Fine-Scale Soil Mapping in Alaska with Multimodal Machine Learning

2025-06-17

AI Summary

Accelerated near-surface permafrost degradation in Alaska due to climate change threatens infrastructure integrity and soil carbon storage, necessitating high-resolution permafrost and soil classification mapping. To address this, we propose MISO—a novel multimodal architecture integrating geospatial foundation models, implicit neural representations (INRs), and contrastive learning. MISO jointly models heterogeneous remote sensing data (e.g., SAR, optical, topographic) and sparse in-situ observations, significantly enhancing generalization to remote, unobserved regions and improving geographic awareness and spatial fidelity. Compared to Random Forest, MISO achieves substantially higher recall in permafrost zones and demonstrates superior cross-regional stability. Systematic validation across Alaska’s major land resource areas confirms its robustness and operational utility. This work advances permafrost dynamics monitoring, vulnerability assessment, and adaptive land-use planning under climate change.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Multimodal Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Fine-scale soil mapping in Alaska, traditionally relying on fieldwork and localized simulations, remains a critical yet underdeveloped task, despite the region's ecological importance and extensive permafrost coverage. As permafrost thaw accelerates due to climate change, it threatens infrastructure stability and key ecosystem services, such as soil carbon storage. High-resolution soil maps are essential for characterizing permafrost distribution, identifying vulnerable areas, and informing adaptation strategies. We present MISO, a vision-based machine learning (ML) model to produce statewide fine-scale soil maps for near-surface permafrost and soil taxonomy. The model integrates a geospatial foundation model for visual feature extraction, implicit neural representations for continuous spatial prediction, and contrastive learning for multimodal alignment and geo-location awareness. We compare MISO with Random Forest (RF), a traditional ML model that has been widely used in soil mapping applications. Spatial cross-validation and regional analysis across Permafrost Zones and Major Land Resource Areas (MLRAs) show that MISO generalizes better to remote, unseen locations and achieves higher recall than RF, which is critical for monitoring permafrost thaw and related environmental processes. These findings demonstrate the potential of advanced ML approaches for fine-scale soil mapping and provide practical guidance for future soil sampling and infrastructure planning in permafrost-affected landscapes. The project will be released at https://github.com/knowledge-computing/Peatland-permafrost.

Problem

• Develop fine-scale soil maps for Alaska's permafrost regions
• Improve permafrost monitoring with advanced machine learning techniques
• Enhance adaptation strategies for climate change impacts

Innovation

• Vision-based ML model for soil mapping
• Implicit neural representations for spatial prediction
• Contrastive learning for multimodal alignment