• Baker-Blocker, A., 1982: Winter weather and cardiovascular mortality in Minneapolis–St. Paul. Amer. J. Public Health, 72, 261265.

  • Balling, R. C., Jr., , and S. W. Brazel, 1987: Diurnal variations in Arizona monsoon precipitation frequencies. Mon. Wea. Rev., 115, 342346.

    • Search Google Scholar
    • Export Citation
  • Bentley, M. L., , W. S. Ashley, , and J. A. Stallins, 2009: Climatological radar delineation of urban convection for Atlanta, Georgia. Int. J. Climatol., 30, 15891594.

    • Search Google Scholar
    • Export Citation
  • Bornstein, R. D., 1968: Observations of the urban heat island effect in New York City. J. Appl. Meteor., 7, 575582.

  • Bornstein, R. D., , and M. LeRoy, 1990: Urban barrier effects on convective and frontal thunderstorms. Extended Abstracts, Conf. on Mesoscale Processes, Boulder, CO, Amer. Meteor. Soc., 25–29.

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

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

    • Search Google Scholar
    • Export Citation
  • Braham, R. R., Jr., , R. G. Semonin, , A. H. Auer, , S. A. Changnon Jr., , and J. M. Hales, 1981: Summary of urban effects on clouds and rain. METROMEX: A Review and Summary, Meteor. Monogr., No. 40, Amer. Meteor. Soc., 141–152.

  • Burian, S. J., , and J. M. Shepherd, 2005: Effect of urbanization on the diurnal rainfall pattern in Houston. Hydrol. Processes, 19, 10891103.

    • Search Google Scholar
    • Export Citation
  • Cain, D. R., cited 2012: Doppler radar frequently asked questions. National Weather Service Southern Region Headquarters. [Available online at http://www.srh.noaa.gov/srh/sod/radar/radinfo/radinfo.html.]

  • Changnon, S. A., Jr., 1968: Precipitation climatology of Lake Michigan Basin. Illinois State Water Survey Bull. 52, Urbana, IL, 46 pp.

  • Changnon, S. A., Jr., 1992: Inadvertent weather modification in urban areas: Lessons for global climate change. Bull. Amer. Meteor. Soc., 73, 619627.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., Jr., 2004: Urban effects on winter snowfall at Chicago and St. Louis. Bull. Ill. Geogr. Soc., 46, 312.

  • Changnon, S. A., Jr., , and N. E. Westcott, 2002: Heavy rainstorms in Chicago: Increasing frequency, altered impacts, and future implications. J. Amer. Water Resour. Assoc., 38, 14671475.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., Jr., , F. A. Huff, , P. T. Schickedanz, , and J. L. Vogel, 1977: Weather anomalies and impacts. Vol. I, Summary of METROMEX, Illinois State Water Survey Bull. 62, Urbana, IL, 260 pp.

  • Changnon, S. A., Jr., , R. T. Shealy, , and R. W. Scott, 1991: Precipitation changes in fall, winter, and spring caused by St. Louis. J. Appl. Meteor., 30, 126134.

    • Search Google Scholar
    • Export Citation
  • Crum, T. D., , R. L. Alberty, , and D. W. Burgess, 1993: Recording, archiving, and using WSR-88D data. Bull. Amer. Meteor. Soc., 74, 645653.

    • Search Google Scholar
    • Export Citation
  • Diem, J. E., , and D. P. Brown, 2003: Anthropogenic impacts on summer precipitation in central Arizona, U.S.A. Prof. Geogr., 55, 343355.

    • Search Google Scholar
    • Export Citation
  • Dixon, P. G., , and T. L. Mote, 2003: Patterns and causes of Atlanta's urban heat island–initiated precipitation. J. Appl. Meteor., 42, 12731284.

    • Search Google Scholar
    • Export Citation
  • Evans, M. S., 1997: A study on the relationship between synoptic-scale model forecasts of vertical motion and snow amounts across southwest Lower Michigan in lake effect snow environments. National Weather Service Central Region Applied Res. Paper 17-11, 17 pp.

  • Evans, M. S., , and R. Murphy, 2008: A proposed methodology for reconciling high-resolution numerical modeling guidance with pattern recognition to predict lake-effect snow. Electron. J. Oper. Meteor., 9 (2). [Available online at http://www.nwas.org/ej/2008/2008.php.]

    • Search Google Scholar
    • Export Citation
  • Gallo, K. P., , and T. W. Owen, 1999: Satellite-based adjustments for the urban heat island temperature bias. J. Appl. Meteor., 38, 806813.

    • Search Google Scholar
    • Export Citation
  • Grillo, J. N., , and J. Spar, 1971: Rain–snow mesoclimatology of the New York metropolitan area. J. Appl. Meteor., 10, 5661.

  • Holroyd, E. W., 1971: Lake-effect cloud bands as seen from weather satellites. J. Atmos. Sci., 28, 11651170.

  • Huff, F. A., , and S. A. Changnon, 1972: Climatological assessment of urban effects on precipitation at St. Louis. J. Appl. Meteor., 11, 823842.

    • Search Google Scholar
    • Export Citation
  • Huff, F. A., , and J. L. Vogel, 1978: Urban, topographic and diurnal effects on rainfall in the St. Louis region. J. Appl. Meteor., 17, 565577.

    • Search Google Scholar
    • Export Citation
  • Huff, T. E., 1986: Urban hydrometeorology review. Bull. Amer. Meteor. Soc., 67, 703711.

  • Jauregui, E., , and E. Romales, 1996: Urban effects on convective precipitation in Mexico City. Atmos. Environ., 30, 33833389.

  • Jones, P. A., , and J. E. Jiusto, 1980: Some local climate trends in four cities of New York State. J. Appl. Meteor., 19, 135141.

  • Kidder, S. Q., , and H.-T. Wu, 1987: A multispectral study of the St. Louis area under snow-covered conditions using NOAA-7 AVHRR data. Remote Sens. Environ., 22, 159172.

    • Search Google Scholar
    • Export Citation
  • Landsberg, H. E., 1981: The Urban Climate. Academic Press, 275 pp.

  • Malevich, S. B., , and K. Klink, 2011: Relationships between snow and the wintertime Minneapolis urban heat island. J. Appl. Meteor. Climatol., 50, 18841894.

    • Search Google Scholar
    • Export Citation
  • Mölders, N., , and M. A. Olson, 2004: Impact of urban effects on precipitation in high latitudes. J. Hydrometeor., 5, 409429.

