During 1998-2002, the U.S. Geological Survey, in cooperation with the South Dakota Department of Transportation (SDDOT), conducted a study to evaluate factors affecting ice forces at selected bridges in South Dakota. The focus of this ice-force evaluation was on maximum ice thickness and ice-crushing strength, which are the most important variables in the SDDOT bridge-design equations for ice forces in South Dakota. Six sites, the James River at Huron, the James River near Scotland, the White River near Oacoma/Presho, the Grand River at Little Eagle, the Oahe Reservoir near Mobridge, and the Lake Francis Case at the Platte-Winner Bridge, were selected for collection of ice-thickness and ice-crushing-strength data. Ice thickness was measured at the six sites from February 1999 until April 2001. This period is representative of the climate extremes of record in South Dakota because it included both one of the warmest and one of the coldest winters on record. The 2000 and 2001 winters were the 8th warmest and 11th coldest winters, respectively, on record at Sioux Falls, South Dakota, which was used to represent the climate at all bridges in South Dakota. Ice thickness measured at the James River sites at Huron and Scotland during 1999-2001 ranged from 0.7 to 2.3 feet and 0 to 1.7 feet, respectively, and ice thickness measured at the White River near Oacoma/Presho site during 2000-01 ranged from 0.1 to 1.5 feet. At the Grand River at Little Eagle site, ice thickness was measured at 1.2 feet in 1999, ranged from 0.5 to 1.2 feet in 2000, and ranged from 0.2 to 1.4 feet in 2001. Ice thickness measured at the Oahe Reservoir near Mobridge site ranged from 1.7 to 1.8 feet in 1999, 0.9 to 1.2 feet in 2000, and 0 to 2.2 feet in 2001. At the Lake Francis Case at the Platte-Winner Bridge site, ice thickness ranged from 1.2 to 1.8 feet in 2001. Historical ice-thickness data measured by the U.S. Geological Survey (USGS) at eight selected streamflow-gaging stations in South Dakota were compiled for 1970-97. The gaging stations included the Grand River at Little Eagle, the White River near Oacoma, the James River near Scotland, the James River near Yankton, the Vermillion River near Wakonda, the Vermillion River near Vermillion, the Big Sioux River near Brookings, and the Big Sioux River near Dell Rapids. Three ice-thickness-estimation equations that potentially could be used for bridge design in South Dakota were selected and included the Accumulative Freezing Degree Day (AFDD), Incremental Accumulative Freezing Degree Day (IAFDD), and Simplified Energy Budget (SEB) equations. These three equations were evaluated by comparing study-collected and historical ice-thickness measurements to equation-estimated ice thicknesses. Input data required by the equations either were collected or compiled for the study or were obtained from the National Weather Service (NWS). An analysis of the data indicated that the AFDD equation best estimated ice thickness in South Dakota using available data sources with an average variation about the measured value of about 0.4 foot. Maximum potential ice thickness was estimated using the AFDD equation at 19 NWS stations located throughout South Dakota. The 1979 winter (the coldest winter on record at Sioux Falls) was the winter used to estimate the maximum potential ice thickness. The estimated maximum potential ice thicknesses generally are largest in northeastern South Dakota at about 3 feet and are smallest in southwestern and south-central South Dakota at about 2 feet. From 1999 to 2001, ice-crushing strength was measured at the same six sites where ice thickness was measured. Ice-crushing-strength measurements were done both in the middle of the winter and near spring breakup. The maximum ice-crushing strengths were measured in the mid- to late winter before the spring thaw. Measured ice-crushing strengths were much smaller near spring breakup. Ice-crushing strength measured at the six sites