• Abdul-Razzak, H., , and S. J. Ghan, 2000: A parameterization of aerosol activation 2. Multiple aerosol types. J. Geophys. Res., 105 (D5), 68376844.

    • Search Google Scholar
    • Export Citation
  • Abdul-Razzak, H., , and S. J. Ghan, 2002: A parameterization of aerosol activation 3. Sectional representation. J. Geophys. Res., 107, 4026, doi:10.1029/2001JD000483.

    • Search Google Scholar
    • Export Citation
  • Andreae, M. O., and Coauthors, 2004: Smoking rain clouds over the Amazon. Science, 303, 13371342.

  • Bell, T. L., and Coauthors, 2008: Midweek increase in U.S. summer rain and storm heights suggests air pollution invigorates rainstorms. J. Geophys. Res., 113, D02209, doi:10.1029/2007JD008623.

    • Search Google Scholar
    • Export Citation
  • Chou, M.-D., , and M. J. Suarez, 1999: A shortwave radiation parameterization for atmospheric studies. NASA Rep. TM-104606, 40 pp.

  • Chou, M.-D., , W. Ridgway, , and M.-H. Yan, 1999: Parameterizations for water vapor IR radiative transfer in both the middle and lower atmospheres. J. Atmos. Sci., 52, 11591167.

    • Search Google Scholar
    • Export Citation
  • Donner, L. J., , C. J. Seman, , and R. S. Hemler, 1999: Three-dimensional cloud system modeling of GATE convection. J. Atmos. Sci., 56, 18851912.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , C. Wang, , J. Wilson, , J. Ström, , and R. Krejci, 2006: Explicit simulation of aerosol physics in a cloud-resolving model: Aerosol transport and processing in the free troposphere. J. Atmos. Sci., 63, 682696.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , A. Engström, , and A. Söderberg, 2011: Impact of two-way aerosol–cloud interaction and changes in aerosol size distribution on simulated aerosol-induced deep convective cloud sensitivity. J. Atmos. Sci., 68, 685698.

    • Search Google Scholar
    • Export Citation
  • Fan, J., and Coauthors, 2009: Dominant role by vertical wind shear in regulating aerosol effects on deep convective clouds. J. Geophys. Res., 114, D22206, doi:10.1029/2009JD012352.

    • Search Google Scholar
    • Export Citation
  • Forster, P., and Coauthors, 2007: Changes in atmospheric constituents and in radiative forcing. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 129–234.

  • Fridlind, A., and Coauthors, cited 2009: ARM/GCSS/SPARC TWP-ICE intercomparison study. [Available online at http://science.arm.gov/wg/cpm/scm/scmic6/index.html.]

  • Ghan, S., , N. Laulainen, , R. Easter, , R. Wagener, , S. Nemesure, , E. Chapman, , Y. Zhang, , and R. Leung, 2001: Evaluation of aerosol direct radiative forcing in mirage. J. Geophys. Res., 106 (D6), 52955316.

    • Search Google Scholar
    • Export Citation
  • Givati, A., , and D. Rosenfeld, 2004: Quantifying precipitation suppression due to air pollution. J. Appl. Meteor., 43, 10381056.

  • Grabowski, W. W., , X. Wu, , and M. W. Moncrieff, 1996: Cloud-resolving modeling of tropical cloud systems during phase III of GATE. Part I: Two-dimensional experiments. J. Atmos. Sci., 53, 36843709.

    • Search Google Scholar
    • Export Citation
  • Grabowski, W. W., , J.-I. Yano, , and M. W. Moncrieff, 2000: Cloud-resolving modeling of tropical circulations driven by large-scale SST gradients. J. Atmos. Sci., 57, 20222039.

    • Search Google Scholar
    • Export Citation
  • Hallet, J., , and S. C. Mossop, 1974: Production of secondary ice particles during the riming process. Nature, 249, 2628.

  • Hobbs, P. V., 1993: Aerosol–Cloud–Climate Interactions. Academic Press, 233 pp.

  • Houze, R. A., 1993: Cloud Dynamics. Academic Press, 573 pp.

  • Khain, A., , and B. Lynn, 2009: Simulation of a supercell storm in clean and dirty atmosphere using weather research and forecast model with spectral bin microphysics. J. Geophys. Res., 114, D19209, doi:10.1029/2009JD011827.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , D. Rosenfeld, , and A. Pokrovsky, 2005: Aerosol impact on the dynamics and microphysics of deep convective clouds. Quart. J. Roy. Meteor. Soc., 131, 26392663.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , N. BenMoshe, , and A. Pokrovsky, 2008: Factors determining the impact of aerosols on surface precipitation from clouds: Attempt of classification. J. Atmos. Sci., 65, 17211748.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , B. Lynn, , and J. Dudhia, 2010: Aerosol effects on intensity of landfalling hurricanes as seen from simulations with WRF model with spectral bin microphysics. J. Atmos. Sci., 67, 365384.

    • Search Google Scholar
    • Export Citation
  • Klemp, J. B., , and R. Wilhelmson, 1978: The simulation of three-dimensional convective storm dynamics. J. Atmos. Sci., 35, 10701096.

  • Koren, I., , Y. J. Kaufman, , D. Rosenfeld, , L. A. Remer, , and Y. Rudich, 2005: Aerosol invigoration and restructuring of Atlantic convective clouds. Geophys. Res. Lett., 32, LI4828, doi:10.1029/2005GL023187.

    • Search Google Scholar
    • Export Citation
  • Kratz, D. P., , M.-D. Chou, , M.-H. Yan, , and C.-H. Ho, 1998: Minor trace gas radiative forcing calculations using the k-distribution method with one-parameter scaling. J. Geophys. Res., 103, 31 64731 656.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , and G. Feingold, 2010: Precipitating cloud-system response to aerosol perturbations. Geophys. Res. Lett., 37, L23806, doi:10.1029/2010GL045596.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , V. T. J. Phillips, , and Y. Ming, 2008a: Examination of aerosol effects on precipitation in deep convective clouds during the 1997 ARM summer experiment. Quart. J. Roy. Meteor. Soc., 134, 12011220.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , V. T. J. Phillips, , and Y. Ming, 2008b: The dependence of aerosol effects on clouds and precipitation on cloud-system organization, shear and stability. J. Geophys. Res., 113, D16202, doi:10.1029/2007JD009224.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , and V. T. J. Phillips, 2009: Sensitivity of aerosol and cloud effects on radiation to cloud types: Comparison between deep convective clouds and warm stratiform clouds over one-day period. Atmos. Chem. Phys., 9, 25552575.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , and J. E. Penner, 2010: Thunderstorm and stratocumulus: How does their contrasting morphology affect their interactions with aerosols? Atmos. Chem. Phys., 10, 68196837, doi:10.5194/acp-10-6819-2010.

    • Search Google Scholar
    • Export Citation
  • Lerach, D. G., , B. J. Gaudet, , and W. R. Cotton, 2008: Idealized simulations of aerosol influences on tornadogenesis. Geophys. Res. Lett., 35, L23806, doi:10.1029/2008GL035617.

    • Search Google Scholar
    • Export Citation
  • Liljegren, J. C., 1994: Two-channel microwave radiometer for observations of total column precipitable water vapor and cloud liquid water path. Proc. Fifth Symp. on Global Change, Nashville, TN, Amer. Meteor. Soc., 266–269.

  • Lin, Y. L., , R. D. Farley, , and H. D. Orville, 1983: Bulk parameterization of the snow field in a cloud model. J. Appl. Meteor., 19, 10651092.

    • Search Google Scholar
    • Export Citation
  • Liu, C.-H., , and M. W. Moncrieff, 1996: A numerical study of the effects of ambient flow and shear on density currents. Mon. Wea. Rev., 124, 22822303.

    • Search Google Scholar
    • Export Citation
  • Lohmann, U., , and K. Diehl, 2006: Sensitivity studies of the importance of dust ice nuclei for the indirect aerosol effect on stratiform mixed-phase clouds. J. Atmos. Sci., 63, 968982.

    • Search Google Scholar
    • Export Citation
  • Lynn, B., , A. P. Khain, , J. Dudhia, , D. Rosenfeld, , A. Pokrovsky, , and A. Seifert, 2005: Spectral (bin) microphysics coupled with a mesoscale model (MM5). Part II: Simulation of a CaPe rain event with squall line. Mon. Wea. Rev., 133, 5971.

    • Search Google Scholar
    • Export Citation
  • Lynn, B., , A. P. Khain, , D. Rosenfeld, , and W. L. Woodley, 2007: Effects of aerosols on precipitation from orographic clouds. J. Geophys. Res., 112, D10225, doi:10.1029/2006JD007537.

    • Search Google Scholar
    • Export Citation
  • Moncrieff, M. W., , and C.-H. Liu, 1999: Convection initiation by density currents: Role of convergence, shear, and dynamical organization. Mon. Wea. Rev., 127, 24552464.

    • Search Google Scholar
    • Export Citation
  • Morrison, H., , and W. W. Grabowski, 2011: Cloud-system resolving model simulations of aerosol indirect effects on tropical deep convection and its thermodynamic environment. Atmos. Chem. Phys. Discuss., 11, 15 57315 629.

    • Search Google Scholar
    • Export Citation
  • Ntelekos, A. A., , J. A. Smith, , L. Donner, , J. D. Fast, , W. I. Gustafson, Jr., , E. G. Chapman, , and W. F. Krajewski, 2009: The effects of aerosols on intense convective precipitation in the northeastern United States. Quart. J. Roy. Meteor. Soc., 135, 13671391.

    • Search Google Scholar
    • Export Citation
  • Rogers, R. R., , and M. K. Yau, 1991: A Short Course in Cloud Physics. Pergamon Press, 293 pp.

  • Rosenfeld, D., , A. Khain, , B. Lynn, , and W. L. Woodley, 2007: Simulation of hurricane response to suppression of warm rain by sub-micron aerosols. Atmos. Chem. Phys., 7, 34113424.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , U. Lohmann, , G. B. Raga, , C. D. O’Dowd, , M. Kulmala, , S. Fuzzi, , A. Reissell, , and M. O. Andreae, 2008: Flood or drought: How do aerosols affect precipitation? Science, 321, 13091313.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , J. B. Klemp, , and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45, 463485.

  • Saleeby, S. M., , and W. R. Cotton, 2004: A large-droplet mode and prognostic number concentration of cloud droplets in the Colorado state university regional atmospheric modeling system (RAMS). Part I: Module description and supercell test simulations. J. Appl. Meteor., 43, 182195.

    • Search Google Scholar
    • Export Citation
  • Seifert, A., , and K. D. Beheng, 2006: A double-moment cloud microphysics parameterization for mixed-phase clouds. Part 2: Maritime vs. continental deep convective storms. Meteor. Atmos. Phys., 92, 6782.

    • Search Google Scholar
    • Export Citation
  • Seifert, A., , C. Köhler, , and K. D. Beheng, 2011: Aerosol-cloud-precipitation effects over Germany as simulated by a convective-scale numerical weather prediction model. Atmos. Chem. Phys. Discuss., 11, 20 20320 243.

