With progressing urbanisation, treatment of urban stormwater is a vital issue that should be addressed to ensure good water quality in receiving water bodies. Treatment may be performed near the source, with different filter systems using various filter materials, or by using an end-of-pipe method, e.g. a stormwater pond. One constraint in the urban environment is the lack of available space in developed areas, where stormwater treatment facilities are needed the most. Methods developed to treat the stormwater runoff have been the focus of previous studies but the increasing standards of water quality and increasing land constraint pressures demand the further development of stormwater treatment systems. Both laboratory and field experiments are necessary to understand and improve the treatment processes as well as to evaluate how the implemented methods perform under field conditions. The aim of the thesis was to increase the knowledge about the components in stormwater treatment systems that can be used in area-efficient treatment facilities. In order to compare four potential stormwater filter materials (peat, bark, air-blown polypropylene and milkweed), column experiments were carried out using synthetic stormwater that simulated road runoff. Experiments were carried out to evaluate the impacts of the ageing of synthetic stormwater quality during laboratory testing, including dissolved metal concentrations and their impact on the estimation of filter efficiency. In a field study, a full-scale application of a zeolite filter installation was investigated, with a focus on service life and the efficiency of treating copper roof runoff. In order to further investigate a novel sedimentation device, a bottom grid structure (BGS), promoting sediment settling in a smaller area of a stormwater pond, a hydraulic modelling study was conducted to investigate the impact of the cell geometry of the structure on sediment settling and the impact of the structure on pond maintenance and sediment resuspension.  The column tests of four different filter materials showed that bark and peat had higher treatment efficiency for dissolved metals than milkweed and polypropylene, with the order of efficiency being peat>bark>milkweed>polypropylene. All four of the filter materials showed a total metal reduction of over 70%, which could be due to the separation of particle-bound metals in the columns. The ageing of the synthetic stormwater showed that dissolved metals, particularly copper, decreased in concentration, quite rapidly. During one experiment run, the dissolved copper concentration was reduced to 15% of its initial value. In order to account for the concentration changes an equation was proposed that normalised the concentration of dissolved metal over the duration of the experiment. During the observation period of 16 months, the zeolite installation removed 52% to 82% and 48% to 94% of total and dissolved copper, respectively. However, the effluent concentrations were still high (360-600 μg/l). There was also an indication of the decreasing filter performance over time with a prediction that the treatment level of total copper would drop to approximately 25% by the end of the service life of three years. The hydraulic experiments on a scaled model of a BGS showed that wider cells were on average 13% more efficient in trapping the particles than the narrower variant. The cell wall angle also had an impact (tilted walls added to the sedimentation efficiency), although the applicability of such cell structures can be questioned, as this cell shape may hinder maintenance efforts. It was also hypothesised that the inclusion of the BGS in the pond reduces the area needed for sediment settling, thus making the pond more area-efficient and easier to include in an urbanised setting.