Estudio de los impactos sobre la zona no saturada derivados del uso de agua desalada procedente de acuíferos salobres continentales

Résumé

The use of desalted water in agriculture is increasing due to the lack of conventional water resources and to improvement in desalination process efficiency. Desalination is not limited only to sea water, but inland brackish aquifer water has become more usual in recent years. The main objective of this thesis is to evaluate the long‐term impacts deriving from the use of the aforementioned water on the vadose zone and its hydraulic properties. An experimental plot was set up at the University of Alicante campus, and grass was cropped and irrigated with desalted water from the nearby inland brackish aquifer of San Vicente del Raspeig. Crop management followed normal procedures in the study area. Monitoring soil volumetric water content and soil pressure head for 20 months was carried out by adopting two different and alternative monitoring strategies: manual and automatic. The experiment was completed with a tracer test from BrLi and was controlled for 160 days. Finally, a complete soil hydraulic laboratory characterisation was done with samples extracted from the plot. Data acquisition was performed manually and automatically, and a flow model was done with the HYDRUS 1D software, calibrated and validated over different periods for the two measured data sets (automatic and manual). The tracer test (conservative transport), also done with HYDRUS 1D, validated the flow model and allowed inferring transport parameters (dispersivity) on the field scale. Finally, a predictive model of reactive transport, including geochemical processes and associated hydraulic parameter changes, was proposed for lengthy periods (30 years). The reactive model was run with HP1, which resulted from coupling the HYDRUS 1D and PHREEQC codes. Besides the predictive model being considered a baseline under current conditions, three different scenarios were proposed according to several criteria. The two flow models from the manual and automatic data acquisition, properly captured the flow dynamics in the unsaturated zone, and it was concluded that both strategies are adequate to reproduce the general trend flow. However, the automatic strategy proved more suitable to detect soil hydraulic properties changes because data acquisition was more accurate and frequent. To obtain transport parameters, the combination of Br determination from X‐ ray in the tracer test and the parameterisation for the inverse method proved a fast and reliable methodology. The predictive models highlight the continuous dissolution process for soil gypsum, as well as the dissolution (in the root zone) and precipitation (below it) of calcite. The simulation results indicate changes in porosity which imply changes in hydraulic conductivity that reached a factor higher than 2. Finally, the study of the different proposed scenarios shows that: lack of irrigation reduces changes in soil; lack of gypsum in the soil profile implies reduced hydraulic conductivity below the root‐zone; and reduction in CO2 partial pressure in the root‐zone dramatically decreases the dissolution of carbonates species.