• Atkins, N. T., , and Wakimoto R. M. , 1991: Wet microburst activity over the southeastern United States: Implications for forecasting. Wea. Forecasting, 6 , 470482.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Balling, R. C., , and Brazel S. W. , 1987: Diurnal variations in Arizona monsoon precipitation frequencies. Mon. Wea. Rev., 115 , 342346.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., and Coauthors, 2005: The Shared Mobile Atmospheric Research and Teaching Radar: A collaboration to enhance research and teaching. Bull. Amer. Meteor. Soc., 86 , 12631274.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biron, P. J., , and Isaminger M. A. , 1989: An analysis of microburst characteristics related to automatic detection from Huntsville, Alabama, and Denver, Colorado. Preprints, 24th Conf. on Radar Meteorology, Tallahassee, FL, Amer. Meteor. Soc., 269–273.

    • Search Google Scholar
    • Export Citation
  • Caracena, F., , and Maier M. W. , 1987: Analysis of a microburst in the FACE meteorological mesonetwork in southern Florida. Mon. Wea. Rev., 115 , 969985.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dodge, J., , Arnold J. , , Wilson G. , , Evans J. , , and Fujita T. T. , 1986: The Cooperative Huntsville Meteorological Experiment (COHMEX). Bull. Amer. Meteor. Soc., 67 , 417419.

    • Search Google Scholar
    • Export Citation
  • Eilts, M. D., , and Doviak R. J. , 1987: Oklahoma downbursts and their asymmetry. J. Climate Appl. Meteor., 26 , 6978.

  • Elmore, K. L., 1986: Evolution of a microburst and bow-shaped echo during JAWS. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 101–104.

    • Search Google Scholar
    • Export Citation
  • Evans, J. E., , and Johnson D. , 1984: The FAA transportable Doppler weather radar. Preprints, 22nd Int. Conf. on Radar Meteorology, Zurich, Switzerland, Amer. Meteor. Soc., 246–250.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1978: Manual of downburst identification for project NIMROD. SMRP Research Paper 156, University of Chicago, 104 pp. [NTIS PB-2860481].

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1979: Objectives, operation, and results of Project NIMROD. Preprints, 11th Conf. on Severe Local Storms, Kansas City, MO, Amer. Meteor. Soc., 259–266.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38 , 15111534.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1985: The downburst. SMRP Research Paper 210, University of Chicago, 122 pp. [NTIS PB-148880].

  • Hales, J. E., 1977: On the relationship of convective cooling to nocturnal thunderstorms at Phoenix. Mon. Wea. Rev., 105 , 16091613.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Heinselman, P. L., , and Schultz D. M. , 2006: Intraseasonal variability of summer storms over central Arizona during 1997 and 1999. Wea. Forecasting, 21 , 559578.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hjelmfelt, M. R., 1984: Radar and surface data analysis of a microburst in JAWS. Preprints, 22nd Conf. on Radar Meteorology, Zurich, Switzerland, Amer. Meteor. Soc., 64–69.

    • Search Google Scholar
    • Export Citation
  • Hjelmfelt, M. R., 1988: Structure and life cycle of microburst outflows observed in Colorado. J. Appl. Meteor., 27 , 900927.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kessinger, C. J., , Roberts R. D. , , and Elmore K. L. , 1986: A summary of microburst characteristics from low-reflectivity storms. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 105–108.

    • Search Google Scholar
    • Export Citation
  • Kingsmill, D. E., , and Wakimoto R. M. , 1991: Kinematic, dynamic, and thermodynamic analysis of a weakly sheared severe thunderstorm over northern Alabama. Mon. Wea. Rev., 119 , 262297.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • MacKeen, P. L., , and Zhang J. , 2000: Convective climatology for central Arizona during the 1999 monsoon. Proc. Southwest Weather Symp., Tucson, AZ, National Weather Service, 64–67.

    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., , McCollum D. M. , , and Howard K. W. , 1995: Large-scale patterns associated with severe summertime thunderstorms over central Arizona. Wea. Forecasting, 10 , 763778.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCarthy, J., , Wilson J. W. , , and Fujita T. T. , 1982: The Joint Airport Weather Studies Project. Bull. Amer. Meteor. Soc., 63 , 15.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCollum, D. M., , Maddox R. A. , , and Howard K. W. , 1995: Case study of a severe mesoscale convective system in central Arizona. Wea. Forecasting, 10 , 643665.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mielke, K. B., , and Carle E. R. , 1987: An early morning dry microburst in the Great Basin. Wea. Forecasting, 2 , 169174.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , and Isaminger M. A. , 1986: Radar characteristics of microbursts in the mid-south. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 116–119.

    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , and Borho A. , 1993: A comparison of the detectability of microbursts in Orlando, Florida, by two C-band Doppler radars. J. Appl. Meteor., 32 , 476489.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , Borho A. , , and Curtiss C. , 1995: Microburst rotation: Simulations and observations. J. Appl. Meteor., 34 , 12671285.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roberts, R. D., , and Wilson J. W. , 1989: A proposed microburst nowcasting procedure using single-Doppler radar. J. Appl. Meteor., 28 , 285303.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Turnbull, D., , McCarthy J. , , Evans J. , , and Zrnic D. , 1989: The FAA Terminal Doppler Weather Radar (TDWR) program. Preprints, Third Int. Conf. on Aviation Weather Systems, Anaheim, CA, Amer. Meteor. Soc., 414–418.

    • Search Google Scholar
    • Export Citation
  • Vasiloff, S. V., 2001: Improving tornado warnings with the Federal Aviation Administration’s Terminal Doppler Weather Radar. Bull. Amer. Meteor. Soc., 82 , 861874.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vasiloff, S. V., , and Howard K. W. , 2009: Investigation of a severe downburst storm near Phoenix, Arizona, as seen by a mobile Doppler radar and the KIWA WSR-88D. Wea. Forecasting, 24 , 856867.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., , and Bringi V. , 1988: Dual-polarization observations of microbursts associated with intense convection: The 20 July storm during the MIST Project. Mon. Wea. Rev., 116 , 15211539.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wallace, C. E., , Maddox R. A. , , and Howard K. W. , 1999: Summertime convective storm environments in central Arizona: Local observations. Wea. Forecasting, 14 , 9941006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Watson, A. I., , López R. E. , , and Holle R. L. , 1994: Diurnal cloud-to-ground lightning patterns in Arizona during the southwest monsoon. Mon. Wea. Rev., 122 , 17161725.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., , Roberts R. D. , , Kessinger C. , , and McCarthy J. , 1984: Microburst wind structure and evaluation of Doppler radar for airport wind shear detection. J. Appl. Meteor., 23 , 898915.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wolfson, M. M., , Distefano J. T. , , and Fujita T. T. , 1985: Low-altitude wind shear characteristics in the Memphis, TN, area based on mesonet and LLWAS data. Preprints, 14th Conf. on Severe Local Storms, Indianapolis, IN, Amer. Meteor. Soc., 322–327.

