Coupling sky images with radiative transfer models: a new method to estimate cloud optical depth

Abstract

Abstract. A method for retrieving cloud optical depth (τ c ) using a UCSD developed ground-based sky imager (USI) is presented. The radiance red-blue ratio (RRBR) method is motivated from the analysis of simulated images of various τ c produced by a radiative transfer model (RTM). From these images the basic parameters affecting the radiance and red-blue ratio (RBR) of a pixel are identified as the solar zenith angle (θ 0 ), τ c , solar pixel angle/scattering angle (ϑ s ), and pixel zenith angle/view angle (ϑ z ). The effects of these parameters are described and the functions for radiance, I λ (τ c , θ 0 , ϑ s , ϑ z ), and RBR(τ c , θ 0 , ϑ s , ϑ z ) are retrieved from the RTM results. RBR, which is commonly used for cloud detection in sky images, provides non-unique solutions for τ c , where RBR increases with τ c up to about τ c = 1 (depending on other parameters) and then decreases. Therefore, the RRBR algorithm uses the measured I meas λ (ϑ s , ϑ z ), in addition to RBR meas (ϑ s , ϑ z ), to obtain a unique solution for τ c . The RRBR method is applied to images of liquid water clouds taken by a USI at the Oklahoma Atmospheric Radiation Measurement (ARM) program site over the course of 220 days and compared against measurements from a microwave radiometer (MWR) and output from the Min et al. (2003)

High‐resolution photography of clouds from the surface: Retrieval of optical depth of thin clouds down to centimeter scales

“…Mejia et al . [] presented images of radiance from which they retrieved images of COD using a RT model, which likewise exhibited strong variation in COD on scales of 15 m (for assumed cloud base height of 1 km). Fine structure in cloud radiance, down to the limit of the resolution, estimated as about 1 m, was seen also in the film‐photographic images of Sachs et al [].…”

“…A fundamental limitation of the approach is the inherent two dimensionality of the measurement, with no characterization of the vertical dimension. Another limitation for quantitative interpretation is that for clouds of optical depth greater than about 3, the inversion of radiance to optical depth becomes much more problematic than for optically thin clouds [ Mejia et al ., ]. Similar considerations obtain for situations with clouds at multiple levels.…”

“…Mejia et al . [] used a three‐color electronic camera with a fish‐eye lens [ Urquhart et al ., ] to obtain sky images from which radiance was inverted to yield cloud optical depth images.…”

“…Ultimately, because of the maximum in the dependence of zenith radiance on COD, the same zenith radiance can result from either a thinner cloud (ascending branch of the curves in Figure ) or a thicker cloud (descending branch of the curves) [ Chiu et al ., ]. Consequently, it is necessary to ascertain whether the radiance measured in a given region of the image is from a cloud of optical depth less than or greater than the COD corresponding to the maximum normalized radiance, the so‐called ambiguity problem in retrieval of COD from zenith radiance [ Chiu et al ., ; Schäfer et al ., ; Niple et al ., ; Mejia et al ., ]. As COD obtained by the inversion is sensitive to zenith radiance only for clouds having COD ≲ 3, above which the COD can increase substantially with little increase in downwelling radiance, we restrict the retrievals for the present to clouds of optical depth less than about 3, depending on zenith angle, indicated by the circular markers.…”

Best Practices Handbook for the Collection and Use of Solar Resource Data for Solar Energy Applications: Second Edition

“…The red-to-blue ratio (RBR; Figure 7-2, middle right) is often used as a main indicator for clouds because of their different spectral-scattering properties (high RBR) and clear-sky (low RBR) conditions (Shields, Johnson, and Koehler 1993;Long and DeLuisi 1998). Using radiative transfer modeling, Mejia et al (2016) demonstrated some ambiguity in the RBR method and proposed the radiance RBR method, which leverages intensities and RBR to identify cloudy pixels. Pixel intensities (Figure 7-2, middle left) are also related to cloud cover and might be exploited as an additional feature for cloud detection.…”