Search

Tropical cyclones affecting Shenzhen city have shown a remarkable tendency to increase in both intensity and quantity, highlighting the urgency of accurate forecasts of storm surges and flooding for effective planning and mitigation. Utilizing satellite and field observations together with the advanced high-resolution baroclinic wave–current model (SCHISM), a comprehensive investigation aimed at storm surge and flooding in Shenzhen was conducted. Statistical work of historical tropical cyclones revealed that Shenzhen was most vulnerable to cyclones propagating from the southeast toward the northwest and passing Shenzhen down the Pearl River Estuary. Thus, a representative, i.e., super typhoon Hato (2017), was selected for further study. Validations of numerical results suggested satisfactory model performance in mapping the wave, tide, and surge processes. Remarkable differences in spatiotemporal distribution and intensity of storm surge and flooding were found along the Shenzhen coast, which was dominated by the propagation of far-field surge and tidal waves, cooperation between wind direction and coastline orientation, estuary morphology, and the land terrain. Intervention of wave–current interaction improved the simulation of the surge and flooding and triggered an earlier occurrence time of the maximum surge in specific areas. The Pearl River discharge significantly elevated the sea level height inside the estuary and contributed to a more severe surge. Given the extremely complicated river networks and huge freshwater flux of Pearl River and the increasing trend of concurrent heavy precipitation of tropical cyclones, future investigations on compound flooding were suggested.

Malware detection methods based on gray images and deep learning have the characteristics of high detection accuracy and no need of feature engineering. Unfortunately, adversarial examples (AEs) can deceive such detection methods. However, it is difficult to reduce the detection accuracy of this kind of detection method greatly without destroying the functional integrity of the original file. By analyzing the structure and loading mechanism of portable executable (PE) files, this paper proposes an unrestricted add-amount bytecode attack (BAUAA). BAUAA generates adversarial samples by adding bytecode to a “section additional space” in the PE file that is scattered after each section and is not loaded into memory, and because of the unlimited amount of this space that can be added, the generated adversarial samples can be transformed into grayscale images that vary in size and texture, which can affect the discrimination accuracy of gray images and deep learning-based malware detection methods. The experimental results show that the detection accuracy of the malware detection method based on gray images and deep learning for the AEs generated by BAUAA is significantly lower than that for the non-AEs. To avoid the abuse of BAUAA in reality, it proposes a targeted AE detection method.