• Albrecht, B. A., 1989: Aerosols, cloud microphysics, and fractional cloudiness. Science, 245 , 12271230.

  • Borys, R. D., , and M. A. Wetzel, 1997: Storm Peak Laboratory: A research, teaching, and service facility for the atmospheric sciences. Bull. Amer. Meteor. Soc., 78 , 21152123.

    • Search Google Scholar
    • Export Citation
  • Borys, R. D., , D. H. Lowenthal, , and D. L. Mitchell, 2000: The relationships among cloud microphysics, chemistry, and precipitation rate in cold mountain clouds. Atmos. Environ., 34 , 25932602.

    • Search Google Scholar
    • Export Citation
  • Borys, R. D., , D. H. Lowenthal, , S. A. Cohn, , and W. O. J. Brown, 2003: Mountaintop and radar measurements of anthropogenic aerosol effects on snow growth and snowfall rate. Geophys. Res. Lett., 30 , 1538. doi:10.1029/2002GL016855.

    • Search Google Scholar
    • Export Citation
  • Cober, S. G., , and R. List, 1993: Measurements of the heat and mass transfer parameters characterizing conical graupel growth. J. Atmos. Sci., 50 , 15911609.

    • Search Google Scholar
    • Export Citation
  • Cotton, W. R., and Coauthors, 2003: RAMS 2001: Current status and future directions. Meteor. Atmos. Phys., 82 , 529.

  • Dusek, U., and Coauthors, 2006: Size matters more than chemistry for cloud-nucleating ability of aerosol particles. Science, 312 , 13751378.

    • Search Google Scholar
    • Export Citation
  • Feingold, G., , W. L. Eberhard, , D. E. Veron, , and M. Previdi, 2003: First measurements of the Twomey indirect effect using ground-based remote sensors. Geophys. Res. Lett., 30 , 1287. doi:10.1029/2002GL016633.

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

  • Givati, A., , and D. Rosenfeld, 2007: Possible impacts of anthropogenic aerosols on water resources of the Jordan River and the Sea of Galilee. Water Resour. Res., 43 , W10419. doi:10.1029/2006WR005771.

    • Search Google Scholar
    • Export Citation
  • Greenan, B. J. W., , and R. List, 1995: Experimental closure of the heat and mass transfer theory of spheroidal hailstones. J. Atmos. Sci., 52 , 37973815.

    • Search Google Scholar
    • Export Citation
  • Harrington, J. Y., 1997: The effects of radiative and microphysical processes on simulated warm and transition season Arctic stratus. Ph.D. dissertation, Colorado State University Atmospheric Science Paper 637, 289 pp.

  • Heymsfield, A. J., , and R. M. Sabin, 1989: Cirrus crystal nucleation by homogeneous freezing of solution droplets. J. Atmos. Sci., 46 , 22522264.

    • Search Google Scholar
    • Export Citation
  • Hindman, E. E., 1986: Characteristics of supercooled liquid water in clouds at mountaintops in the Colorado Rockies. J. Climate Appl. Meteor., 25 , 12711279.

    • Search Google Scholar
    • Export Citation
  • Hindman, E. E., , M. A. Campbell, , and R. D. Borys, 1994: A ten-winter record of cloud-droplet physical and chemical properties at a mountaintop site in Colorado. J. Appl. Meteor., 33 , 797807.

    • Search Google Scholar
    • Export Citation
  • Jirak, I. L., , and W. R. Cotton, 2006: Effect of air pollution on precipitation along the Front Range of the Rocky Mountains. J. Appl. Meteor. Climatol., 45 , 236245.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain-Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Levin, Z., , and W. R. Cotton, 2008: Aerosol Pollution Impact on Precipitation: A Scientific Review. Springer, 386 pp.

  • Lynn, B., , A. 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
  • Meyers, M. P., , R. L. Walko, , J. Y. Harrington, , and W. R. Cotton, 1997: New RAMS cloud microphysics parameterization. Part II. The two-moment scheme. Atmos. Res., 45 , 339.

    • Search Google Scholar
    • Export Citation
  • Mitchell, D. L., , R. Zhang, , and R. L. Pitter, 1990: Mass dimensional relationships for ice particles and the influence of riming on snowfall rates. J. Appl. Meteor., 29 , 153163.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1976: Production of secondary ice particles during the growth of graupel by riming. Quart. J. Roy. Meteor. Soc., 102 , 4557.

    • Search Google Scholar
    • Export Citation
  • Pitter, R. L., , and H. R. Pruppacher, 1974: A numerical investigation of collision efficiencies of simple ice plates colliding with supercooled water drops. J. Atmos. Sci., 31 , 551559.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., 1981: Microphysical processes in two stably stratified orographic cloud systems. M.S. thesis, Dept. of Atmospheric Science, Colorado State University Atmospheric Science Paper 337, 151 pp.

  • Rauber, R. M., , L. O. Grant, , D. Feng, , and J. B. Snider, 1986a: The characteristics and distribution of cloud water over the mountains of northern Colorado during wintertime storms. Part I: Temporal variations. J. Climate Appl. Meteor., 25 , 468488.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., , L. O. Grant, , D. Feng, , and J. B. Snider, 1986b: The characteristics and distribution of cloud water over the mountains of northern Colorado during wintertime storms. Part II: Spatial distribution and microphysical characteristics. J. Climate Appl. Meteor., 25 , 489504.

    • Search Google Scholar
    • Export Citation
  • Reinking, R. F., , J. B. Snider, , and J. L. Coen, 2000: Influences of storm-embedded orographic gravity waves on cloud liquid water and precipitation. J. Appl. Meteor., 39 , 733759.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , and A. Givati, 2006: Evidence of orographic precipitation suppression by air pollution-induced aerosols in the western United States. J. Appl. Meteor. Climatol., 45 , 893911.

    • Search Google Scholar
    • Export Citation
  • 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 descriptions and supercell test simulations. J. Appl. Meteor., 43 , 182195.

    • Search Google Scholar
    • Export Citation
  • Saleeby, S. M., , and W. R. Cotton, 2005: A large-droplet mode and prognostic number concentration of cloud droplets in the Colorado State University Regional Atmospheric Modeling System (RAMS). Part II: Sensitivity to a Colorado winter snowfall event. J. Appl. Meteor., 44 , 19121929.

    • Search Google Scholar
    • Export Citation
  • Saleeby, S. M., , and W. R. Cotton, 2008: A binned approach to cloud-droplet riming implemented in a bulk microphysics model. J. Appl. Meteor. Climatol., 47 , 694703.

    • Search Google Scholar
    • Export Citation
  • Twomey, S., 1977: The influence of pollution on the shortwave albedo of clouds. J. Atmos. Sci., 34 , 11491152.

  • 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
  • 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, P. K., , and W. Ji, 2000: Collision efficiencies of ice crystals at low-intermediate Reynolds numbers colliding with supercooled cloud droplets: A numerical study. J. Atmos. Sci., 57 , 10011009.

    • Search Google Scholar
    • Export Citation
  • Warburton, J. A., , and T. P. deFelice, 1986: Oxygen isotopic composition of central Sierra Nevada precipitation. I: Identification of ice-phase water capture regions in winter storms. Atmos. Res., 20 , 1122.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Model grid configuration. (top) The nested grid setup with grids 1, 2, and 3 denoted by the labels G1, G2, and G3. (middle) A topographic plan view of grid 4 centered over SPL and within G3. The SNOTEL stations of Dry Lake (DRY), Tower (TOW), Rabbit Ears (RAB), Columbine (COL), and Buffalo Park (BUF) are shown for reference. (bottom) A model topographic cross section along the arrowed line in the middle panel; the labels SPL and SPL-4, e.g., represent the locations of SPL and four grid points west of SPL, respectively.

  • View in gallery

    Sum of total measured SWE during January–February 2007 at five measuring sites near SPL.

  • View in gallery

    Time series (UTC) of (a) CCN concentration at 0.3% supersaturation measured at SPL for the duration of the ISPA II field study (January–February 2007), and (b) CCN concentration at 0.3% supersaturation and RH for the snowfall and orographic cloud event of 11–12 Feb 2007.

  • View in gallery

    The 12 Feb 2007 time series (UTC) of NCAR MAPR radar reflectivity, vertical velocity, and wind profiles.

  • View in gallery

    As in Fig. 4, but for 24 Feb 2007.

  • View in gallery

    Storm total accumulated snow water equivalent simulated on grid 4 (mm, shaded) and topography (m, line contours). Note the enhanced precipitation oriented north–south along the Park Range. The simulations shown were initialized with a maximum CCN concentration of 100 cm−3.

  • View in gallery

    Conceptual model of the orographic supercooled water cloud impinging upon the crest of the Park Range of Colorado. Ice particle trajectory and condensate distribution depict the seeder–feeder growth process that occurs because of riming of ice crystals during their passage through the clouds prior to surface deposition (from Borys et al. 2000; adapted from Rauber 1981).

