• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

The direct radiative forcing effects of aerosols on the climate in California

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Hui Du
Abstract:
The Weather Research and Forecast (WRF) model is used to explore the influence of aerosol direct radiative effects on regional climate of California. Aerosol data is provided by the MOZART global chemistry transport model and includes sulfate, black carbon, organic carbon, dust and sea salt. To investigate the sensitivity of aerosol radiative effects to different aerosol species and to the quantity of sulfate and dust, tests are conducted by using different combinations of aerosols and by resetting the quantity of sulfate and dust. The model results show that all the considered aerosols could have a cooling effect of one half to one degree in terms of temperature and that dust and sulfate are the most important aerosols. However, large uncertainties exist. The results suggest that the dust from MOZART is greatly overestimated over the simulation domain. The single scattering albedo (SSA) values of dust used in some global climate models are likely underestimated compared to recent studies on dust optical properties and could result in overestimating the corresponding cooling effects by approximately 0.1 degree. Large uncertainties exist in estimating the roles of different forcing factors which are causing the observed temperature change in the past century in California.

ii Table of contents List of Acronyms…………………………………………………………………………v List of Tables…………………………..………………………………………………. .ix List of Figures and Associate Captions………………………………………………...xi Acknowledgements...………………………………………………………………….xvii Dissertation Abstract…………………………………………………………………xviii 1. Introduction…………………………………………………………………………..1 1.1. Motivation for this study………………………………………………………..1 1.2. General climate change…………………………………………………………1 1.3. Aerosols…………………………………………………………………………..2 1.4. Global studies……………………………………………………………………5 1.4.1 Global observational studies………………………………………………...5 1.4.2 Global modeling studies……………………………………………………..7 1.5. Regional studies…………………………………………………………………9 1.5.1 Regional observational studies……………………………………………...9 1.5.2 Regional modeling studies………………………………………………….11 2. Methodology ………………………………………………………………………..13 2.1 Description of the Model……………………………………………………….14 2.1.1 Advantages of Advanced Research WRF ………………………………...14 2.1.2 Major Features of the ARW System………………………………………15 2.1.3 Model physics……………………………………………………………….16 2.1.4 Radiation scheme…………………………………………………………...18 2.2 Aerosol data……………………………………………………………………..21

iii

2.2.1 Online versus offline chemistry model ……………………………………21 2.2.2 MOZART aerosol data……………………………………………………..22 2.2.3 Introduction of aerosol data into the WRF……………………………….25 2.2.3.1 Overall strategy…………………………………………………….25 2.2.3.2 Aerosol vertical profile in the test…………………………………27 2.2.3.3 Aerosol CCM distribution…………………………………………29 2.3 Meterological Driving data ……………………………………………………31 2.4 Experiment design……………………………………………………………...32 2.4.1 Domain setup ……………………………………………………………….32 2.4.2 Control ensemble…………………………………………………………...33 2.4.3 Test run……………………………………………………………………...34 3. Results ………………………………………………………………………………35 3.1 Basic climatology of the control run …………………………………………..35 3.1.1 Surface Temperature ………………………………………………………36 3.1.2 Precipitation………………………………………………………………...37 3.2 Aerosol optical properties ……………………………………………………..38 3.2.1 Comparison with observations data ……………………………………....38 3.2.2 Aerosol optical properties ………………………………………………....43 3.3 Radiative forcing………………………………………………………………..45 3.4 Temperature response………………………………………………………….47 3.5 Role of relative humidity (RH) in aerosol climate effects……………………48 3.6 Influences of Single scattering albedo of dust………………………………...50 3.7 Sensitivity of radiative forcing to the quantities of sulfate…. ………………52

iv 3.8 Sensitivity of radiative forcing to the quantities of dust……………………..53 4. Discussion and conclusions.…………………………………………………………54 4.1 The effects of aerosol on the near surface temperature in California……....54 4.2 The most important aerosols and associated uncertainties………………….57 4.3 Influence of topography………………………………..………………………58 4.4 Uncertain factors………………………………………..………………………59 REFERENCES: ………………………………………………………………………113 Appendix A: Modified files in the WRF……………………………………………..135 Appendix B: vertical profile of aerosols in the model………………………………185 Appendix C: Microphysics and Cumulus parameterization……………………….187

