The Nature Conservancy (TNC) owned and managed 24,795 acres of mixed wetland, native prairie, farmland and woods east of Crookston, in northwestern Minnesota. The original wetlands and prairies that once occupied this land are being restored by TNC in cooperation with many partners and are becoming part of the Glacial Ridge National Wildlife Refuge. Results of this study indicate that these restorations are likely to have a substantial effect on the local hydrology. Water occurs within the study area on the land surface, in surficial aquifers, and in buried aquifers of various depths, the tops of which are 50 to several hundred feet below the land surface. Surficial aquifers are generally thin (about 20 feet), narrow (several hundred feet), and long (tens of miles). Estimates of the horizontal hydraulic conductivity of surficial aquifers were 2.7?300 feet per day. Buried aquifers underlie much of the study area, but interact with surficial aquifers only in isolated areas. In these areas, water flows directly from buried to surficial aquifers and forms a single aquifer as much as 78 feet thick. The surface?water channel network is modified by several manmade ditches that were installed to remove excess water seasonally and to drain wetlands. The channels of the network lie primarily parallel to the beach ridges but cut through them in places. Back?beach basin wetlands delay and reduce direct runoff to ditches. Recharge to the surficial aquifers (10.97?25.08 inches per year during 2003?5) is from vertical infiltration of rainfall and snowmelt (areal recharge); from surface waters (particularly ephemeral wetlands); and from upward leakage of water from buried aquifers through till confining units (estimated at about 1 inch per year). Areal recharge is highly variable in space and time. Water leaves (discharges from) the surficial aquifers as flow to surface waters (closed basins and ditches), evapotranspiration, and withdrawals from wells. Unmeasured losses (primarily discharge to ungaged (closed) basins) were 53?115 percent of areal recharge during 2003?5, while discharge to ditches that leave the study area was 17?41 percent. Discharge over 100 percent of areal recharge indicates a loss in ground?water storage. During the dry year of 2003, substantial ground water (about one?third of annual areal recharge) was released from aquifer storage but was replenished quickly during the subsequent normal year. Shallow ground?water flow is complex, with water in surficial aquifers, ditches, and wetlands part of a single hydrologic system. The ages determined for surficial ground?water samples were less than 15 years old, and one?third (8 of 24) were less than 5 years old, substantiating the close connection of surficial ground water to the land surface. During the study, 68?81 percent of water left the area through unmeasured surface?water losses (primarily evapotranspiration), which is 2? to 4?times that leaving through the ditch system. Base flow in ditches (ground?water discharge) was 30 to 71 percent of all ditch flow. Mean annual runoff in all gaged basins except SW3 (2.26 inches per year) was similar (3.69?4.12 inches per year). The quality of water samples from surficial aquifers and surface water collected in the study area was generally suitable for most uses but was variable. Most ground? and surface?water samples were dominated by calcium, magnesium, and bicarbonate ions. About one?quarter of surficial ground?water samples contained nitrate at concentrations greater than the U.S. Environmental Protection Agency?s (USEPA) Maximum Contaminant Level for human consumption. The median concentration of dissolved phosphorus ranged from 0.0108 milligrams per liter as phosphorus (mg/L?P) to 0.0293 mg/L?P. Nutrient concentrations in ditches were generally above the USEPA nutrient guidelines for reference streams in the area. Water samples contained detectable concentrations of atrazine, acetachlor, metolachlor, pendimethalin