria, Proteobacteria, Bacteriodetes, and Acidobacteria (Figure D.1), have all been idenitified in large numbers in raw influent, activated sludge, and effluent at a municipal treatment facility (Ye and Zhang, 2013). Proteobacteria and Bacteriodetes were also dominant in the PBR run by Krohn-Molt et al., (2013); however no Cyanobacteria were reported in that study, where inoculation was done specifically with chlorophytes and influent N was in the form of nitrate. Nitrogen in the form of ammonia and urea has been shown to favour non-heterocystous cyanobacteria and chlorophytes, whereas nitrate promoted chlorophytes and only some cyanobacteria, while nitrogen-fixing cyanobacteria showed little response to influent N (Donald et al., 2011). The cyanobacterial genera identified in this study, Tychonema, is not known for heterocystic species (Rudi et al., 1998), and may have been promoted by the form of nitrogen present in the domestic wastewater used in this study. A microbiome analysis of microalgal cultures present in suspended-growth PBRs fed with municipal wastewater have shown that Chlorophytes are present, and Proteobacteria and Bacteriodetes are the dominant bacterial phyla over Cyanobacteria (Carney et al., 2014; Krustok et al., 2015). However, Tychonema has also been identified as a major mat-building genera in benthic cyanobacterial mats on coral reefs (Brocke et al., 2018). Therefore,  Tychonema may be uniquely adapted specifically for wastewater treatment in biofilm PBRs due to their niche role in biofilm (i.e. mat) formation. A species of the Tychonema genera has been shown to be producer of anatoxin-a, a powerful toxin that poses a health risk for humans when present in drinking water sources (Shams et al., 2015). However, anatoxin-a does degrade under normal environmental conditions, including exposure to natural light at a half-life of 150 minutes (Stevens and Krieger, 1991), so its presence in PBR effluent may be naturally limited.     76 Several genera containing known AOB’s and denitrifiers were detected, but only one (Rhodoferax) was in statistically different abundance between the open and closed PBRs, despite the significant differences observed in nitrogen conversion processes. Rhodoferax is a genera known for denitrifying species capable of growth in municipal wastewater treatment bioreactors (McIlroy et al., 2016), and had a significantly higher abundance in the closed versus the open PBR. This finding is consistent with PBR performance, as the closed PBR consistently showed lower DO levels. Thauera, another known denitrifying genera (Foss and Harder, 1998) that was detected in similar abundance in both systems, was identified in a study of 17 WWTP as the most versatile at substrate consumption for denitrification, consuming VFA’s, ethanol, and amino acids (Thomsen et al., 2007).  Nitrosomonas was the only AOB observed, albeit at low abundances, and no known NOB were observed. AOB and NOB are often present in very low numbers, due to their slow growth rate, low biomass yields, and environmental sensitivity. In healthy nitrifying environments, nitrifying fractions have varied from 0.39% in activated sludge (Dionisi et al., 2002) to 18% in a combined activated sludge/rotating biological reactor (You et al., 2003). In some cases, targeted assays are necessary for the detection of Nitrosomonas (Silyn-Roberts and Lewis, 2001). However, given the high rate of nitrification observed in this study, ammonia oxidation may also have been carried out by ammonia oxidizing archaea (AOA). AOA are gaining recognition for their role in wastewater treatment systems, and increased environmental adaptability compared with AOB’s (Yin et al., 2018). The low oxygen levels and pH variability observed over the course of this experiment may have created a selective pressure for the more environmentally adaptable AOA’s. Unfortunately, the primers used in this study had <5% coverage for the Crenarchaeota   77 phyla from which AOA’s are delineated, as confirmed by an in silico PCR of the SILVA r132 database (Klindworth et al., 2013) (data not shown).      78 Chapter 7: Conclusion In summary, this study demonstrated the following key findings: • Microscreens are an effective means of primary treatment of decentralized wastewater. The 54 µm microscreen used in this study performed better than in previous studies using centralized municipal wastewater, removing 70 ± 6% and 38 ± 9% of SS and TS, respectively, and it did so without the formation of a cake.  • The denitrification rate increased significantly in both the open and closed PBRs when operating in a batch feeding regime (during the unlit period) versus continuous feeding. Therefore, the feeding pattern of biofilm PBRs could be adjusted to satisfy desired effluent TN limits.  • The closed reactor had higher relative abundances of Rhodoferax, which are putative denitrifiers in wastewater treatment systems. This higher relative percentage of Rhodoferax was supported with higher fractions of TN removal via denitrification in the closed PBR. • Alkalinity limitation was a large driver of microbial community structure. The stable level of diversity (Shannon Index) in the closed PBR indicates that self-generation of alkalinity via denitrification could modulate the microbial dynamics, providing more stable long-term performance. • The oxygen loading ratio was identified as a potential metric for predicting TKN removal efficiency in the design of new PBRs. For 95% TKN removal, an OLR of 0.4 gCOD/gO2 and 0.8 gCOD/gO2 is recommended in open and closed PBRs, respectively, with a   79 baseline alkalinity concentration of 1-2 gCaCO3/gTN. Greater OLR’s will result in declining TKN removal but more complete denitrification of generated NO3.    7.1 Recommendations The following recommendations for future research are made: • The findings in this study indicated that microscreening is more efficient when applied closer to the wastewater source, however the comparisons between microscreen efficiency for decentralized and centralized wastewater in this study relied on literature for centralized data. It may be deemed appropriate to perform sampling and treatment at varying degrees of decentralization throughout a single sewer collection network within a single study to verify this relationship.   • In order to verify the validity of the oxygen loading ratio as a metric in predicting biofilm PBR TKN and TN removal rates, parallel open and closed PBRs should be run with synthetic wastewater (to reduce variability) at randomly selected COD concentrations and then again at randomly selected irradiances.   • Metagenomic analysis, in addition to amplicon sequencing as conducted here, could be used as a tool to investigate variations in the metabolic capacity of biofilm PBRs, to create a more comprehensive understanding of the microbial communities present and elucidate the impacts of reactor design on the presence of different nitrogen conversion pathways.   80 References Ahn, K.-H., Song, K.-G., 2000. Application of microfiltration with a novel fouling control method for reuse of wastewater from a large-scale resort complex. Desalination 129, 207–216.  Ahn, K.-H., Song, K.-G., Yeom, I.-T., Park, K.-Y., 2001. Performance comparison of direct membrane separation and membrane bioreactor for domestic wastewater treatment and water reuse. Water Sci. Technol. Water Supply 1, 315–323.  American Society of Civil Engineers, 2013. Failure to Act: The Impact of Current Infrastructure Investment on America’s Economic Future. American Society of Civil Engineers, Reston, VA.  BC Ministry of Environment, 2013. Reclaimed Water Guideline.  Biesterfeld, S., Farmer, G., Russell, P., Figueroa, L., 2003. Effect of Alkalinity Type and Concentration on Nitrifying Biofilm Activity. Water Environ. Res. 75, 196–204.  Boelee, N.C., Temmink, H., Janssen, M., Buisman, C.J.N., Wijffels, R.H., 2014. Balancing the organic load and light supply in symbiotic microalgal–bacterial biofilm reactors treating synthetic municipal wastewater. Ecol. Eng. 64, 213–221.  Boelee, N.C., Temmink, H., Janssen, M., Buisman, C.J.N., Wijffels, R.H., 2011. Nitrogen and phosphorus removal from municipal wastewater effluent using microalgal biofilms. Water Res. 45, 5925–5933.  Brocke, H.J., Piltz, B., Herz, N., Abed, R.M.M., Palinska, K.A., John, U., Haan, J. den, de Beer, D., Nugues, M.M., 2018. Nitrogen fixation and diversity of benthic cyanobacterial mats on coral reefs in Curaçao. Coral Reefs 37, 861–874.  Callahan, B., McMurdie, P., Rosen, M., Han, A., Johnson, A., Holmes, S., 2016. DADA2: High-resolution sample inferene from Illumina amplicon data. Nat. Methods 13, 581–583.  Capodaglio, A.G., 2017. Integrated, Decentralized Wastewater Management for Resource Recovery in Rural and Peri-Urban Areas. Resour. Basel 6, 22.   81 Carney, L.T., Reinsch, S.S., Lane, P.D., Solberg, O.D., Jansen, L.S., Williams, K.P., Trent, J.D., Lane, T.W., 2014. Microbiome analysis of a microalgal mass culture growing in municipal wastewater in a prototype OMEGA photobioreactor. Algal Res. 4, 52–61.  