Spatial aggregation trends of cultivated land quality and landscape patterns: A remote sensing analysis

Cultivated land is a vital resource for agricultural production and plays an important role in ensuring food security and maintaining ecological balance. Effective use of cultivated land resources requires long-term, large-scale land quality evaluations that are rapid and accurate, along with targeted management measures. Remote sensing technology has the advantages of a wide monitoring range, short revisit period, and low cost and is a valuable tool for large-scale cultivated land quality monitoring. In this study, Landsat 8 OLI data were used to extract spectra and vegetation indices, and a remote sensing monitoring and evaluation model of cultivated land quality was established using soil sample data in the study area from 2018 and 2021. Based on the optimal random forest model (R 2 = 0.91, RMSE = 0.59, MAE = 0.46), the distribution of cultivated land quality in the study area in 2024 was predicted, enabling analysis of ecological landscape patterns and spatial aggregation trends of land quality. According to these patterns and trends in the study area, management zones were delineated, including restricted construction areas, comprehensive improvement areas, and key improvement areas, with corresponding suggestions for the management measures of cultivated land quality zoning. This study aims to clarify the spatial distribution of cultivated land quality grades over large areas, identify the optimal machine learning model and input parameters for cultivated land quality evaluation, enhance evaluation efficiency, and provide a scientific basis for formulating reasonable land use policies and precise measures to improve cultivated land quality.