Technical Abstract: Background/Questions/Methods Animal agriculture in the Spring Creek watershed of central Pennsylvania contributes sediment to the stream and ultimately to the Chesapeake Bay. Best Management Practices (BMPs) such as streambank buffers are intended to intercept sediment moving from heavy-use areas toward the stream. The placement of BMPs on a farm is generally based on untested assumptions about flow paths. Most often, a straight-line distance from the heavy-use area to the stream is assumed to be correct. Our objective was to compare the straight-line path to hydrologic flow paths calculated from fine-, medium- and coarse-grained Digital Elevation Models (DEMs; 1m, 10m, 30m) for 359 mapped heavy-use points within 100m of the stream. The 30m DEMs are the most widely available and require the least processing time. We anticipated that the flow path distance would be longer than the straight-line distance in all cases, that the finest resolution would lead to the most accurate measurement, but that the difference might not be great enough to justify the increased costs. Understanding the changes in path length and direction calculated using more complex methods and higher-resolution source data will enable us to make recommendations on methods to be used in developing conservation management plans. Results/Conclusions The medium-grain data (10m DEM) had the smallest differences between the hydrologic flow path and straight-line path: median differences in path length of 77m for the 1m DEM; 14m for the 10m DEM; 29m for the 30m DEM. Hydrologic flow paths were significantly longer than straight-line paths for all three scales; BMP placement based on straight-line distances may not be the most effective. Although the overall difference was significantly positive, calculations on the 30m DEMs sometimes produced straight-line paths that were longer than the hydrologic flow paths, apparently due to inaccuracies in the data. Where fine-scale DEMs are available, BMPs might be more effectively situated by considering the corresponding drainage pathways. The very different results produced at the three scales demonstrate that using the finest-grained elevation data may substantially improve placement of BMPs intended to mitigate for heavy animal use areas. The use of 30m DEMs for this purpose should be avoided. Fine-grained data such as 1m-resolution LiDAR-derived DEMs are available for Pennsylvania through PAMAP, and can be incorporated in the planning stages of BMP placement ultimately resulting in reducing agricultural sediment and nutrient loadings into local watersheds and the Chesapeake Bay.