Climate change impacts on magnitude and frequency of urban floods under scenario and model uncertainties.

Many studies have confirmed that climate change leads to frequent urban flooding, which can lead to significant socioeconomic repercussions. However, most existing studies have not evaluated the impacts of climate change on urban flood from both event-scale and annual-scale dimensions. In addition, there are only few studies that simultaneously consider scenario and model uncertainties of climate change, and combine flood risk assessment and uncertainty analysis results to provide practical suggestions for urban drainage system management. This study uses the statistical downscaling method to calculate the design rainfall under ten rainfall return periods of four climate models and three climate change scenarios in 2040s, 2060s, and 2080s in various prefecture-level cities in China. The four climate models are HadGEM2- ES, MPI-ESM-MR, NorESM1-M and FGOALS-g2 models and the three climate change scenarios are constructed by different representative concentration pathways (RCP), namely RCP2.6, RCP4.5 and RCP8.5. On this basis, relying on the generated drainage systems using geographical information and other data, event-scale and annual-scale precipitation are combined to calculate the change ratio of annual flood volume expectation in prefecture-level cities in each future year compared with the current situation. Furthermore, the study evaluates scenario and model uncertainties of climate change, and then comprehensively integrates the flood risk and its uncertainties to provides suggestions for urban drainage system management.

Relational Part-Aware Learning for Complex Composite Object Detection in High-Resolution Remote Sensing Images.

In high-resolution remote sensing images (RSIs), complex composite object detection (e.g., coal-fired power plant detection and harbor detection) is challenging due to multiple discrete parts with variable layouts leading to complex weak inter-relationship and blurred boundaries, instead of a clearly defined single object. To address this issue, this article proposes an end-to-end framework, i.e., relational part-aware network (REPAN), to explore the semantic correlation and extract discriminative features among multiple parts. Specifically, we first design a part region proposal network (P-RPN) to locate discriminative yet subtle regions. With butterfly units (BFUs) embedded, feature-scale confusion problems stemming from aliasing effects can be largely alleviated. Second, a feature relation Transformer (FRT) plumbs the depths of the spatial relationships by part-and-global joint learning, exploring correlations between various parts to enhance significant part representation. Finally, a contextual detector (CD) classifies and detects parts and the whole composite object through multirelation-aware features, where part information guides to locate the whole object. We collect three remote sensing object detection datasets with four categories to evaluate our method. Consistently surpassing the performance of state-of-the-art methods, the results of extensive experiments underscore the effectiveness and superiority of our proposed method.