Les lits de séchage de boues plantés de roseaux (LSPR) existent en France depuis le début des années 1990. Ceux-ci ont principalement été développés pour le traitement de boues issues de station boues activées en aération prolongée. Cependant, ce procédé est trop souvent associé à des problèmes de dysfonctionnements liés à des dimensionnements et des gestions mal maîtrisés. En effet, les installations souvent composées de 4 lits souffrent de performances insuffisantes avec une siccité qui avoisine seulement 15 % contre 30, voire 35 % au Danemark où la technique est largement répandue. Une meilleure connaissance des divers mécanismes en jeux contribuerait sans conteste à accroître les potentialités de cette filière. En outre, avec la mise en place croissante de SPANC, l'augmentation future du flux de matières de vidange à traiter nous a également amené à transposer ce procédé vers ce type de produit. Pour cela le suivi de 16 pilotes de 2 m² en phase de démarrage et d'une installation de taille réelle de type « boues activées en aération prolongée » (13 000 EH, 8 lits de 470m² en fonctionnement depuis 4 ans) a permis d'étudier l'influence de différentes qualités de boues (boues activées épaissies ou non, matières de vidange), de la charge organique, et de la fréquence d'alimentation sur les performances de déshydratation et de minéralisation. Après un an et demi de phase de démarrage, ce travail de thèse se focalise sur le fonctionnement au nominale du système. Les expérimentations menées sur les pilotes ont permis d'identifier les limites du système, et ainsi émettre des préconisations de dimensionnement et de gestion pour chacun des deux types de boue étudiés. Parallèlement, des expériences menées en laboratoire ont permis de d'étudier plus finement les phénomènes en cause. Notamment en répondant aux questions relatives à l'influence de la qualité de la boue. De plus, une étude mécaniste a également été menée pour identifier l'influence du drainage et de l'évapotranspiration des roseaux sur la déshydratation des boues. Enfin, les processus impliqués dans la stabilisation des boues ont également été étudiés. / Sludge drying reed beds exist in France since the early 1990s and have been mainly developed for activated sludge treatment. Nevertheless, these systems often operate badly due to unadapted design and management strategy. The usual design for these facilities consists of 4 beds each being alternately fed during a week. Performances are low with a dryness around only 15% against 30 or 35% in Denmark where this technology is widespread. A better knowledge of the mechanisms will indisputably increase the potentiality of this process. Besides, taking into account the obligation of regulatory control of on-site wastewater treatment, communities are now facing larger volumes of septage. Therefore, one of the main idea of this study is to also use drying reed beds for septage treatment. Thus, 16 pilots of 2m² each and a full–scale plant (13 000 p.e., 8 beds of 470m² in operation for 4 years) have been monitored to study the influence of the sludge quality (activated sludge thickened or not, septage and a mixture of both), the sludge loading rate, and the load frequency on the dewatering and mineralization efficiencies. After one and a half year of commissioning period, this thesis focuses on the nominal operating conditions. Experiments carried out on pilots allow the limits of the system to be identifying. Hence, design and management recommendations for both types of sludge studied can be established. Meanwhile, laboratory experiments have aimed at evaluating more precisely the phenomena involved in the system. We first consider the influence of the sludge quality on its treatment ability on sludge drying reed beds. In addition, a mechanistic study was also conducted to identify the influence of drainage and reeds evapotranspiration on the sludge dewatering. Finally, the processes involved in the sludge stabilization were also investigated.

Reed canarygrass (Phalaris arundinacea) is a grass species that can dominate wet meadow plant communities. This study investigated if grazing by cattle on restored wet meadows suppresses reed canarygrass, thereby promoting the restored plant community. This study was conducted at two locations in northwest Minnesota. Management practices used were a patch-burn grazing treatment and a four-pasture high intensity-short duration grazing rotation. A pretreatment survey was conducted before grazing followed by annual surveys every five years after grazing. Both treatments reduced reed canarygrass canopy cover by 49 percent compared to non-grazed control sites. Grazed patches were moving towards a Carex dominated community. The community not invaded with reed canarygrass had similar native species richness at the end of the experiment in the rotational grazing treatment, and improved plant richness in the patch-burn grazing treatment. This study demonstrates grazing reduces cover of reed canarygrass, while maintaining or increasing native plant species richness.

