Stiff-stemmed grass hedges are a vegetative conservation technique that has been introduced to reduce water runoff and soil erosion from land under row crop management. Planting stiff-stemmed grass hedges in a watershed may reduce water runoff and soil erosion, in part by altering soil macroporosity. The objective of this study was to evaluate differences in Computed Tomography (CT)-measured macroporosity parameters as affected by a 12-year-old perennial grass hedge system relative to row crop management, and to compare CT-measured macroporosity results with values estimated from water retention data. Soil at the study site was a Monona silt loam (fine-silty, mixed, superactive, mesic Typic Hapludolls). The site was under a no-till corn (Zea mays L.)-soybean (Glycine max L.) rotation with switchgrass (Panicum vergatum L.) hedges 1 m wide spaced at 15 m intervals. Three positions were sampled: grass hedge position, deposition zone position 0.5 m upslope from grass hedges, and row crop position which was approximately 7 m upslope from the hedges. Intact core samples (76 mm x 76 mm) were collected from two depths, 0 to 100 mm and 100 to 200 mm, with five replicates per position per depth. Five scans (0.5 mm thick) were taken per soil core with scans separated by 10 mm. Simple thresholding was used to isolate pores in each image using Image-J software. CT-measured macropore parameters including number of pores, soil macroporosity (> 1000 μm diam.), soil mesoporosity (200-1000 μm diam.), and circularity were examined as a function of depth as well as macropore fractal dimension. Number of pores (macro- and meso-), averaged across depths, in the grass hedge were nearly 2.5 times greater than those in the row crop and 5 times greater than in the deposition positions. However, circularity for the grass hedge treatment was 8.8% lower than in the row crop treatment and 2.6% lower than in the deposition position treatment. These differences imply that pore perimeters were greater and more irregular for the grass hedge treatment compared to the row crop treatment where the macropores were more circular. CT-measured macroporosity was significantly greater (P < 0.01) for the grass hedge position (0.056 m³ m^-3) as compared to the row crop (0.014 m³ m^-3) and deposition positions (0.006 m³ m^-3). The macropore fractal dimension (D) was significantly greater (P < 0.01) for the grass hedge position (D = 1.56) than in the row crop (D = 1.31) and the deposition (D = 1.12) positions. The values of all measured pore characteristics decreased with depth. CT-measured macroporosity data were comparable with macroporosity estimated from water retention data. Differences between CT-measured macroporosity and water retention-estimated values were not statistically significant (P > 0.05) for all treatments and depths except for the row crop treatment in the second depth. Results indicate that the CT method with image analysis can obtain similar macroporosity results to data from water retention techniques with bulk core samples. CT-measured macroporosity values were positively correlated (r = 0.95) with core-measured saturated hydraulic conductivity suggesting the possibility that this method may provide a useful index related to soil hydraulic properties. These findings show that perennial grass hedges created more pores and greater area of macroporosity, which will have a significant impact on infiltration and runoff for soils under this management system. The CT-scanning technique combined with image analysis appeared to be a useful method for quantifying macropore parameters. CT-scanning can evaluate the spatial variations in these parameters both across and within core samples which is impossible with more traditional techniques such as the water retention method. This technique can be applied to soil pore characterization studies for different management systems such as riparian zones and soil quality restoration sites as well as geotechnical investigations.