Performance and Efficiency of Climate In-Situ Data Reconstruction: Why Optimized IDW Outperforms kriging and Implicit Neural Representation

TL;DR

This paper compares three climate data reconstruction methods and finds that the simple inverse distance weighting (IDW) method outperforms kriging and neural network models in accuracy and efficiency across multiple datasets.

Contribution

It provides a comprehensive evaluation showing that optimized IDW surpasses more complex methods in climate data reconstruction tasks.

Findings

IDW achieved the lowest RMSE, MAE, and maximum error.

IDW demonstrated the highest R^2 value.

Differences were statistically significant with moderate to large effect sizes.

Abstract

This study evaluates three reconstruction methods for sparse climate data: the simple inverse distance weighting (IDW), the statistically grounded ordinary kriging (OK), and the advanced implicit neural representation model (MMGN architecture). All methods were optimized through hyper-parameter tuning using validation splits. An extensive set of experiments was conducted, followed by a comprehensive statistical analysis. The results demonstrate the superiority of the simple IDW method over the other reference methods in terms of both reconstruction accuracy and computational efficiency. IDW achieved the lowest RMSE ($3.00 \pm 1.93$), MAE ($1.32 \pm 0.77$), and $\Delta_{MAX}$ ($24.06 \pm 17.15$), as well as the highest $R^2$ ($0.68 \pm 0.16$), across 100 randomly sampled sparse datasets from the ECA\&D database. Differences in RMSE, MAE, and $R^2$ were statistically significant and exhibited moderate to large effect sizes. The Dunn post-hoc test further confirmed the consistent superiority of IDW across all evaluated quality measures [...]