Collow, Allison B.. An analysis of the radiation budget in two tropical continental atmospheric columns. Retrieved from https://doi.org/doi:10.7282/T3X92D43
DescriptionMany uncertainties remain in the relationship between clouds, aerosols, and radiation, especially in locations along tropical margins. Two regions of particular interest because of their perceived susceptibility to climate change are the Sahel region of West Africa and the Amazon Rainforest of Brazil. Both the Sahel and the Amazon Rainforest have two distinct seasons with varying cloudiness, but the impacts of this cloudiness upon the regional radiation budgets are only vaguely understood. With this in mind, the Atmospheric Radiation Measurement Program’s Mobile Facility #1 collected a full year of surface based observations in Niamey, Niger in the central Sahel and Manacapuru, Brazil in the Amazon Rainforest. These surface observations were complemented by top of the atmosphere observations from the Geostationary Earth Radiation Budget instrument and Clouds and the Earth's Radiant Energy System. This dissertation presents observations from these deployments with focus on the diurnal cycle of meteorology and the accompanying radiation budget at the surface, top of the atmosphere, and within the atmospheric column itself in the Sahel and upon the seasonal cycle in the Amazon. Cloud radiative effects (CREs) are examined in detail. Although both regions are located in the tropics and share some common features, their climates and radiation budgets are very different. Among the conclusions of this work are that clouds reflect incoming solar insolation and prevent the loss of longwave radiation to space in both locations, though the large loading of water vapor and abundance of clouds in the Amazon create a more substantial impact at the boundaries of the atmospheric column. It is demonstrated that CREs are dominated by the shortwave spectrum in the Sahel and by the longwave spectrum in the Amazon Rainforest. It is further demonstrated that CREs are themselves a diurnally varying signal that cannot be appropriately quantified by instantaneous measurements from polar orbiting satellites, but that can be quantified by geostationary satellites equipped with broadband radiometers. Sensitivity to the averaging period employed in the analysis of CREs is also demonstrated, further reinforcing the notion that instantaneous measurements of CREs are likely to produce significant misrepresentations of these effects.