A greater frequency of sea fog/low clouds may be observed in the Arctic under the effects of global warming, which strongly affects the safety of navigation in the Arctic. Using moderate-resolution imaging spectroradiometer (MODIS) infrared multiband remote sensing data, in this paper, we construct a detection algorithm for sea fog and low clouds in the daytime during the summer period in the Arctic through multiband infrared remote sensing. Physically, the algorithm relies on the spectral variation in the Arctic Sea route region with different solar zenith angles (SZAs). Specifically, for days with strong sunshine (SZA < 70°), the effect of the sea surface is removed by coupling the B7 (2.130 μm) near-infrared band reflectance (R2.130 μm) and the radiation difference (BTD3.750 μm–11.030 μm) between the B20 (3.750 μm) mid-infrared band (BT3.750 μm) and B31 (11.030 μm) thermal infrared band (BT11.030 μm). Then, the sea fog and low-cloud ratio index (SFLCRI) is constructed on the basis of two near-infrared bands, namely, the B5 (1.240 μm) water vapor absorption band and the B9 (0.936 μm) atmosphere window, to remove the middle and high clouds. Finally, the sea fog and low clouds are extracted according to the radiation threshold of BT11.030 μm. At dawn and dusk (SZA > 70°), the SFLCRI is utilized to remove the sea surface and middle and high clouds, and then the undetected middle and high clouds are removed via BT11.030 μm. The validation results reveal that the accuracy of the algorithm is high, and the probability of detection, false alarm ratio, and the critical success index of sea fog and low-cloud detection are 86 %, 13 %, and 76 %, respectively, as a whole. The algorithm accuracy was less affected by time. The research results are of great theoretical and practical importance for Arctic sea fog and low-cloud detection, and for ensuring the safety of Arctic shipping.

中文翻译：

---

通过多源红外遥感对夏季白天北极海雾和低云进行卫星探测


在全球变暖的影响下，北极可能会观察到更频繁的海雾/低云，这严重影响了北极的航行安全。本文利用中分辨率成像光谱仪 （MODIS） 红外多波段遥感数据，通过多波段红外遥感构建了北极夏季白天海雾和低云的探测算法。从物理上讲，该算法依赖于具有不同太阳天顶角 （SZA） 的北冰洋路线区域的光谱变化。具体来说，对于日照强烈的日子 （SZA < 70°），通过耦合 B7 （2.130 μm） 近红外波段反射率 （R2.130 μm） 和 B20 （3.750 μm） 中红外波段 （BT3.750 μm） 和 B31 （11.030 μm） 热红外波段 （BT11.030 μm） 之间的辐射差 （BTD3.750 μm–11.030 μm） 来消除海面的影响。然后，在 B5 （1.240 μm） 水汽吸收波段和 B9 （0.936 μm） 大气窗两个近红外波段的基础上构建海雾低云比指数 （SFLCRI），以去除中高云。最后，根据 BT11.030 μm 的辐射阈值提取海雾和低云。在黎明和黄昏 （SZA > 70°），利用 SFLCRI 去除海面和中高云，然后通过 BT11.030 μm 去除未检测到的中高云。验证结果表明，该算法的准确率较高，海雾和低云检测的检出概率、误报率和临界成功指数总体上分别为 86 %、13 % 和 76 %。算法准确性受时间的影响较小。 研究成果对北极海雾和低云探测，对保障北极航运安全具有重要的理论和实践意义。