Combined use of statistical Bayesian model and strontium isotopes deciphering the high complexity groundwater flow in the Guarani Aquifer System (GAS)

Abstract

The Guarani Aquifer System (GAS) is one of the most important groundwater reservoirs in Latin America. Even though its geological aspects and hydraulic behavior are deeply studied, questions regarding hydraulic connections between the GAS and the overlying (SGAS) and underlying (PRE-GAS) aquifers within the Paraná Sedimentary Basin geological framework are recently being pursued. This paper presents a new tracer approach within the GAS, incorporating strontium isotopes in a statistical Bayesian model. The multi-pronged approach is used to quantify mixing proportions and identify different processes related to water-rock interaction and hydrochemical data. Strontium isotopic ratio ranged from enriched values (>0.711) associated with higher GAS groundwater contribution, to lower values associated with contribution from the SGAS (∼0.707) and PRE-GAS (∼0.709) units, as well as isotopic fingerprint associated with direct rainwater recharge. As groundwater flows through the GAS, the mixing proportions between the end-members changes along with the hydrogeochemical evolution. Near recharge zones, GAS represents the dominant source (38% ± 10%), followed by PRE-GAS (28% ± 16%), SGAS (19% ± 11%) and rainwater (14% ± 10%). In transition waters between outcrop and confined areas, the GAS contribution remains dominant (48% ± 9%), followed by PRE-GAS (37% ± 14%) and SGAS (14% ± 8%). In the GAS confined zone, where contributions from rainwater and SGAS is absent, 87Sr/86Sr ratio is more associated with waters from PRE-GAS (53% ± 16%) than from GAS (46% ± 16%). This behavior indicates that groundwater flow in the GAS represent a more complex and vulnerable groundwater mixing system, closely related to the regional geological context of a closed intracratonic basin, than previously understood. The use of strontium isotopes as a hydrogeochemical tracer applied in a statistical mixing model, contributes to a more refined understanding of regional groundwater flow and water origin in the GAS.