GloMarGridding : A Python Toolkit for Flexible Spatial Interpolation in Climate Applications

Key findings from this study

• The study demonstrates that GloMarGridding decouples spatial interpolation from upstream dataset processing, enabling independent assessment of interpolation-induced structural uncertainty.
• The authors report that the toolkit supports three spatial covariance parametrizations and propagates uncertainty via error covariance matrices and conditional simulation.
• The researchers establish that this framework enables researchers to evaluate how interpolation method choices affect the uncertainty structure in gridded temperature fields.

Overview

GloMarGridding is a Python package that isolates and evaluates structural uncertainty arising from spatial interpolation methods in global temperature dataset construction. The toolkit decouples interpolation from upstream processing stages, enabling independent assessment of how interpolation choices propagate uncertainty into gridded climate outputs.

Methods and approach

The package implements Gaussian Process Regression Modelling to generate spatially complete temperature fields from grid-box averages and point observations. GloMarGridding supports three spatial covariance parametrizations: fixed isotropic variograms, ellipse-based anisotropic approaches, and empirically derived covariance matrices. The framework propagates uncertainty through error covariance matrices and conditional simulation from input ensembles.

Results

GloMarGridding provides a modular framework that separates spatial interpolation from homogenization, quality control, and aggregation stages. This decoupling permits researchers to independently evaluate how upstream processing choices affect gridded outputs. The toolkit generates uncertainty estimates alongside spatially complete temperature fields, enabling quantification of interpolation-specific structural uncertainty contributions.

Implications

By isolating interpolation uncertainty from other processing chain components, GloMarGridding facilitates more transparent evaluation of global temperature dataset construction. The toolkit supports sensitivity analyses of covariance parametrization choices and their downstream effects on climate analyses. This modular approach may reduce ambiguity when comparing structural uncertainty contributions across different dataset production methods.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.