(1.Hubei Province Laboratory for Geographical Process Analyzing & Modeling, Wuhan 430079,China;2.College of Urban and Environmental Sciences, Central China Normal University, Wuhan 430079,China)
Abstract:This study examines digital soil mapping (DSM) technology by systematically reviewing its developmental trajectory, analyzing core methodologies, evaluating application potential, addressing current challenges, and forecasting future trends. Initially constrained by technological limitations, DSM relied on traditional field surveys and manual cartography, resulting in suboptimal outcomes. The introduction of personal computing in the 1970s catalyzed early digitization efforts, leading to transformative advancements in geographic information systems (GIS) and modeling innovations from the 1980s to the 2010s. Post-2010, DSM matured through the fusion of multi-source data, artificial intelligence (AI)-driven innovation, and global collaborations. The technical framework of DSM integrates traditional and modern approaches: data acquisition combines conventional sampling, remote sensing, and proximal sensing to enhance data quality, diversity, operational efficiency, and spatial resolution; data processing applies advanced geostatistics and optimized machine learning algorithms to improve prediction accuracy, model generalizability, and interpretability; and model development merges traditional methods with intelligent frameworks, simulating soil processes and constructing novel algorithms to enhance simulation fidelity and mapping precision. DSM demonstrates extensive applications across multiple domains: precision agriculture — optimizing fertilization, irrigation, and field management to enhance agricultural productivity and sustainability; land resource management — facilitating spatial planning and preventing soil degradation; environmental-climate research — quantifying ecosystem services while safeguarding biodiversity through climate resilience strategies; and soil census innovation — modernizing survey methodologies and unlocking multidimensional value in Third National Soil Census datasets. Current challenges include inconsistent data quality, prediction uncertainties, limited model accuracy, barriers to cross-disciplinary collaboration, and insufficient integration of knowledge. However, the convergence of AI, big data, and Internet of Things (IoT) technologies is expected to drive DSM into a new era of intelligent, refined, and real-time applications: AI unlocks the full potential of data, big data fosters global sharing platforms, and IoT enables real-time sensing and data transmission. These innovations are set to expand DSM’s application scope, deepen interdisciplinary collaboration, and strengthen the foundations of sustainable soil management, ecological stability, green agriculture, and climate resilience.