One way of coping with water shortness is through reusing wastewater. In order to enhance potable water production in sites dealing with water shortness, treated wastewater can be fed into an intermediate storage system, generally a natural water reservoir from which drinking water is produced. This procedure is called indirect potable reuse. Although this technique may be an innovative way of coping with water shortness, little is known about the effect of this procedure concerning environmental and human health. In order to find more about the environmental and health human risks arising from indirect potable reuse technologies a risk assessment was carried over in the French Department of Vendée. This site deals with water shortness from May to October through a significantly enhanced water consumption in the coastal aera of this department caused by the increase of tourism activities and intensified crop irrigation. To achieve this task 35 different organic micropollutants, in its majority pesticides and pharmaceuticals were studied. Therefore, a risk assessment on water reuse technologies was performed at the site of concern. Furthermore, the impact of two tertiary treatments developed within the framework of the European DEMOWARE project on minimising arising risks was also taken into account. Results showed that 16 substances present a potential risk to the environment and/or to human health at the WWTP effluent. Similar results were achieved in fresh water, in this case 14 compounds have a risk quotient higher than 1 indicating a risk for both, health and environment. Furthermore, outcomes from the tertiary treatments risk characterisation showed that none of the studied tertiary techniques is cappable of reducing environmental and health risks for all substances to acceptable levels (RQ < 1). Same results were achieved for the risk characterisation in fresh water.

Powerstep wird auf der KA MS in zwei Etappen realisiert. Der erste Part umfasste die Generierung von maximalen Primärschlammmengen durch den Einsatz einer Flockung und anschließenden Filtration im Anschluss an die mechanische Reinigung. Zielstellung war es mit dem energiereichen Primärschlamm durch Faulung und Verstromung den Endenergiebedarf der KA MS decken zu können und durch die Überführung des wesentlichen Schmutzstoffanteils in den produzierten Schlamm die biologische Abwasserbehandlungsstufe, genauer die Belebung, zu entlasten. Die Reinigungsmethode wurde im Labormaßstab getestet und als erfolgreich bewertet. Allerdings konnte innerhalb der Laboruntersuchungen auf der Kläranlage nachgewiesen werden, dass bei dem neuen Verfahren keine Entfernung des Ammoniums realisiert wird. Da Ammonium einer der wesentlichen Parameter zur Bewertung der Reinigungsleistung einer Kläranlage ist und durch die Abwasserverordnung (AbwV) mit einem Ablaufgrenzwert belegt wurde, wird im zweiten Part von Powerstep eine Technik zur energieeffizienten Ammoniumelimination untersucht. Die Bearbeitung dieser Problematik ist Gegenstand der vorliegenden Arbeit. Das Ziel dieser Arbeit ist es, Methoden zur Ammoniumentfernung bei der Abwasserreinigung zu betrachten und tiefgreifend auf die Elimination durch den Einsatz von Wasserlinsen einzugehen. Dabei werden sowohl ökologische, als auch chemische und technische Belange beleuchtet.