  • View in gallery

    Time series (UTC) of cloud LWC (g m−3) as measured at SPL (light gray) and from RAMS closest grid point to SPL for the simulation initialized with maximum CCN concentration of 100 cm−3 (dark gray) and 1900 cm−3 (black).

  • View in gallery

    As in Fig. 8, but for temperature (°C).

  • View in gallery

    As in Fig. 8, but for RH (%).

  • View in gallery

    As in Fig. 8, but for wind speed (m s−1).

  • View in gallery

    Time series (UTC) of snow water equivalent (mm) from RAMS simulations and the Park Range SNOTEL stations of RAB, COL, TOW, and DRY. The notation PHQ signifies data from the grid point closest to the Patrol Station Headquarters located about 1 km downslope from SPL. Likewise, PHQ-2 signifies data from two grid points to the west. PHQ snow measurements are given by the large black dots.

  • View in gallery

    Variations in storm total SWE (mm) along a west-to-east transect running through SPL. Simulation results with CCN concentrations of 100 and 1900 cm−3 are shown in light and dark shading, respectively.

  • View in gallery

    Simulated total accumulated SWE difference (mm) resulting from an increase in NCCN from clean to polluted conditions. Increased pollution aerosols reduced precipitation on the windward slope and increased precipitation on the leeward slope.

  • View in gallery

    Vertical cross section (denoted in Fig. 1) of RAMS cloud water mixing ratio (g kg−1, shaded), snow (g kg−1 × 100, red solid contours), and graupel (g kg−1 × 100, black dashed contours). Chosen times were near the event maximum in simulated liquid water content at SPL. The event date and maximum CCN concentrations (cm−3) are labeled. (bottom right) Model height is shown in meters.

  • View in gallery

    Time series (UTC) of vertically integrated rimed mass in the gridcell column above SPL for the clean (gray line) and polluted (black line) simulations.

  • View in gallery

    The 11 Feb 2007 time series (UTC) of cloud droplet concentrations from the forward scattering spectrometer probe at SPL and from the model grid point closest to SPL for the clean (RAMS SPL 100) and polluted (RAMS SPL 1900) simulations.

  • View in gallery

    The 11 Feb 2007 time series (UTC) of ice crystal concentrations from the 2D ice probe (2D ice probe counts) at SPL and from the model grid point closest to SPL for the clean (RAMS SPL 100) and polluted (RAMS SPL 1900) simulations. Apparent zero values or data gaps from the 2D probe are missing or unreliable data.

  • View in gallery

    Simulated accumulated SWE difference (mm), from the 11 Feb event, resulting from an increase in NCCN from 100 cm−3 to more polluted conditions with NCCN of (a) 300, (b) 500, (c) 800, and (d) 1100 cm−3.

  • View in gallery

    Time series (UTC) of grid 4 domain-summed (a) cloud nucleated mass and (b) cloud water mixing ratio (LWC) from the 11 Feb event. These are vertically integrated quantities summed across the domain (kg m−2). Plots show the time series for each simulation with different NCCN (concentration labeled; cm−3).

  • View in gallery

    As in Fig. 20, but for (a) cloud droplet vapor diffusion and (b) cloud droplet rimed mass.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 114 114 16
PDF Downloads 166 166 10

Influence of Cloud Condensation Nuclei on Orographic Snowfall

View More View Less
  • 1 Colorado State University, Fort Collins, Colorado
  • | 2 Desert Research Institute, Reno, Nevada
© Get Permissions
Full access

Abstract

Pollution aerosols acting as cloud condensation nuclei (CCN) have the potential to alter warm rain clouds via the aerosol first and second indirect effects in which they modify the cloud droplet population, cloud lifetime and size, rainfall efficiency, and radiation balance from increased albedo. For constant liquid water content, an increase in CCN concentration (NCCN) tends to produce an increased concentration of droplets with smaller diameters. This reduces the collision and coalescence rate, and thus there is a local reduction in rainfall. While this process applies to warm clouds, it does not identically carry over to mixed-phase clouds in which crystal nucleation, crystal riming, crystal versus droplet fall speed, and collection efficiency play active roles in determining precipitation amount. Sulfate-based aerosols serve as very efficient cloud nuclei but are not effective as ice-forming nuclei. In clouds where precipitation formation is dominated by the ice phase, NCCN influences precipitation growth by altering the efficiency of droplet collection by ice crystals and the fall trajectories of both droplet and crystal hydrometeors. The temporal and spatial variation in both crystal and droplet populations determines the resultant snowfall efficiency and distribution. Results of numerical simulations in this study suggest that CCN can play a significant role in snowfall production by winter, mixed-phase, cloud systems when liquid and ice hydrometeors coexist. In subfreezing conditions, a precipitating ice cloud overlaying a supercooled liquid water cloud allows growth of precipitation particles via the seeder–feeder process, in which nucleated ice crystals fall through the supercooled liquid water cloud and collect droplets. Enhanced NCCN from sulfate pollution by fossil fuel emissions modifies the droplet distribution and reduces crystal riming efficiency. Reduced riming efficiency inhibits the rate of snow growth, producing lightly rimed snow crystals that fall slowly and advect farther downstream prior to surface deposition. Simulations indicate that increasing NCCN along the orographic barrier of the Park Range in north-central Colorado results in a modification of the orographic cloud such that the surface snow water equivalent amounts are reduced on the windward slopes and enhanced on the leeward slopes. The inhibition of snowfall by pollution aerosols (ISPA) effect has significant implications for water resource distribution in mountainous terrain.

Corresponding author address: Stephen M. Saleeby, Atmospheric Science Department, Colorado State University, Fort Collins, CO 80523. Email: smsaleeb@atmos.colostate.edu

Abstract

Pollution aerosols acting as cloud condensation nuclei (CCN) have the potential to alter warm rain clouds via the aerosol first and second indirect effects in which they modify the cloud droplet population, cloud lifetime and size, rainfall efficiency, and radiation balance from increased albedo. For constant liquid water content, an increase in CCN concentration (NCCN) tends to produce an increased concentration of droplets with smaller diameters. This reduces the collision and coalescence rate, and thus there is a local reduction in rainfall. While this process applies to warm clouds, it does not identically carry over to mixed-phase clouds in which crystal nucleation, crystal riming, crystal versus droplet fall speed, and collection efficiency play active roles in determining precipitation amount. Sulfate-based aerosols serve as very efficient cloud nuclei but are not effective as ice-forming nuclei. In clouds where precipitation formation is dominated by the ice phase, NCCN influences precipitation growth by altering the efficiency of droplet collection by ice crystals and the fall trajectories of both droplet and crystal hydrometeors. The temporal and spatial variation in both crystal and droplet populations determines the resultant snowfall efficiency and distribution. Results of numerical simulations in this study suggest that CCN can play a significant role in snowfall production by winter, mixed-phase, cloud systems when liquid and ice hydrometeors coexist. In subfreezing conditions, a precipitating ice cloud overlaying a supercooled liquid water cloud allows growth of precipitation particles via the seeder–feeder process, in which nucleated ice crystals fall through the supercooled liquid water cloud and collect droplets. Enhanced NCCN from sulfate pollution by fossil fuel emissions modifies the droplet distribution and reduces crystal riming efficiency. Reduced riming efficiency inhibits the rate of snow growth, producing lightly rimed snow crystals that fall slowly and advect farther downstream prior to surface deposition. Simulations indicate that increasing NCCN along the orographic barrier of the Park Range in north-central Colorado results in a modification of the orographic cloud such that the surface snow water equivalent amounts are reduced on the windward slopes and enhanced on the leeward slopes. The inhibition of snowfall by pollution aerosols (ISPA) effect has significant implications for water resource distribution in mountainous terrain.

Corresponding author address: Stephen M. Saleeby, Atmospheric Science Department, Colorado State University, Fort Collins, CO 80523. Email: smsaleeb@atmos.colostate.edu

1. Introduction

The Park Range of Colorado receives the majority of its annual precipitation in the form of snow during the winter months (Borys and Wetzel 1997). This north–south-oriented mountain range is generally aligned orthogonally to the westerly mean flow that accompanies most synoptic midlatitude cyclones over the western United States. In the absence of blocked flow, deep lifting of a near-surface air mass from the west can be transported over the crest of the Park Range and lead to development of mixed-phase clouds. This deep lifting along the slope provides for enhanced condensate production and surface snowfall. A strong cross-barrier pressure gradient and upslope winds contribute to supercooled liquid water formation in the orographic cloud (Rauber et al. 1986b; Borys et al. 2000; Saleeby and Cotton 2005). Supercooled cloud events are frequently observed during the winter months at the Desert Research Institute (DRI) Storm Peak Laboratory (SPL). SPL is a high-altitude research facility located at 3210 m MSL on the west summit of Mount Werner on the ridge crest of the Park Range in the Routt National Forest permit area of the Steamboat ski area near Steamboat Springs, Colorado (Borys and Wetzel 1997).