    • Search Google Scholar
    • Export Citation
  • Simpson, J., , and W.-K. Tao, 1993: The Goddard Cumulus Ensemble model. Part II: Applications for studying cloud precipitating processes and for NASA TRMM. Terr. Atmos. Ocean. Sci., 4, 73116.

    • Search Google Scholar
    • Export Citation
  • Soong, S.-T., , and Y. Ogura, 1980: Response of trade wind cumuli to large-scale processes. J. Atmos. Sci., 37, 20352050.

  • Stevens, B., , and G. Feingold, 2009: Untangling aerosol effects on clouds and precipitation in a buffered system. Nature, 461, 607613, doi:10.1038/nature08281.

    • Search Google Scholar
    • Export Citation
  • Storer, R. L., , S. C. van den Heever, , and G. L. Stephens, 2010: Modeling aerosol impacts on convection under differing storm environments. J. Atmos. Sci., 67, 39043915.

    • Search Google Scholar
    • Export Citation
  • Tao, W.-K., , and J. Simpson, 1993: The Goddard Cumulus Ensemble model. Part I: Model description. Terr. Atmos. Ocean. Sci., 4, 1954.

  • Tao, W.-K., and Coauthors, 2003: Microphysics, radiation and surface processes in the Goddard Cumulus Ensemble (GCE) model. Meteor. Atmos. Phys., 82, 97137.

    • Search Google Scholar
    • Export Citation
  • Tao, W.-K., and Coauthors, 2006: Retrieval of latent heating from TRMM measurements. Bull. Amer. Meteor. Soc., 87, 15551572.

  • Tao, W.-K., , X. Li, , A. Khain, , T. Matsui, , S. Lang, , and J. Simpson, 2007: The role of atmospheric aerosol concentration on deep convective precipitation: Cloud-resolving model simulations. J. Geophys. Res., 112, D24S18, doi:10.1029/2007JD008728.

    • Search Google Scholar
    • Export Citation
  • Tompkins, A. M., 2000: The impact of dimensionality on long-term cloud resolving model simulations. Mon. Wea. Rev., 128, 15211535.

  • Tompkins, A. M., 2001: Organization of tropical convection in low vertical wind shears: The role of cold pools. J. Atmos. Sci., 58, 16501672.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , and W. R. Cotton, 2007: Urban aerosol impacts on downwind convective storms. J. Appl. Meteor. Climatol., 46, 828850.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , G. G. Carrio, , W. R. Cotton, , P. J. DeMott, , and A. J. Prenni, 2006: Impacts of nucleating aerosol on Florida Storms. Part I: Mesoscale simulations. J. Atmos. Sci., 63, 17521775.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , G. L. Stephens, , and N. B. Wood, 2011: Aerosol indirect effects on tropical convection characteristics under conditions of radiative–convective equilibrium. J. Atmos. Sci., 68, 699718.

    • Search Google Scholar
    • Export Citation
  • Vaughan, G., , K. Bower, , C. Schiller, , A. R. MacKenzie, , T. Peter, , H. Schlager, , N. R. P. Harris, , and P. T. May, 2008: SCOUT-O3/ACTIVE: High-altitude aircraft measurements around deep tropical convection. Bull. Amer. Meteor. Soc., 89, 647662.

    • Search Google Scholar
    • Export Citation
  • Walko, R. L., , W. R. Cotton, , M. P. Meyers, , and J. Y. Harrington, 1995: New RAMS cloud microphysics parameterization: Part I. The single-moment scheme. Atmos. Res., 38, 2962.

    • Search Google Scholar
    • Export Citation
  • Wang, H., , and G. Feingold, 2009: Modeling open cellular structures and drizzle in marine stratocumulus. Part II: The microphysics and dynamics of the boundary region between open and closed cells. J. Atmos. Sci., 66, 32573275.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , and J. B. Klemp, 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110, 504520.

    • Search Google Scholar
    • Export Citation
  • Xie, S., , T. Hume, , C. Jakob, , S. A. Klein, , and R. B. McCoy, 2010: Observed large-scale structures and diabatic heating and drying profiles during TWP-ICE. J. Climate, 23, 5779.

    • Search Google Scholar
    • Export Citation
  • Xu, K. M., and Coauthors, 2002: An intercomparison of cloud-resolving models with the Atmospheric Radiation Measurement summer 1997 intensive observation period data. Quart. J. Roy. Meteor. Soc., 128, 593624.

    • Search Google Scholar
    • Export Citation
  • Zhang, H., , G. McFarquhar, , S. Saleeby, , and W. R. Cotton, 2007: Impacts of Saharan Dust as CCN on the evolution of an idealized tropical cyclone. Geophys. Res. Lett., 34, L14812, doi:10.1029/2007GL029876.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Vertical profiles of initial potential temperature and water vapor mixing ratio.

  • View in gallery

    Horizontal distribution of the domain-averaged background aerosol number concentration over the PBL for the ideal-high and -low runs.

  • View in gallery

    Time series of the area-averaged precipitation rate for observation and the control run.

  • View in gallery

    Contours of cloud liquid mixing ratio (g kg−1) at the time of the occurrence of maximum precipitation rate for the ideal-low run. Contours are at 0.1 and 0.5 g kg−1.

  • View in gallery

    Vertical distribution of the time- and area-averaged vertical velocity (>0 m s−1) for (a),(c) MESO and for (b),(d) LARGER. Here, (a),(b) are for the ideal-high run and the ideal-low run, and (c),(d) are for the ideal-high-whole run and the ideal-low run.

  • View in gallery

    Vertical distribution of the time- and area-averaged heating rate from freezing of cloud liquid for (a) MESO and (b) LARGER.

  • View in gallery

    Wind vector fields (arrows) and contours of condensation rate (solid lines) at (a),(b) 1800 LST 23 Jan and (c),(d) 2200 LST 23 Jan. Here, (a),(c) are for the ideal-high run, and (b),(d) are for the ideal-low run. Contours are at 0.001, 0.002, and 0.003 g kg−1 s−1. Red boxes represent MESO and green boxes in (a),(c) represent the spatial extent of wind fields generated by divergence in MESO around cloud tops. The same green boxes in (a),(c) are displayed in (b),(d), respectively, to compare the divergence and associated wind fields in the ideal-high run to those in the ideal-low run.

  • View in gallery

    Vertical distribution of the time- and area-averaged vertical distribution of condensation rate in 10−5 g kg−1 s−1 at (a),(b) 1800 LST 23 Jan, (c),(d) 2200 LST 23 Jan, and (e) 0000 LST 24 Jan. Here, (a) is for MESO and LARGER in the ideal-high run, and (b),(c) are for MESO and LARGER in the ideal-low run; (d),(e) are for ideal-high and -low runs in LARGER.

  • View in gallery

    Vertical distribution of the area-averaged vertical gradient of potential temperature for LARGER at (a) 2200 LST 23 Jan and (b) 0000 LST 24 Jan.

  • View in gallery

    Wind vector fields (arrows) and contours of cloud liquid mixing ratio (solid lines) at (a),(b) 2200 LST 23 Jan and (c),(d) 0000 LST 24 Jan. Here, (a),(c) are for the ideal-high run, and (b),(d) are for the ideal-low run. For the effective visualization of interactions between vertical components of wind fields and cloud-top height, model-predicted vertical components of wind are multiplied by a factor of 8. Contours are at 0.1, 0.5, 0.7 and 1.0 g kg−1. Red boxes represent MESO and green boxes in (a),(c) represent the spatial extent of wind fields (generated by divergence in MESO) and strong subsidence. The same green boxes in (a),(c) are displayed in (b),(d), respectively, to compare the divergence, associated wind, and subsidence fields in the ideal-high run to those in the ideal-low run.

  • View in gallery

    Time series of difference (high − low) in domain-averaged cumulative precipitation, updraft mass flux, subsidence mass flux, vertical gradient of potential temperature, cumulative condensation minus cumulative evaporation of cloud liquid, and cloud-top height. Here, updraft and subsidence mass fluxes are averaged over a layer between 5 and 12 km.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 48 48 6
PDF Downloads 39 39 7

Effect of Aerosol on Circulations and Precipitation in Deep Convective Clouds

View More View Less
  • 1 NOAA/Earth System Research Laboratory, Chemical Sciences Division, and Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder, Colorado
© Get Permissions
Full access

Abstract

This study examines the effect of a mesoscale perturbation of aerosol on a larger-scale cloud system driven by deep convective clouds. An aerosol-perturbed domain of size 120 km is prescribed in the middle of the larger-scale domain of size 1100 km. Aerosol perturbations in the mesoscale domain result in an intensification of convection in a mesoscale convective system (MCS). This leads to an intensification of the larger-scale circulations, which in turn leads to an intensification of the larger-scale subsidence. While the invigorated convection enhances precipitation in the MCS, the intensified larger-scale subsidence acts to increase the larger-scale stability and thus to suppress convection and precipitation in the larger-scale domain. The suppression of precipitation in the larger-scale domain outweighs the enhancement of precipitation in the mesoscale domain, leading to suppressed precipitation over the entire domain. The ramifications of aerosol perturbations therefore need to be considered on scales much larger than the scale of the perturbation.

Corresponding author address: Seoung Soo Lee, NOAA/Earth System Research Laboratory, 325 Broadway, Boulder, CO 80301. E-mail: seoung.soo.lee@noaa.gov

Abstract

This study examines the effect of a mesoscale perturbation of aerosol on a larger-scale cloud system driven by deep convective clouds. An aerosol-perturbed domain of size 120 km is prescribed in the middle of the larger-scale domain of size 1100 km. Aerosol perturbations in the mesoscale domain result in an intensification of convection in a mesoscale convective system (MCS). This leads to an intensification of the larger-scale circulations, which in turn leads to an intensification of the larger-scale subsidence. While the invigorated convection enhances precipitation in the MCS, the intensified larger-scale subsidence acts to increase the larger-scale stability and thus to suppress convection and precipitation in the larger-scale domain. The suppression of precipitation in the larger-scale domain outweighs the enhancement of precipitation in the mesoscale domain, leading to suppressed precipitation over the entire domain. The ramifications of aerosol perturbations therefore need to be considered on scales much larger than the scale of the perturbation.

