• Boe, B. A., , and P. J. DeMott, 1999: Comparisons of Lohse wing-tip nuclei generators and burn-in-place pyrotechnics in the North Dakota Cloud Modification Project. J. Wea. Modif., 31, 109118.

    • Search Google Scholar
    • Export Citation
  • Breed, D., , R. Rasmussen, , C. Weeks, , B. Boe, , and T. Deshler, 2014: Evaluating winter orographic cloud seeding: Design of the Wyoming Weather Modification Pilot Project (WWMPP). J. Appl. Meteor. Climatol, 53, 282299, doi:10.1175/JAMC-D-13-0128.1.

    • Search Google Scholar
    • Export Citation
  • DeMott, P. J., 1997: Report to the North Dakota Atmospheric Resource Board and Weather Modification Incorporated on tests of the ice nucleating ability of aerosols produced by the Lohse airborne generator—1997. Department of Atmospheric Sciences, Colorado State University, Fort Collins, CO, 14 p.

  • DeMott, P. J., , A. B. Super, , G. Langer, , D. C. Rogers, , and J. T. McPartland, 1995: Comparative characterizations of the ice nucleus ability of AgI aerosols by three methods. J. Wea. Modif., 27, 116.

    • Search Google Scholar
    • Export Citation
  • DeMott, P. J., and Coauthors, 2011: Resurgence in ice nuclei measurement research. Bull. Amer. Meteor. Soc., 92, 16231635, doi:10.1175/2011BAMS3119.1.

    • Search Google Scholar
    • Export Citation
  • Garvey, D. M., 1975: Testing of cloud seeding materials at the cloud simulation and aerosol laboratory, 1971–1973. J. Appl. Meteor., 14, 883889, doi:10.1175/1520-0450(1975)014<0883:TOCSMA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Heimbach, J. A., , A. B. Super, , and J. T. McPartland, 1977: A suggested technique for the analysis of airborne continuous ice nucleus data. J. Appl. Meteor., 16, 255261, doi:10.1175/1520-0450(1977)016<0255:ASTFTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Heimbach, J. A., , A. B. Super, , B. A. Boe, , G. Langer, , and J. T. McPartland, 2008: Comparison of two acoustic ice nucleus counters. J. Wea. Modif., 40, 5463.

    • Search Google Scholar
    • Export Citation
  • Holroyd, E. W., III, , J. T. McPartland, , and A. B. Super, 1988: Observations of silver iodide plumes over the Grand Mesa of Colorado. J. Appl. Meteor., 27, 11251144, doi:10.1175/1520-0450(1988)027<1125:OOSIPO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Holroyd, E. W., III, , J. A. Heimbach, , and A. B. Super, 1995: Observations and model simulation of AgI seeding within a winter storm over Utah’s Wasatch Plateau. J. Wea. Modif., 27, 3556.

    • Search Google Scholar
    • Export Citation
  • Huggins, A. W., 2007: Another wintertime cloud seeding case study with strong evidence of seeding effects. J. Wea. Modif., 39, 936.

  • Langer, G., 1973: Evaluation of NCAR ice nucleus counter. Part I: Basic operation. J. Appl. Meteor., 12, 10001011, doi:10.1175/1520-0450(1973)012<1000:EONINC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Langer, G., , J. Rosinski, , and C. P. Edwards, 1967: A continuous ice nucleus counter and its application to tracking in the troposphere. J. Appl. Meteor., 6, 114125, doi:10.1175/1520-0450(1967)006<0114:ACINCA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Langer, G., , G. Cooper, , C. T. Nagamoto, , and J. Rosinski, 1978: Ice nucleation mechanisms of submicron monodispersed silver iodide, 1,5-dihydroxynaphathalene and phloroglucinol aerosol particles. J. Appl. Meteor., 17, 10391048, doi:10.1175/1520-0450(1978)017<1039:INMOSM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., , and L. O. Grant, 1986: 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, doi:10.1175/1520-0450(1986)025<0489:TCADOC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Reynolds, D. W., 1988: A report on winter snowpack augmentation. Bull. Amer. Meteor. Soc., 69, 12901300, doi:10.1175/1520-0477(1988)069<1290:AROWSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rogers, D. C., , P. J. DeMott, , S. M. Kreidenweis, , and Y. Chen, 2001: A continuous-flow diffusion chamber for airborne measurements of ice nuclei. J. Atmos. Oceanic Technol., 18, 725741, doi:10.1175/1520-0426(2001)018<0725:ACFDCF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., 1990: Winter orographic cloud seeding in the intermountain west. J. Wea. Modif., 22, 106116.

  • Super, A. B., 1999: Summary of the NOAA/Utah Atmospheric Modification Program: 1990-1998. J. Wea. Modif., 31, 5175.

  • Super, A. B., , and B. A. Boe, 1988: Microphysical effects of wintertime cloud seeding with silver iodide over the Rocky Mountains, Part III: Observations over the Grand Mesa, Colorado. J. Appl. Meteor., 27, 11661182, doi:10.1175/1520-0450(1988)027<1166:MEOWCS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , and J. A. Heimbach Jr., 1983: Evaluation of the Bridger Range winter cloud seeding experiment using control gages. J. Climate Appl. Meteor., 22, 19892011, doi:10.1175/1520-0450(1983)022<1989:EOTBRW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , and J. A. Heimbach Jr., 1988: Microphysical effects of wintertime cloud seeding with silver iodide over the Rocky Mountains. Part II: Observations over the Bridger Range, Montana. J. Appl. Meteor., 27, 11521165, doi:10.1175/1520-0450(1988)027<1152:MEOWCS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , J. T. McPartland, , and J. A. Heimbach, 1975: Field observations of the persistence of AgI-NH4I-acetone ice nuclei in daylight. J. Appl. Meteor., 14, 15721577, doi:10.1175/1520-0450(1975)014<1572:FOOTPO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , B. A. Boe, , J. A. Heimbach, , J. T. McPartland, , and G. Langer, 2010: Comparison of silver iodide outputs from two different generators and solutions measured by acoustic ice nucleus counters. J. Wea. Modif., 42, 4960.

    • Search Google Scholar
    • Export Citation
  • Warburton, J. A., , R. H. Stone III, , and B. L. Marler, 1995: How the transport and dispersion of AgI aerosols may affect detectability of seeding effects by statistical methods. J. Appl. Meteor., 34, 19291941, doi:10.1175/1520-0450(1995)034<1929:HTTADO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • WMI, 2005: Wyoming Level II Weather Modification Feasibility Study. Wyoming Water Development Commission Rep., 151 pp. [Available online at http://charybdis.wrds.uwyo.edu/weathermod/Report.pdf.]

  • Xue, L., , X. Chu, , R. Rasmussen, , D. Breed, , B. Boe, , and B. Geerts, 2014: The dispersion of silver iodide particles from ground-based generators over complex terrain. Part II: WRF large-eddy simulations versus observations. J. Appl. Meteor. Climatol., 53, 13421361, doi:10.1175/JAMC-D-13-0241.1.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    The relative locations of the ground-based cloud-seeding (AgI aerosol) generators with respect to terrain, target precipitation measurement sites, and other relevant instrumentation. The identifier of each generator is provided, because generators frequently differed from case to case.

  • View in gallery

    Three periods of IN sampling at the MMC surface detection site during WWMPP RSE cases when the MB was seeded. In each period, seeding began at t = 0 min and ended at t = 240 min (4 h). (a) During this period, with four generators operating, a significant AgI IN plume was observed. Plumes of this magnitude at the surface were observed only infrequently. (b) This plume is more typical of MB cases; with five generators running, a significant plume was observed on the order of an hour after seeding began, but plume meander during the case resulted in a gap, at least at the surface MMC site. (c) This period shows an instance in which very little of the IN plume was observed, even though all eight generators were functioning. The lack of a stronger plume was not readily explainable. Instances such as the one that is depicted in (c) were observed infrequently.