v List of Acronym AERONET AErosol RObotic NETwork AFWA Air Force Weather Agency API Application Program Interface ARW Advanced Research WRF AOD Aerosol Optical Depth AR4 Fourth Assessment Report AVHRR Advanced Very High Resolution Radiometer BC Black Carbon BC1 Hydrophobic Black carbon BC2 Hydrophilic Black carbon BLA Bottom Layer Aerosol CAM Community Atmosphere Model CAPS Center for Analysis and Prediction of Storm CCM Column Cumulative Mass CERES Clouds and the Earth's Radiant Energy System CFCs Chlorofluorocarbons CRU Climate Research Unit CTM Chemistry Transport Models DMS DiMethyl Sulfide DRE Direct Radiative Effect EC Extinction Coefficient ECMWF European Centre for Medium-Range Weather Forecasts

vi ENSO El Niño Southern Oscillation ESMF Earth System Modeling Framework ESRL Earth System Research Laboratory FAA Federal Aviation Administration GCM Global Climate Model GHG Green House Gas GPU Graphics Processing Unit IMPROVE Interagency Monitoring of Protected Visual Environments IPCC Intergovernmental Panel on Climate Change ITCZ Intertropical Convergence Zone TAR Third Assessment Report TOMS Total Ozone Mapping Spectrometer LBC Lateral Boundary Conditions LSM land-surface model MCR Main Control Run MISR Multiangle Imaging Spectroradiometer MMR Mass Mixing Ratio MODIS Moderate Resolution Imaging Spectroradiometer MOZART Model of Ozone and Related Chemical Tracers NARR North American Reginal Re-analysis NASA National Aeronautics and Space Administration NCAR National Center for Atmospheric Research NCEP National Center for Environmental Prediction

vii NMM Nonhydrostatic Mesoscale Model NOAA National Oceanic and Atmospheric Administration NRL Naval Research Laboratory OC Organic Carbon OC1 ` Hydrophobic organic carbon OC2 Hydrophilic organic carbon OMI Ozone Monitoring Instrument PBL Planetary Boundary Layer PM2.5 Particulate Matters with diameters smaller than 2.5µm PM10 Particulate Matters with diameters smaller than 10µm POLDER Polarization and Directionality of the Earth Reflectances PPM Parts Per Million PTR Principal Test Run RF Radiative Forcing RH Relative Humidity RTM Radiation Transfer Models SOx Oxides of Sulfur SSA S ingle Scattering Albedo SSLT Sea salt Tmax Maximum temperature Tmin Minimum temperature TOMS Total Ozone Mapping Spectrometer T2 2-meter temperature

viii

TOA Top of the Atmosphere UCLA University of California, Los Angeles VMR Volume Mixing Ratio WPS WRF Preprocessing System WRF Weather Research & Forecast model WSF WRF Software Framework

ix List of Tables Table 1 Direct effects of aerosols on global climate change from modeling studies (BC: Black Carbon, OC: Organic Carbon)…………………………………………………….63 Table 2 Effects of aerosol on regional climate from modeling studies (BC: Black Carbon, OC: Organic Carbon)…………………………………………………………………….64 Table 3a Aerosol optical properties used in CAM for sulfate, hydrophilic carbon, sea salt (Data are from a file in the CAM radiation scheme package)…………………………...65 Table 3b Aerosol optical properties used in CAM for Dust and Black Carbon (Data are from the same source as that in Table 4a)……………………………………………….66 Table 4 Surface Emissions in MOZART (Source: Table 1 from Horowitz et al., 2006)……………………………………………………………………………………..66 Table 5 Domain averaged aerosol values for August 1996……………………………..67 Table 6 Physics options of control runs (Parameterizations listed use the naming convention in Wang et al., 2008, which also contains the relevant references)…..……..68 Table 7 Physics and dynamics configurations used in MCR (Parameterization names are from Wang et al. 2008) (See Appendix C for more detailed information about these Microphysics and cumulus options)……………………………………………………..69 Table 8 Experiments designs for different cases………………………………………..70 Table 9 Standard errors from spatial variability of MODIS and MISR (calculated based on the observed AOD over the model domain for August and September 2000) and from interannual variations of MODIS and MISR (calculated based on the data from Figure 22)………………………………………………………………………………………..71

