Advancing provenance assignment using machine learning and time series analysis of chemical chronologies in archival tissues

Sep 17, 2026

Uploaded by Theodora

Accurate provenance assignment underpins ecological connectivity and conservation decisions. Natural chemical chronologies preserved in metabolically inert, incrementally growing tissues such as otoliths offer strong promise, yet traditional workflows struggle with biological turnover, force time information into single values, and require subjective interpretation. Arai and colleagues introduce a machine learning time-series framework with automated feature extraction, explicit classification using DTW and KNN, and ensemble modeling, validated on 87Sr/86Sr otolith profiles from California Chinook salmon.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Background and importance of provenance tools

Chemical records in inert incrementally growing tissues

Biological challenges: turnover and maternal effects

Ecological Informatics 93 (2026) 103604 Contents lists available at ScienceDirect Ecological Informatics journal [homepage:](homepage: www.elsevier.com/locate/ecolinf)[ www.elsevier.com/locate/ecolinf](homepage: www.elsevier.com/locate/ecolinf) | Advancing provenance assignment using machine learning and time series analysis of chemical chronologies in archival tissues
Kohma Arai a,*, Malte Willmes b, Rachel C. Johnson a,c, Anna M. Sturrock a,d
a Center for Watershed Sciences, University of California Davis, One Shields Ave., Davis, CA 95616, USA b Norwegian Institute for Nature Research, PO Box 5685, Torgarden, 7485 Trondheim, Norway
c National Marine Fisheries Service, Fisheries Ecology Division, Southwest Fisheries Science Center, 110 McAllister Way, Santa Cruz, CA 95060, USA d School of Life Sciences, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK | | | | | --- | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords:
Otolith chemistry
Strontium isotope ratios (87Sr/86Sr) Chinook salmon
Random forest
Dynamic time warping (DTW) Ensemble methods | | Accurate provenance assignment is critical for understanding ecological connectivity and guiding conservation and management. Natural chemical chronologies stored in metabolically inert, incrementally growing tissues (e. g., otoliths) provide a powerful tool for this purpose. However, traditional approaches face biological challenges, collapse chronological data into single values, and require subjective interpretation, limiting accuracy and scalability. We present a novel framework that integrates machine learning, time series analysis, and ensemble modeling to improve provenance assignment from archival tissue chemistry. Using otolith 87Sr/86Sr profiles from 17 distinct natal sources (n = 255) of California Central Valley Chinook salmon, we developed automated feature extraction, explicit time series classification (including dynamic time warping [DTW] with k-nearest neighbors [KNN]), and ensembles combining multiple classifiers. Furthermore, to address gaps in under-sampled life histories within natal sources, we added simulated chemical profiles to the reference baseline and tested whether they improved model performance. Time series approaches (mean accuracy: 0.56–0.62) consistently outperformed traditional methods (0.48), particularly for sources influenced by maternal signatures or early dispersal. Feature extraction approaches performed best when profiles followed predictable life-stage patterns, while explicit time series classification (DTW + KNN) excelled for distinct profile “shapes”. Ensembles leveraged complementary strengths and outperformed any single method. Our results highlight the advantages of treating archival chemical data as time series and applying machine learning and ensemble strategies to enhance the accuracy and scalability of provenance assignment. This framework is broadly applicable across taxa, tissue types, and chemical markers, offering a roadmap for advancing ecological inference and conservation. | | 1. Introduction Tracing the origins and movements of organisms is key to understanding ecological processes and informing conservation and management. Provenance tools are broadly used in contexts ranging from migration and connectivity studies to wildlife forensics, food traceability, and species management (Bataille et al., 2020; Carter and Chesson, 2017; Hobson, 1999; Reich et al., 2024). Chemical records in metabolically inert, incrementally growing tissues such as bird feathers, mammal teeth, and fish otoliths have been widely used for this purpose (Newsome et al., 2010; Reis-Santos et al., 2023; van Dijk et al., 2014). In aquatic organisms, the chemistry of these tissues reflects the physicochemical properties of ambient water conditions (Hüssy et al., 2020), enabling reconstruction of provenance and movement histories across taxa and ecosystems (Doubleday et al., 2022).

FAQs

Why is provenance assignment important in ecology?

Provenance assignment supports understanding ecological processes and informing conservation and management decisions by tracing origins and movement patterns.

What limitations affect traditional chemical-tag approaches?

They can be limited by biological turnover, reduce chronological information to single values, and rely on subjective interpretation, reducing accuracy and scalability.

What methods does the paper propose to improve provenance assignment?

It integrates machine learning with time-series analysis, including automated feature extraction, explicit time-series classification using DTW with KNN, and ensemble models that combine multiple classifiers.

Sep 17, 2026

Uploaded by Theodora