  • Mote, T. L., , M. C. Lacke, , and J. M. Shepherd, 2007: Radar signatures of the urban effect on precipitation distribution: A case study for Atlanta, Georgia. Geophys. Res. Lett., 34, L20710, doi:10.1029/2007GL031903.

    • Search Google Scholar
    • Export Citation
  • Niziol, T. A., 1987: Operational forecasting of lake effect snowfall in western and central New York. Wea. Forecasting, 2, 310321.

  • Oke, T. R., 1987: Boundary Layer Climates. 2nd ed. Routledge, 435 pp.

  • Rinehart, R. E., 1997: Radar for Meteorologists. 3rd ed. Rinehart, 428 pp.

  • Rosenfeld, D., 1999: TRMM observed first direct evidence of smoke from forest fires inhibiting rainfall. Geophys. Res. Lett., 26, 31053108.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., 2000: Suppression of rain and snow by urban and industrial air pollution. Science, 287, 17931796.

  • Seeley, M. W., , and B. Jensen, 2006: Minnesota Weather Almanac. Minnesota Historical Society Press, 294 pp.

  • Shepherd, J. M., 2005: A review of current investigations of urban-induced rainfall and recommendations for the future. Earth Interact., 9. [Available online at http://EarthInteractions.org.]

    • Search Google Scholar
    • Export Citation
  • Shepherd, J. M., , and S. J. Burian, 2003: Detection of urban-induced rainfall anomalies in a major coastal city. Earth Interact., 7. [Available online at http://EarthInteractions.org.]

    • Search Google Scholar
    • Export Citation
  • Shepherd, J. M., , H. Pierce, , and A. J. Negri, 2002: Rainfall modification by major urban areas: Observations from spaceborne rain radar on the TRMM satellite. J. Appl. Meteor., 41, 689701.

    • Search Google Scholar
    • Export Citation
  • Tardy, A., 2000: Lake effect and lake enhanced snow in the Champlain Valley of Vermont. National Weather Service Eastern Region Tech. Attach. 2000-05, 27 pp.

  • Todhunter, P. E., 1996: Environmental indices for the Twin Cities metropolitan area (Minnesota, USA) urban heat island—1989. Climate Res., 06, 5969.

    • Search Google Scholar
    • Export Citation
  • Van den Berg, W., 2008: Fog turning into snow. Weather, 63, 248248.

  • Vasquez, T., 2002: Weather Forecasting Handbook. 5th ed. Weather Graphics Techologies, 198 pp.

  • Winkler, J. A., , R. H. Skaggs, , and D. G. Baker, 1981: Effect of temperature adjustments on the Minneapolis–St. Paul urban heat island. J. Appl. Meteor., 20, 12951300.

    • Search Google Scholar
    • Export Citation
  • Wood, C. R., , and R. G. Harrison, 2009: Anthropogenic snowfall events in the UK: Examples of urban weather modification? Weather, 64, 277280.

    • Search Google Scholar
    • Export Citation
  • Woollum, C. A., , and N. L. Canfield, 1968: Washington metropolitan area precipitation and temperature patterns. Environmental Science Services Tech. Mem. WBTM 28, Weather Bureau, Eastern Region, Washington, DC, 32 pp.

  • Yuan, F., , K. E. Sawaya, , B. C. Loeffelholz, , and M. E. Bauer, 2005: Land cover classification and change analysis of the Twin Cities (Minnesota) metropolitan area by multitemporal Landsat remote sensing. Remote Sens. Environ., 98, 317328.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Locations of urban (MSP) and rural (all other) thermometers used in the individual event analyses.

  • View in gallery

    (a) Total maximum dBZ sum and (b) total maximum dBZ maximums from composite analyses. The 72- and 64-km-radius city buffers are bisected with lines perpendicular to the 3-h southwest wind direction to denote upwind and downwind urban areas.

  • View in gallery

    The maximum dBZ sum with the 169- and 64-km city buffers for the four events found to have increases in the greatest maximum dBZs downwind: (a) 8 Dec 1995, (b) 13 Mar 1997, (c) 14 Mar 2002, and (c) 4 Dec 2007.

  • View in gallery

    Difference profiles (enhanced − all-event averages) for the 0000 UTC category. Temperature is the darker, black line, and dewpoint is the lighter, gray line.

  • View in gallery

    As in Fig. 4, but for the 1200 UTC category.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 14 14 9
PDF Downloads 5 5 2

A Radar Analysis of Urban Snowfall Modification in Minneapolis–St. Paul

View More View Less
  • 1 Desert Research Institute, Reno, Nevada
  • 2 Mississippi State University, Mississippi State, Mississippi
© Get Permissions
Full access

Abstract

A better understanding of urban snowfall climatology will help to mitigate winter weather hazards in highly populated cities, such as Minneapolis–St. Paul, Minnesota. Winter road maintenance accounts for roughly 25% of Minnesota Department of Transportation maintenance budgets, and state and local agencies spend more than $2.3 billion on snow and ice control operations annually. Urban snowfall has also been shown to enhance health problems and increase mortality rates. Further research on urban snowfall climatology is needed to more accurately diagnose urban regions susceptible to such winter health risks. The winter urban heat island effect has been suggested to reduce snowfall downwind of city centers because of localized energy and moisture flux variations, but previous research lacks spatial detail since it is primarily based on sparse surface observations. This project utilizes radar data for two studies—a 25-event snowfall composite and an individual-event analysis of 13 snow-only events—occurring from 1995 to 2012 and passing over the Minneapolis–St. Paul urban area to quantify the change in radar reflectivity values as a proxy for snowfall intensity downwind of the city. Results show that for the summed maximum composite event snowfall was significantly decreased downwind of the urban region; however, the highest maximum composite event, as well as 4 of the 13 individual snowfall events evaluated, did not have significantly decreased snowfall downwind of the city center, with the highest maximum composite and two of the three individual events having increased reflectivity values downwind. Analysis of related atmospheric variables for events with increased downwind reflectivity suggests that atmospheric instability and convergence may play a critical role in urban snowfall modification.