x Table 10 Domain averaged differences (over the land areas of the model domain) for August and September 1996 of Tmax, T2 and RF between test runs and MCR (All aerosols: ‘PTR’ minus ‘MCR’, All quarterdust: ‘All_QuarterDust’ minus ‘MCR’, BC: ‘PTR’ minus ‘All_Nobc’, OC: ‘PTR’ minus ‘NO_oc’, BC+OC: ‘PTR’ minus ‘No_carbon’, Sulfate: ‘PTR’ minus ‘All_NoSulfate’, Dust: ‘PTR’ minus ‘All_NoDust’, Half Sulfate: “All_HalfSulfate’ minus ‘All_NoSulfate’, Double sulfate: ‘All_DoubleSulfate’ minus ‘All_NoSulfate’, Half dust: ‘All_NoDust’ minus ‘All_HalfDust’, Quarter dust: ‘All_QuarterDust’ minus ‘All_NoDust’, Modified dust SSA: ‘PTR’ minus ‘All_ModifiedSSA’…………………………………………………72 Table 11 Averaged (Over the land area of the model domain) T2, Tmax and Tmin changes for August and September from 1901 to 1970 and 1971 to 2000 ………..……73

xi List of Figures and Associated Captions

Figure 1 Monthly mean atmospheric carbon dioxide from 1998 to 2009 at Mauna Loa Observatory, Hawaii (Pieter Tans, NOAA/ESRL (www.esrl.noaa.gov/gmd/cgg/trends))...............................................................................74 Figure 2 Published records of surface temperature change over large regions. Köppen (1881) tropics and temperate latitudes using land air temperature. Callendar (1938) global using land stations. Willett (1950) global using land stations. Callendar (1961) 60°N to 60°S using land stations. Mitchell (1963) global using land stations. Budyko (1969) Northern Hemisphere using land stations and ship reports. Jones et al. (1986a, b) global using land stations. Hansen and Lebedeff (1987) global using land stations. Brohan et al. (2006) global using land air temperature and sea surface temperature data is the longest of the currently updated global temperature time series. All time series were smoothed using a 13-point filter. The Brohan et al. (2006) time series are anomalies from the 1961 to 1990 mean (°C). Each of the other time series was originally presented as anomalies from the mean temperature of a specific and differing base period. To make them comparable, the other time series have been adjusted to have the mean of their last 30 years identical to that same period in the Brohan et al. (2006) anomaly time series. (Source: Figure 1.3, Chap1, AR4)……………………………………………………….75 Figure 3 Aerosol Global RF effects at TOA (abscissa: the radiative forcing magnitude W m -2 ; ordinate: different forcing factors) based on the studies in Table 1………………... 76 Figure 4 Regional aerosol RF effects at surface (abscissa: the radiative forcing magnitude W m -2 ; ordinate: different forcing factors) based on the studies in Table 2…77

xii Figure 5 WRF system components (Source: Fig. 1.1, Chapter 1, Skamarock et al., 2005)……………………………………………………………………………………..78 Figure 6 ARE η coordinate, P hs is the hydrostatic pressure at surface, P ht is the h ydrostatic pressure at the top of a model (Source: Figure 2.1, Chapter 2, Skamarock et al., 2005)…………………………………………………………………………………79 Figure 7 WRF ARW Modeling System Flow Chart (Source: Wang et al. 2008, Chapter 1, WRF user guide version 2.0)…………………………………………………………….80 Figure 8 The MOZART Sulfate Volume Mixing Ratio at the bottom layer for August 1996 (Unit: Volume Mixing Ratio)……………………………………………………...81 Figure 9 Sulfate Mass Mixing Ratio at the bottom layer of the model domain (Interpolated from MOZART to the model domain) for August 1996 (Unit: Mass Mixing Ratio)…………………………………………………………………………………….82 Figure 10 The aerosol vertical profile we assumed (abscissa: scale factor; ordinate: pressure (mb))……………………………….…………………………………………..83 Figure 11 Meridional average Vertical distribution of the MOZART sulfate aerosol approximately over the model domain for August 1996 (Unit: Volume Mixing Ratio, abscissa: Longitude; ordinate: hybrid_sigma_pressure)………………………………...84 Figure 12 Meridional average MOZART vertical distribution of BC1 (Hydrophobic Black Carbon) approximately over the model domain for August 1996 (Unit: Volume Mixing Ratio; abscissa: Longitude; ordinate: hybrid_sigma_pressure)………………....85 Figure 13 Meridional average MOZART Vertical distribution of Dust1 (size: 0.01-1.0µm for August (Unit: Volume Mixing Ratio, abscissa: Longitude; ordinate: hybrid_sigma_pressure)………………………………………………………………….86