Corresponding author address: Seoung Soo Lee, NOAA/Earth System Research Laboratory, 325 Broadway, Boulder, CO 80301. E-mail: seoung.soo.lee@noaa.gov

1. Introduction

Numerical modeling has suggested that aerosol-induced changes in evaporation and freezing of hydrometeors (i.e., changes in latent heat distribution) affect low-level convergence and parcel buoyancy of mesoscale convective system (MCS) driven by deep convective clouds (Andreae et al. 2004; Givati and Rosenfeld 2004; Khain et al. 2005, 2008; Koren et al. 2005; Ekman et al. 2006; van den Heever et al. 2006; Seifert and Beheng 2006; Lynn et al. 2007; van den Heever and Cotton 2007; Rosenfeld et al. 2008; Tao et al. 2007; Lee et al. 2008a,b; Lerach et al. 2008; Khain and Lynn 2009; Fan et al. 2009; Ntelekos et al. 2009; Lee et al. 2010; Storer et al. 2010; Ekman et al. 2011; Morrison and Grabowski 2011). The MCS comprises multiple deep convective clouds and these clouds grow above the freezing level to reach the tropopause. In spite of using different modeling frameworks and microphysical representations, the above-mentioned studies all show that aerosol-induced enhancement in evaporation develops stronger downdrafts and, when stronger downdrafts descend below cloud base and collide with ambient flow, low-level convergence is intensified; however, it is notable that the intensification of the low-level convergence depends on the magnitude of wind shear and the intensification of the convergence may not lead to that of convection when the interaction between wind shear and cold pools results in more upshear-tilted convection. Increased low-level convergence invigorates the mesoscale convection and circulation, leading to more cloud mass and precipitation. Moreover, there are aerosol-induced increases in freezing due to reduced droplet size and thus reduced conversion of cloud liquid to rain and associated increases in cloud-liquid mass as a source of freezing. This enhances parcel buoyancy from freezing and thus further invigorates the mesoscale convection and circulation. This contributes to more cloud mass and precipitation. This enhancement in parcel buoyancy occurs most efficiently when updrafts are not strong enough to carry a significant portion of rain upward.

Mesoscale circulations are building blocks of large-scale circulations observed in systems such as monsoon and the intertropical convergence zone (ITCZ). Hence, we raise the possibility that aerosol-induced changes in mesoscale circulations might extend to larger-scale systems by affecting large-scale circulations, even though the aerosol perturbation is limited in the mesoscale system. This possibility is hinted at in Bell et al. (2008), Khain et al. (2005), and van den Heever and Cotton (2007). They showed a strong correlation between changes in the strength of a source of aerosol pollution and precipitation changes in a convective system removed from, and, thus, not directly affected by the source. Since these studies did not show a detailed analysis of the mechanisms controlling this correlation, the current study extends these studies to identify the mechanisms.

This study aims to examine aerosol–cloud interactions in a mesoscale domain and their impacts on a larger-scale domain that is ~8 times larger than the mesoscale domain. The mesoscale domain enables the simulation of a MCS and the larger-scale domain corresponds to a mesoscale alpha system. This study aims to identify mechanisms of how aerosol-induced changes in mesoscale circulations over the aerosol perturbation region affect the larger-scale circulations and thus clouds and precipitation far removed from the aerosol perturbation.

2. Cloud-system-resolving model

The Goddard Cumulus Ensemble (GCE) model (Tao et al. 2003), which is a two-dimensional nonhydrostatic compressible model, is used here as a cloud system–resolving model (CSRM). The detailed equations of the dynamical core of the GCE model are described by Tao and Simpson (1993) and Simpson and Tao (1993).

The subgrid-scale turbulence used in the GCE model is based on work by Klemp and Wilhelmson (1978) and Soong and Ogura (1980). In their approach, one prognostic equation is solved for the subgrid-scale kinetic energy, which is then used to specify the eddy coefficients. The effect of condensation on the generation of subgrid-scale kinetic energy is also incorporated into the model.

To represent microphysical processes, the GCE model has been adopted to include the double-moment bulk representation of Saleeby and Cotton (2004), which uses bin model–derived lookup tables for hydrometeor collection processes. Hydrometeor size distributions assume gamma basis functions with fixed breadth. Cloud droplet and ice crystal nucleation also mimic a size-resolved approach (Lee et al. 2010).

A Lagrangian scheme is used for sedimentation of hydrometeors, which is transportation of the mixing ratio and number concentration of each class of hydrometeors from any given grid cell to a lower height in the vertical column, following Walko et al. (1995).

The cloud droplet nucleation parameterization of Abdul-Razzak and Ghan (2000, 2002), which is based on Köhler theory, is used to obtain parameterized supersaturation around cloud base and thus droplet nucleation via primary nucleation around cloud base. This parameterization combines the treatment of multiple aerosol types and a sectional representation of size to deal with arbitrary aerosol mixing states and arbitrary aerosol size distributions. The bulk hygroscopicity parameter for each category of aerosol is the volume-weighted average of the parameters for each component taken from Ghan et al. (2001). In applying the Abdul-Razzak and Ghan parameterization, the size spectrum for each aerosol category is divided into 30 bins. For in-cloud or secondary droplet nucleation above cloud base, supersaturation is not parameterized but predicted supersaturation by the CSRM used in this study is directly used. Aerosol particles whose critical supersaturation is smaller than the predicted supersaturation are activated via secondary droplet nucleation in cloud above cloud base.

Lohmann and Diehl’s (2006) parameterizations, taking into account the dependence of ice nuclei (IN) activation on dust and black carbon (BC) aerosol mass concentration, are used for contact, immersion, and condensation–freezing activation of IN.

Secondary production of ice occurs by the Hallet–Mossop process of rime splintering (Hallet and Mossop 1974) and involves 350 ice splinters emitted for every milligram of rimed liquid at −5.5°C. The number of splinters per milligram of rime liquid is linearly interpolated to zero between −3° and −8°C.

The parameterizations developed by Chou and Suarez (1999) for shortwave radiation and by Chou et al. (1999) and Kratz et al. (1998) for longwave radiation have been implemented in the GCE model. The solar radiation scheme includes absorption due to water vapor, CO2, O3, and O2. Interactions among the gaseous absorption and scattering by cloud particles, molecules, and the surface are fully taken into account. Effective size of cloud particles is calculated using predicted cloud particle number and mass by the double-moment scheme. Reflection and transmission of a cloud layer are computed using the δ–Eddington approximation. Fluxes for a composite of layers are then computed using the two-stream adding approximation. In computing thermal infrared fluxes, the k-distribution method with temperature and pressure scaling is used to compute the transmission function.

3. Case description

a. Control run

For this study, 2-day two-dimensional simulations of an observed MCS are performed. The MCS was observed during the Tropical Warm Pool International Cloud Experiment (TWP-ICE) [1200 LST (local solar time) 23 January–1200 LST 25 January 2006] campaign in Darwin, Australia (12.47°N, 130.85°W). Henceforth, the simulation of this observed case is referred to as “the control run.”

Periodic boundary conditions are set on horizontal boundaries. This study focuses on cloud-scale interactions but not on interactions between clouds and large scales much larger than the model domain where clouds are generated in this study. Hence, observed heat and moisture fluxes are imposed on the model surface and there are no two-way interactions between surface fluxes and clouds. To prevent the reflection of gravity or sound waves from the model top, a damping layer of 5-km depth is applied near the model top.

The TWP-ICE observations provide initial humidity, temperature, and large-scale forcings of humidity and temperature. These observations are processed by variational analysis described in Xie et al. (2010). Vertical profiles of the initial specific humidity and potential temperature applied are shown in Fig. 1. Up to the top of the planetary boundary layer (PBL) around 2 km, potential temperature and humidity do not change significantly, whereas above the PBL top, potential temperature (humidity) increases (decrease) significantly. Large-scale advective tendencies for potential temperature and specific humidity were interpolated from 3-hourly analyses of balloon soundings at every time step. The time- and domain-averaged large-scale forcings of potential temperature and specific humidity are 4.8 K day−1 and 4.3 g kg−1 day−1. The wind in the west–east direction u is westward at most altitudes except for those in the PBL. Hence, the time- and domain-averaged wind is westward with a magnitude of 4.8 m s−1. The details of the procedure for applying large-scale forcing are described in Donner et al. (1999) and are similar to the method proposed by Grabowski et al. (1996). Horizontal momentum was damped to observed values, following Xu et al. (2002).

Fig. 1.
Fig. 1.

Vertical profiles of initial potential temperature and water vapor mixing ratio.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

It is assumed that there are five chemical components of aerosol: dust, sulfate, organics, BC, and sea salt. Depending on predicted aerosol mass within cloud, the total aerosol number for each aerosol species varies and is reset to the background value at all levels outside cloud. Within clouds, aerosol is advected, diffused, and depleted by nucleation (i.e., nucleation scavenging). In this study, we did not include the aerosol removal by precipitation that captures aerosol (i.e., impaction scavenging) since it is known that impaction scavenging only accounts for ~10% or less of the total aerosol removal by scavenging (Hobbs 1993). Initially aerosol mass mixing ratio is everywhere set equal to its background value. Background aerosol number concentrations for all aerosol species in each aerosol size mode are assumed not to vary during time integration.

The control run adopts the background aerosol profiles that are extracted from the Aerosol and Chemical Transport in Tropical Convection (ACTIVE) program (Vaughan et al. 2008) with which the TWP-ICE was coordinated. The size distribution and number concentration of background aerosol is calculated following the methodology described in Fridlind et al. (2009) and aerosol size distributions shown in Fig. 4 in Fridlind et al. (2009) are applied to background aerosol. The number concentration of the background aerosol varies with height as shown in Fridlind et al. (2009). Here, the mode diameter and standard deviation of the distributions, as well as the partitioning among modes for background aerosol follow Fridlind et al. (2009) and are assumed not to vary spatiotemporally.

For the control run, the horizontal domain length is set at 2200 km in the west–east direction while the vertical domain length is set at 20 km to cover the troposphere and the lower stratosphere. The large-scale forcings of temperature and humidity are applied to all parts of this 2200-km domain. However, to minimize the effect of boundary conditions on clouds and associated circulations, an initial water vapor perturbation is applied only to a part of the horizontal domain between 550 and 1650 km. Hence, clouds are only formed over the part of the domain between 550 and 1650 km, that is, a distance of 1100 km. The water vapor perturbations vary in the horizontal but are constant throughout the lowest 1.5 km in each column of the model. The perturbations are horizontally random, generated from a distribution between ±2 g kg−1. These perturbations are similar to those employed by Donner et al. (1999) and are chosen to be random so as not to impose organized structure on the convection when it develops. Also, initial wind fields and the damping of simulated wind momentum to the observed one are imposed only on the part of the domain between 550 and 1650 km and initial wind fields in the rest of the domain are set at 0. The periodic boundary conditions mean that circulations can exit the domain on one side and re-enter the domain on the other side. This can disrupt the circulation and thus makes it difficult to isolate the effect of aerosol on the circulation. The use of this larger 2200-km domain with cloudy area embedded inside it prevents winds from reaching the boundary and thus keeps them within the domain throughout the simulations. Simulation results described in the following sections show that winds are indeed kept within the domain as intended. This facilitates the isolation of the effect of aerosol on circulation by mitigating the effect of boundary condition on circulation.

All analyses (including averaging) of results in this paper are performed and displayed only over the 1100-km domain between 550 and 1650 km where clouds form. Henceforth, the west end of this domain is set to 0 km and the east end to 1100 km and the horizontal domain refers to this 1100-km domain unless otherwise stated. The horizontal domain length is ~6–7 times larger than that adopted for a mesoscale cloud system simulated in Lee et al. (2008a,b), Lee et al. (2009), and Lee et al. (2010). Hence, this size of domain enables the examination of the effect of aerosol–cloud interactions on clouds in a larger-scale domain. The horizontal gird length Δx is 500 m while the constant vertical grid length Δz is 200 m.