Corresponding author address: Nyssa Perryman, Division of Atmospheric Sciences, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512. E-mail: nyssa.perryman@dri.edu

Abstract

A better understanding of urban snowfall climatology will help to mitigate winter weather hazards in highly populated cities, such as Minneapolis–St. Paul, Minnesota. Winter road maintenance accounts for roughly 25% of Minnesota Department of Transportation maintenance budgets, and state and local agencies spend more than $2.3 billion on snow and ice control operations annually. Urban snowfall has also been shown to enhance health problems and increase mortality rates. Further research on urban snowfall climatology is needed to more accurately diagnose urban regions susceptible to such winter health risks. The winter urban heat island effect has been suggested to reduce snowfall downwind of city centers because of localized energy and moisture flux variations, but previous research lacks spatial detail since it is primarily based on sparse surface observations. This project utilizes radar data for two studies—a 25-event snowfall composite and an individual-event analysis of 13 snow-only events—occurring from 1995 to 2012 and passing over the Minneapolis–St. Paul urban area to quantify the change in radar reflectivity values as a proxy for snowfall intensity downwind of the city. Results show that for the summed maximum composite event snowfall was significantly decreased downwind of the urban region; however, the highest maximum composite event, as well as 4 of the 13 individual snowfall events evaluated, did not have significantly decreased snowfall downwind of the city center, with the highest maximum composite and two of the three individual events having increased reflectivity values downwind. Analysis of related atmospheric variables for events with increased downwind reflectivity suggests that atmospheric instability and convergence may play a critical role in urban snowfall modification.

Corresponding author address: Nyssa Perryman, Division of Atmospheric Sciences, Desert Research Institute, 2215 Raggio Parkway, Reno, NV 89512. E-mail: nyssa.perryman@dri.edu

1. Introduction

The urban heat island effect (UHI) occurs when cities remain warmer than surrounding rural areas because of reduced cooling during late afternoon and evening hours under relatively cloudless and calm weather conditions (Bornstein 1968; Oke 1987). This UHI thermal gradient, and other urban characteristics, such as wind convergence and enhanced aerosols for cloud condensation nuclei, can enhance liquid precipitation downwind of cities (Shepherd 2005), where rainfall can be more than 30% heavier than upwind of a city (Mote et al. 2007). Numerous studies have attributed both the initiation and enhancement of rainfall to the UHI (Changnon 1968, 1992; Huff and Changnon 1972; Changnon et al. 1977, 1991; Huff and Vogel 1978; Braham et al. 1981; Landsberg 1981; Huff 1986; Balling and Brazel 1987; Jauregui and Romales 1996; Changnon and Westcott 2002; Diem and Brown 2003; Dixon and Mote 2003; Burian and Shepherd 2005; Shepherd 2005).

Research on the UHI mechanisms influencing urban snowfall is sparse and lacking in detail. Early research on UHI snowfall modification suggests that reduced snowfall downwind of major U.S. cities—such as Washington, D.C. (Woollum and Canfield 1968); Chicago, Illinois (Landsberg 1981); and New York City, New York (Grillo and Spar 1971; Jones and Jiusto 1980)—occurs as a result of higher urban temperatures, but these studies do not consider the warming influence of large water bodies impacting their studies and use only observations taken at the surface. Kidder and Wu (1987) used only two passes of the National Oceanic and Atmospheric Administration-7 (NOAA-7) Advanced Very High Resolution Radiometer (AVHRR) satellite over the city of St. Louis, Missouri, to compare the difference in urban and rural snow cover albedos—finding a 15%–25% in-city decrease in albedo. Grillo and Spar (1971), in compiling a climatology of snowfall for the New York City metropolitan area, found a decrease in snowfall over the lower Manhattan–North Queens–Bronx “heat island”; however, hourly measurements of wind, pressure, and humidity for the surface snowfall measurements were omitted and the authors thus warned that the results were based on relatively small samples of inhomogeneous data and, therefore, open to question (Grillo and Spar 1971). Changnon (2004) also used surface snowfall measurements and found a decrease in snow amount downwind of the St. Louis UHI, but the study warned that the four-winter sample used is not adequate to evaluate urban effects on snowfall with a high degree of reliability. Woollum and Canfield (1968) found a 20% decrease in snowfall in the Washington, D.C., region but comment that fewer than 10 stations provided continuous measurements for the 20 yr of record they considered (Woollum and Canfield 1968). When discussing the snowfall distribution across Washington, D.C., Woollum and Canfield (1968) stated that both elevation and the UHI effect were factors in the pattern of long-term snowfall totals (Woollum and Canfield 1968), further complicating the evaluation of the UHI, alone. The authors of this study also warned that natural variability and measurement difficulties (i.e., distinguishing actual snowfall from snow drifts) cause considerable data variability and make the selection of a single snowiest month quite arbitrary (Woollum and Canfield 1968), showing the complications of determining the impact of the UHI on snowfall amounts by using surface observations alone. More recent studies of snowfall enhancement downwind of urban areas have been generally focused on aerosol effects (Van den Berg 2008; Wood and Harrison 2009), which conflicting research has shown to both enhance (Landsberg 1981; Mölders and Olson 2004) and suppress (Rosenfeld 1999, 2000; Borys et al. 2000, 2003) snowfall downwind of urban centers. Thus, in an attempt to improve upon the methods of these UHI snowfall modification studies, the study presented seeks to include both a large number of events spanning multiple years and several snowfall measuring techniques that will serve to supplement surface observations for an urban region positioned in relatively flat, homogeneous topography. Determining the more complex role of aerosols was excluded in this study.

Ground- and space-based radar data have been implemented in previous UHI rainfall modification studies to aid in addressing the questions of both the mechanisms and magnitude of the urban effect on precipitation (Bornstein and LeRoy 1990; Shepherd et al. 2002; Shepherd and Burian 2003; Mote et al. 2007; Bentley et al. 2009). Similarly, in the study presented, it is hypothesized that by revealing changes in reflectivity, and thus intensity, of snowfall particles above the surface, radar reflectivity data for urban snowfall events can be used to supplement surface snowfall measurements in order to more accurately determine the impact, if any, of UHI mechanisms on urban winter precipitation.

Minneapolis–St. Paul was selected for this study because this region is a large, rapidly growing urban area. According to Yuan et al. (2005), from 1974 to 2000, the population of the Minneapolis–St. Paul metropolitan area increased by 38% while the urban land area increased by 59%, indicating rapid urbanization. Another reason for selecting Minneapolis–St. Paul for this snowfall study is the unexplained correlation between winter temperature and snowfall records that this city exhibits. The long-term average seasonal snowfall in the Minneapolis–St. Paul area since 1885 is 45.8 in. (1 in. = 25.4 mm), and the current 30-yr normal (1971–2000) shows 55.9 in., an upward trend that corresponds with the Minnesota statewide statistic that 5 of the 10 warmest winters since 1895 have occurred since 1987 (Seeley and Jensen 2006). The Twin Cities of Minneapolis–St. Paul have also been the focus of other urban heat island studies because the city is not in close proximity to a large body of water or topography that would influence precipitation and make it difficult to assess only the meteorological effects of the urban environment (Winkler et al. 1981; Todhunter 1996). Recently, Malevich and Klink (2011) implemented inexpensive thermochrons to determine what impact snowfall had on the wintertime Minneapolis–St. Paul UHI, rather than vice versa. The study found that the winter UHI was stronger during the day than at night, and that snow accumulations greater than 5 cm reduced the diurnal UHI (Malevich and Klink 2011). In future work, microscale temperature data can be used to better decipher how the UHI impacts snowfall events, if collected for a longer time period.

