Can we detect centennial sea-level variations over the last three thousand years in Israeli archaeological records?

Archaeological remains are valuable relative sea-level (RSL) indicators in Israel, a tectonically stable coast with minor isostatic inputs. Previous research has used archaeological indicators to argue for centennial sea-level fluctuations. Here, we place archaeological indicators in a quality-controlled dataset where all indicators have consistently calculated vertical and chronological uncertainties, and we subject the data to statistical analysis. We combine the archaeological data with bio-construction data from Dendropoma petraeum colonial vermetids. The final dataset consists of 99 relative sea-level index points and 12 limiting points from the last 4000 a. The temporal distribution of the index points is uneven; Israel has only four index points before 2000 a BP. We apply an Errors-In-Variables Integrated Gaussian Process (EIV IGP) to the index points to model the evolution of RSL. Results show RSL in Israel rose from −0.8 ± 0.5 m at ∼2750 a BP (Iron Age) to 0.0 ± 0.1 m by ∼1850 a BP (Roman period) at 0.8 mm/a, and continued rising to 0.1 ± 0.1 m until ∼1600 a BP (Byzantine Period). RSL then fell to −0.3 ± 0.1 m by 0.5 mm/a until ∼650 a BP (Late Arab period), before returning to present levels at a rate of 0.4 mm/a. The re-assessed Israeli record supports centennial-scale RSL fluctuations during the last 3000 a BP, although the magnitude of the RSL fall during the last 2000 a BP is 50% less. The new Israel RSL record demonstrates correspondence with regional climate proxies. This quality-controlled Israeli RSL dataset can serve as a reference for comparisons with other sea-level records from the Eastern Mediterranean.