SSL-SoilNet: A Hybrid Transformer-Based Framework With Self-Supervised Learning for Large-Scale Soil Organic Carbon Prediction

• Published in: IEEE transactions on geoscience and remote sensing 2024, Vol.62, p.1-15
• Main Authors: Kakhani, Nafiseh, Rangzan, Moien, Jamali, Ali, Attarchi, Sara, Kazem Alavipanah, Seyed, Mommert, Michael, Tziolas, Nikolaos, Scholten, Thomas
• Format: Article
• Language: English
• Subjects: Algorithms; Carbon; Carbon cycle; Carbon sequestration; Climatic data; Contrastive learning; Correlation coefficient; Correlation coefficients; Data models; Decision making; Decision trees; deep learning (DL); Digital imaging; Digital mapping; digital soil mapping (DSM); Ecological function; Ecosystem services; Europe; Information management; Information processing; Land management; LUCAS; Machine learning; Mapping; Meteorology; Mitigation; Nutrient balance; Nutrient cycles; Organic carbon; Organic soils; Remote sensing; Root-mean-square errors; Satellite imagery; Self-supervised learning; self-supervised model; Soil fertility; Soil mapping; soil organic carbon (SOC); Soil properties; spatiotemporal model; Statistical models; Training; Transformers
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2024.3446042

Soil organic carbon (SOC) constitutes a fundamental component of terrestrial ecosystem functionality, playing a pivotal role in nutrient cycling, hydrological balance, and erosion mitigation. Precise mapping of SOC distribution is imperative for the quantification of ecosystem services, notably carbon sequestration and soil fertility enhancement. Digital soil mapping (DSM) leverages statistical models and advanced technologies, including machine learning (ML), to accurately map soil properties, such as SOC, utilizing diverse data sources like satellite imagery, topography, remote sensing indices, and climate series. Within the domain of ML, self-supervised learning (SSL), which exploits unlabeled data, has gained prominence in recent years. This study introduces a novel approach that aims to learn the geographical link between multimodal features via self-supervised contrastive learning, employing pretrained Vision Transformers (ViT) for image inputs and Transformers for climate data, before fine-tuning the model with ground reference samples. The proposed approach has undergone rigorous testing on two distinct large-scale datasets, with results indicating its superiority over traditional supervised learning models, which depends solely on labeled data. Furthermore, through the utilization of various evaluation metrics (e.g., root-mean-square error (RMSE), mean absolute error (MAE), concordance correlation coefficient (CCC), etc.), the proposed model exhibits higher accuracy when compared to other conventional ML algorithms like random forest and gradient boosting. This model is a robust tool for predicting SOC and contributes to the advancement of DSM techniques, thereby facilitating land management and decision-making processes based on accurate information.