Ecological environment quality reflects the overall condition and health of the environment. Analyzing the spatiotemporal dynamics and driving factors of ecological environment quality across large regions is crucial for environmental protection and policy-making. This study utilized the Google Earth Engine (GEE) platform to efficiently process large-scale remote sensing data and construct a multi-scale Remote Sensing Ecological Index (RSEI) based on Landsat and Sentinel data. This approach overcomes the limitations of traditional single-scale analyses, enabling a comprehensive assessment of ecological environment quality changes across provincial, municipal, and county levels in Fujian Province. Through the Mann–Kendall mutation test and Sen + Mann–Kendall trend analysis, the study identified significant change points in the RSEI for Fujian Province and revealed the temporal dynamics of ecological quality from 1987 to 2023. Additionally, Moran’s I statistic and Geodetector were employed to explore the spatial correlation and driving factors of ecological quality, with a particular focus on the complex interactions between natural factors. The results indicated that: (1) the integration of Landsat and Sentinel data significantly improved the accuracy of RSEI construction; (2) the RSEI showed a consistent upward trend across different scales, validating the effectiveness of the multi-scale analysis approach; (3) the ecological environment quality in Fujian Province experienced significant changes over the past 37 years, showing a trend of initial decline followed by recovery; (4) Moran’s I analysis demonstrated strong spatial clustering of ecological environment quality in Fujian Province, closely linked to human activities; and (5) the interaction between topography and natural factors had a significant impact on the spatial patterns of RSEI, especially in areas with complex terrain. This study not only provides new insights into the dynamic changes in ecological environment quality in Fujian Province over the past 37 years, but also offers a scientific basis for future environmental restoration and management strategies in coastal areas. By leveraging the efficient data processing capabilities of the GEE platform and constructing multi-scale RSEIs, this study significantly enhances the precision and depth of ecological quality assessment, providing robust technical support for long-term monitoring and policy-making in complex ecosystems.

Urban green hearts provide essential ecosystem services, including carbon sequestration, water purification, and hydrological regulation. The Green Heart Area of the Chang-Zhu-Tan Urban Agglomeration in Hunan Province, China, is the largest globally, and plays a critical role in regional water management. These functions are increasingly threatened by intensive land-use, while soil, as the foundational ecosystem component, mediates water retention, nutrient cycling, and erosion resistance. This study examined the effects of four land-use types—cropland, plantation, arbor woodland, and other woodland—on soil particle composition and key nutrients (organic carbon, total nitrogen, and total phosphorus). Statistical comparisons among land-use types were performed. Results indicated that silt was the dominant soil fraction across all land-uses (64–72%). Arbor woodland exhibited significantly higher sand content (29%) compared to cropland (19%; p < 0.05), suggesting improved water permeability and erosion resistance. Cropland showed elevated nutrient levels, with TN (1450.32 mg·kg⁻¹) and TP (718.86 mg·kg⁻¹) exceeding both national averages and those in arbor woodland. Coupled with acidic soil conditions (pH 5.23) and lower stoichiometric ratios (C/N: 10.82; C/P: 35.67; N/P: 3.29), these traits indicate an increased risk of nutrient leaching in croplands. In contrast, arbor woodland displayed more balanced C:N:P ratios (C/N: 12.21; C/P: 48.05; N/P: 3.84), supporting greater nutrient retention and aggregate stability. These findings underscore the significant influence of land-use type on soil ecological functions, including water infiltration, runoff reduction, and climate adaptability. The study highlights the importance of adopting conservation-oriented practices such as reduced tillage and targeted phosphorus management in croplands, alongside reforestation with native species, to improve soil structure and promote long-term ecological sustainability.

Cropland non-grain and non-agriculturalization monitoring (CNNM) is of great significance for ensuring national food security. Recently, semantic segmentation has been a promising approach that can fully exploit the advantages of high-resolution remote sensing imagery. It is of great significance to clarify the performance of semantic segmentation algorithms for CNNM and their influencing factors. This paper contributes to research on deep learning-based semantic segmentation models for CNNM by investigating the robustness and generalization ability of detection models in complex scenarios. To investigate these models, seven mainstream segmentation models were examined with respect to two self-constructed unmanned aerial vehicle (UAV)-based cropland monitoring datasets under various experimental settings. The experimental results reveal several key findings: For the datasets employed in the present study, remote sensing semantic segmentation models demonstrate superior performance relative to generic visual models. Increasing backbone depth does not guarantee significant performance improvements. The recognition challenges posed by task-specific categories such as agricultural facility land further underscore the need for tailored deep learning semantic segmentation algorithms for CNNM. These experimental results provide valuable insights into the practical performance of deep learning semantic segmentation models and offer useful guidance for future semantic segmentation research in the field of CNNM.