Organische Spurenstoffe, wie Pharmazeutikrückstände, werden durch die konventionellen Reinigungsschritte der kommunalen Kläranlagen nur unzureichend entfernt und können in die aquatische Umwelt sowie ins Trinkwasser gelangen. Ozon stellt eine Möglichkeit zur Elimination dieser Spurenstoffen aus Kläranlagenabläufen dar. Dabei werden Wasserinhaltsstoffe direkt durch Ozon und indirekt über °OH-Radikale oxidiert und transformiert. Für die Ausstattung einer Kläranlage mit einer Ozonanlage erfolgt oft eine aufwändige und langwierige Pilotierung. Daher ist eine Testmethode im Labormaßstab von Nöten, anhand derer Auslegungsparameter für die Dimensionierung einer Ozonanlage abgelesen werden können. Im Rahmen der vorliegenden Arbeit wurde ein einheitliches Testverfahren entwickelt und der Einfluss der Probenlagerung, der Wassertemperatur bei Versuchsdurchführung, der Wassermatrix und des Gehalts an DOC, Nitrit und abfiltrierbaren Stoffen untersucht. Des Weiteren wurden Methoden zur Vorhersage der Spurenstoffelimination mit Ozon überprüft und bewertet. Die Testmethode eines Semi-Batch-Verfahrens, bei der eine Wasserprobe mit einem Sauerstoff/ Ozon-Gemisch begast wird, wurde mit einem Batch-Verfahren, bei dem eine hochkonzentrierte Ozonlösung der Probe zugegeben wird, verglichen. Batchversuche können mit konstanten oder variablen Verdünnung durchgeführt werden. Dabei wird zu einer gleichbleibenden Menge an Probenlösung eine variierende Menge an Ozonstarkwasser gegeben und diese Lösung bei der konstanten Verdünnung auf einen festgelegten Wert mit voll entsalztem Wasser aufgefüllt, während bei der variablen Verdünnung keine Zugabe von voll entsalztem Wasser erfolgt, so dass variierende Gesamtvolumina entstehen. Zur Bewertung der Ozonung wurde der spektrale Absorptionskoeffizient bei 254nm benutzt, da er den Gehalt an Organik in einer Wasserprobe wiederspiegelt. Es wurde festgestellt, dass beide Versuchs- und Verdünnungsmethoden übereinstimmende Ergebnisse der SAK254-Reduktion und Spurenstoffelimination liefern, das Batchverfahren jedoch einen erheblich geringeren Proben- und Zeitaufwand benötigt. Durch eine zweiwöchige Probenlagerung bei 7 °C konnte ein Anstieg des pH-Wertes beobachtet werden, der jedoch nur eine geringe Erhöhung der Elimination mancher Spurenstoffe, wie Benzotriazol, bewirkte. Des Weiteren führte eine Variation der Wassertemperatur bei der Ozonung von 10 - 30 °C bei Batch-Versuchen zu keiner signifikanten Veränderung der SAK254-Reduktion und Spurenstoffelimination. Zwei Kläranlagenabläufe und Trinkwasser mit unterschiedlichen DOC- und Nitritkonzentrationen wurden verglichen. Dabei wurde festgestellt, dass die Korrektur des Ozoneintrags um die Ozonzehrung durch Nitrit und der Bezug auf den DOC ein gutes Instrument darstellt, um unterschiedliche Wassermatrizen vergleichbar zu machen. Während die Kläranlagenabläufe ähnliche Ergebnisse der Spurenstoffelimination liefern, wurde im Trinkwasser eine erhöhte Entfernung von Acesulfam, Benzotriazol, Candesartan, Gabapentin und Valsartansäure gemessen. Dies lässt sich durch die verschiedene Zusammensetzung des organischen Materials und der Menge an als Radikalfängern agierenden Stoffen im Kläranlagenablauf erklären. Zwischen der Spurenstoffelimination in filtriertem (0,45 µm, Cellulosenitrit) und unfiltrietem (AFS = 4 mg/L) Kläranlagenablauf wurde kein Unterschied festgestellt, während eine Zudosierung von suspendierten Stoffen (30 -75 mg/L) eine Verminderung der SAK254-Reduktion sowie der Spurenstoffelimination um bis zu 20% zur Folge hatte. Dies wird durch die Abreaktion des Ozons und der °OH-Radikale an den Partikeln erklärt. Zur Vorhersage der Spurenstoffelimination konnten Diclofenac, Benzotriazol, Valsartan und Iomperol als Indikatorsubstanzen im Kläranlagenablauf identifiziert werden, über die die Elimination weiterer Spurenstoffe genügend genau abgeschätzt werden kann. Darüber hinaus konnte ein linearer Zusammenhang zwischen delta SAK54 und der Elimination jedes Spurenstoffes aufgestellt werden, mit dem die zu erwartende Entfernung bei einem bestimmten Ozoneintrag berechnet werden kann.

