Abstract

The technique of studying seasonal variations in cloud parameters over regions of Western Siberia according to satellite data is presented. Five natural zones have been identified: tundra, forest tundra, swamps, taiga, and forest steppe. A combined “summer” and “winter” classification of cloudiness is introduced; the classification includes 16 and 12 cloud types, respectively. Cloudiness images are classified using an algorithm based on neural network technologies and fuzzy logic methods. Results of analysis of seasonal variations in some parameters of different types of cloudiness and their repeatability over the considered regions of western Siberia based on MODIS satellite data for 2017 are discussed. The dependences of the seasonal variability of cloudiness parameters agree well with known literature data on ground-based observations, which corroborates the effectiveness of the proposed technique.

中文翻译：

---

根据MODIS卫星数据，西伯利亚自然区不同云类型的参数

摘要

提出了根据卫星数据研究西西伯利亚地区云参数季节性变化的技术。已确定了五个自然区：冻原，森林冻原，沼泽，针叶林和森林草原。引入了多云的“夏季”和“冬季”组合分类；该分类分别包括16种和12种云类型。使用基于神经网络技术和模糊逻辑方法的算法对浊度图像进行分类。讨论了基于MODIS 2017年卫星数据的西伯利亚西部地区不同类型云量的某些参数的季节性变化及其可重复性的分析结果。浊度参数季节性变化的依赖性与地面观测的已知文献数据非常吻合，