基于数据同化与CA模型的包头市热岛模拟预测研究

发布时间：2018-08-30 12:26

【摘要】：以内蒙古第一大工业城市包头市市区作为研究区,以1996年、2001年、2006年和2011年Landsat5的遥感影像以及2016年的Landsat8的遥感影像作为基础研究数据,用IB算法对地表温度进行反演,引入三种常用的数据同化算法对包头市市区进行热岛模拟预测并与传统的CA-Markov模型进行对比,选择最适算法对包头市市区2020年的热岛分布进行模拟预测。结果表明:引入数据同化算法能够提高CA模型模拟精度,En SRF-CA模型在包头市热岛模拟上要优于其他模型,精度能达到88.35%。根据预测得到的热岛模拟影像,到了2020年,包头市市区整体升温效果依旧明显,强绿岛区大幅度向绿岛区转移,强热岛区面积增加了44.21km~2,黄河流域水流量的减少降低了降温效果,应加强对南海湿地生态系统的保护力度以及适当增加城市绿地以减弱热岛效应。
[Abstract]:Taking Baotou City, the largest industrial city of Inner Mongolia, as the research area, taking the remote sensing images of Landsat5 from 1996, 2001, 2006 and 2011 and the remote sensing images of Landsat8 in 2016 as the basic research data, the surface temperature is inversed by IB algorithm. This paper introduces three commonly used data assimilation algorithms to simulate and predict the heat island distribution in Baotou urban area and compares it with the traditional CA-Markov model, and selects the optimal algorithm to simulate and predict the heat island distribution in Baotou urban area in 2020. The results show that the data assimilation algorithm can improve the simulation accuracy of the CA model and the en SRF-CA model is superior to other models in the simulation of Baotou heat island, and the accuracy can reach 88.35. According to the predicted heat island simulation image, by 2020, the overall warming effect in Baotou urban area is still obvious. The Qianggreen Island area has been transferred to the Green Island area by a large margin, the area of the strong Heat Island area has increased by 44.21 km2, and the decrease of water flow in the Yellow River Basin has reduced the cooling effect. The protection of wetland ecosystem in the South China Sea should be strengthened and the urban green space should be increased to reduce the heat island effect.
【作者单位】：北京林业大学精准林业北京市重点实验室;佛罗里达大学地理系;
【基金】：国家自然科学基金项目(41371189) “十二五”国家科技支撑计划项目(2012BAD16B00)
【分类号】：X16

【相似文献】