Mangroves are important ecosystems and their distribution and dynamics can provide an understanding of the processes of ecological change. Meanwhile, mangroves protection is also an important element of the Maritime Silk Road (MSR) Cooperation Project. Large amounts of accessible satellite remote sensing data can provide timely and accurate information on the dynamics of mangroves, offering significant advantages in space, time, and characterization. In view of the capability of deep learning in processing massive data in recent years, we developed a new deep learning model—Capsules-Unet, which introduces the capsule concept into U-net to extract mangroves with high accuracy by learning the spatial relationship between objects in images. This model can significantly reduce the number of network parameters to improve the efficiency of data processing. This study uses Landsat data combined with Capsules-Unet to map the dynamics of mangrove changes over the 25 years (1990–2015) along the MSR. The results show that there was a loss in the mangrove area of 1,356,686 ha (about 21.5%) between 1990 and 2015, with anthropic activities such as agriculture, aquaculture, tourism, urban development, and over-development appearing to be the likely drivers of this decline. This information contributes to the understanding of ecological conditions, variability characteristics, and influencing factors along the MSR.

Mapping water bodies with a high accuracy is necessary for water resource assessment, and mapping them rapidly is necessary for flood monitoring. Poyang Lake is the largest freshwater lake in China, and its wetland is one of the most important in the world. Poyang Lake is affected by floods from the Yangtze River basin every year, and the fluctuation of the water area and water level directly or indirectly affects the ecological environment of Poyang Lake. Synthetic Aperture Radar (SAR) is particularly suitable for large-scale water body mapping, as SAR allows data acquisition regardless of illumination and weather conditions. The two-satellite Sentinel-1 constellation, providing C-Band SAR data, passes over the Poyang Lake about five times a month. With its high temporal-spatial resolution, the Sentinel-1 SAR data can be used to accurately monitor the water body. After acquiring all the Sentinel-1 (1A and 1B) SAR data, to ensure the consistency of data processing, we propose the use of a Python and SeNtinel Application Platform (SNAP)-based engine (SARProcMod) to process the data and construct a Poyang Lake Sentinel-1 SAR dataset with a 10 m resolution. To extract water body information from Sentinel-1 SAR data, we propose an automatic classification engine based on a modified U-Net convolutional neural network (WaterUNet), which classifies all data using artificial sample datasets with a high validation accuracy. The results show that the maximum and minimum water areas in our study area were 2714.08 km${}^2$ on 20 July 2020, and 634.44 km${}^2$ on 4 January 2020. Compared to the water level data from the Poyang gauging station, the water area was highly correlated with the water level, with the correlation coefficient being up to 0.92 and the $R^2$ from quadratic polynomial fitting up to 0.88; thus, the resulting relationship results can be used to estimate the water area or water level of Poyang Lake. According to the results, we can conclude that Sentinel-1 SAR and WaterUNet are very suitable for water body monitoring as well as emergency flood mapping.

Mangrove forests are distributed in intertidal regions that act as a “natural barrier” to the coast. They have enormous ecological, economic, and social value. However, the world’s mangrove forests are declining under immense pressure from anthropogenic and natural disturbances. Accurate information regarding mangrove forests is essential for their protection and restoration. The main objective of this study was to develop a method to improve the classification of mangrove forests using C-band quad-pol Synthetic Aperture Radar (SAR) data (Radarsat-2) and optical data (Landsat 8), and to analyze the spectral and backscattering signatures of mangrove forests. We used a support vector machine (SVM) classification method to classify the land use in Hainan Dongzhaigang National Nature Reserve (HDNNR). The results showed that the overall accuracy using only optical information was 83.5%. Classification accuracy was improved to a varying extent by the addition of different radar data. The highest overall accuracy was 95.0% based on a combination of SAR and optical data. The area of mangrove forest in the reserve was found to be 1981.7 ha, as determined from the group with the highest classification accuracy. Combining optical data with SAR data could improve the classification accuracy and be significant for mangrove forest conservation.

Classification of remote sensing images is a key technology for extracting the information on ground objects using remote sensing methods. Inspired by the great success of deep learning in artificial intelligence, researchers have proposed different algorithums based on deep learning to improve the performance of classification. At present, a deep learning model represented by the Convolutional Neural Networks (CNNs) can extract the abstract feature, but it loses the spatial context of the ground objects. In order To solve the problem of lack of spatial information in CNNs, the feature of the Capsule network takes the form of vectors that convey the location transformation information from part to whole. This study propose using a Capsules-Unet model, which incorporates capsules within the U-net architecture for classification of remote sensing images. The aim of this study is to train better models by encapsulating the multi-dimensional features of the objects in the form of capsules, and to reduce parameter space by improving the dynamic routing algorithm. Experiments are conducted on ISPRS Vaihingen and Potsdam datasets. Capsules-Unet slightly outperforms all other approaches with far fewer parameters, a reduction in parameters of over 81.8% compared with U-net and over 13.8% compared with Capsule network.

Spatio-temporal characteristics are the crucial conditions for Moon-based Earth observations. In this study, we established a Moon-based Earth observation geometric model by considering the intervisibility condition between a Moon-based platform and observed points on the Earth, which can analyze the spatio-temporal characteristics of the observations of Earth’s hemisphere. Furthermore, a formula for the spherical cap of the Earth visibility region on the Moon is analytically derived. The results show that: (1) the observed Earth spherical cap has a diurnal period and varies with the nadir point. (2) All the annual global observation durations in different years show two lines that almost coincide with the Arctic circle and the Antarctic circle. Regions between the two lines remain stable, but the observation duration of the South pole and North pole changes every 18.6 years. (3) With the increase of the line-of-sight minimum observation elevation angle, the area of an intervisible spherical cap on the lunar surface is obviously decreased, and this cap also varies with the distance between the barycenter of the Earth and the barycenter of the Moon. In general, this study reveals the effects of the elevation angle on the spatio-temporal characteristics and additionally determines the change of area where the Earth’s hemisphere can be observed on the lunar surface; this information can provide support for the accurate calculation of Moon-based Earth hemisphere observation times.