Fu, QiangBalmes, Kelly Ann2021-08-262021-08-262021-08-262021Balmes_washington_0250E_22717.pdfhttp://hdl.handle.net/1773/47327Thesis (Ph.D.)--University of Washington, 2021The clear-sky aerosol direct radiative effect was estimated at the Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) site in Lamont, Oklahoma and Tropical Western Pacific (TWP) site in Darwin, Australia. The NASA Langley Fu-Liou radiative transfer (RT) model was used with observed inputs including aerosol vertical extinction profile from the Raman lidar (RL); spectral aerosol optical depth (AOD), single-scattering albedo and asymmetry factor from Aerosol Robotic Network (AERONET); temperature and water vapor profiles from radiosondes; and surface shortwave spectral albedo from radiometers. A radiative closure experiment was conducted for clear-sky conditions. The mean differences of modeled and observed surface downwelling shortwave total fluxes were 1 W m−2 at SGP and 2 W m−2 at TWP, which are within observational uncertainty. At SGP, the estimated annual mean clear-sky aerosol DRE is -3.00±0.58 W m−2 at the top of atmosphere (TOA) and -6.85±1.00 W m−2 at the surface. The strongest aerosol DRE of -4.81 (-10.77) W m−2 at the TOA (surface) are in the summer when AODs are largest. The weakest aerosol DRE of -1.28 (-2.77) W m−2 at the TOA (surface) are in November-January when AODs and single-scattering albedos are lowest. At TWP, the annual mean clear-sky DRE is -2.82 W m−2 at the TOA and -10.34 W m−2 at the surface. The strongest aerosol DRE of -5.95 (-22.20) W m−2 at the TOA (surface) are in November (October) due to the biomass burning season’s peak. The weakest aerosol DRE of -0.96 (-4.16) W m−2 at the TOA (surface) are in March (April) when AODs are smallest. The all-sky aerosol DRE was also estimated for the first time at the SGP site. To estimate the aerosol DRE under all-skies, the observations used to estimate the clear-sky aerosol DRE were supplemented with cloud vertical extinction profile from the RL and cloud water content profiles from the Ka-band Zenith ARM radar (KAZR) and the millimeter cloud radar (MMCR). A radiative closure experiment was also conducted for cloudy-sky conditions. The relative mean differences of modeled and observed surface downwelling shortwave total fluxes were 6% (7%) for transparent (opaque) cloudy-skies. The estimated annual mean all-sky aerosol DRE is -2.13±0.54 W m−2 at the TOA and -5.95±0.87 W m−2 at the surface. The seasonal cycle of all-sky aerosol DRE is similar to that of the clear-sky, except with secondary influences of the clouds: The cloud radiative effect (CRE) seasonal cycle is strongest (most negative) in the spring, which corresponds to additional weakening of the all-sky aerosol DRE. The relative uncertainties in all-sky aerosol DRE due to measurement errors are generally comparable to those in clear-sky conditions except that the relative uncertainty in the TOA aerosol DRE due to aerosol single-scattering albedo in all-sky is larger than that in clear-sky, leading to a larger total relative uncertainty in the all-sky aerosol TOA DRE. The measurement errors in cloud properties have a small effect on the all-sky aerosol DRE. Also examined is the application of the daytime-mean and insolation-weighted-mean solar zenith angles to calculate the diurnally-averaged aerosol DRE for observed clear-sky conditions at the SGP and TWP sites. The diurnally-averaged aerosol DRE are too strong (i.e., more negative) using daytime-mean solar zenith angles (DTMSZA) and are too weak (i.e., less negative) using insolation-weighted-mean solar zenith angles (IWMSZA). The relative biases for DTMSZA and IWMSZA are about 25-30% at the TOA and 10-20% at the surface. By introducing an adjusted insolation-weighted-mean solar zenith angle, the biases in the diurnally-averaged aerosol DRE at the TOA reduces by a factor of 7-10. In addition, the 24-hour diurnal variability of the AOD is examined for observed clear-sky and cloudy-sky conditions at the SGP site. The daytime AOD variation from the Aerosol Robotic Network (AERONET) showed an excellent agreement with RL observations, which demonstrates that the RL AOD is relatively unaffected by solar background contamination. The daytime-mean AOD is only slightly larger than the nighttime-mean AOD by ∼0.01 for both clear-skies and cloudy-skies in a climatological annual-mean sense. The annual- mean clear-sky AOD diurnal minimum occurs overnight and in the early morning while the maximum occurs in the afternoon and evening. For a given day the difference between the daytime- and nighttime-mean AOD can be large, with 95% of days within about 0.2. The seasonal AOD diurnal range relative to the seasonal diurnal mean AOD was ∼10-15% except in the winter when it was ∼44%. The seasonal-mean cloudy-sky AOD diurnal variation is similar to that for clear-sky, except that the AODs are larger: the annual-mean daily-mean cloudy-sky AOD is larger than the clear-sky AOD by about 24%. The diurnal variation of the lidar ratio and its seasonal dependence are also examined. The diurnal range of the lidar ratio is found to be ∼10-20% for all seasons with a minimum near 9 am to 15 pm for all seasons except winter. Also presented is the nighttime and daytime AOD from the Cloud–Aerosol Lidar and Infrared Pathfinder Satellite (CALIPSO). The annual-mean daytime CALIPSO AOD is smaller than those from both the RL and AERONET by 0.05 to 0.1. The nighttime CALIPSO Level 3 AOD is larger than the RL AOD and might be related to the low bias mitigation methods used by CALIPSO Level 3. CALIPSO Level 2 AOD and RL observations largely agreed, which suggests that the bias due to the CALIPSO lidar ratio and sensitivity might largely cancel each other.application/pdfen-USnoneAtmospheric sciencesAtmospheric sciencesThesis