基于GIS的空气质量指数空间插值方法研究

发布时间：2018-06-19 01:15

本文选题：AQI + 空间自相关　；参考：《昆明理工大学》2015年硕士论文

【摘要】：伴随着我国近年来频现的雾霾天气,空气质量愈来愈多得到公众的关注。2012年出台的空气质量新标准,用空气质量指数(AQI)替代了原有的空气污染指数(API)对空气质量状况进行了定量描述。将与灰霾的形成密切相关的PM2.5,以及反映机动车尾气造成的光化学污染的臭氧指标,均纳入到AQI的评价体系中。江苏是中国的人口大省,其社会经济发展现状及城市规划的需要,都对江苏省的空气质量提出了更高要求。因此,探索江苏省空气质量指数的时空分布特点,建立全省范围内的空气质量指数预测模型,有着重要的现实意义。本文根据2013年1月-2014年2月江苏省的日均AQI数据,对全省的空气质量进行分析。首先,以省会南京为例分析AQI在不同季节的变化特点、工作日与周末的差别,并运用定性分析与定量分析相结合的方法考虑了气温、降水量因素对AQI的影响；其次,对AQI的空间自相关性进行了探索,了解全省AQI空间聚集特征,并通过直方图、正态QQ分布图、趋势分析的方法,对2013年的AQI数据进行探索分析；最后,比较多种空间插值方法,根据预测精度选择最合适的模型来分析全省的AQI空间分布规律,并对AQI的达标率进行预测。主要得到以下结论：2013年夏季,是全年中江苏省空气质量最好的时段,南京市的情况也不例外。南京市周末的AQI远高于工作日,存在着周末效应。并且,南京市的气温与AQI、降水量与AQI之间均具有线性相关性。江苏省各监测站点间的AQI具有很强的空间自相关特性。苏州、泰州、南通三城市,在2013年7月、8月连续呈现出“高-高”空间集聚的情况。江苏省AQI整体趋势为由西向东先逐渐降低然后略有上升,南北方向上则比较稳定。对各种插值模型的精度评价显示,全局多项式的RMS最小,克里金法生成的表面可以更清楚地描述出局部细节。全省AQI的分布特征是,沿着海岸线方向由内陆向沿海地区逐渐降低,最高值在在徐州地区。创建出AQI超出临界值100的概率图,其最大特点就是能轻松识别出AQI超标的区域,使公众对空气状况有更加直观的感受,也为政府部门制定空气质量预测、预警提供有效的参考。
[Abstract]:With the frequent haze weather in China in recent years, the air quality is getting more and more public attention. The new air quality standard was introduced in 2012. Air quality index (AQI) was used instead of the original air pollution index (API) to describe the air quality quantitatively. PM2.5 which is closely related to haze formation and ozone index which reflects photochemical pollution caused by vehicle exhaust are all included in AQI evaluation system. Jiangsu is one of the most populous provinces in China. Its social and economic development and the needs of urban planning have put forward higher requirements for air quality in Jiangsu Province. Therefore, it is of great practical significance to explore the spatial and temporal distribution of air quality index in Jiangsu Province and to establish a prediction model of air quality index in Jiangsu province. Based on the daily AQI data of Jiangsu Province from January 2013 to February 2014, the air quality of Jiangsu Province is analyzed. First of all, taking Nanjing as an example to analyze the variation characteristics of AQI in different seasons, the difference between weekdays and weekends, and to consider the effects of temperature and precipitation factors on AQI by combining qualitative analysis with quantitative analysis. This paper explores the spatial autocorrelation of AQI, understands the spatial aggregation characteristics of AQI in the province, and explores and analyzes the AQI data in 2013 through histogram, normal QQ distribution map and trend analysis method. According to the prediction precision, the most suitable model is chosen to analyze the spatial distribution law of AQI in the province, and the rate of reaching AQI is forecasted. The main conclusions are as follows: summer 2013 is the best time for air quality in Jiangsu Province and Nanjing is no exception. The AQI of Nanjing weekend is much higher than that of working day, and there is weekend effect. Moreover, there is a linear correlation between air temperature and AQI, precipitation and AQI in Nanjing. AQI between monitoring stations in Jiangsu Province has strong spatial autocorrelation characteristics. Suzhou, Taizhou, Nantong three cities, in July and August 2013, a continuous "high-high" spatial agglomeration. The overall trend of AQI in Jiangsu Province is that the trend of AQI decreases gradually from west to east, then increases slightly, and is stable in north and south direction. The accuracy evaluation of various interpolation models shows that the RMS of the global polynomial is the smallest and the surface generated by the Crekin method can describe the local details more clearly. The distribution characteristic of AQI in the whole province is that the distribution of AQI decreases gradually from inland to coastal area along the coastline direction, and the highest value is in Xuzhou area. A probability map of AQI exceeding the critical value of 100 is created, the biggest characteristic of which is that it can easily identify the area where AQI exceeds the standard, which makes the public feel more intuitively about the air condition, and also provides an effective reference for government departments to make air quality prediction and early warning.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X823

【参考文献】