Tendo em vista a crescente problemática quanto à eutrofização de corpos dágua em países em desenvolvimento, devido ao lançamento de esgoto mesmo que tratado em nível secundário, o processo de lodo ativado vem sendo estudo e aprimorado para efetuar a remoção conjunta de matéria orgânica, nitrogênio e o fósforo, nutrientes limitantes para ocorrência da eutrofização. Neste contexto, o presente trabalho pretende contribuir com informações quanto à remoção biológica de nutrientes de esgoto sanitário através de estudo em escala piloto em um sistema cíclico de lodo ativado. Este sistema, operado em regime de bateladas sequenciais, difere da configuração habitual, pois possui três compartimentos internos, zonas 1, 2 e 3 que apresentam condições anaeróbias/anóxicas, anóxicas e aeróbias respectivamente, além do retorno de lodo da zona 3 para zona 1 durante o período de reação, sob vazão de 20% da vazão afluente a unidade. Cada batelada possui duração de 4 horas, distribuídas em 4 etapas de 1h cada, sendo elas o enchimento com aeração, reação, sedimentação e descarga. Com essas características o sistema cíclico de lodo ativado do presente estudo, demonstrou no tratamento de esgoto sanitário, elevada eficiência na remoção de matéria orgânica, sempre acima de 90%, além da boa eficiência de remoção nitrogênio através do processo de nitrificação e desnitrificação, cuja concentração média de amônia e nitrato no efluente final foram respectivamente de 2,8 e 11,4 mgN.L-1 para idade de lodo de 25 dias e quando operado com 12 dias de idade, o efluente apresentou concentração de 6,6 e 12,4 mgN.L-1. Também foram determinadas via respirometria as taxas máximas de crescimento (n,max) para bactérias nitrificantes e nitratantes, os valores obtidos foram respectivos, de 0,45 d-1 e 0,37 d-1 para idade de lodo de 12 dias, resultados próximos a literatura. A remoção biológica de fósforo foi obtida com melhor resultado na segunda etapa, superando 95% de eficiência, produzindo efluente final com teor de fósforo total inferior a 0,5 mgP.L-1. / Given the growing problem regarding the eutrophication of water bodies in developing countries, due to discharge of sewage treated even at the secondary level, the activated sludge process has been enhanced to study and perform the joint removal of organic matter, nitrogen and phosphorus, the limiting nutrient for the occurrence of eutrophication. In this context, this work aims to contribute information for biological nutrient removal from wastewater by pilot scale study in a cyclic activated sludge system. This system operated in sequential batch system differs from typical configuration because it has three internal compartments, zones 1, 2 and 3 show that anaerobic / anoxic, aerobic and anoxic respectively, beyond the conditions of return sludge from zone 3 to zone 1 during the reaction, flow rate under 20% of the influent flow unit. Each batch has a duration of 4 hours, distributed into 4 stages each 1H, they aeration and filling, reaction, settling and discharge. With these characteristics the cyclic activated sludge system of the present study showed for the treatment of sewage, high efficiency in the removal of organic matter, always above 90%, besides the good efficiency of nitrogen removal by nitrification and denitrification, which average concentration of ammonia and nitrate in the final effluent were respectively 2.8 and 11.4 mgN.L-1 for sludge age of 25 days, when operated at 12 days of age, the effluent showed a concentration of 6.6 to 12 4 mgN.L-1. Were also determined via respirometry maximum growth rates (n, max) for nitrifying bacteria and nitratantes, the respective value obtained were, 0.45 d-1 and 0.37 d-1 for sludge age of 12 days, results Nearby literature. Biological phosphorus removal was obtained with better results in the second stage, surpassing 95% efficiency, producing the final content of the effluent below 0.5 mgP.L-1 total phosphorus.