A seeder–feeder mechanism, involving the sedimentation of higher-altitude snow crystals through the low-level orographic cloud, produces greater precipitation amounts near the mountaintop due to ample riming of cloud droplets in the lowest 2 km of the cloud (Rauber et al. 1986a; Reinking et al. 2000). This low-level riming process enhances the precipitation efficiency, such that the rimed ice may compose 20%–50% of the final snow mass that reaches the surface (Mitchell et al. 1990; Borys et al. 2003). Enhanced riming increases the mass growth rate of snow crystals as well as the fall speed. Larger snowfall amounts can thus be expected along windward slopes (Hindman 1986). Slower-falling unrimed snow crystals can be carried over to leeward slopes where subsidence allows sublimation, a reduction in total surface snowfall, and the disappearance of the “feeder” cloud (Rauber et al. 1986a,b).

Characteristics of the cloud droplet spectra in the orographic cloud are related to riming efficiency. Aerosols, primarily those with the composition to act as cloud condensation nuclei (CCN), that enter a cloud typical of winter orographic systems in the Park Range can influence the resulting cloud droplet number concentration (CDNC) and droplet size. A change in the droplet size impacts ice particle riming efficiencies (Hindman et al. 1994). Pitter and Pruppacher (1974) identified 10 μm as the cloud droplet diameter below which ice crystal riming efficiencies for all crystal habits are near zero. Wang and Ji (2000) further examined rime accretion on various snow crystal habits and found that the rime collection efficiency does not completely decrease to zero, but it does tend rapidly toward zero below 10 μm. Borys et al. (2000) document results of aerosol chemistry and orographic cloud microphysical properties from a wintertime field campaign at SPL in January 1995. Their observational analysis revealed that an increase in aerosol sulfate concentration led to higher concentrations of smaller cloud droplets, a reduction in efficiency of ice crystal riming, and thus an inhibition of crystal growth rate. Further, Lynn et al. (2007) performed 2D simulations of orographic precipitation across the Sierra Nevada of California using a spectral bin microphysics model and compared simulations representing marine and continental air masses. Their polluted continental-type simulations tended to suppress precipitation, shift the precipitation maximum downstream, and reduce the presence of graupel.

Annual water supplies in Colorado rely heavily on the water content of the winter snowpack, which is dependent on in-cloud crystal riming to enhance precipitation efficiency. The Inhibition of Snowfall by Pollution Aerosols (ISPA) II field campaign was conducted at SPL during January and February 2007 to investigate the cloud microphysical and cloud dynamical processes that control snow growth in winter orographic systems. The Colorado State University (CSU) Regional Atmospheric Modeling System (RAMS) was used in a real-time forecast mode to provide model guidance for potential orographic cloud formation and snowfall totals during the field program and in postevent simulation mode to examine aerosol effects on mixed-phase precipitation formation mechanisms. The observational goal of the field project was to characterize and compare the properties of mixed-phase clouds in clean and polluted environments during which the riming mechanism was active. Cloud and ice particle probes provided information concerning the cloud droplet, ice crystal, and snow size spectra. Data were also obtained for NCCN, non-CCN aerosol concentrations, meteorological conditions, acoustically measured time series accumulation of snowfall, mass gage measurement of snow water equivalent, and daily manual measurement of snowfall depth at several elevations relative to SPL. Rime ice and snow were collected for chemical analysis to quantify anthropogenic pollution and determine the riming altitude using oxygen isotopic ratios (Warburton and deFelice 1986). The National Center for Atmospheric Research (NCAR) Multiple Antenna Profiler Radar (MAPR) was operated at a site near the base of Mount Werner to provide time series analysis of precipitation intensity and vertical profiles of wind during storm events. The numerical modeling component of this study was carried out at CSU to examine the impact of NCCN upon orographic clouds, the riming growth process, and orographic snowfall along the Park Range, including SPL. Several heavy riming and heavy snowfall cases were identified from data collected during ISPA II. Following the field work, these events were simulated at high resolution using CSU RAMS.

2. Model and simulation description

The RAMS is utilized for performing sensitivity simulations with varying NCCN. The nonhydrostatic, compressible version of RAMS is configured on an Arakawa-C grid and sigma-z terrain-following coordinate system (Cotton et al. 2003). For these simulations, the model uses two-way nesting with a nested four-grid arrangement centered over Colorado. The outer grid 1 covers the continental United States with 60-km grid spacing (62 × 50 grid points), grid 2 covers Colorado and the adjacent surrounding states with 15-km grid spacing (54 × 50 grid points), grid 3 encompasses much of Colorado with 3-km grid spacing (97 × 82 grid points), and grid 4 covers the north–south-oriented Park Range from the city of Hayden in the west to the city of Walden in the east with 750-m grid spacing (114 × 114 grid points) (Fig. 1). Within each grid there are 40 vertical levels with a minimum of 75-m grid spacing. The model uses vertical grid stretching with a a stretch ratio of 1.12 and a maximum vertical grid spacing of 750 m.

The RAMS model contains a bin-emulating, bulk microphysics package that predicts two moments of the hydrometeor distributions (mixing ratio and number concentration) for rain, pristine ice, snow, aggregates, graupel, and hail (Walko et al. 1995; Meyers et al. 1997; Cotton et al. 2003). Saleeby and Cotton (2004) extended the two-moment approach to the cloud droplet distribution via a parameterization for the formation of cloud droplets from activation of CCN within a lifted parcel. The Lagrangian parcel model of Heymsfield and Sabin (1989) was used to determine the percent of user-specified CCN that would deliquesce, activate, and grow by condensation into cloud droplets for a given ambient temperature, vertical velocity, NCCN, and median radius of the CCN distribution. It is assumed that CCN are ammonium sulfate. Sulfate tends to be a dominant base chemistry of hygroscopic aerosols but not the only possible constituent. However, Dusek et al. (2006) stressed that the aerosol size distribution is of much greater importance than aerosol chemistry for droplet nucleation. Ice nuclei (IN) are represented by a density-weighted vertically decaying concentration profile of NIN × ρ5.4 with a maximum IN number concentration, NIN, of 100 L−1, where ρ is the air density. The total number of newly formed ice crystals is dependent upon the ice nuclei concentration and supersaturation with respect to ice and is given as
i1558-8432-48-5-903-e1
with Nice as the number of nucleated ice crystals, Si as the computed supersaturation with respect to ice, and T as the ambient temperature (Meyers et al. 1997). Furthermore, homogeneous freezing of cloud droplets, contact nucleation, and secondary ice production, based on Mossop (1976), are simulated. Saleeby and Cotton (2008) introduced a binned approach to riming within the bulk microphysics framework in which realistic collection efficiencies, from Wang and Ji (2000), Cober and List (1993), and Greenan and List (1995), are used in the computation of the collision–coalescence of ice crystals and cloud droplets. The hydrometeor gamma distributions, used for representing hydrometeor spectra in the bulk microphysics, are temporarily decomposed into 36 size bins for riming computations of all possible size interactions. This method is highly beneficial in winter orographic simulations, and is much improved over the bulk riming method that applied a single collection efficiency to the full size distributions. The 32-km North American Regional Reanalysis (NARR) was used for model initialization and nudging of the lateral boundaries. The model also made use of a two-stream, hydrometeor-sensitive radiation scheme (Harrington 1997) and the Kain–Fritsch cumulus parameterization on the outermost grid (Kain and Fritsch 1993).

The case studies were run to simulate clean and polluted aerosol environments. The NCCN were initialized as horizontally homogeneous with a vertical profile that decreases linearly with height up to 4 km AGL. In the clean and polluted simulations, the surface NCCN was initialized with maximum surface values of 100 and 1900 cm−3, respectively. The minimum initial concentration allowed at any grid point was 100 cm−3. The aerosol spectrum was represented by a polydisperse field with a lognormal distribution with a median CCN dry radius of 0.04 μm. While an NCCN of 1900 cm−3 would represent a highly polluted event, the given distribution represents the potential CCN and would require rather high supersaturation for all of the particles to activate and lead to droplet formation. As such, we will examine how this aerosol specification modifies the cloud properties, which ultimately alter the riming efficiency. As a source/sink function, CCN were depleted upon droplet nucleation and replenished upon droplet evaporation. These sensitivity simulations in clean and polluted environments were performed on four cases.

Two significant events occurred during ISPA II in February 2007. The precipitation system that occurred during 11–13 February was characterized by conditions of heavy riming while 23–25 February was characterized by light riming. Simulations of these events were initialized at 0000 UTC 11 February and 1200 UTC 23 February 2007 and run for approximately 39 h to demonstrate the influence of NCCN under differing conditions of cloud LWC. Following a review of the observations and model results, the 11 February 2007 event was selected for further study because of the extended coexistence of ice and supercooled cloud liquid water during the measurement time period. This event had a persistent mixed-phase cloud with ample amounts of cloud water and snowfall both produced by the model simulations and observed by field measurements. Further simulations were carried out to examine the impact of CCN on cloud microphysical interactions. In these model experiments, maximum NCCN of 100, 300, 500, 800, 1100, 1500, and 1900 cm−3 were applied, using the same vertically decreasing concentration profile with height. This ensemble provides an estimate of the potential change in the crystal growth and resultant snowfall distribution as a function of aerosol loading. All model results shown herein come from the finest nested grid at 750-m grid spacing.

