Woodchip bioreactors: biological nitrogen removal for cleaning agricultural runoff
26 July 2016
Covering the woodchips with a geo-textile fabric before laying the soil cover at a bioreactor installation (courtesy of the Iowa Soybean Association Environmental Programs and Services)
Biological nitrogen removal that pairs nitrification and denitrification is a widely practised treatment process for municipal wastewater. The process of denitrification — that is, the natural microbially-driven reduction of nitrate to nitrogen gas — is typically fuelled by soluble carbon sources like methanol, which are readily available for denitrifying communities living in these wastewater-treatment reactors.
While treatment of nitrate at wastewater plants provides significant water-quality benefits, many areas around the world are now struggling to address nitrate loadings from non-point or diffuse sources. Agriculture is a major non-point source contributor to many nitrogen related, water-quality challenges globally.
In the United States, the hypoxic zone that forms seasonally in the Gulf of Mexico is fed by both nitrogen and phosphorus entering the Gulf via the Mississippi River. The Mississippi River basin is one of the largest watersheds in the world and covers one of the most agriculturally productive regions of the US.
Agriculture in US Midwest is heavily underpinned by artificial subsurface drainage networks, also known as tile drains (Figure 1). Drainage improvement projects in this part of the country date to the mid-1800s, when European settlers first began to farm the very rich prairie soils.
Today, more than 38 million acres (15 million hectares) of the American Midwest are tile drained to improve trafficability of otherwise wet fields and to improve crop growth and yield. However, tile drainage also serves as conduit for soluble nutrients, primarily nitrate, to move directly from the rich soil to streams. This nitrate transported by midwestern tile drainage is known to be a major contributing factor to the Gulf of Mexico hypoxic zone.
Unlike runoff over the surface of fields, tiles drainage pipes provide an ideal centralised location for targeted edge-of-field biological nitrogen removal before the water moves downstream. However, the use of a soluble carbon source, such as methanol, for tile drainage would be highly impractical due to both cost and real-time dosing rate determination since tile drainage flows are highly variable.
Denitrifying woodchip bioreactors
The idea of an enhanced-denitrification reactor using a low-cost and practical solid organic carbon source, such as woodchips, began to take shape approximately two decades ago. Today, denitrifying woodchip bioreactors consist of a long trench filled with woodchips, through which nitrate-laden agricultural effluents and drainage waters are routed (Figure 2).
Generally, flow manifolds at the bioreactor inlet and outlet distribute and collect the flow, respectively, and the flow is gravity-driven so the system remains relatively inexpensive. Unlike wastewater treatment reactors, these systems have never required denitrifier inoculation. Moreover, start-up is rapid with nitrate removal observed almost immediately.
The major start-up challenge is that the woodchips elute a flush of nutrients and chemical oxygen demand in initial tea-coloured effluent which typically lasts a few days to weeks. Bioreactor total annual nitrate load reductions generally range from 15-90% and volumetric removal rates are generally on the order of 5-10 g N removed per cubic meter of bioreactor per day.
Denitrifying woodchip bioreactors have been trialled around the world for a variety of nitrate-laden waters including small-scale treatment of municipal wastewater, removal of nitrate in drainage from mines using ammonium nitrate explosives, and as a low-cost nitrate mitigation option for land-based aquaculture facilities. The most common bioreactor application is the treatment of nitrate in agricultural drainage water.
Bioreactors designed for this purpose (Figure 1) have an estimated design life of approximately ten years and cost around $10,000 USD (€9,000) to treat roughly 30-80 drained acres (12-32 hectares). In this agricultural field setting, bioreactors provide focused nitrate removal without requiring land to be removed from production and with very little annual maintenance. Both benefits are of significant interest to farmers.
Woodchip bioreactor nitrate removal is strongly influenced by both water temperature and hydraulic retention time (or, the length of time the water stays inside the bioreactor which depends on the water flow rate and the size of the bioreactor). Overcoming the disconnect between when tile drainage bioreactors work best (i.e. late summer with lower drainage flow rates and slightly higher groundwater temperatures) and when the majority of the drainage nitrate load occurs (i.e. spring) would be an important advancement in bioreactor design.
A bioreactor’s by-pass flow line (Figure 1) is an essential design component to prevent reduction of the drainage system’s flow capacity, but this means that a portion of the total annual drainage flow from the field by-passes the bioreactor untreated.
Current woodchip bioreactor design models
Current design models attempt to balance the desire to treat a significant proportion of the annual drainage volume (i.e. minimise by-pass flow) with maintenance of a sufficient minimum bioreactor retention time (i.e. the minimum time required to produce anoxic conditions required for denitrification). Most fields would require an impractically large bioreactor to treat all of the annual drainage volume, with such a bioreactor vastly overdesigned for the low flow-rates predominant much of the year.
In addition to hydraulics and reactor design, woodchip bioreactor fill media presents new questions for biological and ecological engineers (Figure 3). There has been no discernible difference in nitrate removal capability between hardwood versus softwood chips, although faster-growing tree species that contain relatively greater amounts of nitrogen (willow, poplar) are not recommended due to their leaching potential.
Particle size and surface area are known to be important in suspended growth processes used in wastewater treatment, but the impact of woodchip particle size is thought to be minimal as water permeates the chips and creates an active microbiological layer several millimetres into the wood. While woodchips are generally the most practical bioreactor fill, agricultural residues such as maize stover or wheat straw provide much higher nitrate removal rates although they degrade much faster.
The new agricultural water quality-improvement efforts across states in the US Midwest call for tens of thousands of woodchip bioreactors to be installed to achieve nutrient loss reduction targets. For this region and for many other nitrogen-impaired regions around the world, there is a key research opportunity to further develop denitrifying woodchip bioreactor technology to provide a practical and low-cost water quality solution.
The 2nd International Conference on Pollution Control and Resource Recovery for the Livestock Sector takes place in NUI Galway next month, providing a forum to present, discuss and develop innovative technologies and practices for managing livestock waste and recovering resources. Dr Laura Christianson is set to speak at the event on the topic: ‘Scientifically advanced woody media for improved water quality from livestock woodchip heavy-use areas.’ She will also present a keynote speech on ‘Innovation in Biogas’.
Dr Laura Christianson, Assistant Professor of Water Quality, Department of Crop Sciences, University of Illinois at Urbana-Champaign. Email: firstname.lastname@example.org. See http://draindrop.cropsci.illinois.edu/.