O aumento na produção de carne bovina aumentou a concentração e volume dos resíduos líquidos produzidos durante seu beneficiamento, conhecidos como água residuária de matadouros (ARMV). Isso vem estimulando o desenvolvimento de processos que operem em alta carga com alta eficiência para seu tratamento. Visto o exposto, objetivou-se avaliar um sistema de tratamento para a ARMV operando em condições anaeróbia/aeróbia/anóxica de forma conjugada visando a remoção de matéria orgânica, nutrientes e toxidade. As unidades que compunham o sistema estudado eram três reatores com biofilme denominados: Reator Anaeróbio Híbrido (RAH), Reator Aeróbio de Leito Móvel (MBBR) e Reator Anóxico com Biofilme (RAB). A ARMV in natura era coletada em um frigorífico e matadouro bovino do vale do Paraíba e caracterizada em termos físicos, químicos e toxicológicos a nível agudo e crônico. O experimento durou aproximadamente 370 dias, operando sob variações de carga orgânica. A coleta ao longo do sistema foi feita semanalmente e as amostras caracterizadas em termos físico-químicos e toxicológicos tanto a nível agudo quanto crônico. Nos reatores RAH, MBBR e RAB foram realizadas testes hidrodinâmicos em condições abiótico e biótico, respectivamente. Com os dados obtidos nas três fases estudadas foram levantados os parâmetros cinéticos de crescimento da biomassa dispersa, além da caracterização microbiológica da mesma e do biofilme. Os resultados obtidos com base na caracterização físico-química e toxicológica da ARMV, in natura, revelaram altas concentrações em toda série de sólidos, de ácidos voláteis totais, alcalinidades, macro e micronutrientes, matéria orgânica em termos de DBO520°C e DQO nas formas total, solúvel e particulada e também de COD. Apresentou-se como um efluente extremamente tóxico a nível agudo para os organismos, bactérias P. putida e E. coli e microcrustaceo D. similis, e extremamente tóxico a nível crônico para os organismos, microcrustaceios C. silvestri e C. dúbia, bactérias E. coli e P putida e alga P. subcaptata. Com base nos testes toxicológicos, concluiu-se que os microcrustaceos e algas foram mais sensíveis as toxinas da ARM testada que as bactérias. O reator RAH operando sob choques orgânicos demonstrou ótimo desempenho operacional. As cargas orgânicas (COV) aplicadas ao RAH foram 608,9; 3.030,4 e 9.581,5 mg L-1 d-1 em termos de DQO para as fases I, II e III, estatisticamente diferente entre si. Apresentando para as três fases eficiências estatisticamente iguais. O reator MBBR operando sob choques orgânicos demonstrou ótimo desempenho operacional. As cargas orgânicas (COV) aplicadas ao MBBR foram 286,5; 2085,2 e 3889,6 mg L-1 d-1 em termos de DQO para as fases I, II e III, estatisticamente diferente entre si. Apresentando para as fases I, II e III eficiências de 76,9%; 60% e 81%, respectivamente, e estatisticamente iguais. Conclui-se com a pesquisa realizada que os reatores RAH e MBBR foram capazes de absorver choques orgânicos e hidráulicos submetidos a biomassa mantendo-se em altos valores de eficiência. O reator MBBR também apresentou bom desempenho no processo de nitrificação com eficiências de 61,2%; 68,1%; 50,7% para as fases I, II e III, respectivamente. A concentração média de OD de 3 mg L-1 mantida no MBBR apresentou-se acima do suficiente. O reator RAB operando sob choques orgânicos demonstrou ótimo desempenho operacional. As cargas orgânicas (COV) aplicadas ao RAB foram 194,9; 1769,1 e 2230,6 mg L-1 d-1 em termos de DQO para as fases I, II e III, estatisticamente diferente entre si. As fases I, II e III apresentaram eficiências de 70,7%; 46,4% e 69,8%, respectivamente, e estatisticamente, iguais entre as fases I e III. Em termos de toxidade, os parâmetros estudados nas fases I e II, mostraram-se ideais para remoção de toxidade, sendo que no final de ambas a ARM tratada apresentou-se livre de