  • View in gallery

    In some cases in which the SM was seeded and prevailing winds were southwesterly, the plume lingered in the MB longer than expected. In the example shown here, the plume was still present through the 120-min buffer that ended at t = 0 of the next case, when seeding began again. To address this situation, the buffer period was doubled to 240 min in the finalization of the RSE design. The dashed trace indicates the placement of a total aerosol filter over the AINC intake. The immediate drop to zero (0.1 on the ordinate, indicating no IN were detected during that minute) confirmed that IN were still present in the ambient air. Ice nucleus concentrations on the logarithmic ordinate shown as 0.02 indicate that the AINC was not collecting data.

  • View in gallery

    Soundings that were released from KSAA at 2200 UTC 16 Feb 2011 (red) and at 0100 UTC 17 Feb 2011 (blue). The day was characterized by a well-mixed layer from the surface to well above crest height, where a weak inversion strengthened during the flight.

  • View in gallery

    The stratocumulus deck present at 0017 UTC 17 Feb 2011, as photographed from the copilot seat by the flight scientist (T. Krauss) during pass 7, heading northwest. The snowcapped crest of the Snowy Range (the highest portion of the MB) appears on the far-right horizon, just below the cloud base.

  • View in gallery

    The flight track for the 16 Feb 2011 mission (a) in plan view, with respect to the active ground-based IN generator sites, and (b) in a vertical profile facing east on a plane aligned north–south. In (a), passes 1 and 2 are those located farthest west and the other six were flown near the crest line, the location of the MMC IN sampling site is shown by the black circle, and the two sets of target precipitation gauges are shown by the open circles with pluses. The plotted concentrations are lagged by 80 s, the mean time required for IN to enter the instrument, nucleate, grow, and be counted. As suggested by the rapidly varying altitudes seen in (b), the turbulence was considerable.

  • View in gallery

    The IN plume observed on the surface at the MMC site is shown for 16 Feb 2011. The AINC unit 1 began collecting IN data about 55 min after the five ground-based IN generators were activated (2148 UTC). The maximum 10-s mean IN concentrations measured by AINC unit 2 during each of the eight passes are also plotted (black diamonds). Ice nucleus concentrations on the logarithmic ordinate shown as 0.02 indicate that the AINC was not yet collecting data. Values of 0.1 indicate no IN were detected during that minute. The airborne concentrations decrease with time as aircraft altitude was increasing and the plumes thinned. Seeding ended at 0110 UTC, 198 min after seeding start. The plume decay at the MMC began shortly thereafter.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 42 42 11
PDF Downloads 39 39 6

The Dispersion of Silver Iodide Particles from Ground-Based Generators over Complex Terrain. Part I: Observations with Acoustic Ice Nucleus Counters

View More View Less
  • 1 * Weather Modification, Inc., Fargo, North Dakota
  • 2 Department of Atmospheric Sciences, University of North Carolina at Asheville, Asheville, North Carolina, and Springvale, Maine
  • 3 Krauss Weather Services, Inc., Red Deer, Alberta, Canada
  • 4 National Center for Atmospheric Research, Boulder, Colorado
  • 5 ** Department of Atmospheric Science, University of Wyoming, Laramie, Wyoming
  • 6 U.S. Bureau of Reclamation, Evergreen, Colorado
© Get Permissions
Full access

Abstract

Part I of this paper presents the results from a series of plume-tracing flights over the Medicine Bow and Sierra Madre Ranges in south-central Wyoming. These flights, conducted during February and early March of 2011, were part of the Wyoming Weather Modification Pilot Project. Effective targeting of ground-based silver iodide plumes to supercooled clouds has long been a problem for winter orographic cloud-seeding projects. Surface-based ice nucleus (IN) measurements made at a fixed location near the Medicine Bow Range target area had confirmed the effective transport of IN plumes in many cases, but not all. Airborne plume tracing, undertaken to further illuminate the processes involved, provided additional insight into the plume behavior while providing physical measurements that were later compared with large-eddy-simulation modeling (Part II). It was found that the plumes were most often encountered along the flight paths set out in the experimental designs and, in the absence of convection, appear to be mostly confined to the lowest 600 m above the highest terrain. All passes above 600 m above ground level revealed IN concentrations greater than background levels, however. An estimate of IN flux measured over the Medicine Bow Range was approximately 85% of that produced by the five ground-based IN generators active at the time.

Retired.

The National Center for Atmospheric Research is sponsored by the National Science Foundation.

Corresponding author address: Bruce Boe, Weather Modification, Inc., 3802 20th St. N, Fargo, ND 58102. E-mail: bboe@weathermodification.com

Abstract

Part I of this paper presents the results from a series of plume-tracing flights over the Medicine Bow and Sierra Madre Ranges in south-central Wyoming. These flights, conducted during February and early March of 2011, were part of the Wyoming Weather Modification Pilot Project. Effective targeting of ground-based silver iodide plumes to supercooled clouds has long been a problem for winter orographic cloud-seeding projects. Surface-based ice nucleus (IN) measurements made at a fixed location near the Medicine Bow Range target area had confirmed the effective transport of IN plumes in many cases, but not all. Airborne plume tracing, undertaken to further illuminate the processes involved, provided additional insight into the plume behavior while providing physical measurements that were later compared with large-eddy-simulation modeling (Part II). It was found that the plumes were most often encountered along the flight paths set out in the experimental designs and, in the absence of convection, appear to be mostly confined to the lowest 600 m above the highest terrain. All passes above 600 m above ground level revealed IN concentrations greater than background levels, however. An estimate of IN flux measured over the Medicine Bow Range was approximately 85% of that produced by the five ground-based IN generators active at the time.

Retired.

The National Center for Atmospheric Research is sponsored by the National Science Foundation.

Corresponding author address: Bruce Boe, Weather Modification, Inc., 3802 20th St. N, Fargo, ND 58102. E-mail: bboe@weathermodification.com

1. Introduction and background

Wintertime orographic snow augmentation cloud-seeding operations are common in much of the American West. During the winter of 2012/13, such operations were ongoing in California, Colorado, Idaho, Nevada, Utah, and Wyoming (see online at http://weathermodification.org/projectlocations.php). For the most part these programs are operational in nature; that is, they are conducted for effect and not for proof of concept. Sponsors include public utilities, municipalities, irrigation districts, and state agencies. Many have been ongoing for decades.

Reynolds (1988) and Super (1990) suggested that a significant problem in seeding winter orographic clouds is the uncertainty of adequate targeting zones of supercooled liquid water (SLW). Some weather modification experiments may have failed because the targeting issue was not adequately addressed. There is a large body of transport and dispersion studies that have been conducted in the western United States. Among the earlier of these is the Bridger Range Experiment (BRE) of Montana (Super and Heimbach 1983), a randomized exploratory experiment that was conducted during the winters from 1969 to 1972. During the early winters, airborne silver iodide (AgI) sampling was done over the north–south-oriented Main Bridger Ridge in “visual flight rules” (VFR) flight conditions. The VFR conditions allowed sampling at near-surface altitudes. The Main Bridger Ridge crest line is approximately 2600 m above mean sea level (MSL), and the target area is approximately 2000 m MSL, which allowed sampling over the target at the approximate elevation of the Main Bridger Ridge in 1985 (Super and Heimbach 1988). Two AgI generators in 1970–72 were located two-thirds of the way up the west slope of the Main Bridger Ridge. Plumes from these two generators were consistently traced over the Main Bridger Ridge toward the target. Plume widths for the southern generator site were 10°–30°. Insufficient data precluded estimation of widths for the northern seeding site.

In January of 1985 airborne physical measurements that included acoustic ice nucleus counter (AINC) data were taken in the same experimental area (Super and Heimbach 1988). Reynolds (1988) pointed out that at that time the BRE was the only statistical exploratory winter experiment that was verified with physically consistent measurements. In 1985, tracing flights were conducted under “instrument flight rules” (IFR) conditions over the north–south-oriented target range about 17 km to the east of the randomized experiment’s southern seeding site. By Federal Aviation Administration (FAA) regulations, IFR sampling is allowed 600 m above ground level (AGL) above the highest terrain within 9.3 km (5 nautical mi.). A special FAA waiver allowed sampling down to 300 m AGL. In 1985, plumes from the southern seeding site were tracked over the target area. In general, the bulk of the plumes remained below 300 m AGL. The plume widths ranged from 5 to 8 km over the target. The ice nucleus (IN) concentrations were sufficiently high to effectively seed winter orographic clouds (≥10 L−1).

