Accurate and timely extraction and evaluation of sandy land are essential for ecological environmental protection; it is urgent to do the research to support the sustainable development goals (SDGs) of Land Degradation Neutrality. This study used Sentinel-1 Synthetic Aperture Radar (SAR) data and Landsat 8 OLI multispectral data as the main data sources. Combining the rich spectral information from optical data and the penetrating advantages of radar data, a feature-level fusion method was employed to unveil the intrinsic nature of vegetative cover and accurately identify sandy land. Simultaneously, leveraging the results obtained from training with measured data, a comprehensive desertification assessment model was proposed, which combines multiple indicators to achieve a thorough evaluation of sandy land. The results showed that the method based on feature-level fusion achieved an overall accuracy of 86.31% in sandy land detection in Gansu Province, China. The integrated multi-indicator model C22_C/FVC is the ratio of correlation texture features of VH to vegetation cover based on which sandy land can be classified into three categories. When C22_C/FVC is less than 2.2, the pixel is classified as fixed sandy land. Pixels of semi-fixed sandy land have an indicator value between 2.2 and 5.2. Shifting sandy land has values greater than 5.2. Results showed that shifting sandy land and semi-fixed sandy land are the predominant types in Gansu Province, with 85,100 square kilometers and 87,100 square kilometers, respectively. The acreage of fixed sandy land was the least, 51,800 square kilometers. The method presented in this paper is robust for the detection and evaluation of sandy land from satellite imageries, which can potentially be applied for conducting high-resolution and large-scale detection and evaluation of sandy land.

Evapotranspiration (ET) is an important part of the water, carbon, and energy cycles in ecosystems, especially in the drylands. However, due to the particularity of sparse vegetation, the estimation accuracy of ET has been relatively low in the drylands. Therefore, based on the dry climate and sparse vegetation distribution characteristics of the drylands, this study optimized the core algorithms (canopy boundary resistance, aerodynamic resistance, and sparse vegetation coverage) and explored an ET estimation method in the Shuttleworth–Wallace two-layer model (SW model). Then, the Beijing–Tianjin sandstorm source region (BTSSR) was used as the study area to evaluate the applicability of the improved model in the drylands. Results show that: (1) The R² value of the improved model results was increased by 1.4 and the RMSE was reduced by 1.9 mm, especially in extreme value regions of ET (maximum or minimum). (2) Regardless of the spatial distribution and seasonal changes of the ET (63–790 mm), the improved ET estimation model could accurately capture the differences. Furtherly, the different vegetation regions could stand for the different climate regions to a certain extent. The accuracy of the optimized model was higher in the semi-arid region (R² = 0.92 and 0.93), while the improved model had the best improvement effect in the arid region, with R² increasing by 0.12. (3) Precipitation was the decisive factor affecting vegetation transpiration and ET, with R² value for both exceeding 0.9. The effect of vegetation coverage (VC) was less. This method is expected to provide a more accurate and adaptable model for the estimation of ET in the drylands.

Net primary productivity (NPP) is an important ecological indicator to evaluate ecosystem, and it is useful for land degradation assessment and monitoring. However, owing to dryland's particularity, retrieving vegetation properties from satellite remote sensing presents some significant challenges in sparse vegetation area. In this study, based on the wildly used time-series GF-1 data, the NPP estimation in sparse vegetation area was analyzed. Results showed that GF-1 data have high spatial and high temporal resolution characteristics, it is useful to distinguish land cover types in semi-arid areas based on NDVI time series data, the accuracy was 83.37% and Kappa coefficient was 0.79. Some key parameters of grassland were simulated and optimized based on CASA model. Compared with the measured data, the result was R ² with 0.71, and results indicated that NPP estimation by GF-1 data based on the new parameters in semi-arid area is feasible.

Effective dynamic agricultural planning is crucial for optimising resource allocation and ensuring income stability, yet conventional methods often face limitations in adapting to the complex and variable conditions of mountainous regions, particularly under fluctuating climate and market pressures. Therefore, this study introduces a novel multi-stage dynamic optimization framework specifically designed for crop planning in such challenging terrains. This framework is underpinned by a sophisticated model integrating advanced monitoring systems with a Hybrid Simulated Annealing Genetic Algorithm (H-SAGA), further enhanced by neural network-driven real-time predictions. The H-SAGA component optimises planting strategies by synergistically combining global exploration (SA) and local refinement (GA) capabilities, while the neural network dynamically adjusts revenue forecasts based on climatic and market data, significantly improving the model's responsiveness and adaptability. We rigorously evaluated the applicability and effectiveness of this model through extensive simulations across 7,290 mu (1,201 acres) of diverse farmland in mountainous Northern China. The results demonstrate that the proposed H-SAGA approach consistently achieves 5-10 percentage points higher profit increment ratios than other benchmark optimization algorithms (such as GA, SA, PSO, and ACO), alongside faster convergence and notable robustness against environmental and economic variability. This research establishes an integrated "monitoring-modelling-decision" paradigm, driven by multi-source data and machine learning, offering a practical and robust tool that provides valuable guidance for enhancing resource allocation efficiency and promoting sustainable precision agriculture in complex topographical regions, thereby holding significant reference value for optimising agricultural production nationwide.

The identification of ecosystem types is important in ecological environmental assessment. However, due to cloud and rain and complex land cover characteristics, commonly used ecosystem identification methods have always lacked accuracy in subtropical urban agglomerations. In this study, China’s Guangdong–Hong Kong–Macao Greater Bay Area (GBA) was taken as a study area, and the Sentinel-1 and Sentinel-2 data were used as the fusion of active and passive remote sensing data with time series data to distinguish typical ecosystem types in subtropical urban agglomerations. Our results showed the following: (1) The importance of different features varies widely in different types of ecosystems. For grassland and arable land, two specific texture features (VV_dvar and VH_diss) are most important; in forest and mangrove areas, synthetic-aperture radar (SAR) data for the months of October and September are most important. (2) The use of active time series remote sensing data can significantly improve the classification accuracy by 3.33%, while passive time series remote sensing data improves by 4.76%. When they are integrated, accuracy is further improved, reaching a level of 84.29%. (3) Time series passive data (NDVI) serve best to distinguish grassland from arable land, while time series active data (SAR data) are best able to distinguish mangrove from forest. The integration of active and passive time series data also improves precision in distinguishing vegetation ecosystem types, such as forest, mangrove, arable land, and, especially, grassland, where the accuracy increased by 21.88%. By obtaining real-time and more accurate land cover type change information, this study could better serve regional change detection and ecosystem service function assessment at different scales, thereby supporting decision makers in urban agglomerations.