Sea level change is a key indicator of climate change, and the prediction of sea level rise is one of most important scientific issues. In this paper, the gridded sea level anomaly (SLA) data from satellite altimetry are used to analyze the sea level variations around Shandong Peninsula from 1993 to 2016. Based on the Complete Ensemble Empirical Mode Decomposition (CEEMD) method and Radial Basis Function (RBF) network, the paper proposes an improved sea level multi-scale prediction approach, namely, CEEMD-RBF combined model. Firstly, the multi-scale frequency oscillatory modes (intrinsic mode functions (IMFs)) representing different oceanic processes are extracted by CEEMD from the highest frequency to the lowest frequency oscillating mode. Secondly, RBF network is used to establish prediction models for various IMF components to predict their future trends, and each IMF is used as an input factor of the RBF network separately. Finally, the prediction results of each IMF component with RBF network are reconstructed to obtain the final predictions of sea level anomalies. The results shows that CEEMD is particularly suitable for analyzing nonlinear and non-stationary time series and RBF network is applicable for regional sea level prediction at different scales.

As an important part of urban ecosystems, urban green spaces play a key role in ecological environmental protection and urban spatial structure optimization. However, due to the complex morphology and high degree of fragmentation of urban green spaces, it is still challenging to effectively distinguish urban green space types from high spatial resolution images. To solve the problem, a Contextual Transformer and Squeeze Aggregated Excitation Enhanced DeepLabV3+ (CTSA-DeepLabV3+) model was proposed for urban green space classification based on Gaofen-2 (GF-2) satellite images. A Contextual Transformer (CoT) module was added to the decoder part of the model to enhance the global context modeling capability, and the SENetv2 attention mechanism was employed to improve its key feature capture ability. The experimental results showed that the overall classification accuracy of the CTSA-DeepLabV3+ model is 96.21%, and the average intersection ratio, precision, recall, and F1-score reach 89.22%, 92.56%, 90.12%, and 91.23%, respectively, which is better than DeepLabV3+, Fully Convolutional Networks (FCNs), U-Net (UNet), the Pyramid Scene Parseing Network (PSPNet), UperNet-Swin Transformer, and other mainstream models. The model exhibits higher accuracy and provides efficient references for the intelligent interpretation of urban green space with high-resolution remote sensing images.

Zhao, J., 2018. Design and implementation of geo-spatial information technology in marine ecological environment monitoring. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 50–56. Coconut Creek (Florida), ISSN 0749-0208. With the traditional ocean monitoring technology, marine monitoring task can be accomplished, but there is the shortcoming of long reaction time for detecting pollution information. To address this problem, geo-spatial information technology in marine ecological environment monitoring is designed in this paper. SQL Server database and Arc SDE database to accelerate the data management. The three-layer architecture system of data layer, business logic layer and performance layer is simplified to complete system software acceleration design. Map module is designed by using GIS geographic information technology and Pan method of Map Control. Rapid monitoring of geo-spatial information based on the geometric relationship of the elements is achieved. The multi-dimensional analysis of Spatial Analyst is used to realize the design of geo-spatial information technology in marine ecological environment monitoring. Experimental results show that the average deviation of the monitoring technology is 82.79% higher than the traditional monitoring technology, and the average time of detecting the pollution information is increased by 78.82%.