Deep classifier kriging for probabilistic spatial prediction of air quality index

2025-12-29

AI Summary

Traditional kriging methods suffer from modeling inaccuracies when applied to non-Gaussian and highly nonlinear air quality index (AQI) spatial fields. To address this, we propose the Deep Classifier Kriging (DCK) framework—a novel probabilistic interpolation paradigm that unifies classifier-driven distribution estimation, spatial covariance learning, and multi-source data fusion (ground station observations and numerical model outputs). DCK explicitly relaxes the linearity and Gaussianity assumptions inherent in classical kriging, enabling non-collocated bivariate modeling and yielding high-accuracy, well-calibrated full predictive distributions. Evaluated on both synthetic and real-world AQI datasets, DCK significantly improves point prediction accuracy and uncertainty quantification quality compared to conventional approaches. It produces spatially resolved predictive distributions, facilitating practical applications such as exceedance probability mapping and risk assessment of extreme pollution events.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate spatial interpolation of the air quality index (AQI), computed from concentrations of multiple air pollutants, is essential for regulatory decision-making, yet AQI fields are inherently non-Gaussian and often exhibit complex nonlinear spatial structure. Classical spatial prediction methods such as kriging are linear and rely on Gaussian assumptions, which limits their ability to capture these features and to provide reliable predictive distributions. In this study, we propose extit{deep classifier kriging} (DCK), a flexible, distribution-free deep learning framework for estimating full predictive distribution functions for univariate and bivariate spatial processes, together with a extit{data fusion} mechanism that enables modeling of non-collocated bivariate processes and integration of heterogeneous air pollution data sources. Through extensive simulation experiments, we show that DCK consistently outperforms conventional approaches in predictive accuracy and uncertainty quantification. We further apply DCK to probabilistic spatial prediction of AQI by fusing sparse but high-quality station observations with spatially continuous yet biased auxiliary model outputs, yielding spatially resolved predictive distributions that support downstream tasks such as exceedance and extreme-event probability estimation for regulatory risk assessment and policy formulation.

Problem

• Develops deep classifier kriging for non-Gaussian spatial prediction
• Enables data fusion of heterogeneous air pollution sources
• Provides probabilistic AQI distributions for regulatory risk assessment

Innovation

• Deep classifier kriging for non-Gaussian spatial prediction
• Data fusion integrates heterogeneous air pollution sources
• Distribution-free deep learning framework for predictive distributions