MODELING TERMINAL LAKE TEMPERATURE AND RUNOFF USING OSTRACOD PALEOCLIMATE RECORDS

Lake systems are an important paleoclimate archive used to test ecosystem and hydrologic responses to important periods of Earth history such as rapid CO₂ increases and glacial-interglacial cycles. Measurements of carbonate ostracod valve geochemistry (e.g. δ¹⁸O, δ¹³C, ⁸⁷Sr/⁸⁶Sr, [Sr], and [Mg]) provide indicators of hydrologic variability in lake systems throughout the geologic record. In this study we present a new terminal lake modeling approach that employs a system of total differential equations describing how δ¹⁸O, ⁸⁷Sr/⁸⁶Sr, Sr/Ca, and Mg/Ca of the ostracod valves change as a function of temperature, runoff, lake evaporation and changing source area. Using equally spaced time steps these equations are simultaneously solved to constrain the hydrologic parameters of the lake that are recorded in the ostracod valves. We use a Monte Carlo approach to account for the uncertainty in the input parameters (e.g., Mg/Ca and δ¹⁸O temperature relationships and watershed solute chemistry).

We use this model to quantify the hydrologic changes observed in a recently published ostracod valve dataset from the Cretaceous Songliao Basin, northeast China. Spanning the Upper Cretaceous, the ostracod valves were measured at < 3 m resolution through lacustrine facies of the SK-1 drill cores. The Cretaceous Ocean Anoxic Event 3 (OAE 3) and the top of OAE 2—events of probable, rapid CO₂ increase—are recorded in the Songliao Basin. Additionally, this lake record demonstrates large excursions and variations in δ¹⁸O, typically associated with temperature or runoff driven hydrologic variability. Focusing on a short (< 5 Myr) portion of the ostracod record, we find that during OAE 3 the Songliao Basin watershed experienced an ~ 3 °C increase in temperature and an ~ 10 % increase in runoff. Incorporating equations to also describe the ⁸⁷Sr/⁸⁶Sr variations over > 5 Myr timescales we find that the modeled shift in source area contributions to the lake is consistent with outcrop evidence. The modeling approach presented in this study can be applied to other terminal lake records, could be modified for other calcifying species (i.e., gastropods or mollusks) or inorganic lacustrine carbonate, and holds promise for extension with additional proxy measurements (i.e., δD or U/Ca).