toxidade a nível agudo e crônico. Os reatores RAH, MBBR e RAB demonstraram ótimo desempenho hidrodinâmico e cinético. Concluiu-se que o sistema anaeróbio/aeróbio/anóxico estudado foi eficiente no tratamento da ARMV, aliando condições de fácil monitoramento, rapidez no processo e ótimo desempenho. / Development in bovine meat production has increased volume and concentration of liquid residues produced during their improvement, known as slaughterhouses wastewaters (SW). This fact has been stimulating the develpment of processes operating in highly charge and efficience for their treatment. As seen, we focused to develop a treatment system for the SW operating in a anaerobic/aerobic/anoxic condition in a conjugated form, aiming the remotion of organic matter, nutrients and toxicity. The units evolving the studied system were three biofilms reators named: Anaerobic Hibrid Reactor (AHR), Moving Bed Biofilm Reactors (MBBR), and Anoxic Biofilm Reactor (AnBR). The SW in natura was collected in the slaughterhouse of the Paraíba Valley, and were characterized in physical, chemical and toxically in a highly sharp and cronically levels. The experiment has lasted for 370 days, operating under organic load rate (OLR). The withdraws along the system were done weekly and the samples were characterized in physical-chemical and toxicological terms and in both sharp and cronical ways. Hidrodinamic tests were realized in reactors AHR, MBBR and AnBR in abiotic and biotic condictions, repectivelly. With the data obtained in the three studied phases, knetics parameters were collected for the dispersal biomass, besides its biofilm and biomass characterization. The results, based on SW physico - chemical and toxicological characterization, showed high concentrations in all solids series, in total volatic acids, alcalinity, macro and micro nutrients, organic matter in terms of BOD520°C and COD in their solute and particulate forms and also in DOC. It was presented as highly toxix effluent in a sharp level for organisms, bacteria P. putida ans E. coli, and D. similis microcrustacean. And extremely toxic to organisms chronic C. silvestri and C. dúbia microceustacean and P. subcaptata algae. According with toxicological tests it was concluded that the microcrustaceos and a algae were more sensitive to the SSW toxines than bacteria. The AHR reator operating under organic shocks showed excelent operational development. The OLR applied to the AHR were 608.9; 3,030.4 and 9,581.5 mg L-1 d-1 in terms of COD for the phases I, II, III, statiscally different among it, showing for the three phases scores of 76,9%; 60% and 81%, respectivelly, and the same, statistically speaking. We may conclude with this research that the AHR and AMBR reators were capable to absorb hidraulic and organic shocks submitted to biomass keeping high levels of efficiency. The MBBR reator has shown also a good performance in nitrification process, scoring 61.2%; 68.1%; 50.7% in effectiveness for phases I, II and III, respectively. The average DO concentration of 3 mg L-1 maintained in MBBR showed over sufficience. The AnBR reator under organic shocks showed highly operational performance. The OLR applied to AnBR were 194.9; 1,769.1 e 2,230.6 mg L-1 d-1 in terms of COD for the phases I, II, III statistically different among them. The phases I, II, III presented efficiences of 70.7%; 46.4% and 69.8%, respectively, and statiscally the same between phases I and III. In terms of toxicity, the studied parameters in phases I and II showed to be ideal to remove the toxicity from both sharp and chronical levels. The reators AHR, MBBR and AnBR showed excelent hidraulic and kinetic performances. We may conclude that the studied anaerobic/aerobic/ anoxic system was efficient in the treatment of SW, joining, easy feasible conditions, velocity in processing and high performance.