Redox condition, in particular the amount of oxygen in groundwater used for drinking water supply, is a key factor for the drinking water quality as well as for the production well’s lifecycle. Thus, a process-based and quantitative understanding about the oxygen fluxes in groundwater systems is fundamental in order to predict e.g. the removal capacity of pollutants or in particular the likelihood of iron-related well clogging. Such well ageing is a major thread for well operators and objective in practice and science. The formation of iron oxides responsible for well clogging is mainly known for wells abstracting groundwater from unconsolidated aquifers with a distinct redox zonation. The accumulation of precipitates is primarily taking place at the slots of the well screens, but also affects aquifers, pumps and collector pipes. Several studies already identified interacting hydro-chemical and microbiological processes as major cause for the development of iron oxides in wells. They develop in the presence of dissolved species of iron and oxygen in the water. The co-occurrence of both, the dissolved iron and oxygen, is the result of a mixing of groundwater with different redox states. The abstraction of groundwater by wells is known to promote such mixing processes. Particularly, frequent water table oscillations with high amplitudes in contrast to natural conditions and managed aquifer recharge measures may deliver oxygen to groundwater. But the impact of different well management strategies on the sources and rates of oxygen delivery to aquifers was not studied in detail so far. Within the thesis presented here, oxygen fluxes to groundwater were qualified and quantified based on statistical, modelling, laboratory and field site studies and their impact on well performance was determined for different well operation schemes and different hydrogeological conditions. Processes were exemplarily investigated for the quaternary aquifers of Berlin, which are the exclusive source for the drinking water supply of the German capital. Analysis of design, operation, geological setting, hydrochemical composition and maintenance activities of Berlin’s drinking water wells illustrated the vulnerability of wells for clogging processes and revealed the relevance of detailed investigations on this topic. A general estimation of the two main oxygen delivering processes influencing groundwater aeration, air entrapment and bank filtration, was done by a generic transport model. Simulation of oxygen fluxes with regard to different hydrogeological and operational boundary conditions revealed air entrapment as major source. Oxygen delivery by bank filtration was subsidiary and strongly depending on flow gradients and permeability of the banks. Air entrapment due to oscillating water tables was quantified by aeration tests in column experiments under laboratory conditions. Results pointed at a downward shift of oxygen caused by repeated oscillations as a consequence of oxygen dissolution and advective transport of dissolved oxygen inside the column. A downward propagation of oxygen into the permanently water-saturated zone was not observed for switching intervals shorter than 24 hours. Such repeated short-termed oscillations led to an enrichment of oxygen, but with a constantly decreasing increment per oscillation. Oxygen degradation was not accounted for in simulation and inhibited in laboratory studies. But, in situ monitoring of oxygen at three selected well sites in Berlin provided a real insight into oxygen fluxes and their effects on well ageing processes under field conditions. The monitoring network included multi-level observation wells and vertical strings of oxygen sensors installed in the aquifer and inside the wells. Thus, it was feasible to measure changes in hydraulic conditions and redox dynamics. Oxygen distribution could be observed as a function of depth and recharge source in a high temporal and spatial resolution for the first time. It was possible to detect traces of oxygen in the well-near aquifer and inside the wells, which are sufficient to oxidize high loads of dissolved iron when supplied constantly. All three well sites showed oxygen distribution patterns, which significantly differed from the others. These variations referred not only to the initial distribution, sampled at idle equilibrium, but also to the progression of oxygen saturation during abstraction and recovery phases. Enrichment and downward propagation of oxygen as result of abstracting water could be observed at all well sites, although absolute concentrations varied strongly between the well sites. By this, it was possible to correlate oxygen variations to hydrogeological boundary conditions. Infiltrating oxic surface water via river, lake or artificial pond banks delivers high amounts of oxygen to the groundwater and can cause an enormous widening of the oxic zone towards the abstracting well. As a result, the oxic/anoxic interface moves downward close to the well once water is abstracted. But, clogging of wells abstracting bank filtrate or artificial recharge strongly depends on the residence times of the filtrate, the hydraulic connection between banks and groundwater and seasonal variations. Only under certain conditions a significant enhancement of clogging can be expected. To directly link well operation, oxygen delivery and ochre formation with well performance development, a well model scaled up to realistic proportions was designed, built and operated with natural groundwater. The tank experiment enabled to study distribution patterns of ochre formation with regard to the different structural zones of the well, including aquifer, filter pack and screen slots and its influence on pressure losses and well performance. It