The project undertaken in this thesis was performed at MittSverige Vatten och Avfall (MSVA) in cooperation with the Swedish Environmental Research Institute (IVL). The effect of temperature on nitrogen removal using moving bed biofilm reactor (MBBR) was controlled using two separate lines by investigating the influence of temperature on nitrification, pre-denitrification, and post- denitrification in a pilot plant scale. By monitoring the process using chemical analysis, batch tests and online systems, it was possible to observe the changes occurring in the different reactors along with different parameter set-ups and react to the undesirable changes in the system. Due to some limitations, the master's thesis couldn’t fully answer the objective of the study. Nonetheless, the processes were investigated and evaluated to better understand the possibilities and limitations of nitrogen removal with MBBR with treating cold wastewater. During the study period, the inflow coming to the pilot plant, recirculation flow, external carbon dosing, set points of dissolved oxygen and temperature values were controlled. By manipulating these different parameters, it was possible to observe their impacts on the processes. The inflow was variable between 0.42 and 0.67 m3/h, recirculation flow was set to 1.2 m3/h at the beginning then changed to function same as the inflow after being reduced to 0.84 m3/h in between. DO level was variable throughout the entire study period with an average concentration of 4.2mg/l in line 1 and 4.43 mg/l in line2 in all the aerated zones for each line. Temperature of incoming water varied between 5.94 and 9.3 oC with an average temperature of 7.73 oC while the average temperature of the water in the system in line 1 was 13.3 oC and 12.62 oC in line 2. During the 5 months study period, the system achieved an average total nitrogen removal efficiency of 46% in line 1 and 60% in line 2. / Projektet i detta examensarbete utfördes vid MittSverige Vatten och Avfall (MSVA) i samarbete med Svenska Miljöinstitutet (IVL). Effekten av temperatur på kväveborttagning med användning av biofilmreaktorer med rörliga bärare (MBBR, eng. för Moving Bed Biofilm Reactor) kontrollerades genom att undersöka temperaturens inverkan på nitrifikation, pre-denitrifikation och post- denitrifikation i två separata linjer i en pilotanläggningsskala. Genom att övervaka processen med hjälp av kemiska analyser, batchtester och online-system, var det möjligt att observera de förändringar som sker i de olika reaktorerna tillsammans med olika parameteruppsättningar och reagera på de oönskade förändringarna i systemet. På grund av vissa begränsningar kunde masteruppsatsen inte helt svara på syftet med studien. Ändå har processerna undersökts och utvärderats för att bättre förstå möjligheterna och begränsningarna av kväverening med MBBR vid rening av kallt avloppsvatten. Under studieperioden kontrollerades inflödet till pilotanläggningen, recirkulationsflöde, extern koldosering, börvärden för löst syre och temperaturvärden. Genom att ändra dessa olika parametrar var det möjligt att observera deras inverkan på processerna. Inflödet var variabelt mellan 0,42 och 0,67 m3 /h, recirkulationsflödet sattes till 1,2 m3/h i början och ändrades sedan till att fungera som inflödet efter att ha reducerats till 0,84 m3 /h däremellan. DO- nivån var variabel under hela studieperioden med en genomsnittlig koncentration på 4,2 mg/l i linje 1 och 4,4 mg/l i linje 2 i alla luftade zoner för varje linje. Temperaturen på inkommande vatten varierade mellan 5,9 och 9,3 oC med en medeltemperatur på 7,7 oC medan medeltemperaturen på vattnet i systemet i linje 1 var 13,3 oC och 12,6 oC i linje 2. Under den 5 månader långa studieperioden uppnådde systemet en genomsnittlig total kvävereduktionseffektivitet på 46 % i linje 1 och 60 % i linje 2.

