Diurnal Variability of Cloud Optical Thickness Derived from DSCOVR/EPIC Observations

When discussing the potential effects of cloud evolution in climate modulation, we can distinguish between inter- and intraday variability. Historically, the focus has been placed on studying the changes in large time scales while ignoring the intraday cycles of cloud properties, partially due to the limited availability of datasets to carry out those studies. In this regard, DSCOVR's vantage point at Lagrange L1 point overcomes the spatiotemporal limitations of polar orbiters and geostationary satellites, allowing us to characterize the daytime properties of clouds as well as their statistical distributions using a single sensor. In previous analyses, we characterized the diurnal cycles of cloud height and cloud fraction, and here we expand our study to investigate these types of cycles for cloud optical thickness. We observe a predictable pattern in cloud thickness for different latitudinal zones that reaches a maximum around noon, and show that the statistical distribution of cloud optical depths fits a lognormal distribution. Accumulating years of observation, we discuss the effect of this distribution on the non-linearity of the relationship between reflectances and optical depths, and discuss the accuracy of the mean as the descriptor of clouds optical depths.