    • Search Google Scholar
    • Export Citation
  • Wolfson, M. M., and Coauthors, 1990: Characteristics of thunderstorm-generated low altitude wind shear: A survey based on nationwide Terminal Doppler Weather Radar testbed measurements. Proc. 29th IEEE Conf. on Decision and Control, Honolulu, HI, Institute of Electrical and Electronics Engineers, 682–688.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Photograph of a Sonoran Desert microburst on 13 Jul 2008 that occurred near overhead electrical power lines. (Courtesy of K. Howard)

  • View in gallery

    (a) Map of the topography of Arizona. The elevated terrain of the Colorado Plateau, the White Mountains, and the Mogollon Rim creates a sharp topographic gradient from northeast to southwest across central Arizona. The box at center encompasses the region over which analysis occurred. (b) The analysis region with the Phoenix metropolitan area outlined and radar locations indicated, encircled by 50-km range rings. Most of the viewable area from SR1 is overlapped by PHX and KIWA.

  • View in gallery

    The 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0122 UTC; (b) velocity at 0123 UTC; (c) reflectivity at 0127 UTC; (d) velocity at 0127 UTC, where maximum outbound velocity was 10.5 m s−1 and maximum inbound velocity was −18.5 m s−1; (e) reflectivity at 0132 UTC; and (f) velocity at 0132 UTC. The 0.9° tilt was used here because the 0.5° tilt was mostly blocked by terrain. The 0.9° tilt was still partially blocked, as evidenced by the missing data (black patches) in the images. In the radial velocity images, red colors indicate inbound velocities and green colors indicate outbound velocities.

  • View in gallery

    Microburst detection locations for summer 2008 (region encompassed by box in Fig. 2a). Red markers indicate a microburst observed from KIWA; blue markers indicate PHX. The four microbursts that were viewed from both KIWA and PHX are in purple. The radar locations are labeled. SR1 data were not used in the microburst count and characterization.

  • View in gallery

    Diurnal variation of the Sonoran microbursts observed during summer 2008 (MST is 7 h behind UTC).

  • View in gallery

    Distribution of Sonoran microburst base radar reflectivity observed during summer 2008. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum still within a 1.5 interquartile range. The six dots to the left of the diagram represent outliers (data outside the 1.5 interquartile range). Sonoran microbursts are moderate to high reflectivity, or wet, microbursts. The median base reflectivity was 57 dBZ.

  • View in gallery

    Distribution of Sonoran microburst size observed during summer 2008. Microburst size was measured from the maximum approaching radial velocity to the maximum receding radial velocity. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum. The median size was 4 km.

  • View in gallery

    Distribution of Sonoran microburst maximum differential velocity observed during summer 2008. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentile, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum still within a 1.5 interquartile range. The six dots to the right of the diagram represent outliers (data outside the 1.5 interquartile range). This histogram has a peak frequency of 20 to 25 m s−1, similar to microbursts from Memphis, Tennessee (Wolfson et al. 1990). The median differential velocity was 24 m s−1.

  • View in gallery

    The 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of a gust front moving north from the location of a previous microburst and initiating new convection, showing (a) reflectivity at 0141 UTC; (b) velocity at 0141 UTC; (c) reflectivity at 0159 UTC; (d) velocity at 0200 UTC; (e) reflectivity at 0218 UTC, where new convective development is indicated by the white arrows; and (f) velocity at 0218 UTC. Gust-front outflow boundaries are indicated by the blue front symbol, where the symbol orientation refers to the direction the gust front is moving. The gust-front boundary is easier to see in the 0.9° elevation angle than the 0.5°.

  • View in gallery

    The new convection from Fig. 9e produced two new microbursts. Shown are three consecutive 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0222 UTC; (b) velocity at 0223 UTC, with microburst A circled; (c) reflectivity at 0227 UTC; (d) velocity at 0227 UTC; (e) reflectivity at 0232 UTC; and (f) velocity at 0232 UTC, with microburst B circled. Microburst B was severe, with a maximum radial velocity of 26 m s−1.

  • View in gallery

    After microburst B in Fig. 10, the microburst outflow transitioned to a gust front and propagated outward from the high-reflectivity core of the storm. Shown are three consecutive 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0236 UTC, (b) velocity at 0237 UTC, (c) reflectivity at 0241 UTC, (d) velocity at 0241 UTC, (e) reflectivity at 0245 UTC, and (f) velocity at 0246 UTC. Gust-front outflow boundaries are indicated by the blue front symbol.

  • View in gallery

    Comparison of maximum radial Doppler velocities observed at sites damaged by gust-front winds vs sites damaged by microburst winds by box-and-whisker diagrams for the summer 2008 damage reports. The bottom and top of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum.

  • View in gallery

    Radar observations around the time of the damage report at 0233 UTC 22 Jul 2008. All subfigures cover the same area. (a) SR1 0.5° velocity at 0230:08 UTC with microburst signatures A1, A2, and B indicated; (b) KIWA 0.5° velocity at 0231:44 UTC; (c) SR1 0.5° velocity at 0235:06 UTC; (d) KIWA 0.5° velocity at 0236:21 UTC; and (e) KIWA 0.5° reflectivity at 0231:26 UTC. Damage site is indicated by the red dot.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 64 64 6
PDF Downloads 53 53 5

Characteristics of Sonoran Desert Microbursts

View More View Less
  • 1 Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, and NOAA/OAR/National Severe Storms Laboratory, Norman, Oklahoma
  • | 2 Department of Geography and Meteorology, Valparaiso University, Valparaiso, Indiana
  • | 3 NOAA/OAR/National Severe Storms Laboratory, Norman, Oklahoma
  • | 4 Salt River Project, Scottsdale, Arizona
© Get Permissions
Full access

Abstract

During the 2008 North American monsoon season, 140 microburst events were identified in Phoenix, Arizona, and the surrounding Sonoran Desert. The Sonoran microbursts were studied and examined for their frequency and characteristics, as observed from data collected from three Doppler radars and electrical power infrastructure damage reports. Sonoran microburst events were wet microbursts and occurred most frequently in the evening hours (1900–2100 local time). Stronger maximum differential velocities (20–25 m s−1) were observed more frequently in Sonoran microbursts than in many previously documented microbursts. Alignment of Doppler radar data to reports of wind-related damage to electrical power infrastructure in Phoenix allowed a comparison of microburst wind damage versus gust-front wind damage. For these damage reports, microburst winds caused more significant damage than gust-front winds.

