The lithium isotope system is increasingly being applied to a variety of Earth science studies, as the burgeoning literature attests; over 180 papers have been published in the last twelve years that report lithium isotope data, including five review papers that cover different aspects of lithium isotope applications (Elliott et al. 2004; Tomascak 2004; Tang et al. 2007b; Burton and Vigier 2011; Schmitt et al. 2012), and a book (Tomascak et al. 2016). The upswing in lithium isotope studies over the past decade reflects analytical advances that have made Li measurements readily obtainable. These include the use of multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) for relatively precise solution measurements (Tomascak et al. 1999a) and secondary ion mass spectrometry (SIMS) for high spatial resolution measurements (Chaussidon and Robert 1998; Kasemann et al. 2005; Bell et al. 2009). In addition, lithium isotope studies are motivated by the large variety of problems for which they may provide insight, including crust–mantle recycling, silicate weathering, fluid–rock interaction, as well as geospeedometry. The great interest in the Li system that spurred the development of these new analytical methods was initiated by the pioneering work of Lui-Heung Chan, who demonstrated not only that Li isotopic fractionation can be very large at or near the Earth’s surface (Chan and Edmond 1988), but also that Li isotopes are strongly fractionated during seawater-basalt interaction (Chan et al. 1992). This discovery naturally led to the search for a recycled slab signature in Li isotopes of arc lavas (some of the earlier studies include Moriguti and Nakamura 1998a; Chan et al. 1999, 2002b; Tomascak et al. 2000, 2002; Leeman et al. 2004; Moriguti et al. 2004), as well as more deeply derived intraplate basalts (e.g., Chan and Frey 2003 …

Strontium isotope ratios (⁸⁷Sr/⁸⁶Sr) are gaining considerable interest as a geolocation tool and are now widely applied in archaeology, ecology, and forensic research. However, their application for provenance requires the development of baseline models predicting surficial ⁸⁷Sr/⁸⁶Sr variations (“isoscapes”). A variety of empirically-based and process-based models have been proposed to build terrestrial ⁸⁷Sr/⁸⁶Sr isoscapes but, in their current forms, those models are not mature enough to be integrated with continuous-probability surface models used in geographic assignment. In this study, we aim to overcome those limitations and to predict ⁸⁷Sr/⁸⁶Sr variations across Western Europe by combining process-based models and a series of remote-sensing geospatial products into a regression framework. We find that random forest regression significantly outperforms other commonly used regression and interpolation methods, and efficiently predicts the multi-scale patterning of ⁸⁷Sr/⁸⁶Sr variations by accounting for geological, geomorphological and atmospheric controls. Random forest regression also provides an easily interpretable and flexible framework to integrate different types of environmental auxiliary variables required to model the multi-scale patterning of ⁸⁷Sr/⁸⁶Sr variability. The method is transferable to different scales and resolutions and can be applied to the large collection of geospatial data available at local and global levels. The isoscape generated in this study provides the most accurate ⁸⁷Sr/⁸⁶Sr predictions in bioavailable strontium for Western Europe (R² = 0.58 and RMSE = 0.0023) to date, as well as a conservative estimate of spatial uncertainty by applying quantile regression forest. We anticipate that the method presented in this study combined with the growing numbers of bioavailable ⁸⁷Sr/⁸⁶Sr data and satellite geospatial products will extend the applicability of the ⁸⁷Sr/⁸⁶Sr geo-profiling tool in provenance applications.