The roles of scattering, absorption, and air mass on the diffuse-to-global correlations
The roles of scattering, absorption, and air mass on the diffuse-to-global correlations
复制标题
散射、吸收和空气质量对扩散到全局相关性的作用
DOI：
10.1016/0038-092x(91)90077-a
复制
发表时间：
1991
期刊：
影响因子：
--
通讯作者：
R. Perez
中科院分区：
文献类型：
--
作者：
M. Vázquez;V. Ruiz;R. Perez
日散射辐射与日总辐射之比Kd与日晴朗指数Kt的月线图表明，大Kt的点几乎都位于一条直线Kt + Kd = C（常数C< 1）上。这条直线在一年中的不同时期表现出平行位移。同样的线也出现在每小时K d与K t值的月度图中。对这种线性关系的一个可能的物理解释是，在低云量的日子里，近一半的直接太阳辐照度被云量阻止，向下散射，近一半向上散射。然而，本文中的云量包括浊度（霾）和实际云。结果表明，整个形状的Kd与Kt的关系可以解释由散射单独。由空气质量或吸收成分的变化引起的吸收变化可以解释季节变化和太阳高度的变化。一个明显的异常，在很高的K-T伴随着高K-D被证明是一个自然的结果，在部分多云的条件下，无阴影的太阳周期。这个相关区域也倾向于一条直线，但斜率为正：Kt-Kd = C。结果表明，对于空间或时间平均数据，这种“异常”拉伸趋于消失。对局部相关和区域相关进行了区分，前者表现出“异常”延伸，而后者则没有。作者建议使用Kd = HdHo与Kt的图，而不是K= HdHt与Kt的未来调查。
Monthly plots of the ratio K d of daily diffuse to daily total extraterrestrial irradiation versus the daily clearness index K t, have shown that points of large K t tend to be placed nearly on a straight line K t+ K d= C (constant C< 1). This straight line exhibits a parallel displacement for different periods of the year. The same line is found in monthly plots of hourly K d versus K t values. A possible physical interpretation for this linear relationship is that on days of low cloudiness level, nearly half of the direct solar irradiance stopped by cloudiness is scattered downward and nearly half upward. However, cloudiness in this paper includes both turbidity (haze) and actual clouds. It is shown that the entire shape of the K d versus K t relationship can be explained by scattering alone. Variations in absorption caused by air mass, or by changes in absorbing constituents, account for seasonal variability, and for variations as a function of solar altitude. An apparent anomaly at very high K t accompanied by high K d is shown to be a natural consequence of unshaded sun periods during partly cloudy conditions. This region of the correlation is also found to tend to a straight line but with a positive slope: K t-K d= C. It is shown that for space or time averaged data this “anomalous” stretch tends to disappear. A useful distinction is made between local correlations and regional correlations, the former display the “anomalous” extension but the latter do not. The authors recommend the use of the K d= H d H o versus K t graphs instead of the K= H d H t versus K t for future investigations.