Search

Abstract We examine the spectral effects of tropospheric absorbing aerosol types on backscattered UV radiation from 300 to 400 nm. First, using satellite observations, we show a non‐linear spectral dependence below 340 nm for scenes with significant aerosol absorption. Lambert‐equivalent reflectivity also follows a non‐linear behavior and may become negative, therefore non‐physical, at shorter wavelengths. We then present radiative transfer calculations of UV radiances accounting for ozone and aerosol attenuation. The effects are quantified as spectral residuals, calculated by comparing radiances simulated with aerosols to those with a molecular atmosphere assuming spectrally independent Lambert‐equivalent surface reflectivity. Simulations are carried out for carbonaceous smoke and mineral dust aerosols using the present‐day optical models applied successfully in the Ozone Monitoring Instrument aerosol algorithm, OMAERUV. The spectral residual has a complex dependence on wavelength, aerosol optical and physical properties, and surface conditions. In all simulated cases, the spectral residual has a maximum around 320 nm and then diminishes rapidly at shorter wavelengths due to increased photon absorption by ozone in the stratosphere. However, we find the relationship between the spectral residual and that at 340 nm nearly linear for a wide range of these conditions. The relationship depends somewhat on the total column ozone. The theoretical results presented offer a means for correcting the effects of UV‐absorbing aerosols in trace gas retrievals without a priori knowledge of aerosol and surface properties. Since our technique of estimating spectral effects of aerosols applies to radiance residuals, and not a specific retrieval algorithm, it may be used broadly to correct the effects in various retrieval algorithms.

Black-sky cloud albedo (BCA) is derived from satellite UV 340 nm observations from NOAA and NASA satellites to infer long-term (1980–2018) shortwave cloud albedo variations induced by volcano eruptions, the El Niño–Southern Oscillation, and decadal warming. While the UV cloud albedo has shown no long-term trend since 1980, there are statistically significant reductions over the North Atlantic and over the marine stratocumulus decks off the coast of California; increases in cloud albedo can be seen over Southeast Asia and over cloud decks off the coast of South America. The derived BCA assumes a C-1 water cloud model with varying cloud optical depths and a Cox–Munk surface BRDF over the ocean, using radiances calibrated over the East Antarctic Plateau and Greenland ice sheets during summer.