成都O3逐日污染潜势关键时段优选的GAM模型

利用成都市2016~2018年O₃逐时监测数据以及该时段同时次的地面气象观测资料,通过对O₃日变化特征的分析,确定了表征研究区O₃逐日污染潜势的四个关键时段,即全天时段（00：00~24：00）、日间时段（05：00~20：00）、O₃超标时段（11：00~19：00）以及O₃峰值时段（15：00~16：00）.基于广义可加模型（Generalized Additive Model,GAM）分别构建了O₃日最大8h滑动平均浓度（O_3-8h）与上述四个时段气象要素之间的函数关系,分析了时间尺度变化对O₃逐日污染潜势的影响.结果表明：GAM模型可以很好地表征O_3-8h与不同时段多气象要素之间的非线性关系.O₃超标时段气象要素对O₃逐日污染潜势具有最佳的指示意义,对应GAM模型的调整判定系数R²和方差解释率IRV分别为0.81和81.4%,模型模拟值与观测值的压轴回归决定系数R²为0.805.太阳辐射、相对湿度和气温是决定O₃逐日污染潜势最重要的气象要素,但三者在GAM模型中的重要性排序会因时间尺度的变化而有所差异.

China's industrial green total factor water efficiency under the constraints of environment and resource

The RAM-SFA-RAM three-stage industrial green total factor water efficiency measurement model was constructed, and the efficiency values of different levels were evaluated and compared. Moreover, the driving mechanism of industrial green total factor water efficiency was empirically analyzed based on Tobit model. The efficiency measurement results showed that environmental factors and statistical noise were important factors that should not be ignored in the calculation of efficiency values. In the period of 2000~2017, the efficiency value of the national sample continued to rise, but there was still a big gap between the perfect development status, and its development stage could be specifically divided into "stable increase period", "vibration period" and "speed increase period". In the regional efficiency analysis, the eastern efficiency had a relatively significant leading edge, while the efficiency difference between the central and northeastern regions was smaller, and the variation curve had obvious "interactive" phenomenon, while the overall efficiency was lower in the west. In the inspection of major economic belts, the efficiency of the Beijing-Tianjin-Hebei region was higher than that in the Pan-Pearl River Delta and Pan-Yangtze River Delta regions. Additionally, the driving mechanism analysis showed that water resources endowment, enterprise technology innovation, water consumption structure and environmental regulation based on the national sample dimension had positively promoted the industrial green total factor water efficiency, while the industrial structure and urbanization rate showed a negative inhibition effect. In the regional sample analysis, there were some differences in the effects of the various influencing factors between different regions, but the industrial structure adjustment and urbanization level should be taken seriously, and different regions needed to take targeted improvement measures based on their actual characteristics.