<p dir="ltr">Excess nitrogen loading to surface waters and groundwater from intensive agriculture threatens human and ecosystem health and economic prosperity within and downstream of the Mississippi River Basin. Restoring wetlands and floodplains reduces nitrogen export, but nitrogen export from the Mississippi River Basin remains elevated. Engineering restored wetlands and floodplains to have higher areal denitrification rates is necessary to advance toward nitrogen reduction goals. Environmental controls of denitrification in restored ecosystems must be further investigated to determine under what conditions denitrification is highest and to link these optimal conditions to restoration approaches. Yet, restoration efforts to reduce nitrogen export may inadvertently increase phosphorus export and greenhouse gas emissions. We evaluated different restoration design approaches and identified environmental controls of denitrification, phosphorus release, and greenhouse gas production to advance knowledge of how floodplain and wetland restorations can be designed and managed to maximize denitrification while also constraining phosphorus release and greenhouse gas production. Comparisons of different restoration design approaches in the Wabash River Basin in Indiana, U.S.A., demonstrated that a hydrologically connected floodplain with row crop agriculture provides limited N treatment. Floodplain restorations that involved structural modifications to enhance hydrologic connectivity supported higher denitrification than restorations that only reestablished native vegetation. Investigations of the plot- and field-scale drivers of denitrification indicated that enhanced hydrologic connectivity and specific native wetland and prairie vegetation types were associated with soil conditions that supported high denitrification potential, mainly sufficient soil moisture and bioavailable organic matter. These same soil conditions were associated with increased risks of phosphorus release and greenhouse gas production. However, artificial flooding experiments showed that preventing prolonged flooding has a strong potential to reduce phosphorus export from floodplains with limited impacts on nitrogen treatment. Microcosm experiments with plant litter and wetland soils indicated that certain wetland vegetation types may reduce greenhouse gas production without sacrificing nitrogen removal capacity based on differences in plant biomass composition.</p>

Examensarbetets syfte var att undersöka möjligheter till denitrifikation på sandfilter (SF) i industriell skala och hur det jämför sig med konventionella metoder för att uppnå denitrifikation och effekten av förändringen på processen vid Bromma reningsverk. Parametrar som undersöktes var rening av ammonium och nitrat, luftning, temperatur, pH, rening av fosfor, rening av organiskt material, rening av suspenderat material, förbigång av vatten vid SF och aktivslamanläggning (ASA), drifttider för SF:arna och total vattenbelastning på ASA och SF. Mätningarna gjordes med aCurve, ett online internt program vid Bromma reningsverk där onlineanalysatorer och driftsparametrar loggades. Nitrat- och nitritvärdena kontrollerades med standardiserade kyvettmetoder. Reningen av totalkväve uppnådde värden på 0,62-0,94 när denitrifikation på SF implementerades. Ökad igensättning av SF inträffade på grund av dosering av metanol på SF och tillväxt av mikroorganismer som orsakade en ökad risk för förbigång av SF, särskilt under hög vattenbelastning. Reningen av suspenderat material, organiskt material och fosfat påverkades inte av implementeringen av denitrifikation på SF. Resultaten av examensarbetet kan användas för att ytterligare optimera konventionella metoder för kväverening i reningsverk. För ytterligare studier kunde igensättning av SF vid användning av metanoldosering på sandfilter undersökas och metoderna kan jämföras i en kontrollerad miljö. / This thesis work aimed to investigate opportunities of denitrification on sand filters (SFs) at industrial scale and how it relates to conventional methods of achieving denitrification and its effect on the process at Bromma WWTP. Investigated parameters were ammonium and nitrate removal, oxygen and aeration, temperature, pH, phosphorus removal, removal of organic material, removal of suspended solids, water bypass of SF and activated sludge facility (ASF), operational times of the SFs and total water load on ASF and SF. The measurements were made with aCurve, an online internal program at Bromma WWTP where online analyzers and operational parameters were logged. The online values for nitrate and nitrite were controlled using standard cuvettes methods. Tot-N removal achieved values of 0.62-0.94 for denitrification on SF. Increased clogging of SFs occured due to dosage of methanol on SFs and growth of microorganisms causing an increased risk of SF bypass especially during high water load. Removal of TSS (total suspended solids), BOD7 and phosphate were unaffected by the implementation of denitrification on SFs.  The findings of the thesis can be used to further optimize conventional methods of nitrogen removal in WWTPs. For further studies the clogging of SFs when using methanol dosage on sand filters could be investigated as well as comparing the methods in a controlled environment.

