Spatial Distribution-Shift Aware Knowledge-Guided Machine Learning

TL;DR

This paper introduces SDSA-KGML, a machine learning approach that incorporates spatial heterogeneity and location-dependent parameters to improve land emission predictions across diverse regions, addressing limitations of traditional models.

Contribution

The paper presents a novel spatial distribution-shift aware knowledge-guided machine learning model that effectively captures spatial heterogeneity for land emission prediction.

Findings

Higher local accuracy in Midwest land emission predictions

Effective modeling of spatial heterogeneity in soil moisture

Improved prediction performance over traditional methods

Abstract

Given inputs of diverse soil characteristics and climate data gathered from various regions, we aimed to build a model to predict accurate land emissions. The problem is important since accurate quantification of the carbon cycle in agroecosystems is crucial for mitigating climate change and ensuring sustainable food production. Predicting accurate land emissions is challenging since calibrating the heterogeneous nature of soil properties, moisture, and environmental conditions is hard at decision-relevant scales. Traditional approaches do not adequately estimate land emissions due to location-independent parameters failing to leverage the spatial heterogeneity and also require large datasets. To overcome these limitations, we proposed Spatial Distribution-Shift Aware Knowledge-Guided Machine Learning (SDSA-KGML), which leverages location-dependent parameters that account for significant spatial heterogeneity in soil moisture from multiple sites within the same region. Experimental results demonstrate that SDSA-KGML models achieve higher local accuracy for the specified states in the Midwest Region.