This study focuses on aerosol effects on the nucleation of cloud particles and thereby cloud microphysical and radiative properties and, thus, does not take into account aerosol direct effects on radiation.

b. Idealized cases

In this study, we limit our interests to the effect of aerosol on latent heat distribution and circulations, which is initiated by the effect of aerosol on freezing as described in Andreae et al. (2004), Khain et al. (2005), van den Heever et al. (2006), Rosenfeld et al. (2008), Lee et al. (2010), van den Heever et al. (2011), and Seifert et al. (2011). Thus, we do not concern ourselves with the effect of wind shear on aerosol–cloud interactions as identified by Lee et al. (2008b) who showed that wind shear caused an aerosol-induced increase in the number of secondary clouds. Instead, we focus on the effect of aerosol on circulations via its effect on parcel buoyancy in individual clouds. As mentioned in the introduction, aerosol-induced increases in freezing enhance parcel buoyancy and this invigorates convection and circulation; however, this increase in parcel buoyancy can be offset by aerosol-induced increase in loading by cloud particles and the invigoration can be prohibited in case the loading effect outweighs the freezing effect. This is because an increase in aerosol suppresses autoconversion and thus increases cloud liquid available for its transportation to the freezing level.

To isolate the effect of aerosol on the mesoscale and larger-scale circulations through the effect of aerosol on freezing, the control run is repeated by setting the background wind field to zero to remove wind shear throughout the domain. Since background wind fields and associated wind shear are known to predominantly affect cloud evolution as compared to cloud-generated wind shear (e.g., Weisman and Klemp 1982; Houze 1993), it is expected that this will exclude most of the effect of wind shear on cloud evolution with good confidence. This experiment is referred to as “the ideal-low run.” The ideal-low run is repeated but with one difference: aerosol mass and number are increased by a factor of 10 between 490 and 610 km in the horizontal domain but is kept the same as in the ideal-low run in the rest of the domain. We thus have a 120-km region subject to a mesoscale aerosol perturbation, embedded in a larger-scale domain, whose horizontal scale is 1100 km. This repeated simulation is referred to as “the ideal-high run.” Henceforth, the domain between 490 and 610 km is referred to as “MESO” and the entire domain except for MESO is referred to as “LARGER.” The repeated ideal-high runs with aerosol concentration increased by a factor 2, 4, and 7 (only in MESO) as compared to that in MESO in the ideal-low run show that the qualitative nature of findings between the ideal-low and -high runs does not depend on the magnitude of aerosol perturbation in MESO. Figure 2 shows the horizontal distribution of the initial aerosol number averaged over the PBL for the ideal-low and ideal-high runs. A comparison between the ideal-low and ideal-high runs identifies the effect of the mesoscale aerosol perturbation on the larger-scale cloud system in LARGER as well as on the mesoscale cloud system in MESO.

Fig. 2.
Fig. 2.

Horizontal distribution of the domain-averaged background aerosol number concentration over the PBL for the ideal-high and -low runs.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

A control run, in which the observed ambient wind field is used, is performed to identify the role of the wind field on the cloud system and to test the behavior of the new CSRM. However, this identification is not the main goal of this study and, thus, its analysis and implications are described briefly. Table 1 summarizes simulations in this study.

Table 1.

Summary of simulations.

Table 1.

4. Results

a. Control

Precipitation rate and cumulative precipitation

Figure 3 depicts the time series of the area-mean precipitation rate smoothed over 3 h in the control run. The comparison of precipitation between observation and the control run in Fig. 3 demonstrates that precipitation is simulated reasonably well because the water budget is well constrained mostly by large-scale forcings of humidity and temperature. Also, the simulated averaged liquid water path (LWP) and cloud fraction (CF) are within ~10% of the observed averaged LWP (52 g m−2) and CF (65%); LWP is observed by the microwave radiometer and corrections are made to eliminate the contamination by raindrops on the instrument as described in Liljegren (1994). This indicates that model used here is able to simulate precipitation with a good confidence, and is therefore a suitable framework in which to explore the goals of the study.

Fig. 3.
Fig. 3.

Time series of the area-averaged precipitation rate for observation and the control run.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

b. Idealized simulations

1) Precipitation rate and cumulative precipitation

The precipitation event simulated here is driven by deep convective clouds as shown in Fig. 4, which depicts contours of mixing ratios of cloud liquid obtained around the occurrence of maximum precipitation rate in the ideal-low run.

Fig. 4.
Fig. 4.

Contours of cloud liquid mixing ratio (g kg−1) at the time of the occurrence of maximum precipitation rate for the ideal-low run. Contours are at 0.1 and 0.5 g kg−1.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

Domain-averaged cumulative precipitation at the last time step is 80.81 and 82.66 mm in the ideal-high and ideal-low runs, respectively. In the absence of wind shear, both ideal runs result in smaller precipitation than the control run; the control-run cumulative precipitation is 85.12 mm. As shown in Lee et al. (2008b), wind shear facilitates evaporation of hydrometeors and thus intensifies downdrafts, cold-pool, and low-level convergence, which leads to the development of secondary clouds and thus enhances precipitation. This is closely linked to the fact that wind shear acts to organize convection with downdraft–updraft separation and, thus, to strengthen the local cold pool circulations (Rotunno et al. 1988; Liu and Moncrieff 1996; Moncrieff and Liu 1999; Tompkins 2001). Hence, the cumulative number of convective cores at the last time step in the control run (where background wind shear is considered) is 126 195, which is larger than 104 725 in the ideal-low run. A convective core satisfies at least one of the following three conditions: 1) maximum cloud draft strength wmax is larger than the average over grid columns within 4 km with w > 1 m s−1, 2) wmax > 3 m s−1, or 3) precipitation rate exceeds 25 mm h−1. Here, w represents vertical wind velocity.

The division of the domain into MESO and LARGER facilitates exploration of how the aerosol perturbation in the mesoscale domain (i.e., MESO) affects a MCS in that domain and then how subsequent changes in the mesoscale circulations affect the larger-scale circulations and clouds in LARGER.

In MESO (LARGER), the domain-averaged cumulative precipitation at the last time step is 91.44 (79.51) and 83.97 (82.50) mm for the ideal-high and ideal-low runs, respectively. Thus precipitation is enhanced in MESO and suppressed in LARGER in spite of the aerosol perturbation being confined to MESO. The suppression dominates the enhancement, which leads to smaller precipitation in the ideal-high run than in the ideal-low run over the whole domain.

Microphysical processes leading to the aerosol-induced differences in precipitation are examined by obtaining differences in the domain-averaged cumulative sources and sinks of the sum of precipitable hydrometeors between the ideal-high and ideal-low runs (high − low). These differences are obtained for each of MESO and LARGER. For this, production equations for the sum of precipitable hydrometeors are integrated over each of MESO and LARGER for the duration of the simulations. The time- and domain-average tendency is zero since the storage of the hydrometeors is zero at the end of simulation. Among the sources and sinks, autoconversion and terms associated with accretion of cloud liquid predominantly account for precipitation differences to yield the following approximate difference equation for both MESO and LARGER:
e1
where volume and area integrations are denoted by 〈 ⋅ 〉 and ||⋅||, respectively:
e2
Here, Lx is the domain length, which is 120 and 980 km for MESO and LARGER, respectively. In Eq. (1), the mixing ratios of cloud liquid, cloud ice, aggregates, rain, and hail are represented by qc, qi, qa, qr, and qh, respectively, and Au and A represent autoconversion and accretion, respectively. The quantity Pr is precipitation. Notation for terms in the budget equations obeys the following conventions: the variable before the semicolon in each term indicates the quantity whose mixing ratio is changed by the source or sink. Following the semicolon, quantities that merge or separate in the source or sink are indicated by a “|” between them. A single variable following a semicolon indicates a quantity whose mixing ratio is changed by a phase transition; this last convention is used in the following Eq. (3).

The terms on the right hand side of Eq. (1) are differences (high − low) in autoconversion, accretion of cloud liquid by rain to form rain, accretion of cloud liquid by hail to form hail, accretion of cloud liquid by cloud ice to form cloud ice, accretion of cloud liquid by aggregates to form hail, accretion of cloud liquid by cloud ice to form hail, and accretion of cloud liquid by aggregates to form aggregates, respectively, between the ideal-high and -low runs. The sources and sinks excluded from Eq. (1) contribute about one order of magnitude less to the differences in precipitation than sources retained in Eq. (1). Note that excluded sources and sinks include those of ice hydrometeors, which are deposition and sublimation. Budget numbers for Eq. (1) are shown in Table 2 for each of MESO and LARGER and for the whole domain, that is, MESO + LARGER.

Table 2.

Domain-averaged differences in the accumulated sources and sinks of precipitation, retained in the approximated Eqs. (1) and (3), and the domain-averaged differences in the accumulated evaporation of rain between the ideal-high and ideal-low runs.

Table 2.

In MESO, the decrease in autoconversion is outweighed by the increase in accretion of cloud liquid, enabling precipitation enhancement [see Eq. (1) and Table 2]. However, in LARGER with increasing aerosol in MESO, autoconversion and accretion both decrease with increasing aerosol. There are no processes that are able to offset the decrease in autoconversion in LARGER (Table 2). This leads to precipitation suppression in LARGER and over the whole domain [see Eq. (1) and Table 2].

Whether the change in accretion of cloud liquid is able to offset the autoconversion decrease is determined by how much cloud liquid mass increases with increasing aerosol. This is because cloud liquid is the source of accretion. To examine the source of cloud liquid, budget terms controlling the evolution of cumulative cloud liquid mass (i.e., ) are added to those in the production equation for the sum of precipitable hydrometeors. Terms associated with cloud liquid in the production equation for the sum of precipitable hydrometeors are canceled out by this addition. Then, it is found that differences in condensation and evaporation of cloud liquid are 1–3 orders of magnitude larger than the other terms (Khain et al. 2008; Lee et al. 2008a,b); also, note that condensation and evaporation of cloud liquid explain the precipitation variation much more significantly than evaporation of rain by one to two orders of magnitude. Therefore, the difference in precipitation is approximated as follows for both MESO and LARGER:
e3
The qυ, C, and E represent water vapor mixing ratio, condensation, and evaporation, respectively. The terms on the right-hand side of Eq. (3) are differences (high − low) in condensation and evaporation of cloud liquid, respectively. Budget numbers for Eq. (3) are also shown in Table 2.