xiii

Figure 14a Aerosol cumulative column burden (kg m -2 ) for sulfate, sea salt, dust1 (0.01- 1µ m), dust2 (1.0-1.8µm), for August 1996 simulated by MOZART (Interpolated to the model domain)…………………………………………………………………………...87 Figure 14b Aerosol column burden (kg m -2 ) for Hydrophobic Black Carbon (BC1), H ydrophilic Black Carbon (BC2), Hydrophobic Organic Carbon (OC1), Hydrophilic Organic Carbon (OC2) for August 1996 simulated by MOZART (Interpolated to the model domain)…………………………………………………………………………...88 Figure 15 "Observed" (analyzed) precipitation (Top) assimilated by the NARR, and NARR precipitation,(bottom) averaged for January 1998 (inches/month) over most areas of North America. White indicates no available observations. . (Source: Figure 3, Mesinger et al. 2006)…………………………………………………………………….89 Figure 16 Topography height (meters) of the model domain…………………………...90 Figure 17 Modeled (top) T2, observed (middle) T2 and their differences (‘MCR’ minus ‘CRU’ (bottom) averaged over July, August and September 1996 (left) and December 1995, January and February 1996 (right) ) (CRU: Climate Research Unit)……...……...91 Figure 18 Modeled (top), observed (middle) average Tmax (a) and Tmin (b), and their differences (bottom) averaged over July, August and September 1996…………………92 Figure 19 Modeled (a) and the NCEP (b) average water mixing ratio at 2 meters (kg kg - 1 ) ) for July August and September 1996………… ……………………………………...93 Figure 20 Modeled (a) and CRU (b) average precipitation (mm day -1 ) for December 1995, J anuary and February 1996………………………………………………………..94

xiv Figure 21 Statewide PM2.5 and SOx emission trend for 1975-2020 (abscissa: Year; ordinate: Annual average (tons/day)) (Sources: Fig. 3-8 for (a) and Fig. 3-14 for (b), Chapter 3, Cox et al. 2009)……..………………………………………………………..95 Figure 22 Annual changes of the MODIS and the MISR AOD (average of August and September monthly mean) from 2000 to 2008 for the land/sea areas of the model domain……………………………………………………………………………………96 Figure 23 Simulated (PTR) AOD (Aerosol Optical Depth) (a) for August and September 1996 and AOD observations from MODIS (b) and from MISR (c) for August and September 2000 (observed AODs of MODIS and MISR have been interpolated into the model domain)…………………………………………………………………………...97 Figure 24 AOD Differences between the runs of ‘PTR’, ‘All_HalfSulfate’, ‘All_HalfDust’ and ‘All_QuarterDust’ and the MODIS AOD for August and September (The simulated AOD is for 1996 and the MODIS AOD is for 2000)……………………98 Figure 25 AOD Differences between the runs of ‘PTR’, ‘All_HalfSulfate’, ‘All_HalfDust’ and ‘All_QuarterDust’ and the MISR AOD for August and September (The simulated AOD is for 1996 and the MISR AOD is for 2000)……………………..99 Figure 26 Averaged AOD over the land and the ocean areas in our domain for MODIS, MISR, ‘PTR’, ‘All_HalfSulfate’, ‘All_DoubleSulfate’, ‘All_HalfDust’ and ‘All_QuarterDust’ (The simulated AOD is for 1996 and the MODIS and the MISR AOD is for 2000)……………………………………………………………………………...100 Figure 27 Modeled average AOD for sulfate, dust, carbon and sea salt for August and September 1996………………………………………………………………………...101

