Application of High-Precision Classification Method Based on Spatiotemporal Stable Samples and Land Use Policy in Oasis–Desert Mosaic Landscape Areas

Highlights: What are the main findings? A policy-aware method integrating spatiotemporally stable samples (two-step filter), land-use policy constraints, and a new Canopy Growth Index achieves 91.9% overall accuracy in a complex oasis–desert landscape. Incorporating policy constraints markedly improves spatiotemporal consistency, reducing pixels with unreasonable repetitive land-cover changes from 56.67% to 34.03% (−22.64 percentage points). What are the implications of the main findings? The workflow is simple and transferable; when large-scale products show poor local spatiotemporal consistency or low overall accuracy, this workflow supports high-precision mapping, and time-series aggregation enables post-classification correction and cross-year sample reuse. The resulting consistent, high-precision maps provide a reliable basis for evaluating ecological restoration and guiding sustainable land-use planning in ecologically fragile arid regions. Land cover products are essential tools in environmental and ecological research. However, limited attention has been paid to their data quality issues. Many existing products suffer from pronounced spatiotemporal inconsistencies, characterized by frequent and repetitive classification fluctuations in specific regions and years, which substantially compromise the accuracy of analyses and models that rely on them. To address these challenges, this study introduces a method for deriving spatiotemporally stable samples to support high-precision land cover classification. The approach integrates national and regional land-use policies to assess temporal stability and incorporates advanced time-series processing techniques together with innovative vegetation indices to facilitate effective sample reuse. Experimental results show that this method markedly improves classification accuracy across vegetation types and reduces the extent of areas prone to frequent land-cover changes by 22.64%. Compared with existing products of similar spatial resolution, our approach achieves an overall classification accuracy of 91.1%, providing stable, high-quality input data that underpin precise and reliable regional-scale environmental and ecological modeling.