Daigger, G., 2009. Evolving Urban Water and Residuals Management Paradigms: Water Reclamation and Reuse, Decentralization, and Resource Recovery. Water Environ. Res. 81, 809–823.  Dionisi, H.M., Layton, A.C., Harms, G., Gregory, I.R., Robinson, K.G., Sayler, G.S., 2002. Quantification of Nitrosomonas oligotropha-Like Ammonia-Oxidizing Bacteria and Nitrospira spp. from Full-Scale Wastewater Treatment Plants by Competitive PCR. Appl. Environ. Microbiol. 68, 245–253.  Donald, D.B., Bogard, M.J., Finlay, K., Leavitt, P.R., 2011. Comparative effects of urea, ammonium, and nitrate on phytoplankton abundance, community composition, and toxicity in hypereutrophic freshwaters. Limnol. Oceanogr. 56, 2161–2175.  Eriksson, Ö., 1974. Mikrosilning av obehandlat avloppsvatten samt dag- och bräddvatten. Feo, G.D., Antoniou, G., Fardin, H.F., El-Gohary, F., Zheng, X.Y., 2014. The Historical Development of Sewers Worldwide. Sustainability 6.  Foladori, P., Petrini, S., Nessenzia, M., Andreottola, G., 2018. Enhanced nitrogen removal and energy saving in a microalgal-bacterial consortium treating real municipal wastewater. Water Sci. Technol. 78, 174–182.  Foss, S., Harder, J., 1998. Thauera linaloolentis sp. nov. and Thauera terpenica sp. nov., Isolated on Oxygen-containing Monoterpenes. Syst. Appl. Microbiol. 21, 365–373.  Franchi, A., Santoro, D., 2015. Current status of the rotating belt filtration (RBF) technology for municipal wastewater treatment. Water Pract. Technol. Lond. 10, 319–327.  Harder, R., 2012. Source-separation in the urban water infrastructure. STOWA.  ISO, 2015. Guidelines for treated wastewater use for irrigation projects.    82 Kalogo, Y., Monteith, H., Eng, P., 2008. State of Science Report: Energy and Resource Recovery From Sludge. GWRC.  Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., Glöckner, F.O., 2013. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 41, e1–e1.  Könneke, M., Bernhard, A.E., Torre, J.R. de la, Walker, C.B., al,  et, 2005. Isolation of an autotrophic ammonia-oxidizing marine archaeon. Nat. Lond. 437, 543–6.  Kozich, J.J., Westcott, S.L., Baxter, N.T., Highlander, S.K., Schloss, P.D., 2013. Development of a Dual-Index Sequencing Strategy and Curation Pipeline for Analyzing Amplicon Sequence Data on the MiSeq Illumina Sequencing Platform. Appl Env. Microbiol 79, 5112–5120.  Krohn-Molt, I., Wemheuer, B., Alawi, M., Poehlein, A., Güllert, S., Schmeisser, C., Pommerening-Röser, A., Grundhoff, A., Daniel, R., Hanelt, D., Streit, W.R., 2013. Metagenome Survey of a Multispecies and Alga-Associated Biofilm Revealed Key Elements of Bacterial-Algal Interactions in Photobioreactors. Appl Env. Microbiol 79, 6196–6206.  Krustok, I., Truu, J., Odlare, M., Truu, M., Ligi, T., Tiirik, K., Nehrenheim, E., 2015. Effect of lake water on algal biomass and microbial community structure in municipal wastewater-based lab-scale photobioreactors. Appl. Microbiol. Biotechnol. 99, 6537–6549.  Larocque, G., McMurtrie, R., Payne, M., Ralston, I., 2018. Onsite Sewerage Systems (Professional Practice Guidelines). Engineers and Geoscientists of British Columbia.  Lateef, S.K., Soh, B.Z., Kimura, K., 2013. Direct membrane filtration of municipal wastewater with chemically enhanced backwash for recovery of organic matter. Bioresour. Technol. 150, 149–155.  Lema, J.M., Martinez, S.S., 2017. Innovative Wastewater Treatment & Resource Recovery Technologies: Impacts on Energy, Economy and Environment. IWA Publishing.    83 Leu, K.-L., Hsu, B.-D., 2005. A programmed cell disintegration of Chlorella after heat stress. Plant Sci. 168, 145–152.  Levine, A.D., Tchobanoglous, G., Asano, T., 1985. Characterization of the Size Distribution of Contaminants in Wastewater: Treatment and Reuse Implications. Water Pollut. Control Fed. 57, 805–816.  Ljunggren, M., 2006. Micro screening in wastewater treatment – an overview. Vatten 171–177.  Love, M.I., Huber, W., Anders, S., 2014. Moderated estimation of fold change and dispersion for RNA-seq data with   DESeq2. Genome Biol. 15, 550.  Lv, P., Luo, J., Zhuang, X., Zhang, D., Huang, Z., Bai, Z., 2017. Diversity of culturable aerobic denitrifying bacteria in the sediment, water and biofilms in Liangshui River of Beijing, China | Scientific Reports. Sci. Rep. 7.  McIlroy, S.J., Starnawska, A., Starnawski, P., Saunders, A.M., Nierychlo, M., Nielsen, P.H., Nielsen, J.L., 2016. Identification of active denitrifiers in full-scale nutrient removal wastewater treatment systems. Environ. Microbiol. 