Corresponding author address: Katherine M. Willingham, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. Email: katherine.willingham@noaa.gov

Abstract

During the 2008 North American monsoon season, 140 microburst events were identified in Phoenix, Arizona, and the surrounding Sonoran Desert. The Sonoran microbursts were studied and examined for their frequency and characteristics, as observed from data collected from three Doppler radars and electrical power infrastructure damage reports. Sonoran microburst events were wet microbursts and occurred most frequently in the evening hours (1900–2100 local time). Stronger maximum differential velocities (20–25 m s−1) were observed more frequently in Sonoran microbursts than in many previously documented microbursts. Alignment of Doppler radar data to reports of wind-related damage to electrical power infrastructure in Phoenix allowed a comparison of microburst wind damage versus gust-front wind damage. For these damage reports, microburst winds caused more significant damage than gust-front winds.

Corresponding author address: Katherine M. Willingham, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. Email: katherine.willingham@noaa.gov

1. Introduction

During the North American monsoon season, convective storms in the southwestern U.S. Sonoran Desert produce damaging microbursts (e.g., Vasiloff and Howard 2009; Fig. 1). Severe winds from microbursts can threaten aviation operations, damage property, and interrupt communications, transportation, and electrical power transmission. Because of rapid urban expansion and the rising population density of central Arizona, specifically in the Phoenix metropolitan area, the potential socioeconomic impacts of Sonoran microbursts have increased. Sonoran microbursts pose a critical challenge for forecasters. Skill scores for forecasting the Arizona monsoon thunderstorms that can generate microbursts are low (Maddox et al. 1995), and the short temporal and small spatial scales over which microbursts occur make detection and nowcasting difficult.

A dearth of any previous systematic documentation established a need to document frequency and characteristics of Sonoran microbursts (Vasiloff and Howard 2009). During the summer of 2008, the National Severe Storms Laboratory deployed a Shared Mobile Atmospheric Research and Teaching Radar [SMART-R (SR1); Biggerstaff et al. 2005] near Phoenix to augment measurements from the local National Weather Service Weather Surveillance Radar-1988 Doppler (WSR-88D; KIWA) and the Federal Aviation Administration (FAA) Terminal Doppler Weather Radar (TDWR; PHX) (Table 1). In addition, The Salt River Project, an Arizona utility company, made available high-resolution reports of wind-related damage incidents to their electrical power network. Together, these observations represented a unique dataset to examine Sonoran microburst frequency, distribution, and intensity.

2. Background

a. Arizona

The terrain of Arizona is complex (Fig. 2a), with the majority of the land in northeastern Arizona above 1500 m in elevation. The elevated terrain includes the Kaibab Plateau, the Colorado Plateau, and the White Mountains; the Mogollon Rim runs diagonally northwest to southeast across north-central Arizona and averages 2000 m in elevation. Much of southwestern Arizona, including Phoenix and the Sonoran Desert, lies below 600 m. The resultant topographic gradient slopes down from northeast to southwest and is sharpest in central Arizona.

Because of this complex terrain, central Arizona experiences a pronounced diurnal cycle of precipitation and convective activity during the monsoon season (Hales 1977; Balling and Brazel 1987; Watson et al. 1994; Wallace et al. 1999; MacKeen and Zhang 2000; Heinselman and Schultz 2006). Over the central and eastern mountains, thunderstorm activity peaks in the midafternoon, attributable to peak solar heating. The storms tend to move progressively off the higher terrain into the lower Sonoran Desert during the early evening, where a hot, moist, and unstable atmospheric environment is supportive of continued convective activity (McCollum et al. 1995; Maddox et al. 1995; Wallace et al. 1999; Heinselman and Schultz 2006).

b. Microbursts

Fujita (1981) defined a microburst as a small downburst with a divergent outflow of winds at or near the surface with a diameter range of 0.4 to 4 km. The Northern Illinois Meteorological Research on Downbursts (NIMROD) near Chicago (Fujita 1978, 1979, 1985) and the Joint Airport Weather Studies (JAWS) near Denver (McCarthy et al. 1982; Wilson et al. 1984; Fujita 1985) were designed to study microbursts and the associated aircraft hazard from wind shears and identified microburst characteristics. A modified microburst definition was developed for Doppler radar observations that used a differential velocity across the divergent microburst center greater than 10 m s−1 (Wilson et al. 1984). Differential velocity is defined as the algebraic difference of the maximum receding radial velocity and the maximum approaching radial velocity across the microburst center.

Motivated by the need to mitigate the low-level wind shear threat microbursts posed to aviation operation, a transportable version of the prototype TDWR (Evans and Johnson 1984) [operated by a joint venture between the Massachusetts Institute of Technology (MIT) Lincoln Laboratory and the University of North Dakota] was deployed to several airports for testing and used to measure thunderstorm-generated low-level wind shear (Wolfson et al. 1990). Data were collected in Memphis, Tennessee, as a part of the multiyear FAA–Lincoln Laboratory Operational Weather Studies (FLOWS) project (Wolfson et al. 1985; Rinehart and Isaminger 1986). In Huntsville, Alabama, the FLOWS project joined with the Microburst and Severe Thunderstorm (MIST) project (Wakimoto and Bringi 1988; Atkins and Wakimoto 1991) in the Cooperative Huntsville Meteorological Experiment (COHMEX) (Dodge et al. 1986) to collect data on microbursts there. Data were also collected on microbursts in Denver, Colorado (1987–88); Kansas City, Missouri (1989); and Orlando, Florida (1990–91) (Hjelmfelt 1988; Wolfson et al. 1990; Rinehart and Borho 1993). The datasets from the MIT Lincoln Laboratory studies provided an extensive collection of microburst data in a wide range of environments, from the humid southeastern states to the semiarid high plains. Microbursts have also been documented in other parts of the country including Oklahoma (Eilts and Doviak 1987) and Utah (Mielke and Carle 1987).

Hjelmfelt (1988) categorized microbursts into two types with data from JAWS: individual microbursts and microburst lines. A single microburst with low-level flow diverging outward in all directions is an individual microburst, and a line of two or more microbursts with combined outflows forming a divergent line with length at least twice the width is a line microburst.