Nitrous oxide (N2O) is a greenhouse gas (GHG) with 300 times the radiative forcing in the atmosphere of carbon dioxide (CO2), and has recently become a subject of great concern because the nitrogen (N) fertilizers which have been necessary to increase agricultural productivity have also dramatically increased N2O emissions from agroecosystems. Many N control practices have been suggested and implemented in agroecosystems, but their ability to simultaneously remove reactive N from the environment and prevent the production of N2O is, at best poorly understood. The goal of this work is to characterize environmental controls on production of N2O in denitrifying bioreactors. The review portion of this work first discusses the geologic history of the N cycle, how its past and present processes differ, and how it is being affected by human activity. It then explores the N cycle's biochemical pathways, reviews the controls for each of its steps, and discusses the environmental drivers of these controls. The review closes with a discussion of environmental N management strategies. The experimental portion of this work further explores these concepts by observing how biochar amendment and the modification of pH affect N2O production in the denitrification pathway in denitrifying bioreactors. Both pH and biochar have previously been shown to affect N2O production and many N management practices utilize biochar or manipulate pH to increase N retention. The objectives of the experiment were to: 1) Examine headspace N2O concentration in sealed, biochar-amended, denitrifying bioreactors; 2) Determine if the effects of pH on N2O production differ in biochar-amended systems versus controls (under acidic, unbuffered, and buffered conditions); 3) Quantify key denitrification genes (nirK, nirS, nosZ) in each treatment combination. Experimental results showed biochar treatment to significantly increase N2O emissions, a result which runs contrary to most, but not all studies regarding its effects on N2O production. Differences between treatments decreased with increasing pH levels. Biochar did not exhibit significant effects on individual denitrification genes, but it did show influence on the ratios of their populations. On the other hand, pH was found to have significant effects on nirS and nosZ populations. Differences in N2O production between biochar and controls were thus explained by biochar's chemical effects, likely its ability to increase denitrification activity. Developing an understanding of the mechanisms behind these differences will require using a combination of isotope tracing, enzyme assays, and mass balance approaches. Future microbial work in biochar-amended systems should attempt to characterize differences in gene expression, overall community structure, and long-term population trends in the genes of interest. The combination of these approaches should allow researchers to better predict where N2O production will occur and develop strategies to mitigate it while simultaneously increasing food production to meet the demands of a growing population. / Master of Science

Human land use practices such as urbanization and agriculture contribute excess nutrients (nitrogen and phosphorus) and runoff volumes to rivers that degrade aquatic ecosystems and cause a loss of river functions such as nutrient processing and flood attenuation. Floodplain restoration increases floodplain exchange and is commonly implemented to improve water quality and reduce flood impacts at watershed scales. However, the effect of multiple restoration projects at the watershed scale is not well studied. We addressed this knowledge gap by two studies. The first study evaluated the impact of cumulative and spatially varying Stage-0 and bankfull floodplain restoration on nitrate removal in a generic 4th-order Virginia Piedmont watershed for small and sub-annual storm sizes (i.e. 2-year, 1-year, half-year, and monthly recurrence intervals). We used HEC-RAS hydraulics results from a prior study together with a nitrate removal model coded in R. Results indicated that watershed nitrate removal varied depending on the location of restoration in the watershed and where removal was evaluated. The greatest reductions in nitrate loads were observed in the same part of the river network where restoration occurred, with diminished impacts downstream. Removal also increased with increasing stream order/river size. However, removal was generally of small magnitude, with up to 1% or 19% of the watershed load removed for median or 90th-percentile removal rates, respectively. We estimated removal for our restoration scenarios under the Chesapeake Bay Program Protocols and found the removal rate to also be a critical factor in determining the efficiency of restoration project. Other controlling factors for nitrate removal were the amount of restoration and storm size. The second study entailed modeling cumulative