xv Figure 28 Simulated AODs (blue) and Observed AOD from AERONET at HJAndrews (Marked as a triangle on ‘All_dust’ in Fig. 27) for August 1996………………………102 Figure 29 Forcing (W m -2 ) of downward short wave flux at the ground surface over the l and areas for all aerosols, sulfate, dust, BC+OC, BC and OC averaged over August and September 1996 (BC: Black Carbon; OC: Organic Carbon)…………………………...103 Figure 30 Forcing of downward short wave flux at the ground surface over the land area for all aerosols with quarter dust averaged over August and September 1996…………104 Figure 31 Tmax (ºC) differences (the cooling effects) over the land areas from all aerosols, sulfate, dust, BC+OC, BC and OC for August and September 1996………...105 Figure 32 Tmax difference (the cooling effects) over the land areas from all aerosols with quarter dust for August and September 1996……………………………………..106 Figure 33 Simulated AOD from sulfate (a) and RH (b) at 00:00:00 (Pacific Standard Time), August 16, 1996………………………………………………………………...107 Figure 34 Vertical distribution of RH cross around latitude of 33 degree (the red line in Fig. 33) at 00:00:00 (Pacific Standard Time), August 16, 1996 (abscissa: grid number of model from west to east; ordinate: Height (km))……………………….………………108 Figure 35 Tmax difference (a) at 2 meter and Temperature difference at 850hPa between PTR and ‘All_modifiedSSA’ (PTR minus All_modifiedSSA) (ºC) averaged over August and September, 1996……………………………………………………………………109 Figure 36 Response of Tmax (ºC) at the surface (Over the land area) to different amounts of sulfate and dust for August and September 1996……………………………………110 Figure 37 The summer (August and September) CRU T2 (top), Tmax (middle) and Tmin (bottom) changes from 1901 to 2000 (left) and from 1991 to 2000 (right) (Only over the

xvi land areas of the model domain). Only changes significant at greater 95% significance are plotted………………………………………………………………………………......111 Figure 38 Ten-year running average of temperature anomalies (degrees C) for California relative to the 1961–1990 base period average using winter (dashed curve), spring (thin solid curve), summer (bold solid curve), and fall (dotted curve) means. The time series were computed from the Hamlet and Lettenmaier [2005] monthly 1/8-degree gridded meteorological data set (Duffy et al. 2007)……… …………………………………….112 Figure 39 T ime series of annual BC concentration in San Francisco Bay Area (SFBA), San Joaquin (SJV) and Sacramento Valleys (SACV and average (AVG) (Source: Novakov et al., 2008)…………………………………………………………113

xvii Acknowledgements

First, I would like to thank my advisor Dr. Bryan Weare for his countless hours of guidance throughout my Ph.D. study. I would also thank every committee member for their invaluable time. Special thanks to Dr. Shu-hua Chen for her help on the set up of NARR data in the WRF. I would also like to thank IT people Terry Chrudinsky and Richard Jacobsen for their continuous help on resolving the computer problems. I would like to thank Dr. Larry Horowitz from NOAA for providing me the MOZART aerosol data, thank Dr. Jianjun Chen for providing me the emission documents. I would also thank the people from WRF help, NCAR, NOAA and PNL for helping me to resolve many technical problems. Finally, special thanks to my family. Thank my wife for her trust, for her patience, and for her sacrifice in past several years. Thank my parents for their loves and continuous support. The climate observational data are from the CRU at http://www.cru.uea.ac.uk. Meteorological data are from NARR at http://www.emc.ncep.noaa.gov/mmb/rreanl/.

xviii

Dissertation abstract The Weather Research and Forecast (WRF) model is used to explore the influence of aerosol direct radiative effects on regional climate of California. Aerosol data is provided by the MOZART global chemistry transport model and includes sulfate, black carbon, organic carbon, dust and sea salt. To investigate the sensitivity of aerosol radiative effects to different aerosol species and to the quantity of sulfate and dust, tests are conducted by using different combinations of aerosols and by resetting the quantity of sulfate and dust. The model results show that all the considered aerosols could have a cooling effect of one half to one degree in terms of temperature and that dust and sulfate are the most important aerosols. However, large uncertainties exist. The results suggest that the dust from MOZART is greatly overestimated over the simulation domain. The single scattering albedo (SSA) values of dust used in some global climate models are likely underestimated compared to recent studies on dust optical properties and could result in overestimating the corresponding cooling effects by approximately 0.1 degree. Large uncertainties exist in estimating the roles of different forcing factors which are causing the observed temperature change in the past century in California.