Roberts and Wilson (1989) classified microbursts by radar reflectivity values, with low-, moderate-, and high-reflectivity microbursts. The characteristics of low-reflectivity microbursts (or dry microbursts, with reflectivity less than 35 dBZ) were well documented by Wilson et al. (1984), Hjelmfelt (1984, 1988), Fujita (1985), Elmore (1986), and Kessinger et al. (1986). Moderate- and high-reflectivity microbursts, also known as wet microbursts, were commonly observed in moist environments and had reflectivity values greater than 35 dBZ (Wilson et al. 1984; Fujita 1985; Wolfson et al. 1985; Rinehart and Isaminger 1986; Caracena and Maier 1987; Wakimoto and Bringi 1988; Biron and Isaminger 1989; Wolfson et al. 1990; Kingsmill and Wakimoto 1991; Atkins and Wakimoto 1991; Rinehart and Borho 1993; Rinehart et al. 1995). Wet microbursts were observed in Memphis, Huntsville, Kansas City, and Orlando. Wet microburst outflows, which are almost always cold, often evolved rapidly into larger gust-front features that continued to propagate outward after the initial microburst downdraft dissipated (Roberts and Wilson 1989; Wolfson et al. 1990).

3. Radar data

a. Data

Figure 2b shows that KIWA, PHX, and SR1 were located at points that surrounded the Phoenix metropolitan area. The complex terrain of the Phoenix area meant radar beam height varied across the city. The radar beam height increased as range from the radar increased, but the beam height was also dependent on topographic variation (the radars are located at relatively low elevations compared to the elevated terrain around Phoenix; see Table 1). Partial or complete blockage of the radar beam from terrain at the lowest elevation angles limited the area where microbursts were detectable. Lower tilts in partially blocked areas were sometimes still useful for microburst detection, since the divergent velocity signature was still observed despite the partial blockage of the beam.

From the 2008 monsoon season, 14 storm days were selected during which convection occurred in the greater Phoenix area (Table 2). KIWA data were available for all 14 days, SR1 data for 12 days (for limited time periods), and PHX data for 7 days. More microbursts in this dataset were observed from KIWA than from PHX because of the longer period of data availability from KIWA. SR1 data were not used to count or characterize the microbursts because of limited hours of collection on storm days. The SR1 data were incorporated on a case-by-case basis in the analysis of damage reports in section 5 when data were available.

Accurate depiction of microburst outflows greatly depended on sampling height, radar resolution, and frequency of lowest elevation angle scans. Sampling height relative to the ground needed to be as low as possible to observe the low-level microburst outflow effectively. PHX microburst observations were, on average, 100 m lower in altitude than KIWA observations because of the difference in elevation of the radar locations. The differential winds within JAWS microbursts peaked at 75 m above ground level and decreased rapidly above 200 m (Wilson et al. 1984). Many microburst observations in this study, especially those from KIWA, were taken higher than 400 m above ground level. The higher-altitude observations may have limited the accuracy of the microburst wind measurements. Measured gradients of wind shear and peak wind speeds are reduced when they occur over small distances compared to radar beam resolution volume size. In the lowest elevation angle scan, KIWA had a range gate resolution of 250 m for radial velocity, whereas PHX and SR1 had a finer range gate resolution of 150 m. PHX also had a narrower beamwidth of 0.5°, compared with the 1° beamwidth of KIWA and 1.5° of SR1. Scan frequency of the lowest elevation angle was important when observing rapidly evolving phenomena, such as microbursts. The TDWR was specifically designed to detect low-altitude wind shear (Turnbull et al. 1989), but limited data availability meant most of the microbursts in this study were observed from KIWA. PHX completed a lowest-level “microburst” scan every 1 min, compared with KIWA and SR1 lowest-level scan updates every 5 min. The lower temporal resolution of WSR-88D and SMART-R observations could potentially miss the point of peak intensity of the relatively short-lived microburst. Vasiloff (2001) documented differences between WSR-88D and TDWR observations of tornadoes, and Vasiloff and Howard (2009) noted important differences between WSR-88D and SMART-R observations of a severe downburst storm.

b. Analysis methods

Developing convective cells were initially identified in reflectivity imagery and monitored through subsequent scans for the occurrence of a microburst. Microbursts were identified using the low-level divergent velocity signature in the lowest elevation angle radial velocity scans. Microburst peak intensity was defined as the time when the maximum differential velocity was measured within the low-level divergent outflow. At peak intensity, the microburst base reflectivity and size were recorded, where size was defined as the distance between the point of maximum approaching velocity and point of maximum receding velocity in the radar data.

A microburst analysis example is shown in Fig. 3. At 0122 UTC 22 July 2008, a small area of convective development was detected in KIWA reflectivity imagery south of the radar at a range of 10 to 15 km (Fig. 3a). The corresponding radial velocity imagery showed flow toward the radar (southerly) around 5 m s−1 (Fig. 3b). In the next scan, at 0127 UTC, the base reflectivity increased markedly in intensity and size (Fig. 3c). Radial velocity imagery contained the classic signature of diverging Doppler velocities in a microburst (Fig. 3d). The maximum approaching and maximum receding radial velocity measurements were −18.5 and 10.5 m s−1, respectively. Differential velocity within this microburst was 29 m s−1. Measured distance between maximum approaching and maximum receding velocity observations was 2.5 km, recorded as the microburst size. The maximum base reflectivity at the center of the microburst was 56 dBZ.

While individual and line microbursts were observed during the study, this microburst was an individual microburst. Reflectivity from the next scan, at 0132 UTC, showed that the high reflectivity area of the storm had increased in size (Fig. 3e). Corresponding radial velocity indicated that the outflow from the microburst had expanded outward and weakened slightly (Fig. 3f), where the divergent outflow was 6 km in size with a differential velocity of 28 m s−1.

4. Characteristics of Sonoran Desert microbursts

a. Spatial and temporal distribution

A total of 140 unique microbursts were observed over the greater Phoenix metropolitan area during the days examined. Figure 4 is a map of microburst detection locations. Only four low-level microburst outflows were detectable by both KIWA and PHX, although not simultaneously because of differences in radar scan times. Poor temporal resolution of KIWA data, limited availability for PHX data, and differences in radar beam sampling height between KIWA and PHX all likely contributed to this very small number of common detections. Even at the lowest tilts, KIWA could not usually observe low-level divergent outflows detected by PHX, and vice versa, because of differences in sampling height between the two radars.

A histogram of microburst timing (Fig. 5) shows a significant peak in the diurnal variation of Sonoran microburst activity occurred between 1900 and 2100 local time [mountain standard time (MST), 7 h behind UTC]. The Sonoran microburst timing was different from the Denver JAWS and Alabama MIST events, where the peak in microburst activity coincided with the afternoon peak in solar heating (Fujita 1985; Atkins and Wakimoto 1991). The NIMROD, Memphis FLOWS, and Orlando, Florida, data did contain smaller, secondary peaks that suggested some nocturnal microburst activity, but the majority of events occurred in the afternoon (Fujita 1985; Wolfson et al. 1985; Rinehart and Isaminger 1986; Rinehart and Borho 1993). Uniquely, Sonoran Desert microbursts were most frequent during the evening hours.

