SC-LUFormer: A Scenario-Conditioned Transformer for Long-Term Land-Use Prediction from Remote Sensing Data in Mountainous Urban Agglomerations
Open Access
- 12 June 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. PP (19391404) , 1-17
Abstract
Long-term land-use prediction based on remote sensing-derived data remains challenging in mountainous urban agglomerations because fragmented landscapes, complex terrain constraints, and scenario-dependent differences jointly affect spatial allocation and future projection reliability. To address these challenges, this study proposes a unified scenario-conditioned Transformer framework (SC-LUFormer) for long-term land-use prediction in mountainous urban agglomerations. Rather than introducing an entirely new neural primitive, the proposed framework integrates non-local spatial representation, scenario-aware feature modulation, and change-oriented evaluation into a single modeling framework, in which future socio-environmental scenarios are incorporated into the deep feature space via feature-wise linear modulation (FiLM). Multi-temporal land-use maps derived from remote sensing observations, together with multi-source biophysical and socioeconomic drivers, were used to validate the model through a 2020 backcasting experiment and to generate scenario-based land-use projections for 2040–2100 under multiple CMIP6-based Shared Socioeconomic Pathway (SSP) scenarios in the Central Yunnan Urban Agglomeration, China. Comparative experiments against representative convolutional neural network, machine-learning, and rule-based models show that SC-LUFormer achieves better backcasting performance, with an overall accuracy of 94.88%, a mean intersection-over-union of 85.71%, an area deviation of 3.17%, and a figure of merit of 0.2917 in reconstructing the year 2020. Further interpretability analysis using Grad-CAM and SHAP provides post-hoc indications that SC-LUFormer focuses on urban expansion boundaries and transition zones and produces scenario-responsive driver-attribution patterns under alternative future assumptions.Keywords
Funding Information
- National Natural Science Foundation of China (42561060)
- Yunnan Fundamental Research Projects (202401AT070360)
This publication has 41 references indexed in Scilit:
- Multi-Scenario Flood Susceptibility Projections in the Eastern Hindu Kush: Integrating Machine Learning, CMIP6, and Land Use ChangeEarth Systems and Environment, 2025
- GreenNet: A dual-encoder network for urban green space classification using high-resolution remotely sensed imagesInternational Journal of Applied Earth Observation and Geoinformation, 2025
- Overview of Land Use and Land Cover Change and Its Impacts on Natural Resources2025
- Assessment of ecological environment quality and their drivers in urban agglomeration based on a novel remote sensing ecological indexEcological Indicators, 2025
- GlobalMind: Global multi-head interactive self-attention network for hyperspectral change detectionISPRS Journal of Photogrammetry and Remote Sensing, 2024
- Multi-scale telecoupling effects of land use change on ecosystem services in urban agglomerations --A case study in the middle reaches of Yangtze River urban agglomerationsJournal of Cleaner Production, 2023
- Simulation of future land use/cover change (LUCC) in typical watersheds of arid regions under multiple scenariosJournal of Environmental Management, 2023
- The impact of multi-scenario land use change on the water conservation in central Yunnan urban agglomeration, ChinaEcological Indicators, 2023
- Explanation of machine learning models using shapley additive explanation and application for real data in hospitalComputer Methods and Programs in Biomedicine, 2022
- Understanding the drivers of sustainable land expansion using a patch-generating land use simulation (PLUS) model: A case study in Wuhan, ChinaComputers, Environment and Urban Systems, 2021