could be shown, that groundwater was enriched with oxygen during the tank passage by oscillating water tables and that permeability and specific well yield generally decreased over time. The distribution of ochre deposits in the well tank showed a distinct mineral zonation with high deposition rates of manganese and iron in the filter pack at the top of the well screen. Further, interfaces of aquifer and/or filter pack were strongly affected by iron deposits. Thus, preventing ochre formation is an appropriate measure. The preventive treatment of wells with hydrogen peroxide could be such a measure, but could also be a potential source for oxygen in well and filter pack. By reviewing the latest research activities and operator’s data and by investigating at laboratory and field site scale, the current treatment procedure was evaluated. Investigations revealed a clear improvement potential for the treatment with hydrogen peroxide. Impacts of the treatment were however low, especially if incrustations were already established. Results of column batch studies and field tests did not fully prove the effectiveness of the preventive treatment, but indicated that with higher concentrated solutions and an improved treatment procedure ochre formation can be retarded and rehabilitation potential can be improved. Another approach to prevent ochre formation is the classification of well sites considering their ageing vulnerability and the development of adapted operation schedules. At least such a measure can support a sustainable construction, operation and maintenance of wells. A statistical approach was used to quantify well ageing and to identify factors promoting well performance loss. Most appropriate clogging indicators could be identified and were used to analyse worst and best site conditions with regard to their impact on ochre formation. Accordingly, a well in high distance to the next surface water with a thick groundwater layer above the well screen situated in a confined aquifer with high redox potential gains the lowest ageing potential. Compared to worst site conditions and calculated for the mean life time of a typical Berlin drinking water well, this can account for a difference in well capacity of up to 90%. In addition to that, optimized rehabilitation intervals for the identified well classes based on their ageing potential could be exemplarily determined. Based on the results of this thesis, strategies for an optimized monitoring of well ageing processes and strategies for an adapted well management aiming at the reduction of ochre formation can be developed.

The aim of this master thesis was to investigate the transferability of the innovative treatment concept CARISMO on a full-scale wastewater treatment plant in a rural area. The idea of CARISMO is integrated in the EU-funded research project Powerstep and focuses on an increased separation of solid substances and their use for energy production with digestion. In the course of this work, principles and conditions for the change of occupancy were determined. Furthermore, preliminary tests were carried out to analyse the wastewater characteristics and the necessary resources for the projected flocculation and filtration stage. The technical feasibility and cost-effectiveness of the treatment concept has been shown with balances of material flows, energy and costs.

The combined sewer overflow issue in the city of Berlin is becoming an increasing threat to the water quality of the surface water bodies, as the number and volumes of combined sewer overflow (CSO) events occurring per year may be on the rise due to climate variations among other aspects. For this reason, a case study was formulated to investigate the implementation of storage tanks in one of Berlin’s sub catchments, Wilmersdorf, in order to reduce the Occurrence of CSO to a once per year event on average. The investigation was made using InfoWorks Collection system (CS), one dimensional urban planning software used widely for sewer system modelling. The network of the Wilmersdorf catchment (majorly consisting of combined sewers, with small portions of separate rain and separate foul sewers) was modelled with the aid of InfoWorks. The implementation of tanks in the network was divided into two main parts: centralized and decentralized tanks. The centralized tanks addressed the issue of CSO, in order to reduce the CSO occurrence to once per year, firstly by using a short design storms representing a one year return period, to implement initial storage volumes, then this network was validated using rain series records for one year (1990) and for thirty years (1980-2010). The decentralized tanks were implemented at much smaller storage volumes compared to the centralized tanks, in localized locations to solve small surface floods in the separate rain sewer system, or to reduce the pollutant load Biological oxygen demand (BOD) of the CSO, by storing water from the separate foul system in the catchment. The results obtained for the centralized storage tanks show major reductions in CSO, with four centralized tanks implemented in central parts of the catchment. The target of once per year CSO event was achieved for the one year rain series (1990), but not for the thirty years rain series (1980-2010). Results for the decentralized storage tanks show reduction of surface flooding for the studied local areas in the catchment, with sometimes a reduction of surface floods also downstream of the targeted areas. On the other hand, the pollutant load (BOD) was reduced by negligible amounts with decentralized tanks at the studied separate foul system locations, with results showing that the overall BOD load reduction in the overflow volume is also accompanied with CSO overall volume reduction.

