The δ³⁴S and δ¹⁸O values for dissolved sulfate in groundwater are commonly used in aquifer studies to identify sulfate reservoirs and describe biogeochemical processes. The utility of these data, however, often is compromised by mixing of sulfate sources within reservoirs and isotope fractionation during sulfur redox cycling. Our study shows that, after all potential sulfate sources are identified and isotopically characterized, the δ³⁴S_SO4 and δ¹⁸O_SO4 values differentiate processes such as sulfate-source mixing, sulfide oxidation, barite dissolution, and organosulfur decomposition. During bacterial reduction of sulfate, the values reflect kinetic sulfur isotope fractionation and exchange of oxygen isotopes between sulfate and water. Detailed analysis of the chemistry (Cl and SO₄ concentrations) and isotopic composition (δ²H_H2Oand δ¹⁸O_H2O) of groundwater in an alluvial aquifer in Central Oklahoma, USA allowed the identification of five distinct end members that supply water to the aquifer (regional groundwater flowing into the study area, river water, leachate from a closed landfill that operated within the site, rain, and surface runoff). The δ³⁴S_SO4 and δ¹⁸O_SO4 values in each end member differentiated three sources of sulfate: sulfate dissolved from Early to Late Permian rocks within the drainage basin (δ³⁴S_SO4 = 8–12‰ and δ¹⁸O_SO4 = 10‰), iron sulfides oxidized by molecular oxygen during low water-table levels (δ³⁴S_SO4 = − 16‰ and δ¹⁸O_SO4 = 10‰), and organosulfur compounds (predominately ester sulfates) from decomposition of vegetation on the surface and from landfill trash buried in the alluvium (δ³⁴S_SO4 = 8‰ and δ¹⁸O_SO4 = 6‰). During bacterial reduction of these sulfate sources, similar isotope fractionation processes are recorded in the parallel trends of increasing δ³⁴S_SO4 and δ¹⁸O_SO4 values. When extensive reduction occurs, the kinetic sulfur isotope fractionation (estimated by ε_H2S–SO4 = − 23‰) results in the steady increase of δ³⁴S_SO4values to greater than 70‰. Equilibrium isotope fractionation during exchange of sulfate oxygen and water oxygen, a process not commonly observed in field-based studies, is documented in δ¹⁸O_SO4 values asymptotically approaching 21‰, the value predicted for conditions at the study site (ε_SO4–H2O = 27‰). These results show that recognition of all potential sulfate sources is a critical first step to resolving complexities in δ³⁴S_SO4 and δ¹⁸O_SO4 data. The approach taken in this study can be used in other aquifer systems where the identification of multiple sulfate sources and sulfur redox cycling is important to understanding natural processes and anthropogenic influences.