Keith E. Schilling

Targeting land-use change for nitrate-nitrogen load reductions in an agricultural watershed

The research was conducted as part of the USDAs Conservation Effects Assessment Project. The objective of the project was to evaluate the environmental effects of land-use changes, with a focus on understanding how the spatial distribution throughout a watershed influences their effectiveness. The Soil and Water Assessment Tool (SWAT) water quality model was applied to the Squaw Creek watershed, which covers 4,730 ha (11,683 ac) of prime agriculture land in southern Iowa. The model was calibrated (2000 to 2004) and validated (1996 to 1999) for overall watershed hydrology and for streamflow and nitrate loadings at the watershed outlet on an annual and monthly basis. Four scenarios for land-use change were evaluated including one scenario consistent with recent land-use changes and three scenarios focused on land-use change on highly erodible land areas, upper basin areas, and floodplain areas. Results for the Squaw Creek watershed suggested that nitrate losses were sensitive to land-use change. If land-use patterns were restored to 1990 conditions, nitrate loads may be reduced 7% to 47% in the watershed and subbasins, whereas converting row crops to grass in highly erodible land, upper basin, and floodplain areas would reduce nitrate loads by 47%, 16%, and 8%, respectively. These SWAT model simulations can provide guidance on how to begin targeting land-use change for nitrate load reductions in agricultural watersheds.

Conservation practice establishment in two northeast Iowa watersheds: strategies, water quality implications, and lessons learned.

Coldwater trout streams are important natural resources in northeast Iowa. Extensive efforts have been made by state and federal agencies to protect and improve water quality in northeast Iowa streams that include Sny Magill Creek and Bloody Run Creek, which are located in Clayton County. A series of three water quality projects were implemented in Sny Magill Creek watershed during 1988 to 1999, which were supported by multiple agencies and focused on best management practice (BMP) adoption. Water quality monitoring was performed during 1992 to 2001 to assess the impact of these installed BMPs in the Sny Magill Creek watershed using a paired watershed approach, where the Bloody Run Creek watershed served as the control. Conservation practice adoption still occurred in the Bloody Run Creek watershed during the 10-year monitoring project and accelerated after the project ended, when a multiagency supported water quality project was implemented during 2002 to 2007. Statistical analysis of the paired watershed results using a pre/post model indicated that discharge increased 8% in Sny Magill Creek watershed relative to the Bloody Run Creek watershed, turbidity declined 41%, total suspended sediment declined 7%, and NOx-N (nitrate-nitrogen plus nitrite-nitrogen) increased 15%. Similar results were obtained with a gradual change statistical model. The weak sediment reductions and increased NOx-N levels were both unexpected and indicate that dynamics between adopted BMPs and stream systems need to be better understood. Fish surveys indicate that conditions for supporting trout fisheries have improved in both streams. Important lessons to be taken from the overall study include (1) committed project coordinators, agency collaborators, and landowners/producers are all needed for successful water quality projects; (2) smaller watershed areas should be used in paired studies; (3) reductions in stream discharge may be required in these systems in order for significant sediment load decreases to occur; (4) long-term monitoring on the order of decades can be required to detect meaningful changes in water quality in response to BMP implementation; and (5) all consequences of specific BMPs need to be considered when considering strategies for watershed protection.

Variability of nitrate-nitrogen load estimation results will make quantifying load reduction strategies difficult in Iowa

Many states within the Upper Mississippi River Basin are developing strategies to reduce nutrient loads to rivers. Reliable load estimation methods are needed to track progress toward nutrient reduction goals. We evaluated the variability of commonly used interpolation and extrapolation models to estimate nitrate-nitrogen (NO3-N) loads in 11 Iowa rivers. Results showed that the overall consistency between models of annual mean daily loads was low. Differences among the methods were particularly pronounced in May when the greatest NO3-N loads normally occur. The disparity in N load estimation among different methods is troubling given that states, federal agencies, or interest groups must have confidence in NO3-N load estimation procedures if the public is to believe that the load reductions strategies are working.