Influence of time on measurement results of erosion plot studies

Previous studies of erosion plot data of six consecutive years from the Is Olias experimental site, Sardinia, Italy, suggest that soil erosion decreases with time. Data were collected in 18 plots set up on three hillslopes under different land use and vegetation (shrubs, burned macchia, and Eucalyptus sp. stand). The objective of this study was to evaluate the decreasing trend with time in the soil erosion rate from the 18 erosion plots and to check variable influence of erosion factors over time. Statistical analyses of the data showed that rainfall and rainfall erosivity values were not significantly different during the investigation period, while soil erosion significantly decreased over time. ANOVAs to examine the possible variable influence of four soil erosion factors (vegetation, position on slope, surface stoniness, and slope) over time highlighted a significantly growing influence of the surface stoniness and slope factors on the erosion rates. The behaviour of these two factors supported the assumption that the cause of decreasing erosion was a reduced sediment availability, since there was no indication of a reduction of transport capacity during the investigation period. Results suggested that erosion rates from plot experiments without agricultural practices contain a high level of temporal uncertainty which affects the use of erosion rates for conservation measures and model calibration. A comprehensive examination of the plot system processes is recommended for erosion plot studies in non-agricultural areas. Also, measurement periods longer than 3 years might be needed.