3. Characteristics of orographic precipitation events

Each of the examined snowfall events occurred under conditions of westerly flow with the passage of a wintertime baroclinic system. Southerly and easterly flow conditions along the Park Range are unfavorable for the formation of a mountaintop cloud layer because of blocking by the Front Range (to the east) and the Flat Tops Wilderness Area (to the south). Flow from the northwest is dominant during wintertime snowfall events (Hindman et al. 1994; Borys et al. 2000).

With westerly to northwesterly winds, the flow is nearly orthogonal to the Park Range barrier, providing for strong upward forcing and production of enhanced condensate in the form of supercooled liquid water as well as snow crystal mass. As such, there is often a sharp gradient in snowfall totals between the base of the Park Range and SPL. Figure 2 depicts the total measured snow water equivalent (SWE) during January and February 2007 for each of five snow measuring stations along the western slope of Mount Werner. Generally, the SWE increases as elevation increases. The one exception is the site MOR, which is located on the lee slope of the ridge crest. This station likely experiences a precipitation lag effect in which condensate produced by orographic lifting is advected downstream of the ridge before surface deposition. This is likely a function of the strength of the flow above the ridge.

CCN spectra were obtained from a Droplet Measurement Technologies CCN counter that was run nearly continuously for the duration of the field project from early January to February. This instrument determines the number concentration of CCN by drawing in ambient air and moistening the chamber to the desired level of supersaturation and then counting the number of droplets that form to a minimum diameter of 2 μm. In most instances, higher imposed supersaturation provides higher counts of CCN. Figure 3a displays the 2-month time series of CCN concentration measured at 0.3% supersaturation. This time series includes both clear air and cloudy periods and reveals substantial time variability. Figure 3b displays the CCN concentration at 0.3% supersaturation and the corresponding relative humidity for an orographic cloud event beginning 10 February 2007, which will be discussed below. This figure demonstrates the nucleation scavenging of the CCN as the relative humidity exceeds saturation and the orographic cloud begins to form.

The selected case study periods, 11–13 and 23–25 February, had significantly different dynamic and microphysical characteristics. Time series from the NCAR MAPR instrument system for 12 February (Fig. 4) indicated relatively smooth west-northwest flow. The MAPR figures (Figs. 4, 5) display the time series (increasing time from left to right) of reflectivity, vertical velocity, and the averaged winds. Cloud sampling at the SPL site observed orographic cloud with high liquid water contents and near-continuous precipitation during the latter period of the event. Observations made at SPL showed that precipitation was dominated by warm crystal habits including needles. Heavy rime was evident on the crystals. High liquid water contents contributed to the large droplet sizes (greater than 15-um diameter) in the presence of few CCN (less than 50 cm−3 droplet concentrations were observed). The RAMS simulations produced similar wind, temperature, and humidity conditions within the system, although the model overestimated LWC (0.7 g m−3 versus observed 0.5 g m−3) during the period of maximum values. In contrast, the 23–25 February 2007 event (Fig. 5) contained two strong wind shift events during the period 0000–0400 UTC 24 February. This was accompanied by a sharp drop in temperature and a shallow cloud layer that was not fully represented by the model. Low liquid water contents were observed at SPL with high CDNC (occasionally exceeding 230 cm−3). Ice crystal riming was thus limited, and the observed crystal type was primarily unrimed dendrites. As with the composite elevation–snowfall relationship shown in Fig. 2, measurements at snow sampling sites indicated greater snowfall at MOR (1–2 grid points to the lee side of the ridge) than was deposited at the ridgeline [Patrol Station Headquarters (PHQ) site] for both the 11–12 and 23–24 February event periods. The leeside snowfall at MOR had an 11% larger SWE accumulation than PHQ for an early period of 11 February during low ice crystal riming.

Plots of simulated accumulated SWE within grid 4 for each case show the strong orographic enhancement of snowfall along the Park Range (Fig. 6). The precise maximum in precipitation varies by event depending on the flow direction and available moisture. Furthermore, impinging orographic flow often results in maximum concentrations of condensate on the windward slope very near the ridge (Fig. 7). The seeder–feeder process enhances the surface accumulation of orographic precipitation through the riming process. Thus, maximum precipitation efficiency relative to available moisture should occur under conditions of deep, strong, moist, upslope flow that favors the coexistence of ice and liquid clouds in which the seeder–feeder process is active, and riming occurs with high efficiency.

Measurements of cloud droplet size distribution and CDNC were made on the SPL observation deck using a Droplet Measurement Technology (DMT) Scattering Particle Probe (SPP)-100 (Borys et al. 2000). The cloud liquid water content was estimated from the measured droplet distribution. In Fig. 8, the smoothed, 1-min average LWC is compared to the RAMS-predicted cloud LWC at the grid point closest to SPL for the clean and polluted simulations. (Aerosol effects will be discussed in the following section.) Although the model-simulated magnitudes show a tendency for overprediction of LWC during the periods of higher observed LWC, the time series show a correspondence for the time periods during which LWC reached maximum values. The RAMS cloud water time series indicates an extended period during the 11 February event with LWC > 0.1 g m−3. Heavy riming and droplets larger than 10 μm are more likely under these conditions (Borys et al. 2000). Field observations for the 11 February event confirm the occurrence of heavy riming during an extended time period. Snow density (ratio of melted water equivalent depth to total snow depth) values measured at PHQ and MOR sites were 0.13 for the 11 February case. The measurements and RAMS results each indicated high values of cloud LWC for a long period for the 11–13 February event. As such, we expect this event to be more sensitive to simulated changes in CCN loading, as presented in a later section. The model time series for LWC during the 23 February simulation shows lower overall magnitudes; 0.1 g m−3 is exceeded for only a short time period. SPL measurements also indicated low LWC conditions. Snow density measurements at PHQ and MOR for this event (0.04) suggested limited crystal riming, corresponding to visual observations of crystals at SPL.

For both snowfall events the modeled and observed time series of temperature, relative humidity (RH), and wind speed for the SPL location are depicted in Figs. 9 –11, respectively. Time series of both the clean and polluted simulations are included for examining potential dynamic and thermodynamic feedbacks due to modified aerosol concentrations. The temperature and RH at SPL are well simulated for the 11 February event. For 23 February, the simulated temperature was too warm and the RH was too moist relative to observations. The wind speed appears more difficult to simulate, such that the observations report much more variability over time than the model. However, the time-averaged SPL-predicted wind speed during the 11 February event (7.5 m s−1) was in close agreement with the observed average (7.7 m s−1). For the 23 February event the model overpredicted the wind (9.6 m s−1) compared to that observed (7.5 m s−1). Comparisons of these meteorological fields as well as the cloud LWC indicate better model to observation agreement for the 11 February event. The simulations of 23 February generally produce stronger than observed winds, warmer temperatures, and greater saturation at the SPL location. Regardless of the forecast skill, both simulated cases reveal negligible feedbacks from the modified CCN field to the ambient environment. This is not unexpected given that winter orographic clouds tend to be shallow with modest values of vertical velocity and latent heat release (due to droplet nucleation) as compared to deep convective events where strong feedbacks may occur from buoyancy perturbations. Further discussion in the next section will demonstrate that the aerosol concentration strongly impacts the CDNC and degree of riming, moderately impacts the total precipitation beneath the orographic cloud, and weakly modifies the cloud LWC.

The good prediction of the atmospheric fields for 11 February and the biased prediction for 23 February directly impact the forecast skill of the simulated snowfall totals. Periodic accumulated SWE at the PHQ site, and time series of SWE for RAMS grid points relative to PHQ as well as local snow telemetry (SNOTEL) sites along the Park Range, are compared in Fig. 12. The SNOTEL stations are automated observing units that measure and report snow depth and SWE using a snow pillow and storage precipitation gauge. The RAMS time series for 11 February represent the grid point closest to PHQ while the time series from 23 February represent the along-slope grid point relative to PHQ that best agrees with the PHQ observation and SNOTEL sites. The time series shown here are from the clean simulations for each case. For the 23 February event, the model overpredicted the orographic enhancement of snow, such that the model prediction at a lower elevation 1.5 km upstream of PHQ agreed best in terms of precipitation amount. One potential mechanism leading to the precipitation difference is the strength of the cross-barrier flow. It is plausible that the overpredicted horizontal wind speed, discussed above, led to greater convergence, vertical velocity, and snow condensate production. The orographic enhancement of precipitation is quite sensitive to the large-scale flow.