Direct observations of surface-released AgI that are germane to this paper’s design were made in February and March of 1986. Holroyd et al. (1988) and Super and Boe (1988) describe AgI plume tracing over the Grand Mesa of Colorado using an airborne AINC. Both aerial and ground releases were tracked, but the aerial releases are not discussed. Like the BRE, surface releases were from sites that were located more than midway up the windward side of the mesa. For the surface-released plumes, the median instantaneous width was 15° and the median meandering width was 38°. The median plume height was greater than 0.5 km. The plumes were discernible up to 40 km downwind and 80 min after release. The character of AgI plumes and resulting IN was analogous to those of the BRE. Likewise, there was consistent delivery of seeding material to regions of orographic cloud SLW.

Warburton et al. (1995) documented how inadequate targeting of seeding material degraded the detection of a seeding signal in two target areas of the central Sierra Nevada. Snow chemistry was used to define areas of surface-released AgI above a background of 2 parts per trillions (ppt) in and around the Truckee–Tahoe and Lake Almanor target areas. Westerly winds produced statistical results that were in line with physical expectations; the southerly wind partition, in contrast, produced little silver in the Lake Almanor snow samples but gave evidence of upwind control sites being contaminated. This result suggests that improper targeting may explain why the Lake Almanor seeding did not produce a seeding effect.

Direct observations of surface-released AgI were made in the early winters of 1991 and 1994 over the Wasatch Plateau of central Utah under the auspices of the National Oceanic and Atmospheric Administration/Utah Atmospheric Modification Program (Super 1999; Huggins 2007). The target area was the north–south Wasatch Plateau with an elevation of approximately 2900 m MSL, 1200 m AGL above the windward San Pete Valley. To the west of the valley are the San Pitch Mountains, about 800 m in elevation above the valley. Part of the field program involved aerial sampling of eight multiple operational surface seeding generators in the San Pete Valley during 1991 and 1994. There were two surface AINCs: one sampled at a high-altitude observation site near the crest of the plateau and the other was mobile, being in an instrumented van. A third AINC was mounted in an aircraft along with cloud physics instrumentation. A special exemption allowed IFR flights down to 300 m above nearby highest terrain. The high-altitude generators’ plumes traversed the plateau more reliably and with higher concentrations than were associated with the valley generators. Eighty percent of the valley plumes traversed the plateau, but the number of IN that could be activated at warmer temperatures were too few to be effective below 300 m AGL.

a. The Wyoming program

After being approached by conservation districts, the Wyoming Water Development Commission in 2004 decided to explore the potential for such operations in that state, beginning with a feasibility study. The results of that study, as reported by Weather Modification, Inc. (WMI 2005), suggested that potential existed, but the water development commission decided that the efficacy should be demonstrated before any state-sponsored operations were undertaken. Therefore, the Wyoming Weather Modification Pilot Project (WWMPP; Breed et al. 2014) was created to obtain additional information about the physical processes involved and to quantify the changes in precipitation thus produced.

The WWMPP was funded for nine winter seasons, beginning with the winter of 2005/06, through the winter season of 2013/14. The first three seasons were used to complete environmental work, equipment deployment, and collection of physical data prior to the initiation of a randomized statistical experiment (RSE). The WWMPP applied ground-based, remote-controlled AgI IN generators manufactured by WMI, operated upwind of three Wyoming mountain ranges (Breed et al. 2014). This winter orographic seeding strategy relies upon ambient winds to transport and mix the ice nuclei upward into clouds bearing supercooled liquid water, where snow formation occurs.

The primary purpose of this paper is to present observations of AgI plumes produced by the WWMPP ground-based IN generators, as measured above the Medicine Bow Range (MB). The airborne observations provide a basis for validation of high-resolution large-eddy simulation (LES) (Xue et al. 2014, hereinafter Part II) and also provide some physical evidence of the generator IN output and transport and dispersion to the regions above the MB that would be occupied by supercooled cloud during seeding.

b. The ice nucleus generators

The IN generators used for the WWMPP burn an AgI solution that after combustion yields a nucleus of AgI0.8Cl0.2-NaCl formulation (DeMott 1997), shown to function effectively by the condensation–freezing mechanism. All generator components through which the solution flows are corrosion resistant, being either stainless steel or appropriate plastics. Solar panels provide the power. Connection for activation and monitoring is done via satellite. When a generator is activated, the propane burner is ignited within the wind shroud, open at the top and bottom. Once flame temperature confirms propane ignition, solution flows through an Intek, Inc., Rheotherm flowmeter, up to the burner head, where it passes through a nozzle and is atomized into the propane flame and burned. Combustion occurs at 5.7 m AGL. When the seeding is complete, the solution flow ceases, and the solution line is purged with nitrogen. When the purge is complete, the propane flow ends, and the burner goes out. Real-time flow rates are provided by the Rheotherms, and flame temperatures are measured by thermocouples. In addition, system pressure (to push solution up to the burner) and battery voltage are monitored. Four of the generator sites also have Vaisala, Inc., WXT-510 weather stations that add surface weather observations to the generator data streams. Thus, ignition is always confirmed, and flow rate known for each generator, in each seeding event.

c. The ice nucleus counters

One AINC was flown and another operated simultaneously at the surface. These two units were previously run side by side while sampling from a common source (Heimbach et al. 2008). The comparative tests showed reproducible and quantitative agreement. Using ratios of total IN counts for each test adjusted for the difference in sample rates gave ratios of the airborne AINC to the ground-based AINC ranging from 1.04 for condensation–freezing nuclei to 1.66 for contact nuclei. The IN produced by the WMI generators and solution used in Wyoming function by the condensation–freezing process (DeMott 1997).

2. Measurements with the AINC

Detailed specifications of the two AINCs used in this study are presented in Table 1 of Heimbach et al. (2008). Following their convention, unit 1 herein refers to an AINC built by Langer, the inventor of the instrument (Langer 1973), and unit 2 refers to one that was built by the second author.

Sample air is drawn into the AINC’s humidifier at approximately 10 L min−1 and a side stream of sodium chloride (NaCl) cloud condensation nuclei (CCN) are added to the humidified air prior to entering the cloud chamber. The sample, now fortified with additional moisture and CCN, flows into the cloud chamber where a supercooled cloud is maintained at −20°C, as measured in the lower chamber. There, those IN active at the cloud’s temperature or warmer nucleate ice crystals that grow to detectable size: ~20-μm diameter. Ice crystals exiting the base of the cloud chamber rapidly accelerate as they pass through a glass Venturi tube, producing an audible “click.” The sound is detected by a microphone connected to an electronic signal processor. Each legitimate count triggers a transistor–transistor logic (TTL) signal that is sent to a data system for real-time counting, display, and archiving at 1 Hz.

In this effort, the AINC chambers were operated at temperatures of −18° and −20°C in AINC unit 1 and unit 2, respectively. These temperatures are easily cold enough to grow ice particles nucleated from AgI, which can be activated at −8°C. These crystals have ample time to grow to detectable sizes (20 μm) before exiting the chamber through the sensor (Heimbach et al. 2008). Most natural IN activate at significantly colder temperatures, however, and often do not have time to grow to detectable sizes in the AINC chamber. To verify that this is indeed the case in the MB and Sierra Madre Range (SM) of Wyoming, the ground-based AINC (unit 1) was operated at −20°C for many hours in the absence of seeding during several winters (2008–11), and the airborne unit likewise was operated more or less continuously from −18° to −20°C while in flight. In all cases, the “background” was observed to be very low, well less than 0.1 L−1. Many times, tens of minutes passed with nothing detected at all. Thus, “IN plumes” as reported herein are anything more than single isolated IN detections. The discernment of “plumes” defined in this way is not difficult [e.g., see the “total counts” column of Table 3 (described below in more detail) in which raw counts (individual IN detections not converted to IN per liter) are reported]. The weakest plume observed in this case study produced 118 counts in a single pass, which, corrected for sample rate, translated to a peak concentration of 87 L−1.

