Application of the statistical optimization method to the inversion algorithm for analyzing aerosol optical properties from sun and sky radiance measurements

In this paper, we describe a new algorithm for analyzing aerosol optical properties from sun and sky radiance measurements. The Skyrad package (SKYRAD.pack) is one of the well-known software which analyzes the sky radiometer data. The SKYRAD.pack has been improved continuously since the first version appeared in 1996, and the programming procedures have become increasingly more complicated. Therefore, we rewrote the software by applying the statistical optimization method to its inversion algorithm to make it easier to improve, and extend the analytic capabilities in the future. In this method, the normal distribution was assumed as a probability density function of measurement error, and the maximum likelihood method was applied. Our new software can retrieve the columnar volume size distribution, and the complex refractive index as well as the SKYRAD.pack. We tested it using the simulated data from aerosol models, and it was able to retrieve the size distributions more accurately and more stably than the SKYRAD.pack. We also confirmed that the new program could analyze the real observed data more stably than the SKYRAD.pack with little difference between the results of the two programs.