This study has provided fundamental information on the fate of urine nitrogen (N) when applied to pasture soils. In this work the three pasture soils used were a Bruntwood silt loam (BW), an old well-developed (lime and fertilizer incorporated and farmed for more than 20 years) peat soil (OP) and a young peat (YP) which was less developed (farmed for about 10 years). Initial soil chemical and physical measurements revealed that the peat soils were acidic, had higher cation exchange capacities, had greater carbon:nitrogen ratios and were better buffered against changes in soil pH than the BW soil. However, the BW soil was more fertile with a higher pH. The peat soils had lower bulk densities and higher porosities. Four experiments were performed. In the first experiment ¹⁵N-labelled urine was applied at 500 kg N ha⁻¹ to intact soil cores of the three soils. Treatments imposed were the presence and absence of a water table at two temperatures, 8°C or 23° C, over 11-14 weeks. ¹⁵N budgets were determined. This first experiment showed that the nitrification rate was faster in the BW soil and was retarded with a water table present. Significant leaching of nitrate occurred at 8°C in the BW soil without a water table. This was reduced when a water table was present. Leaching losses of urine-N were lower in the peat soils than in the BW soil. Apparent denitrification losses (i.e. calculated on a total-N recovery basis) ranged from 18 to 48 % of the ¹⁵N-applied with the greatest losses occurring in the peat soils. The second experiment examined denitrification losses, over 30 days, following the application of synthetic urine-N at 420 kg N ha⁻¹ to small soil cores situated in growth cabinets. The effects of temperature (8°C or 18°C) and synthetic urine (presence or absence) were measured on the BW and OP soils. Nitrous oxide (N₂0) measurements were taken from all soil cores and a sub-set of soil cores, at 18°C, had ¹⁵N-labelled synthetic urine-N applied so that ¹⁵N-labelled nitrogen gases could be monitored. This experiment showed that the application of synthetic urine and increased soil temperature enhanced denitrification losses from both soils. Denitrification losses, at 18°C, as ¹⁵N-labelled nitrogen gases accounted for 24 to 39 % of the nitrogen applied. Nitrous oxide comprised less than half of this denitrification loss. Losses of N₂0 in leachate samples from the soil cores accounted for less than 0.1 % of the nitrogen applied. A third experiment, using Iysimeters, was performed over a 150 day period in the field. The six treatments consisted of the 3 soils with applied synthetic urine, with or without a simulated water table; each replicated three times. Lysimeters were installed in the field at ground level and ¹⁵N-labelled synthetic urine-N was applied (500 kg N ha⁻¹) on June 4 1992 (day 1). Nitrification rates differed between the soils following the trend noticed in the first experiment. As in the first experiment, nitrate was only detected in the leachate from the BW soil and the inclusion of a water table reduced the concentration of nitrate. In the BW soil, the leachate nitrate concentrations exceeded the World Health Organisation's recommended limit (< 10 mg N L-1) regardless of water table treatment. No nitrate was detected in the leachates from the peat soils but there was some leaching of organic-N (< 5 % of N added) in all the peat soil treatments. Denitrification losses were monitored for the first 100 days of the experiment. In the BW soil without a water table, N₂0 production peaked at approximately day 20 and accounted for 3 % of the nitrogen applied. In the peat soils the measured denitrification losses accounted for less than 1 % of the nitrogen applied. Apparent denitrification losses in the peats were, however, calculated to be approximately 50 % of the ¹⁵N-labelled synthetic urine-N applied. It is postulated that the difference between apparent denitrification losses and those measured could have been due to; loss of dinitrogen in leachate, protracted production of dinitrogen below detectable limits, production of denitrification gases after measurements ceased (i.e. days 100 to 150) and entrapment of dinitrogen in soil cores. Due to the apparent denitrification losses being so high, further research into this nitrogen loss pathway was performed. The fourth and final experiment measured denitrification directly using highly enriched (50 atom %) ¹⁵N-labelled synthetic urine-N. It was performed in a growth cabinet held initially at 8°C. The ¹⁵N-labelled synthetic urine was applied at 500 kg N ha⁻¹ to small soil cores of each soil type. Fluxes of N₂0 and ¹⁵N-labelled gases were measured daily for 59 days. On day 42 the temperature of the growth cabinet was increased to 12°C in an attempt to simulate the mean soil temperature at the end of the field experiment. Up to this time, production of nitrogenous gases from the YP soil had been very low. Interpretation of gaseous nitrogen loss in the YP soil was difficult due to the possibility of chemodenitrification occurring. However, in the OP and BW soils, gaseous losses of nitrogen (determined as ¹⁵N-labelled gas) represented 16 and 7 % of the nitrogen applied respectively. Nitrous oxide comprised approximately half of this gaseous nitrogen loss, in both the OP and BW soils. This work implies that urine-N applied to the mineral soil (BW) could potentially threaten the quality of ground water due to nitrate contamination through leaching. In contrast, denitrification appears to be the major loss mechanism from the peat soils, with the production of nitrous oxide being the primary focus for any environmental concern. Future work should examine the fate of the nitrate leached from the BW soil and the potential for dilution, plant uptake or denitrification below a 30 cm soil depth. A better understanding of the denitrification mechanisms could help reduce denitrification and thereby improve the efficiency of nitrogen use and reduce the output of nitrous oxide.