The WMI IN generators have not been directly calibrated for yield of IN because of the unavailability of a suitable U.S. testing facility such as the Colorado State University isothermal cloud chamber (ICC) previously used for this purpose (e.g., DeMott et al. 1995). Although other, similar generators had been tested prior to the decommissioning of the ICC, the AgI-complex solution and generators presently being used in the WWMPP were not. A nearly identical solution was tested at the ICC when burned with an airborne-style IN generator (DeMott 1997), however, although not with wind-tunnel drafts at speeds commensurate with those typical of ground-based generators.

Neither the ICC nor the AINCs precisely mimic typical winter orographic clouds (Boe and DeMott 1999). Liquid water content (LWC) within the ICC was generally reported for two values, 0.5 and 1.5 g m−3, with corresponding cloud droplet concentrations of about 2100 and 4300 droplets per cubic centimeter, respectively (Garvey 1975). New cloud droplets were continuously introduced to maintain LWC, and ice crystals were frequently collected on microscope slides for up to 50 min after aerosol introduction. The ICC droplet concentration and LWC values were consistently greater than most winter measurements within orographic clouds of the Intermountain West (e.g., Rauber and Grant 1986).

Greater droplet concentrations are required within AINC cloud chambers to enhance the probability of nucleation and ice crystal growth within the limited chamber residence time—typically about 1 min. Table 2 of Langer (1973) indicates that, for cloud and humidifier temperatures employed in this effort, LWC varied from about 1.5 g m−3 at the cloud-chamber top inlets to about 2 g m−3 by the bottom exit. Calculations and observations suggested that typical droplet concentrations were in the range from 3 × 104 to 8 × 104 cm−3. The AINC functions by forcing nucleation by whatever process to maximize detection of AgI aerosol concentrations. These and other differences from natural clouds suggest some caution should be used in applying ICC or AINC results to winter orographic clouds. Despite these limitations, AINCs have been successfully used in many studies to document the presence and extent of AgI plumes (Holroyd et al. 1988, 1995; Super et al. 1975).

Although the AINC does not provide the quality of IN measurement afforded by more sophisticated instrumentation such as the continuous-flow diffusion chamber (Rogers et al. 2001; DeMott et al. 2011), it is small enough to be flown effectively in a small aircraft. When operated with the complete knowledge of their principles and implementation of proper procedural safeguards, it provides a reliable and reproducible means by which AgI-complex IN plumes can be consistently measured. The interested reader is referred to Langer et al. (1967) and Langer (1973) for further details of the design, functionality, and operation of the AINC.

Before ambient air sampling began in each airborne experiment, phloroglucinol mist, an organic nucleant, was injected into the sample airstream to verify the functionality of the AINC. This procedure was repeated at the conclusion of each sampling period. As shown by Langer et al. (1978), phloroglucinol is an excellent surrogate for AgI. It has a very short persistence time and therefore affords minimal contamination potential. The AINC operators continuously monitored all instrumentation parameters throughout each flight, thus quickly identifying problems if and when they occurred.

3. Ice nucleus measurements in the WWMPP

Physical measurements and numerical modeling are part of the WWMPP evaluation (Breed et al. 2014). The IN output produced by the generators was established (DeMott et al. 1995; Super et al. 2010), but additional verification of the desired transport and dispersion of these plumes was also deemed necessary to verify targeting in these ranges. Of specific interest was the time required for residual seeding plumes to vacate both the target and control areas in the randomized, crossover design. The most affordable avenue was to measure the IN in the MB, which sometimes is downwind of the SM, which was also targeted. Surface-based IN measurements were conducted initially, because they allowed continuous sampling over many hours.

Table 1 provides the relative elevations of the key project facilities. All of the airborne-plume-tracing efforts reported herein used at least some IN generators sited at relatively high elevations—some less than 400 m in elevation below that of the mean crest line (3180 m). Some sites are farther removed from the crest (MB09, MB10, MB11, SM09, SM10, and SM11) and thus are lower, but none are sited near the valley floor. None of these plume-tracing efforts utilized only the lower-elevation IN generators, but it would be interesting to do this in the future to delineate the flow regimes in which they are effective. All flights during the 2011 intensive operations period (IOP) were conducted in well-mixed conditions at low levels with a stable layer above save one, that being the late portion of the 18 February 2011 case when the atmosphere became unstable.

Table 1.

The site identifiers (ID), types, names, and elevations for the utilized field facilities. Here, GLEES is the Glacier Lakes Environmental Experimental Site.

Table 1.

a. Surface IN measurements

To begin the verification of IN targeting in Wyoming, the unit-1 AINC was initially deployed during February of 2008, at 3028 m MSL, on the eastern slope of the MB, within ~1 km of the target-area precipitation gauges. The surface sampling site was selected because of the availability of lodging and power and because of fiscal constraints. A site nearer the crest line or even upwind of it (beneath the aircraft passes) would have been preferable.

This unit was operated during seeding events to begin characterization of the IN plumes released from the upwind (western) side of the MB. The relative locations of all of the ground-based generator sites, the high-resolution precipitation gauges used for WWMPP evaluation, the ground-based AINC site, and the WWMPP rawinsonde release site are shown in Fig. 1.

Fig. 1.
Fig. 1.

The relative locations of the ground-based cloud-seeding (AgI aerosol) generators with respect to terrain, target precipitation measurement sites, and other relevant instrumentation. The identifier of each generator is provided, because generators frequently differed from case to case.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

The initial surface measurements obtained in 2008 produced mixed results. Ice nucleus plumes were observed at the surface sampling site during some of the 10 ground-based seeding events observed but not all. The reasons for the plume absence in some cases were not established.

The original experimental design required a 120-min no-seed “buffer” between cases to allow residual IN to be flushed from both the MB and SM target areas, but the early surface IN sampling revealed that, in some cases in which the SM was seeded, the IN plume was detected in the MB well after the 120 min had elapsed. This resulted in the buffer period being doubled to 240 min prior to the beginning of the RSE.

When not detected at the surface sampling site, the plume may have been nucleated or scavenged, and thus removed from the free air, or perhaps preferentially channeled by the terrain away from the surface sampling site. Alternatively, the IN may have been carried farther aloft and past the target well above it. In some cases, the flow may have simply been insufficient for transport over the barrier. Taken together, these uncertainties prompted expanded surface-based IN sampling during the three subsequent winter seasons, during which an AINC was operated at the Mountain Meadow Cabins (MMC; see Table 1) during most seeding events, regardless of the seeding decision. Plots of three typical plume traces are provided in Fig. 2. These observations did not fully alleviate the lingering uncertainties about consistent targeting.

Fig. 2.
Fig. 2.

Three periods of IN sampling at the MMC surface detection site during WWMPP RSE cases when the MB was seeded. In each period, seeding began at t = 0 min and ended at t = 240 min (4 h). (a) During this period, with four generators operating, a significant AgI IN plume was observed. Plumes of this magnitude at the surface were observed only infrequently. (b) This plume is more typical of MB cases; with five generators running, a significant plume was observed on the order of an hour after seeding began, but plume meander during the case resulted in a gap, at least at the surface MMC site. (c) This period shows an instance in which very little of the IN plume was observed, even though all eight generators were functioning. The lack of a stronger plume was not readily explainable. Instances such as the one that is depicted in (c) were observed infrequently.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

In some cases in which the randomized seeding was conducted in the SM and flow was southwesterly, IN from the SM generators were observed to linger within the MB (at the AINC site) longer than the 2-h buffer period between cases that was initially allotted (Fig. 3).

Fig. 3.
Fig. 3.