A better understanding of urban snowfall climatology, especially in a region prone to regular snowfall events such as Minneapolis–St. Paul, will help to mitigate winter weather hazards in this highly populated metropolis. According to the Minnesota Department of Transportation (Mn/DOT) (http://www.newsline.dot.state.mn.us), snow and ice operations account for approximately 25% of the Mn/DOT annual maintenance budget. Urban snowfall has also been shown to increase mortality rates. Baker-Blocker (1982) found that snow events in Minneapolis–St. Paul impacted mortality rates on the day of occurrence, as well as up to 2 days following the snowfall, and that snowfall is also somewhat more important in triggering deaths from heart disease than is air temperature.

As previously stated, past UHI snowfall modification studies have concluded that snowfall is decreased downwind of the urban region due to melting by the warmer urban environment (Woollum and Canfield 1968; Landsberg 1981; Grillo and Spar 1971; Jones and Jiusto 1980). However, when reviewing the dynamic and thermodynamic characteristics of the UHI environment, many similarities were found between these characteristics and those of a classic lake-enhanced snowfall environment. A comparison of the wintertime UHI environment to a lake-enhanced snowfall environment led to the research hypothesis that there were enough similarities between these atmospheric settings to expect similar snowfall enhancements downwind of a city.

Similarities to a lake-enhanced snowfall environment

Three major similarities exist between lake-enhanced snow environments and the winter UHI: instability due to the vertical temperature gradient, moisture input, and convergence (Tardy 2000; Oke 1987). Thus, in this study, the idea that the UHI environment may act as a lake-enhanced snowfall environment, and possibly enhance urban snowfall downwind of the city center, is presented.

Urban temperatures have been shown to exceed rural temperatures in all seasons, including winter (Gallo and Owen 1999; Seeley and Jensen 2006). The distribution of the urban temperature through the vertical profile of the atmosphere over the UHI is comparable to that over an unfrozen lake (Dixon and Mote 2003; Tardy 2000). When determining the likelihood of a lake-effect snow event to occur, the stability over a lake surface is calculated using differences in vertical temperature gradients of the lake surface and 850 hPa. The most common temperature threshold cited for lake-effect events is a difference of 13°C (Holroyd 1971; Niziol 1987); however, the threshold difference to produce lake-enhanced snow events has been found to be less than 13°C (Evans and Murphy 2008), with an average difference between 8° and 13°C (Evans 1997). On average, the 850-hPa level can be found at around 1.5 km above the mean sea level surface (Vasquez 2002), and it fluctuates near the border between the planetary boundary layer and the free atmosphere. During the day, the influence of a large city may extend up to 0.6–1.5 km above the surface, which is virtually throughout the entire planetary boundary layer (Oke 1987). During both the day and night, the mixing of the urban layer tends to destroy strong temperature gradients (Oke 1987) and the warmer urban temperature often extends to the 850-hPa level. This well-mixed, warm layer could possibly set up the same mixed, warm environment found in lake-enhanced snowfall events (Tardy 2000).

The urban atmosphere can also be more humid during winter days in cold regions (Oke 1987). In such regions, the rural source of water vapor (i.e., evapotranspiration) during the winter is virtually eliminated because the ground may be covered with snow or be frozen and vegetation is dormant. In the city, however, the anthropogenic releases from combustion provide a significant vapor input (Oke 1987). The urban environment during winter thus provides moisture to snow events passing over the UHI, in a process similar to that which fuels lake-enhanced snow events, as moisture is made available over the lake (Tardy 2000).

Normal daytime convection is augmented by both mechanical and thermal convection from the rougher, warmer city (Oke 1987). Buildings are the main roughness elements of a city, and create increased drag and turbulence as air masses move across the city (Oke 1987). Increased drag and turbulence result in a deeper zone of frictional influence within which wind speeds are reduced in comparison with those at the same height in rural areas, and the local inhibition of the airflow causes it to “pile up” (i.e., converge) over the city, relieved by uplift (Oke 1987). At night, the bulk of the planetary boundary layer is stable and this suppresses vertical transfer, but the combination of urban warmth at the surface and increased forced convection is capable of eroding the stability of rural air as it advects over the city (Oke 1987). The low-level convergence created by the UHI is comparable to the convergence found in lake-enhanced snow events. In two of the three lake-enhanced snow events studied by Tardy (2000), a low-level convergence zone was present: frictional convergence due to water–land interactions, focused convergence due to a land-breeze boundary, or convergence due to a surface trough. Convergence-enhanced convection led to enhanced snowfall downwind of the lake (Tardy 2000), in much the same way that urban cities enhance rainfall downwind of the urban center. Urban areas may also enhance snowfall through convergence in a similar fashion. Lake-enhanced snow events have also been more frequently evaluated using radar reflectivity to determine the amount of snowfall throughout the lower atmosphere instead of only at the surface, providing a framework for some of the methods proposed in this study.

2. Methods

a. Event selection

The study was broken into two parts: a composite analysis of snowfall events and a second, more detailed analysis of individual snowfall-only events. Events were selected from the winter months of November–March for the years 1995–2012. For the individual-event analysis, the daily weather occurrence (DYSW) report from the Minneapolis–St. Paul International Airport Automated Surface Observing System (ASOS) station was used to distinguish snow-only days from days with both snow and rain in order to eliminate days with melting. Snow-only events were selected to reduce error resulting from radar “bright banding,” an effect that occurs when the outer coating of snowfall melts, as a result of falling through a warmer layer, and reflects a higher radar-reflectivity-decibel (hereinafter dBZ) value than the snowfall (Rinehart 1997). One event per year was then sought that additionally met the criteria of having a daily snow total greater than 2 in. (5.08 cm) or, for years with missing daily snow total values, having the highest value of daily precipitation. Upper-air soundings from the Chanhassen, Minnesota, National Weather Service rawinsonde site for each of these snow-only days were also analyzed to ensure that the vertical temperature profile was well below the freezing point and, thus, that no melting layers were present to produce bright banding for these events. (Note that, for the 1996 event, the 0000 UTC sounding was not available. The event was selected based on the fact that the maximum surface temperature never reached above freezing and the fact that the two 1200 UTC atmospheric soundings before and after the event had temperature profiles well below freezing.)

