Towards variance-conserving reconstructions of climate indices with Gaussian process regression in an embedding space

Klockmann, Marlene; von Toussaint, Udo; Zorita, Eduardo

doi:https://doi.org/10.5194/gmd-17-1765-2024

Articles | Volume 17, issue 4

https://doi.org/10.5194/gmd-17-1765-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

Benchmark datasets and machine learning algorithms for Earth...

https://doi.org/10.5194/gmd-17-1765-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 17, issue 4

Development and technical paper

|

28 Feb 2024

Development and technical paper |

| 28 Feb 2024

Towards variance-conserving reconstructions of climate indices with Gaussian process regression in an embedding space

Marlene Klockmann, Udo von Toussaint, and Eduardo Zorita

Supplement

https://doi.org/10.5194/gmd-17-1765-2024-supplement

Data sets

A global multiproxy database for temperature reconstructions of the Common Era Hali Kilbourne et al. https://doi.org/10.6084/m9.figshare.c.3285353.v2

CCSM4 coupled simulation for CMIP5 past 1000 years (850-1850) with natural forcings Bette Otto-Bliesner https://doi.org/10.1594/WDCC/CMIP5.NRS4pk

CCSM4 coupled run for CMIP5 historical (1850-2005) Jerry Meehl https://doi.org/10.1594/WDCC/CMIP5.NRS4hi

Short summary

Reconstructions of climate variability before the observational period rely on climate proxies and sophisticated statistical models to link the proxy information and climate variability. Existing models tend to underestimate the true magnitude of variability, especially if the proxies contain non-climatic noise. We present and test a promising new framework for climate-index reconstructions, based on Gaussian processes, which reconstructs robust variability estimates from noisy and sparse data.