Albedo, atmospheric solar absorption and heating rate measurements with stacked UAVs

Abstract This paper reports unique measurements of albedo, atmospheric solar absorption, and heating rates in the visible (0.4 to 0.7 µm) and broadband (0.3 to 2.8 µm) spectral regions using vertically stacked multiple lightweight autonomous unmanned aerial vehicles (UAVs). The most significant finding of this study is that when absorbing aerosols and water vapour concentrations are measured accurately and accounted for in models, and when heating rates are measured directly with stacked aircraft, the simulated clear sky heating rates are consistent with the observed broadband heating rates within experimental errors (about 15%). We conclude that there is no need to invoke anomalous or excess absorption or unknown physics in clear skies. Aerosol–radiation–cloud measurements were made over the tropical Indian Ocean within the lowest 3 km of the atmosphere during the Maldives Autonomous UAV Campaign (MAC). The UAVs and ground‐based remote sensing instruments determined most of the parameters required for calculating the albedo and vertical distribution of solar fluxes. The paper provides a refined analytical procedure to reduce errors and biases due to the offset errors arising from mounting of the radiometers on the aircraft and due to the aircraft attitude. Measured fluxes have been compared with those derived from a Monte‐Carlo radiative transfer algorithm which can incorporate both gaseous and aerosol components. Under cloud‐free conditions the calculated and measured incoming fluxes agree within 2–10 W m −2 (<1%) depending upon the altitudes. Similarly, the measured and calculated reflected fluxes agreed within 2–5 W m −2 (<5%). The analysis focuses on a cloud‐free day when the air was polluted due to long‐range transport from India, and the mean aerosol optical depth (AOD) was 0.31 and mean single scattering albedo was 0.92. The UAV‐measured absorption AOD was 0.019 which agreed within 20% of the value of 0.024 reported by a ground‐based instrument. The observed and simulated solar absorption agreed within 5% above 1.0 km and aerosol absorption accounted for 30% to 50% of the absorption depending upon the altitude and solar zenith angle. Thus there was no need to invoke spurious or anomalous absorption, provided we accounted for aerosol black carbon. The diurnal mean absorption values for altitudes between 0.5 and 3.0 km above mean sea level were observed to be 41 ± 3 W m −2 (1.5 K/day) in the broadband region and 8 ± 2 W m −2 (0.3 K/day) in the visible region. The contribution of absorbing aerosols to the heating rate was an order of magnitude larger than the contribution of CO 2 and one‐third that of the water vapour. In the lowest 3 km of the tropical atmosphere, aerosols accounted for more than 80% of the atmospheric absorption in the visible region. Copyright © 2007 Royal Meteorological Society