Houssou et al. (2026) Spatial pattern regression for meteorological fields interpolation

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-09-15
• Authors: Vihotogbé Houssou, Julie Carreau
• DOI: 10.5194/hess-30-5791-2026

Research Groups

• Department of Mathematics and Industrial Engineering, Polytechnique Montréal
• GERAD – Groupe d’Études et de Recherche en Analyse des Décisions
• Mila – Quebec Artificial Intelligence Institute
• IVADO – Institute for Data Valorization

Short Summary

This paper introduces Spatial Pattern Regression (SPR), a data-driven method that reconstructs gridded meteorological fields by combining spatial information extracted from high-resolution regional climate model simulations with station observations. SPR outperforms baseline interpolation methods, particularly under sparse network conditions.

Objective

• To develop an interpolation method that exploits the spatial structure of RCM simulations through spatial patterns.
• To evaluate and compare the performance of SPR against commonly used baseline interpolation methods (IDW, OK, and KED).

Study Configuration

• Spatial Scale: Regional climate model simulations cover a North American domain discretized on a regular grid of 280×280 cells, with a horizontal resolution of approximately 11 km.
• Temporal Scale: The study considers daily precipitation, minimum temperature, and maximum temperature over the auxiliary period (1980–2009) and interpolation period (2000–2009).

Methodology and Data

• Models used: Spatial Pattern Regression (SPR), Inverse Distance Weighting (IDW), Ordinary Kriging (OK), and Kriging with External Drift (KED).
• Data sources: Regional climate model simulations, station observations from Environment and Climate Change Canada.

Main Results

• SPR outperforms baseline interpolation methods in reconstructing gridded meteorological fields under sparse network conditions.
• The method is particularly effective for precipitation, minimum temperature, and maximum temperature variables.
• Sensitivity analyses highlight the dominant role of station density and location on interpolation accuracy.

Contributions

• This paper contributes to the development of an interpolation method that exploits the spatial structure of RCM simulations through spatial patterns.
• SPR provides a physically-constrained spatial reconstruction framework that bridges the gap between traditional statistical interpolation and model-based reanalyses.

Funding

• This research was supported by the Natural Sciences and Engineering Research Council (NSERC) of Canada.

Citation

@article{Houssou2026Spatial,
  author = {Houssou, Vihotogbé and Carreau, Julie},
  title = {Spatial pattern regression for meteorological fields interpolation},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-5791-2026},
  url = {https://doi.org/10.5194/hess-30-5791-2026}
}