Global Simulation of Nitrate and Ammonium Aerosols and Their Radiative Effects and Comparison of Satellite-Based and Modeled Aerosol Indirect Forcing.

Abstract

Large uncertainties exist in estimates of aerosol direct and indirect radiative effects on climate. In this dissertation, we improve the representation of nitrate and ammonium in the global model and evaluate their radiative effects. We also improve understanding why satellite-based estimates of aerosol indirect forcing are consistently lower than global model estimates. The formation of nitrate and ammonium on five types of externally mixed pre-existing aerosols is taken into account in the global chemistry transport model. Compared with the treatment assuming the internal mixture, this work predicts lower burden, higher sources (sinks) and therefore shorter lifetimes of aerosol nitrate and ammonium. Nitrate and ammonium are found to exhibit two counteracting direct effects on pre-existing aerosols. The inclusion of ammonium and nitrate can boost scattering efficiency of sulfate and organic matter but lower scattering characteristics of sea salt due to their lower hygroscopicity than that of sea salt. The anthropogenic forcing of nitrate and ammonium at the top of atmosphere (TOA) are estimated to be -0.11 W m-2. Nitrate and ammonium also affect aerosol activation and the reflectivity of clouds. The first aerosol indirect forcing induced by anthropogenic nitrate and ammonium is estimated to be -0.1 W m-2 at TOA. By using an offline radiative transfer model combined with statistical slopes of a proxy of cloud properties (ln(Nd)) versus a proxy of aerosol properties (ln(AOD) or ln(AI)) based on either only PD values of aerosol and cloud properties or PD and PI values, we found that statistical slopes based on the temporal variation of PD and PI values of Nd and AOD would be steeper than those only based on the spatial variation of PD values because the spatial variation of PD values does not include magnitude of AOD and Nd as small as those in the PI simulation. As a result, the associated error in aerosol indirect forcing can be between a factor of 3 to more than a factor of 6 on a global average basis if one uses slopes of ln(Nd) versus ln(AOD) to estimate PI Nd or about ±25-35% if one uses ln(AI).