Nutrients and sediment in surface runoff from frozen agricultural fields were monitored within three small (16.0 ha [39.5 ac] or less), adjacent basins at a no-till farm in southwest Wisconsin during four winters from 2003 to 2004 through 2006 to 2007. Runoff depths and flow-weighted constituent concentrations were compared to determine the impacts of surface-applied liquid-dairy or solid-beef manure to frozen and/or snow-covered ground. Despite varying the manure type and the rate and timing of applications, runoff depths were not significantly different among basins within each winter period. Sediment losses were low (generally less than 22 kg ha⁻¹ [20 lb ac⁻¹] in any year) and any statistical differences in sediment concentrations among basins were not related to the presence or absence of manure or the amount of runoff. Concentrations and losses of total nitrogen and total phosphorus were significantly increased in basins that had either manure type applied less than one week preceding runoff. These increases occurred despite relatively low application rates. Lower concentrations and losses were measured in basins that had manure applied in fall and early winter and an extended period of time (months) had elapsed before the first runoff event. The highest mean, flow-weighted concentrations of total nitrogen (31.8 mg L⁻¹) and total phosphorus (10.9 mg L⁻¹) occurred in winter 2003 to 2004, when liquid-dairy manure was applied less than one week before runoff. On average, dissolved phosphorus accounted for over 80% of all phosphorus measured in runoff during frozen-ground periods. The data collected as part of this study add to the limited information on the quantity and quality of frozen-ground runoff at field edges, and the results highlight the importance of manure management decisions during frozen-ground periods to minimize nutrients lost in surface runoff.