18, 50–64.  Metcalf and Eddy, Inc., Tchobanoglous, G., Stensel, D., Tsuchihashi, R., Burton, F., 2013. Wastewater Engineering: Treatment and Resource Recovery.  Ministry of Agriculture, n.d. B.C. Irrigation Management Guide. Government of British Columbia.  MOBIO Laboratories Inc, n.d. PowerMag Soild DNA Isolation Kit (Instruction Manual).  Muga, H.E., Mihelcic, J.R., 2008. Sustainability of wastewater treatment technologies. J. Environ. Manage. 88, 437–447.  Nogueira, R., Lazarova, V., Manem, J., Melo, L.F., 1998. Influence of dissolved oxygen on the nitrification kinetics in a circulating bed biofilm reactor. Bioprocess Eng. 19, 441–449.    84 NSF, n.d. NSF/ANSI 40: Residential Onsite Systems [WWW Document]. URL http://www.nsf.org/services/by-industry/water-wastewater/onsite-wastewater/residential-wastewater-treatment-systems (accessed 1.22.19).  Oksanen, J., Blanchet, F., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P., O’Hara, R., Simplson, G., Solymos, P., Stevens, M., Szoecs, E., Wagner, H., 2018. Vegan: Community Ecology Package.  Parks, D., Tyson, G., Hugenholtz, P., Beiko, R., 2014. STAMP: Statiscial analysis of taxonomic and functional profiles. Bioinformatics 3123–3124.  Petterson, F., 2004. Mikrosilning som förbehandlingsmetod av hushållsavloppsvatten 78.  Posadas, E., García-Encina, P.-A., Soltau, A., Domínguez, A., Díaz, I., Muñoz, R., 2013. Carbon and nutrient removal from centrates and domestic wastewater using algal–bacterial biofilm bioreactors. Bioresour. Technol. 139, 50–58.  Poseidon Resources, 2019. Water Filtration Infographic. URL http://stevecookillustration.com/portfolio/infographic-12/ (accessed 3.6.19).  Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J., Glöckner, F.O., 2013. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, 590–596.  Ramakrishna, V., Babu, B.V., 1999. Fresh Water Depletion - A Crisis: Causes and Remediation. ResearchGate.  Rohart, F., Gautier, B., Singh, A., Le Cao, K.-A., 2017. mixOmics: An R package for ’omics feature selection and multiple data integration. PLoS Comput. Biol. 13.  Roni, P., Max, M., François-Gaël, M., Andreas, S., Jiande, Z., Markus, H., 2019. Quantifying physical disintegration of faeces in sewers: Stochastic model and flow reactor experiments. Water Res. 152, 159–170.    85 Rossello-Mora, R., Wagner, M., Amann, R., Schleifer, K., 1995. The abundance of Zoogloea ramigera in sewage treatment plants. Appl. Environ. Microbiol. 61, 702–707.  Rudi, K., Skulberg, O.M., Jakobsen, K.S., 1998. Evolution of Cyanobacteria by Exchange of Genetic Material among Phyletically Related Strains. J. Bacteriol. 180, 3453.  Rutledge, K., McDaniel, M., Boudreau, D., Ramroop, T., Teng, S., 2011. Dead Zone [WWW Document]. Natl. Geogr. URL https://www.nationalgeographic.org/encyclopedia/dead-zone/ (accessed 12.17.18).  Salsnes Filter AS., 2017. Salsnes Filter: Eco-Efficient Solids Separation.  Särner, E., 1978. Silning av obehandlat avloppsvatten. Vatten.  Särner, E., 1976. Silning som förbehandlingsprocess. Bull. Ser. VA.  Seviour, R., Nielsen, P.J., 2010. Microbial Ecology of Activated Sludge. IWA Publishing.  Shams, S., Capelli, C., Cerasino, L., Ballot, A., Dietrich, D., Sivonen, K., Salmaso, N., 2015. Anatoxin-a producing Tychonema (Cyanobacteria) in European waterbodies. - PubMed - NCBI. Water Res. 69, 68–79.  Sharma, R., Gupta, S.K., 2004. Influence of chemical oxygen demand/total kjeldahl nitrogen ratio and sludge age on nitrification of nitrogenous wastewater. Water Environ. Res. 76, 155–161.  Shea, T.G., Males, R.M., 1971. Investigation of Response Surfaces of the Microscreen Process, Water Pollution Control Research Series.  Silyn-Roberts, G., Lewis, G., 2001. In situ analysis of Nitrosomonas spp. in wastewater treatment wetland biofilms. Water Res. 35, 2731–2739.  Standard Methods for the Examination of Water and Wastewater, 20th Edition. ed, 2003. . APHA, Washington, DC, USA.   86 Stevens, D.K., Krieger, R.I., 1991. Stability studies on the cyanobacterial nicotinic alkaloid anatoxin-A. Toxicon 29, 167–179.  Thomsen, T., Kong, Y., Nielsen, P.H., 2007. Ecophysiology of abundant denitrifying bacteria in activated sludge | FEMS Microbiology Ecology | Oxford Academic. FEMS Microbiol. Ecol. 60, 370–382.  Todt, D., Heistad, A., Jenssen, P.D., 2015. Load and distribution of organic matter and nutrients in a separated household wastewater stream. Environ. Technol. 36, 1584–1593.  USEPA, 2012a. Guidelines for Water Reuse 90.  USEPA, 2012b. Guidelines for Water Reuse 131.  