Simulating chemistry–aerosol–cloud–radiation–climate feedbacks over the continental U.S. using the online-coupled Weather Research Forecasting Model with chemistry (WRF/Chem)

Abstract

The chemistry–aerosol–cloud–radiation–climate feedbacks are simulated using WRF/Chem over the continental U.S. in January and July 2001. Aerosols can reduce incoming solar radiation by up to −9% in January and −16% in July and 2-m temperatures by up to 0.16 °C in January and 0.37 °C in July over most of the continental U.S. The NO₂ photolysis rates decrease in July by up to −8% over the central and eastern U.S. where aerosol concentrations are high but increase by up to 7% over the western U.S. in July and up to 13% over the entire domain in January. Planetary boundary layer (PBL) height reduces by up to −23% in January and −24% in July. Temperatures and wind speeds in July in big cities such as Atlanta and New York City reduce at/near surface but increase at higher altitudes. The changes in PBL height, temperatures, and wind speed indicate a more stable atmospheric stability of the PBL and further exacerbate air pollution over areas where air pollution is already severe. Aerosols can increase cloud optical depths in big cities in July, and can lead to 500–5000 cm⁻³ cloud condensation nuclei (CCN) at a supersaturation of 1% over most land areas and 10–500 cm⁻³ CCN over ocean in both months with higher values over most areas in July than in January, particularly in the eastern U.S. The total column cloud droplet number concentrations are up to 4.9 × 10⁶ cm⁻² in January and up to 11.8 × 10⁶ cm⁻² in July, with higher values over regions with high CCN concentrations and sufficient cloud coverage. Aerosols can reduce daily precipitation by up to 1.1 mm day⁻¹ in January and 19.4 mm day⁻¹ in July thus the wet removal rates over most of the land areas due to the formation of small CCNs, but they can increase precipitation over regions with the formation of large/giant CCN. These results indicate potential importance of the aerosol feedbacks and an urgent need for their accurate representations in current atmospheric models to reduce uncertainties associated with climate change predictions.

Introduction

The complex feedback mechanisms among chemistry–aerosol–cloud–radiation–climate exist ubiquitously in the Earth systems and represent one of the most uncertain research areas in understanding climate change and its potential impact on atmosphere (Jacobson, 2002, IPCC, 2007, Zhang, 2008, Jacob and Winner, 2009). The feedbacks of aerosols may include a reduction of downward solar radiation (direct effect); a decrease in surface/near surface temperature and wind speed, as well as planetary boundary layer (PBL) height but an increase in relative humidity (RH) and atmospheric stability (semi-direct effect), a decrease in cloud drop size but an increase in cloud droplet number concentrations (CDNC) via serving as cloud condensation nuclei (CCN) (first indirect effect or cloud albedo effect), as well as an increase in liquid water content, cloud coverage, and lifetime of low-level clouds and either suppression or enhancement of precipitation (the second indirect effect or cloud lifetime effect). These feedbacks have been observed in numerous field experiments or through analyses of long-term historic surface and satellite observational data. For example, smoke from rain forest fires over Amazon and Indonesia and burning of agricultural vegetations can inhibit rainfall by shutting off warm rain-forming processes over these regions (Warner, 1968, Kaufman and Fraser, 1997, Rosenfeld and Lensky, 1998, Rosenfeld, 1999, Rosenfeld and Woodley, 1999). The suppression in orographic precipitation by anthropogenic aerosols was found to be 15–25% of the annual precipitation in hilly areas in California and Israel (Givati and Rosenfeld, 2004, Givati and Rosenfeld, 2005, Rosenfeld et al., 2008a). The orographic precipitation observed at Mt. Hua near Xi’an in China decreased by 30–50% during hazy conditions in the presence of high levels of aerosols and small CCN based on the analyses of more than 50-year observations (Rosenfeld et al., 2007a). Enhanced rainfall, on the other hand, was found in (Braham et al., 1981, Cerveny and Bailing, 1998) and downwind (Eagen et al., 1974, Jauregui and Romales, 1996) of major urban areas and paper mills, suggesting that giant CCN can enhance precipitation. The two opposite effects of aerosols on precipitation are results of different aerosol radiative properties and CCN potentials under different conditions (Rosenfeld et al., 2008b). For example, atmospheric aerosols decrease net downward solar radiation reaching surface, causing less heat available for water evaporation thus suppressing precipitation. On the other hand, the strongly-absorbing aerosols such as mineral dust and particles from heavy smoke have been found to invigorate and restructure convective clouds due to the solar heating and induced convection by these aerosols (Levin et al., 1996, Levin et al., 2005, Rudich et al., 2003, Miller et al., 2004, Koren et al., 2005, Klüser et al., 2008, Levin and Brenguier, 2009), thus enhancing precipitation.Aerosols such as mineral dust and black carbon can also alter atmospheric circulation (e.g., Zanis, 2009) through changing the atmospheric heating and stability to affect the monsoons (Lau et al., 2006, Lau and Kim, 2006) and severe storms (Rosenfeld, 2006, Zhang et al., 2007, Bell et al., 2008, Ramanathan and Carmichael, 2008 and references therein).

Accurately simulating these feedbacks requires the use of online-coupled meteorology-chemistry models; among which the NOAA Weather Research and Forecasting Model with Chemistry (WRF/Chem) of Grell et al. (2005) represents a state-of-the-science online model. While most air quality modeling has focused on accessing the models’ capability in capturing past pollution episodes and forecasting short-term (2–4 days) air quality, there have been fewer studies on simulating the feedbacks among atmospheric components and/or processes. Simulating hurricane Katrina using WRF, Rosenfeld et al. (2007b) reported a 25% reduction in the radius of hurricane force winds in response to warm rain suppression by sub-micron aerosols. By coupling a cloud microphysics module with WRF, Lynn et al. (2007) illustrates the suppression of precipitation by continental aerosols in the Sierra Nevada Mountains. Using a global-through-urban model, GATOR-GCMOM, over the Los Angeles basin, Jacobson et al. (2007) found that aerosol particles and their precursor gases reduce net downward surface total solar irradiance, near-surface temperature, and surface wind speed; increase RH, aerosol optical depth (AOD), and cloud optical thickness (COT), cloud fractions, cloud liquid water; and either increase or decrease precipitation depending on location and magnitude of precipitation intensity. Applying WRF/Chem over the eastern Texas in August 2000, Zhang (2008) showed that the presence of aerosols leads to a decrease in temperature by up to 0.18 °C at/near surface but an increase by 0.16 °C aloft in PBL (defined as the height from surface to ∼2.9 km above the ground level (AGL)) at a site in the coastal area of the Galveston Bay, and Zhang et al. (in press) reported reduction of the domain-wide mean precipitation by 0.22–0.59 mm day⁻¹ over the eastern Texas. In this work, WRF/Chem simulations are conducted at a horizontal grid spacing of 36 km for January and July 2001 over North America that covers the contiguous U.S. (CONUS), southern Canada, and northern Mexico to examine the importance of the aforementioned feedbacks. Seasonal variations in aerosol direct, semi-direct, and indirect feedbacks are analyzed and contrasted. Limitations and uncertainties in accurately representing such feedbacks to be addressed in future online-coupled model development and improvement are discussed.