Augmenting Ground-Level PM2.5 Prediction via Kriging-Based Pseudo-Label Generation

TL;DR

This paper introduces a data augmentation method using kriging-based pseudo-labels from satellite data to improve ground-level PM2.5 prediction models, leading to better spatial correlation and lower errors.

Contribution

It presents a novel approach to augment training data with pseudo-labels generated via kriging, enhancing model performance in climate modeling.

Findings

Improved spatial correlation in PM2.5 predictions

Reduced prediction error with augmented data

Enhanced CNN-RF model performance

Abstract

Fusing abundant satellite data with sparse ground measurements constitutes a major challenge in climate modeling. To address this, we propose a strategy to augment the training dataset by introducing unlabeled satellite images paired with pseudo-labels generated through a spatial interpolation technique known as ordinary kriging, thereby making full use of the available satellite data resources. We show that the proposed data augmentation strategy helps enhance the performance of the state-of-the-art convolutional neural network-random forest (CNN-RF) model by a reasonable amount, resulting in a noteworthy improvement in spatial correlation and a reduction in prediction error.