Despite the snowfall total differences, the simulated cases performed well with regard to the timing and intensity of snowfall periods compared with the SNOTEL time series and PHQ measurements. This is quite encouraging in our efforts to better predict and understand topographically modified precipitation.

4. Results for clean versus polluted air masses

CCN impact orographic precipitation by activation to form cloud droplets resulting in a supercooled liquid water cloud. Sulfate-based CCN do not readily act as ice forming nuclei, nor do ice nuclei act as effective CCN. As such, an augmentation of pollution aerosols acting as CCN will impact the orographic cloud LWC, but will have minimal influence in clouds having little or no supercooled cloud water. Thus, it is possible in some snowfall events, for which polluted air advects in from the west, to have a negligible impact on the cloud microphysics at mountaintop.

West-to-east histogram cross sections through SPL indicate the effects of NCCN on simulated accumulated orographic snowfall (Fig. 13). The CCN effects are identified by comparison among adjoining columns in the figure, with the darker columns representing the polluted simulations. In both events there exists a decrease in total precipitation on the upwind side, and an overall increase on the downwind side of the orographic barrier. The CCN-attributed changes in precipitation are not directly related to the highest point in the terrain or ridgeline, but rather to the cross section of SWE, and this demarcation may be upwind or downwind of the ridgeline depending on horizontal advection of falling snow.

In each of the simulated events the snowfall amounts at SPL and at several grid points adjacent to SPL on both the upwind and downwind slopes were reduced for simulations of increased CCN. The horizontal range of effect was greater for the 11 February case (extending from gridpoint SPL − 8 to SPL + 12), perhaps due to the extended time period of high LWC conditions during this event. Snowfall increased at grid points farther downwind on the lee side (beyond gridpoint SPL + 12 for 11 February case and beyond gridpoint SPL + 10 for 23 February case). The horizontal displacement of maximum snowfall amount was more strongly influenced by CCN in the 11 February case. The increase in precipitation for the downwind zone appears inconsistent with traditional notions of the aerosol indirect effect, but winter orographic cloud systems have more complex precipitation mechanisms and water mass budgets. Hindman (1986) discussed the mass budget of the Park Range winter cloud systems, including blown-over precipitation that falls downwind of the ridge. A reduction in efficiency of snow growth can increase this component and thus modify the relative balance of upwind versus downwind precipitation distribution.

Evaluation of the model results for increased NCCN in both the 11 and 23 February cases indicates a significant difference in both upwind and downwind snowfall amounts for clean versus polluted airmass conditions. Increased NCCN leads to an abundance of smaller cloud droplets with lower riming efficiencies and a consequent inhibition of the snow growth rate during crystal trajectories through a mixed-phase cloud. This process will be termed the “ISPA effect.” The ISPA effect results in snow crystals that have a reduced rime fraction, smaller overall mass, slower fall speeds, and longer horizontal trajectories, which are likely to reach the surface farther downwind. In the case of orographic cloud systems, the longer crystal trajectory often augments displacement of the snowfall beyond mountain ridges. This not only reduces snowpack on the upwind side, but causes an overall reduction in snowpack at the highest (coldest) elevations, which leads to an overall shorter life of the snowpack and reduced storage potential in a lower elevation zone (due to earlier spring melt).

A broader view of impacts of the ISPA effect is seen in Fig. 14 in a surface plan view plot of the simulated snowfall difference between the polluted and clean simulations. The 11 February event shows a broad expanse along the full extent of the mountain range that is impacted by an increase in NCCN. The 23 February event is only impacted by CCN over a small region of the southern Park Range that happens to include SPL. However, in both events the impacted areas display a windward decrease and leeward increase in total precipitation relative to the ridge. The remarkable difference between cases, with respect to CCN loading, is due to the presence of an orographic supercooled liquid water cloud or lack thereof. Recall that LWC measured and simulated at SPL was substantially higher and persisted longer at high values in the 11 February event, while the 23 February event displayed considerably less-persistent cloud water. The seeder–feeder process is primarily effective in modifying precipitation if supercooled water is present in large amounts for an extended period of time. CCN are only effective at modifying precipitation in these conditions if upslope flow becomes water saturated so that droplets begin to nucleate.

A west to east cross section of hydrometeor condensate along the transect through SPL displays the differences in ice and liquid cloud that occur between the clean and polluted simulations (Fig. 15). For both events, the cross section is plotted at the time of maximum cloud water at SPL. This view of the 11 February event further shows the greater cloud LWC, and thus riming potential, as compared with 23 February. However, both events exhibit an ISPA effect in the condensate fields. Comparisons of the clean and polluted simulations for both events reveal several significant modifications. When NCCN was increased, the mountaintop cloud LWC increased because of a reduction in riming. Likewise, the snow mixing ratio increased above the ridge top and the lee slope, and the graupel mixing ratio subsequently decreased in these regions. Here it should be noted that the model microphysics distinguishes graupel from rimed snow by the degree of riming. When riming occurs, the model computes the thermodynamic internal energy of the coalesced snow and liquid water and determines if a liquid layer should occur or if the droplet should freeze instantly to the snow crystals. In conditions of light riming, instantaneous freezing on the snow crystals dominates, and the snow crystals largely retain their shape. However, when riming is heavy, the crystals lose their distinguishing shape and a liquid layer may occur that is due to the addition of substantial amounts of cloud water. If this occurs, snow is transferred to the graupel category and assumes the mass and fall speed characteristics of graupel.

The time series of rimed mass that is removed from the orographic cloud enveloping SPL provides further insight into the impact of increased CCN on precipitation efficiency. Figure 16 displays the vertically integrated collected cloud water that is rimed and removed from the gridcell column above SPL. The time series traces are shown for the clean and polluted simulations for both events. In both cases, the clean simulations undergo greater riming over the duration of the events. When compared to the time series of cloud LWC (Fig. 8), it is evident that the clean and polluted simulations greatly deviate when cloud LWC is high. Low concentrations of cloud water limit the ISPA effect. Earlier figures have shown a relatively small ISPA effect during the 23 February event due to the reduced LWC and shorter cloud lifetimes. This is mirrored in the riming time series by overall lesser amounts of rimed mass in both the clean and polluted simulations when compared to 11 February.

Since we have established that the ISPA effect is most prominent for higher ice water content and liquid water content, we will train our focus primarily on the 11 February event. The observed orographic cloud exhibited variability in CDNC over the period of the event as seen in Fig. 17. Concentrations varied from a few tens of droplets to upward of 250 cm−3. As such, it would be expected that the clean simulations would agree more closely with the observations. The model-simulated time series of droplet number in Fig. 17 reveal several features regarding the impact of CCN initialization upon predicted cloud properties. In the clean case most of the CCN were activated to form droplets, and the time series approximately fits the mean observed droplet values. Without having variable CCN input we cannot expect the model to predict the observed microscale variability in cloud number. While the polluted case was initialized with a maximum NCCN of 1900 cm−3, the predicted supersaturation supported droplet formation up to only half of the available initial maximum number concentration. The resulting droplet concentration in the polluted event is rather high later in the simulation, but given that SPL has observed events with concentrations above 500 cm−3, it is not out of the realm of possibility if aerosol emissions increase in the future. It is possible that these simulated droplet number time series could represent future clean and polluted events. If this is so, the simulated precipitation modification gives us an indication of the potential changes in water resources near regions of sustained orographic liquid water clouds.

The 11 February event was characterized by an extended period in which the temperature at SPL was within the −4° to −8°C range, which is optimal for the ice splintering process. This process can alter the number of ice crystals available for riming. However, the droplet collection efficiency for snow crystals is relatively small below about 100 μm (Wang and Ji 2000). Small ice splinters would have to grow by vapor diffusion prior to surface deposition to have a substantial impact. Figure 18 displays the time series of ice crystal concentration from the model and from the 2D ice crystal probe. Care should be taken when comparing these datasets; the model ice categories contributing to this number concentration constitute snow and aggregates with sizes greater than 125 μm, whereas the 2D probe results omit the first channel and report on crystals larger than about 200 μm. Therefore, the model will tend to report higher numbers compared to the observations. Despite this discrepancy, the time series trend in ice crystal number is quite similar and denotes similar periods of increased ice crystal production. The two model time series from the clean and polluted simulations reveal only minor differences throughout the lifetime of the event; it is unlikely that this minor variability substantially impacts the amount of accrued rime ice.

