Bayesian Synchronization of Proxy Paleorecords with Reference Chronologies
Marco A. Aquino–López Francesco Muschitiello Matthew Osman · 2026
Read Bayesian Synchronization of Proxy Paleorecords with Reference Chronologies by Marco A. Aquino–López Francesco Muschitiello Matthew Osman (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-10218.txt.
Introduction
Aquino-López, Muschitiello and Osman bring out BSync, a Bayesian tool for aligning paleoclimate proxy records to set chronologies. Proxy series sit on depth or rough age scales that must map onto calendar time, and old tools return one alignment with little sense of doubt. BSync treats the monotone time map as inference, with a link that locally stretches and squeezes the input scale. Readable rate terms let priors encode how sediment builds up.
Catalog record
| CREATOR | Marco A. Aquino–López Francesco Muschitiello Matthew Osman |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71636 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.10218
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bayesian Synchronization of Proxy Paleorecords with Reference Chronologies Marco A. Aquino–López Francesco Muschitiello Matthew Osman Abstract Many scientific fields compare two or more noisy time series that integrate the same underlying process but are recorded on different time scales. In paleoclimate studies, for example, proxy measurements are collected versus stratigraphic depth in a climate archive and then converted to calendar time. Synchronizing two proxy records often requires estimating an alignment that maps the depth (or preliminary age) of an input record onto the calendar–time scale of an absolutely–dated target record so that corresponding proxy signals line up. Existing alignment approaches are generally optimization–based and return a single transformation, providing limited formal uncertainty quantification. Here, we introduce BSync , a Bayesian synchronization framework that treats alignments as inference over a monotone time–mapping function to match an input to a target record. The alignment is expressed as a transformation of the input depth (or age) scale to match the target record, achieved through a link function that locally expands and compresses the input scale. The model is parameterized through interpretable local rate parameters, enabling the specification of priors on deposition times to regularize the alignment toward physically plausible deformations. BSync jointly infers the aligned chronology and provides posterior uncertainty for the time–warping function and the resulting age scale. In synthetic data experiments and a real–data case study, BSync yields well–calibrated credible intervals for the aligned time scale and achieves more accurate alignments than a state–of–the–art automated method, particularly when independent age constraints are sparse. Keywords: Bayesian statistics, paleoclimatology, climate proxies, age–depth models, statistical modeling in environmental science, uncertainty quantification 1 Introduction Sedimentary sequences recovered from the ocean floor, lakes, peat bogs, and other depositional environments preserve physical, biological and chemical fingerprints whose variations are often linked to past climate and environmental changes. Since direct observations of key climate variables are only available for recent decades — a period covering less than 10^{-7} of Earth’s history ( 11 ) — these fingerprints, or “proxies", are our only window into Earth’s deep climate history. The stratigraphic alignment of these climate proxy profiles is thus an essential application in for paleoclimate research: by matching the distinctive “wiggles" found across different proxy records, researchers can place multiple climate archives onto a shared, internally consistent age scale. This permits environmental signals to be robustly compared across different regions and environmental domains well beyond the instrumental era. The stratigraphic alignment (or synchronization) of climate records has historically been pursued by manually identifying and matching tie points marking sharp or distinctive transitions in the proxy stratigraphies in both an undated input record and an absolutely–dated target record (e.g. ( 46 ) ). Figure 1 shows an illustration of this practice, which assumes that climate transitions are synchronous in both input and target proxy signals. This tie–point synchronization is especially prevalent in paleoceanographic studies, where the lack of independent chronological constraints can hinder the construction of precise timescales ( 37 ) , particularly beyond the radiocarbon calibration window (ca. 50 ka before 1950 AD; ( 39 ; 7 ) ). Even when distinct proxy features exist — for example, abrupt climate shifts such as Dansgaard–Oeschger events ( 22 ; 13 ) — this approach remains limited because it relies on few discrete stratigraphic markers, which inhibits continuous quantification of alignment uncertainties. Additionally, tie–point alignments introduce subjective constraints on sedimentation, often resulting in “step changes" in deposition times that are difficult to validate or reconcile with realistic sedimentation processes. The alignment process is further complicated by the fact that sediments compact, expand, and accumulate unevenly over space and time ( 42 ) , resulting in markedly different accumulation histories across sites and depositional environments, regardless of proxy data resolution, or degree of proxy “noise". Figure 1: Illustration of stratigraphic alignment based on “manual” synchronization of common transitions (i.e., “tie–points”) determined between two climate proxy signals. The gray curve (Signal A) serves as an absolutely dated target record, while the blue curve (Signal B) represents the input record placed on a preliminary age scale before synchronization. Each record is divided into a set of stratigraphic periods, shown as horizontal bars: black bars for Signal A and red bars for Signal B. These bars represent the time intervals between successive transitions, and the midpoint of each interval defines the tie–point used for alignment. Dashed lines connect corresponding midpoints where the timing of both signals is assumed to be synchronous. Despite the circularity that climate synchronization entails (e.g. 9 ), proxy alignment remains a widely used and often necessary step in paleoclimate research. In fact, there is an urgent need to develop alignment methods that minimize subjectivity, ensure reproducibility, and allow quantifying uncertainties associated with the alignment process. In this study, while acknowledging the potential for interpretive circularity, we assume that tuning itself does not compromise the integrity of results; we thus focus on developing a more robust and reproducible alignment tool for geophysical problems where stratigraphic alignment is required. Beyond tie–point synchronization, more objective methods have been developed to estimate alignments by means of nonlinear optimization ( 12 ; 31 ; 28 ; 33 ; 29 ; 27 ) . These approaches typically maximize an overall goodness–of–fit metric by deforming the entirety of a proxy time series onto a target record. However, because the optimization is subject to constraints and penalties designed to impose sedimentation realism and guard against undesirable fitting behavior, these methods do not provide a strictly formal quantification of alignment uncertainty. Attempts have been made to quantify alignment uncertainty in a continuous fashion (e.g., 32 ). However, these methods often entail subjective and time–consuming “link functions", where the depth scale of the input stratigraphy must be aligned to a target record via manual selection of tie points prior to alignment. Over the past decade, significant progress has been made in addressing these limitations using Bayesian techniques. The benefits of these methods are manifold: they produce physically grounded alignments based on prior expectations of sediment accumulation rates variability ( ( 26 ; 24 ; 36 ) ), and provide credible intervals associated with the alignment process. However, despite their clear potential, these methods still have limitations. For instance, most have been designed for aligning benthic foraminifera–derived \delta^{18} O data, which typically have low temporal resolution and high signal–to–noise ratios. The most recent alignment method –BIGMACS– 24 , uses Gaussian Process regression; while powerful, this technique incorporates smoothing techniques that may dampen important information in the proxy data. Consequently, uncertainty estimates may be underestimated in certain cases, leading to overconfidence in the precision of the alignment. In addition, BIGMACS uses an empirical sedimentation prior based on a limited set of radiocarbon dates from low–latitude ocean sediment cores. Although this approach may work well in some contexts, the empirical prior reflects assumptions specific to those environments that do not account for radiocarbon reservoir effects and are entirely absent for polar oceans, so it may misrepresent the alignment characteristics of the sediment core being analyzed. To address some of these limitations and to improve the uncertainty quantification of the alignment process, here we present a new Bayesian synchronization method named BSync . BSync provides an integrated and generalized approach for aligning proxy time series