restoration in a case study watershed to assess the impacts on nutrient removal and flood attenuation. We built a 1D HEC-RAS model of the 4th-order Gwynns Falls watershed near Baltimore MD using georeferenced HEC-RAS model geometries from the Maryland Department of the Environment and simulated unsteady stormflow hydraulics due to cumulative Stage-0 floodplain restoration for small and sub-annual storms. Restoration actually increased peak flow on the main channel (up to 0.9%) due to slowing of the flood wave on the main channel which was then better synchronized with tributary inflows. Restoration increased nitrate removal but at low levels (up to 0.12% or 2.6% removal for a median and 90th-percentile removal rate respectively) due to the small footprint of restoration in the watershed (up to 21.4% of the main channel was restored). These small and sometimes adverse outcomes occurred in response to what would be expensive restoration. Therefore, we argue for large-scale solutions to address watershed-scale water quality and flooding issues yet acknowledge re-evaluation of restoration goals against other societal priorities may be necessary. Overall, our results highlight the potential value and limitations of floodplain restoration in reducing flooding and nitrate exports at the channel network scale and provide practical insight for application of floodplain modeling at the watershed scale. / Master of Science / Human land use practices such as building cities and farms adds nutrients (nitrogen and phosphorus) and increase storm flows in rivers downstream. While nutrients and flows are needed for humans and wildlife, too much of either can harm aquatic organisms and endanger people and property. Floodplain restoration is a common river engineering technique that increases exchange between the river channel and low-lying areas next to rivers known as floodplains. Floodplains are natural features, but people have reduced river flows between channels and floodplains in many ways. For example, by allowing sediments to build up in floodplains or building levees that separate channels from adjacent floodplains. Increasing floodplain exchange by floodplain restoration is commonly implemented to improve water quality and reduce the impact of flooding in watersheds, which are large areas that drain to a single river. However, while the goals of restoration are often at watershed scales, the effect of multiple restoration projects at that watershed scale is not well studied. We addressed this knowledge gap by two studies. The first study evaluated the impact of multiple restoration projects and project locations in a generic (average/typical) watershed on nitrate removal. We used a nitrate removal model and the results from a prior study that modeled the stormflow behavior resulting from floodplain restoration. Results indicated that watershed nitrate removal varied depended on the location of restoration in the watershed and where removal was evaluated. The most nitrate was removed where restoration occurred, with less removal downstream in the watershed. Removal also increased with increasing river size. However, removal was generally small with up to 1% or 19% of the watershed load removed for a smaller and larger nitrate removal rate, respectively. Other factors that changed the amount of nitrate removed were the amount of restoration, nitrate removal rate in the floodplains, and storm size. The second study entailed modeling cumulative restoration in a case study watershed to assess the impacts on nitrate removal and reducing flooding. We modeled stormflow for multiple hypothetical restoration projects in the Gwynns Falls watershed and found that restoration can actually increase peak flow when placed in certain locations. Restoration increased removal but at low levels (up to 0.12% or 2.6% for a smaller and larger removal rate) due to the small amount of restoration simulated. These small and sometimes adverse outcomes occurred in response to what would be expensive restoration projects to construct. Therefore, we argue for large-scale solutions to address watershed-scale water quality and flooding issues yet acknowledge that re-evaluation of restoration goals against other societal priorities may be necessary. Overall, our results highlight the potential value and limitations of using floodplain restoration to reduce flooding and nutrient exports and provide practical insight for using our modeling techniques in managing watershed flows and pollution.