5. Sensitivity of ISPA effect to CCN concentration

Thus far, we have examined the relative impacts of CCN on orographic snowfall for conditions of clean and polluted CCN profiles. The two snowfall periods on 11 and 23 February represent opposing conditions, with 11 February characterized by heavy riming and 23 February characterized by limited riming. Earlier figures have shown a widespread ISPA effect for 11 February under highly polluted conditions and a minimal effect for 23 February due to a lack of persistent cloud water. As such, the 11 February event was further examined, and a suite of simulations was run to assess the concentration of pollution aerosols that must be present to initiate a discernable aerosol influence. A new set of simulations was run with maximum NCCN set at 100, 300, 500, 800, 1100, 1500, and 1900 cm−3 with a linearly decreasing NCCN vertical profile in each case. Figure 19 depicts the accumulated precipitation difference for maximum NCCN increasing from 100 to 300, 500, 800, and 1100 cm−3. The influence of CCN concentration is observed along the Park Range in Fig. 19, with a precipitation decrease on the windward slope and increase on the leeward slope. The change that results from an increase to 300 cm−3 is quite small and localized and is comparable to what was observed between the clean and polluted simulations from 23 February. However, an NCCN increase to 500 cm−3 was sufficient to initiate a considerable precipitation modification along the north–south extent of the Park Range. Beyond a concentration of 500 cm−3, the ISPA effect continues to expand in area and magnitude, but the relative precipitation change begins to slow as NCCN increases further. At some point the change due to increased CCN concentration likely becomes asymptotic. While this effect is case-specific, once the CCN alter the cloud properties such that the droplet sizes decrease below 10 μm, riming would be dramatically reduced since collection efficiencies between droplets of this size and snow crystals are near zero. High LWC clouds are likely susceptible to higher pollution contents until the presence of the aerosols reduces typical droplet sizes below the riming cutoff.

Another method of examining the relative changes in orographic precipitation is to examine the ratio between upwind and downwind accumulations, and how this ratio varies as pollution aerosols increase in concentration. Givati and Rosenfeld (2004), Rosenfeld and Givati (2006), and Jirak and Cotton (2006) specifically examined the orographic enhancement factor, Ro, which is the ratio of precipitation over elevated terrain to that over relatively flat upwind terrain. Care should be taken when examining Ro since increases or decreases in precipitation in the upwind or downwind sites will modify this ratio. With this in mind, it is still useful to examine the magnitude and location of regional relative precipitation. These studies showed many locations in which the ratio had decreased over long time periods. They suggested that this change occurred because of increases in pollution loading over time, which tended to suppress orographic precipitation. We apply this precipitation ratio in a slightly different manner by comparing the ratio of precipitation on the Park Range lee slope to that on the windward slope for the various pollution concentrations during the 11 February event; we will refer to this ratio as the windward to leeward slope precipitation ratio (RWL). In our analysis, the Park Range was divided along a longitude line of −106.70°W, which runs down the center of the north–south-oriented range. The windward and leeward slope precipitation was summed within the grid cells 17 km to the west and east, respectively, of the ridge demarcation and spanned the north–south length of the domain. The results are summarized in Table 1. The division of the Park Range into the western and eastern slopes at the −106.70°W longitude line resulted in a west–east RWL of 0.98 for the cleanest conditions, which indicates evenly distributed precipitation between the two slopes. As NCCN was increased, RWL subsequently decreased up to the highest CCN values. However, the relative change in RWL with NCCN diminished with increasing CCN. The greatest percent change occurred with an increase in NCCN from 100 to 300 cm−3. The lower rows of Table 1 reveal that for increases in NCCN, there is a systematic loss of precipitation on the windward slope, a gain on the leeward slope, and an increasingly negative percent change in RWL. The largest increase in NCCN results in a −13% shift in snowfall from the western slope to the eastern slope. Despite the substantial downwind shift in precipitation, the overall domainwide precipitation change does not exceed ±1% for any increase in NCCN from 100 up to 1900 cm−3 (Table 1). Furthermore, there is no clear trend in the change in domain total precipitation as the aerosol concentration is increased.

Even for a relatively small increase in NCCN, from 100 to 300 cm−3 there is a loss in surface snowpack of over 10 900 acre-feet of water equivalent from the windward slope. Given that the Park Range marks the local continental divide, the western slope runoff flows into the Colorado River basin. The Elkhead Reservoir and Wolford Reservoir are located west of the Park Range in the Colorado River basin and hold 25 550 and 66 000 acre-feet of water, respectively. To put the ISPA effect into perspective, modification of precipitation in this one snowfall event by slightly higher aerosol concentrations would fill roughly 43% of Elkhead Reservoir or 17% of Wolford Reservoir. Determining a change in runoff is not as simple as filling a reservoir with a given amount of snow water, but it gives one an idea of the potential ISPA impact and prompts the need to examine how the predicted change in snowpack would affect streamflow and reservoir resources through use of a local hydrology model. Givati and Rosenfeld (2007) also address the issue of relating precipitation inhibition by aerosol loading as it pertains to the water resource issues of Israel. They were able to demonstrate that lowered water reserves were not solely the result of increased consumption, but rather, the reservoirs are lower because of reduced regional precipitation stemming from higher aerosol concentrations.

An examination of several bulk microphysical quantities provides further insight into the causal mechanisms of this cloud-modifying ISPA effect. For each simulation performed with various initial NCCN, the domain-summed quantities for the microphysical processes were computed at each time step to compare relative contributions from each process over time. While the cloud processes and properties vary over time because of variations in the model dynamics and thermodynamics, we wanted to focus on the general comparisons among simulations with various NCCN and then draw conclusions based on these bulk comparative time series. These comparisons provide some insight into the primary cloud processes that lead to the ISPA effect on precipitation distribution.

Figure 20 depicts the time series of the domain-summed values (per unit area) of the mass of nucleated cloud water and the total cloud LWC. The domain-summed “nucleated cloud water” was computed by considering only the nucleation source contribution toward the total LWC, whereas vapor diffusion, evaporation, and collection also affect the value of LWC at a given time. The amount of the cloud mass contributed by droplet nucleation alone is directly impacted by NCCN. The time series are shown for each simulation of various NCCN profiles. For the full duration of the simulations, the trend is for higher NCCN to lead to lower cloud water nucleation. In such a polluted situation a large portion of available vapor is consumed by the vapor diffusion growth process once the available CCN are utilized for droplet nucleation. There exists a competition between vapor available for the nucleation of new droplets versus diffusion growth of existing droplets; since riming is reduced in a polluted event, existing droplets remain in the cloud longer and less vapor is used for the nucleation of new droplets. The greatest change in nucleation occurs for an increase in NCCN between 100 and 800 cm−3. Increases to even higher concentrations still lead to a change in nucleation, but the change is proportionally less for the given NCCN increase. Total cloud LWC shows that the cloud water is greatest at nearly all times for the most polluted case with a consistent trend toward lesser cloud water as the NCCN is reduced. In this mixed-phase cloud situation, the cloud nucleated mass and cloud LWC display opposite trends with respect to increasing NCCN, such that higher NCCN leads to reduced cloud droplet nucleation but greater cloud LWC. The relationship between the nucleation amount and the amount of cloud LWC is closely tied to the cloud water removal processes.

Figure 21 displays the corresponding domain-summed time series of cloud droplet vapor diffusion and cloud droplet riming of ice crystals. Along with nucleation, these are two of the key processes determining the amount of supercooled cloud water that is present. The time series of droplet vapor diffusional growth exhibit a trend toward greater growth as NCCN increases, with the greatest change occurring for an increase from 100 to 800 cm−3. The impact diminishes for further increases in CCN loading. An increase in CDNC results in more growth sites for vapor diffusion and a greater droplet surface area to volume ratio; this means that more total droplet surface area is available for vapor growth when CDNC is high. The computed rimed mass in Fig. 16 shows that the greatest riming occurs for the lowest NCCN, with the greatest difference among simulations when riming is maximized because of a strong seeder–feeder process. At moderate riming levels, the difference in rimed mass for the various NCCN simulations is almost nondiscernable beyond a concentration of 500 cm−3, suggesting that a modest increase in pollution aerosols above a clean background state can substantially modify the cloud properties and inhibit snow crystal growth. These results are consistent with observations at SPL by Borys et al. (2003) where riming inhibition, caused by an increase in aerosol sulfate concentration of only 1 μg m−3, and an increase in CDNC from 74 to 310 cm−3 were associated with a decrease in water equivalent snowfall rate of 50%.

The data presented in Table 1 indicate that more than 10 000 acre-feet of water can be displaced from the windward to the leeward side of the Park Range during one event with an increase of NCCN from 100 to 300 cm−3. The Park Range is on the continental divide and at the headwaters of the Colorado River drainage basin. Thus, water in the form of snowpack is taken from the Colorado River basin and displaced to the Platte River basin, on the eastern slope of the Colorado Rockies. This is approximately the amount of water consumed by all the residences, assuming a family of four, in the Yampa River Valley downstream of the Park Range assuming one acre-foot per family per year and the latest population figures for the Yampa Valley of 36 000. On a human scale, the ISPA effect is not trivial.

Though not explored here, there should be a limiting effect in which very little change in precipitation would occur as more NCCN are added to the environment. While this is very case-specific, once the CCN alter the cloud properties, such that the droplet sizes decrease below 10 μm, the riming process would nearly shut down since collection efficiencies between snow and droplets of this size are nearly zero. One would suspect that higher LWC clouds would be susceptible to the highest pollution contents until they modify the droplet sizes below the riming cutoff.