In some cases in which the SM was seeded and prevailing winds were southwesterly, the plume lingered in the MB longer than expected. In the example shown here, the plume was still present through the 120-min buffer that ended at t = 0 of the next case, when seeding began again. To address this situation, the buffer period was doubled to 240 min in the finalization of the RSE design. The dashed trace indicates the placement of a total aerosol filter over the AINC intake. The immediate drop to zero (0.1 on the ordinate, indicating no IN were detected during that minute) confirmed that IN were still present in the ambient air. Ice nucleus concentrations on the logarithmic ordinate shown as 0.02 indicate that the AINC was not collecting data.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

After three winter seasons of ground-based AINC operations at the MMC, the uncertainties remained. To be specific, knowledge of actual plume behavior above the ground and in cloud was still lacking. Although transport from the ground-based generators over the MB to a single surface site near the target gauges had been demonstrated, there were still occasions on which plumes were not observed.

At that time, modeling of WWMPP plume behavior had been limited to runs of a Weather Research and Forecasting (WRF) model real-time, four-dimensional (x, y, z, and t) data-assimilation setup on a cluster at the National Center for Atmospheric Research. The horizontal resolution of the innermost grid was 2 km (Breed et al. 2014) The obvious solution was to fly an AINC to obtain some airborne measurements.

b. The airborne plume measurement effort

Two AINCs were available. For the 2011 IOP presented here, AINC unit 1 was deployed at the same MMC site used during previous seasons while unit 2 was installed in a twin-engine turboprop aircraft based in nearby Cheyenne, Wyoming.

The instrumented aircraft, a Piper Cheyenne II, contained a Science Engineering Associates, Inc., M300 data acquisition system that recorded date, time, GPS position, aircraft heading and attitude, true airspeed, altitude, wind speed and direction (u, υ, and w), total temperature, dewpoint temperature, and cloud liquid water content. In addition, the number of ice nuclei counted each second by the AINC was also recorded. The data-system observations were further complemented by notes kept by the flight scientist and the AINC operator.

There were some limitations on the extent and character of the airborne measurements. Actual seeding events were, of course, conducted during storm conditions, when the mountains were enshrouded by orographic cloud. Flight operations at altitudes low enough to sample the plumes released from the surface are not possible in such conditions; IFR are then operative, and regulations do not allow flight at altitudes below ~600 m (2000 ft) above the highest terrain in the area.

Also, IN sampling with an AINC is continuous, but the response is lagged and smoothed. Each IN has to enter the AINC cloud chamber, nucleate an ice crystal, and then grow large enough to be detected when it exits the chamber. The fastest that this sequence can occur is typically 25 s (in a dense IN plume), but the average time to detect an IN is 80 s (Heimbach et al. 1977). In addition, when a dense seeding plume is encountered during flight so many IN activate that up to 5 min may be required to purge the AINC chamber once out of the plume. This purging makes the exact time at which a plume is exited uncertain. To better define actual plume boundaries, reciprocal passes are generally flown, and the entry point of each pass is then used to define the horizontal plume extent. This approach implies an assumption of steady-state transport and diffusion. Heimbach et al. (1977) describe a method of removing the AINC-induced variance and lag.

Ice nucleus plumes were measured in conditions that closely approximated storm conditions but without cloud, that is, in lower humidity regimes. If winds and stability were similar, then flow would be similar to storm conditions. Seeding conditions in the context of the WWMPP tend to be stable, as documented by rawinsondes released at the beginning of each case. To establish the stability during plume-tracing flights, soundings were released from nearby Saratoga, Wyoming, at the beginning and conclusion of each flight. To further guard against the presence of significant convective transport of the plumes, flights were conducted when mid- and upper-level cloud layers were present to reduce the effects of insolation. All of the flights were by necessity conducted when cross-barrier winds were present; hence turbulence was omnipresent and often sufficient to move unsecured items around in the cabin.

The initial flights were conducted with a single generator, MB05, operating in the MB; this approach is designated experiment 1. Additional flights were conducted with five of the eight MB generators operating, to better simulate actual WWMPP seeding events. These are designated experiment 2. For the last two flights in the IOP, only SM generators were active (experiment 3).

Table 2 lists the flight operations conducted on each day during which usable data were collected. Plumes were consistently observed near the crest lines on all flights; plume densities vary considerably, however. In total, 65 sampling passes were flown. Of these, 16 were flown immediately downwind of the active generator(s), 42 were at varying altitudes just upwind of the crest lines, 6 were downwind of the crest lines, and the last was along the north end of the SM and detected few IN. The majority of passes were flown mostly just upwind of the crest line because the primary purpose was to gain an improved understanding of the transport and dispersion as it relates to targeting of the orographic clouds over the mountain ridge. The case day having the most favorable conditions for this type of investigation was 16 February 2011. The observations of that date are examined in more detail below.

Table 2.

Summary of plume-tracing efforts for each of the seven days during which airborne sampling was successfully conducted during the 2011 IOP. Soundings were released from Saratoga at the beginning and end of each mission, from which the Froude and Richardson numbers were calculated. Reported plume concentrations are 10-s running means (~1-km flight path) for the airborne sampling, and 1-min averages for ground-based measurements at the MMC site. Here, ggens indicates ground-based IN generators. For the experiment type, 1 is a single generator active upwind of MB, 2 indicates multiple generators are active upwind of MB, and 3 indicates multiple generators are active upwind of SM. The strongest plume is of all plumes observed on that day and is most always close (immediately downwind) of the generator(s). The strongest crest plume and the highest plume are of those plumes sampled during passes nearest to the crest line, i.e., not close to or immediately downwind of the generators. Date and beginning and ending times of sampling are UTC.

Table 2.

1) Experiment 1

The first three flights (9, 12, and 13 February 2011; experiment 1) were designed to sample plumes produced by a single generator, specifically, the MB05 site (elevation: 2752 m) on the western flank of the MB (see Fig. 1). The flight plan for experiment 1 was designed to fly an initial pass as low as safely possible immediately downwind of the generator (MB05) and perpendicular to the wind direction to determine a close-in plume density and then to follow the initial passes with pairs of passes farther downwind, near the crest line. Passes were flown approximately parallel to the crest line, roughly perpendicular to the mean wind at the flight level. The crest line passes began at the lowest safe altitude (approximately 150 m AGL) and then stepped upward in 150-m increments.

The experiment-1 flights consistently encountered plumes; this approach did not replicate actual seeding procedures, however. Thus, focus was changed to multiple-generator seeding in all subsequent flights.

2) Experiment 2

The next two flights, flown on 16 and 18 February 2011, utilized the five MB generators that would have been operated for the seeding of clouds in similar flow regimes. These generators were MB04, MB05, MB06, MB08, and MB09 (Fig. 1 and Table 1). These two flights detected AgI on every pass. Sampling flight paths were perpendicular to the winds at aircraft altitude. After initial low-altitude passes within 1–3 km of the generators, subsequent passes were flown as for experiment 1, in pairs near the crest line, at increasing altitudes.

3) Experiment 3

The two type-3 experiments utilized different flight plans. After initial sampling over the upwind (west) side of the SM, passes were attempted above the Sierra Madre and then downwind, over the North Platte River valley to the east, as the plumes moved toward the MB. Again, sample passes were flown perpendicular to the winds at aircraft level. These experiments were attempted at the very end of the IOP, when the meteorological conditions, especially the winds, were not as favorable. Though IN plumes were initially detected on both missions, the attempts to follow them past the SM were unsuccessful.

4. A case study—16 February 2011

a. Flight plan and instrumentation status

The flight plan for the 16 February 2011 mission started with an initial pair of passes at low altitude (just above terrain) within a kilometer or so downwind of the generators closest to the crest line, followed by a series of passes near the crest line on the upwind side, beginning near to the surface and then stepping up in ~150-m (500 ft) increments, making pairs of passes at each altitude as described previously. Altitude steps would continue until little plume was encountered, thus providing an approximate vertical cross section of the plume. The major difference between this flight and those preceding is that the five ground-based IN generators thought to be upwind of the target precipitation gauges were activated (the MB04, MB05, MB06, MB08, and MB09 generators shown in Fig. 1), as would be the case during operational experimental seeding, rather than the single generator.

