CONICET | Buscador de Institutos y Recursos Humanos

Mountain forests and their soils provide ecological services such as maintenance of biodiversity, provision of clean water, carbon capture and forage for livestock rearing, which is one of the principal economic activities in mountain areas. However, surprisingly little is known about livestock impact in South American mountain forest soils. With the aim of understanding how livestock and topography influence patterns of forest cover, soil compaction, soil loss and soil chemical properties, we analyzed these parameters in 100 Polylepis australis woodland plots situated in the humid subtropical mountains of Central Argentina. We used distance from the nearest ranch as an objective index of historical livestock impact and measured standard topographic variables. We also measured indicators of soil erosion and soil properties in 3 topographically similar valleys: (1) With traditional livestock management; (2) a 10 year livestock exclusion and re-forestation with Polylepis trees; (3) a reference valley with Polylepis forests. Forest recovery potential as influences by soil degradation was measured by evaluating soil erosion and seedling presence in 285 1 m2 plots. Our main results reveal that distance from ranch in all cases partly explains tree canopy cover, soil loss, soil compaction and soil chemical properties; suggesting a strong negative effect of livestock. Intermediate altitudes had more tree canopy cover, while landscape roughness a measure of the variability in slope inclination and aspect - was negatively associated to soil impedance and acidity, and positively associated to soil organic matter content. Finally, flatter areas were more acid. The comparison of the valleys with and without livestock showed reduced soil erosion and superficial soil compaction (but not bellow 5 cm) without livestock, however the reference valley had no erosion at all and soil compaction was lowest by far, indicating soil restoration is a slow process. Plots with Polylepis seedlings were found in sites with less erosion than plots without seedlings, even after controlling for distance to seeder tree and other relevant variables. We conclude that livestock has had a substantial influence on forest soil degradation in the Mountains of Central Argentina, that soil restoration is a slow process and that forest recovery capacity is reduced by degraded soils. We suggest a similar situation may be found in many other similar South American mountains. Soil degradation should be incorporated into decision making when considering long-term forest sustainability, or when taking into account retaining livestock for biodiversity conservation reasons. Where soil loss and degradation are ongoing, we recommend drastic reductions in livestock density.