Periodic releases from an upstream dam cause rapid stage fluctuations in the Colorado River near Austin, Texas. These daily pulses modulate fluid exchange and residence times in the hyporheic region, where biogeochemical reactions are pronounced. We installed two transects of wells perpendicular to the river to examine in detail the reactions occurring in this zone of surface-water and groundwater exchange. One well transect recorded physical water level fluctuations and allowed us to map hydraulic head gradients and fluid movement. The second transect allowed for water sample collection at three discrete depths. Samples were collected from 12 wells every 2 hours for a 24-hour period and were analyzed for nutrients, carbon, major ions, and stable isotopes. The results provide a detailed picture of biogeochemical processes in the bank environment during low flow/drought conditions in a regulated river. Findings indicate that a pulse that causes a change in river stage of approximately 16-centimeters does not cause significant mixing in the bank. Under these conditions, the two systems act independently and exhibit only slight mixing at the interface. / text

<p>This master thesis in energy optimization was made during the autumn of 2006 at Käppala wastewater treatment plant in Lidingö, Stockholm. A preceding thesis, where all electricity consumption was mapped, showed that the aeration in the biological treatment is the single largest consumer in the plant, and it is therefore of interest to reduce this cost. The oxygen control strategy used at Käppala WWTP is working well from a nutrient removal point of view, but not from an economic one. The last aerobic zones have a very low oxygen consumption during low loading periods which give rise to enhanced dissolved oxygen concentrations with excessive costs and reduced denitrification as a result. But also during periods of normal loading unnecessary high oxygen concentration are sometimes given.</p><p>By modifying the aeration control strategy three full-scale experiments have been made, with the intention to reduce the air consumption. The experiments were carried out during week 37-50 in the autumn of 2006 and showed that savings could be made.</p><p>The regular oxygen control at Käppala WWTP controls the oxygen level in the aerobic compartment with two DO-setpoints; one in the first aerobic zone and one in the last. The zones in between are controlled by an airflow fractionation depending on the oxygen level in the first and last zone. In the first strategy to be evaluated, all four zones in the aerated part were individually controlled with its own setpoint. Two different setpoint combinations were tested. By using the fact that the efficiency in the oxygen transfer rate was higher at low airflows, savings of approximately 16 % were achieved. In the second strategy tested, an ammonia-feedback control combined with a DO-feedback controlled the DO-set point in the first aerobic zone. This strategy adjusted the DO- set points to the loading variations, and this gave a decreased airflow of approximately 9 %. Finally the two strategies were combined. All zones were then controlled individually with DO-set points set by an ammonium-feedback and a DO-feedback. The strategy gave savings in the airflow of approximately 18 %. In all three trials the aerated zones were more efficiently used, and the estimated savings are 550 000 SEK/year, and with a preserved nutrient removal efficiency.</p> / <p>Under hösten 2006 har ett examensarbete om energieffektivisering på Käppalaverket på Lidingö utförts. Ett föregående examensarbete där all elenergiförbrukning kartlades visade att blåsmaskinerna i biosteget står för den enskilt största förbrukningen i verket och det är därför av intresse att minska denna kostnad. Syrestyrningsstrategin som används på Käppalaverket fungerar mycket bra ur reningssynpunkt, men är inte optimal ur energisynpunkt. Dels luftas de första aeroba zonerna för mycket vid låg belastning vilket ger upphov till kraftigt förhöjda syrekoncentrationer i de sista aeroba zonerna med höga luftningskostnader och risk för försämrad denitrifikation, men även under normal belastning har det visat sig att onödigt höga syrekoncentrationer ibland ges.</p><p>Tre fullskaliga optimeringsförsök har utförts, med syfte att minska luftförbrukningen med bibehållen reningsgrad. Försöken pågick från vecka 37 till 50 hösten 2006, och visade att det finns möjlighet att spara energi genom att modifiera syrestyrningsstrategin.</p><p>Den reguljära syreregleringen i Käppalaverket styr syrehalten i den aeroba bassängen mot två syrebörvärden; ett i den första luftade zonen och ett i den sista. Luftflödet till de mellanliggande zonerna styrs av luftflödesandelar beroende på syrehalten i dessa två zoner. Den första strategin som utvärderades styrde istället samtliga zoner individuellt med egna börvärden, där två olika strukturer på de satta börvärdena användes. Genom att utnyttja en högre effektivitet i syreöverföringshastigheten vid låga luftflöden uppnåddes luftflödesbesparingar på ca 16 % i första försöket.</p><p>I den andra strategin styrdes syrebörvärdet i den första luftade zonen med hjälp av två återkopplingar, en från utgående ammoniumhalt och en från syrehalten i den sista luftade zonen.</p><p>Tack vare att strategin anpassade syrebörvärdena efter belastningen av syretärande ämnen erhölls luftflödesbesparingar på ca 9 %. Slutligen kombinerades de två strategierna; samtliga zoner styrdes individuellt med börvärden satta av en ammonium-återkoppling och en syre-återkoppling. Strategin medförde luftflödesbesparingar på ca 18 %. I samtliga försök utnyttjades de luftade zonerna bättre, och besparingspotentialen uträknad från 2005 års elpriser blev som mest 550 000 SEK/år, detta med en bibehållen reningsgrad.</p>