USEPA, 2012c. Guidelines for Water Reuse 148.  USEPA, 2000. Technology Fact Sheet: Ammonia Stripping.  Verardo, D.J., Froelich, P.N., McIntyre, A., 1990. Determination of organic carbon and nitrogen in marine sediments using the Carlo Erba NA-1500 Analyzer. Deep Sea Res. 37, 157–165.  Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naïve Bayesian Classifier for Rapid Assignment of rRNA Sequences into the New Bacterial Taxonomy. Appl. Environ. Microbiol. 73, 5261–5267.  Westerling, K., 2014. Biological Treatment 101 Suspended Growth Vs. Attached Growth [WWW Document]. Water Online. URL https://www.wateronline.com/doc/biological-treatment-suspended-growth-vs-attached-growth-0001 (accessed 12.17.18).  Ye, L., Zhang, T., 2013. Bacterial communities in different sections of a municipal wastewater treatment plant revealed by 16S rDNA 454 pyrosequencing - ProQuest. Appl. Microbiol. Biotechnol. 97, 2681–2690.    87 Yin, Z., Bi, X., Xu, C., 2018. Ammonia-Oxidizing Archaea (AOA) Play with Ammonia-Oxidizing Bacteria (AOB) in Nitrogen Removal from Wastewater. Archaea.  You, S.J., Hsu, C.L., Chuang, S.H., Ouyang, C.F., 2003. Nitrification efficiency and nitrifying bacteria abundance in combined AS-RBC and A2O systems. Water Res. 37, 2281–2290.  Zamalloa, C., Boon, N., Verstraete, W., 2013. Decentralized two-stage sewage treatment by chemical–biological flocculation combined with microalgae biofilm for nutrient immobilization in a roof installed parallel plate reactor. Bioresour. Technol. 130, 152–160.  Zaneveld, J.R., McMinds, R., Thurber, R.V., 2017. Stress and stability: applying the Anna Karenina principle to animal microbiomes. Nat. Microbiol. 2, 17121.      88 Appendix A: Microscreen Performance   Figure A.1 - COD removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences.  ●●●●●●●●●Median= 40.3 %0255075100COD  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons  89  Figure A.2 - SS removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences.  Figure A.3 - TKN removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ●●●●●●●●●Median= 68 %0255075100SS  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons●●●●●Median= 8.19 %0255075100TKN  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons  90  Figure A.4 - TN removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences.  Figure A.5 - TP removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences. ●●●●●Median= 7.89 %0255075100TN  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons●●●●●●Median= 26.5 %0255075100TP  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons  91  Figure A.6 - Turbidity removal efficiency (%) of a 54 µm microscreen on wastewater from two UBC residences.  ●●●●●●●● ●Median= 20 %0255075100Turbidity  Removal Efficiency  %SamplingLocation●Acadia ParkOrchard Commons  92 Appendix B: PBR Performance  Figure B.1 - Alkalinity concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ●I = 188A = −4.69 B = −3.41020040020 25 30 35A● ●●●I = 280A = 157 B = 193020040040 50 60 70B●● ● ● ● ●●●●I = 263A = 114 B = 191020040070 80 90 100DayExperimentalCondition●Influent Open ClosedCAlk  ( mg/L )  93  Figure B.2 - BOD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.3 - COD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ●● ● ● ● ●● ● ● ● ●●I = 66.2A = 3.36 B = 1.6301002003004000 10 20 30A● ● ● ● ● ● ● ● ● ● ●I = 153A = 5.28 B = 5.12010020030040040 50 60 70B● ● ● ● ● ●I = 144A = 4.04 B = 6.31010020030040070 75 80 85 90 95DayExperimentalCondition●Influent Open ClosedCBOD  ( mg/L )● ● ● ● ● ● ● ● ● ● ● ● ●I = 220A = 63.3 B = 49.902505007500 10 20 30A● ● ● ● ● ●● ● ● ●I = 401A = 68 B = 57025050075040 50 60 70B● ● ● ● ● ● ●●I = 329A = 61.9 B = 68.2025050075070 80 90 100DayExperimentalCondition●Influent Open ClosedCCOD  ( mg/L )  94  Figure B.4 - Filtered COD concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.5 - NH4-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ● ● ● ● ●●● ●I = 132A = 46.2 B = 34.902004000 10 20 30A● ● ● ● ● ●● ● ● ●I = 204A = 60.3 B = 38.2020040040 50 60 70B● ● ● ● ● ● ●●I = 204A = 70 B = 53.5020040070 80 90 100DayExperimentalCondition●Influent Open ClosedCCOD.F  ( mg/L )● ● ● ●●●● ●● ●● ● ●I = 38.2A = 18.3 B = 1202040600 10 20 30A● ● ● ● ● ● ● ●●● I = 26.3A = 4.57 B = 6.77020406040 50 60 70B● ●●● ●●●●I = 32A = 9.11 B = 18.3020406070 80 90 100DayExperimentalCondition●Influent Open ClosedCNH3  ( mg/L )  95  Figure B.6 - NO2-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.7 – NO3-N concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ● ● ● ●●● ● ●I = 0.4A = 0.37B = 2.1010200 10 20 30A● ● ● ● ● ● ● ● ● ●I = 0.085A = 0.14B = 0.0270102040 50 60 70B● ●●●● ● ● ●I = 0.039A = 0.2B = 20102070 80 90 100DayExperimentalCondition●Influent Open ClosedCNO2  ( mg/L )●● ● ● ● ●● ●●●● ●●I = −0.19A = 20 B = 13−10010200 10 20 30A●● ● ● ●●● ●●● I = 0.13A = 9.8B = 7.7−100102040 50 60 70B● ●● ● ● ● ● ●I = −0.012A = 9.8 B = 0.63−100102070 80 90 100DayExperimentalCondition●Influent Open ClosedCNO3  ( mg/L )  96  Figure B.8 – PO4-P concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.9 - Suspended solids concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ● ● ● ●● ● ● ●I = 4.5A = 2.57 B = 2.6501020300 10 20 30A● ● ● ● ● ● ● ●●● I = 4.37A = 1.85 B = 2.05010203040 50 60 70B● ● ● ● ● ● ●●I = 7.47A = 3.68 B = 3.56010203070 80 90 100DayExperimentalCondition●Influent Open ClosedCPO4  ( mg/L )● ● ● ● ● ● ● ● ● ● ● ● ●I = 50.5A = 7.33 B = 5.270501001502002500 10 20 30A●● ● ● ● ● ● ● ● ●I = 97.1A = 7.09 B = 7.2805010015020025040 50 60 70B●●● ● ● ● ●I = 72.6A = 6.86 B = 905010015020025070 80 90 100DayExperimentalCondition●Influent Open ClosedCSS  ( mg/L )  97  Figure B.10 - TKN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.11 - Filtered TKN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ● ● ● ● ●● ● ●I = 62A = 19.8 B = 16.90501001502000 10 20 30A● ● ● ● ● ● ● ●●● I = 64.2A = 9.58 B = 1305010015020040 50 60 70B● ● ● ● ● ●●I = 94.4A = 12.8 B = 23.305010015020070 80 90 100DayExperimentalCondition●Influent Open ClosedCTKN  ( mg/L )● ● ● ●●●● ● ●● ● ● ●I = 40.1A = 15.9 B = 10.502550750 10 20 30A● ● ● ● ● ● ● ●●●I = 30.1A = 5.23 B = 8.91025507540 50 60 70B● ●●● ●●●●I = 39A = 10.3 B = 21.2025507570 80 90 100DayExperimentalCondition●Influent Open ClosedCTKN.F  ( mg/L )  98  Figure B.12 - TN concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.13 - TP concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ●● ● ● ●● ● ●I = 62.2A = 39.6 B = 32.40501001502000 10 20 30A● ● ● ● ● ● ● ●●● I = 64.4A = 19.5 B = 20.805010015020040 50 60 70B● ● ●● ● ● ●●I = 94.4A = 21.2 B = 2305010015020070 80 90 100DayExperimentalCondition●Influent Open ClosedCTN  ( mg/L )● ●● ● ● ● ● ●● ●●●●I = 2.36A = 1.05 B = 0.7910.02.55.07.510.00 10 20 30A●● ●●●●●● ●● I = 2.89A = 1.32 B = 2.080.02.55.07.510.040 50 60 70B●● ● ● ● ●●I = 3.54A = 1.32 B = 1.880.02.55.07.510.070 80 90 100DayExperimentalCondition●Influent Open ClosedCTP  ( mg/L )  99  Figure B.14 - TS concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.15 - Turbidity (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means. ●● ● ●● ● ● ●● ●●● ●I = 186A = 177 B = 1532505007500 10 20 30A●●●● ●● ●● ●●I = 483A = 308 B = 24425050075040 50 60 70B● ● ● ● ● ● ●I = 405A = 231 B = 21125050075070 80 90 100DayExperimentalCondition●Influent Open ClosedCTS  ( mg/L )● ● ● ● ● ● ● ● ● ● ● ● ●I = 40.9A = 1.91 B = 1.5601002003004000 10 20 30A● ● ● ● ● ● ● ● ● ●I = 133A = 2.23 B = 2.07010020030040040 50 60 70B● ● ● ●I = 75.6A = 1.82 B = 2.83010020030040070 80 90 100DayExperimentalCondition●Influent Open ClosedCTurb  ( NTU )  100  Figure B.16 - Transmittance at 254nm (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the influent (I) and effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.17 - DO concentrations (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ●● ● ● ●● ● ● ● ● ●●I = 23.1A = 60.2 B = 66.40204060800 10 20 30A●● ● ● ● ●●● ● ●I = 15.3A = 50.8 B = 59.702040608040 50 60 70B●● ● ● ●I = 5.22A = 56.5 B = 55.902040608070 80 90 100DayExperimentalCondition●Influent Open ClosedCX254  ( % )●●●●● ●●●● ●●●●●●● ● ●●●●●A = 6.9 B = 7.20510150 10 20 30A●●●●●● ●●●●●●● ● ● ● ●● ● ●●A = 8.9B = 6.205101540 50 60B●●● ● ● ● ● ●●● ●● ●● ● ● ●● ● ● ●A = 6.4B = 2.805101570 80 90DayExperimental Condition ●Open ClosedCDO  ( mg/L )  101  Figure B.18 - PH (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means.  