The quantitative inversion of the concentrations of water quality parameters could clarify the temporal and spatial distribution characteristic, migration, and conversion of water quality parameters. This study took the Danjiangkou Reservoir as the research object, and established an inversion model based on the reflectance of different band combinations of remote sensing analyses on Sentinel-2 images, combined with the water quality monitoring data of total nitrogen (TN) and ammonia nitrogen (NH₃-N) of the sampling sites in February 2016. The inversion results of TN and NH₃-N in 2020 were obtained, the variation of TN and NH₃-N concentrations in the reservoir area were analyzed, and the factors accounting for the variation were discussed. The results indicated that the fitting accuracy using the established model was high for both TN and NH₃-N, and R² was 0.782 for TN and 0.851 for NH₃-N, respectively, showing high predication accuracy, which could be suitable for remote sensing inversion of TN and NH₃-N concentrations in the Danjiangkou Reservoir. The NH₃-N concentration of the Danjiangkou Reservoir was in line with Class I from 2016 to 2020, while the TN concentration was between Class III and IV. The inter-annual changes indicated that the overall water quality had an upward trend. The main tributary in the northern of the Danjiangkou Reservoir had a heavy load of TN, and after entering the reservoir, the flow velocity decreased, which caused nitrogen to accumulate at the river entrance, leading to a high TN concentration. The large slope of the mountainous area cause soil erosion. The lost soil and water carried a large amount of pesticides and fertilizers, and the ground runoff carried a large amount of nitrogen into water body, which could account for the high NH₃-N concentration on the east and west sides of the southern part of the Danjiangkou Reservoir.

Image super-resolution (SR) techniques can improve the spatial resolution of remote sensing images to provide more feature details and information, which is important for a wide range of remote sensing applications, including land use/cover classification (LUCC). Convolutional neural networks (CNNs) have achieved impressive results in the field of image SR, but the inherent localization of convolution limits the performance of CNN-based SR models. Therefore, we propose a new method, namely, the dilated Transformer generative adversarial network (DTGAN) for the SR of multispectral remote sensing images. DTGAN combines the local focus of CNNs with the global perspective of Transformers to better capture both local and global features in remote sensing images. We introduce dilated convolutions into the self-attention computation of Transformers to control the network’s focus on different scales of image features. This enhancement improves the network’s ability to reconstruct details at various scales in the images. SR imagery provides richer surface information and reduces ambiguity for the LUCC task, thereby enhancing the accuracy of LUCC. Our work comprises two main stages: remote sensing image SR and LUCC. In the SR stage, we conducted comprehensive experiments on Landsat-8 (L8) and Sentinel-2 (S2) remote sensing datasets. The results indicate that DTGAN generates super-resolution (SR) images with minimal computation. Additionally, it outperforms other methods in terms of the spectral angle mapper (SAM) and learned perceptual image patch similarity (LPIPS) metrics, as well as visual quality. In the LUCC stage, DTGAN was used to generate SR images of areas outside the training samples, and then the SR imagery was used in the LUCC task. The results indicated a significant improvement in the accuracy of LUCC based on SR imagery compared to low-resolution (LR) LUCC maps. Specifically, there were enhancements of 0.130 in precision, 0.178 in recall, and 0.157 in the F1-score.

The landscape of a coal gangue backfi lling reclamation region in the Yang Quan First mine has experienced remarkable land backfilling reclamation over the past few decades. Effective management strategies are needed to protect the environment in these areas as well as the sustainable development of land resources. In this paper, a study of pattern change is proposed by using five landscape metrics and a stability analysis of features derived from land cover maps in order to investigate their magnitude and its comprehensive ecological stability in the gangue backfilling reclamation region between a separate time period (2000-14). The results illustrate that: The artificial reclamation project disturbed the landscape so that patching and ecological stability has changed. The vegetation landscape patch indices all decreased during the study period, showing that the reclamation had a greater impact on the vegetation. The comprehensive index of ecological effects of landscape ecology was 0.3615, showing that the area’s landscape remained stable during the study period. Backfill landscape stability is lowest, which indicates that landscape stability is mainly affected by human disturbance, while the influence of natural and semi natural landscapes is relatively low.

After large-scale land consolidation in hilly loess region of the Loess Plateau in China, land subsidence has a wide affecting area and considerable difficulty of prevention. Hence, large-scale, stabilized, and continuous deformation monitoring is urgently needed for slopes. In this study, land consolidation zone in the loess platform area of Weinan, China, was selected as the object, and the 30-scene Sentinel-1A data in Jan, 2018 to Dec, 2019 were analyzed. The mean annual velocity of ground deformation was from -6.19 mm∙a^-1 to 3.86 mm∙a^-1, and Accumulated deformation velocity was within -8.49 mm∙a^-1 to 7.24 mm∙a^-1. Accumulated deformation of land consolidation changed with the seasons changing. The interrelationship between the spatiotemporal variations in ground subsidence and the precipitation, ground water, loess engineering properties was also discussed. Accumulated deformation of land consolidation changed with the seasons changing. The precipitation accelerated the subsidence by unexpected strong precipitation reflects that the infiltration of rainwater can lead to compacted loess deformation which caused by moistening effect. Under varying ground water environment, external loads may lead to soil collapse, resulting in non-uniform land subsidence. Co-compression deformation of original loess and compacted loess is main influencing factors of subsidence. These findings have important implications and significant positive effects on the prevention of potential hazard such as subsidence and side slope slip.