For the composite analysis, days with a mixture of snow, ice, and liquid precipitation were considered in order to maximize the number of snow events, and thus soundings for the composite events were not considered. However, all composite events met the minimum daily snow total (>5.08 cm).

The ASOS hourly precipitation data report was then used to determine a period of three consecutive hours of most intense snowfall for each event in both the composite and individual analyses, narrowing down the large number of radar volume scans to be used in the radar analysis. For each of the 3-h event periods, level-III, base-reflectivity data from the Chanhassen Next Generation Weather Radar (NEXRAD) radar site (KMPX) were initially examined to validate the presence of precipitation in the radar area for these hours, to ensure no obvious brightband occurrence for the mixed-precipitation events, and to ensure that the radar was both operational and in precipitation mode for the entire event period. Level-III data were also examined to ensure storm motions were generally unidirectional for the entire area and over the entire period in order to designate a “downwind” and “upwind” region for each event. For the composite analysis, the primary direction of snowfall movement was from the southwest to northeast; thus, all composite events selected were required to have a general movement in this direction.

Table 1 lists the 25 events selected for composite analysis, along with the daily precipitation totals and generalized storm motions for each event. Table 2 lists the 13 events selected for the individual event analyses, including event hours and precipitation totals for each event. For the individual event analyses, the years 1998, 2004, and 2009–11 did not have all-snow events that also met all of the previously listed criteria.

Table 1.

List of the 25 event dates used in the composite study along with daily snow and precipitation totals and the generalized storm motion for each event.

Table 1.
Table 2.

List of the 13 event dates and hours used in the individual event analyses along with daily snow, precipitation, and heaviest 3-h precipitation totals.

Table 2.

b. Temperature, wind, and radar data

The Minneapolis–St. Paul urban heat island has been documented year-round, including during the winter months (Winkler et al. 1981; Seeley and Jensen 2006), where it usually results in higher overnight minimum temperatures than those of the surrounding countryside (Seeley and Jensen 2006). For the individual event analyses, seven rural sites, all located approximately 50 km or greater from the city center (Todhunter 1996) and representing all cardinal directions, were used to compile and average daily maximum and minimum temperatures for each event; this dataset was then compared with the urban maximum and minimum temperatures in order to determine the strength of the UHI for each event (Fig. 1). By averaging the maximum and minimum temperatures from all seven rural sites, possible factors influencing temperature such as differing latitudes and observation times were mitigated. Using multiple rural sites also provides a more complete temperature dataset, as some sites have missing temperature data.

Fig. 1.
Fig. 1.

Locations of urban (MSP) and rural (all other) thermometers used in the individual event analyses.

Citation: Journal of Applied Meteorology and Climatology 52, 7; 10.1175/JAMC-D-12-090.1

The upper-air soundings from Chanhassen, used to ensure snow-only events, were also used to determine the vertical temperature gradient for each event in the individual analyses. Soundings were selected based on the relationship between the event hours and the daily standard sounding times (0000 and 1200 UTC). Table 3 lists the hours for each event, the relationship between the event hours and the standard sounding times, and the resulting sounding(s) used to determine the surface–850-hPa temperature difference for each event. The difference between the surface temperature and the 850-hPa temperature was calculated to determine the vertical instability present during each event.

Table 3.

Sounding-selection process based on event hours relative to standard sounding times.

Table 3.

Also, all of the individual-analysis event soundings, except for the missing 1996 event sounding, were averaged using vertical pressure bins to compile an average vertical profile for the event environment. After determining whether dBZs were enhanced downwind for each event, the enhanced event soundings were averaged and the difference between the average sounding of all events and the averaged sounding for only enhanced events was calculated. The 2006 event fell between the two standard sounding times, so the 0000 and 1200 UTC soundings were implemented into each sounding category.

For both the composite and individual event analyses, NEXRAD level-II reflectivity data from KMPX for the lowest tilt angle (0.5°) were downloaded from the National Climatic Data Center (NCDC) for these events. Level-II data have a greater intensity resolution and are less processed than level-III data (Crum et al. 1993) and were thus used to quantify dBZ changes downwind.

Each event had approximately 9–12 radar volume scans per hour, characteristic of a Weather Surveillance Radar-1988 Doppler (WSR-88D) operating in precipitation mode. The lower limit of reflectivity values reported when the radar is in precipitation mode is 5 dBZ (Cain 2012); thus, for all events, the radar data were filtered to include only reflectivity values of 5 dBZ or greater and were further fitted onto a standardized latitude–longitude Network Common Data Form (NetCDF) grid via the NCDC Weather and Climate Toolkit (WCT; available online). Because radar decibels are based on a logarithmic scale (Rinehart 1997; Cain 2012), only maximum dBZ values were isolated and used in the analysis, a similar technique to that used in a previous urban precipitation enhancement study by Bentley et al. (2009).

c. Data analysis

For both the individual and composite analyses, the urban boundary was defined using Minneapolis–St. Paul's metropolitan statistical area as determined by the 2000 U.S. Census Tiger Dataset. A circular buffer with a 169-km (105 mi) radius, along with an oval buffer with a 64-km (40 mi) radius, both extending from the city center, were created around the urban center, and a bisecting line was drawn through this buffer perpendicular to the primary storm motion for the composite and each 3-h event. The smaller, oval buffer was created in an attempt to detect small-scale spatial patterns of higher dBZ values transecting the city that could be overshadowed in the total grid summation. The radii lengths selected were simply products of these established city buffers. The total dBZ values for each city region (i.e., upwind and downwind) were then summed to determine the change in snowfall dBZ values downwind of the Minneapolis–St. Paul urban center.

To determine whether there is a statistically significant difference (α = 0.05) between upwind and downwind summed dBZ totals, two types of means tests were considered based on distribution assumptions: a paired-sample t test and a Wilcoxon signed-rank test, similar to the radar analysis methods of Mote et al. (2007). A Shapiro–Wilk W test was run on the 169- and 64-km buffers for all events to test for normality, and only those events with no statistical significance were considered to have a normal distribution. Thus, for these events, the t test was applied, while the Wilcoxon signed-rank test was applied to all others.

