Effect of bioenergy crops and fast growing trees on hydrology and water quality in the Little Vermilion River Watershed

Energy security and sustainability require a suite of biomass crops, including woody species. Short rotation woody crops (SRWCs) such as Populus have great potential as biofuel feedstocks. Quantifying biomass yields of bioenergy crop and hydrologic and water quality responses to growth is important should it be widely planted in the Midwestern U.S. Subsurface tile drainage systems enable the Midwest area to become highly productive agricultural lands, but also create environmental problems like nitrate-N contamination of the water it drains. The Soil and Water Assessment Tool (SWAT) has been used to model watersheds with tile drainage, but the new tile drainage routine in SWAT2012 has not been fully tested. The objectives of this study were to develop algorithms and growth parameters of Populus in Agricultural Land Management Alternative with Numerical Assessment Criteria (ALMANAC) and SWAT models, compare performance of tile drainage routines in SWAT2009 and SWAT2012 in simulating tile drainage, and simulate biomass yields of bioenergy crops and the impacts of their impacts on water quantity and quality for a typical tile-drained watershed in the Midwest USA. The functional components and parameters of hybrid poplar Tristis #1 (Populus balsamifera L. × P.tristis Fisch) and eastern cottonwood (Populus deltoides Bartr.) were determined, and related algorithms improved in ALMANAC and SWAT based on improved simulation of leaf area, plant biomass and biomass partitioning. Long-term (1991-2003) field site and river station data from the Little Vermilion River (LVR) watershed in Illinois were used to evaluate performance of tile drainage routines in SWAT2009 revision 528 (the old routine) and SWAT2012 revision 615 and 645 (the new routine). Calibrated monthly tile flow, surface flow, nitrate in tile and surface flow, sediment and annual corn and soybean yield results at field sites, and flow, sediment load and nitrate load at the river station for the old and new tile drainage routines were compared with observed values. Crop residue from corn stover, perennial grasses, switchgrass and Miscanthus, and hybrid poplar trees were considered as potential bioenergy crops for the LVR watershed. SWAT2012 (Revision 615) with the new tile drainage routine (DRAINMOD routine) and improved perennial grass and tree growth simulation was used to model long-term annual biomass yields, flow, tile flow, sediment load, total nitrogen, nitrate load in flow, nitrate in tile flow, soluble nitrogen, organic nitrogen, total phosphorus, mineral phosphorus and organic phosphorus under various bioenergy scenarios in the LVR watershed. Simulated results from different bioenergy crop scenarios were compared with those from the baseline. Tree growth calibration and validation results showed that improved algorithms of leaf area index (LAI) and biomass simulation and suggested values and potential parameter range for hybrid poplar Tristis #1 and Eastern cottonwood ( Populus deltoides Bartr.) were reasonable, and performance of the modified ALMANAC in simulating LAI, aboveground biomass and root biomass of Populus was good. Performance of the modified SWAT simulated hybrid poplar LAI and aboveground woody biomass (PBIAS: -57 ~ 7%, NSE: 0.94 ~ 0.99, and R2: 0.74 ~ 0.99), and cottonwood aboveground biomass, seasonal mean runoff, mean sediment, mean nitrate-N and total nitrate-N were satisfactory (PBIAS: -39 ~ 11%, NSE: 0.86 ~ 0.99, and R2: 0.93 ~ 0.99). Additionally, tile drainage calibration and validation results indicated that the new routine provides acceptable simulated tile flow (NSE = 0.50 ~ 0.68), and nitrate in tile flow (NSE = 0.50 ~ 0.77) for field sites, while the old routine simulated tile flow (NSE = -0.77~ -0.20) and nitrate in tile flow (NSE = -0.99 ~ 0.21) for the field site with constant tile spacing were unacceptable. The new modified curve number calculation method in revision 645 (NSE = 0.56 ~ 0.82) better simulated surface runoff than revision 615 (NSE = -5.95 ~ 0.5). Bioenergy crop simulation results showed that 38% corn stover removal (66,439 Mg/yr) with combination of Miscanthus both on highly erodible areas and marginal land (19,039 Mg/yr) provided the highest biofeedstock production. Flow, tile flow, erosion and nutrient losses were slightly reduced under bioenergy crop scenarios of Miscanthus, switchgrass, and hybrid poplar on highly erodible areas, marginal land and marginal land with forest. The increase in sediment load and nutrient losses resulting from corn stover removal could be offset under scenarios with various combinations of bioenergy crops. Corn stover removal with bioenergy crops both on highly erodible areas and marginal land could provide more biofuel production relative to the baseline, and was beneficial to hydrology and water quality at the watershed scale. The modified ALMANAC and SWAT can be used for biofeedstock production modeling for Populus. The modified SWAT model can be used for Populus biofeedstock production modeling and hydrologic and water quality response to its growth. The improved algorithms of LAI and biomass simulation for tree growth should also be useful for other process based models, such as SWAT, EPIC and APEX. Tile drainage calibration and validation results provided reasonable parameter sets for the old and new tile drainage routines to accurately simulate hydrologic processes in mildly-sloped watersheds. Bioenergy crop simulation results provided guidance for further research on evaluation of bioenergy crop scenarios in a typical extensively tile-drained watershed in the Midwestern US