AbstractUsing data collected from collocated hillslopes in central Iowa, the United States, the authors (1) explored the spatial variability of runoff coefficient at the event scale by examining the relationships between the standard deviation and coefficient of variation of runoff coefficient and the mean and (2) analyzed the temporal persistence of spatial pattern of runoff coefficient using Spearman rank and Pearson correlation coefficient. This study considered 12 cropland hillslopes with 0–20% native prairie vegetation coverage distributed at different hillslope locations. Seventy runoff events over the period 2008–2011 were investigated, of which 51 occurred during crop active growing season, when the hydrologic responses of crops and prairie vegetation are similar. For these events, the spatial coefficient of variation had a median value of 0.80, which indicate high variation of event‐scale runoff coefficients across neighboring hillslopes. This spatial variation largely cannot be consistently explained by the individual hillslope structural properties investigated. The standard deviation and mean of runoff coefficient showed a convex upward relationship across the range of runoff coefficients, with the maximum standard deviation value at the mean runoff coefficient of about 0.48. The coefficient of variation exponentially decreased with increasing runoff coefficient. For 71% of the cases, the results of both correlation analyses were statistically significant (p ≤ 0.05), which indicate stable spatial pattern of runoff coefficient across events. This temporal persistence could be disrupted under extremely dry and wet conditions. The spatial variation‐mean empirical relation and the temporal persistence of spatial pattern provide insight for parameterizing spatial variability of runoff coefficient in distributed hydrologic models.