In the event that both condensation and evaporation increase with increasing aerosol, the combination of Eqs. (1) and (3) indicates that whether precipitation increases or decreases with increasing aerosol is determined by whether an increase in the production of cloud liquid by condensation (leading to an increase in accretion) is larger than an increase in the loss of cloud liquid by evaporation (leading to a decrease in accretion). In MESO, the increased condensation of cloud liquid is greater than the increased evaporation of cloud liquid, resulting in the greater ideal-high precipitation than the ideal-low precipitation (Table 2). This is despite smaller precipitation efficiency in the ideal-high run (41%) than in the ideal-low run (52%) at the last time step in MESO. Precipitation efficiency is defined as cumulative precipitation normalized by cumulative condensation. In LARGER, condensation and evaporation both decrease with increasing aerosol. Decrease in condensation reduces cloud liquid as a source of evaporation, leading to decrease in evaporation; the cause of decrease in condensation is described in the following section 4b(2). Also, condensation decreases more than does evaporation (Table 2). In LARGER, precipitation efficiency at the last time step is nearly identical between the ideal-high (48%) and -low (47%) runs with only ~1% difference. Hence, the ratio of cumulative evaporation to cumulative condensation at the last time step is similar between the two runs based on Eq. (3); note that Eq. (3) shows that condensed water not converted into precipitation is evaporated. Cloud precipitation properties (represented by precipitation efficiency and the ratio of cumulative evaporation to cumulative condensation at the last time step) do not change much between the ideal-high and -low runs in LARGER. With the similar fraction of condensed water that is evaporated between the runs, decreased cumulative condensation in the ideal-high run leads to the situation where decrease in cumulative condensation is larger than decrease in cumulative evaporation at the last time step in LARGER. The fraction of condensed water that is evaporated is ~50% for both the ideal-high and -low runs. This leads to decrease in cumulative evaporation that is roughly 50% of decrease in cumulative condensation at the last time step in the ideal-high run. This means that decrease in condensation and, thus, in cloud liquid as a source of accretion of cloud liquid (thus leading to decrease in accretion of cloud liquid), outweighs increase in cloud liquid (leading to increase in accretion of cloud liquid) because of the decrease in evaporation of cloud liquid. As a result, there is a decrease in accretion of cloud liquid with increasing aerosol in LARGER. Hence, in LARGER, at high aerosol, decrease in accretion contributes to the precipitation suppression in tandem with decrease in autoconversion, leading to precipitation suppression at high aerosol in LARGER (Table 2).

While the aerosol perturbation acts to enhance precipitation in places where the perturbation occurs, it acts to suppress precipitation in places not directly affected by the perturbation. Thus the effect of an aerosol perturbation of limited spatial extent propagates into a domain beyond that of the perturbation.

2) Dynamic aspects

In MESO, there is an intensification of the vertical velocity (>0 m s−1) with increasing aerosol throughout altitudes (Fig. 5a). A comparison between runs with no ice physics [described in section 4b(3)] and with ice physics shows that this aerosol-induced intensification of vertical velocity in MESO is caused by an increase in parcel buoyancy because of an increase in latent heat from cloud particle freezing (Givati and Rosenfeld 2004; Khain et al. 2005; Koren et al. 2005; Ekman et al. 2006; van den Heever et al. 2006; van den Heever and Cotton 2007; Rosenfeld et al. 2008; Tao et al. 2007; Lee et al. 2008a,b; Lerach et al. 2008; Khain and Lynn 2009; Fan et al. 2009; Ntelekos et al. 2009; Lee et al. 2010; Storer et al. 2010; Ekman et al. 2011; van den Heever et al. 2011; Seifert et al. 2011). Delayed autoconversion as shown in Table 2 enhances the amount of cloud liquid below the freezing level, which enables more transportation of cloud liquid to levels above the freezing level and thus the increased freezing of cloud liquid mostly via riming of liquid particles onto solid particles throughout altitudes above the freezing level as shown in Fig. 6a; the intensification of the vertical velocity leads to an increase in condensation in MESO. This increase in condensation establishes a positive feedback with the vertical velocity to further enhance condensation, which results in an increase in condensation larger than the increase in evaporation. This leads to precipitation enhancement with increasing aerosol in MESO (Table 2).

Fig. 5.
Fig. 5.

Vertical distribution of the time- and area-averaged vertical velocity (>0 m s−1) for (a),(c) MESO and for (b),(d) LARGER. Here, (a),(b) are for the ideal-high run and the ideal-low run, and (c),(d) are for the ideal-high-whole run and the ideal-low run.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

Fig. 6.
Fig. 6.

Vertical distribution of the time- and area-averaged heating rate from freezing of cloud liquid for (a) MESO and (b) LARGER.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

In LARGER, the vertical velocity is smaller in the ideal-high run than in the ideal-low run throughout altitudes (Fig. 5b). This is due in part to the absence of the increase in cloud liquid freezing in the ideal-high run in LARGER in most of altitudes above the freezing level as shown in Fig. 6b. This absence of the increase in freezing is mainly because there is no prescribed increase in background aerosol in the ideal-high run in LARGER. Thus, there is no substantial suppression of autoconversion as shown in Table 2 and thus no substantial increase in cloud liquid below the freezing level. This contributes to no increase in cloud liquid transported above the freezing level in the ideal-high run in LARGER. While explaining why the ideal-high run does not have higher vertical velocity than the ideal-low run, the absence of the increase in freezing is not able to explain why the ideal-high run has lower vertical velocity than the ideal-low run in LARGER. To explain the smaller vertical velocity in the ideal-high run than in the ideal-low run, aerosol-induced changes in dynamics and circulations and associated changes in latent-heat distribution are examined. Condensation and associated latent heating predominantly control the dynamic response as well as the precipitation response to aerosol. The condensation rate is about 1–2 orders of magnitude larger than the other sources of latent heating, which are deposition and freezing in most of cloud layers (Tao et al. 2006; Lee et al. 2008a, 2010). Thus, henceforth, the focus of this paper is on condensation to explain the response of latent heat distribution and circulation to aerosol.

Figure 7 shows condensation rate and wind fields and Fig. 8 shows the vertical distribution of the area-averaged condensation rate at various stages of the simulation. Because of increasing heating and vertical velocity from freezing in MESO between 490 and 610 km in the ideal-high run, condensation becomes significantly larger in MESO than in LARGER for the ideal-high run at 1800 LST 23 January throughout the cloud layer (Figs. 7a and 8a). Red boxes in Fig. 7 represent MESO. Figure 7a (ideal high) shows that stronger updrafts in the invigorated convection mostly because of the increased condensation result in divergence around cloud tops between ~(12–16) km in MESO. This divergence generates wind fields spreading horizontally and their spatial extent is represented by green boxes in the ideal-high runs in Fig. 7. The same green boxes in Figs. 7a and 7c (for the ideal-high runs) are displayed in Figs. 7b and 7d (for the ideal-low runs), respectively, to compare the divergence fields in MESO and associated spreading wind fields in the ideal-high run to those in the ideal-low run. Divergence originating from aerosol-invigorated convection in MESO is much stronger than that in LARGER since convection (and thus the divergence it powers) is weaker in LARGER than in MESO. This enables the divergence from MESO to form wind fields in the green box that spread out to LARGER (Fig. 7a). The wind fields generated by divergence in MESO are to the west and the east relative to MESO (Fig. 7a; see the green box). The west branch of these wind fields reaches ~300 km and the east branch ~800 km. As time progresses, the west (east) branch of divergence-generated wind fields spread out further to ~100 (1000) km as shown in the green box in Fig. 7c at 2200 LST 23 January. In the ideal-low run, due to no increase in aerosol and thus in freezing in both MESO and LARGER, there are no significant differences in condensation and thus convection between MESO and LARGER up to 2200 LST 23 January throughout the cloud layer as shown in Figs. 7b, 7d, 8b, and 8c. Hence, there are no strong convection and thus divergence in MESO in the ideal-low run. This results in no wind fields (generated by divergence in MESO) spreading to LARGER around cloud tops in the ideal-low run up to 2200 LST 23 January as shown in the green boxes in Figs. 7b and 7d. The divergence-generated wind fields turn into strong subsidence nearly over the whole LARGER to complete circulation between ~10 and ~12 km in the ideal-high run at 2200 LST January, as shown in the lower-half part of the green box in Fig. 7c.

Fig. 7.
Fig. 7.

Wind vector fields (arrows) and contours of condensation rate (solid lines) at (a),(b) 1800 LST 23 Jan and (c),(d) 2200 LST 23 Jan. Here, (a),(c) are for the ideal-high run, and (b),(d) are for the ideal-low run. Contours are at 0.001, 0.002, and 0.003 g kg−1 s−1. Red boxes represent MESO and green boxes in (a),(c) represent the spatial extent of wind fields generated by divergence in MESO around cloud tops. The same green boxes in (a),(c) are displayed in (b),(d), respectively, to compare the divergence and associated wind fields in the ideal-high run to those in the ideal-low run.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

Fig. 8.
Fig. 8.

Vertical distribution of the time- and area-averaged vertical distribution of condensation rate in 10−5 g kg−1 s−1 at (a),(b) 1800 LST 23 Jan, (c),(d) 2200 LST 23 Jan, and (e) 0000 LST 24 Jan. Here, (a) is for MESO and LARGER in the ideal-high run, and (b),(c) are for MESO and LARGER in the ideal-low run; (d),(e) are for ideal-high and -low runs in LARGER.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

We now explore the manifestation of aerosol-induced changes in the heating profiles. Subsidence warms the air by transporting larger θ from higher altitudes. Here, θ is potential temperature. Figure 9 depicts the vertical distribution of the vertical gradient of potential temperature dθ/dz. Figure 9a indicates that the strong subsidence (warming the air) induces stronger stability between ~10 and ~12 km in the ideal-high run than in the ideal-low run at 2200 LST 23 January in LARGER. This increased stability suppresses clouds and thus lowers cloud-top heights down to ~10 km in LARGER in the ideal-high run, while cloud-top height is maintained around 12 km in LARGER in the ideal-low run with no strong subsidence at 2200 LST 23 January (Figs. 10a and 10b); Fig. 10 depicts cloud and wind fields. In Fig. 10, red and green boxes represent MESO and the spatial extent of wind fields and subsidence powered by the intensified convection in MESO (demonstrated by much larger cloud-liquid mixing ratio in the ideal-high run than in the ideal-low run in MESO), respectively. The identical green boxes in Figs. 10a and 10c (for the ideal-high runs) are displayed in Figs. 10b and 10d (for the ideal-low runs), respectively, for comparisons of subsidence and associated wind fields between the ideal-high run and the ideal-low run. The suppressed clouds have suppressed condensation in the ideal-high run in LARGER at 2200 LST 23 January (Fig. 8d).

Fig. 9.
Fig. 9.

Vertical distribution of the area-averaged vertical gradient of potential temperature for LARGER at (a) 2200 LST 23 Jan and (b) 0000 LST 24 Jan.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

Fig. 10.
Fig. 10.