The aircraft instrumentation was all fully functional. The unit-2 AINC parameters were given the exclusive attention of its operator and were monitored throughout the flight. Prior to the beginning of sampling passes and again after the conclusion, samples of surrogate IN, phloroglucinol, were injected in the AINC intake to confirm instrument functionality. The unit-1 AINC was operated at the MMC site throughout the flight, obtaining IN plume measurements at the surface.

b. Stability and sky condition

Soundings released from Saratoga (KSAA) at 2200 UTC 16 February 2011 (red) and at 0100 UTC 17 February 2011 (blue) are shown in Fig. 4. The day was characterized by a well-mixed (near adiabatic) layer from the surface to above crest height, where a weak turbulence inversion strengthened during the flight. A scattered-to-broken stratocumulus deck was present throughout. Winds measured by the aircraftborne Aventech Research, Inc., AIMMS-20 probe during the six passes near the crest line ranged from 17.8 to 21.7 m s−1 and from 223° to 229°, varying little throughout. The sky condition late in the flight is shown in Fig. 5.

Fig. 4.
Fig. 4.

Soundings that were released from KSAA at 2200 UTC 16 Feb 2011 (red) and at 0100 UTC 17 Feb 2011 (blue). The day was characterized by a well-mixed layer from the surface to well above crest height, where a weak inversion strengthened during the flight.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

Fig. 5.
Fig. 5.

The stratocumulus deck present at 0017 UTC 17 Feb 2011, as photographed from the copilot seat by the flight scientist (T. Krauss) during pass 7, heading northwest. The snowcapped crest of the Snowy Range (the highest portion of the MB) appears on the far-right horizon, just below the cloud base.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

c. Observations

Eight passes were made, six of them near the crest line. The passes are summarized in Table 3. The mission duration was limited primarily by the turbulence, which adversely affected the stamina of the crew members, and not by aircraft on-station time or instrumentation limitation. A plan view of the flight path of the 16 February 2011 mission is shown in Fig. 6a. The first pass was flown from southeast to northwest, following the terrain and passing just east of two of the active generators (MB05 and MB06), encountering dense plumes near both and moderate turbulence throughout. The crew deemed this pass too turbulent to safely repeat, and therefore the reciprocal pass (pass 2) was flown farther east. This strong turbulent kinetic energy close to the ground, diminishing with increasing altitude, matches well with the LES shown in Part II. Pass 2 was thus flown approximately midway (in plan view) between the generators closest to the crest line and the crest line itself, but about 400 m higher, again following terrain at approximately 150 m AGL. These passes, and those that followed, are shown in vertical cross section in Fig. 6b. High concentrations of IN were again encountered.

Table 3.

Summary of the eight passes flown on 16 Feb 2011. Passes 3–8 were those used to calculate an approximation of the IN flux over the barrier during the day; steady-state conditions were assumed. Start and end times of passes are UTC. The altitude, heading, and wind speed and direction columns reflect pass-average characteristics measured by the aircraft instrumentation. “First plume encounter” gives the location of first encounter with the plume, i.e, the location of “entry” into the plume. “Total counts” gives the total number of raw “counts” recorded by the AINC during the entire pass. “Peak concentration” gives the maximum 10-s (~1 km) average IN concentration encountered during the pass. The entries in the rightmost column indicate whether the pass was flown immediately downwind of the generators (“ggen”) or near the crest line (“crest”; as revealed by Fig. 6a, the last six passes were flown partially downwind and partially upwind of the crest line).

Table 3.
Fig. 6.
Fig. 6.

The flight track for the 16 Feb 2011 mission (a) in plan view, with respect to the active ground-based IN generator sites, and (b) in a vertical profile facing east on a plane aligned north–south. In (a), passes 1 and 2 are those located farthest west and the other six were flown near the crest line, the location of the MMC IN sampling site is shown by the black circle, and the two sets of target precipitation gauges are shown by the open circles with pluses. The plotted concentrations are lagged by 80 s, the mean time required for IN to enter the instrument, nucleate, grow, and be counted. As suggested by the rapidly varying altitudes seen in (b), the turbulence was considerable.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

The third pass was again flown following the terrain, but closer still to the crest line. The remaining five passes were all flown close to the crest line. Passes 4 and 5 were reciprocal passes at ~3500 m MSL, as were passes 6 and 7, at ~3700 m. Pass 8 was flown higher, at just above 3800 m.

Plumes were encountered on every pass, and in approximately the same positions, reflecting the minimal variation in wind direction. As expected, IN concentrations decreased with increasing altitude, but an IN plume was measured even on pass 8, flown at 3820 m, more than 600 m (~2000 ft) above the average crest-line height. Medicine Bow Peak is the highest at 3660 m (12 013 ft). Sampling safely even at this altitude would not be possible in IFR flight conditions.

The IN plume measured at the MMC ground sampling site on 16 February 2011 is shown in Fig. 7. Although AINC unit 1 at the MMC had not cooled enough to begin sampling when seeding began (t = 0 in Fig. 7), indications of a weak plume having concentrations of 10–20 L−1 were briefly present. This early pulse of IN then faded to background levels before the plume arrived in earnest, shortly after being observed aloft just upwind.

Fig. 7.
Fig. 7.

The IN plume observed on the surface at the MMC site is shown for 16 Feb 2011. The AINC unit 1 began collecting IN data about 55 min after the five ground-based IN generators were activated (2148 UTC). The maximum 10-s mean IN concentrations measured by AINC unit 2 during each of the eight passes are also plotted (black diamonds). Ice nucleus concentrations on the logarithmic ordinate shown as 0.02 indicate that the AINC was not yet collecting data. Values of 0.1 indicate no IN were detected during that minute. The airborne concentrations decrease with time as aircraft altitude was increasing and the plumes thinned. Seeding ended at 0110 UTC, 198 min after seeding start. The plume decay at the MMC began shortly thereafter.

Citation: Journal of Applied Meteorology and Climatology 53, 6; 10.1175/JAMC-D-13-0240.1

d. Approximation of IN flux

Fluxes of IN were calculated using 10-s running means of concentrations, lagged by the 80-s mean time for sampling an IN to its detection (see section 3b). The observed IN concentrations along each pass are shown in Figs. 6a and 6b. The maximum running-mean IN concentration for all eight passes was 697 L−1. Applying the AINC unit-2 sample rate of 8.7 L min−1 and assuming a 1:10 IN detection ratio as determined by Langer (1973) makes this value correspond to 10.1 IN s−1, indicated by the AINC during its operation in the field.

The pass numbers are entered near the respective tracks in Fig. 6a. Passes 1 and 2 are not used in the flux calculations because they were near the generator sites. Only passes 3–8 are used, because all were farther downwind and nearer the crest line. Passes 4–5 and 6–7 were pairs flown at the same altitude, and each pair was averaged for the flux analysis below.

From Super et al. (1975), the flux of IN (s−1) for one pass through the layer Δz is
e1
where N is the number of samples of IN concentration Ci taken during one pass over time period T. The term ∑Ci/N represents the mean IN concentration for the pass. A nonorthogonal plane of sampling is compensated for by the cosθ term, where θ is the angle of the pass to the crosswind plane. The mean wind speed for the layer is u. The aircraft speed is υ, and the width of the vertical plane is υT. The total number of IN for each pass, signified by S (unitless), was also recorded, allowing simplification of computations. In the pass, the total of sample air processed by the AINC is Tq, where q is the sample flow rate = 8.7 L min−1 for AINC unit 2, and
e2
Then,
e3
The approximate flux for each pass, in IN per second (effective at −20°C), is tabulated in Table 4.
Table 4.

