A 40 ka high-resolution soil carbonate record from southern Utah: proxy development, paleohydrology, and paleoecology

Terrestrial paleoclimate records are critically important for testing hypotheses of climate dynamics and verifying climate simulations. However, unlike their oceanic counterparts, terrestrial records are short, more commonly discontinuous, and require specific geographic conditions not necessarily ideal for proposed questions (e.g., speleothem records must come from wherever a cave occurs). We instead utilize laminated soil carbonate rinds as a high resolution (100s yr) paleoclimate archive. Soil carbonate rinds can represent 10s-100s kyr, are apparently continuous over relevant timescales, maintain stratigraphic order, and are common in arid and semi-arid regions. We demonstrate the utility of this methodology at Torrey, Utah which is at the northern edge of the North American Monsoon (NAM). Sample rinds form on the bottom of large (≈1 m) boulders. The rinds are 0.5-2 cm thick and have visually and microscopically identifiable stratigraphy. Radiocarbon dates are in order and suggest a nearly constant growth rate from ≈40 ka to 3 ka, when the record ends. However, the pendants have significant pore space, so secondary carbonate has the potential to bias sample ages to be too young by 100s-1000s years. Precise sampling may be able to mitigate this bias. In spite of the potential secondary carbonate bias, δ13C and δ18O analyses show trends consistent with regional records. Secondary carbonate is therefore a concern, but apparently does not overpower the primary isotopic signals. A calibration study identified soil carbonate formation during the peak growing season (i.e., JAS, during the NAM), so we interpret our records as summer signals modulated by soil `memory' effects. The δ18O record has low variability (±0.5 ‰) but mimics regional NAM records, suggesting similar climate drivers as far north as Utah. The δ13C record shows some correlation with δ18O, which is reasonable given that the strength of the NAM can drive ecologic responses. However, the data are a more reasonable match for insolation, implicating it as a more important factor in regional ecology. For example, high insolation correlates with a C4 dominated landscape, even during the height of the Last Glacial Maximum. These data highlight the potential for soil carbonate rinds to provide long, continuous paleorecords at specific locations.