The use of artificial ponds for the temporary storage of urban stormwater runoff is a commonly used environmental engineering practice in North America. By releasing runoff at a rate slower than the initial generation rate, on-line flood control is achieved. Urban runoff typically has a high eutrophication potential, so that single unit detention ponds may sustain excess algal/macrophyte growth within only a few years of construction. A research project was undertaken between 1992 and 1995 on stormwater detention ponds in the Province of Saskatchewan, Canada. Three ponds aged 15 to 17 years old are described in this thesis. The focus of the study was to describe the nutrient characteristics and associated phytoplankton cycles within these systems, and to further identify potential management options for water quality improvement. Four to five months of permanent winter ice cover occurs in the central zone of the Province. Of the three systems reported here, complete winter anoxia is typical in two (1.7 to 1.8 m deep - mixed), while partial or complete anoxia occurs in the third pond according to the timing of snowmelt (2.7 m deepthermally stratified in summer). Therefore, biological community structure is limited by the overwintering potential. Accumulated nitrogen, phosphorus, and silica sustained green algae and diatom blooms following ice-melt, and pH values of >9.5 often occurred by late April. During the open water season, hypereutrophic conditions were sustained and the systems typically oscillated between blue-green and green algal dominance according to flushing, external N- loading and mixing. Dissolved inorganic nitrogen:phosphorus (DIN:DIP) ratios of stormwater were typically below Redfield stoichiometry. Therefore, inpond DIN:DIP ratios of <3 were typical, with one system frequently <1. These low ratios were the result of both internal P-Ioading and N loss mechanisms during dry weather storage periods, and DIP >0.25 mg/L sometimes occurred within the euphotic zone. Dense N-fixing Anabaena sp. blooms periodically developed under suitable climatic conditions. Nitrogen fixed into the system in turn supported non-fixing species as a subsiding bloom was mineralized. Nutrient and phytoplankton cycles fluctuated within short time scales, according to physical disturbances and algal self-shading at peak biomass (sometimes > 100 mm3/L). Average algal biomass levels in the stratifying pond were lower than the other ponds on account of sedimentary losses to an anaerobic bottom zone. xv Grazing by herbivorous zooplankton was generally not significant in promoting phytoplankton species successions. The zooplankton of all three systems were dominated by cyclopoid copepods and rotifers. Predation by fish (minnows), food quality (blue-green algae dominance), and losses during large flushing events are among factors which may suppress desirable large-bodied cladoceran zooplankton in these systems. The mean surficial (1 cm) sediment iron content was lowest (",,27 mg Fe/g dry wt VS. 38 mg Fe/g dry wt) and the organic content was highest ("" 18% LOI VS. 11 % LOI) in the stratifying pond compared to the two non-stratifying ponds. Also, the mean sediment depth in the stratifying pond (not including the littoral slope sediments) was highest (",,21 cm VS. 9 to 13 cm), when compared to the two non-stratifying ponds. Iron may be lost from the stratifying pond by flushing of anaerobic hypolimnetic waters during stormflows. Groundwater inflows to the stratified pond resulted in a higher alkalinity system and consequently the highest mean sediment calcium content of the three ponds (",,82 mg Calg dry wt, VS. 31 to 34 mg Calg dry wt). The surficial sediment total-P [TP] of the three ponds ranged from 0.97 to 1.26 mg/g dry wt, and in all cases 11 to 15% of this was associated with inorganic extractable P, with more in the calcium than the iron/aluminium bound fraction. At peak water temperatures (s:26°C) internal P loading rates >30 mg/m2/day were calculated from field data in the non-stratifying ponds. An average P release rate of 15 mg/m2/day was measured during anaerobic incubation of sediment cores from one of these ponds at 20°C. However, aerobic incubation of these sediments showed that Fe concentrations were sufficient to provide high P uptake potential when oxidized. Anaerobic incubations of intact cores from the stratifying pond gave average release rates of 5 to 16 mg/m2/day from 5 to 20°C. Field data also showed that net internal P loads were reflected by changes in the surficial sediment P pool. A nutrient input budget for the stratifying pond showed that groundwater baseflow supplied a massive amount of DIN (as N03) relative to stormflows. If the seasonal stormwater DIP load was expressed as an averaged areal mass/day, the average seasonal internal P loading (-8 to 10 mg/m2/day) was four times higher. Theoretical P removal efficiencies of 63 to 80% were calculated for the ponds, but resuspension and flushing of internally loaded P accumulated during dry weather may reduce these values. A TP mass export of 0.25 to 0.3 kg/ha impervious/0.58 year from 250 mm precipitation was calculated from runoff studies in Saskatoon. In experimental work, inorganic nitrogen additions to the most N-limited pond were carried out from May to July 1994. Complete dominance of the spring to mid-summer phytoplankton by green algae and diatoms was maintained. However, warming water increased P recycling, and during a period of lower than average wind speeds a non-fixing blue-green algae bloom developed in place of the usual N-fixing algae bloom. No significant alteration to the zooplankton species composition was evident despite structural changes to the vernal phytoplankton composition. Phosphorus inactivation with aluminium sulphate was successful in improving water quality for a six week period during which the control pond developed a dense N-fixing algae bloom. Sediment surface oxidation was promoted by the reduction of productivity, and P adsorption to sediment iron complexes was an important secondary benefit. Several very large storms were ultimately responsible for exchanging approximately 100% of the storage volume, after which bloom conditions were restored. The procedure may be an effective short-term measure, but benefits will not extend beyond major exchange events. Management options for aesthetic improvement are very limited in these hypereutrophic ponds. External DIP loads will continue to be at least 5 to 10 times greater than threshold values for nuisance algal growth, and seasonal internal loading of P is high. The inability of increased N availability to prevent blue-green algae bloom formation, together with high exchange volumes and a general lack of herbivorous zooplankton, suggest that top-down management interventions (limited by overwintering) to control zooplanktivores are unlikely to prevent algae bloom formation in ponds with lower volume:catchment area ratios. More work is required with regard to nutrient budgets if pond operational efficiencies are to be accurately assessed. In addition, measurement of primary productivity would provide invaluable information for any attempt to model algal growth in these ponds. Sediment removal is ultimately required as a long-term maintenance measure, but more information on the incorporation of P inactivation agents directly into the sediment structure is needed as a means to retard internal P loading.