Climate change introduces significant uncertainty when assessing the risk of soil salinity in water-scarce regions. We combine a soil–water-salinity–sodicity model (SOTE) and a weather generator model (AWE-GEN) to develop a framework for studying salinity and sodicity dynamics under changing climate definitions. Using California’s San Joaquin Valley as a case study, we perform first-order sensitivity analyses for the effect of changing evapotranspiration (ET) rates, length of the rain season, and magnitude of extreme rainfall events. Higher aridity, through increased ET, shorter rainy seasons, or decreased magnitude of extreme rainfall events, drives higher salinity — with rising ET leading to the highest salinity levels. Increased ET leads to lower levels of soil hydraulic conductivity, while the opposite effect is observed when the rainfall season length is shortened and extreme rainfall events become less intense. Higher ET leads to greater unpredictability in the soil response, with the overall risk of high salinity and soil degradation increasing with ET. While the exact nature of future climate changes remains unknown, the results show a serious increase in salinity hazard for climate changes within the expected range of possibilities. The presented results are relevant for many other salt-affected regions, especially those characterized by intermittent wet–dry seasons. While the San Joaquin Valley is in a comparatively strong position to adapt to heightened salinity, other regions may struggle to maintain high food production levels under hotter and drier conditions.