<p>The euthropthication of the Baltic Sea is a threat that is beginning to be taken seriously by the governments concerned. In Sweden, regulations concerning the allowed nitrogen (N) concentration in the effluent water from wastewater treatment plants are being tightened up. The Rimbo wastewater treatment plant has been imposed to reduce the annual mean concentration of total N in the effluent water to levels below 15 mg l-1. A more and more common way to reduce the nitrogen level in wastewater is to let the water pass through a wetland. This study investigates the possibility to build this kind of wetland at the outlet of the Rimbo wastewater treatment plant.</p><p>A prestudy of the topography, soil characteristics and groundwater flow indicates that the land area in question is well suited for the construction of a wetland. A proposal for the design has been made by using a physically based computer model developed in the PRIMROSE project (PRocess based Integrated Management of constructed and Riverine wetlands for Optimal control of wastewater at catchment ScalE), which is financed by the EU. Analysis of the residence time distribution (RTD) is a tool for understanding wetland design characteristics and can be used for wetland engineering such as optimizing design for best possibleefficiency in nitrogen removal. In order to characterize the performance of a wetland, it is useful to translate the RTD to a key figure representing the treatment efficiency. In this work, two types of such key figures have been used. Key figure 1 gives the hydraulic efficiency and Key figure 2 gives an estimation of the nitrogen retention by an integration of hydraulic characteristics and the chemical transformation of nitrogen.</p><p>The results of this study show that constructing a wetland in Rimbo probably would be an efficient way to reduce the nitrogen level at the effluent of the wastewater plant below the limits of the regulations. In addition, a wetland would form a nice place of recreation for the people in Rimbo and also make a good habitat for birds.</p> / <p>Övergödningen i Östersjön är ett problem som uppmärksammas alltmer. Ett led i att minska kvävebelastningen på Östersjön är ökade krav på rening i de svenska kommunala reningsverken. Rimbo avloppsreningsanläggning har blivit ålagd ett riktvärde för totalkväve på 15 mg l-1 i utgående vatten, vilket motsvarar en reningsgrad som inte uppnås idag. Ett alltmer vanligt sätt att minska föroreningshalterna är att anlägga våtmarker i anslutning till reningsverken för att efterpolera spillvattnet. Det här arbetet är en del av en förstudie till en sådan våtmark i Rimbo.</p><p>En förundersökning av topografi, jordart och grundvattenflöden indikerar att det område som föreslagits i anslutning till reningsverket i Rimbo lämpar sig väl för ett våtmarksbygge. Förslag till utformning har tagits fram med hjälp av en fysikaliskt baserad modell över vattenströmning, utvecklad inom det EU-finansierade projektet PRIMROSE (PRocess based Integrated Management of constructed and Riverine wetlands for Optimal control of wastewater at catchment ScalE). Analys av vattnets uppehållstidsfördelning ger förståelse för våtmarkens egenskaper och kan därför användas vid t ex optimering av våtmarksdesign med avseende på kväverening. För att på ett enkelt sätt kunna jämföra olika våtmarkers effektivitet är det praktiskt att översätta uppehållstidsfördelningen till ett nyckeltal för reningseffekten. I det här arbetet har två olika sådana nyckeltal beräknats. Det första ger den hydrauliska effektiviteten och det andra bygger på en metod där våtmarkens interna hydraulik integreras med den kemiska omvandlingen av kväve. Nyckeltal 1 ger ett mått på hur stor del av volymen i våtmarken som används för kväverening, medan Nyckeltal 2 ger ett mått på den procentuella kväveavskiljningen.</p><p>Den här förstudien visar att en våtmark sannolikt skulle vara ett utmärkt sätt att klara riktvärdet för kvävehalten vid reningsverket i Rimbo. Därutöver skulle en våtmark kunna utgöra ett positivt inslag i landskapet och öka den biologiska mångfalden, inte minst vad gäller fågelliv.</p>