The aim of this thesis was to identify the denitrifying microbial communities in activated sludge from full-scale treatment plants and from small-scale reactors exposed to acetate or methanol as external carbon sources. Biological denitrification is currently the most widely used, sustainable and cost-effective process to remove nitrogen from wastewater. Increasingly strict effluent discharge standards are posing significant challenges to plant operators to reduce effluent NO3--N concentrations to levels as low as 2-3 mg L-1 or even lower. The lack of sufficient influent carbon in many municipal wastewater treatment plants makes it very difficult to achieve such low NO3--N concentrations in the effluent. An effective solution to the problem is to introduce additional external carbon sources to enhance denitrification. The selection of external carbon sources is not purely based on costs but is also dependent on the possible microbial transformations that these carbon sources may bring about in activated sludge. The most common carbon source used is methanol due to its low cost, but it has been found to cause long delays until an improvement in denitrification performance is observed. On the other hand, acetate has been found to improve denitrification almost instantaneously when added, but it has a significantly higher cost. In this study, methanol and acetate utilising denitrifiers were investigated in activated sludge with and without enrichment in laboratory scale bioreactors. The relevant denitrifiers were identified and evaluated in situ using culture independent methods particularly stable isotope probing (SIP), 16S rDNA cloning, fluorescence in situ hybridisation (FISH) and microautoradiography (MAR). Activated sludge collected from a biological nutrient removal plant exhibiting good denitrification was enriched in an anoxically-operated sequencing batch reactor (SBR) by feeding methanol as the sole carbon source and nitrate as the electron acceptor. The SBR was operated over a duration of 7 months and the SBR denitrification rate improved from 0.02 mg NO3--N mg mixed liquor volatile suspended solids (MLVSS)-1 h-1 to a steady-state value of 0.06 mg NO3-N mg MLVSS-1 h-1. At steady state operation the enriched biomass was subjected to SIP with 13C-methanol to biomark the denitrifiers capable of utilising methanol under anoxic conditions. The separated 12C-DNA and 13C-DNA fractions from the SIP experiment were individually subjected to full cycle rRNA analysis. The dominant 16S rRNA gene phylotype (Group-A clones) in the 13C-library was closely related to the obligate methylotrophs Methylobacillus and Methylophilus in the order Methylophilales of the Betaproteobacteria (96-97% sequence identities), while the most abundant clone groups in the 12C-library mostly belonged to the family Saprospiraceae in the Bacteroidetes phylum. Oligonucleotide probes were designed for FISH to target the Group-A clones and Methylophilales (probes DEN67 and MET1216, respectively) and the Saprospiraceae clones (probe SAP553). Application of these probes on SBR biomass over the enrichment period demonstrated a strong correlation between the level of SBR denitrification and relative abundance of DEN67-targeted bacteria in the SBR community. By contrast, no correlation was found between denitrification rate and the relative abundances of the well known denitrifying genera Hyphomicrobium and Paracoccus nor the Saprospiraceae-clones visualised by FISH in the SBR biomass. FISH combined with microautoradiography independently confirmed that the DEN67-targeted cells were the dominant bacterial group capable of anoxic [14C] methanol uptake in the enriched biomass. As observed in full-scale operations, the methanol-fed SBR experienced a lag period of several weeks before denitrification performance increased. Using FISH quantification, it was shown that this coincided with the lag phase in the growth of the DEN67-targeted denitrifying population. It was therefore concluded that the Methylophilales bacteria dominant in our SBR system are likely to be important in full-scale methanol-fed denitrifying sludges. The acetate utilising microbial consortium in activated sludge was investigated without prior enrichment using stable isotope probing (SIP). 13C-acetate was used in SIP to biomark the DNA of the denitrifiers. The extracted 13C-DNA fraction was subjected to a full cycle rRNA analysis. The dominant 16S rRNA gene phylotypes in the 13C-library were closely related to bacterial families Comamonadaceae and Rhodocyclaceae of class Betaproteobacteria (96-97% sequence identities). Seven oligonucleotide probes (DEN444, DEN220, DEN581, DEN441, DEN124, DEN220a and DEN1454) for use in FISH was designed to specifically target the identified phylotypes. Application of these probes on the sludge of a continuously fed denitrifying sequencing batch reactor (CFDSBR) operated over a duration of 16 days indicated a strong correlation between the level of CFDSBR denitrification and relative abundance of all probe-targeted bacteria in the CFDSBR community. FISH combined with microautoradiography (FISH-MAR) further confirmed that the DEN581- and DEN124-targeted cells dominating the CFDSBR were capable of taking up [14C] acetate under anoxic conditions. The initial occurrence of the DEN444- and DEN1454-targeted bacteria and the final dominance of DEN581- and DEN124-targeted bacteria in the CFDSBR community were likely related to the changing in-reactor nitrite concentrations during the first few days of CFDSBR operation. Hence, the DEN444- and DEN1454-targeted bacteria were hypothesised to have low affinities for nitrite while DEN124- and DEN581-targeted bacteria have higher nitrite affinities. However, it was clear that all probe-targeted bacteria were denitrifiers capable of utilising acetate as a carbon source. The rapid increase in numbers of the probe-targeted organisms positively correlates with the immediate increase in denitrification rates. The rapid response and community shifts observed when acetate was used to enhance denitrification suggest that an intermittent application of acetate is quite effective to temporarily enhance the denitrification capacity of a treatment plant. However, the importance of a bacterial impact assessment of activated sludge subjected to intermittent acetate supplementation is recommended prior to the wide use of acetate in the wastewater industry. The acetate utilising denitrifying microbial communities investigated in the previous chapter were characterised according to their eco-physiological properties using the r- and K-selection criteria. The electron donor (acetate) and acceptor (nitrite) affinities of these probe-identified denitrifiers were used as traits for this characterisation. The substrate to microorganism (S/M) ratio was manipulated to provide high and low substrate concentrations in the reactor to create conditions favourable for r- and K-strategists, respectively. Two factors, namely feeding regimes and sludge retention times, were studied to achieve the desired S/M ratios and enable r/K characterisation. The high substrate affinities and high specific growth rates of two probe-identified denitrifiers (DEN124 and DEN581) did not enable resolution of these two organisms with the feeding regimes used in this study. However, the application of different sludge retention times as a control strategy to maintain constant high and low in-reactor S/M ratios enabled characterisation of the two probe-targeted denitrifiers DEN124 and DEN581 as K- and r-strategists, respectively. The in-reactor S/M ratios applied in this study did not facilitate the characterisation of populations targeted by probes DEN444 and DEN1454. The minor fluctuations of the S/M ratios during a cycle in the SBR operation was considered as a drawback, but conclusive results could still be obtained from the study. A chemostat reactor operation with constant loading and variable flow rates is suggested as an alternative. Conclusively, this study was able to identify specific groups of denitrifying microorganisms in activated sludge when exposed to acetate and methanol. Unlike most previous studies, which relied on culture dependent methods, this study adopted a pure culture independent approach to identify microorganisms in relation to their function, i.e. denitrification. Moreover, acetate denitrifiers were in situ characterised based on eco-physiological properties. The identification of denitrifying communities in this study has paved the way to a larger research project on the optimisation of denitrification processes with external acetate, methanol and other carbon supplements. As such, this study has contributed significantly to the understanding of the denitrification processes by linking process data with microbial investigations.