Wind vector fields (arrows) and contours of cloud liquid mixing ratio (solid lines) at (a),(b) 2200 LST 23 Jan and (c),(d) 0000 LST 24 Jan. Here, (a),(c) are for the ideal-high run, and (b),(d) are for the ideal-low run. For the effective visualization of interactions between vertical components of wind fields and cloud-top height, model-predicted vertical components of wind are multiplied by a factor of 8. Contours are at 0.1, 0.5, 0.7 and 1.0 g kg−1. Red boxes represent MESO and green boxes in (a),(c) represent the spatial extent of wind fields (generated by divergence in MESO) and strong subsidence. The same green boxes in (a),(c) are displayed in (b),(d), respectively, to compare the divergence, associated wind, and subsidence fields in the ideal-high run to those in the ideal-low run.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

Figure 10c shows that subsidence penetrates further down to ~(8–9) km in the ideal-high run at 0000 LST 24 January (see the green box) Fig. 9b shows that this further penetration of subsidence increases stability in the ideal-high run further down to ~8 km. This results in ~4 km lower-cloud-top height in the ideal-high run than in the ideal-low run at 0000 LST 24 January in LARGER as shown in Figs. 10c and 10d. This explains the weakened updrafts (as shown in Fig. 5b) and leads to further suppression of condensation in the ideal-high run in LARGER as shown in a comparison of condensation in LARGER between the ideal-high and -low runs in Fig. 8e. The decreased condensation results in precipitation suppression in the ideal-high run in LARGER, which in turn results in precipitation suppression in the ideal-high run in the whole domain (Table 2).

Figure 11 shows the time series of the domain-averaged differences (high − low) in terms associated with the precipitation suppression in LARGER. Subsidence becomes persistently stronger in the ideal-high run than in the ideal-low run at ~2130 LST 23 January. Here, subsidence is averaged between 5 and 12 km where the effect of subsidence on the suppression of cloud-top height is strong (Fig. 9). Thereafter the stability (dθ/dz) starts to be persistently larger in the ideal-high run. The ideal-low run does not experience the same degree of stabilization and cloud-top height becomes larger than that in the ideal-high run around 2200 LST 23 January. The suppressed convection leads to reduced updrafts (see also Fig. 5b) and thus reduced condensation. The reduced condensation at high aerosol leads to the negative difference (high − low) in the domain-averaged cumulative condensation whose magnitude is larger than the negative difference in the domain-averaged cumulative evaporation around 0300 LST 24 January (thin blue line in Fig. 11). This means that in LARGER, accretion of cloud liquid decreases due to the decrease in condensation as a source of accretion of cloud liquid with increasing aerosol. This together with suppressed autoconversion leads to precipitation suppression in the ideal-high run, starting around 0400 LST 24 January (thick blue line in Fig. 11). Also, it can be seen that the integrated difference in updraft mass fluxes over the entire simulation period is nearly identical to that in downdraft mass fluxes. This is associated with the fact that downdrafts and updrafts mass fluxes balance each other nearly exactly in each of the ideal-high and -low runs, which conserves air mass in the 1100-km domain reasonably well. This time sequence demonstrates how an aerosol perturbation within MESO induces in LARGER increased subsidence, increased stability, and suppressed condensation, and, finally, reduced precipitation.

Fig. 11.
Fig. 11.

Time series of difference (high − low) in domain-averaged cumulative precipitation, updraft mass flux, subsidence mass flux, vertical gradient of potential temperature, cumulative condensation minus cumulative evaporation of cloud liquid, and cloud-top height. Here, updraft and subsidence mass fluxes are averaged over a layer between 5 and 12 km.

Citation: Journal of the Atmospheric Sciences 69, 6; 10.1175/JAS-D-11-0111.1

3) Sensitivity test

The ideal-high run is repeated by increasing the aerosol number in the whole domain (i.e., not only in MESO but also in LARGER) relative to the ideal-low run. This repeated run is referred to as “the ideal-high-whole run” (Table 1). By increasing aerosol over the entire domain, the difference in the aerosol perturbation between MESO and LARGER in the ideal-high-whole run disappears. By comparing differences in results between the ideal-high-whole and ideal-low runs to those between the ideal-high and ideal-low runs, we can see how spatial differences in the aerosol perturbation affect the response of precipitation to the perturbation.

The increase in buoyancy and thus in updraft and cloud-top height (associated with the invigoration of convection) due to increasing freezing occurs across the whole domain since background aerosol increases throughout the whole domain in the ideal-high-whole run; see Figs. 5c and 5d for the difference in updrafts between the ideal-high-whole run and the ideal-low run and the averaged cloud-top height is 10.1 (9.9) and 9.5 (9.2) km in MESO (LARGER) for the ideal-high-whole run and the ideal-low run, respectively. Hence, there are no substantially stronger or weaker cells and thus the strength of divergence and associated wind fields from each of the cells do not vary significantly among cells in the ideal-high-whole run. This leads to no suppression of convective cells by substantially increased intensity of divergence, associated wind fields, and subsidence from invigorated cells by aerosol. This in turn leads to larger precipitation in the ideal-high-whole run than in the ideal-low run throughout the whole domain. Domain-averaged difference (ideal-high-whole − ideal-low) in cumulative precipitation is 6.92 and 6.10 mm for MESO and LARGER, respectively. This is in contrast to the findings in the previous sections where the aerosol increase is not omnipresent, indicating that there should be a spatial variation of the aerosol perturbation to enable the suppression of cells by subsidence.

To isolate the effect of aerosol-induced increase in freezing on results here, the ideal-high and -low runs are repeated with latent heat release from freezing turned off. However, in these repeated runs, microphysical formation of ice particles is allowed. These repeated runs are referred to as the ideal-high-no-freeze run and the ideal-low-no-freeze run. It is found that there are no invigoration of clouds in MESO and associated larger-scale circulations in LARGER in the ideal-high-no-freeze run. Owing to no invigoration and associated feedbacks among aerosol, microphysics, and dynamics, just suppressed autoconversion with increasing aerosol leads to suppressed precipitation in MESO. Because of no suppression of clouds by subsidence (driven by stronger convection in MESO), precipitation is much less suppressed in LARGER with increasing aerosol. Domain-averaged difference (ideal-high-no-freeze − ideal-low-no-freeze) in cumulative precipitation is −2.13 and −0.34 mm for MESO and LARGER, respectively. These simulations with no latent heat release from freezing demonstrate that it is aerosol effects on latent heat release from freezing that induce invigoration of clouds and precipitation enhancement in MESO and larger-scale circulations and subsidence (which suppress precipitation further) in LARGER.

5. Summary and conclusions

This study shows that the effect of mesoscale aerosol perturbations on clouds can extend into a larger domain through aerosol-induced changes in the larger-scale circulations as suggested by the satellite study by Bell et al. (2008).

In other words, these changes in the large-scale circulations enable the effect of aerosol to be teleconnected to cloud systems away from the aerosol pollution. When aerosol increases in the mesoscale domain, it invigorates the MCS in the mesoscale domain by enhancing latent heat from the freezing of cloud liquid and thus updrafts. This results in precipitation enhancement over the mesoscale domain. These enhanced updrafts in turn induce enhanced intensity of divergence of air around cloud tops in the mesoscale domain. Horizontally spreading airflows (generated by divergence in the mesoscale domain) eventually descend over the larger-scale domain. More divergence of air with increasing aerosol in the mesoscale domain leads to stronger descent in the larger-scale domain. This stronger subsidence increases stability, which suppresses clouds and thus precipitation in the larger-scale domain. The net result is a suppression of precipitation with increasing aerosol over the entire model domain.

Numerous previous studies of MCSs have demonstrated features of aerosol–cloud interactions that are not seen in simulations of a single cloud system in a small-scale domain (e.g., Andreae et al. 2004; Givati and Rosenfeld 2004; Khain et al. 2005; Koren et al. 2005; Ekman et al. 2006; van den Heever et al. 2006; Seifert and Beheng 2006; Lynn et al. 2007; van den Heever and Cotton 2007; Rosenfeld et al. 2008; Khain et al. 2005, 2008; Tao et al. 2007; Lee et al. 2008a,b, 2010; Lee and Feingold 2010; van den Heever et al. 2011; Seifert et al. 2011). The increasing aerosol affects low-level convergence of convective clouds in the MCSs and this changes the dynamic, thermodynamic, and microphysical development of subsequent secondary clouds in the presence of wind shear. The effect of the changes in subsequent secondary clouds on MCSs is shown to be substantiated ~(12–18) h after the beginning of simulations (Lee et al. 2008a,b; Lee and Feingold 2010). We are not able to see this effect in simulations of a single cloud where the simulation of secondary clouds is not viable and the simulation period is generally shorter than a few hours. However, this study demonstrates that there are also different features of aerosol–cloud interactions over the larger-scale domain that we have not been able to see in simulations of MCSs. This is because mesoscale domains for the MCSs are not able to resolve aerosol-induced changes in the larger-scale circulations and subsidence. These changes in circulations and associated changes in precipitation in the larger-scale domain suggest that we need to consider a much larger domain for a better assessment of the effect of pollution on precipitation.

Assessment of the effect of pollution on precipitation only in a domain where pollution takes place can be misleading since results here show the different response of precipitation to aerosol between MESO and LARGER. This is in line with Wang and Feingold (2009), although Wang and Feingold (2009) simulated stratocumulus clouds trapped in the PBL in a mesoscale domain. They showed that aerosol-induced changes in circulations in a part of domain with higher aerosol concentrations (high-aerosol domain) affected clouds in the other parts of domain with lower aerosol concentrations (low-aerosol domain). This leads to a different sign of aerosol-induced changes in cloud liquid content between the high-aerosol domain and the low-aerosol domain. Results here are also in line with Lynn et al. (2005), Rosenfeld et al. (2007), Zhang et al. (2007), and Khain et al. (2010). Lynn et al. (2005), van den Heever et al. (2011), and Seifert et al. (2011) showed that, in spite that aerosol tended to intensify squall line, the total precipitation over large computational area remained unchanged or slightly decreased. Rosenfeld et al. (2007), Zhang et al. (2007), and Khain et al. (2010) showed that aerosol-intensified convection at the periphery of a tropical cyclone competed with that in the eyewall convection, which led to the weakening of the tropical cyclone and to a decrease in precipitation over the entire region of the tropical cyclone.

As shown by Lee and Feingold (2010), the microphysical pathways in clouds tend to compensate in ways that result in a small overall effect and represent further evidence of a buffered aerosol–cloud–precipitation system (Stevens and Feingold 2009). This study also exhibits buffering of a different kind: competition between the effect of aerosol on precipitation in the mesoscale system and that in the larger-scale system in ways that result in the smaller aerosol-induced precipitation difference over the whole domain than that in each of the mesoscale and larger-scale domains. In addition to the microphysical buffering found in Lee and Feingold (2010) and Lee et al. (2008a,b), this buffering associated with the effect of aerosol on larger-scale circulations further makes it difficult to discern a clear signal of the aerosol effect on precipitation in convective systems. Hence, buffering here makes it more difficult to separate aerosol-induced changes in cloud systems from meteorological influences.