Dans le cadre de l’évaluation multicritère des systèmes en protection intégrée contre les adventices (PIC-Adventices), cette étude a eu pour objectif d’étudier l’impact de la combinaison de pratiques alternatives à l’usage d’herbicides sur les émissions de N2O par les sols. L’étude a été conduite sur quatre systèmes de culture, un système de référence en agriculture raisonnée (S1) et trois systèmes PIC-Adventices (S2, S3, S5), sur le site expérimental Dijon-Epoisses (47°20'N, 5°2'E). La mise en place de dispositifs de mesure à haute résolution, alliant la méthode des chambres d’accumulation de gaz en surface du sol aux analyses de gaz par infrarouge, a permis d’estimer l’intensité des émissions de N2O pour les sols argilo-calcaires caractéristiques de la plaine dijonnaise, avec des flux moyens compris entre -6 et 257 g N-N2O ha-1 jour-1. Cette intensité s’est avérée étroitement liée à certains paramètres environnementaux (température, part de porosité occupée par l’eau, teneur en azote), et certaines pratiques agricoles, induisant des différences significatives entre les systèmes. Les émissions de N2O ont notamment été affectées par le mode de gestion des sols et la nature des cultures constituant la succession de chaque système. L’absence de travail du sol depuis 2008 dans le système S2 s’est traduite par une amplification des émissions de N2O d’un facteur multiplicatif d’au moins six par rapport aux autres systèmes pour l’année 2012. Par ailleurs, ces émissions équivalentes à 0.7 t C-CO2 ha-1 pourraient avoir compensé la séquestration de carbone additionnelle induite par le travail réduit du sol pendant la période de mesure. De la même manière, la dégradation des résidus de culture légumineuse, dans les systèmes S3 et S5 ont respectivement stimulé les émissions de N2O en 2011 et en 2012, en comparaison avec le système de référence S1. Par ailleurs, le suivi mensuel de la taille des communautés microbiennes du sol a également permis de mettre en évidence une variabilité temporelle liée aux pratiques culturales (travail du sol, récolte), en plus de la variabilité saisonnière. En revanche, dans cette étude les variations de taille des communautés microbiennes nitrifiantes et dénitrifiantes ne sont pas apparues comme un facteur d’explication de la variabilité des émissions de N2O, quel que soit le système. Enfin, l’approche par la modélisation des émissions de N2O sur les quatre systèmes, à l’aide des modèles NOE et DNDC, a permis d’identifier les principaux facteurs de régulation de ces émissions. De surcroît, le modèle NOE a bien identifié le système de culture S2, sans travail du sol, comme étant le système le plus émetteur de N2O, conformément aux observations de terrain. Cette étude renforce donc la pertinence de la modélisation pour estimer et analyser les émissions de N2O dans différents systèmes de culture. / In the context of the multicriteria evaluation of Integrated Weed Management (IWM) cropping systems, our study highlights the impacts of the combination of alternative agricultural practices to herbicide reliance on N2O emissions from soils. The study was located at the INRA experimental site of Dijon-Epoisses (47 ° 20'N, 5 ° 2'E) and considered 4 cropping systems, a reference system designed to maximize financial returns (S1) and three IWM systems (S2, S3, S5). The setup of high resolution measuring devices, combining the non-steady state chamber method with infrared gas analysis, allowed to estimate the intensity of N2O emissions from the calcareous clay soil, which is specific of Dijon plain, with average fluxes laying between -6 and 257 g N2O-N ha-1 day-1. The intensity of fluxes was closely related to environmental parameters (temperature, water filled pore space, inorganic N amount) and particular agricultural practices, leading to significant differences between systems. N2O emissions were notably affected by soil management and nature of crops. The conversion of S2 system to no-till in 2008 increased the N2O emissions, fluxes being at least 6 times more important than those emitted by the other cropping systems, during the year 2012. Moreover, these emissions, equivalent to 0.7 ha-1 C-CO2 ha-1, could have offset the soil additional carbon sequestration induced by reduced tillage during the studying period. Similarly, the degradation of legume crop residues in the S3 and S5 systems respectively enhanced N2O emissions in 2011 and 2012, by comparison with the S1 reference system. Elsewhere, the monthly monitoring of the soil microbial communities sizes has also helped to highlight temporal variability induced by the agricultural practices (tillage, harvest) as well as the seasonal variability. However, in this study the fluctuations of nitrifying and denitrifying communities sizes observed did not explain the variability of N2O emissions, whatever the considered cropping system. Finally, the modeling of N2O emitted from the 4 cropping systems, using NOE and DNDC models, allowed to identify the mains factors regulating the emissions. In addition, the NOE model accurately identified the S2 system, without tillage, as the most N2O emitter, in accordance with field observations. Thus, our study reinforces the relevance of modeling to estimate and explain N2O emissions in different cropping systems.

