Deji Wuyun

Soil salinity mapping in heterogeneous irrigation districts faces a dual challenge: the high dimensionality of hyperspectral data leads to redundancy, while the scarcity of ground-truth samples restricts the generalization of data-driven models. Traditional regression methods often struggle to capture non-linear spectral responses under such “small-sample” conditions. To address these limitations, this study proposes a semi-supervised retrieval framework coupling Optimal Band Combination Analysis (OBCA) with a Spectral Wasserstein GAN with Gradient Penalty (S-WGAN-GP). We constructed a robust feature set via cross-scenario evaluation and developed a rigorous “Uncertainty-Aware Filtering” protocol to screen synthetic samples generated by a teacher mechanism. The OBCA screening revealed that salinity-sensitive features are robustly clustered in the Green (550–570 nm) and Near-Infrared (NIR, 880–950 nm) regions, with NIR bands demonstrating superior stability across different sites. The proposed S-WGAN-GP successfully densified the feature manifold by generating 1186 high-fidelity synthetic samples. By incorporating these augmented data, the inversion accuracy was substantially improved: the R² of the optimal SVR model increased from 0.36 (baseline) to 0.60 (+66.7%), and the RMSE decreased from 7.06 to 5.57 dSm⁻¹. This study confirms that physically constrained generative augmentation, when combined with rigorous quality control, effectively bridges the distribution gap in limited datasets. The proposed framework offers a transferable and accurate solution for fine-scale soil salinity monitoring in data-scarce arid regions.

Amid growing global food security concerns and frequent armed conflicts, real-time monitoring of abandoned cropland is essential for strategic planning and crisis management. This study develops a method to map abandoned cropland accurately, crucial for maintaining the food supply chain and ecological balance. Utilizing Sentinel-1/2 satellite data, we employed multi-feature stacking and machine learning to create the ARCC10-IM (Abandoned and Reclaimed Cropland Classification at 10-meter resolution in Inner Mongolia) dataset, which tracks annual cropland activity. A novel temporal segmentation algorithm was developed to extract cropland abandonment and reclamation patterns annually, using sliding time windows over several years. This research differentiates cropland states—active cultivation, unstable fallowing, continuous abandonment, and reclamation—providing continuous, regional-scale maps with 10-meter resolution. ARCC10-IM is crucial for land planning, environmental monitoring, and agricultural management in arid areas like Inner Mongolia, enhancing decision-making and technology in land use tracking.

The prompt and precise identification of corn and soybeans are essential for making informed decisions in agricultural production and ensuring food security. Nonetheless, conventional crop identification practices often occur after the completion of crop growth, lacking the timeliness required for effective agricultural management. To achieve in-season crop identification, a case study focused on corn and soybeans in the U.S. Corn Belt was conducted using a crop growth curve matching methodology. Initially, six vegetation indices datasets were derived from the publicly available HLS product, and then these datasets were integrated with known crop-type maps to extract the growth curves for both crops. Furthermore, crop-type information was acquired by assessing the similarity between time-series data and the respective growth curves. A total of 18 scenarios with varying input image numbers were arranged at approximately 10-day intervals to perform identical similarity recognition. The objective was to identify the scene that achieves an 80% recognition accuracy earliest, thereby establishing the optimal time for early crop identification. The results indicated the following: (1) The six vegetation index datasets demonstrate varying capabilities in identifying corn and soybean. Among those, the EVI index and two red-edge indices exhibit the best performance, all surpassing 90% accuracy when the entire time-series data are used as input. (2) EVI, NDPI, and REVI2 indices can achieve early identification, with an accuracy exceeding 80% around July 20, more than two months prior to the end of the crops’ growth periods. (3) Utilizing the same limited sample size, the early crop identification method based on crop growth curve matching outperforms the method based on random forest by approximately 20 days. These findings highlight the considerable potential and value of the crop growth curve matching method for early identification of corn and soybeans, especially when working with limited samples.

The desert steppe serves as a transitional zone between grasslands and deserts, and long-term monitoring of aboveground biomass (AGB) in the desert steppe is essential for understanding grassland changes. While AGB observation techniques based on multisource remote-sensing data and machine-learning algorithms have been widely applied, research on monitoring methods specifically for the desert steppe remains limited. In this study, we focused on the desert steppe of Inner Mongolia, China, as the study area and used field sampling data, MODIS data, MODIS-based vegetation indices (VI), and environmental factors (topography, climate, and soil) to compare the performance of four commonly used machine-learning algorithms: multiple linear regression (MLR), partial least-squares regression (PLS), random forest (RF), and support vector machine (SVM) in AGB estimation. Based on the optimal model, the spatial–temporal characteristics of AGB from 2000 to 2020 were calculated, and the driving forces of climate change and human activities on AGB changes were quantitatively analyzed using the random forest algorithm. The results are as follows: (1) RF demonstrated outstanding performance in terms of prediction accuracy and model robustness, making it suitable for AGB estimation in the desert steppe of Inner Mongolia; (2) VI contributed the most to the model, and no significant difference was found between soil-adjusted VIs and traditional VIs. Elevation, slope, precipitation, and temperature all had positive effects on the model; (3) from 2000 to 2020, the multiyear average AGB in the study area was 58.34 g/m², exhibiting a gradually increasing distribution pattern from the inner region to the outer region (from north to south); (4) from 2000 to 2020, the proportions of grassland with AGB slightly and significantly increasing trend in the study area were 87.08% and 5.13%, respectively, while the proportions of grassland with AGB slightly and significantly decreasing trend were 7.76% and 0.05%, respectively; and (5) over the past 20 years, climate change, particularly precipitation, has been the primary driving force behind AGB changes of the study area. This research holds reference value for improving desert steppe monitoring capabilities and the rational planning of grassland resources.