The identical surface fluxes from observation are prescribed for the ideal-high and -low runs. Therefore, the surface fluxes do not contribute to the different precipitation response to aerosol perturbation between MESO and LARGER. In this study, we focused on how aerosol affects clouds and precipitation for an identical observed net heat and moisture supplied to or removed from the domain by large-scale flow and surface fluxes in deep convective clouds. Although feedbacks from differences in clouds onto the large-scale flow and surface fluxes cannot be captured by this design, this isolates interactions between aerosol, microphysics, and dynamics and enables the identification of microphysics–aerosol interactions on the scale of cloud systems. However, it is notable that changes in environment (represented by large-scale forcings and the surface fluxes) can change cloud types and the effect of aerosol on precipitation (Lee et al. 2008b; Lee et al. 2009, 2010; van den Heever et al. 2011; Seifert et al. 2011). Hence, it merits future study how changing environment and associated changes in cloud types affect interactions between aerosol, circulations and precipitation.

For the mesoscale system in MESO, it takes around 8 h to establish aerosol-induced precipitation enhancement. Figure 11 demonstrates that if simulations lasted for less than 8 h, we would only observe precipitation enhancement for LARGER. To see the effect of aerosol-induced changes in circulations and their effect on precipitation suppression in LARGER with confidence, simulations need to last at least 16 h according to Fig. 11. This indicates that the time scale of the response of the cloud system precipitation to aerosol gets longer as the spatial scale of the system gets larger. Thus, we need to have much longer time-scale considerations as the spatial scale of a system of interest gets larger to gain a correct assessment of aerosol effects on the system.

Use of a two-dimensional, rather than three-dimensional, cloud system model affords substantial computational advantages but, as Tompkins (2000) notes, some aspects of the divergence and subsidence differ in two- and three-dimensional models. A three-dimensional version of the ideal-high and ideal-low runs has also been conducted with single-moment microphysics, similar to Lin et al. (1983), and coarser resolutions; the horizontal and vertical grid resolutions are 2 km and 500 m. The spatial scale of the wind fields in LARGER (generated by convection and divergence in MESO) in this three-dimensional version of the ideal-high run is ~70% of that simulated with the two-dimensional domain. Also, the averaged strength of the divergence in MESO and associated wind fields in LARGER decreases by ~15% in this three-dimensional version of the ideal-high run as compared to the strength in the two-dimensional domain. However, precipitation enhancement in MESO and precipitation suppression in LARGER are still simulated, resulting in precipitation suppression over the whole domain in the three-dimensional version of the ideal-high run as compared to the three-dimensional version of the ideal-low run. Hence, the qualitative nature of results here is found to be robust to the choice of domain dimensionality.

To see the effect of background wind shear here, the ideal-high and -low runs are repeated with observed wind fields as in the control run. Comparisons between this repeated ideal-high run and the ideal-high run showed that background wind shear acted to increase the intensity and number of subsequently developing clouds in MESO. This increase in the intensity of subsequent clouds leads to larger increases in updrafts and precipitation in MESO and in the strength of horizontally spreading wind fields (generated by convective divergence in MESO) from MESO to LARGER as compared to those with no background wind shear. This leads to precipitation increase and decrease in MESO and LARGER with increasing aerosol, respectively, which are larger with background wind shear than with no background wind shear. Also, this increase in precipitation in MESO is offset by this decrease in precipitation in LARGER to result in smaller precipitation variation over the whole domain than that over each of MESO and LARGER in the repeated ideal-high run. As simulated in the ideal-high run, this offset leads to aerosol-induced precipitation suppression over the whole domain in the repeated ideal-high run. This demonstrates that results here are fairly robust to whether background wind shear is considered or not.

This study does not consider aerosol returning to the atmosphere via evaporation and sublimation of cloud particles. If aerosol were returned to the environment, the divergence in MESO, associated spreading wind fields from MESO to LARGER, and subsidence in LARGER would have more aerosol returning to the atmosphere in themselves in the high-aerosol case than in the low-aerosol case in LARGER. This would result in larger aerosol differences between the high-aerosol and low-aerosol cases than simulated here particularly in divergence in MESO, associated spreading wind fields from MESO to LARGER and subsidence in LARGER. However, it should be stressed that the strong divergence, associated wind fields and subsidence are mostly limited above the freezing level around 4–5 km in the high-aerosol case up to the time when feedbacks leading to precipitation suppression in LARGER are established as shown in Figs. 7 and 10. Lee et al. (2008a,b, 2009) have shown that convective intensity is mostly controlled by aerosol variation below the freezing level, leading to the variation of primary nucleation around cloud base, and the role of the aerosol variation above the freezing level in the convective intensity is negligible; note that cloud droplet number concentration is mostly controlled by primary nucleation (Rogers and Yau 1991). The variation in cloud-droplet number concentration below the freezing level induced by variation in cloud condensation nuclei (CCN) from the boundary layer and the associated variation in evaporation, condensation and homogeneous freezing of droplets nucleated on CCN basically determine changes in updrafts and the convective intensity. Hence, it is not likely that aerosol returning to the atmosphere in the divergence, associated wind fields, and subsidence above the freezing level does have a significant impact on the convective intensity in MESO and LARGER and thus the relative intensity of convection in LARGER to that in MESO in the high-aerosol case. This indicates that stronger convection in MESO than in LARGER, which generates the wind fields spreading from MESO to LARGER and thus subsidence in LARGER, does not depend on the consideration of the aerosol release to the atmosphere.

The propagation of gravity waves from a strong convective region (i.e., MESO) to LARGER in the ideal-high run can have an impact on large-scale circulations and associated cloud and precipitation evolutions in LARGER as suggested by Grabowski et al. (2000). Grabowski et al. (2000) showed that this propagation of gravity waves can generate quasi-two-day oscillations of convective intensity and precipitation. Simulations only last two days and thus the effect of the propagation of gravity waves on cloud and precipitation responses to aerosol in LARGER are not to be shown clearly in this study. Longer simulations will identify interactions between the propagation of gravity waves and large-scale circulations in a more convincing way and thus give us more confident understanding of aerosol effects on large-scale circulations and associated clouds and precipitation.

Acknowledgments

The author thanks NOAA’s Climate Goal Program for supporting this work and the HPCC for computing support. He also thanks Graham Feingold for review.

REFERENCES

  • Abdul-Razzak, H., , and S. J. Ghan, 2000: A parameterization of aerosol activation 2. Multiple aerosol types. J. Geophys. Res., 105 (D5), 68376844.

    • Search Google Scholar
    • Export Citation
  • Abdul-Razzak, H., , and S. J. Ghan, 2002: A parameterization of aerosol activation 3. Sectional representation. J. Geophys. Res., 107, 4026, doi:10.1029/2001JD000483.

    • Search Google Scholar
    • Export Citation
  • Andreae, M. O., and Coauthors, 2004: Smoking rain clouds over the Amazon. Science, 303, 13371342.

  • Bell, T. L., and Coauthors, 2008: Midweek increase in U.S. summer rain and storm heights suggests air pollution invigorates rainstorms. J. Geophys. Res., 113, D02209, doi:10.1029/2007JD008623.

    • Search Google Scholar
    • Export Citation
  • Chou, M.-D., , and M. J. Suarez, 1999: A shortwave radiation parameterization for atmospheric studies. NASA Rep. TM-104606, 40 pp.

  • Chou, M.-D., , W. Ridgway, , and M.-H. Yan, 1999: Parameterizations for water vapor IR radiative transfer in both the middle and lower atmospheres. J. Atmos. Sci., 52, 11591167.

    • Search Google Scholar
    • Export Citation
  • Donner, L. J., , C. J. Seman, , and R. S. Hemler, 1999: Three-dimensional cloud system modeling of GATE convection. J. Atmos. Sci., 56, 18851912.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , C. Wang, , J. Wilson, , J. Ström, , and R. Krejci, 2006: Explicit simulation of aerosol physics in a cloud-resolving model: Aerosol transport and processing in the free troposphere. J. Atmos. Sci., 63, 682696.

    • Search Google Scholar
    • Export Citation
  • Ekman, A. M. L., , A. Engström, , and A. Söderberg, 2011: Impact of two-way aerosol–cloud interaction and changes in aerosol size distribution on simulated aerosol-induced deep convective cloud sensitivity. J. Atmos. Sci., 68, 685698.

    • Search Google Scholar
    • Export Citation
  • Fan, J., and Coauthors, 2009: Dominant role by vertical wind shear in regulating aerosol effects on deep convective clouds. J. Geophys. Res., 114, D22206, doi:10.1029/2009JD012352.

    • Search Google Scholar
    • Export Citation
  • Forster, P., and Coauthors, 2007: Changes in atmospheric constituents and in radiative forcing. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 129–234.

  • Fridlind, A., and Coauthors, cited 2009: ARM/GCSS/SPARC TWP-ICE intercomparison study. [Available online at http://science.arm.gov/wg/cpm/scm/scmic6/index.html.]

  • Ghan, S., , N. Laulainen, , R. Easter, , R. Wagener, , S. Nemesure, , E. Chapman, , Y. Zhang, , and R. Leung, 2001: Evaluation of aerosol direct radiative forcing in mirage. J. Geophys. Res., 106 (D6), 52955316.

    • Search Google Scholar
    • Export Citation
  • Givati, A., , and D. Rosenfeld, 2004: Quantifying precipitation suppression due to air pollution. J. Appl. Meteor., 43, 10381056.

  • Grabowski, W. W., , X. Wu, , and M. W. Moncrieff, 1996: Cloud-resolving modeling of tropical cloud systems during phase III of GATE. Part I: Two-dimensional experiments. J. Atmos. Sci., 53, 36843709.

    • Search Google Scholar
    • Export Citation
  • Grabowski, W. W., , J.-I. Yano, , and M. W. Moncrieff, 2000: Cloud-resolving modeling of tropical circulations driven by large-scale SST gradients. J. Atmos. Sci., 57, 20222039.

    • Search Google Scholar
    • Export Citation
  • Hallet, J., , and S. C. Mossop, 1974: Production of secondary ice particles during the riming process. Nature, 249, 2628.

  • Hobbs, P. V., 1993: Aerosol–Cloud–Climate Interactions. Academic Press, 233 pp.

  • Houze, R. A., 1993: Cloud Dynamics. Academic Press, 573 pp.

  • Khain, A., , and B. Lynn, 2009: Simulation of a supercell storm in clean and dirty atmosphere using weather research and forecast model with spectral bin microphysics. J. Geophys. Res., 114, D19209, doi:10.1029/2009JD011827.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , D. Rosenfeld, , and A. Pokrovsky, 2005: Aerosol impact on the dynamics and microphysics of deep convective clouds. Quart. J. Roy. Meteor. Soc., 131, 26392663.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , N. BenMoshe, , and A. Pokrovsky, 2008: Factors determining the impact of aerosols on surface precipitation from clouds: Attempt of classification. J. Atmos. Sci., 65, 17211748.

    • Search Google Scholar
    • Export Citation
  • Khain, A., , B. Lynn, , and J. Dudhia, 2010: Aerosol effects on intensity of landfalling hurricanes as seen from simulations with WRF model with spectral bin microphysics. J. Atmos. Sci., 67, 365384.

    • Search Google Scholar
    • Export Citation
  • Klemp, J. B., , and R. Wilhelmson, 1978: The simulation of three-dimensional convective storm dynamics. J. Atmos. Sci., 35, 10701096.

