Urban construction sites often have excessive erosion and compaction after topoil removal, scraping, and grading. The purpose of this study is to determine if compaction remediation efforts are effective on a simulated urban site. The sod and topsoil were removed, the area was graded to a 1% slope, and the subsoil was compacted by trafficking. The untreated topsoil was re-applied to the control area about 5 cm deep, and compost:topsoil mixture (2:1) was applied to the improved area to 15 cm depth. A lawn mixture was planted to the control side (C-3 grasses). A mixture of buffalo grass and blue gamma grass was planted on the treatment side. Frequently, a Hydra probe was used to manually determine soil water content (θ) at ten spots for each treatment. A rainfall simulator was used to evaluated infiltration, runoff, and sediment loss. After three years, blocks of soils were sampled to examine roots and soil structure. The improved area held more water than the control area. Seasonal surface water contents showed greater soil water use in the spring and fall for the control area (C-3 grasses), and the improved area showed greater water use in the mid summer (C-4 grasses). Runoff was greater for the improved area than the control area, but sediment loss was reduced. In both control and improved areas, roots penetrated into the dense subsoil around clods or structural units, but the improved area showed more roots penetrating into the subsoil.