Knowledge of grassland classification in a timely and accurate manner is essential for grassland resource management and utilization. Although remote sensing imagery analysis technology is widely applied for land cover classification, few studies have systematically compared the performance of commonly used methods on semi-arid native grasslands in northern China. This renders the grassland classification work in this region devoid of applicable technical references. In this study, the central Xilingol (China) was selected as the study area, and the performances of four widely used machine learning algorithms for mapping semi-arid grassland under pixel-based and object-based classification methods were compared: random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), and naive Bayes (NB). The features were composed of the Landsat OLI multispectral data, spectral indices, Sentinel SAR C bands, topographic, position (coordinates), geometric, and grey-level co-occurrence matrix (GLCM) texture variables. The findings demonstrated that (1) the object-based methods depicted a more realistic land cover distribution and had greater accuracy than the pixel-based methods; (2) in the pixel-based classification, RF performed the best, with OA and Kappa values of 96.32% and 0.95, respectively. In object-based classification, RF and SVM presented no statistically different predictions, with OA and Kappa exceeding 97.5% and 0.97, respectively, and both performed significantly better than other algorithms. (3) In pixel-based classification, multispectral bands, spectral indices, and geographic features significantly distinguished grassland, whereas, in object-based classification, multispectral bands, spectral indices, elevation, and position features were more prominent. Despite the fact that Sentinel 1 SAR variables were chosen as an effective variable in object-based classification, they made no significant contribution to the grassland distinction.

Soil salinization is a widespread environmental hazard and a major abiotic constraint affecting global food production and threatening food security. Salt-affected cropland is widely distributed in China, and the problem of salinization in the Hetao Irrigation District (HID) in the Inner Mongolia Autonomous Region is particularly prominent. The salt-affected soil in Inner Mongolia is 1.75 million hectares, accounting for 14.8% of the total land. Therefore, mapping saline cropland in the irrigation district of Inner Mongolia could evaluate the impacts of cropland soil salinization on the environment and food security. This study hypothesized that a reasonably accurate regional map of salt-affected cropland would result from a ground sampling approach based on PlanetScope images and the methodology developed by Sentinel multi-sensor images employing the machine learning algorithm in the cloud computing platform. Thus, a model was developed to create the salt-affected cropland map of HID in 2021 based on the modified cropland base map, valid saline and non-saline samples through consistency testing, and various spectral parameters, such as reflectance bands, published salinity indices, vegetation indices, and texture information. Additionally, multi-sensor data of Sentinel from dry and wet seasons were used to determine the best solution for mapping saline cropland. The results imply that combining the Sentinel-1 and Sentinel-2 data could map the soil salinity in HID during the dry season with reasonable accuracy and close to real time. Then, the indicators derived from the confusion matrix were used to validate the established model. As a result, the combined dataset, which included reflectance bands, spectral indices, vertical transmit–vertical receive (VV) and vertical transmit–horizontal receive (VH) polarization, and texture information, outperformed the highest overall accuracy at 0.8938, while the F1 scores for saline cropland and non-saline cropland are 0.8687 and 0.9109, respectively. According to the analyses conducted for this study, salt-affected cropland can be detected more accurately during the dry season by using just Sentinel images from March to April. The findings of this study provide a clear explanation of the efficiency and standardization of salt-affected cropland mapping in arid and semi-arid regions, with significant potential for applicability outside the current study area.

The cropland in the farming-pastoral ecotone of Northern China is highly unstable owing to environmental restoration projects, poor soil fertility, poverty, and rural labor loss, and it is characterized by a large number of fallow fields. Mapping fallow fields in a farming-pastoral ecotone can help in evaluating the impact of complex cropland landscapes on the environment and food security. To map the fallow fields using the Sentinel-1 and Sentinel-2 archives, a multi-metric dataset of vertical transmit–vertical receive + vertical transmit–horizontal receive polarization was established. Moreover, spectral bands and vegetation index datasets were established using Google Earth Engine to classify cropped and fallow fields using the Random Forest classifier. The overall accuracies and Kappa coefficients of different datasets were assessed to examine the dataset with the highest overall accuracy in the main growing season. A 10 m resolution fallow map for 2020 was then generated based on the combined Sentinel-1 and Sentinel-2 datasets with the highest overall accuracy and Kappa coefficients were 95.82% and 0.92, respectively. In addition, the time-series characteristics of the entropy eigenvalues generated via dual-polarization decomposition were quantitatively evaluated to clarify the contribution of the Sentinel-1 synthetic-aperture radar archive to the fallow field mapping. The eigenvalues were more sensitive to the phenological characteristics of cropped and fallow fields than the original backscatter signal of the Sentinel-1 data. Moreover, the mapping method was tested at different time intervals by gradually aggregating the results across an increasing number of months to optimize the fallow field monitoring using the minimum number of observations possible within a short period. Data aggregated over August achieved the highest one-month accuracy; it was also very close to the observations from the whole growing season. The results further emphasize the influence of Sentinel-1 archives on fallow field mapping. Overall, this study clarifies the potential applicability of Sentinel archives for monitoring and mapping managing patterns of agricultural land in a region.