Sites with five periods of consecutive measurements from the Kaikoura Peninsula data were used to assess the influence of erosional history and surface topography upon subsequent erosion rates. Analysis of multilevel models suggests that initial erosion rates and initial surface topography have a significant, but decreasing influence upon subsequent erosion rates. Surface topography and erosion rates immediately prior to each measurement period have a consistent positive and negative relationship respectively to erosion rates in the subsequent measurement periods. These data suggest that there is a short-term memory influencing erosion rates at any specific measurement period. Use of traversing microerosion meter data may, however, not be the most appropriate means of analysing the causes of the variance of erosion at the intra-site scale.