Summary of passes used for the estimation of ice nucleus flux over the crest line. When multiple passes were available at the same altitude (i.e., passes 4 and 5 and, later, passes 6 and 7) two-pass averages were used, as indicated by the identical values in the fourth column. The flux is calculated from Eq. (3), using the total number of IN sampled on each pass S, the aircraft speed υ, a Δz determined by the increases in aircraft altitudes between passes, the sampling rate q, and the environmental wind orthogonal to the passes u. The total flux (3.5 × 1013 IN s−1) is the sum of the flux calculated for pass 3, passes 4 and 5 (averaged because they were flown at the same altitude), passes 6 and 7 (again averaged), and pass 8. The total IN estimated to have been produced by the combined five active IN generators is 4.2 × 1013 IN s−1, ~20% greater than what was observed. Passes 1 and 2 were not included because of their proximity to the generator sites (Fig. 6a) and not the crest line, whereas all subsequent passes approximated the same vertical plane.

Table 4.

The number of IN being produced by the five IN generators (MB04, MB05, MB06, MB08, and MB09) can be estimated knowing the flow rates and the aerosol production curve reported by DeMott (1997), who states an average activity for this solution of 1.2 × 1015 IN per gram of AgI at the chamber temperature typical of the AINCs. Each generator burns approximately 25 g AgI h−1, as derived from the measured seeding-solution flow rate of 0.40 gal h−1 (1 gal ≈ 3.785 L). This translates to (0.0069 g AgI s−1) × (1.2 × 1015 IN g−1) = 8.3 × 1012 IN s−1. Since five generators were in operation, the total output was 5 times that, or about 4.14 × 1013 IN s−1, as compared with the observed IN flux of 3.5 × 1013 nuclei s−1, or about 85% of the expected value.

e. Modeling

To further understand the transport and dispersion of the IN plumes on this date, a high-resolution (10- m horizontal grid spacing) LES was run (Part II). The model revealed nuances about the development of the plume, and the airborne sampling herein aids validation of the model. Modeling of the 16 February 2011 case indicated that the planetary boundary layer over the range should have been very turbulent, which indeed was the case. The reader is referred to Part II for modeling details.

5. Summary

Airborne plume tracing, primarily in stable conditions analogous to those experienced during seeding experiments in the WWMPP, consistently revealed IN plumes at altitudes at which supercooled clouds would exist during actual seeding events (see again Table 2). At least 10 IN L−1 are needed to produce enough additional ice to increase precipitation (Reynolds 1988). In these flights, however, the AINCs were operated with chamber temperatures at or near −20°C, whereas the clouds being seeded are typically closer to −10°C, at least at the altitudes affected by the plumes. This translates to approximately an order-of-magnitude difference, and so to ensure effectiveness the authors like to see 100 IN L−1 in the AINC measurements. This noted, the maximum observed plume concentrations exceeded this threshold on all days sampled. Plumes were most dense at lower altitudes and thinned with height.

These flights, conducted over complex terrain at low altitudes in cross-barrier flow, were very turbulent. Although it was not a safety issue, the entire flight crew, and especially the data-system and AINC operators, who were seated aft in the aircraft, were consistently pushed to their limits. This made it difficult to continue with sampling beyond about 90 min, typically sufficient for perhaps 8–12 passes, depending upon the horizontal extent and densities of the plumes each day. As a result, passes were never made completely above the plumes.

Forty-two of the 65 plume-sampling passes were flown near the crest lines of the two ranges. During the 20 such passes flown for experiment type 1 (just one generator active), the maximum plume height did not exceed 520 m above the mean crest-line elevation. Fourteen passes were flown near the crest line for experiment type 2, when five generators were active, during which plumes were detected as high as 640 m above the mean crest line elevation.

An additional 16 near-crest-line passes were flown for experiment type 3 over the SM, but observed plumes were significantly less dense and shallow and were never observed more than 100 m above the mean crest-line elevation. Winds were not as steady on these type-3 days, which may have contributed to weaker plumes. Much of each type-3 flight was spent seeking plumes beyond the SM in the North Platte Valley and along the north end of the range, where it appeared (to the flight crew, in real time) that the plumes on both flights were headed.

On 16 February 2011, five ground-based IN generators were activated between 2148 and 2152 UTC. Ground-based sampling at the MMC began at 2243 UTC, 55 min after seeding began and 36 min before the first aircraft pass. Although small concentrations of IN were detected at the surface when sampling began, the plume was not sampled at maximum density at the MMC until more than 2 h after seeding commenced. The development of this plume is explored through high-resolution LES in Part II. Aircraft sampling began at 2319 UTC, 88 min after the first generator was ignited. Individual plume boundaries were too complex to resolve when multiple generators were in use, except in passes made within a few kilometers of the generators. A series of transects at increasingly higher altitudes made an estimation of total IN flux possible. The output of five ground-based generators is estimated to be about 4.2 × 1013 nuclei per second, about 20% greater than the flux of 3.5 × 1013 that was observed for the passes flown near the crest line on 16 February 2011. The strongest plume of those passes flown at the highest altitudes on each flight (165 IN L−1 at 3750 m MSL) was also recorded on 16 February 2011.

6. Conclusions

This effort differs from previous similar research by virtue of the simultaneous operations of both airborne and ground-based IN counters to determine the time required for residual IN to be flushed from both RSE target ranges, to verify targeting, and (with Part II) to provide measurements useful in model validation. In experiment types 2 and 3, multiple IN generators were operated at once, as is commonly done in most operational winter orographic cloud-seeding programs. Except for the initial type-1 experiments, the behavior and interactions of IN plumes from multiple sources are investigated.

Ground-based IN plumes were consistently transported over the MB and were detected in concentrations sufficient to influence the precipitation development (10 L−1; Reynolds 1988). The IN concentrations measured near the crest line on all days exceeded the 100 L−1 desired for AINC measurements, made with chamber temperatures of ~−20°C.

The IN concentrations observed at the MMC site were often less than those observed aloft above the windward slope to the west, and they frequently lagged behind the plume aloft. These findings together suggest that the lesser surface plume densities typified in Figs. 2b and 2c should not cause concern with regard to targeting.

The limited flight time (because of turbulence during sampling) precluded the determination of the maximum vertical extents of any of the plumes; that is, no passes were flown high enough to completely overfly the plumes.

These observations show that transport by prevailing winds, aided by dispersion and terrain-induced mixing, is sufficient to produce broad but shallow (<0.5 km AGL) IN plumes that pass over and beyond the crest line. The measurements were made in VFR conditions, in the absence of cloud and precipitation that could nucleate and/or scavenge the nuclei, but in otherwise similar conditions. Supercooled liquid cloud would normally occupy the sampled volumes during storm conditions.

The IN that were sampled when multiple generators were operated in concert, especially in type-2 experiments, revealed complex plumes that spread and intermingled with each other. The existing spacing of the generators (Fig. 1) appeared to be adequate to target much of the desired “cloud” volume. Only 20 flight hours were flown, however—many in similar flow regimes, and therefore this may not always be true, especially in stronger (>25 m s−1) flow.

The authors recommend that additional airborne plume tracing be done to better resolve transport from ground generators sited at lower elevations, farther upwind of the crest lines. Many, if not most, operational (nonresearch) orographic seeding programs rely on IN generators sited at such elevations, more than a kilometer below the crest elevations that their plumes must traverse.

Also, further examination of the IN transport from the SM to the MB when the SM is the seeded range and the MB is the control (experiment type 3) would be desirable. The lack of consistent plume detection in the lee of the SM during the 28 February 2011 and 1 March 2011 flights suggest that optimum flight paths may not have been flown on those days. Flight at lower altitudes is certainly possible.

The IN flux estimated from the six passes flown nearest the MB crest line on 16 February 2011 is close to the estimated IN output from the five generators active at the time, suggesting that the AINC measurements are reasonable and have quantitative value. The AINC can be flown on a light aircraft and thus is more likely to be affordable to those wishing to better understand AgI IN plume behavior. The instrument is also a viable means to gather data needed for the validation of numerical models being developed to predict the evolution of IN plumes over complex terrain.