The diagnostic technique, polarization and depolarization current (PDC) is useful for insulation testing. It requires applying a DC step voltage to the test sample and measuring the current. To measure fast PDC phenomena a fast step is needed. One way of applying a fast high voltage step is to use power electronic switches. Series connection can be used to increase the voltage limit, but this result in unequal voltage sharing unless equipped with voltage balancing. In this work a high voltage vacuum reed switch is investigated as a simple and low-cost alternative to power electronic switches, handling up to 10 kV with a single device. The switch turn on and off behavior was studied. It was found that the initial turn-on is good, in the range of nanoseconds, but there is a problem with the vacuum recovering its insulating properties at low currents before the contacts fully close. The required output voltage level is therefore obtained only after a further settling time that increases with increased input voltage and is much longer than the initial breakdown, e.g. 20 µs for the case of 4.5 kV input voltage. Other limitations of the fast high voltage PDC were also studied. The output voltage was measured across the test sample without adding an intentional resistor in the circuit. There were large oscillations for 1 µs but these oscillations are damped due to inherent resistance of the connecting leads, series resistance of the capacitors and resistance of the reed switch. A comparison is made between the measured and the simulated results using MATLAB to see the effect of parasitic inductance. A damping resistor was added in the circuit and the output results were again compared. With the addition of the damping resistor, the number of oscillations were reduced and their time scale was limited to 0.1 µs . An analysis is made at the end which describes the limitation occurring in determining the high frequency component of PDC. The current during the step is many orders of magnitude higher than the polarization current even at 1 µs , so measurement of the current and protection of the apparatus is not trivial.

The invasion of riparian habitats by the exotic plant, Arundo donax L., commonly known as the giant reed is one of the greatest threats to ecosystems in Central and Southern California. There have been several efforts to eradicate this plant with varying control methods. This study evaluated some previously known techniques of controlling the giant reed in addition to some novel methodologies to determine the most effective approach. The study site is a section of the Lower Calaveras River that transects University of the Pacific's campus and is a prime example of the devastation caused by the giant reed. The project site was cut and cleared over a period of eight months and divided into fourteen plots. Seven different techniques including a control were selected and each treatment was randomly applied to two plots. The methods chosen were: "cut, resprout and spray", "cut-stem, spray", "chip and compost", "chip and tarp", "compost" and "compost and tarp". The response variables were the "mean regrowth height", "mean circumference of stalks" and the "number of resprouted stalks". A repeated measures ANOVA was used to analyze the response variables. The "compost and tarp" method had the lowest number of resprouted stalks and was also significantly different from the control treatment. None of the other techniques were significantly different from the control but the "compost" method showed some promise. Based on this study, the "compost and tarp" technique is the most effective treatment. The data collection for this study was limited and future experiments should be conducted on a longer time scale to assess the effectiveness of these methods. Further research should also be conducted to evaluate the effectiveness of various biotic factors on the growth of A. donax, which could enhance the efficacy of methodologies currently used to control this introduced invasive plant.

The purpose of this study was to determine how methods of support functioned in the senate debate on the seating of Reed Smoot. In order to clarify the directions of the study, answers to the following questions were sought:1. How extensively were methods of support used by each side?2. What was the frequency of supports per assertion by each side?3. How was the use of support materials distributed among the speakers?4. How many different methods of support were used by each side?5. What were the most frequently used methods of support by each side?6. What were the least frequently used methods of support by each side?7. Did the methods of support which were used differ from the methods discussed in public speaking textbooks?8. Did the winning side differ in the use of methods of support from the losing side?

Nutrient leakage into streams and rivers can lead to eutrophication, which negatively impact aquatic ecosystems. A commonly used tool to mitigate nutrient leakage is the implementation of wetlands into the polluted system. This study focuses on the levels of N and P in the river Fyris.n in Uppsala, Sweden, and the possible effects from restoration of a wetland near the river’s outlet. The results show a potential reduction of 41% and 45% for N and P respectively. The results point to a need for more wetlands further up in the catchment system to mitigate the nutrient load in the subcatchments, as the high nutrient levels there would not be affected by the proposed wetland. There are also several uncertainties due to the lack of reference material, and the results proves the need for further studies on this subject.

No description available.

No description available.

No description available.

No description available.