A reproducible, data-centric exploration using LinkedEarth Tools
This repository contains a collection of Jupyter notebooks that explore the relationship between external climate forcings—in particular solar variability and orbital forcing—and paleoclimate variability across a range of timescales, from the last two millennia to the late Pleistocene.
The notebooks are designed as a narrative introduction to modern paleoclimate data analysis, combining standardized data formats, programmatic data discovery, and transparent time-series analysis in a fully reproducible environment. These notebooks are meant to be used as both research and teaching resources.
Understanding how external forcings influence climate variability is a central question in paleoclimatology. However, detecting these signals in proxy records is challenging due to age uncertainty and irregular sampling, proxy-specific sensitivities, internal climate variability, and the risk of false positives in spectral analysis.
This project emphasizes statistical rigor, careful significance testing, and physical interpretation, rather than simple pattern matching.
The notebooks address questions such as:
- Is there robust evidence for a solar influence on climate variability over the Common Era?
- What periodicities emerge in large paleoclimate temperature compilations?
- How consistent are spectral signals across archives and regions?
- How do orbital-scale forcings manifest in individual high-resolution records?
- What are the limitations of common spectral and coherence methods in paleoclimate applications?
The notebooks are intended to be read roughly in the following order.
01-solar_climate2k.ipynb
Explores the relationship between solar forcing and temperature reconstructions over the last two millennia, including exploratory analysis, visual comparison, and discussion of coherence, significance, and interpretation.
2. Exploring the relationship between insolation and millennium to orbital climate variability at scale
02a-query_lipd_graph.ipynb
Introduces graph-based discovery of paleoclimate datasets using LiPDGraph and SPARQL, demonstrating how metadata-driven queries can be used to assemble analysis-ready datasets.
02b-spectral_analysis_temp12k.ipynb
Applies spectral analysis techniques to temperature records from the Temp12k database, focusing on methodological choices and their implications.
02c-spectral_analysis_peaks_temp12k.ipynb
Identifies and evaluates spectral peaks, with emphasis on red-noise benchmarks, uncertainty, and archive-dependent behavior.
03-EpicaDomeC_explore.ipynb
A deep dive into the EPICA Dome C ice core, exploring long-term variability, spectral structure, and interpretive challenges.
04-IODP339-U1385.ipynb
Analyzes orbital-scale variability in a marine sediment core, linking observed periodicities to orbital forcing while highlighting uncertainty and methodological caveats.
This project builds on the open paleoclimate software and data ecosystem, including:
- LiPD (Linked Paleo Data) for standardized data representation
- LiPDGraph for metadata querying using graph technologies
- PyLiPD for opening and manipulating LiPD files.
- Pyleoclim for time-series, spectral, and coherence analysis
- Public paleoclimate data repositories such as NOAA NCEI and PANGAEA
All analyses are designed to be transparent, inspectable, and reusable.
This material is intended for:
- Graduate students learning paleoclimate data analysis
- Researchers transitioning to Python-based, reproducible workflows
- Instructors seeking worked examples of spectral analysis and forcing attribution
- Researchers interested in large-scale, metadata-driven paleoclimate synthesis
Some familiarity with time-series analysis and basic climate concepts is helpful, but the notebooks are written to be pedagogical and exploratory.
- All analyses are performed in Jupyter notebooks
- Code and narrative are interleaved to support transparency
- The repository is suitable for use as a Jupyter Book
- Users are encouraged to adapt workflows to their own datasets
If you reuse this material, please cite the relevant datasets and software packages.
To run the notebooks locally:
git clone <repo-url>
cd <repo>
conda env create -f environment.yml
conda activate <env-name>
jupyter labAll notebooks herein are provided under an Apache 2.0 license.
We needn't tell you that making research tools accessible requires time and effort. If you find any of these resources useful and use them in your own research, please do us the kindness of one or more citations. Notebooks in this collection are registered on Zenodo, and associated with a digital object identifier (DOI). A ready-to-use citation is provided on this GitHub repository in APA and BibTex (in the "About" section on the right panel, click on "Cite this repository"). If you use any of the software, please cite them as well. It will make us (and our sponsors) very happy to hear that these investments spawned more research.
This material is based upon work supported by the National Science Foundation under Grant Number ICER-2126510. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the National Science Foundation.