In recent years several ways of recovering phosphorous from municipal wastewater have been developed. Depending on the applied technology the recovered products vary significantly concerning the concentrations of heavy metals and organic residues. Within the boundaries of data quality and present uncertainties a comparative risk assessment of seven secondary phosphorus fertilizers, sewage sludge, raw ash and triple super phosphate has been conducted for PCDD/Fs, PCBs, PAHs, As, Cd, Cr, Cu, Hg, Ni, Pb and Zn. Local exposure assessment was done using the kinetic model of the European Union’s Technical Guidance Document for all substances accounting for both fertilization and average atmospheric deposition. For substances of concern (Cd and Zn) the exposure was additionally refined using a solute transport model (HYDRUS-1D) and a precipitation model (Visual MINTEQ-software). An annual fertilizer amount equivalent to 60 kg P2O5/ ha × year by these products is assumed. In order to account for potential accumulation a time span of 100 years is modelled. Results indicate that out of the selected 11 (groups of) chemicals only cadmium and zinc are of concern. Regarding soil organisms, zinc is of concern for sludge, raw ash and one of the seven secondary phosphate fertilizers in case of soil-pH above pH 6.0. Regarding groundwater, cadmium and zinc are of concern below pH 6.0 since mobilization at this pH level increase significantly. No risk is expected regarding the endpoint humans. Among the investigated products struvites have shown the lowest phosphorus-specific heavy metal contents. For ash related products more data from full scale operations are needed to reduce still existing uncertainties like the influence of raw wastewater quality and WWTP operation on the final product.

Being a measure to counteract an increasing water scarcity in multiple regions of the world, water reuse is the subject of numerous investigations. The pan-European project DEMOWARE aims at tackling barriers related to water reuse to support further applications. At the project site of Braunschweig in Germany wastewater is reused historically to irrigate agricultural fields. Against the background of an ongoing debate in Germany to establish a legal basis for water reuse, options for a disinfection of secondary treated wastewater were tested at the wastewater treatment plant Steinhof. This thesis investigates the disinfection performance of two pilot scale plants (performic acid (PFA) dosage and UV irradiation) against the background of the compliance with wastewater-related standards, especially DIN 19650 and a WHO guideline regarding wastewater reuse. In order to meet recommendations of the latter, a noro- and rotavirus removal of 1.5 log was recommended by (1) for the site of Braunschweig. E. coli, intestinal enterococci (IE) and Clostridium perfringens were analyzed regarding their abundance and log removal concerning physicochemical properties. Serving as indicator organisms, they were related to the log removal of noro- and rotavirus. Differing doses were tested. The impact of the contact time was investigated for PFA by the presence or absence of sodium thiosulfate (STS) in the sampling bottles whereat the presence of STS terminates the disinfection process. The application of STS in the PFA sampling bottles caused a performance difference of 1.32 log for E. coli, 0.78 log for IE and 0.28 log for C. perfringens. Hence, the PFA reaction was not completed at the effluent sampling spot of the reactor (stirrer tank) and the determined minimum contact time of 3.5 min is not sufficient for PFA to develop its complete disinfection performance. The disinfection reactor did not provide the minimum recommended RT (10 min). Possibly, short circuits and zones with limited substance exchange occur. For full scale implementations, the PFA contact basins need to be designed carefully. A flow channel as used at full scale reference applications is preferable here. The PFA plant showed break-ins of the disinfection performance (to values < 0.5 log) for definable and indefinable reasons. Hence, its disinfection performance is not constant. Further investigations are necessary. Of the 3 doses per method deployed, a UV dose of 44 Wh/m³ and a PFA dose of 2 ppm (= 10 min, without STS) is proper to achieve quality class 3 of the DIN 19650 which is required for the present conditions and applications in Braunschweig. The requirements for a noro- and rotavirus removal of 1.5 log units according to the WHO guideline can be satisfied by a dose of 2 ppm of PFA and 35 Wh/m³, respectively, using an evaluation based on the mean value of the disinfection performance (both PFA and UV). By the application of a threshold-based evaluation the requirements are not satisfied for UV irradiation. Regarding PFA dosing, a dose of 2 ppm is sufficient as long as a sufficient contact time (= 10 min) is provided. The order of sensitivity against the disinfection methods was found for both UV irradiation and PFA dosage the same: E. coli > E. cocci > C. perfringens. Clostridium being used as an indicator for endospore-formers showed a remarkably lower sensitivity against both methods. A doseperformance-linearity can be suggested for the present range of dosage for E. coli and IE. C. perfringens does not show a correlation between the PFA/UV dose and the disinfection performance.

