The Markagunt Plateau, in southwestern Utah, lies at an altitude of about 9,500 feet, largely within Dixie National Forest. The plateau is capped primarily by Tertiary- and Quaternary-age volcanic rocks that overlie Paleocene- to Eocene-age limestone of the Claron Formation, which forms escarpments on the west and south sides of the plateau. In the southwestern part of the plateau, an extensive area of sinkholes has formed that resulted primarily from dissolution of the underlying limestone and subsequent subsidence and (or) collapse of the basalt, producing sinkholes as large as 1,000 feet across and 100 feet deep. Karst development in the Claron Formation likely has been enhanced by high infiltration rates through the basalt. Numerous large springs discharge from the volcanic rocks and underlying limestone on the Markagunt Plateau, including Mammoth Spring, one of the largest in Utah, with discharge that ranges from less than 5 to more than 300 cubic feet per second (ft³/s). In 2007, daily mean peak discharge of Mammoth Spring was bimodal, reaching 54 and 56 ft³/s, while daily mean peak discharge of the spring in 2008 and in 2009 was 199 ft³/s and 224 ft³/s, respectively. In both years, the rise from baseflow, about 6 ft³/s, to peak flow occurred over a 4- to 5-week period. Discharge from Mammoth Spring accounted for about 54 percent of the total peak streamflow in Mammoth Creek in 2007 and 2008, and about 46 percent in 2009, and accounted for most of the total streamflow during the remainder of the year. Results of major-ion analyses for water samples collected from Mammoth and other springs on the plateau during 2006 to 2009 indicated calcium-bicarbonate type water, which contained dissolved-solids concentrations that ranged from 91 to 229 milligrams per liter. Concentrations of major ions, trace elements, and nutrients did not exceed primary or secondary drinking-water standards; however, total and fecal coliform bacteria were present in water from Mammoth and other springs. Temperature and specific conductance of water from Mammoth and other springs showed substantial variance and generally were inversely related to changes in discharge during snowmelt runoff and rainfall events. Over the 3-year study period, daily mean temperature and specific conductance of water from Mammoth Spring ranged from 3.4 degrees Celsius (°C) and 112 microsiemens per centimeter (μS/cm) during peak flow from snowmelt runoff to 5.3°C and 203 μS/cm during baseflow conditions. Increases in specific conductance of the spring water prior to an increase in discharge in 2008–09 were likely the result of drainage of increasingly older water from storage. Variations in these parameters in water from two rise pools upstream from Mammoth Spring were the largest observed in relation to discharge and indicate a likely hydraulic connection to Mammoth Creek. Variations in water quality, discharge, and turbidity indicate a high potential for transport of contaminants from surface sources to Mammoth and other large springs in a matter of days. Results of dye-tracer tests indicated that recharge to Mammoth Spring largely originates from southwest of the spring and outside of the watershed for Mammoth Creek, particularly along the drainages of Midway and Long Valley Creeks, and in the Red Desert, Horse Pasture, and Hancock Peak areas, where karst development is greatest. A significant component of recharge to the spring takes place by both focused and diffuse infiltration through the basalt and into the underlying Claron limestone. Losing reaches along Mammoth Creek are also a source of rapid recharge to the spring. Maximum groundwater travel time to the spring during the snowmelt runoff period was about 7 days from sinking streams as far as 9 miles away and 1,900 feet higher, indicating a velocity of more than a mile per day. Response of the spring to rainfall events in the recharge area, however, indicated potential lag times of only about 1 to 2 days. Samples collected from Mammoth Spring during baseflow conditions and analyzed for tritium and sulfur-35 showed that groundwater in storage is relatively young, with apparent ages ranging from less than 1 year to possibly a few tens of years. Ratios of oxygen-18 and deuterium also showed that water from the spring represents a mixture of waters from different sources and altitudes. On the basis of evaluating results of dye-tracer tests and relations to adjacent basins, the recharge area for Mammoth Spring probably includes about 40 square miles within the Mammoth Creek watershed as well as at least 25 square miles outside and to the south of the watershed. Additional dye-tracer tests are needed to better define boundaries between the groundwater basins for Mammoth Spring and Duck Creek, Cascade, and Asay Springs.