Northern Mexico's future hydroclimate remains uncertain, with climate models showing discrepancies in projected drying. This uncertainty has significant social, economic, and ecological implications, considering the region's vulnerability to droughts. Paleoclimate records offer critical constraints on regional precipitation variability and can improve climate projections. However, few such records exist in Northern Mexico. Speleothems, with their precise dating and multiple hydrologically sensitive proxies, are ideal for reconstructing past hydroclimate. This study aims to address the knowledge gap by providing a high-resolution speleothem record from Tamaulipas, Mexico, to understand the interplay of dynamic and thermodynamic controls on precipitation variability in the region over orbital to millennial timescales. The Caribbean Low-Level Jet (CLLJ), a major moisture transport system, and its response to both orbital (insolation) and millennial (Heinrich Stadial) climate changes are investigated.

Previous studies in Northeast Mexico have shown unclear evidence of a strong insolation control on hydroclimate, despite proposals linking it to ITCZ migration and CLLJ strengthening. While some records show a positive correlation between summer insolation and runoff, others suggest autumn or spring insolation as more significant drivers. In contrast, the role of insolation is better understood in Northwest and Southern Mexico, where multiple records show a strong positive correlation with precipitation. Millennial-scale precipitation variability has been linked to CLLJ strength, but interpretations vary, highlighting the need for further investigation. Several studies have linked hydroclimate variations across Mesoamerica and the Caribbean to SST changes, but the influence of SST variations on millennial and orbital timescales remains poorly understood. The inconsistent responses of lake sediment records in northern Mexico to Heinrich Stadials underscore the need for more detailed paleoclimate data from the region, particularly in Northeast Mexico. Studies of extreme precipitation events in NE Mexico have suggested links between increased clay mineral concentrations (indicating high-intensity rainfall) and warm Gulf of Mexico SSTs, but the overall correlation between SST and precipitation remains weak or inconsistent, demanding further research.

A ~57,000-year multiproxy record (δ¹⁸O, δ¹³C, Mg/Ca) was developed from a stalagmite (CB2) collected from Cueva Bonita, Tamaulipas, Mexico. The chronology was established using a U-series age-depth model constrained by 33 ²³⁰Th-²³⁴U ages and 2000 Monte Carlo simulations with COPRA software. The average temporal resolution is ~36 years. Modern precipitation isotope systematics were analyzed using 15 years of moisture-bearing air trajectories and observational precipitation δ¹⁸O data from near the cave, which showed a strong correlation between δ¹⁸O and precipitation amount. A simple proxy system model demonstrated minimal cave environmental influence on speleothem δ¹⁸O. The δ¹⁸O record was interpreted as a proxy for precipitation amount, while δ¹³C was considered to be more locally influenced, reflecting vegetation changes, soil respiration, temperature, and CO₂ degassing. Mg/Ca ratios, also analyzed, provide additional insights into past precipitation and precipitation amount. To evaluate the climate dynamics associated with Heinrich Stadials, an isotope-enabled Earth System Model (iCESM1) simulation was conducted, adding freshwater to the North Atlantic on a glacial background state. This simulation examined SST changes in the Atlantic and Pacific, changes in wind patterns (easterlies), moisture flux, and the stable oxygen isotope ratio of precipitation (δ¹⁸O). Moisture budget analysis using the method of Seager & Henderson was employed to decompose precipitation changes into thermodynamic and dynamic components. The model results were compared to merged model-observation data from the Global Precipitation Climatology Project (GPCP).

The CB2 δ¹⁸O record shows millennial-scale variability, with significant positive excursions during Heinrich Stadials (HS1, 3–6, and Younger Dryas), indicating drier conditions. In contrast to some other tropical records, CB2 shows no clear precessional signal on orbital timescales. The δ¹³C record covaries with δ¹⁸O, exhibiting similar millennial-scale variations and a decrease across the deglaciation. Mg/Ca ratios also show similar millennial-scale trends, particularly during glacial periods, consistent with the impact of precipitation amount on calcite precipitation. The lack of a strong correlation between CB2 δ¹⁸O and summer insolation, along with stronger relationships with global temperature and atmospheric pCO₂, suggests that thermodynamic controls on orbital timescales may be more important for influencing precipitation at the study site. The iCESM1 model simulation showed significant SST cooling in the North Atlantic during Heinrich Stadials, coupled with a strengthening of tropical easterlies and increased δ¹⁸O in Northeast Mexico, confirming the drying conditions observed in the speleothem record. The model indicates that the combination of stronger easterlies and cooler SSTs reduces moisture transport and convection in the region. This drying pattern is spatially coherent with other records across Mesoamerica and the Caribbean. The study also found evidence for increased water balance during HS 2 and the LGM, which could be attributed to a weaker HS 2 event and/or reduced evapotranspiration due to colder temperatures. The speleothem record from Cueva Bonita and model simulation suggest that both dynamical and thermodynamical mechanisms play significant roles in driving hydroclimate variability in Northeast Mexico.

The findings of this study significantly advance our understanding of past hydroclimate changes in Northeast Mexico. The lack of a clear insolation signal challenges existing hypotheses that primarily emphasize insolation's role in driving regional precipitation. The strong influence of SSTs, particularly the Atlantic-Pacific inter-basin gradient, highlights the importance of considering both thermodynamic and dynamic factors in climate modeling. The consistent drying during Heinrich Stadials and the model's prediction of similar drying conditions caused by SST reduction and stronger easterlies strengthen the understanding of the dynamical and thermodynamical mechanisms. The study also emphasizes the importance of considering broader regional climate dynamics and highlights the interconnectedness of hydroclimate patterns across Mesoamerica and the Caribbean. The study’s results will enhance the accuracy of future climate projections for Northern Mexico, particularly by improving parameterizations of the CLLJ and its response to dynamic and thermodynamic changes.

This study presents the first high-resolution multiproxy speleothem record of hydroclimate variability in Northeast Mexico over the past ~57,000 years, revealing a stronger influence of SSTs than previously thought. The model-data comparison clarifies the roles of both thermodynamic (SST) and dynamic (easterlies) mechanisms in driving hydroclimate changes, especially during Heinrich Stadials. This work improves our understanding of the regional climate system and has implications for refining future climate projections for this vulnerable region. Future research could focus on expanding the spatial coverage of paleoclimate records in Northern Mexico to further refine regional climate models and improve predictions of future hydroclimate change.

While this study significantly advances our knowledge of Northeast Mexico's hydroclimate, certain limitations exist. The influence of factors like local vegetation changes and cave environmental variability on speleothem δ¹³C values is not fully resolved. Although efforts were made to address these through multiple proxy analysis and modeling, additional studies could help to refine the interpretations. The climate model used does not include tropical cyclones, which can significantly affect precipitation in the region. The impact of these events on past hydroclimate warrants further study. The limited spatial extent of the study also restricts the ability to generalize findings to the entire region of Northern Mexico, although the similarities to other records in Mesoamerica provide some support for larger-scale applications. Finally, while this study focuses on the influence of Atlantic and Pacific SSTs, other potential drivers such as changes in atmospheric circulation patterns and other climate systems may also play a role.