3. Results and discussion

a. Composite analysis

After bisecting the 169- and 64-km city buffers with a line perpendicular to the southwestern radar motion and summing all 25 events (Fig. 2a), a decrease in summed maximum reflectivity totals was found downwind of the urban center. A Shapiro–Wilk W test revealed that the composite dataset does not display a normal distribution, and the Wilcoxon signed-rank test showed that summed reflectivity values were significantly decreased from upwind to downwind of the city center. Table 4 shows the results of these tests along with upwind and downwind summed dBZ totals. Although a significant decrease in snowfall (rather than an increase) is seen in the composite analysis, an interesting spatial pattern of reflectivity values can be found downwind of the urban area (Fig. 2a). For this reason, a composite of the greatest maximum dBZ value found for each grid cell and all 25 events was also created (Fig. 2b). This composite of greatest maximum values was analyzed with the same statistical techniques as the summed composite (Table 5), and a statistically significant increase in maximum dBZ downwind of the city center was found in two of the three directional analyses. Although the highest dBZs can be seen to the northwest and southeast of the city in this composite, a region of slightly higher maximum dBZs can be seen directly downwind of the city, perhaps suggesting that the UHI decreases the downwind area receiving snow but increases intensity over a smaller area.

Fig. 2.
Fig. 2.

(a) Total maximum dBZ sum and (b) total maximum dBZ maximums from composite analyses. The 72- and 64-km-radius city buffers are bisected with lines perpendicular to the 3-h southwest wind direction to denote upwind and downwind urban areas.

Citation: Journal of Applied Meteorology and Climatology 52, 7; 10.1175/JAMC-D-12-090.1

Table 4.

A summary of the summed composite-analysis results. Statistically significant (α = 0.05) p values are set in boldface.

Table 4.
Table 5.

As in Table 4, but for the greatest maximum dBZ composite analysis.

Table 5.

b. Individual event analysis

1) Radar analysis results

Total dBZ values for downwind and upwind portions of the 169-km radius city buffer were found by summing all 3 h of each event. Of the 13 events, only 4–8 December 1995, 13 March 1997, 14 March 2002, and 4 December 2007 were shown to have more dBZ values downwind than upwind of the urban area (Fig. 3), making events with an increase in total dBZ downwind only 31% of the total events.

Fig. 3.
Fig. 3.

The maximum dBZ sum with the 169- and 64-km city buffers for the four events found to have increases in the greatest maximum dBZs downwind: (a) 8 Dec 1995, (b) 13 Mar 1997, (c) 14 Mar 2002, and (c) 4 Dec 2007.

Citation: Journal of Applied Meteorology and Climatology 52, 7; 10.1175/JAMC-D-12-090.1

The Shapiro–Wilk W test showed a statistically significant difference in means–medians for all events except for the 169-km circle buffer 2007 event, the only event with greater dBZ maximum sums downwind than upwind found for this buffer. Thus, all other 169-km circle buffer events were found to be significantly decreased downwind (Table 6).

Table 6.

Summed downwind − upwind maximum dBZ differences for each event as well as the t-test and Wilcoxon signed-rank test results with p values denoted for both the 169-km circular and 64-km oval buffers. Positive, italicized dBZ differences indicate greater dBZ sums downwind than upwind, and statistically significant (α = 0.05) p values are set in boldface.

Table 6.

For the 64-km oval city buffer, four events were found to have greater dBZ sums downwind than upwind. A significant difference in upwind and downwind dBZ totals was found for two of the five events with increased dBZs downwind (1997 and 2007), while the other two events with increased downwind dBZ values (1995 and 2002) showed no significant difference in dBZ change downwind. The remaining eight events showed a significant decrease in maximum dBZ values downwind.

2) Urban–rural temperature difference

Table 7 shows the averaged maximum and minimum rural temperatures for all seven rural sites and the urban–rural temperature difference for each event. Note that for the 2012 event, the Santiago and Ellsworth stations were replaced with nearby Milanca and Baldwin station data, respectively, due to missing original station data. When comparing the tests of significance on the 64-km buffer with the urban–rural temperature difference for each event, several interesting results are found. The 2007 event, found to be significantly enhanced for both the 169- and 64-km buffers, had an urban temperature 7.78°C greater than the rural minimum temperature, the highest urban–rural temperature difference found in the individual event analysis. For the 1995 event, found to be significantly enhanced for the 64-km buffer, the urban maximum temperature was found to be slightly higher than the rural maximum temperature, with a higher rural minimum temperature. For the 1995 and 2002 events, found to have an insignificant increase in summed maximum dBZs downwind, the urban temperature is up to 1.11°C greater than the rural maximum temperature; however, for both of these events, the minimum rural temperature was found to be up to 3°C greater than the urban temperature, limiting urban heat island development. It is important to note that the winter UHI could be sensitive to cloud cover and other local variables, and approaching clouds that reach the urban area could lead to cooler urban temperatures and a weakened UHI.

Table 7.

Daily maximum and minimum temperatures (temp) averaged for all seven rural locations, the MSP daily maximum and minimum temperatures (representing the urban location), and the urban − rural temperature differences for daily maximum and minimum temperatures. All temperatures are in degrees Celsius. Boldface values are from enhanced events for the 64-km-radius buffer.

Table 7.

The remaining events that were significantly decreased downwind of the urban center had a variety of urban–rural temperature differences, ranging from −3.89°C (2008) to 3.89°C (2001). The previously discussed daily urban–rural temperature difference considers only the extreme temperatures for each day, and even with hourly temperature measurements, the exact timing of the maximum urban–rural temperature gradient and the resulting forcing mechanism is difficult to quantify for each event and was thus not considered in this study; however, in future studies, this temperature difference should be better quantified to allow for a more detailed analysis of the UHI.

3) Vertical temperature results

To determine the thermal instability that may have enhanced snowfall for each event, the vertical temperature gradient was calculated. Table 8 shows the surface–850-hPa temperature difference calculated from each sounding. All of the enhanced events are shown to have a vertical temperature gradient ranging from 1.8° to 4.4°C. The events with significant decreases in dBZs downwind exhibit vertical temperature gradients from −3.6° to 7.3°C. None of the events exhibited the 8°–13°C surface–850-hPa temperature difference necessary to produce lake-enhanced snowfall, but the event exhibiting a gradient closest to this threshold was the 2001 event, an event having a significant decrease in dBZs downwind.

Table 8.

Calculated differences between surface and 850-hPa temperatures for each event in order to determine vertical instability. Enhanced events are set in boldface.

Table 8.

