Submitted as: development and technical paper
05 Apr 2022
Submitted as: development and technical paper |  | 05 Apr 2022
Status: a revised version of this preprint is currently under review for the journal GMD.

Towards variance-conserving reconstructions of climate indices with Gaussian Process Regression in an embedding space

Marlene Klockmann, Udo von Toussaint, and Eduardo Zorita

Abstract. We present a new framework for the reconstruction of climate indices based on proxy data such as tree rings. The framework is based on the supervised learning method Gaussian Process Regression (GPR) and designed to preserve the amplitude of past climate variability and to adequately handle noise-contaminated proxies and variable proxy availability in time. To this end, the GPR is performed in a modified input space, termed embedding space. We test the new framework for the reconstruction of the Atlantic Multi-decadal Variability (AMV) in a controlled environment with pseudoproxies derived from coupled climate-model simulations. In this test environment, the GPR outperforms benchmark reconstructions based on multi-linear Principle Component Regression. On AMV-relevant timescales, i.e., multi-decadal timescales, the GPR is able to reconstruct the true magnitude of variability even if the proxies contain a non-climatic noise signal and become sparser back in time. Thus, we conclude that the embedded GPR framework is a highly promising tool for climate-index reconstructions.

Marlene Klockmann et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2022-32', Anonymous Referee #1, 06 Jul 2022
  • RC2: 'Comment on gmd-2022-32', Anonymous Referee #2, 04 Aug 2022
  • AC1: 'Comment on gmd-2022-32', Marlene Klockmann, 30 Aug 2022

Marlene Klockmann et al.

Marlene Klockmann et al.


Total article views: 982 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
810 151 21 982 31 7 7
  • HTML: 810
  • PDF: 151
  • XML: 21
  • Total: 982
  • Supplement: 31
  • BibTeX: 7
  • EndNote: 7
Views and downloads (calculated since 05 Apr 2022)
Cumulative views and downloads (calculated since 05 Apr 2022)

Viewed (geographical distribution)

Total article views: 906 (including HTML, PDF, and XML) Thereof 906 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 22 Oct 2023
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.