6. Summary and conclusions

The aerosol indirect effects, as discussed by Twomey (1977) and Albrecht (1989), hypothesize that an increase in pollution aerosols leads to an increase in CCN, an increase in CDNC, a decrease in droplet size (first indirect effect), suppression of precipitation due to increased colloidal stability in the cloud, and increased cloud-top albedo due to increased cloud size and lifetime (second indirect effect). These effects modify both precipitation efficiency and net radiative fluxes. These processes can be expected to occur in any liquid-phase cloud system, at least in the formative stages of cloud development prior to possible additional effects from dynamic and thermodynamic feedbacks (e.g., Saleeby and Cotton 2004, 2005; van den Heever and Cotton 2007; Feingold et al. 2003). Wintertime mixed-phase orographic clouds are also sensitive to the degree of pollution aerosols given that they are quasi-stationary features that continually entrain new air parcels as flow lifts and saturates over steep terrain and subsides and dries on the leeward slope (Levin and Cotton 2008). Given their nearly stationary condition, they can act as a consistent feeder source for riming by snowfall seeder clouds aloft, thus modifying the amount of snow water equivalent that reaches the surface (Rauber et al. 1986a,b). Increases in hygroscopic aerosol content that is transported into the orographic cloud can reduce the SWE that accumulates at the surface by modifying the cloud droplet distribution (Borys et al. 2000, 2003).

This work examines the aerosol indirect effect through a modeling study centering on the Park Range of north-central Colorado. SPL is a mountaintop facility located near Steamboat Springs at the crest of the Park Range that is often enshrouded in an orographic cloud containing supercooled liquid water during the winter months. The ISPA II field campaign took place during January–February 2007 at SPL and made measurements of aerosol, CCN, droplet spectra, and snowfall. Two events were selected for modeling the aerosol effect on snowfall: 11–13 and 23–25 February. The former was noted for its long-lived and high–liquid water content orographic feeder cloud and the other for its significant accumulation of low-density snowfall. Simulations of these events were run on telescoping nested grids down to 750-m grid spacing using the CSU RAMS. Simulations were initially run with prescribed clean (100 cm−3) and polluted (1900 cm−3) vertical profiles of NCCN to essentially assess a maximum aerosol indirect effect. Follow-up simulations were executed over a range of NCCN (100, 300, 500, 800, 1100, 1500, and 1900 cm−3) to further delineate what levels of pollution would be required to substantially modify the cloud and surface accumulated snowfall. The primary results from these sensitivity tests are as follows:

  1. For maximized increase in SWE due to the seeder–feeder process, there needs to be a long-lived supercooled orographic cloud with high LWC and large droplet sizes as well as heavy snowfall through the feeder cloud. This will achieve maximum precipitation efficiency for the given available vapor, however, it will also be susceptible to cloud and precipitation modification by pollution aerosols.
  2. Given the presence of a strong seeder–feeder process, the following are true:
    • The addition of NCCN modifies the cloud by increasing the CDNC and reducing the droplet size. This reduces the droplet riming efficiency of ice crystals. Unrimed crystals have a slower fall speed and are advected farther downstream prior to surface deposition. An increase in NCCN therefore results in decreased precipitation on the windward slope and increased precipitation on the leeward slope.
    • For the orographic clouds in this study, the greatest change in the cloud microphysical processes and resulting precipitation per the change in NCCN occurs for an increase in NCCN from roughly 100 to 500 cm−3. While further change occurs for increasing NCCN, the amount of modification becomes increasingly less. For example, the change caused by increasing from 100 to 500 cm−3 was greater than that due to increasing NCCN resulting from a change from 1500 to 1900 cm−3. The variation of precipitation with respect to aerosol loading is not linear.
    • There is no clear trend in the total domain-summed precipitation on the fine grid resulting from increased NCCN (over the range from 100 to 1900 cm−3), and the total change does not exceed ±1%.

This study solely investigates the influence of increasing CCN concentration on winter orographic precipitation. Though not addressed here, an increase in pollution sources may lead to increases in giant CCN (GCCN) and ice nuclei (IN) concentrations as well, though measurements of these aerosols are seldom collected and are not available from the ISPA field study. They remain a greater unknown relative to CCN, though their impact may certainly be of great importance. While increasing NCCN tends to suppress precipitation, it is likely that the addition of NGCCN may have an opposing effect, and an addition of NIN may modify the seeder–feeder process by introducing a greater number of rimers within the cloud. Interactive impacts of these three cloud-altering aerosol types will be investigated in future modeling sensitivity simulations.

Acknowledgments

This research was supported by the National Science Foundation under Grants ATM-0451439 (CSU) and ATM-0451374 (DRI). This project was also funded in part by Subawards S04-44700 (CSU) and S04-44698 (DRI) with UCAR under the sponsorship of NOAA/DOC as part of the COMET Outreach Program. Logistical assistance from the Steamboat Ski and Resort Corporation is greatly appreciated. The Desert Research Institute is an equal opportunity service provider and employer and is a permittee of the Medicine-Bow and Routt National Forests. We acknowledge the NCAR-EOL MAPR and ISS facilities and William Brown and his support staff in the field for their efforts in providing an excellent set of sounding and surface data during our field campaign.

REFERENCES

  • Albrecht, B. A., 1989: Aerosols, cloud microphysics, and fractional cloudiness. Science, 245 , 12271230.

  • Borys, R. D., , and M. A. Wetzel, 1997: Storm Peak Laboratory: A research, teaching, and service facility for the atmospheric sciences. Bull. Amer. Meteor. Soc., 78 , 21152123.

    • Search Google Scholar
    • Export Citation
  • Borys, R. D., , D. H. Lowenthal, , and D. L. Mitchell, 2000: The relationships among cloud microphysics, chemistry, and precipitation rate in cold mountain clouds. Atmos. Environ., 34 , 25932602.

    • Search Google Scholar
    • Export Citation
  • Borys, R. D., , D. H. Lowenthal, , S. A. Cohn, , and W. O. J. Brown, 2003: Mountaintop and radar measurements of anthropogenic aerosol effects on snow growth and snowfall rate. Geophys. Res. Lett., 30 , 1538. doi:10.1029/2002GL016855.

    • Search Google Scholar
    • Export Citation
  • Cober, S. G., , and R. List, 1993: Measurements of the heat and mass transfer parameters characterizing conical graupel growth. J. Atmos. Sci., 50 , 15911609.

    • Search Google Scholar
    • Export Citation
  • Cotton, W. R., and Coauthors, 2003: RAMS 2001: Current status and future directions. Meteor. Atmos. Phys., 82 , 529.

  • Dusek, U., and Coauthors, 2006: Size matters more than chemistry for cloud-nucleating ability of aerosol particles. Science, 312 , 13751378.

    • Search Google Scholar
    • Export Citation
  • Feingold, G., , W. L. Eberhard, , D. E. Veron, , and M. Previdi, 2003: First measurements of the Twomey indirect effect using ground-based remote sensors. Geophys. Res. Lett., 30 , 1287. doi:10.1029/2002GL016633.

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

  • Givati, A., , and D. Rosenfeld, 2007: Possible impacts of anthropogenic aerosols on water resources of the Jordan River and the Sea of Galilee. Water Resour. Res., 43 , W10419. doi:10.1029/2006WR005771.

    • Search Google Scholar
    • Export Citation
  • Greenan, B. J. W., , and R. List, 1995: Experimental closure of the heat and mass transfer theory of spheroidal hailstones. J. Atmos. Sci., 52 , 37973815.

    • Search Google Scholar
    • Export Citation
  • Harrington, J. Y., 1997: The effects of radiative and microphysical processes on simulated warm and transition season Arctic stratus. Ph.D. dissertation, Colorado State University Atmospheric Science Paper 637, 289 pp.

  • Heymsfield, A. J., , and R. M. Sabin, 1989: Cirrus crystal nucleation by homogeneous freezing of solution droplets. J. Atmos. Sci., 46 , 22522264.

    • Search Google Scholar
    • Export Citation
  • Hindman, E. E., 1986: Characteristics of supercooled liquid water in clouds at mountaintops in the Colorado Rockies. J. Climate Appl. Meteor., 25 , 12711279.

    • Search Google Scholar
    • Export Citation
  • Hindman, E. E., , M. A. Campbell, , and R. D. Borys, 1994: A ten-winter record of cloud-droplet physical and chemical properties at a mountaintop site in Colorado. J. Appl. Meteor., 33 , 797807.

    • Search Google Scholar
    • Export Citation
  • Jirak, I. L., , and W. R. Cotton, 2006: Effect of air pollution on precipitation along the Front Range of the Rocky Mountains. J. Appl. Meteor. Climatol., 45 , 236245.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain-Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Levin, Z., , and W. R. Cotton, 2008: Aerosol Pollution Impact on Precipitation: A Scientific Review. Springer, 386 pp.