Averaged event soundings were made based on the event hour proximity to the standard sounding. Differences between the total and enhanced soundings for both the 0000 and 1200 UTC categories were then graphed, with each showing a much steeper low-level lapse rate for enhanced events (Figs. 4 and 5). A layer of up to 4°C colder air was found in the averaged enhanced soundings from the surface to 850 hPa, along with a very weak increase in surface temperature, which together yield greater lapse rates during enhanced events. The 1200 UTC difference sounding also reveals a 2°C cooler surface with a peak of 2°C warmer air around 950 hPa, a profile that could indicate slightly increased lapse rates near the surface to promote a little extra instability found in the enhanced event. A cooling trend can also be seen from approximately 950 to 700 hPa, with a very moist atmospheric profile from the surface to midlevels. Both the 1200 and 0000 UTC average soundings suffer from small sample sizes, so conclusions based on these diagrams are tenuous.

Fig. 4.
Fig. 4.

Difference profiles (enhanced − all-event averages) for the 0000 UTC category. Temperature is the darker, black line, and dewpoint is the lighter, gray line.

Citation: Journal of Applied Meteorology and Climatology 52, 7; 10.1175/JAMC-D-12-090.1

Fig. 5.
Fig. 5.

As in Fig. 4, but for the 1200 UTC category.

Citation: Journal of Applied Meteorology and Climatology 52, 7; 10.1175/JAMC-D-12-090.1

4. Summary and conclusions

Twenty-five snowfall events, including ice and liquid precipitation, for the composite study and 13 snow-only events for the individual case analyses were selected from 1995 to 2012 for Minneapolis–St. Paul, Minnesota. Each event displayed uniform storm motions across the area, with all composite cases having a generalized southwesterly flow, for the three heaviest snowfall hours recorded. Level-II radar data maxima were composited by both summation and by greatest grid maximum for all composite events. For the individual event analyses, radar data maxima were summed through all three event hours. The summed composite showed a significant decrease in summed radar reflectivity values downwind. However, the greatest maximum grid analysis showed a significant increase downwind of maximum dBZ values for two of the three buffers used, and out of the 13 events in the individual analysis, 4 were found to have larger maximum dBZ totals downwind when evaluated with the 64-km city buffer. Two of these events (1997 and 2007) were found to be significantly enhanced downwind, and the other two events (1995 and 2002) showed no significant difference in dBZ change downwind. All other events showed a significant decrease in maximum dBZ values downwind of the city.

While the majority of these results suggest a significant decrease in downwind dBZ values, the detection of some enhanced events in this study provides enough evidence to call into question the assumption stated in the hypothesis that the urban heat island simply decreases all snowfall events. The variability in UHI and the low-level instability suggest that local effects (cloud cover, preexisting snow cover, etc.) may be responsible for modifying the winter UHI and the effect of the urban area upon snowfall distribution and intensity.

There are some limitations with this research, and efforts have been made to minimize some of the potential issues in this project, including bright banding and the use of level-II data instead of level-III data. Radar beam height is one issue not accounted for in this research. As the beam of energy travels away from the radar, it spreads vertically and horizontally so that the lower-resolution radar grid cells at farther distances are also at higher altitudes than those higher-resolution grid cells closer to the radar site (Cain 2012).

REFERENCES

  • Baker-Blocker, A., 1982: Winter weather and cardiovascular mortality in Minneapolis–St. Paul. Amer. J. Public Health, 72, 261265.

  • Balling, R. C., Jr., , and S. W. Brazel, 1987: Diurnal variations in Arizona monsoon precipitation frequencies. Mon. Wea. Rev., 115, 342346.

    • Search Google Scholar
    • Export Citation
  • Bentley, M. L., , W. S. Ashley, , and J. A. Stallins, 2009: Climatological radar delineation of urban convection for Atlanta, Georgia. Int. J. Climatol., 30, 15891594.

    • Search Google Scholar
    • Export Citation
  • Bornstein, R. D., 1968: Observations of the urban heat island effect in New York City. J. Appl. Meteor., 7, 575582.

  • Bornstein, R. D., , and M. LeRoy, 1990: Urban barrier effects on convective and frontal thunderstorms. Extended Abstracts, Conf. on Mesoscale Processes, Boulder, CO, Amer. Meteor. Soc., 25–29.

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

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

    • Search Google Scholar
    • Export Citation
  • Braham, R. R., Jr., , R. G. Semonin, , A. H. Auer, , S. A. Changnon Jr., , and J. M. Hales, 1981: Summary of urban effects on clouds and rain. METROMEX: A Review and Summary, Meteor. Monogr., No. 40, Amer. Meteor. Soc., 141–152.

  • Burian, S. J., , and J. M. Shepherd, 2005: Effect of urbanization on the diurnal rainfall pattern in Houston. Hydrol. Processes, 19, 10891103.

    • Search Google Scholar
    • Export Citation
  • Cain, D. R., cited 2012: Doppler radar frequently asked questions. National Weather Service Southern Region Headquarters. [Available online at http://www.srh.noaa.gov/srh/sod/radar/radinfo/radinfo.html.]

  • Changnon, S. A., Jr., 1968: Precipitation climatology of Lake Michigan Basin. Illinois State Water Survey Bull. 52, Urbana, IL, 46 pp.

  • Changnon, S. A., Jr., 1992: Inadvertent weather modification in urban areas: Lessons for global climate change. Bull. Amer. Meteor. Soc., 73, 619627.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., Jr., 2004: Urban effects on winter snowfall at Chicago and St. Louis. Bull. Ill. Geogr. Soc., 46, 312.

  • Changnon, S. A., Jr., , and N. E. Westcott, 2002: Heavy rainstorms in Chicago: Increasing frequency, altered impacts, and future implications. J. Amer. Water Resour. Assoc., 38, 14671475.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., Jr., , F. A. Huff, , P. T. Schickedanz, , and J. L. Vogel, 1977: Weather anomalies and impacts. Vol. I, Summary of METROMEX, Illinois State Water Survey Bull. 62, Urbana, IL, 260 pp.

  • Changnon, S. A., Jr., , R. T. Shealy, , and R. W. Scott, 1991: Precipitation changes in fall, winter, and spring caused by St. Louis. J. Appl. Meteor., 30, 126134.

    • Search Google Scholar
    • Export Citation
  • Crum, T. D., , R. L. Alberty, , and D. W. Burgess, 1993: Recording, archiving, and using WSR-88D data. Bull. Amer. Meteor. Soc., 74, 645653.

    • Search Google Scholar
    • Export Citation
  • Diem, J. E., , and D. P. Brown, 2003: Anthropogenic impacts on summer precipitation in central Arizona, U.S.A. Prof. Geogr., 55, 343355.