Figure B.19 - Temperature (sampled at hour 17 in Phases A and B, and hour 24 in Phase C) of the effluent from the open (A) and closed (B) reactors. Annotations denote means. ● ● ● ● ● ●● ●●●● ● ● ●● ●● ●● ● ●●A = 5 B = 5.45790 10 20 30A● ●● ●● ● ● ●● ● ● ●● ● ● ● ● ● ●● ● ●A = 8.5B = 7.357940 50 60B●●●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ●A = 7.8 B = 7.357970 80 90DayExperimental Condition ●Open ClosedCpH  (  )●●●●●●●●● ●● ●●●●● ● ●●●● ●A = 21.4B = 22.7182124270 10 20 30A● ● ●●● ● ●●●● ● ●● ● ●● ●●●●●A = 22.8 B = 241821242740 50 60B●●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ●● ●A = 24.2 B = 25.51821242770 80 90DayExperimental Condition ●Open ClosedCTemp  (  )  102  Figure B.20 - The COD removal rate (g/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant.   0 5 10 15 20 25 30−20246DayCOD Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●●● ●● ●●●●● ●45 50 55 60 65−50510152025DayCOD Removal Rate  (g/m2/day)OpenClosedOpen−Closed●● ●●●●● ●● ●80 85 90 950510DayCOD Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●  103  Figure B.21 - The denitrification rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant.   0 5 10 15 20 25 30−0.50.00.51.01.5DayDenitrification Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●●● ●●45 50 55 60 65−0.50.00.51.01.52.02.5DayDenitrification Rate  (g/m2/day)OpenClosedOpen−Closed●● ●●● ●● ●●●80 85 90 9501234DayDenitrification Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●  104  Figure B.22 - The nitrification rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. 0 5 10 15 20 25 30−0.50.00.51.01.52.0DayNitrification Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●●● ● ●45 50 55 60 65−0.50.00.51.01.52.02.5DayNitrification Rate  (g/m2/day)OpenClosedOpen−Closed● ● ●● ●● ● ●●●80 85 90 9501234DayNitrification Rate  (g/m2/day)OpenClosedOpen−Closed●●●● ●●●  105   Figure B.23 - The assimilation rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant.    0 5 10 15 20 25 30−0.20.00.20.40.60.8DayAssimilation Rate  (g/m2/day)OpenClosedOpen−Closed● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ●40 45 50 55 60 650.00.20.40.60.8DayAssimilation Rate  (g/m2/day)OpenClosedOpen−Closed● ● ● ●● ● ● ●● ● ●● ● ● ● ●70 75 80 85 90 950.00.20.40.60.81.0DayAssimilation Rate  (g/m2/day)OpenClosedOpen−Closed● ●● ● ●● ● ●● ● ●  106  Figure B.24 - The TKN removal rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant.   0 5 10 15 20 25 300.00.51.01.52.0DayTKN Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●●●● ●45 50 55 60 650.00.51.01.52.02.53.0DayTKN Removal Rate  (g/m2/day)OpenClosedOpen−Closed● ● ●● ●● ● ●●●80 85 90 9501234DayTKN Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●● ●●●  107  Figure B.25 - The TN removal rate (gN/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. 0 5 10 15 20 25 30−0.50.00.51.01.5DayTN Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●●●●●45 50 55 60 65−0.50.00.51.01.52.02.53.0DayTN Removal Rate  (g/m2/day)OpenClosedOpen−Closed●● ●●●●● ●●●80 85 90 9501234DayTN Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●●●●●  108  Figure B.26 - The TP removal rate (gP/m2/day) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases A (top left), B (top right), and C (bottom left). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant. 0 5 10 15 20 25 300.00.10.20.30.40.50.6DayTP Removal Rate  (g/m2/day)OpenClosedOpen−Closed●● ● ● ●● ●● ●●●●45 50 55 60 650.00.10.20.30.40.50.6DayTP Removal Rate  (g/m2/day)OpenClosedOpen−Closed●● ● ●●●●●●80 85 90 950.00.20.40.6DayTP Removal Rate  (g/m2/day)OpenClosedOpen−Closed●●●● ●●●  109  Figure B.27 - The specific alkalinity demand (gCaCO3/gN) in the open (■) and closed (•) are shown with 95% confidence shaded intervals for Phases B (left) and C (right). If the confidence interval of the differential (▲) overlaps with zero, the difference between the rates in the open and closed reactors is not significant.   50 55 60 65−505DaySpecific Alkalinity Demand  (gCaCO3/gN)OpenClosedOpen−Closed●●●●80 85 90 95−2−1012DaySpecific Alkalinity Demand  (gCaCO3/gN)OpenClosedOpen−Closed●●● ●●●●  110 Appendix C: PBR Biomass Characteristics   Figure C.1 - Cartenoid content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ●●●● ●● ●●A = 206B = 14710020030040 50 60 70B●●●●●●A = 180B = 15310020030070 75 80 85 90DayExperimental Condition ●Open ClosedCCart  ( mg/m2 )  111  Figure C.2 – Total chlorophyll content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means.  Figure C.3 - Chlorophyll A content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means. ●●●●●●●●A = 598B = 437250500750100040 50 60 70B●●●●●●A = 438 B = 417250500750100070 75 80 85 90DayExperimental Condition ●Open ClosedCCHL  ( mg/m2 )●●●●● ● ●●A = 429B = 34620040060040 50 60 70B●●●●●●A = 350 B = 35120040060070 75 80 85 90DayExperimental Condition ●Open ClosedCCHLA  ( mg/m2 )  112  Figure C.4 - Chlorophyll B content (mg/m2) of biomass scrapped from the surface of the open and closed reactors in Phases B and C. Annotations denote means.  