Microbursts were usually first detected on radar during the afternoon hours along the edge of higher elevation that surrounded Phoenix and the Sonoran Desert to the north, east, and southeast. During the late afternoon and early evening hours, the storm activity propagated into the lower desert where microbursts impacted the Phoenix metropolitan area. The timing of the Sonoran Desert microburst activity was consistent with the established diurnal cycle of central Arizona monsoon convection (McCollum et al. 1995; Maddox et al. 1995; Wallace et al. 1999; Heinselman and Schultz 2006). The peak in thunderstorm activity over the lower desert during the evening hours explained the timing of the evening peak in microburst activity.

b. Size, intensity, and evolution

The monsoon circulation that transported low-to-midlevel moisture to central Arizona transformed the otherwise arid environment into an environment supportive of deep, moist convection. Initiation of low-level divergent microburst outflow always followed, or was coincident with, increased low-level reflectivity (see Figs. 3a,c). Figure 6 shows a histogram and a box-and-whisker diagram of Sonoran microburst base radar reflectivity. Sonoran microbursts were wet microbursts with a median base reflectivity of 57 dBZ at peak intensity. A previous analysis of a Sonoran downburst storm by Vasiloff and Howard (2009) was also a wet microburst. Figure 7 shows that the size of Sonoran microbursts ranged from 1 to 8 km with a peak between 2 and 4 km. The median size was 4 km.

The frequency of maximum differential velocity for all microbursts detected is shown in Fig. 8. The histogram peaked between 20 and 25 m s−1. The box-and-whisker plot above the histogram in Fig. 8 indicates with a black bar the median differential velocity, which was 24 m s−1. Stronger maximum differential velocities were observed more frequently with Sonoran microbursts than in many previously documented microburst study areas. Microbursts observed in Huntsville, Denver, Kansas City, and Orlando all had peak frequency of maximum differential velocity between 15 and 20 m s−1 (Wolfson et al. 1990). Similar to the Sonoran microbursts, microbursts observed in Memphis had a peak frequency of maximum differential velocity between 20 and 25 m s−1 (Wolfson et al. 1990). There were 24 Sonoran microbursts observed in this study that had maximum differential velocities greater than 30 m s−1. Of those 24 microbursts, seven were severe by the formal definition with a maximum radial velocity of 26 m s−1 or greater. No correlation was found between maximum microburst differential velocity and reflectivity or microburst size.

Microburst lines grew larger than individual microbursts and were associated with higher maximum differential velocities. On average, the line microbursts in this study were 3 km larger in size and had 5 m s−1 stronger differential velocities. Differences between the maximum differential velocities of microbursts detected by PHX and those detected by KIWA were not found to be statistically significant.

Sonoran microburst divergent outflows in this study were frequently observed to continue expansion outward as density currents (gust fronts). Outflows from previously observed wet microbursts were almost without exception cold (Wolfson et al. 1990). It was likely that strong temperature contrasts between the cold microburst outflow and the hot Sonoran Desert surface boundary layer air led to the microburst outflow transition to a gust front after the dissipation of the initial microburst downdraft. For example, after a brief microburst at 0127 UTC 22 July 2008 (Fig. 3d), the microburst divergent outflow quickly expanded (Fig. 3f). Figure 9 shows the continued propagation outward of the microburst outflow as a gust front away from the high-reflectivity core of the storm. At 0141 UTC, the gust front was evidenced by a visible boundary in KIWA reflectivity imagery; it is marked with the blue front symbol encircling the storm cell in the bottom of the figure (Fig. 9a). The gust front was also visible in KIWA radial velocity data (Fig. 9b). The north–south-oriented gust front marked on the right-hand side of Figs. 9a and 9b was initially generated by previous convection to the east. At 0159 UTC, the southern gust front had propagated well northward, across the KIWA site, and away from the parent storm cell whose intensity had decayed from previous scans (Figs. 9c,d). Along the gust-front boundary new convection was initiated at 0218 UTC, visible in reflectivity imagery and indicated by two white arrows (Fig. 9e). Similar development of new convection along gust-front boundaries generated by microburst outflows was common, consistent with previous studies (Roberts and Wilson 1989; Wolfson et al. 1990).

Frequently, new microbursts occurred as a result of new convection initiated from such a gust front. The 22 July 2008 example is continued in Fig. 10. Between 0218 and 0222 UTC the new convection intensified rapidly from Figs. 9e to 10a. At the same time, the classic divergent microburst signature appeared in radial velocity (marked by the white circle as microburst A; Fig. 10b). In the next scan, at 0227 UTC, the core high-reflectivity area of the right-hand cell expanded (Fig. 10c), and the divergent microburst outflow propagated outward and velocity increased (Fig. 10d). At 0232 UTC, reflectivity imagery indicated the two individual cells had merged into one larger cell (Fig. 10e). The larger storm cell produced a second microburst on the southwestern side of the storm, marked by the white circle as microburst B (Fig. 10f). Microburst B was severe, with a maximum base reflectivity of 63 dBZ, maximum radial velocity of 26 m s−1, and a maximum differential velocity of 38.5 m s−1 across a radial distance of 3.75 km. At 0237 UTC, divergent outflow from microbursts A and B merged to form a microburst line and rapidly expanded outward to a size of 8 km, and maintained a severe 26 m s−1 maximum radial velocity (Figs. 11a,b). Similar to the initial microburst and resultant gust front, shown between Figs. 3 and 9, the outflow from microbursts A and B quickly propagated outward from the parent storm as another gust front, shown in Figs. 11c–f.

5. Analysis of damage reports

a. Data and analysis

The Salt River Project (SRP) compiled a detailed list of wind-related damage incidents to overhead electrical transmission lines and poles during the 2008 summer monsoon season. Reports were automatically generated by SRP’s distribution and transmission systems. When high winds downed a power line or pole and cut electrical power transmission, SRP’s monitoring systems produced an interruption report that noted the exact time of the incident. Later, SRP engineers assessed the damage to provide details on the nature of the electrical pole or line damage. Eighty-two reports of wind-related damage occurred over the 14 storm days analyzed in section 4 (Table 2). These proprietary damage reports were not part of the Storm Prediction Center’s archived storm reports and provided valuable and precise damage information for comparison with storm and microburst wind intensity.

Radar characteristics and storm attributes that lead up to each damage report, and those at the time of the report, were documented relative to the site of the report (composite reflectivity, radial velocity). Using reflectivity and radial velocity observations at the time and location of the incident report, damage was classified as caused either by a microburst or by gust-front winds. Microburst winds resulted in 22 damage reports, and gust-front winds resulted in 60 damage reports. The microbursts that caused damage are a subset of the microbursts discussed in section 4.

