Soil carbonates are widely used in paleoclimate studies because their stable isotopes record a valuable, ground-level perspective of environmental conditions including moisture availability, temperatures, concentrations of CO2, and overlying vegetation. Despite their importance as an archive and decades of study, we do not understand how soil carbonates form. This uncertainty clouds our ability to use soil carbonates to quantitatively constrain temperature and water fluxes in ancient soils. In this talk, I will discuss ongoing research conducted at the University of Michigan that seeks to fill in this knowledge gap by studying modern soil carbonates and soil water in the Western US. We are using two innovative stable isotope techniques, clumped and triple oxygen isotopes, to constrain the environmental factors that drive carbonate formation in dryland soils.