Variations in speleothem calcium isotope ratios (δ44Ca) are thought to be uniquely controlled by prior carbonate precipitation (PCP) above a drip site and, when calibrated with modern data, show promise as a semi-quantitative proxy for paleorainfall. However, few monitoring studies have focused on δ44Ca in modern cave systems. We present a multi-year comparative study of δ44Ca, carbon isotopes (δ13C), and trace elemental ratios from cave drip waters, modern calcite, and host rocks from two cave systems in California - White Moon Cave (WMC) and Lake Shasta Caverns (LSC). Drip water and calcite δ44Ca from both caves indicate PCP-driven enrichment, and we use a simple Rayleigh fractionation model to quantify PCP variability over the monitoring period. Modern calcite δ44Ca from deeper sites at WMC display a larger PCP signal than shallower sites, indicating that longer flow paths allow for more PCP under the same hydroclimate conditions. At both WMC and LSC, we observe an inverse relationship between PCP and rainfall amounts, though this relationship is variable across individual drip sites. Our modeled data suggest that WMC experiences ~20% more PCP than LSC, consistent with the fact that WMC receives less annual rainfall. This work supports speleothem δ44Ca as an independent constraint on PCP that can aid in the interpretation of other hydrologically sensitive proxies and provide quantitative estimates of paleorainfall. Additional, long-term monitoring studies from a variety of climate settings will be key for understanding δ44Ca variability in cave systems more fully and better constraining the relationship between PCP and rainfall.

During the mid-Holocene (MH: ~6,000 years BP) and Last Interglacial LIG (LIG: ~129,000-116,000 years BP) differences in the seasonal and latitudinal distribution of insolation drove northern hemisphere high-latitude warming comparable to that projected in end-21st century low emissions scenarios, making these intervals potential analogs for future climate change in North America. However, terrestrial precipitation during past warm intervals is not well understood and PMIP4 models produce variable regional moisture patterns in North America during both intervals. To investigate the extent to which the latest generation of models reproduces moisture patterns indicated by proxy records, we compare hydroclimate output from 17 PMIP4 models with networks of moisture-sensitive proxies compiled for North America during the LIG (39 sites) and MH (257 sites). Agreement is lower for the MH, with models producing wet anomalies across the western United States (US) where a high concentration of proxies indicate aridity. The models that agree most closely with the LIG proxies differ from the PMIP4 ensemble by showing relative wetness in the eastern US and dryness in the northwest and central US. An assessment of atmospheric dynamics using an ensemble subset of the three models with the highest agreement suggests that LIG precipitation patterns are driven by weaker winter North Pacific pressure gradients and steeper summer North Pacific and Atlantic gradients. Comparison of this LIG subset ensemble with simulations of future low emissions scenarios indicates that the LIG may not be a sufficient analog for projected, end-21st century hydroclimatic change in North America.