Vid Hammarby Sjöstadsverk drivs en pilotanläggning som är en mindre skala av det framtida avloppsreningsverket i Henriksdal. Delar av reningsprocessen består av membranbioreaktorer. I pilotanläggningen finns en zon kallad RAS-DeOx dit returslammet från membrantankarna och rejektvattnet från slambehandlingen leds. Luftning av membrantankarna gör att returslammet är syrerikt och rejektvattnet innehåller mycket ammonium.Zonen fungerar som ett nitrifikationssteg då syret i returslammet kan oxidera ammoniumet från rejektvattnet. Dessutom kan zonen minska syrehalterna i returslammet för att undvika att det hamnar i pilotanläggningens fördenitrifikation. För att få bättre uppfattning om vad som sker i zonen och hur denna fungerar som ett nitrifikationssteg studerades nitrifikation, denitrifikation och syreförbrukningen i zonen. För att utvärdera RAS-DeOx-zonen belastades den med olika nivåer av ammonium från rejektvattnet vid olika hög luftning i membrantankarna. Detta utfördes både experimentellt direkt på pilotanläggningen och i en simuleringsstudie där processimuleringar genomfördes i en simuleringsmodell. I modellen utvärderades även två styrstrategier för zonen.Resultaten från studierna visade att både nitrifikation och denitrifikation förekom i zonen. Jämfört med simuleringsstudien varierade omfattningen av nitrifikation och denitrifikation mer i den experimentella studien. Båda studierna visade att det fanns risk att syre hamnande i pilotanläggningens fördenitrifikation. Styrstrategierna visade att det inte var fördelaktigt för pilotanläggningens resurseffektivitet att zonen luftades vid hög ammoniumbelastning från rejektvattnet. Det visade sig också att det var ingen större skillnad när det gällde pilotanläggningens prestation vid högt, lågt eller styrt returflöde för zonen. Däremot skiljde sig prestationen mer avseende luftning och koldosering.Utöver att det förekom nitrifikation och denitrifikation i zonen bedömdes den även fungera som ett nitrifikationssteg för ammoniumet i rejektvattnet. Dessutom minskade den syrehalterna i returslammet. Hur väl ammonium nitrifierades och syre förbrukades i zonen berodde på förhållandena i pilotanläggningen och förhållandet mellan mängden ammonium och syre i zonen. / At Hammarby Sjöstadsverk there is a pilot plant which is a smaller version of the future wastewater treatment plant at Henriksdal. Parts of the treatment process steps are membrane bioreactors. There is a zone in the pilot plant which is called RAS-DeOx to where the return activated sludge from the membrane tanks and the reject water from the sludge treatment are pumped. The return activated sludge contains oxygen since the membranetanks are aerated and the reject water has high contents of ammonium.The zone works as a nitrification step for the ammonium in the reject water, and also reduces the oxygen levels in the return sludge to avoid oxygen in the pilot plant's pre-denitrification step. To aquire a better understanding of what occurs in the zone and how it functions as a nitrification step; nitrification, denitrification and oxygen consumption was studied in the zone. To evaluate the RAS-DeOx-zone it was loaded with different loads of ammonium from the reject water at different levels of aeration in the membrane tanks. This was evaluated both experimentally at the pilot and in a simulation model of the pilot. In the simulation model two control strategies for the zone were also evaluated.It was shown that both nitrification and denitrification occurred in the zone. Furthermore, both studies showed that there is a risk that oxygen occurs in pilot's pre-denitrification step if the ammonium load in zone was low. Then, all of oxygen in the zone was not consumed. As for the control strategies, it was shown that the benefits were low for the resource efficiency of the pilot when the zone was aerated during high loads of ammonium. The performance of the pilot was similiar for high, low and regulated return flows for the zone, but the aeration and carbon dosage in the pilot differed.Besides that both nitrification and denitrification occurred in the zone, it was also functioning as a nitrification step for the ammonium from the reject water. The oxygen from the return sludge was reduced as well. The extent of nitrification and consumption of oxygen in the zone depended on the circumstances in the pilot and the relationship between the amount of oxygen and ammonium.