The aim of this thesis was to identify the denitrifying microbial communities in activated sludge from full-scale treatment plants and from small-scale reactors exposed to acetate or methanol as external carbon sources. Biological denitrification is currently the most widely used, sustainable and cost-effective process to remove nitrogen from wastewater. Increasingly strict effluent discharge standards are posing significant challenges to plant operators to reduce effluent NO3--N concentrations to levels as low as 2-3 mg L-1 or even lower. The lack of sufficient influent carbon in many municipal wastewater treatment plants makes it very difficult to achieve such low NO3--N concentrations in the effluent. An effective solution to the problem is to introduce additional external carbon sources to enhance denitrification. The selection of external carbon sources is not purely based on costs but is also dependent on the possible microbial transformations that these carbon sources may bring about in activated sludge. The most common carbon source used is methanol due to its low cost, but it has been found to cause long delays until an improvement in denitrification performance is observed. On the other hand, acetate has been found to improve denitrification almost instantaneously when added, but it has a significantly higher cost. In this study, methanol and acetate utilising denitrifiers were investigated in activated sludge with and without enrichment in laboratory scale bioreactors. The relevant denitrifiers were identified and evaluated in situ using culture independent methods particularly stable isotope probing (SIP), 16S rDNA cloning, fluorescence in situ hybridisation (FISH) and microautoradiography (MAR). Activated sludge collected from a biological nutrient removal plant exhibiting good denitrification was enriched in an anoxically-operated sequencing batch reactor (SBR) by feeding methanol as the sole carbon source and nitrate as the electron acceptor. The SBR was operated over a duration of 7 months and the SBR denitrification rate improved from 0.02 mg NO3--N mg mixed liquor volatile suspended solids (MLVSS)-1 h-1 to a steady-state value of 0.06 mg NO3-N mg MLVSS-1 h-1. At steady state operation the enriched biomass was subjected to SIP with 13C-methanol to biomark the denitrifiers capable of utilising methanol under anoxic conditions. The separated 12C-DNA and 13C-DNA fractions from the SIP experiment were individually subjected to full cycle rRNA analysis. The dominant 16S rRNA gene phylotype (Group-A clones) in the 13C-library was closely related to the obligate methylotrophs Methylobacillus and Methylophilus in the order Methylophilales of the Betaproteobacteria (96-97% sequence identities), while the most abundant clone groups in the 12C-library mostly belonged to the family Saprospiraceae in the Bacteroidetes phylum. Oligonucleotide probes were designed for FISH to target the Group-A clones and Methylophilales (probes DEN67 and MET1216, respectively) and the Saprospiraceae clones (probe SAP553). Application of these probes on SBR biomass over the enrichment period demonstrated a strong correlation between the level of SBR denitrification and relative abundance of DEN67-targeted bacteria in the SBR community. By contrast, no correlation was found between denitrification rate and the relative abundances of the well known denitrifying genera Hyphomicrobium and Paracoccus nor the Saprospiraceae-clones visualised by FISH in the SBR biomass. FISH combined with microautoradiography independently confirmed that the DEN67-targeted cells were the dominant bacterial group capable of anoxic [14C] methanol uptake in the enriched biomass. As observed in full-scale operations, the methanol-fed SBR experienced a lag period of several weeks before denitrification performance increased. Using FISH quantification, it was shown that this coincided with the lag phase in the growth of the DEN67-targeted denitrifying population. It was therefore concluded that the Methylophilales bacteria dominant in our SBR system are likely to be important in full-scale methanol-fed denitrifying sludges. The acetate utilising microbial consortium in activated sludge was investigated without prior enrichment using stable isotope probing (SIP). 13C-acetate was used in SIP to biomark the DNA of the denitrifiers. The extracted 13C-DNA fraction was subjected to a full cycle rRNA analysis. The dominant 16S rRNA gene phylotypes in the 13C-library were closely related to bacterial families Comamonadaceae and Rhodocyclaceae of class Betaproteobacteria (96-97% sequence identities). Seven oligonucleotide probes (DEN444, DEN220, DEN581, DEN441, DEN124, DEN220a and DEN1454) for use in FISH was designed to specifically target the identified phylotypes. Application of these probes on the sludge of a continuously fed denitrifying sequencing batch reactor (CFDSBR) operated over a duration of 16 days indicated a strong correlation between the level of CFDSBR denitrification and relative abundance of all probe-targeted bacteria in the CFDSBR community. FISH combined with microautoradiography (FISH-MAR) further confirmed that the DEN581- and DEN124-targeted cells dominating the CFDSBR were capable of taking up [14C] acetate under anoxic conditions. The initial occurrence of the DEN444- and DEN1454-targeted bacteria and the final dominance of DEN581- and DEN124-targeted bacteria in the CFDSBR community were likely related to the changing in-reactor nitrite concentrations during the first few days of CFDSBR operation. Hence, the DEN444- and DEN1454-targeted bacteria were hypothesised to have low affinities for nitrite while DEN124- and DEN581-targeted bacteria have higher nitrite affinities. However, it was clear that all probe-targeted bacteria were denitrifiers capable of utilising acetate as a carbon source. The rapid increase in numbers of the probe-targeted organisms positively correlates with the immediate increase in denitrification rates. The rapid response and community shifts observed when acetate was used to enhance denitrification suggest that an intermittent application of acetate is quite effective to temporarily enhance the denitrification capacity of a treatment plant. However, the importance of a bacterial impact assessment of activated sludge subjected to intermittent acetate supplementation is recommended prior to the wide use of acetate in the wastewater industry. The acetate utilising denitrifying microbial communities investigated in the previous chapter were characterised according to their eco-physiological properties using the r- and K-selection criteria. The electron donor (acetate) and acceptor (nitrite) affinities of these probe-identified denitrifiers were used as traits for this characterisation. The substrate to microorganism (S/M) ratio was manipulated to provide high and low substrate concentrations in the reactor to create conditions favourable for r- and K-strategists, respectively. Two factors, namely feeding regimes and sludge retention times, were studied to achieve the desired S/M ratios and enable r/K characterisation. The high substrate affinities and high specific growth rates of two probe-identified denitrifiers (DEN124 and DEN581) did not enable resolution of these two organisms with the feeding regimes used in this study. However, the application of different sludge retention times as a control strategy to maintain constant high and low in-reactor S/M ratios enabled characterisation of the two probe-targeted denitrifiers DEN124 and DEN581 as K- and r-strategists, respectively. The in-reactor S/M ratios applied in this study did not facilitate the characterisation of populations targeted by probes DEN444 and DEN1454. The minor fluctuations of the S/M ratios during a cycle in the SBR operation was considered as a drawback, but conclusive results could still be obtained from the study. A chemostat reactor operation with constant loading and variable flow rates is suggested as an alternative. Conclusively, this study was able to identify specific groups of denitrifying microorganisms in activated sludge when exposed to acetate and methanol. Unlike most previous studies, which relied on culture dependent methods, this study adopted a pure culture independent approach to identify microorganisms in relation to their function, i.e. denitrification. Moreover, acetate denitrifiers were in situ characterised based on eco-physiological properties. The identification of denitrifying communities in this study has paved the way to a larger research project on the optimisation of denitrification processes with external acetate, methanol and other carbon supplements. As such, this study has contributed significantly to the understanding of the denitrification processes by linking process data with microbial investigations.