Composite reflectivity over the damage sites at the time of the damage, where composite reflectivity was the maximum reflectivity observed in the vertical column above the damage site, varied over a wide range, 5 to 65 dBZ. These measurements came from the site of the damage, not the center of the microburst. In each case where the composite reflectivity was relatively low (<35 dBZ), the damage site was located where the microburst outflow expanded just outside the high reflectivity core of the parent storm.

Doppler radar radial wind velocity measured at a point (in this case, a damage site) was highly dependent on viewing angle (Hjelmfelt 1988). The principle that the microburst outflow diverges radially outward from the center in all directions was used, so it was possible to choose the radar with the best viewing angle for each damage site. The staggered nature of the lowest-level scans for the three radars made analysis possible at time scales less than 5 min when data from more than one radar were available.

Sites damaged by microburst outflows had lower average observed maximum wind velocities than the sites damaged by gust-front winds, shown in Fig. 12. Maximum radial velocities measured at microburst damage sites varied from 6.5 to 27 m s−1 with a mean of 18.8 m s−1. The maximum radial wind velocities measured at damage sites impacted by gust-front wind varied from 6 to 42 m s−1 with a mean of 25 m s−1.

The altitude of microburst velocity observations above the damage sites averaged 400 m above ground level. Operational scanning strategies with the lowest-level beam tilt of 0.5° and poor temporal resolution limited low-altitude measurements around damage report times. Without focused study closer to the ground where the damage occurred, wind speeds within the microburst outflows that caused damage could not be more accurately measured. Complex terrain also limited low-altitude radar observations at damage sites.

The electrical transmission infrastructure damage reports were classified into three categories: minor damage, where one or more lines were reported down; moderate damage, where one pole and lines were reported down; and major damage, where more than one pole and lines were reported down. Of the total microburst wind damage reports, 32% were moderate damage. Only 22% of total gust-front wind damage reports caused moderate damage. Of the total microburst damage reports, 18% had major damage (more than one pole down). Only 8% of gust-front wind events caused major damage. In these damage reports, the microburst winds caused more significant damage than the gust-front winds.

b. Damage example from 22 July 2008

Eight damage reports on 22 July 2008 resulted from microburst B shown in Fig. 10f. Table 3 lists these damage reports. The first damage report occurred at 0233 UTC, when four power poles were brought down (location is shown in Fig. 13 by the red dot). Shown in Fig. 13a, radial velocity imagery from SR1 at 0230 UTC contains three distinct microburst divergence signatures, marked by the white circles, 3 min prior to the damage (data from SR1 were of good quality for this case). Microburst signatures A1 and A2 were collocated with microburst A initially detected at 0223 UTC in KIWA imagery (Figs. 10b,d). The higher-resolution SR1 radar observed two distinct microburst divergence signatures, where KIWA observed only one larger microburst divergence signature. Microburst signature B was the developing divergence signature of the microburst B shown in KIWA imagery (Fig. 10f). In Fig. 13a, the damage site was located along the zero line of Doppler velocity at the center of microburst B, 3 min prior to the damage report.

Two minutes prior to the power pole damage, KIWA radial velocity imagery at 0231 UTC in Fig. 13b showed how the microburst outflow expanded outward and intensified. Wind velocity at the damage site increased to 10 m s−1 when the zero line of Doppler velocity at the center of the microburst shifted south slightly from its location in Fig. 13a to its location in Fig. 13b. Two minutes after the damage occurred, SR1 radial velocity at 0235 UTC in Fig. 13c indicated a radial velocity of 23.5 m s−1 above the damage site. In Fig. 13c, the center of the microburst was located about 1 km southeast of the damage site. The power pole damage occurred between the available radial velocity scans in Figs. 13b and 13c, so there was uncertainty about the winds at the exact time of the damage. The radar beam height for SR1 was 150 m above the ground at the damage site, and 200 m for KIWA. Two minutes prior to the damage incident, base reflectivity at the damage site at 0231 UTC was 56 dBZ (Fig. 13e). Composite reflectivity was 65 dBZ. This indicated very heavy rainfall at the damage site.

6. Summary and conclusions

The characteristics of 140 Sonoran microbursts were observed over 14 storm days during the 2008 summer monsoon season in central Arizona. A unique nocturnal maximum in diurnal frequency of microburst activity suggested that the timing of Sonoran microbursts was connected to the diurnal cycle of monsoon convection in the central Arizona. All of the Sonoran microbursts observed were wet microbursts, and many of the divergent outflows rapidly expanded outward into larger gust-front features. Growth and expansion of outflows into gust fronts frequently contributed to initiation of new convection.

Higher maximum differential velocities occurred more frequently in Sonoran microbursts than in microbursts previously observed in many other areas. There were also several strong and some severe microbursts observed in this study. However, many of the Sonoran microbursts were observed from KIWA, for which the poor temporal resolution of lowest elevation angle scans and higher altitude of sampling likely limited the accuracy of the wind velocity measurements within the microbursts. More detailed study of the Sonoran microbursts that were well observed by multiple radars below 200 m would provide for a more rigorous comparison of the relative strength of Sonoran microbursts compared to those previously documented.

The Sonoran microbursts observed in this study caused more significant damage to overhead electrical transmission infrastructure than gust-front wind events, despite lower observed peak wind speeds. This was not a surprise given the physical nature of microbursts, which caused the most intense winds closest to the surface. Based on the severity of damage caused by microbursts, it is speculated that the radar velocity observations within the microbursts at the damage sites may be underrepresentative of the true winds at the surface. Operational radar scanning strategies did not allow intensive study near ground level, and infrequent scanning at the lowest elevation angles limited observations of microburst evolution.

A follow-up experiment is desirable to better focus on collecting high temporal and spatial resolution observations near the surface to obtain more accurate wind observations of Sonoran microburst outflows and to perform a more detailed comparison of their intensity to surface damage.

Acknowledgments

The SMART-R deployment and data collection and the power transmission damage summary were facilitated by the Salt River Project. The authors thank Charles Ester of the Salt River Project Water Operations Group for his continued support. The authors also thank Ami Arthur for providing the maps and hybrid scan height figures. This research was supported in part by an appointment to the National Oceanic and Atmospheric Administration Ernest F. Hollings Undergraduate Scholarship Program through a grant award to Oak Ridge Institute for Science and Education.

