Artificial subsurface agricultural drainage helps increase agricultural productivity by improving the timeliness of field operations and the workability of the soil in many locations around the world. However, the detrimental water quality impacts precipitated by such drainage systems are a concern in many of these locations. For example, in the US Midwest region, artificial agricultural drainage has been done for over 100 years, but over the past several decades, nitrogen loadings from drainage in this area have been causally implicated in one of the United States' largest water quality concerns: the hypoxic zone in the Gulf of Mexico. Additionally, local water quality impairment stemming from agricultural drainage is beginning to be targeted by regulatory agencies around the world. Denitrification bioreactors for agricultural drainage are one of the newest technologies being investigated for practical edge-of-field nitrate-nitrogen (NO3-N) reduction. Promising early results from these systems have led to increased attention over the past few years from many groups including governmental agencies, researchers, and agricultural organizations. The overall goal of this dissertation was to investigate the design, performance, and cost-efficiency of denitrification bioreactors for agricultural drainage. This Ph. D. work is the culmination of 3.5 yr of work both in Iowa, USA and in Palmerston North, New Zealand. In short, pilot-scale investigations of bioreactor performance (Chapters 2 and 3) influenced the design and field-scale evaluation of several bioreactors in Iowa (Chapter 4), and as interest in denitrification bioreactors grew over this Ph. D. period, a comprehensive cost comparison of bioreactors and other water quality improvement options became necessary (Chapter 5). Major conclusions related to the importance of retention time and water temperature in bioreactor nitrate removal performance. Additionally, the hydraulics of water moving through a denitrification bioreactor was shown to be important as increased flow rates during drainage hydrographs caused decreased retention times and reduced nitrate removal. In terms of cost, this "apples to apples" comparison showed that bioreactors were comparable to other water quality practices at $1.44/kg N removed. While this dissertation concluded denitrification bioreactors were a viable option for reducing the amount of nitrate in agricultural drainage, more research on their design and performance is needed.