张宸豪^1,2, 冯曦^1,2, 冯卫兵², 刘涛^1,2, 丁坤^1,2
1.海岸灾害及防护教育部重点实验室, 江苏 南京 210098;2.河海大学 港口海岸与近海工程学院, 江苏 南京 210098

摘要:
为研究全球变暖与极寒天气间的关系，对加拿大13个省代表性测站10年的观测数据进行时空变化趋势分析，采用经验正交函数（EOF）寻找海洋表面温度历史数据的变化规律。另外利用BP神经网络建立了年平均温度、日降水量与地球吸热、散热、海表面温度、当地纬度间的关系，预测未来25年气候的变化，并建立了“极寒天气”与气候变化的关系模型。研究表明：高纬度地区温度、降水量普遍较低，同经度地区的温度差异较小且降水量变化不大；加拿大地区温度呈周期性变化，符合北半球的季节变化特征；北大西洋的东部与其他海洋的温度是反相关的，西太平洋南北回归线附近的海洋表面温度升高；“极寒天气”出现频率与气候变化有一定关系，局地极寒现象与全球变暖的大趋势并不矛盾。本研究为人们认识和理解“全球变暖”提供了一个新的思路。
关键词:全球变暖极寒天气EOF法BP神经网络气候变化模型
DOI：10.11759/hykx20191122004
分类号:P717
基金项目:国家自然科学基金青年基金（51709091）；江苏省自然科学基金青年基金（BK20170874）；中央高校基金（2017B00514）

Climate model based on big-data analysis
ZHANG Chen-hao^1,2, FENG Xi^1,2, FENG Wei-bing², LIU Tao^1,2, DING Kun^1,2
1.Key Laboratory of Coastal Disaster and Protection of Ministry of Education(Hohai University), Nanjing 210098, China;2.College of Harbor, Coastal and Offshore Engineering, Hohai University, Nanjing 210098, China
Abstract:
To study the relationship between global warming and extreme cold weather, we performed spatial and temporal variation trend analyses on 10-year observation data from representative stations in 13 Canadian provinces. We then used the empirical orthogonal function (EOF) to determine the variation rule of historical ocean-surface-temperature data. In addition, we used a BP neural network to establish the relationship between the annual average temperature, daily precipitation, and the Earth's heat absorption, heat dissipation, sea surface temperature, and local latitude to predict climate change in the next 25 years and to establish a relationship model between “extremely cold weather” and climate change. The results show that the temperature and precipitation in high latitudes are generally low, the temperature difference with longitude is small, and the precipitation changes only slightly. The temperature in Canada changes periodically, which is consistent with seasonal changes in the northern hemisphere. The eastern part of the North Atlantic is inversely related to the temperature of the other oceans. The frequency of “extreme cold weather” has a certain relationship with climate change, and the local extreme cold phenomenon is not inconsistent with the general trend of global warming. This study provides a new way to understand “global warming”.
Key words:global warmingextreme cold weatherEOF methodBP neural networkclimate change





PDF全文下载地址:

点我下载PDF