REFERENCES

  • Atkins, N. T., , and Wakimoto R. M. , 1991: Wet microburst activity over the southeastern United States: Implications for forecasting. Wea. Forecasting, 6 , 470482.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Balling, R. C., , and Brazel S. W. , 1987: Diurnal variations in Arizona monsoon precipitation frequencies. Mon. Wea. Rev., 115 , 342346.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., and Coauthors, 2005: The Shared Mobile Atmospheric Research and Teaching Radar: A collaboration to enhance research and teaching. Bull. Amer. Meteor. Soc., 86 , 12631274.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Biron, P. J., , and Isaminger M. A. , 1989: An analysis of microburst characteristics related to automatic detection from Huntsville, Alabama, and Denver, Colorado. Preprints, 24th Conf. on Radar Meteorology, Tallahassee, FL, Amer. Meteor. Soc., 269–273.

    • Search Google Scholar
    • Export Citation
  • Caracena, F., , and Maier M. W. , 1987: Analysis of a microburst in the FACE meteorological mesonetwork in southern Florida. Mon. Wea. Rev., 115 , 969985.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dodge, J., , Arnold J. , , Wilson G. , , Evans J. , , and Fujita T. T. , 1986: The Cooperative Huntsville Meteorological Experiment (COHMEX). Bull. Amer. Meteor. Soc., 67 , 417419.

    • Search Google Scholar
    • Export Citation
  • Eilts, M. D., , and Doviak R. J. , 1987: Oklahoma downbursts and their asymmetry. J. Climate Appl. Meteor., 26 , 6978.

  • Elmore, K. L., 1986: Evolution of a microburst and bow-shaped echo during JAWS. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 101–104.

    • Search Google Scholar
    • Export Citation
  • Evans, J. E., , and Johnson D. , 1984: The FAA transportable Doppler weather radar. Preprints, 22nd Int. Conf. on Radar Meteorology, Zurich, Switzerland, Amer. Meteor. Soc., 246–250.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1978: Manual of downburst identification for project NIMROD. SMRP Research Paper 156, University of Chicago, 104 pp. [NTIS PB-2860481].

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1979: Objectives, operation, and results of Project NIMROD. Preprints, 11th Conf. on Severe Local Storms, Kansas City, MO, Amer. Meteor. Soc., 259–266.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38 , 15111534.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1985: The downburst. SMRP Research Paper 210, University of Chicago, 122 pp. [NTIS PB-148880].

  • Hales, J. E., 1977: On the relationship of convective cooling to nocturnal thunderstorms at Phoenix. Mon. Wea. Rev., 105 , 16091613.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Heinselman, P. L., , and Schultz D. M. , 2006: Intraseasonal variability of summer storms over central Arizona during 1997 and 1999. Wea. Forecasting, 21 , 559578.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hjelmfelt, M. R., 1984: Radar and surface data analysis of a microburst in JAWS. Preprints, 22nd Conf. on Radar Meteorology, Zurich, Switzerland, Amer. Meteor. Soc., 64–69.

    • Search Google Scholar
    • Export Citation
  • Hjelmfelt, M. R., 1988: Structure and life cycle of microburst outflows observed in Colorado. J. Appl. Meteor., 27 , 900927.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kessinger, C. J., , Roberts R. D. , , and Elmore K. L. , 1986: A summary of microburst characteristics from low-reflectivity storms. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 105–108.

    • Search Google Scholar
    • Export Citation
  • Kingsmill, D. E., , and Wakimoto R. M. , 1991: Kinematic, dynamic, and thermodynamic analysis of a weakly sheared severe thunderstorm over northern Alabama. Mon. Wea. Rev., 119 , 262297.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • MacKeen, P. L., , and Zhang J. , 2000: Convective climatology for central Arizona during the 1999 monsoon. Proc. Southwest Weather Symp., Tucson, AZ, National Weather Service, 64–67.

    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., , McCollum D. M. , , and Howard K. W. , 1995: Large-scale patterns associated with severe summertime thunderstorms over central Arizona. Wea. Forecasting, 10 , 763778.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCarthy, J., , Wilson J. W. , , and Fujita T. T. , 1982: The Joint Airport Weather Studies Project. Bull. Amer. Meteor. Soc., 63 , 15.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCollum, D. M., , Maddox R. A. , , and Howard K. W. , 1995: Case study of a severe mesoscale convective system in central Arizona. Wea. Forecasting, 10 , 643665.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mielke, K. B., , and Carle E. R. , 1987: An early morning dry microburst in the Great Basin. Wea. Forecasting, 2 , 169174.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , and Isaminger M. A. , 1986: Radar characteristics of microbursts in the mid-south. Preprints, 23rd Conf. on Radar Meteorology, Snowmass, CO, Amer. Meteor. Soc., 116–119.

    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , and Borho A. , 1993: A comparison of the detectability of microbursts in Orlando, Florida, by two C-band Doppler radars. J. Appl. Meteor., 32 , 476489.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rinehart, R. E., , Borho A. , , and Curtiss C. , 1995: Microburst rotation: Simulations and observations. J. Appl. Meteor., 34 , 12671285.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roberts, R. D., , and Wilson J. W. , 1989: A proposed microburst nowcasting procedure using single-Doppler radar. J. Appl. Meteor., 28 , 285303.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Turnbull, D., , McCarthy J. , , Evans J. , , and Zrnic D. , 1989: The FAA Terminal Doppler Weather Radar (TDWR) program. Preprints, Third Int. Conf. on Aviation Weather Systems, Anaheim, CA, Amer. Meteor. Soc., 414–418.

    • Search Google Scholar
    • Export Citation
  • Vasiloff, S. V., 2001: Improving tornado warnings with the Federal Aviation Administration’s Terminal Doppler Weather Radar. Bull. Amer. Meteor. Soc., 82 , 861874.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vasiloff, S. V., , and Howard K. W. , 2009: Investigation of a severe downburst storm near Phoenix, Arizona, as seen by a mobile Doppler radar and the KIWA WSR-88D. Wea. Forecasting, 24 , 856867.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., , and Bringi V. , 1988: Dual-polarization observations of microbursts associated with intense convection: The 20 July storm during the MIST Project. Mon. Wea. Rev., 116 , 15211539.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wallace, C. E., , Maddox R. A. , , and Howard K. W. , 1999: Summertime convective storm environments in central Arizona: Local observations. Wea. Forecasting, 14 , 9941006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Watson, A. I., , López R. E. , , and Holle R. L. , 1994: Diurnal cloud-to-ground lightning patterns in Arizona during the southwest monsoon. Mon. Wea. Rev., 122 , 17161725.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., , Roberts R. D. , , Kessinger C. , , and McCarthy J. , 1984: Microburst wind structure and evaluation of Doppler radar for airport wind shear detection. J. Appl. Meteor., 23 , 898915.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wolfson, M. M., , Distefano J. T. , , and Fujita T. T. , 1985: Low-altitude wind shear characteristics in the Memphis, TN, area based on mesonet and LLWAS data. Preprints, 14th Conf. on Severe Local Storms, Indianapolis, IN, Amer. Meteor. Soc., 322–327.