1 1. Introduction 1.1 Motivation for this study. Climate change plays a significant role in affecting human lives. Observations have shown increases in global average air and ocean temperatures, widespread melting of snow and ice and rising global average sea level; therefore, warming of the climate system is unequivocal (Solomon et al. 2007; hereafter AR4). AR4 suggests that glacier retreat, ice shelf disruption, sea level rise, changes in rainfall patterns, and increased intensity and frequency of extreme weather events are partially due to global warming. Other effects include drought and flooding in different regions due to the decreased or increased precipitation and extensive health effects from warmer temperatures (McMichael et al., 2006). Aerosols, in particular, anthropogenic aerosols, which have been brought to attention for the last decades, may be an important factor in modifying climate change by a series of processes involving radiation. Understanding the radiative effects of aerosols is, therefore, important to comprehending global and regional climate change, accordingly developing and adopting corresponding adaptation strategies 1.2 General Climate change. Since pre-industry (AD 1000-1750), the composition of the atmosphere has been considerably altered by human-beings. Various measurements confirm that the atmospheric mixing ratio of CO 2, a dominant anthropogenic greenhouse gas (GHG), has i ncreased globally by about 100 parts per million (PPM) (36%) over the last 250 years, from about 280 to 379 PPM (AR4). Figure 1 (Pieter Tans, NOAA/ESRL ( www.esrl.noaa.gov/gmd/cgg/trends ) shows the change of CO 2 fraction by volume at

2 Mauna Loa Observatory, Hawaii, since 1960. Having been affecting global climate by altering the radiation balance of the earth-atmosphere system, GreenHouse Gases (Hereafter GHGs) are well known for their warming effects. With the increase of CO 2 , g lobal mean surface temperature has risen by 0.74°C ± 0.18°C over the last 100 years (1906–2005) (AR4). The temperature changes from various studies have been shown in Figure 2 (Chap1, AR4). Figure 1 and 2 clearly present the increasing trends for both volume mixing ratio of CO 2 and temperature, in particular, for the period between 1960 a nd 2008, indicating a strong link between CO 2 and temperature change. F igure 2 also shows an evident downward trend during the period 1940-1960. This downward trend is obviously opposed to the GHGs upward trend. Therefore, besides the well-mixed GHGs, it is very likely that there have been other important forcing factors modifying the radiation of the earth-atmosphere system, accordingly affecting climate change. These factors could include variations in solar output, land-use changes (e.g., urbanization and agriculture activities), volcanic activities and emissions of different types of aerosols into the atmosphere. To understand the process behind the temperature signals and to estimate the climate change in both global and regional-scale, it is necessary to understand all of the above forcing factors among which large uncertainties exist. Aerosol may be an important factor affecting both global and regional climate. 1.3 Aerosols

Aerosols, mainly including sulfate, black carbon (BC), organic carbon (OC), dust

and sea salt, can have natural and anthropogenic sources (Haywood and Boucher, 2000). Fossil fuel burning constitutes the main source of anthropogenic sulfate aerosol, and

3 dimethyl sulfide (DMS) and volcanic emissions are the main natural contributors (Haywood and Boucher, 2000). BC is mainly from the burning of biomass and fossil fuels, including the emissions from automobiles and aircraft (Badarinath et al., 2009). OC can be formed by both primary particles (fossil fuel and biomass burning, debris, pollen, spores) and by secondary production from gaseous compounds (Haywood and Boucher, 2000). For dust, given the condition soil skin is dry enough with high surface wind speed, dust production can originate from different sources such as deserts or sparsely vegetated areas, cultivated land outside of the growing season or areas where the soil surface is freshly exposed to wind erosion after the vegetation cover is removed (Tegen and Fung,1995). The sea salt aerosols originate from the open ocean under high wind speed conditions, so the source is found closely related to the wind speed at 10m (Gong et al., 1997). Aerosols affect climate by radiative forcing which is divided into direct, semi- direct, and indirect effects. Direct effects alter the radiative balance of the Earth- Atmosphere system by scattering and absorbing short and long wave radiation. For semi- direct effect, absorbing aerosols such as black carbon and mineral dust, absorb solar radiation or longwave radiation and therefore the atmosphere is heated, and the t emperature profile, circulation and cloud cover may be altered (Hansen et al., 1997). Heating of the atmosphere could cause evaporation of the cloud and thus reduction of cloud cover, leading to warmer climate. For indirect effect, aerosols alter the amount and lifetime of clouds by modifying the microphysical properties of clouds such as the size of cloud drops. As a result, the radiative properties of clouds are changed. The indirect effects are further sub-classified