Continental margin sediments play an important role in marine biogeochemical cycles, partly due to high primary production rates and efficient export of organic matter to sediments in margin environments. This thesis presents studies of solute exchange in fine-grained sediments representative of slope and rise environments, and carbon and nitrogen cycling in sandy sediments dominant in continental shelves worldwide. Results of these studies advance understanding of the role of benthic processes on marine ecosystems. In fine-grained sediments, solute exchange by diffusion, biological mixing and bioirrigation can be quantified using in situ flux chambers with inert tracer additions. Mechanistic models of chamber tracer transport across the seabed indicate that in organic-rich sediments, bioirrigation and mixing dominate over a wide range of bottom water oxygen levels, reflecting the patchiness and versatility of benthic macrofaunal communities. Positive correlations between benthic oxygen and tracer fluxes appear site-specific. Reliable chamber volume estimates derived from mechanistic models reveal that empirical fits to chamber tracer datasets may overestimate chamber volume and benthic solute fluxes. The biogeochemistry of sandy, highly permeable sediments that dominate continental shelves is largely unknown because of the difficulty in sampling and studying them under natural conditions. Novel sediment reactors were developed and used to mimic in situ porewater advection and natural sedimentary conditions. Compositional changes of natural seawater, with and without the addition of

Το Πι-κριτήριο των Bittanti et al. (1973) έχει χρησιμοποιηθεί εκτενώς σε εφαρμογές με στόχο την πρόβλεψη ενδεχόμενης βελτίωσης της απόδοσης ενός μη γραμμικού συστήματος υπό περιοδική είσοδο. Το κριτήριο όμως έχει τοπική ισχύ και περιορίζεται σε περιοδικές διαταραχές μικρού πλάτους. Η παρούσα εργασία αναπτύσσει μια μέθοδο προσδιορισμού διορθώσεων υψηλότερης τάξης στο πι-κριτήριο, προερχόμενη από βασικά αποτελέσματα της θεωρίας κεντρικής πολλαπλότητας (Center Manifold theory). Η προτεινόμενη μέθοδος βασίζεται στην επίλυση της μερικής διαφορικής εξίσωσης της κεντρικής πολλαπλότητας με χρήση δυναμοσειρών. Το τελικό αποτέλεσμα της προτεινόμενης προσέγγισης είναι ο κατά προσέγγιση υπολογισμός του δείκτη απόδοσης υπό μορφή σειράς, η οποία παρέχει ακριβή αποτελέσματα σε μεγαλύτερα εύρη. Η προτεινόμενη μέθοδος εφαρμόζεται σε έναν συνεχή αντιδραστήρα πλήρους ανάδευσης (CSTR), όπου στόχος είναι η μεγιστοποίηση της παραγωγής του επιθυμητού προϊόντος. Κατασκευάστηκε αλγόριθμος που προβλέπει την μόνιμη κατάσταση στην οποία καταλήγει ένα σύστημα απομάκρυνσης αζώτου που αποτελείται από αντιδραστήρα εμβολικής ροής και δεξαμενή δευτεροβάθμιας καθίζησης με ανακύκλωση. Με χρήση υπολογιστικού μοντέλου βασιζόμενο στο ASM3 υπολογίστηκαν οι μόνιμες καταστάσεις αυτού του συστήματος για ένα εύρος καταστάσεων λειτουργίας. Βρέθηκαν οι βέλτιστες τιμές των βαθμών ελευθερίας για την ελαχιστοποίηση του συνολικού αερισμού και για την ελαχιστοποίηση του συνολικού αζώτου στην απορροή. Και στις δύο περιπτώσεις στις βέλτιστες μόνιμες καταστάσεις παρατηρήθηκε έκπλυση των Nitrobacter δηλαδή παράκαμψη της παραγωγής των νιτρικών. / The frequency-dependent Pi criterion of Bittanti et al. (1973) has been used extensively in applications to predict potential performance improvement under periodic forcing in a nonlinear system. The criterion, however, is local in nature and is limited to periodic forcing functions of small magnitude. The present work develops a method to determine higher-order corrections to the pi criterion, derived from basic results of Center Manifold theory. The proposed method is based on solving the Center Manifold partial differential equation via power series. The end result of the proposed approach is the approximate calculation of the performance index in the form of a series expansion, which provides accurate results under larger amplitudes. The proposed method is applied to a continuous stirred tank reactor, where the yield of the desired product must be maximized. An algorithm was constructed, that predicts the steady state of a nitrogen removal system consisting of a plug flow reactor and a secondary clarifier with recycle. Using a numerical model based on ASM3 and a grid of degrees of freedom, the steady states of this system were calculated. The optimal values for minimizing the total aeration were found, as well as those for minimizing the total nitrogen exit flow. In both cases the Nitrobacter bacteria were washed out thus indicating the bypassing of nitrate production.

