南京市极端温度与经济增长变化趋势分析及风险预测研究

发布时间：2018-05-10 15:34

本文选题：南京市 + 极端气温　；参考：《南京信息工程大学》2014年硕士论文

【摘要】：本文根据南京市1951-2011年逐日的温度观测资料,采用线性趋势分析、灰色关联理论、灰色马尔可夫预测模型,详细分析了南京市极端气温的变化特征、极端温度时间序列与南京市GDP的关联分析以及改变点探索,同时预测2012年的南京市高低温事件发生的风险。其中所涉及的气温要素包括：平均气温、平均最高气温、平均最低气温、极端最高气温、极端最低气温、极端最高气温频数、极端最低气温频数、高温日数、低温日数共9个要素。 (1)对南京市1951-2011年极端温度时间序列的趋势分析：首先根据南京市的逐日最高气温、逐日最低气温、平均气温资料构建极端气温指数,从而分析南京市61年来的极端温度变化趋势,可以得出南京市年平均气温、最高气温与最低气温均呈上升趋势,且在20世纪90年代后增加明显,而年平均最低气温增加幅度明显超过年平均最高气温。年平均最高气温以及年平均最低气温的季节变化趋势除夏季年平均最高气温呈下降趋势外,其余均为上升趋势。年极端最高气温呈平稳趋势,年极端最低气温呈显著地增温趋势。南京市极端最高气温频数以及高温日数减少的趋势并不明显,呈平稳的状态,而极端最低气温频数以及低温日数具有显著的减少趋势,说明南京的各种温度指数都以升温为主要趋势。 (2)对南京市1951-2011年极端温度时间序列与GDP的关联关系以及改变点探索：南京市极端气温变化与GDP增长率的关联度较大。其中最强的是高温日数与低温日数,关联度为0.7819和0.7798。也就是说南京市的高低温日数与GDP的关联度最大,表明当地的短期气候变化对经济发展具有较强的敏感性。反之,短期气候变化也会对经济发展造成影响。南京市GDP增长率与平均气温、极端气温和极端气温频数时间序列的改变点与变化趋势基本一致,也就是说南京的经济发展与气候变化之间存在着密切的联系。 (3)基于1951-2011年的高低温日数预测2012年的高低温风险：基于加权马尔可夫和灰色加权马尔可夫预测方法,通过对高温日数与低温日数的持续时间划分,构建高低温赋权指数,对南京市高低温事件发生的风险进行预测,得出南京市2012年低温和高温风险状态都为一般风险状态,产生高低温事件的风险较小。
[Abstract]:Based on the daily temperature observation data of Nanjing from 1951 to 2011, using linear trend analysis, grey correlation theory and grey Markov prediction model, the variation characteristics of extreme temperature in Nanjing are analyzed in detail. The correlation analysis between extreme temperature time series and GDP in Nanjing and the exploration of change point are also used to predict the risk of high and low temperature events in Nanjing in 2012. The temperature elements involved include: average temperature, mean maximum temperature, average minimum temperature, extreme maximum temperature, extreme minimum temperature, extreme maximum temperature frequency, extreme minimum temperature frequency, high temperature days, The number of low temperature days is 9 elements. (1) the trend analysis of the time series of extreme temperatures in Nanjing from 1951 to 2011: firstly, according to the daily maximum temperature, the daily minimum temperature and the average temperature data of Nanjing, the extreme temperature index was constructed. By analyzing the trend of extreme temperature variation in Nanjing over the past 61 years, it can be concluded that the annual average temperature, the highest temperature and the lowest temperature in Nanjing City are all on the rise, and they have increased obviously since the 1990s. However, the annual mean minimum temperature increased more than the annual mean maximum temperature. The seasonal variation trend of the annual mean maximum temperature and the annual mean minimum temperature except the annual average maximum temperature in summer showed a downward trend and the rest were all upward trends. The annual extreme maximum temperature showed a steady trend, and the annual extreme minimum temperature showed a significant warming trend. The decreasing trend of the frequency of extreme maximum temperature and the number of days of high temperature in Nanjing is not obvious, but the frequency of extreme minimum temperature and the number of days of low temperature have a significant decreasing trend. It shows that the temperature rise is the main trend of all kinds of temperature index in Nanjing. (2) the relationship between extreme temperature time series and GDP from 1951 to 2011 in Nanjing and its change point are explored. The correlation between extreme temperature change and GDP growth rate in Nanjing is great. Among them, the strongest is the high temperature days and the low temperature days, the correlation degree is 0.7819 and 0.7798. In other words, the correlation between high and low temperature days and GDP is the largest in Nanjing, which indicates that local short-term climate change is sensitive to economic development. Conversely, short-term climate change will also have an impact on economic development. The time series of GDP growth rate and mean temperature, extreme temperature and extreme temperature frequency in Nanjing are basically consistent with the change trend, that is to say, there is a close relationship between the economic development of Nanjing and climate change. Forecast of high and low temperature risk for 2012 based on the number of days of high and low temperature from 1951 to 2011: based on weighted Markov and grey weighted Markov forecasting methods, we construct a high and low temperature weighting index by dividing the duration of high temperature days and low temperature days. The risk of high and low temperature events in Nanjing is forecasted. It is concluded that the state of low temperature and high temperature in Nanjing in 2012 are both general risk states, and the risk of producing high and low temperature events is relatively small.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F127;P49

【参考文献】