The application of 137Cs to post-fire erosion in north-west Spain

Abstract Wildfires are a frequent occurrence in the Cantabrian mountain range in north-west Spain, causing erosion and soil degradation. With the aim of measuring this erosion, 43 soil cores were taken from four slopes with different degrees of fire frequency and 137 Cs concentration. The four slopes are located in the same geographical area. They have a quartzitic substrate, with slopes of between 20° and 30° and share the same type of soil, elevation and orientation. However, vegetation cover is different as a result of the action of fire. Maps of the area, showing the burnt areas over the period 1984 to 2004 (remote sensing data), confirm the effects that fire has had. Soil cores were taken from transects on each slope and three reference samples were also taken from a high plateau in order to establish a reference value for 137 Cs inventory. Soil redistribution rates (erosion and deposit) for the slopes could thus be calculated, according to whether 137 Cs inventory was higher or lower than the reference value. The values obtained show that 137 Cs inventory is lower for the slopes with shrubby vegetation than for the reference samples and its distribution indicates higher erosion rates for the upper slope samples. Mean erosion rates obtained for these slopes were 6.75 t ha − 1 year − 1 and 6.6 t ha − 1 year − 1 . These data are compared with erosion rates for 137 Cs obtained in areas with different climates and patterns of land use. 137 Cs redistribution was not found on slopes with tree cover, suggesting that surface erosion has not occurred, or at least not during the time period for which 137 Cs can be measured. 137 Cs distribution was compared with the distribution of other variables and topographical indices (distance from the hilltop, accumulated flow, slope, curvature and CTI).The application of 137 Cs in studies of post-fire erosion is relatively novelty and it a useful technique in post-fire erosion studies over long timescales and for study sites larger than point and plot scales.