Updated multi-method estimates of Lake Titicaca's radiocarbon reservoir offset

Abstract

Inaccurate lake chronologies are often an outcome of unresolved radiocarbon (¹⁴C) reservoir effects. These phenomena may impose considerable challenges to the ¹⁴C dating of lacustrine samples, limiting the accuracy of environmental and archaeological research. In South America, the importance of such studies around the Lake Titicaca basin, which presents a vast biodiversity and a rich archaeological heritage, cannot be overstated. However, the lack of preserved terrestrial organic macrofossils in the region hinders the derivation of age offsets for the lake. Despite a previous attempt to quantify the ¹⁴C reservoir effect associated with Lake Titicaca, the phenomenon requires further investigation. Here, we used a combination of new and published ¹⁴C dates to recalculate the offset using Bayesian approaches, obtaining a value of R = 163 ± 83 ¹⁴C yr for the period between 4000 and 3000 cal BP. Moreover, in the present study we employed a novel approach using mercury peaks as a chrono-tracer across different cores to construct a Bayesian chronological model. This allowed us to derive an R value of 315 ± 33 ¹⁴C yr for the southern basin of the lake at 1590 CE, indicating temporal variation in the radiocarbon reservoir effect. The same approach allowed us to calculate offsets of 166 ± 53 ¹⁴C yr and 335 ± 29 ¹⁴C yr for the shallow and deep parts of the basin, respectively, showing evidence of depth-dependent variation. The methods are thoroughly described and the potential explanations for these values are discussed.