The monthly/seasonal characteristics of the concentration and aerosol optical depth (AOD) of four aerosol components (water-soluble, insoluble, black carbon (BC), and sea-salt) and their direct radiative forcing (DRF) were investigated at three environmental locations in Southeast Texas. We used fine particulate matter (PM2.5) samples measured at one urban residential (Aldine (AD)) and two suburban (Deer Park (DP) and West Liberty (WL)) sites located around Houston during 2016-2017, and performed model-based analysis using the mass concentrations of the four aerosol components to evaluate their impact on the DRF. Overall, the concentrations, AODs, and DRFs of all four aerosol components at AD were higher than those at DP and WL during the study period. In particular, the water-soluble component was the most dominant contributor, except for absorbing BC. The monthly AOD patterns of the four individual aerosol components (especially, water-soluble and BC) at the three sites were found to have strong associations with their concentrations and/or relative humidity (RH). The DRFs at the top of the atmosphere (DRFTOA) and surface level (DRFSFC) for most of the aerosol components were found to be highest in winter 2017 (AD), spring 2016 and winter 2017 (DP), and winter 2016 and fall 2017 (WL). The exceptions were sea-salt and insoluble components, which showed a peak in summer 2016 and no distinct monthly variation, respectively. Uncertainties in the DRFs of the four target aerosol components calculated using in-situ RH measurements were found to be less than 20%, with the exception of the water-soluble component at WL (24%). A sensitivity test showed that the DRFs of the aerosol components were slightly and significantly influenced by changes in AOD and single scattering albedo, respectively; additionally, sensitively changed with RH.

We estimated the current (base years) and future (2021-2100) direct radiative forcing ( DRF) of four aerosol components (water-soluble, insoluble, black carbon (BC), and sea-salt) at urban (Yeonsan (Busan) and Gwangjin (Seoul)) and background sites (Aewol and Gosan (Jeju Island)), based on a modeling approach. The analysis for base years was conducted using PM2.5 samples measured at two urban and two background sites (Yeonsan and Gwangjin: 2016, Aewol and Gosan: 2014). The future DRFs were estimated according to changes in relative humidity (RH) of RCP8.5 climate change scenario at the same sites during four different periods (PI: 2021 similar to 2040, PII: 2041 similar to 2060, PIII: 2061 similar to 2080, and PIV: 2081 similar to 2100). In addition, we compared the differences between the DRFs of future (PI similar to PIV) and base years (2016 and 2014). Overall, the water-soluble component was predominant over all other components in terms of the concentrations, optical parameters (e.g., AOD), and DRFs, regardless of sites. For the base years, the monthly patterns of total DRFs for all components and the DRFs for the water-soluble component varied with sites, and months of their highest and lowest DRFs were different depending on sites. This might be due to the combined effect of the monthly patterns of the concentrations and RHs for each site. For the differences between the DRFs of future and base years, the highest future DRFs at Yeonsan and Aewol ranged from -59 to -63 W/m(2) increasing -20 (July in PII) to -28 W/m(2) (August in PIII) compared to the base years and from -73 to -74 W/m(2) increasing -31 (July in PII) to -41 W/m(2) (September in PIV), respectively. These DRFs at Gwangjin and Gosan ranged from -79 to -84 W/m(2) increasing -29 (June in PII and PIII) to -34 W/m(2) (June in PI) and from -58 to -92 W/m(2) increasing -14 (July in PII) to -26 W/m(2) (May in PI), respectively. The high heating rates at Yeonsan (up to 4.4 K/day in November) and Aewol (up to 3.7 K/day in February) of BC component might be caused by its strong radiative absorption.

The concentration characteristics of four aerosol chemical components (water-soluble, insoluble, black carbon (BC), and sea-salt) and their direct radiative forcing (DRF) were analyzed using the daily or hourly data (PM2.5) measured at urban (Yeonsan, Gwangbok, Hakjang, and Gijang in Busan) and background sites (Aewol in Jeju Island) during haze events, based on a modeling approach. Overall, the concentrations of water-soluble component and its impact on the DRF were predominant over all other components at most of the sites (especially at Aewol, Gwangbok and/or Hakjang). The DRFs at the surface (DRFSFC), top of the atmosphere (DRFTOA), and the atmosphere (DRFATM) for most aerosol components (except for BC) at most of the sites (except for Yeonsan) were high in spring or winter and low in summer or fall. Meanwhile, the DRFs at Yeonsan were highest in summer (for DRFTOA) or fall (for DRFSFC) and lowest in spring (for both). These seasonal DRF characteristics in the study sites might be closely related to the seasonal patterns of aerosol component concentrations and/ or meteorological conditions (e.g., relative humidity). In addition, the positive DRFATM of BC in the study sites was highest among the all aerosol components due to strong radiative absorption. The differences in DRFs for water-soluble component between haze and non-haze periods were largest in the all study sites. In particular, the DRFTOA (and DRFSFC) of water-soluble at the sites of Gwangbok and Aewol during the haze periods were higher by a factor of 1.8 and 2.3 (and a factor of 1.9 and 2.4) than those during the non-haze periods.