Evaluating the Alignment Between GeoAI Explanations and Domain Knowledge in Satellite-Based Flood Mapping

2026-04-28

AI Summary

This study addresses the opacity of deep learning models in satellite-based flood mapping and the absence of systematic evaluation methods that align model explanations with domain knowledge, such as surface spectral characteristics. To bridge this gap, the authors propose the ADAGE framework—the first explanation alignment evaluation system tailored for GeoAI. By introducing Channel-Group SHAP, the method quantifies the consistency between model interpretations and remote sensing prior knowledge, integrating domain expertise into the explainability validation pipeline. Evaluated on two flood mapping tasks, ADAGE effectively measures explanation alignment and aids experts in identifying model biases, thereby substantially enhancing the trustworthiness of GeoAI models in both scientific research and operational applications.

Abstract

The increasing number of satellites has improved the temporal resolution of Earth observation, making satellite-based flood mapping a promising approach for operational flood monitoring. Deep learning-based approaches for flood mapping using satellite imagery, an important application within Geospatial Artificial Intelligence (GeoAI), have shown improved predictive performance by learning complex spatial and spectral patterns from large volumes of remote sensing data. However, the opaque decision-making processes of deep learning models remain a major barrier to their integration into critical scientific and operational workflows. This highlights the need for a systematic assessment of whether model explanations align with established domain knowledge in remote sensing. To address this research gap, this study introduces the ADAGE (Alignment between Domain Knowledge And GeoAI Explanation Evaluation) framework. The proposed framework is designed to systematically evaluate how well explanations of deep learning models align with established remote sensing knowledge, particularly regarding the distinctive spectral properties of the Earth's surface. The ADAGE framework employs Channel-Group SHAP (SHapley Additive exPlanations) method to estimate the contributions of grouped input channels to pixel-level predictions. Experiments on two satellite-based flood mapping tasks demonstrate that the ADAGE framework can (1) quantitatively assess the alignment between model explanations and reference explanations derived from domain knowledge and (2) help domain experts identify misaligned explanations through alignment scores. This study contributes to bridging the gap between explainability and domain knowledge in GeoAI for Earth observation, enhancing the applicability of GeoAI models in scientific and operational workflows.

Problem

• GeoAI
• flood mapping
• model explainability
• domain knowledge
• satellite imagery

Innovation

• ADAGE framework
• Channel-Group SHAP
• GeoAI explainability
• domain knowledge alignment
• satellite-based flood mapping