Effects of spatial orientation of prairie vegetation in an agricultural landscape on curve number values

Abstract:

The simplicity of the Curve Number (CN) method, which summarizes an area’s hydrologic soil group, land cover, treatment, and hydrologic condition into a single number, make it a consistently popular choice for modelers. When multiple land cover types are present, a weighted average of the CNs is used. However, the weighted CN does not account for the spatial distribution of different land cover types within the watershed. To overcome this limitation, it becomes necessary to discretize the model into homogenous subunits, perhaps even to the hillslope scale, leading to a more complex model application. The objective of this study is to empirically derive CN values that reflect the effects of placements of native prairie vegetation (NPV) within agricultural landscapes. We derived CN values using precipitation and runoff data from (May 1 – Sept 30 over a 7 year period (2008 – 2014) for 9 ephemeral watersheds in Iowa (USA) ranging from 0.47 to 3.19 ha. The watersheds were planted with varying extents of NPV (0%, 10%, 20%) in different watershed positions (footslope vs. contour strips), with the rest of the watershed as row crop. The derived CN values from watersheds with all row crop were consistent with published values and watersheds with NPV had an average CN reduction of 6.4%, with a maximum reduction of 11.6%. Four of the six sites with treatment had a lower CN than one calculated using a weighted average of look-up values, indicating that accounting for placement of vegetation within the landscape is important for modeling runoff with the CN method. The derived CNs were verified using the leave-one-year-out method (computing CN using data from 6 of the 7 years, and then estimating runoff on the seventh year with that CN). Nash-Sutcliffe Efficiency (NSE) values for the estimated runoff typically ranged from 0.4-0.6. Our results suggest that the new CNs could confidently be used in future modeling studies to explore the hydrologic impacts of the NPV treatments at increasingly larger watershed scales.