Urban stormwater runoff is a potential entrance pathway for a wide range of anthropogenic trace pollutants, like biocides, plasticizers, heavy metals or flame retardants, to urban lakes and rivers. However, little is known on dependencies of the occurrence of these trace pollutants on rain event characteristics and climate or seasonal influences. Furthermore, the importance of such dependencies for the calculation of loads and the uncertainties involved are unclear. This thesis evaluates possible correlations between trace pollutant concentrations in urban stormwater runoff and rain event characteristics together with further climate and seasonal influences, based on a large set of measurements from the project “Trace organics in Berlin stormwater runoff (OgRe)”. Here, samples were taken in a one-year monitoring program for five stormwater catchments representing specific urban structure types. Additionally, this thesis investigates whether the consideration of those correlations is necessary for the calculation of loads or whether the use of a mean concentration is sufficient. A method for the correlation analysis is developed and applied to the data, under the requirement to use just one influencing factor (predictor) per correlation and to keep the models simple. Regression models are fitted with regard to normal and log-normal error distributions. The models are evaluated regarding their goodness of fit using the Nash-Sutcliffe efficiency, the log-likelihood ratio, and the prediction coefficient of determination. For 45 out of 48 of the considered substances at least one correlation (i.e. in one of the five catchments) with rain or climate predictors is found. In addition, it is demonstrated that seasonal influences have an effect on substance concentrations for 25 out of 48 substances. Thus, the selected predictor values prove useful to explain the measured concentrations. Only 11 substances show the same correlation with a rain/climate predictor in four catchments and none in all five catchments. So, while concentrations for single events in one catchment can be well explained by the correlations, overall concentration patterns seem to be strongly influenced by the catchment, i.e. its urban structure type. Furthermore, it is shown that the assumption of a normally distributed error does not represent the data adequately in most cases. Consideration of a log-normal error distribution improves most regression models significantly. Regarding single substances, the correlation analysis helps to explain observed patterns. For instance, terbuthylazine, an agricultural pesticide, was only detected during typical application months of May and June, with the same observation in all five catchments. Accordingly, atmospheric deposition from the agricultural surroundings seems a reasonable explanation. In a second example, nicotine was found at very high concentrations in four catchments for low rain event durations, showing a strong decrease with increasing duration. This behavior can be explained by the fast elution of nicotine from cigarette butts within the first minutes of a rain event, followed by dilution during longer rain events. An exemplary load estimation based on a 30-year rain series for Berlin using a Monte Carlo simulation demonstrates that the use of regression models versus mean concentrations can lead to very different results. The reason lies in the selection of sampled rain events which are not distributed according their contribution to the total runoff volume (there should be more small to medium rain events, which contribute more to the total runoff volume). In conclusion, errors in the load estimation can result from i) using a mean concentration instead of a valid correlation, but also from ii) using a non-valid correlation. This underlines the importance of performing a correlation analysis before load calculations, but also the importance of a critical evaluation of the sample data and the correlations. For the latter, a combined evaluation along several goodness-of-fit metrics is suggested, together with plausibility checks of the correlation and of the considered range of values within which the regression model is applied.