A pseudoproxy emulation of the PAGES 2k database using a hierarchy of proxy system models

Area of Science:

• Paleoclimatology
• Climate modeling
• Data science

Background:

• Paleoclimate reconstructions are crucial for climate assessments.
• Evaluating reconstruction methods requires rigorous benchmarking.
• Existing methods lack transparency and sensitivity analysis.

Purpose of the Study:

• To develop a robust dataset for evaluating paleoclimate reconstruction techniques.
• To create realistic synthetic proxies using proxy system models (PSMs).
• To assess the impact of various errors and biases on climate reconstructions.

Main Methods:

• Utilizing proxy system models (PSMs) to emulate climate signal to proxy data translation.
• Applying PSMs to the PAGES 2k dataset with realistic spatiotemporal and error structures.
• Employing a hierarchical approach to generate dataset variants, isolating specific error sources.

Main Results:

• Developed the "pseudoPAGES2k" dataset, a comprehensive resource for pseudoproxy experiments (PPEs).
• Demonstrated the dataset's utility in isolating impacts of sampling bias, representation error, and other assumptions.
• Showcased a framework for conducting PPEs with emerging climate field reconstruction techniques.

Conclusions:

• The pseudoPAGES2k dataset offers a transparent and tractable test bed for paleoclimate reconstruction methods.
• This resource enhances the evaluation of climate assessment methodologies and proxy network sensitivities.
• Facilitates improved understanding and application of paleoclimate data in climate science.