  • Koren, I., , Y. J. Kaufman, , D. Rosenfeld, , L. A. Remer, , and Y. Rudich, 2005: Aerosol invigoration and restructuring of Atlantic convective clouds. Geophys. Res. Lett., 32, LI4828, doi:10.1029/2005GL023187.

    • Search Google Scholar
    • Export Citation
  • Kratz, D. P., , M.-D. Chou, , M.-H. Yan, , and C.-H. Ho, 1998: Minor trace gas radiative forcing calculations using the k-distribution method with one-parameter scaling. J. Geophys. Res., 103, 31 64731 656.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , and G. Feingold, 2010: Precipitating cloud-system response to aerosol perturbations. Geophys. Res. Lett., 37, L23806, doi:10.1029/2010GL045596.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , V. T. J. Phillips, , and Y. Ming, 2008a: Examination of aerosol effects on precipitation in deep convective clouds during the 1997 ARM summer experiment. Quart. J. Roy. Meteor. Soc., 134, 12011220.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , V. T. J. Phillips, , and Y. Ming, 2008b: The dependence of aerosol effects on clouds and precipitation on cloud-system organization, shear and stability. J. Geophys. Res., 113, D16202, doi:10.1029/2007JD009224.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , and V. T. J. Phillips, 2009: Sensitivity of aerosol and cloud effects on radiation to cloud types: Comparison between deep convective clouds and warm stratiform clouds over one-day period. Atmos. Chem. Phys., 9, 25552575.

    • Search Google Scholar
    • Export Citation
  • Lee, S. S., , L. J. Donner, , and J. E. Penner, 2010: Thunderstorm and stratocumulus: How does their contrasting morphology affect their interactions with aerosols? Atmos. Chem. Phys., 10, 68196837, doi:10.5194/acp-10-6819-2010.

    • Search Google Scholar
    • Export Citation
  • Lerach, D. G., , B. J. Gaudet, , and W. R. Cotton, 2008: Idealized simulations of aerosol influences on tornadogenesis. Geophys. Res. Lett., 35, L23806, doi:10.1029/2008GL035617.

    • Search Google Scholar
    • Export Citation
  • Liljegren, J. C., 1994: Two-channel microwave radiometer for observations of total column precipitable water vapor and cloud liquid water path. Proc. Fifth Symp. on Global Change, Nashville, TN, Amer. Meteor. Soc., 266–269.

  • Lin, Y. L., , R. D. Farley, , and H. D. Orville, 1983: Bulk parameterization of the snow field in a cloud model. J. Appl. Meteor., 19, 10651092.

    • Search Google Scholar
    • Export Citation
  • Liu, C.-H., , and M. W. Moncrieff, 1996: A numerical study of the effects of ambient flow and shear on density currents. Mon. Wea. Rev., 124, 22822303.

    • Search Google Scholar
    • Export Citation
  • Lohmann, U., , and K. Diehl, 2006: Sensitivity studies of the importance of dust ice nuclei for the indirect aerosol effect on stratiform mixed-phase clouds. J. Atmos. Sci., 63, 968982.

    • Search Google Scholar
    • Export Citation
  • Lynn, B., , A. P. Khain, , J. Dudhia, , D. Rosenfeld, , A. Pokrovsky, , and A. Seifert, 2005: Spectral (bin) microphysics coupled with a mesoscale model (MM5). Part II: Simulation of a CaPe rain event with squall line. Mon. Wea. Rev., 133, 5971.

    • Search Google Scholar
    • Export Citation
  • Lynn, B., , A. P. Khain, , D. Rosenfeld, , and W. L. Woodley, 2007: Effects of aerosols on precipitation from orographic clouds. J. Geophys. Res., 112, D10225, doi:10.1029/2006JD007537.

    • Search Google Scholar
    • Export Citation
  • Moncrieff, M. W., , and C.-H. Liu, 1999: Convection initiation by density currents: Role of convergence, shear, and dynamical organization. Mon. Wea. Rev., 127, 24552464.

    • Search Google Scholar
    • Export Citation
  • Morrison, H., , and W. W. Grabowski, 2011: Cloud-system resolving model simulations of aerosol indirect effects on tropical deep convection and its thermodynamic environment. Atmos. Chem. Phys. Discuss., 11, 15 57315 629.

    • Search Google Scholar
    • Export Citation
  • Ntelekos, A. A., , J. A. Smith, , L. Donner, , J. D. Fast, , W. I. Gustafson, Jr., , E. G. Chapman, , and W. F. Krajewski, 2009: The effects of aerosols on intense convective precipitation in the northeastern United States. Quart. J. Roy. Meteor. Soc., 135, 13671391.

    • Search Google Scholar
    • Export Citation
  • Rogers, R. R., , and M. K. Yau, 1991: A Short Course in Cloud Physics. Pergamon Press, 293 pp.

  • Rosenfeld, D., , A. Khain, , B. Lynn, , and W. L. Woodley, 2007: Simulation of hurricane response to suppression of warm rain by sub-micron aerosols. Atmos. Chem. Phys., 7, 34113424.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , U. Lohmann, , G. B. Raga, , C. D. O’Dowd, , M. Kulmala, , S. Fuzzi, , A. Reissell, , and M. O. Andreae, 2008: Flood or drought: How do aerosols affect precipitation? Science, 321, 13091313.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , J. B. Klemp, , and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45, 463485.

  • Saleeby, S. M., , and W. R. Cotton, 2004: A large-droplet mode and prognostic number concentration of cloud droplets in the Colorado state university regional atmospheric modeling system (RAMS). Part I: Module description and supercell test simulations. J. Appl. Meteor., 43, 182195.

    • Search Google Scholar
    • Export Citation
  • Seifert, A., , and K. D. Beheng, 2006: A double-moment cloud microphysics parameterization for mixed-phase clouds. Part 2: Maritime vs. continental deep convective storms. Meteor. Atmos. Phys., 92, 6782.

    • Search Google Scholar
    • Export Citation
  • Seifert, A., , C. Köhler, , and K. D. Beheng, 2011: Aerosol-cloud-precipitation effects over Germany as simulated by a convective-scale numerical weather prediction model. Atmos. Chem. Phys. Discuss., 11, 20 20320 243.

    • Search Google Scholar
    • Export Citation
  • Simpson, J., , and W.-K. Tao, 1993: The Goddard Cumulus Ensemble model. Part II: Applications for studying cloud precipitating processes and for NASA TRMM. Terr. Atmos. Ocean. Sci., 4, 73116.

    • Search Google Scholar
    • Export Citation
  • Soong, S.-T., , and Y. Ogura, 1980: Response of trade wind cumuli to large-scale processes. J. Atmos. Sci., 37, 20352050.

  • Stevens, B., , and G. Feingold, 2009: Untangling aerosol effects on clouds and precipitation in a buffered system. Nature, 461, 607613, doi:10.1038/nature08281.

    • Search Google Scholar
    • Export Citation
  • Storer, R. L., , S. C. van den Heever, , and G. L. Stephens, 2010: Modeling aerosol impacts on convection under differing storm environments. J. Atmos. Sci., 67, 39043915.

    • Search Google Scholar
    • Export Citation
  • Tao, W.-K., , and J. Simpson, 1993: The Goddard Cumulus Ensemble model. Part I: Model description. Terr. Atmos. Ocean. Sci., 4, 1954.

  • Tao, W.-K., and Coauthors, 2003: Microphysics, radiation and surface processes in the Goddard Cumulus Ensemble (GCE) model. Meteor. Atmos. Phys., 82, 97137.

    • Search Google Scholar
    • Export Citation
  • Tao, W.-K., and Coauthors, 2006: Retrieval of latent heating from TRMM measurements. Bull. Amer. Meteor. Soc., 87, 15551572.

  • Tao, W.-K., , X. Li, , A. Khain, , T. Matsui, , S. Lang, , and J. Simpson, 2007: The role of atmospheric aerosol concentration on deep convective precipitation: Cloud-resolving model simulations. J. Geophys. Res., 112, D24S18, doi:10.1029/2007JD008728.

    • Search Google Scholar
    • Export Citation
  • Tompkins, A. M., 2000: The impact of dimensionality on long-term cloud resolving model simulations. Mon. Wea. Rev., 128, 15211535.

  • Tompkins, A. M., 2001: Organization of tropical convection in low vertical wind shears: The role of cold pools. J. Atmos. Sci., 58, 16501672.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , and W. R. Cotton, 2007: Urban aerosol impacts on downwind convective storms. J. Appl. Meteor. Climatol., 46, 828850.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , G. G. Carrio, , W. R. Cotton, , P. J. DeMott, , and A. J. Prenni, 2006: Impacts of nucleating aerosol on Florida Storms. Part I: Mesoscale simulations. J. Atmos. Sci., 63, 17521775.

    • Search Google Scholar
    • Export Citation
  • van den Heever, S. C., , G. L. Stephens, , and N. B. Wood, 2011: Aerosol indirect effects on tropical convection characteristics under conditions of radiative–convective equilibrium. J. Atmos. Sci., 68, 699718.

    • Search Google Scholar
    • Export Citation
  • Vaughan, G., , K. Bower, , C. Schiller, , A. R. MacKenzie, , T. Peter, , H. Schlager, , N. R. P. Harris, , and P. T. May, 2008: SCOUT-O3/ACTIVE: High-altitude aircraft measurements around deep tropical convection. Bull. Amer. Meteor. Soc., 89, 647662.

    • Search Google Scholar
    • Export Citation
  • Walko, R. L., , W. R. Cotton, , M. P. Meyers, , and J. Y. Harrington, 1995: New RAMS cloud microphysics parameterization: Part I. The single-moment scheme. Atmos. Res., 38, 2962.

    • Search Google Scholar
    • Export Citation
  • Wang, H., , and G. Feingold, 2009: Modeling open cellular structures and drizzle in marine stratocumulus. Part II: The microphysics and dynamics of the boundary region between open and closed cells. J. Atmos. Sci., 66, 32573275.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , and J. B. Klemp, 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110, 504520.

    • Search Google Scholar
    • Export Citation
  • Xie, S., , T. Hume, , C. Jakob, , S. A. Klein, , and R. B. McCoy, 2010: Observed large-scale structures and diabatic heating and drying profiles during TWP-ICE. J. Climate, 23, 5779.

    • Search Google Scholar
    • Export Citation
  • Xu, K. M., and Coauthors, 2002: An intercomparison of cloud-resolving models with the Atmospheric Radiation Measurement summer 1997 intensive observation period data. Quart. J. Roy. Meteor. Soc., 128, 593624.

    • Search Google Scholar
    • Export Citation
  • Zhang, H., , G. McFarquhar, , S. Saleeby, , and W. R. Cotton, 2007: Impacts of Saharan Dust as CCN on the evolution of an idealized tropical cyclone. Geophys. Res. Lett., 34, L14812, doi:10.1029/2007GL029876.

    • Search Google Scholar
    • Export Citation
Save