The impact of deforestation on hydrology and soil salinity : a case study over the South American Dry Chaco

The Dry Chaco is a semi-arid ecoregion in South America, covering parts of Argentina, Paraguay, and Bolivia. To date, the region hosts one of the largest dry forests in the world, but expansion of dryland agriculture and cattle ranching led to gradual conversion of native vegetation to anthropogenic land cover. The potential impact of these newly established agricultural lands on the surrounding environment is of great concern. Local studies have shown that deforestation leads to changes in the soil-water balance. In the longer term, this can lead to a gradual rise of the groundwater table, mobilizing water-soluble salts to the surface, and affecting plant growth and crop productivity. Therefore, the main objective of this study was to gain insight into spatio-temporal patterns of soil hydrology and salinity of the Dry Chaco using land surface models (LSMs), remote sensing data, and in situ measurements of soil salinity. The central hypotheses were that (i) the impact of deforestation on soil salinity varies with existing climatic and topographic gradients, and (ii) the susceptibility to severe soil salinization following deforestation is spatially-dependent. First, the impact of deforestation on the different water budget components over the entire Dry Chaco was analyzed using three state-of-the-art LSMs, grouped within the Land Information System (LIS) of NASA. Because most LSMs use climatological vegetation data and static land cover parameters, they lack the ability to correctly represent the ongoing land cover changes in the Dry Chaco. Therefore, land cover changes were implemented using time-varying satellite-derived vegetation indices, together with annually updated land cover information. Our results showed that there are large regional and long-term differences in the simulated water budget components between the three LSMs, mainly due to differences in LSM structures and model-specific parameters. The different LSMs also redistribute water differently in response to the updated vegetation and land cover treatment. Second, the link between hydrology and soil salinity was studied based on in situ salinity data and environmental variables. In situ salinity data were collected along topographic and climatic gradients in the dry season of 2019. The LSM output at a regional scale, satellite-derived vegetation indices, and topographic data were related to soil salinity data to produce a first physically-based regional estimate of salinity patterns in the central Argentinean Dry Chaco. In the driest parts of the study area, salts are retained in the soil profile as salt leaching is prevented. Primary soil salinization occurs in topographic depressions where groundwater is present near the surface and high evapotranspiration rates enhance the formation of evaporites. With greater precipitation, salts are increasingly leached to deeper layers resulting in low natural soil salinity values. Also, in regions with high hydraulic gradients, salts are effectively removed by subsurface flow and regional groundwater transport. In the last chapter of this dissertation, we took a closer look at the chemical composition of water-soluble salts in soils of the central Argentinean Dry Chaco, and the link between the chemistry of salt-affected soils and deforestation. Previous studies mainly measured the electrical conductivity or chloride content to quantify soil salinity, and little was known about the chemistry of water-soluble salts that contribute to salt stress. We filled this knowledge gap by analyzing the full spectrum of water-soluble salts based on paired soil samples collected in forested and deforested areas. A geochemical analysis revealed that chlorides contribute only 25% to total salinity and that sulphates, calcium, and sodium are equally important. We further showed that the effects of deforestation on soil salinity are spatially dependent and closely related to the geomorphological setting.