  • Lynn, B., , A. 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
  • Meyers, M. P., , R. L. Walko, , J. Y. Harrington, , and W. R. Cotton, 1997: New RAMS cloud microphysics parameterization. Part II. The two-moment scheme. Atmos. Res., 45 , 339.

    • Search Google Scholar
    • Export Citation
  • Mitchell, D. L., , R. Zhang, , and R. L. Pitter, 1990: Mass dimensional relationships for ice particles and the influence of riming on snowfall rates. J. Appl. Meteor., 29 , 153163.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1976: Production of secondary ice particles during the growth of graupel by riming. Quart. J. Roy. Meteor. Soc., 102 , 4557.

    • Search Google Scholar
    • Export Citation
  • Pitter, R. L., , and H. R. Pruppacher, 1974: A numerical investigation of collision efficiencies of simple ice plates colliding with supercooled water drops. J. Atmos. Sci., 31 , 551559.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., 1981: Microphysical processes in two stably stratified orographic cloud systems. M.S. thesis, Dept. of Atmospheric Science, Colorado State University Atmospheric Science Paper 337, 151 pp.

  • Rauber, R. M., , L. O. Grant, , D. Feng, , and J. B. Snider, 1986a: The characteristics and distribution of cloud water over the mountains of northern Colorado during wintertime storms. Part I: Temporal variations. J. Climate Appl. Meteor., 25 , 468488.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., , L. O. Grant, , D. Feng, , and J. B. Snider, 1986b: The characteristics and distribution of cloud water over the mountains of northern Colorado during wintertime storms. Part II: Spatial distribution and microphysical characteristics. J. Climate Appl. Meteor., 25 , 489504.

    • Search Google Scholar
    • Export Citation
  • Reinking, R. F., , J. B. Snider, , and J. L. Coen, 2000: Influences of storm-embedded orographic gravity waves on cloud liquid water and precipitation. J. Appl. Meteor., 39 , 733759.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., , and A. Givati, 2006: Evidence of orographic precipitation suppression by air pollution-induced aerosols in the western United States. J. Appl. Meteor. Climatol., 45 , 893911.

    • Search Google Scholar
    • Export Citation
  • 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 descriptions and supercell test simulations. J. Appl. Meteor., 43 , 182195.

    • Search Google Scholar
    • Export Citation
  • Saleeby, S. M., , and W. R. Cotton, 2005: A large-droplet mode and prognostic number concentration of cloud droplets in the Colorado State University Regional Atmospheric Modeling System (RAMS). Part II: Sensitivity to a Colorado winter snowfall event. J. Appl. Meteor., 44 , 19121929.

    • Search Google Scholar
    • Export Citation
  • Saleeby, S. M., , and W. R. Cotton, 2008: A binned approach to cloud-droplet riming implemented in a bulk microphysics model. J. Appl. Meteor. Climatol., 47 , 694703.

    • Search Google Scholar
    • Export Citation
  • Twomey, S., 1977: The influence of pollution on the shortwave albedo of clouds. J. Atmos. Sci., 34 , 11491152.

  • 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
  • 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, P. K., , and W. Ji, 2000: Collision efficiencies of ice crystals at low-intermediate Reynolds numbers colliding with supercooled cloud droplets: A numerical study. J. Atmos. Sci., 57 , 10011009.

    • Search Google Scholar
    • Export Citation
  • Warburton, J. A., , and T. P. deFelice, 1986: Oxygen isotopic composition of central Sierra Nevada precipitation. I: Identification of ice-phase water capture regions in winter storms. Atmos. Res., 20 , 1122.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Model grid configuration. (top) The nested grid setup with grids 1, 2, and 3 denoted by the labels G1, G2, and G3. (middle) A topographic plan view of grid 4 centered over SPL and within G3. The SNOTEL stations of Dry Lake (DRY), Tower (TOW), Rabbit Ears (RAB), Columbine (COL), and Buffalo Park (BUF) are shown for reference. (bottom) A model topographic cross section along the arrowed line in the middle panel; the labels SPL and SPL-4, e.g., represent the locations of SPL and four grid points west of SPL, respectively.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 2.
Fig. 2.

Sum of total measured SWE during January–February 2007 at five measuring sites near SPL.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 3.
Fig. 3.

Time series (UTC) of (a) CCN concentration at 0.3% supersaturation measured at SPL for the duration of the ISPA II field study (January–February 2007), and (b) CCN concentration at 0.3% supersaturation and RH for the snowfall and orographic cloud event of 11–12 Feb 2007.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 4.
Fig. 4.

The 12 Feb 2007 time series (UTC) of NCAR MAPR radar reflectivity, vertical velocity, and wind profiles.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 5.
Fig. 5.

As in Fig. 4, but for 24 Feb 2007.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 6.
Fig. 6.

Storm total accumulated snow water equivalent simulated on grid 4 (mm, shaded) and topography (m, line contours). Note the enhanced precipitation oriented north–south along the Park Range. The simulations shown were initialized with a maximum CCN concentration of 100 cm−3.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 7.
Fig. 7.

Conceptual model of the orographic supercooled water cloud impinging upon the crest of the Park Range of Colorado. Ice particle trajectory and condensate distribution depict the seeder–feeder growth process that occurs because of riming of ice crystals during their passage through the clouds prior to surface deposition (from Borys et al. 2000; adapted from Rauber 1981).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 8.
Fig. 8.

Time series (UTC) of cloud LWC (g m−3) as measured at SPL (light gray) and from RAMS closest grid point to SPL for the simulation initialized with maximum CCN concentration of 100 cm−3 (dark gray) and 1900 cm−3 (black).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 9.
Fig. 9.

As in Fig. 8, but for temperature (°C).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 10.
Fig. 10.

As in Fig. 8, but for RH (%).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 11.
Fig. 11.

As in Fig. 8, but for wind speed (m s−1).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 12.
Fig. 12.

Time series (UTC) of snow water equivalent (mm) from RAMS simulations and the Park Range SNOTEL stations of RAB, COL, TOW, and DRY. The notation PHQ signifies data from the grid point closest to the Patrol Station Headquarters located about 1 km downslope from SPL. Likewise, PHQ-2 signifies data from two grid points to the west. PHQ snow measurements are given by the large black dots.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 13.
Fig. 13.

Variations in storm total SWE (mm) along a west-to-east transect running through SPL. Simulation results with CCN concentrations of 100 and 1900 cm−3 are shown in light and dark shading, respectively.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 14.
Fig. 14.

Simulated total accumulated SWE difference (mm) resulting from an increase in NCCN from clean to polluted conditions. Increased pollution aerosols reduced precipitation on the windward slope and increased precipitation on the leeward slope.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 15.
Fig. 15.

Vertical cross section (denoted in Fig. 1) of RAMS cloud water mixing ratio (g kg−1, shaded), snow (g kg−1 × 100, red solid contours), and graupel (g kg−1 × 100, black dashed contours). Chosen times were near the event maximum in simulated liquid water content at SPL. The event date and maximum CCN concentrations (cm−3) are labeled. (bottom right) Model height is shown in meters.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 16.
Fig. 16.

Time series (UTC) of vertically integrated rimed mass in the gridcell column above SPL for the clean (gray line) and polluted (black line) simulations.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 17.
Fig. 17.

The 11 Feb 2007 time series (UTC) of cloud droplet concentrations from the forward scattering spectrometer probe at SPL and from the model grid point closest to SPL for the clean (RAMS SPL 100) and polluted (RAMS SPL 1900) simulations.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 18.
Fig. 18.

The 11 Feb 2007 time series (UTC) of ice crystal concentrations from the 2D ice probe (2D ice probe counts) at SPL and from the model grid point closest to SPL for the clean (RAMS SPL 100) and polluted (RAMS SPL 1900) simulations. Apparent zero values or data gaps from the 2D probe are missing or unreliable data.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 19.
Fig. 19.

Simulated accumulated SWE difference (mm), from the 11 Feb event, resulting from an increase in NCCN from 100 cm−3 to more polluted conditions with NCCN of (a) 300, (b) 500, (c) 800, and (d) 1100 cm−3.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 20.
Fig. 20.

Time series (UTC) of grid 4 domain-summed (a) cloud nucleated mass and (b) cloud water mixing ratio (LWC) from the 11 Feb event. These are vertically integrated quantities summed across the domain (kg m−2). Plots show the time series for each simulation with different NCCN (concentration labeled; cm−3).

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Fig. 21.
Fig. 21.

As in Fig. 20, but for (a) cloud droplet vapor diffusion and (b) cloud droplet rimed mass.

Citation: Journal of Applied Meteorology and Climatology 48, 5; 10.1175/2008JAMC1989.1

Table 1.

(top) Results from simulations with the (left column) maximum CCN concentration specified. The domain-wide, windward, and leeward precipitation totals and RWL are given. The windward and leeward classifications are described in the text. (bottom) Percent changes in the quantities from the top section that occur because of the increase in CCN, specified in the “CCN change” column.

Table 1.
Save