Figure C.5 - Dry weight (g/m2/day) of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means. ●● ●●●●●●A = 169B = 90.9010020030040040 50 60 70B●●●●●●A = 88.2 B = 65.5010020030040070 75 80 85 90DayExperimental Condition ●Open ClosedCCHLB  ( mg/m2 )●●●● ●●●●A = 2.5 B = 3.60.02.55.07.510.012.50 10 20 30A● ● ●●● ●●●A = 4.5 B = 4.10.02.55.07.510.012.540 50 60 70B● ● ● ●●●A = 7.3B = 3.80.02.55.07.510.012.570 75 80 85 90DayExperimental Condition ●Open ClosedCDWM2D  ( g/m2/day )  113  Figure C.6 - Dry to wet weight ratio of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means.  Figure C.7 - Photosynthetic efficiency (%) of biomass scrapped from the surface of the open and closed reactors in the three experimental phases. Annotations denote means. ●●●●●●●●A = 0.06 B = 0.060.0500.0750.1000 10 20 30A● ● ● ●● ● ●●A = 0.06 B = 0.070.0500.0750.10040 50 60 70B● ●● ●●●A = 0.09 B = 0.090.0500.0750.10070 75 80 85 90DayExperimental Condition ●Open ClosedCDWR  (  )●●●● ●●●●A = 2 B = 2.80.02.55.07.510.00 10 20 30A● ● ●●● ●●●A = 3.5 B = 3.20.02.55.07.510.040 50 60 70B● ● ● ●●●A = 5.8B = 30.02.55.07.510.070 75 80 85 90DayExperimental Condition ●Open ClosedCPE  ( % )  114 Appendix D: 16s rRNA Gene Abundance Analysis Results   Figure D.1 - Relative abundance (%) of the 12 most common phyla observed in triplicate at the end of the startup and three experimental phases. 0 28 64 93ClosedOpen1 2 3 1 2 3 1 2 3 1 2 302550751000255075100RepRelative Abundance (%)PhylumFusobacteriaEpsilonbacteraeotaDeinococcus−ThermusActinobacteriaVerrucomicrobiaArmatimonadetesFirmicutesPlanctomycetesAcidobacteriaBacteroidetesProteobacteriaCyanobacteriaDay  115   Figure D.2 - Principle coordinate analysis of the relative abundance of microbes at the end of the startup phase and each of the three experimental phases. Top-left graph shows the 1st (35% explained variance) and 2nd (12% explained variance) principle coordinates, top-right shows 1st and 3rd (10% explained variance), and bottom-left shows 1st and 4th (9% explained variance).  ●●●●●●PCA sample plot−60 −30 0 30 60−40040xyGroup●●ClosedOpenDay● 0286493●●●●●●PCA sample plot−60 −30 0 30 60−60−30030xyGroup●●ClosedOpenDay● 0286493●●●● ●●PCA sample plot−60 −30 0 30 60−25025xyGroup●●ClosedOpenDay● 0286493  116    Figure D.3 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of the startup phase. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively.    117  Figure D.4 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of the Phase A. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively.    118  Figure D.5 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of Phase B. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively.    119  Figure D.6 - STAMP analysis comparing relative abundance of genera in the open and closed reactors at the end of Phase C. Bars to the left and right of center indicate significantly greater abundance in the open and closed reactors, respectively.       120 Appendix E: Continuous Diel Monitoring Experiment Results  Figure E.1 - Concentrations of key parameters during the diel monitoring study at the end of Phase A with the reactors operating under the “continuous” feed condition.   Open Closed●●●●●●●●●●●●●●●●●● ●●●●● ● ●● ● ●●●●●●020400 5 10 15 20●●●●●●●●●●●●●●●●●● ●●●●● ● ●● ● ●●●●●●020400 5 10 15 20Parameter●TN  ( mg/L )TKN.F  ( mg/L )NH3  ( mg/L )NO2  ( mg/L )NO3  ( mg/L )●●●●●●●●●●●●●●●●●●●●●●●● ●● ●●●●●●● ●●01020300 5 10 15 20●●●●●●●●●●●●●●●●●●●●●●●● ●● ●●●●●●● ●●01020300 5 10 15 20Parameter●DO  ( mg/L )Temp  ( *C )pH  (  )●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●1002003004000 5 10 15 20●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●1002003004000 5 10 15 20Parameter● COD  ( mg/L )ALK  ( mg/L )01020300 5 10 15 2001020300 5 10 15 20ParameterTP.F  ( mg/L )Experiment Hour  121  Figure E.2 - Concentrations of key parameters during the diel monitoring study at the end of Phase B with the reactors operating under the “continuous” feed condition.  Open Closed●●●●●●●●●●●●●● ●●●●●● ●● ●●● ●●●●●●●●●●010203040500 5 10 15 20●●●●●●●●●●●●●● ●●●●●● ●● ●●● ●●●●●●●●●●010203040500 5 10 15 20Parameter●TN  ( mg/L )TKN.F  ( mg/L )NH3  ( mg/L )NO2  ( mg/L )NO3  ( mg/L )●●●●●●●●●●●●●●●●●●●●●●●● ●●● ●●●●●●●●●●●010200 5 10 15 20●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●010200 5 10 15 20Parameter●DO  ( mg/L )Temp  ( *C )pH  (  )●●●●●●●●●●●●●●●●●●●●●●● ●●● ●●●●●●●●●●●02550750 5 10 15 20 25●●●●●●●●●●●●●●●●●●●●●●● ●●● ●●●●●●●●●●●02550750 5 10 15 20 25Parameter● COD  ( mg/L )ALK  ( mg/L )01020300 5 10 15 2001020300 5 10 15 20ParameterTP.F  ( mg/L )Experiment Hour

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0380207/manifest

Comment

Related Items