4 into the ‘first’ and the ‘second’ indirect effects. For the first indirect effect or ‘cloud albedo effect’ (Lohmann and Feichter, 2005), given the condition of constant liquid water, more aerosols result in more cloud drops with smaller size and hence the cloud optical properties alters (Ramaswamy et al., 2001). As a result, it would be more difficult to have precipitation clouds with smaller cloud drops. For the ‘second indirect effect’ (Ramaswamy et al., 2001), also called aerosols the ‘cloud lifetime effect’ (Lohmann and Feichter, 2005), aerosols can affect the liquid water content, cloud height, and lifetime of clouds. As a result, changed clouds alter the radiation balance. The net effect from all aerosol types is to cool the climate system (AR4). Obviously, this cooling effect is opposed to the well-known warming effect of the GHGs. The above three processes are determined by a series of complex physical and optical properties, which depend on the aerosol composition, size, shape and ambient environment. Unlike the GHGs, which are well mixed and evenly distributed, the concentrations of different aerosols considerately vary by region and time, and it is difficult to obtain relatively accurate measurements. Another important property for s ome aerosols is water uptake. In other words, some aerosols swell by absorbing water. The growth will depend on chemical composition, the size of the particle and ambie nt relative humidity (Li et al., 2001). Despite the efforts spent on the effects of aerosols on climate in the past years, many uncertainties still remain due to all these challenges. Nonetheless, in the past decades, especially in recent years, we have gained a better understanding of the aerosol effects on global and regional climate system. Section 1.4 and 1.5 review the literature about the effects of aerosols on global and regional climate from the aspects of both modeling studies and observational studies

5 1.4 Global studies. Global aerosol studies enable us to approximately investigate the overall effects of aerosols on climate and to estimate its importance relative to the effects of GHGs on a global scale. Global observational studies usually employ the facilities like satellites and surface instruments to estimate the aerosol radiative forcing, geographical distribution, and composition and temporal variations. Such observations are also important references for evaluating the simulations of aerosols from chemistry transport models. Global modeling studies usually employ Global Climate Model (GCM), radiation transfer models (RTM) and chemistry transport models (CTM) which can be used not only as estimating aerosol amount, composition and radiative forcing but also as projecting the effects of aerosols on climate change in the future under different scenarios. 1.4.1 Global Observational studies Many observation-based studies have been conducted by using satellite data and surface-based data. These studies measure parameters such as aerosol optical depth, composition, size distribution and provide direct valuable information for estimating the aerosol effects on climate and validating the quality of global chemistry transport models. However, observation-based studies have limitations. In situ studies such as AErosol RObotic NETwork (AERONET) observe only scattered samples in space and time, which may not represent even a regional average, in particular in a region with complex topography. Most surface-based studies only observe the aerosol information at the surface. However, the vertical profiles of aerosols are also important, especially when it comes to the interaction with clouds. Although, surface instruments like lidar have the ability to retrieve some vertical information of aerosols (Turner et al., 2002), the number

6 of those sites is still very limited and not enough for observing the geographical distribution of aerosols. Another disadvantage is that in situ studies often observe only the aerosol optical depth instead of separating the effects from different aerosol species. As a result, they do not provide clues on the relative importance for different aerosols. For satellite data, they have the advantages of global coverage and high resolution. Their main limitation is that most of them provide only a measure of aerosol optical depth, and thus give little information about species and size. Also, most instruments are incapable of estimating aerosol properties as a function of height above the surface. They are usually affected by highly variable weather condition and different characteristic of land surfaces (Hauser et al., 2005). For example, it is usually difficult to separate the scattering from bright surfaces and that from aerosols. Therefore, some studies (Remer et al., 2005) can only have relatively quality-assured observations over dark land areas. Nonetheless, observational studies provide us with valuable information for validating the results of chemistry transport model and estimating the effects of aerosols on climate change. Bellouin et al. (2008) use Aerosol Optical Depth (AOD) and accumulation-mode fractions retrieved by the satellite instrument of Moderate Resolution Imaging Spectroradiometer (MODIS) to estimate the anthropogenic aerosol direct forcing in the shortwave spectrum. The direct forcing is estimated at -1.30 and -0.65 Wm -2 on a global a verage for clear-sky and all sky respectively for 2002. Quaas et al. (2008) estimate the aerosol radiative forcing by using the information from the Clouds and the Earth's Radiant Energy System (CERES) and MODIS. They derive a statistical relationship between planetary albedo and cloud properties and between the cloud properties and

