香河地区冬季NO2光解速率的变化特征及其估算方法的建立

NO₂的光解速率j(NO₂)是对流层化学研究中的一个重要的参数,它是对流层臭氧（O₃）和OH自由基的形成关键影响因子之一。本文利用香河站2017年11月至2018年1月的观测资料对NO₂的光解速率j(NO₂)、太阳辐射的变化特征,及光解速率与分波段辐射相关性进行分析,在此基础上建立了一套适合于构建香河地区j(NO₂)的参数化方案。结果表明,香河地区的j(NO₂)与太阳辐射的变化趋势基本一致,日变化均呈正午高、早晚低的单峰型。香河地区冬季j(NO₂)的变化范围是0.00046～0.0044 s-1,平均值为0.0029 s-1。j(NO₂)与晴空指数（K_S）和定义的紫外辐射晴空指数（K_UV）存在很好的相关性,因此,利用j(NO₂)与大气质量数和晴空指数的相互依赖关系,基于K_UV构建了适合于香河地区j(NO₂) 的估算方程。由于总辐射是气象站的常规观测要素,晴空指数K_S更容易获取,继而建立了通过K_S参数化计算j(NO₂)的估算方程。利用K_UV和K_S建立的j(NO₂)估算方程的瞬时估算值与观测值的平均相对误差分别为6.5%和13.9%,均方根误差分别为0.00029和0.00051。

Variation Characteristics and Estimation Method of NO2 Photolysis Rate in Xianghe during Winter

The NO₂ photolysis frequency j(NO₂) is an important parameter in tropospheric chemistry. It is one of the key factors affecting the formation of ozone and OH radicals in the troposphere. The variations of j(NO₂) and solar radiation and the correlation between photolysis rate and sub-band solar radiation were analyzed using the observational data in Xianghe from November 2017 to January 2018. On this basis, a parameterized scheme suitable for reconstructing j(NO₂) in the Xianghe area was constructed. The results indicate that the variation trends of j(NO₂) and solar radiation in Xianghe are basically the same, and the diurnal variations are all unimodal with high noon and small morning and evening. The variation range of j(NO₂) in winter is 0.00046-0.0044 s^-1, with a mean of 0.0029 s^-1. Since j(NO₂) correlates well with the clear sky index (K_S) and transmittance of ultraviolet radiation (K_UV), an experimental estimating equation based on K_UV was established using the correlation between j(NO₂), air mass number, and clear air index. Considering the clear sky index K_S is more easily obtained and global radiation is a conventional observation of meteorological stations, an experimental estimating equation was established based on K_S. The instantaneous value of j(NO₂) can be estimated by the equations established using K_UV and Ks, and the mean absolute percentage errors between the instantaneous values obtained by the two equations and the observed value are 6.5% and 13.9%, and the root mean square errors are 0.00029 and 0.00051, respectively.