Correlated Short Term Fluctuations in Aerosol Optical Thickness and Shortwave Radiative Quantities

Close examination of direct normal solar irradiance (DNSI) and downwelling diffuse irradiance (DDI) on cloud-free days at the Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) site reveals anticorrelated shortterm (several minute) fluctuations, especially prominent in the hours around local solar noon; i.e., DDI increases as DNSI decreases. These fluctuations are correlated or anticorrelated with other direct and derived radiometric quantities, including aerosol optical thickness (AOT) from multifilter rotating shadowband radiometer (MFRSR) and Ångström exponent, evaluated as -d ln (AOT)/d ln wavelength. These fluctuations are attributed to increased relative humidity associated with intermittent rising air parcels. Examples of correlations will be given together with analysis of relative magnitudes of fluctuations in the several quantities establishing the consistency of this interpretation. Motivation • Aerosols are the greatest contributor to uncertainty and variability in the shortwave radiation budget for cloudfree skies. At 60° solar zenith angle, a change in AOT at 550 nm of 0.01, the present limit of accuracy in measurement, corresponds to a change in DNSI of 14 W m (Halthore et al. 1997). • Shortwave radiative forcing of climate change by anthropogenic aerosols is considered to be a potentially substantial offset of greenhouse gas warming over the industrial period, but is highly uncertain (IPCC 1996; Schwartz and Andreae 1996). • Accurate characterization of aerosol scattering and absorption properties is required to constrain models of atmospheric absorption. An overestimate of 0.02 nm in AOT at 550 nm, compensated by a corresponding underestimate in direct beam absorption, leads to an underestimation of atmospheric absorption by 5% (Halthore et al. 1998). Observations The ARM complement of shortwave instrumentation provides a unique opportunity for characterization of aerosol influences on shortwave radiation. The following are the radiometric quantities used here and the instruments employed: • DNSI: normal incidence pyrheliometer (NIP). • Direct Beam Extinction: MFRSR. • Total and DDI: Precision Spectral Pyranometers (unshaded, shaded). Results are reported here for April 18, 1996, characterized by an absence of clouds throughout the day. Comments on the Observations Figure 1 shows time series of AOT and other radiometric quantities at the SGP site on April 18, 1996; local standard time = UT 6 h; solar noon ca. 1830 UT. AOT at a given wavelength, τλ, is from MFRSR. Path extinction is evaluated as -ln (V/V0), where V is measured voltage and V0 is voltage corresponding to top of atmosphere as inferred from Langley plots. Vertical extinction is path extinction divided by airmass. AOT is vertical extinction minus Rayleigh scattering optical thickness and ozone absorption optical thickness. The points at 1730-1800 show DNSI calculated with the Moderate Resolution Atmospheric Radiance and Transmittance Model (MODTRAN)-3. The smooth black curve denoted “Modeled DNSI” passing through these points