The accumulation of reactive nitrogen (Nr) in terrestrial and aquatic environments is a global environmental issue that causes or contributes to climate change, stratospheric ozone depletion, and deterioration of coastal and terrestrial waters. Point source discharges of Nr from municipal and septic treatment systems, agricultural tile drainage, and industrial discharges contribute to these issues. Practical, low-cost methods are needed to reduce the Nr load into the environment from small-volume point source discharges. Denitrification beds are one such method. Improving the nitrate removal rate of denitrification beds will lead to reduced bed volumes, lower construction costs that likely facilitate greater uptake of the technology and reduced accumulation of Nr in the environment. The main objective of this thesis was to test a number of approaches that might increase the rate of nitrate removal rate in a denitrification bed under non-nitrate limiting conditions, including: manipulation of carbon source, temperature and hydraulic flow. To date, operational denitrification beds have used wood media as the carbon source which sustains nitrate removal rates of between 2–10 g N m-3 of media d-1 and relatively high permeability. While previous laboratory experiments have investigated the potential of alternative carbon sources, these studies were typically of short duration and small scale and did not necessarily provide reliable information for denitrification bed design purposes. To address this issue, nitrate removal, hydraulic and nutrient leaching characteristics of nine different carbon substrates were compared in 0.2 m3 barrels, at 14oC and 23.5oC over a 23 month period. The relationship between hydraulic efficiency and nitrate removal of the different media was also investigated. Findings from the barrel trial were field tested in pilot scale (2.9 m3) denitrification beds receiving municipal effluent dosed with KNO3, over a 15 month period. The pilot scale trial tested whether nitrate removal could be improved by using an alternative carbon media (maize cobs) and increasing bed temperature through passive solar heating. The influence of bed flow regime (horizontal-point, horizontal-diffuse, downflow and upflow) on hydraulic efficiency and nitrate removal was also investigated. This thesis demonstrated that more labile carbon sources, such as maize cobs, had significantly higher nitrate removal rates (15.0 to 21.8 g N m-3 d-1) than wood media (3.0 to 4.9 g N m-3 d-1) over the duration of the barrel trial. Nitrate removal rates increased with increasing temperature with mean Q10 of 1.6 for all media. The hydraulic efficiency of fragmented wood media decreased with increasing grain-size. However, nitrate removal rate was not dependent on hydraulic efficiency of the media, which was attributed to the significant secondary porosity of the media allowing denitrification to occur both on the surface and within the media particle. In the pilot scale trial, bed temperature increased by 3.4oC due to passive solar heating, but did not cause a measureable increase in nitrate removal rate due to variability in removal rates and possibly low temperature responsiveness of maize cobs for removing nitrate. Flow regime affected the hydraulic efficiency of denitrification beds and nitrate removal rates were lower in flow regimes with poor hydraulic efficiency. This was attributed to short-circuit flow reducing the bed volume that contributed to nitrate removal. The results indicate that a four-fold reduction in denitrification bed size could potentially be achieved by using maize cobs as the carbon substrate, as opposed to wood fragments, and increasing bed temperature by incorporating passive solar heating techniques. The findings of this thesis indicate that future research on improving the nitrate removal rate of denitrification beds under non-nitrate limiting conditions should focus on carbon substrates, increasing bed temperature, and hydraulic design of beds rather than on hydraulic efficiency of media. For example, research on coupling improved solar heating design with an appropriate inlet/outlet structure and location.