Author:

Keywords:

Index of connectivity, Sediment trap effectiveness, RMMF model, Land abandonment, Soil erosion, Spanish Pyrenees, Science & Technology, Physical Sciences, Life Sciences & Biomedicine, Geosciences, Multidisciplinary, Soil Science, Water Resources, Geology, Agriculture, HYDROLOGICAL CONNECTIVITY, AGRICULTURAL TERRACES, SEMIARID CATCHMENT, SCALE, SPAIN, LANDSCAPE, BASIN, ABANDONMENT, SYSTEMS, YIELD, 0403 Geology, 0406 Physical Geography and Environmental Geoscience, 0503 Soil Sciences, Geochemistry & Geophysics, 3707 Hydrology, 3709 Physical geography and environmental geoscience, 4106 Soil sciences

Abstract:

Overland flow connectivity and runoff and sediment trap effectiveness are currently the cutting edge topics in soil erosion research. The effect of agricultural terraces, irrigation channels and trails on runoff and soil erosionmodelling at catchment scale is still a remaining research question. In this study we run the index of connectivity of Borselli et al. (2008) and a modified version of the revised Morgan, Morgan and Finney (RMMF) model to predict the hydrological connectivity and the rates of soil erosion under four different scenarios of land uses and land abandonment. This goal is achieved by using geographic information systems (GIS) in the Estanque de Arriba catchment (74 ha; Spanish Pre-Pyrenees) where 83 soil samples were collected. The different maps of hydrological connectivity were used to create runoff and sediment trap effectiveness masks that were included in the assessment of the effective cumulative runoff. The results showed that the index of connectivity was very high in the irrigation channels and walls of the agricultural terraces for the past and current scenarios. The runoff and sediment connectivity of the catchment from the hillsides to the lake decreased with an increasing vegetation cover and a decreasing number of linear landscape elements. The connectivity decreased from the past to the current scenario and from the current to the future scenario with vegetation recovery in the abandoned fields. The vegetation factor appeared to be more important than the disappearance of the agricultural terraces and channels to explain the changes in the connectivity at catchment scale. Random changes in the IC values appeared between the current and the future scenario without vegetation recovery in some parts of the catchment and were related to the mathematical procedure of the model that combines the upslope and downslope components at each pixel. This information is of special relevance in areas disturbed by humans and those with complex topography. The average erosion rates ranged between 1.5 and 3.7 Mg ha−1 year−1 for the different land use scenarios decreasing with the increase of the vegetation recovery in the abandoned fields. A significant percentage of the study area will suffer more erosion in the early stage of future land abandonment with an increasing volume of overland flow in the fields at the bottom of the catchment whereas soil erosion will decrease significantly after long-term land abandonment with vegetation recovery. Further research is proposed to better account the changes that happen in the soil and vegetation parameters at the early stages of land abandonment. The results of this study suggest that current active systems could be stabilized by trapping and retaining eroded sediment by increasing the vegetation cover and preserving the agricultural terraces.