7 column aerosol concentration and estimate an anthropogenic radiative forcing of - 0.9 ± 0.4 Wm -2 for the aerosol direct effect and -0.2 ± 0.1 Wm -2 for the cloud albedo effect. P atadia et al. (2008) use one year (2000-2001) of Multiangle Imaging Spectroradiometer (MISR), MODIS and CERES data sets from NASA’s Terra satellite to estimate the Top Of the Atmosphere (TOA) cloud-free direct radiative effect (DRE) of aerosols over global land areas. The global mean shortwave DRE is -5.1 ± 1.1 Wm -2 . T he above studies are all satellite-based with a global coverage. In situ studies will be discussed for regional studies. In summary, observational studies usually estimate the Radiative Forcing (RF) by observing AOD. They agree at the forcing direction (negative or cooling). However, the uncertainties of the magnitudes, partly from the discrepancies of AOD, vary from -5.1 to -0.1 Wm -2 among these studies. The discrepancies of AOD are due to different a ssumptions in the cloud clearing algorithms (approaches to separate the AOD from clouds and the AOD from aerosols), aerosol models, different wavelengths and viewing geometries used in the retrievals, different parameterizations of ocean surface reflectance, etc (AR4). Therefore, large uncertainties would enter if these results are used to estimate the response of climate change. 1.4.2 Global Modeling studies Modeling studies have advantages such as global coverage, longer time span and parameters control. For example, in a modeling study, the response of climate to the quantity of aerosols can be estimated by resetting the aerosol inputs. However, limitations also exist for modeling studies. They may come from low resolution, simplified physics schemes, uncertainties of initial and boundary conditions and the uncertainties of

8 estimation of aerosols from chemistry transport models. In the last decades, considerable efforts have been spent on global modeling studies. Kirkevag et al. (2008) integrate a new aerosol module in the atmospheric GCM CAM-Oslo to estimate the effect of anthropogenic aerosols. They include sea salt, mineral dust, sulfate, black carbon and particulate organic matter, which compose most of the important aerosol species. The results show that anthropogenic aerosols produce a global near-surface radiative forcing of -3.88 Wm -2 and a 5.5% precipitation decrease by c ombined direct and indirect effects . There are many other studies which are s ummarized in Table 1. The studies in Table 1 clearly show cooling effects of aerosols on global climate. In addition to the direct cooling effects, aerosols may have modified global and regional circulations. Gu et al. (2006) show that inclusion of a background aerosol optical depth of 0.2 reduces the global mean net surface solar flux by about 5 Wm -2 and produces a de crease in precipitation in the tropics as a result of decreased temperature contrast between this area and the middle to high latitudes. Kristjansson et al. (2005) found similar results and the southward shift of the Intertropical Convergence Zone (ITCZ) due to a large cooling effect at high latitude. This implies that aerosols not only cool the c limate system, but also possibly influence the atmospheric circulation. Sulfate and black carbon have received the most attention and are considered in most studies in Table 1; however, few studies include all of the important aerosols. Large discrepancies also exist between different studies and are demonstrated by Figure 3, a bar plot for the estimated ranges of RF from both the global observational and modeling studies we have discussed. In Figure 3, the uncertainties from all aerosols and sulfate are

Full document contains 208 pages
Abstract: The Weather Research and Forecast (WRF) model is used to explore the influence of aerosol direct radiative effects on regional climate of California. Aerosol data is provided by the MOZART global chemistry transport model and includes sulfate, black carbon, organic carbon, dust and sea salt. To investigate the sensitivity of aerosol radiative effects to different aerosol species and to the quantity of sulfate and dust, tests are conducted by using different combinations of aerosols and by resetting the quantity of sulfate and dust. The model results show that all the considered aerosols could have a cooling effect of one half to one degree in terms of temperature and that dust and sulfate are the most important aerosols. However, large uncertainties exist. The results suggest that the dust from MOZART is greatly overestimated over the simulation domain. The single scattering albedo (SSA) values of dust used in some global climate models are likely underestimated compared to recent studies on dust optical properties and could result in overestimating the corresponding cooling effects by approximately 0.1 degree. Large uncertainties exist in estimating the roles of different forcing factors which are causing the observed temperature change in the past century in California.