From the Proceedings of the 1977 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 15-16, 1977, Las Vegas, Nevada / This study, conducted to determine the interrelationships among nitrogen transformations, infiltration rates, and development of the black layer found in the Santa Cruz River downstream of the Tucson (Arizona) sewage treatment plant, tested these interrelationships by percolating sewage effluent through clear acrylic columns uniformly packed with river sand for the first run, with gravel for the second run. Sewage effluent was continuously applied to three of the columns for 28 and 64 days during the first and second runs respectively. The remaining column was continuously flooded with tap water to serve as a control. Infiltration rates decreased rapidly upon application of the sewage, and within a few days a black layer developed, its thickness inversely related to the infiltration rate but not a cause of reduced flow, which is attributed, rather, to clogging of the surface by suspended solids. There was an average reduction in total nitrogen of 62.9% for the first run, and 15.9% for the second. The mechanisms of removal for run 1 were predominately absorption and denitrification, whereas the predominate removal mechanism in run 2 was filtering of organic nitrogen with adsorption and denitrification also playing an important role.

No description available.

Les sédiments sont des sites importants d’élimination d’azote (N) puisqu’ils possèdent des gradients d’oxydoréduction leur conférant les conditions idéales pour les réactions microbiennes de transformation de N. L’eutrophisation des régions côtières peut altérer ces gradients, par des changements dans la concentration d’oxygène (O2) de l’eau interstitielle, et modifier l’importance relative des processus transformant le N. Afin de mieux comprendre comment l’O2 pourrait influencer les transformations de N, nous avons mesuré les flux diffusifs de diazote (N2), nitrate (NO3-), oxygène et ammonium (NH4+) dans les sédiments de l’Estuaire Maritime du St-Laurent (EMSL), et nous avons estimé les taux de dénitrification. L’importance du couple nitrification-dénitrification en fonction d’un gradient de concentrations d’O2 dans la zone d’hypoxie de l’EMSL fut aussi évaluée. La concentration des gaz dissous fut mesurée en utilisant une nouvelle approche développée dans cette étude. Les flux diffusifs de N2, O2, NO3- et NH4+ variaient de 5.5 à 8.8, de -37.1 à -84.8, de -4.0 à -5.8 et de 0.6 à 0.8 μmol N m-2 h-1 respectivement. Les concentrations de N2 et NO3- dans l’eau porale et les flux de NO3- et de N2 des sédiments, suggèrent que la diffusion de NO3- provenant de l’eau à la surface des sédiments ne peut pas expliquer par elle-même la production de N2 observée. En utilisant une approche stoichiométrique, les taux de nitrification potentielle estimés comptent pour 0.01 à 52% du flux total de NO3 nécessaire pour aboutir aux flux de N2 observés et diminuent avec l’augmentation de l’hypoxie. / Sediments display strong redox gradients and represent important sites of various microbially mediated nitrogen (N) transformation pathways resulting in the loss of fixed N. Increased eutrophication of coastal systems will likely impact the overall function of these sediments via changes in the oxygen (O2) concentration in the overlying water, thus influencing the redox gradient and the relative importance of different N processes. To have a better understanding of how O2 could influence N transformations, we measured dinitrogen (N2), nitrate (NO3-), oxygen and ammonium (NH4+) diffusive fluxes in the sediments of the Lower St. Lawrence Estuary (LSLE) and estimated denitrification rates. The importance of the nitrification-denitrification couple as a function of a gradient of hypoxic O2 concentrations was also evaluated. Dissolved gas concentrations were measured using a new approach developed in this study. N2, O2, NO3- and NH4+ diffusive fluxes varied respectively from 5.5 to 8.8, from -37.1 to -84.8, from -4 to -5.8 and from 0.6 to 0.8 μmol N m-2 h-1 among the five stations sampled. N2 and NO3- concentrations in sediment pore water and measured fluxes of NO3- in and N2 out of the sediments suggested that NO3- diffusion from the overlying water alone could not support N2 production. Using a stoichiometric approach, estimated potential nitrification rates varied among sites, from 0.01 to 52% of the total NO3 supply rate. The relative importance of nitrification decreased with increasing hypoxia.