Heteroscedastic Gaussian Process Regression on the Alkenone over Sea Surface Temperatures
This work addresses the calibration of proxies for sea surface temperature reconstruction, which is incremental as it builds on existing Gaussian process methods with specific enhancements for noise and outliers.
The authors tackled the problem of restoring historical sea surface temperatures by developing a heteroscedastic Gaussian process regression model for alkenone proxies, which handles variable noise patterns and includes a Bayesian outlier classification method.
To restore the historical sea surface temperatures (SSTs) better, it is important to construct a good calibration model for the associated proxies. In this paper, we introduce a new model for alkenone (${\rm{U}}_{37}^{\rm{K}'}$) based on the heteroscedastic Gaussian process (GP) regression method. Our nonparametric approach not only deals with the variable pattern of noises over SSTs but also contains a Bayesian method of classifying potential outliers.