Cloud cover is inevitable in optical remote sensing (RS) imagery on account of the influence of observation conditions, which limits the availability of RS data. Therefore, it is of great significance to be able to reconstruct the cloud-contaminated ground information. This paper presents a sparse dictionary learning-based image inpainting method for adaptively recovering the missing information corrupted by thick clouds patch-by-patch. A feature dictionary was learned from exemplars in the cloud-free regions, which was later utilized to infer the missing patches via sparse representation. To maintain the coherence of structures, structure sparsity was brought in to encourage first filling-in of missing patches on image structures. The optimization model of patch inpainting was formulated under the adaptive neighborhood-consistency constraint, which was solved by a modified orthogonal matching pursuit (OMP) algorithm. In light of these ideas, the thick-cloud removal scheme was designed and applied to images with simulated and true clouds. Comparisons and experiments show that our method can not only keep structures and textures consistent with the surrounding ground information, but also yield rare smoothing effect and block effect, which is more suitable for the removal of clouds from high-spatial resolution RS imagery with salient structures and abundant textured features.

Oceanic dissolved oxygen (DO) is crucial for oceanic material cycles and marine biological activities. However, obtaining subsurface DO values directly from satellite observations is limited due to the restricted observed depth. Therefore, it is essential to develop a connection between surface oceanic parameters and subsurface DO values. Machine learning (ML) methods can effectively grasp the complex relationship between input attributes and target variables, making them a valuable approach for estimating subsurface DO values based on surface oceanic parameters. In this study, the potential of ML methods for subsurface DO retrieval is analyzed. Among the selected ML methods, namely support vector regression (SVR), random forest (RF) regression, and extreme gradient boosting (XGBoosting) regression, the RF method generally demonstrates superior performance. As the depth increases, the accuracy of DO estimates tends to initially decrease, then gradually improve, with the poorest performance occurring at the depth of 600 dbar. The range of determination coefficients (R²) and root mean square error (RMSE) values based on the test dataset at different depths lies between 0.53 and 47.59 μmol/kg to 0.99 and 4.01 μmol/kg. In addition, compared to sea surface salinity (SSS) and sea surface chlorophyll-a (SCHL), sea surface temperature (SST) plays a more significant role in DO retrieval. Finally, compared to the pelagic interactions scheme for carbon and ecosystem studies (PISCES) model, the RF method achieves higher retrieval accuracies at depths above 700 dbar. In the deep ocean, the primary differences in DO values obtained from the RF method and the PISCES model-based method are noticeable in the vicinity of the equatorial region.

This research analyzes extreme precipitation events in the Huaihe River Basin in China, a densely populated region with a history of human settlements and agricultural activities. This study aims to explore the impact of extreme precipitation index changes and provide decision-making suggestions for flood early warning and agricultural development in the Huaihe River Basin. The study utilizes the NEX-GDDP-CMIP6 climate model dataset and daily value dataset (V3.0) from China’s national surface weather stations to investigate temporal and spatial changes in the extreme precipitation indices from 1960 to 2014 and future projections. At the same time, this study adopts the RclimDex model, Taylor diagram, and Sen+Mann–Kendall trend analysis research methods to analyze the data. The results reveal a slight increase in extreme precipitation indices from the northwest to southeast within the basin, except for the CDD, which shows a decreasing trend. Regarding the spatial variation, the future increase in extreme precipitation in the Huaihe River Basin shows a spatial variation characteristic that decreases from the northwest to southeast. These findings suggest that extreme precipitation events are intensifying in the region. Understanding these trends and their implications is vital for adaptation strategy planning and mitigating the risks associated with extreme precipitation events in the Huaihe River Basin.

Intra-urban surface water (IUSW) is an indispensable resource for urban living. Accurately acquiring and updating the distributions of IUSW resources is significant for human settlement environments and urban ecosystem services. High-resolution optical remote sensing data are used widely in the detailed monitoring of IUSW because of their characteristics of high resolution, large width, and high frequency. The lack of spectral information in high-resolution remote sensing data, however, has led to the IUSW misclassification problem, which is difficult to fully solve by relying only on spatial features. In addition, with an increasing abundance of water products, it is equally important to explore methods for using water products to further enhance the automatic acquisition of IUSW. In this study, we developed an automated urban surface-water area extraction method (AUSWAEM) to obtain accurate IUSW by fusing GaoFen-1 (GF-1) images, Landsat-8 Operational Land Imager (OLI) images, and GlobeLand30 products. First, we derived morphological large-area/small-area water indices to increase the salience of IUSW features. Then, we applied an adaptive segmentation model based on the GlobeLand30 product to obtain the initial results of IUSW. Finally, we constructed a decision-level fusion model based on expert knowledge to eliminate the problem of misclassification resulting from insufficient information from high-resolution remote sensing spectra and obtained the final IUSW results. We used a three-case study in China (i.e., Tianjin, Shanghai, and Guangzhou) to validate this method based on remotely sensed images, such as those from GF-1 and Landsat-8 OLI. We performed a comparative analysis of the results from the proposed method and the results from the normalized differential water index, with average kappa coefficients of 0.91 and 0.55, respectively, which indicated that the AUSWAEM improved the average kappa coefficient by 0.36 and obtained accurate spatial patterns of IUSW. Furthermore, the AUSWAEM displayed more stable and robust performance under different environmental conditions. Therefore, the AUSWAEM is a promising technique for extracting IUSW with more accurate and automated detection performance.