Work

Geo-Mamba: A data-driven Mamba framework for spatiotemporal modeling with multi-source geographic factor integration

Year: 2025

Type: article

Abstract: • Geo-Mamba: the framework integrating Mamba and KANs for efficient geospatial-temporal modeling. • Systematic classification and unified integration of dynamic, static, and categorical geographic factors. • Mamba-based selective encoder efficiently models complex associations among diverse geographic features. Earth science data exhibit inherent complexity characterized by heterogeneous spatiotemporal attributes, high collinearity among variables, and diverse input formats. Despite rapid advancements in deep learning, geographic modeling lacks unified frameworks for integrating heterogeneous spatiotemporal data and diverse factor types. While Mamba architecture has demonstrated efficiency in large language models, computer vision and remote sensing, its applicability to geographic modeling remains unexplored. This study introduces Geo-Mamba, a novel framework addressing these challenges through three key innovations. First, we propose a systematic geographical factor classification method that categorizes elements into dynamic, static, and categorical factors, enabling standardized integration of heterogeneous data within a unified paradigm. Second, we design a selective encoder module based on Mamba architecture that leverages its linear complexity and scanning mechanism to establish selective state spaces for geographical inputs, revealing intricate associations between diverse feature types. Third, we incorporate Kolmogorov-Arnold Network (KAN) layers as intermediate components replacing multilayer perceptron linear layers, enhancing numerical regression accuracy in geographical applications. Experimental validation across three tasks demonstrates Geo-Mamba’s effectiveness: in net ecosystem exchange modeling (R2 = 0.92, RMSE = 0.37 μmol·m –2 ·s –1 ), groundwater storage anomaly downscaling (R 2 = 0.95, RMSE = 1.916 cm), and land cover classification (accuracy = 88.12 %, F1-Score = 84.27 %). These results confirm Geo-Mamba as an efficient unified framework for complex Earth science modeling, while establishing its viability for geographical data processing and factor integration.

Source: International Journal of Applied Earth Observation and Geoinformation

DOI: 10.1016/j.jag.2025.104854

Authors: Zongwen Shi, Chengyi Zhao, Kaixin Wang, Xiangkui Kong, Jianting Zhu

Institutions: Nanjing University of Information Science and Technology, State Key Laboratory of Geodesy and Earth's Dynamics, Innovation Academy for Precision Measurement Science and Technology, CAS, China University of Geosciences (Beijing), University of Wyoming

Language: English

Topic: Spatial and Panel Data Analysis

Subfield: Economics and Econometrics

Field: Economics, Econometrics and Finance

Domain: Social Sciences

Open Access status: gold

Estimated APC paid: $3,550

Funder: National Key Research and Development Program of China