Disentangling the effect of past and present agroforestry practices in modifying landscapes of Mediterranean mountains

Soil erosion and sediment delivery to rivers are important drivers for land degradation and environmental change in mountain agroecosystems.  These factors are especially severe in areas affected by intermittent heavy rainfalls after dry periods, and human practices such as deforestation or clearcutting practices. Many Mediterranean mountain environments underwent conversion of rangelands into croplands during the previous centuries increasing the risk of erosion. After land abandonment the process was gradually reversed during the middle of the 20^thcentury, allowing the recovery of natural land cover and reduced soil erosion rates. To further control the high erosion rates, several afforestation programs introduced pine forests at the headwater of most Mediterranean mountain catchments transforming the landscape by terracing and reducing the sediment connectivity. However, nowadays, forests’ different management could lead to high erosion rates and subsequent landscape modifications. To understand the possible effect of these practices together with the current agricultural management, we have combined the strength of empirical data and spatially distributed modelling. Surface soil samples from different land uses were collected in a representative catchment at the foot of Santo Domingo range at the border of the central Ebro River valley. The study catchment was mostly cultivated at the beginning of the 19_th century but changed to rangeland and pine afforestation in the last 50 years. The remaining croplands are predominated by rainfed agriculture that leaves soils mostly unprotected from June to January when erosive storms occur. The main land uses are croplands, pine afforestation, scrubland and Mediterranean forest.

In this study, we propose an ensemble technique composed of ¹³⁷Cs derived soil redistribution rates as specific point values and as a calibration tool for the widely used WaTEM/SEDEM sediment delivery model. Thus, by the use of ground truth and modelled data we aim to i) apply the WaTEM/SEDEM model at the catchment scale and calibrate it with ¹³⁷Cs derived soil redistribution rates for finding an optimal set of input parameters; ii) examine the effect of clearcutting and agricultural practices on the total erosion; iii) compare the modelled results with recently calculated sediment apportionments by using the sediment fingerprinting technique.

Our findings highlight the use of spatially distributed models combined with ¹³⁷Cs derived rates as a powerful tool to understand the driving factors of soil erosion in the last decades and to delineate the hotspot areas that could suffer high erosion if subjected to certain management practices.