Acknowledgments

This research was supported by the Wyoming Water Development Commission and the Wyoming Water Development Office (WWDO) under Contract 05SC0292770. The authors greatly appreciate the confidence expressed by the commission and the guidance and support provided by the WWDO program manager, Barry Lawrence. Special thanks are extended to Drs. Gary Langer and Arlin Super for their counsel and suggestions during the field effort and to Dr. Super for the use of AINC unit 1. The authors greatly appreciate the dedication and skills of the pilot Jody Fischer, the aircraft data system operator Erin Fischer, and the IOP forecasters Daniel Gilbert and Bradley Waller, who did double duty as rawinsonde technicians. The authors also greatly appreciate the detailed comments of the reviewers, which aided significantly in the preparation of the final typescript. All rights to the underlying data collected and/or generated with funding from the WWDO from which this paper was created remain with the WWDO. This paper does not constitute the opinions of the State of Wyoming, the Wyoming Water Development Commission, or the WWDO.

REFERENCES

  • Boe, B. A., , and P. J. DeMott, 1999: Comparisons of Lohse wing-tip nuclei generators and burn-in-place pyrotechnics in the North Dakota Cloud Modification Project. J. Wea. Modif., 31, 109118.

    • Search Google Scholar
    • Export Citation
  • Breed, D., , R. Rasmussen, , C. Weeks, , B. Boe, , and T. Deshler, 2014: Evaluating winter orographic cloud seeding: Design of the Wyoming Weather Modification Pilot Project (WWMPP). J. Appl. Meteor. Climatol, 53, 282299, doi:10.1175/JAMC-D-13-0128.1.

    • Search Google Scholar
    • Export Citation
  • DeMott, P. J., 1997: Report to the North Dakota Atmospheric Resource Board and Weather Modification Incorporated on tests of the ice nucleating ability of aerosols produced by the Lohse airborne generator—1997. Department of Atmospheric Sciences, Colorado State University, Fort Collins, CO, 14 p.

  • DeMott, P. J., , A. B. Super, , G. Langer, , D. C. Rogers, , and J. T. McPartland, 1995: Comparative characterizations of the ice nucleus ability of AgI aerosols by three methods. J. Wea. Modif., 27, 116.

    • Search Google Scholar
    • Export Citation
  • DeMott, P. J., and Coauthors, 2011: Resurgence in ice nuclei measurement research. Bull. Amer. Meteor. Soc., 92, 16231635, doi:10.1175/2011BAMS3119.1.

    • Search Google Scholar
    • Export Citation
  • Garvey, D. M., 1975: Testing of cloud seeding materials at the cloud simulation and aerosol laboratory, 1971–1973. J. Appl. Meteor., 14, 883889, doi:10.1175/1520-0450(1975)014<0883:TOCSMA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Heimbach, J. A., , A. B. Super, , and J. T. McPartland, 1977: A suggested technique for the analysis of airborne continuous ice nucleus data. J. Appl. Meteor., 16, 255261, doi:10.1175/1520-0450(1977)016<0255:ASTFTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Heimbach, J. A., , A. B. Super, , B. A. Boe, , G. Langer, , and J. T. McPartland, 2008: Comparison of two acoustic ice nucleus counters. J. Wea. Modif., 40, 5463.

    • Search Google Scholar
    • Export Citation
  • Holroyd, E. W., III, , J. T. McPartland, , and A. B. Super, 1988: Observations of silver iodide plumes over the Grand Mesa of Colorado. J. Appl. Meteor., 27, 11251144, doi:10.1175/1520-0450(1988)027<1125:OOSIPO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Holroyd, E. W., III, , J. A. Heimbach, , and A. B. Super, 1995: Observations and model simulation of AgI seeding within a winter storm over Utah’s Wasatch Plateau. J. Wea. Modif., 27, 3556.

    • Search Google Scholar
    • Export Citation
  • Huggins, A. W., 2007: Another wintertime cloud seeding case study with strong evidence of seeding effects. J. Wea. Modif., 39, 936.

  • Langer, G., 1973: Evaluation of NCAR ice nucleus counter. Part I: Basic operation. J. Appl. Meteor., 12, 10001011, doi:10.1175/1520-0450(1973)012<1000:EONINC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Langer, G., , J. Rosinski, , and C. P. Edwards, 1967: A continuous ice nucleus counter and its application to tracking in the troposphere. J. Appl. Meteor., 6, 114125, doi:10.1175/1520-0450(1967)006<0114:ACINCA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Langer, G., , G. Cooper, , C. T. Nagamoto, , and J. Rosinski, 1978: Ice nucleation mechanisms of submicron monodispersed silver iodide, 1,5-dihydroxynaphathalene and phloroglucinol aerosol particles. J. Appl. Meteor., 17, 10391048, doi:10.1175/1520-0450(1978)017<1039:INMOSM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rauber, R. M., , and L. O. Grant, 1986: 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, doi:10.1175/1520-0450(1986)025<0489:TCADOC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Reynolds, D. W., 1988: A report on winter snowpack augmentation. Bull. Amer. Meteor. Soc., 69, 12901300, doi:10.1175/1520-0477(1988)069<1290:AROWSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rogers, D. C., , P. J. DeMott, , S. M. Kreidenweis, , and Y. Chen, 2001: A continuous-flow diffusion chamber for airborne measurements of ice nuclei. J. Atmos. Oceanic Technol., 18, 725741, doi:10.1175/1520-0426(2001)018<0725:ACFDCF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., 1990: Winter orographic cloud seeding in the intermountain west. J. Wea. Modif., 22, 106116.

  • Super, A. B., 1999: Summary of the NOAA/Utah Atmospheric Modification Program: 1990-1998. J. Wea. Modif., 31, 5175.

  • Super, A. B., , and B. A. Boe, 1988: Microphysical effects of wintertime cloud seeding with silver iodide over the Rocky Mountains, Part III: Observations over the Grand Mesa, Colorado. J. Appl. Meteor., 27, 11661182, doi:10.1175/1520-0450(1988)027<1166:MEOWCS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , and J. A. Heimbach Jr., 1983: Evaluation of the Bridger Range winter cloud seeding experiment using control gages. J. Climate Appl. Meteor., 22, 19892011, doi:10.1175/1520-0450(1983)022<1989:EOTBRW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , and J. A. Heimbach Jr., 1988: Microphysical effects of wintertime cloud seeding with silver iodide over the Rocky Mountains. Part II: Observations over the Bridger Range, Montana. J. Appl. Meteor., 27, 11521165, doi:10.1175/1520-0450(1988)027<1152:MEOWCS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , J. T. McPartland, , and J. A. Heimbach, 1975: Field observations of the persistence of AgI-NH4I-acetone ice nuclei in daylight. J. Appl. Meteor., 14, 15721577, doi:10.1175/1520-0450(1975)014<1572:FOOTPO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Super, A. B., , B. A. Boe, , J. A. Heimbach, , J. T. McPartland, , and G. Langer, 2010: Comparison of silver iodide outputs from two different generators and solutions measured by acoustic ice nucleus counters. J. Wea. Modif., 42, 4960.

    • Search Google Scholar
    • Export Citation
  • Warburton, J. A., , R. H. Stone III, , and B. L. Marler, 1995: How the transport and dispersion of AgI aerosols may affect detectability of seeding effects by statistical methods. J. Appl. Meteor., 34, 19291941, doi:10.1175/1520-0450(1995)034<1929:HTTADO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • WMI, 2005: Wyoming Level II Weather Modification Feasibility Study. Wyoming Water Development Commission Rep., 151 pp. [Available online at http://charybdis.wrds.uwyo.edu/weathermod/Report.pdf.]

  • Xue, L., , X. Chu, , R. Rasmussen, , D. Breed, , B. Boe, , and B. Geerts, 2014: The dispersion of silver iodide particles from ground-based generators over complex terrain. Part II: WRF large-eddy simulations versus observations. J. Appl. Meteor. Climatol., 53, 13421361, doi:10.1175/JAMC-D-13-0241.1.

    • Search Google Scholar
    • Export Citation
Save