    • Search Google Scholar
    • Export Citation
  • Wolfson, M. M., and Coauthors, 1990: Characteristics of thunderstorm-generated low altitude wind shear: A survey based on nationwide Terminal Doppler Weather Radar testbed measurements. Proc. 29th IEEE Conf. on Decision and Control, Honolulu, HI, Institute of Electrical and Electronics Engineers, 682–688.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Photograph of a Sonoran Desert microburst on 13 Jul 2008 that occurred near overhead electrical power lines. (Courtesy of K. Howard)

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 2.
Fig. 2.

(a) Map of the topography of Arizona. The elevated terrain of the Colorado Plateau, the White Mountains, and the Mogollon Rim creates a sharp topographic gradient from northeast to southwest across central Arizona. The box at center encompasses the region over which analysis occurred. (b) The analysis region with the Phoenix metropolitan area outlined and radar locations indicated, encircled by 50-km range rings. Most of the viewable area from SR1 is overlapped by PHX and KIWA.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 3.
Fig. 3.

The 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0122 UTC; (b) velocity at 0123 UTC; (c) reflectivity at 0127 UTC; (d) velocity at 0127 UTC, where maximum outbound velocity was 10.5 m s−1 and maximum inbound velocity was −18.5 m s−1; (e) reflectivity at 0132 UTC; and (f) velocity at 0132 UTC. The 0.9° tilt was used here because the 0.5° tilt was mostly blocked by terrain. The 0.9° tilt was still partially blocked, as evidenced by the missing data (black patches) in the images. In the radial velocity images, red colors indicate inbound velocities and green colors indicate outbound velocities.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 4.
Fig. 4.

Microburst detection locations for summer 2008 (region encompassed by box in Fig. 2a). Red markers indicate a microburst observed from KIWA; blue markers indicate PHX. The four microbursts that were viewed from both KIWA and PHX are in purple. The radar locations are labeled. SR1 data were not used in the microburst count and characterization.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 5.
Fig. 5.

Diurnal variation of the Sonoran microbursts observed during summer 2008 (MST is 7 h behind UTC).

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 6.
Fig. 6.

Distribution of Sonoran microburst base radar reflectivity observed during summer 2008. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum still within a 1.5 interquartile range. The six dots to the left of the diagram represent outliers (data outside the 1.5 interquartile range). Sonoran microbursts are moderate to high reflectivity, or wet, microbursts. The median base reflectivity was 57 dBZ.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 7.
Fig. 7.

Distribution of Sonoran microburst size observed during summer 2008. Microburst size was measured from the maximum approaching radial velocity to the maximum receding radial velocity. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum. The median size was 4 km.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 8.
Fig. 8.

Distribution of Sonoran microburst maximum differential velocity observed during summer 2008. A box-and-whisker diagram is shown above the histogram. The left and right ends of the box represent the 25th and 75th percentile, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum still within a 1.5 interquartile range. The six dots to the right of the diagram represent outliers (data outside the 1.5 interquartile range). This histogram has a peak frequency of 20 to 25 m s−1, similar to microbursts from Memphis, Tennessee (Wolfson et al. 1990). The median differential velocity was 24 m s−1.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 9.
Fig. 9.

The 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of a gust front moving north from the location of a previous microburst and initiating new convection, showing (a) reflectivity at 0141 UTC; (b) velocity at 0141 UTC; (c) reflectivity at 0159 UTC; (d) velocity at 0200 UTC; (e) reflectivity at 0218 UTC, where new convective development is indicated by the white arrows; and (f) velocity at 0218 UTC. Gust-front outflow boundaries are indicated by the blue front symbol, where the symbol orientation refers to the direction the gust front is moving. The gust-front boundary is easier to see in the 0.9° elevation angle than the 0.5°.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 10.
Fig. 10.

The new convection from Fig. 9e produced two new microbursts. Shown are three consecutive 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0222 UTC; (b) velocity at 0223 UTC, with microburst A circled; (c) reflectivity at 0227 UTC; (d) velocity at 0227 UTC; (e) reflectivity at 0232 UTC; and (f) velocity at 0232 UTC, with microburst B circled. Microburst B was severe, with a maximum radial velocity of 26 m s−1.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 11.
Fig. 11.

After microburst B in Fig. 10, the microburst outflow transitioned to a gust front and propagated outward from the high-reflectivity core of the storm. Shown are three consecutive 22 Jul 2008 KIWA radar observations at 0.9° elevation angle of (a) reflectivity at 0236 UTC, (b) velocity at 0237 UTC, (c) reflectivity at 0241 UTC, (d) velocity at 0241 UTC, (e) reflectivity at 0245 UTC, and (f) velocity at 0246 UTC. Gust-front outflow boundaries are indicated by the blue front symbol.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 12.
Fig. 12.

Comparison of maximum radial Doppler velocities observed at sites damaged by gust-front winds vs sites damaged by microburst winds by box-and-whisker diagrams for the summer 2008 damage reports. The bottom and top of the box represent the 25th and 75th percentiles, and the dark band near the middle of the box is the median. The ends of the whiskers represent the lowest and highest datum.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Fig. 13.
Fig. 13.

Radar observations around the time of the damage report at 0233 UTC 22 Jul 2008. All subfigures cover the same area. (a) SR1 0.5° velocity at 0230:08 UTC with microburst signatures A1, A2, and B indicated; (b) KIWA 0.5° velocity at 0231:44 UTC; (c) SR1 0.5° velocity at 0235:06 UTC; (d) KIWA 0.5° velocity at 0236:21 UTC; and (e) KIWA 0.5° reflectivity at 0231:26 UTC. Damage site is indicated by the red dot.

Citation: Weather and Forecasting 26, 1; 10.1175/2010WAF2222388.1

Table 1.

Characteristics of the KIWA, PHX, and SR1 radars.

Table 1.
Table 2.

Case days for 2008 where analysis was performed; “NA” indicates data were not available. SR1 data were only available for limited hours during these days. Most microbursts occurred in the evening, so the cases continued past 0000 UTC and bridged two calendar days.

Table 2.
Table 3.

Damage reports to SRP electrical power infrastructure (overhead power lines and poles) from 22 Jul 2008. Composite reflectivity and maximum velocity measurements were taken at the damage site. Composite reflectivity is the maximum reflectivity value in the vertical column above the damage site. Maximum velocity is the peak radar radial velocity observed above the damage site in the closest available radar scan to the time of the damage.

Table 3.
Save