摘要
气温变化对经济、社会、生态文明等各个领域有着显著影响,成为国内外的研究热点和重点,在环境建设中起重要保障作用。然而实际获取的气温数据具有明显的相关性、多样性和复杂性,为不同地区气温空间建模带来困难。空间变系数回归模型中的地理加权回归模型(Geographical Weighted Regression,GWR)可以很好的解决这一问题。文章主要利用1961-2000年的年均气温数据,在考虑地形影响的条件下,建立全国气温空间GWR模型,并对模拟精度进行评价,说明各地形因子的影响程度。通过与OLS模型对比,显示出GWR的优越性。
Temperature change has a significant impact on economic, social, ecological civilization and other fields. It has become a research hotspot and focus at home and abroad, and plays an important role in environmental construction. However, the actual temperature data have obvious correlation, diversity and complexity, which brings difficulties to the spatial modeling of air temperature in different areas. The geographical weighted regression model (GWR) in the spatial variable coefficient regression model can solve this problem very well. In this paper, based on the annual temperature data from 1961 to 2000, the GWR model of national air temperature space is established under the condition of considering the influence of topography, the simulation accuracy is evaluated, and the influence degree of each topographic factor is explained. Compared with the OLS model, the superiority of GWR is shown.
出处
《科技创新与应用》
2019年第13期11-15,共5页
Technology Innovation and Application
关键词
气温
GWR模型
地形因子
模拟精度
air temperature
GWR model
terrain factor
simulation accuracy