    • Search Google Scholar
    • Export Citation
  • Dixon, P. G., , and T. L. Mote, 2003: Patterns and causes of Atlanta's urban heat island–initiated precipitation. J. Appl. Meteor., 42, 12731284.

    • Search Google Scholar
    • Export Citation
  • Evans, M. S., 1997: A study on the relationship between synoptic-scale model forecasts of vertical motion and snow amounts across southwest Lower Michigan in lake effect snow environments. National Weather Service Central Region Applied Res. Paper 17-11, 17 pp.

  • Evans, M. S., , and R. Murphy, 2008: A proposed methodology for reconciling high-resolution numerical modeling guidance with pattern recognition to predict lake-effect snow. Electron. J. Oper. Meteor., 9 (2). [Available online at http://www.nwas.org/ej/2008/2008.php.]

    • Search Google Scholar
    • Export Citation
  • Gallo, K. P., , and T. W. Owen, 1999: Satellite-based adjustments for the urban heat island temperature bias. J. Appl. Meteor., 38, 806813.

    • Search Google Scholar
    • Export Citation
  • Grillo, J. N., , and J. Spar, 1971: Rain–snow mesoclimatology of the New York metropolitan area. J. Appl. Meteor., 10, 5661.

  • Holroyd, E. W., 1971: Lake-effect cloud bands as seen from weather satellites. J. Atmos. Sci., 28, 11651170.

  • Huff, F. A., , and S. A. Changnon, 1972: Climatological assessment of urban effects on precipitation at St. Louis. J. Appl. Meteor., 11, 823842.

    • Search Google Scholar
    • Export Citation
  • Huff, F. A., , and J. L. Vogel, 1978: Urban, topographic and diurnal effects on rainfall in the St. Louis region. J. Appl. Meteor., 17, 565577.

    • Search Google Scholar
    • Export Citation
  • Huff, T. E., 1986: Urban hydrometeorology review. Bull. Amer. Meteor. Soc., 67, 703711.

  • Jauregui, E., , and E. Romales, 1996: Urban effects on convective precipitation in Mexico City. Atmos. Environ., 30, 33833389.

  • Jones, P. A., , and J. E. Jiusto, 1980: Some local climate trends in four cities of New York State. J. Appl. Meteor., 19, 135141.

  • Kidder, S. Q., , and H.-T. Wu, 1987: A multispectral study of the St. Louis area under snow-covered conditions using NOAA-7 AVHRR data. Remote Sens. Environ., 22, 159172.

    • Search Google Scholar
    • Export Citation
  • Landsberg, H. E., 1981: The Urban Climate. Academic Press, 275 pp.

  • Malevich, S. B., , and K. Klink, 2011: Relationships between snow and the wintertime Minneapolis urban heat island. J. Appl. Meteor. Climatol., 50, 18841894.

    • Search Google Scholar
    • Export Citation
  • Mölders, N., , and M. A. Olson, 2004: Impact of urban effects on precipitation in high latitudes. J. Hydrometeor., 5, 409429.

  • Mote, T. L., , M. C. Lacke, , and J. M. Shepherd, 2007: Radar signatures of the urban effect on precipitation distribution: A case study for Atlanta, Georgia. Geophys. Res. Lett., 34, L20710, doi:10.1029/2007GL031903.

    • Search Google Scholar
    • Export Citation
  • Niziol, T. A., 1987: Operational forecasting of lake effect snowfall in western and central New York. Wea. Forecasting, 2, 310321.

  • Oke, T. R., 1987: Boundary Layer Climates. 2nd ed. Routledge, 435 pp.

  • Rinehart, R. E., 1997: Radar for Meteorologists. 3rd ed. Rinehart, 428 pp.

  • Rosenfeld, D., 1999: TRMM observed first direct evidence of smoke from forest fires inhibiting rainfall. Geophys. Res. Lett., 26, 31053108.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., 2000: Suppression of rain and snow by urban and industrial air pollution. Science, 287, 17931796.

  • Seeley, M. W., , and B. Jensen, 2006: Minnesota Weather Almanac. Minnesota Historical Society Press, 294 pp.

  • Shepherd, J. M., 2005: A review of current investigations of urban-induced rainfall and recommendations for the future. Earth Interact., 9. [Available online at http://EarthInteractions.org.]

    • Search Google Scholar
    • Export Citation
  • Shepherd, J. M., , and S. J. Burian, 2003: Detection of urban-induced rainfall anomalies in a major coastal city. Earth Interact., 7. [Available online at http://EarthInteractions.org.]

    • Search Google Scholar
    • Export Citation
  • Shepherd, J. M., , H. Pierce, , and A. J. Negri, 2002: Rainfall modification by major urban areas: Observations from spaceborne rain radar on the TRMM satellite. J. Appl. Meteor., 41, 689701.

    • Search Google Scholar
    • Export Citation
  • Tardy, A., 2000: Lake effect and lake enhanced snow in the Champlain Valley of Vermont. National Weather Service Eastern Region Tech. Attach. 2000-05, 27 pp.

  • Todhunter, P. E., 1996: Environmental indices for the Twin Cities metropolitan area (Minnesota, USA) urban heat island—1989. Climate Res., 06, 5969.

    • Search Google Scholar
    • Export Citation
  • Van den Berg, W., 2008: Fog turning into snow. Weather, 63, 248248.

  • Vasquez, T., 2002: Weather Forecasting Handbook. 5th ed. Weather Graphics Techologies, 198 pp.

  • Winkler, J. A., , R. H. Skaggs, , and D. G. Baker, 1981: Effect of temperature adjustments on the Minneapolis–St. Paul urban heat island. J. Appl. Meteor., 20, 12951300.

    • Search Google Scholar
    • Export Citation
  • Wood, C. R., , and R. G. Harrison, 2009: Anthropogenic snowfall events in the UK: Examples of urban weather modification? Weather, 64, 277280.

    • Search Google Scholar
    • Export Citation
  • Woollum, C. A., , and N. L. Canfield, 1968: Washington metropolitan area precipitation and temperature patterns. Environmental Science Services Tech. Mem. WBTM 28, Weather Bureau, Eastern Region, Washington, DC, 32 pp.

  • Yuan, F., , K. E. Sawaya, , B. C. Loeffelholz, , and M. E. Bauer, 2005: Land cover classification and change analysis of the Twin Cities (Minnesota) metropolitan area by multitemporal Landsat remote sensing. Remote Sens. Environ., 98, 317328.

    • Search Google Scholar
    • Export Citation
Save