Quantifying Past ENSO Variability From <i>Tridacna</i> δ ¹⁸ O Using a Pseudo‐Proxy Baseline Approach

Abstract Understanding the long‐term variability of the El Niño‐Southern Oscillation (ENSO) is essential for assessing its response to external forcings, yet instrumental records are too short to capture its full range of behavior. The giant clam ( Tridacna ) has emerged as a promising archive for high‐resolution ENSO reconstructions, but the scarcity of long modern Tridacna δ 18 O records—due to overfishing—has hindered efforts to establish reliable baselines for past variability. Here we develop a pseudo‐proxy baseline approach to quantitatively assess ENSO variability from fossil Tridacna δ 18 O records. To precisely simulate the behavior of Tridacna δ 18 O records, we advance the Tridacna proxy system model (PSM) by integrating ShellChron, a novel algorithm that accurately reconstructs shell growth information at sub‐annual timescales. The improved PSM is validated against the modern Tridacna δ 18 O record and used to generate a pseudo‐δ 18 O ensemble to serve as a modern baseline. This ensemble is driven by modern temperature and salinity data, but mirrors the sampling resolution, analytical error, and growth characteristics of the fossil record. Using this pseudo‐proxy baseline approach, we quantitatively estimate that ENSO variability was 43% (6%–66%, 95% confidence interval) lower in around 3300 BP compared to modern conditions, based on a 38‐year fossil Tridacna record in the northern South China Sea. This result is consistent with central equatorial Pacific coral records and Peruvian bivalve records. Our approach not only establishes a quantitative framework for Tridacna ‐based ENSO reconstructions, but also points to a potential pathway for assessing past climate variability when sufficiently long modern proxy records are unavailable.