The radiation reconstruction model (Version 2) and its application for estimating variability of ultraviolet radiation over the 1968–2024 period in Moscow

Solar radiation is a vital component of the climate system, and knowledge of its temporal variation is of great importance. Ultraviolet (UV) radiation, due to its high photon energy, has a strong impact on the biosphere and human health. In this work, a reconstruction model (version 2) for estimating surface solar radiation long-term variability was proposed. It has been configured to assess the variability of erythemal UV radiation and UV-A radiation within the 315–400 nm spectral range. In this version of the model, physical parameterizations of UV radiation with respect to various factors have been refined, and an additional factor, which accounts for nitrogen dioxide influence on UV radiation, has been added. The model was validated against long-term UV measurements at the Moscow State University Meteorological Observatory, which has the longest continuous record of UV observations in the world. The mean RMSE of the differences between model simulations and measurements is on the order of 3% for annual values, which is comparable to the measurement uncertainty. High correlation coefficients (about 0.8) were observed between the modeled and measured values of the UV temporal variability for almost all periods. The analysis of surface UV radiation variability for the 1968–2024 period revealed noticeable UV fluctuations, which were characterized by a minimum in the late 1970s and early 1980s and a maximum after 2010. For erythemal UV radiation, the maximum amplitude of annual changes is about 30%, and for UV-A radiation it is about 19%. Much stronger changes were observed for erythemal UV radiation compared to UV-A radiation owing to the additional influence of ozone variability.