Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Person-place-time analysis of vehicle fatalities caused by flash floods in Texas

Person-place-time analysis of vehicle fatalities caused by flash floods in Texas Geomatics, Natural Hazards and Risk Vol. 3, No. 4, November 2012, 311–323 Person-place-time analysis of vehicle fatalities caused by flash floods in Texas HATIM O. SHARIF*, Md. MOAZZEM HOSSAIN, TERRANCE JACKSON and SAZZAD BIN-SHAFIQUE Department of Civil and Environmental Engineering, University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249, USA (Received 6 July 2011; in final form 15 August 2011) A significant number of crashes on roads are caused by adverse weather conditions. Among the most serious consequences of rainfall and flooding in regard to road safety are the motor vehicle-related flash flood fatalities. These fatalities are of particular concern in Texas. Information on motor vehicle fatalities caused by flash floods was extracted from the National Climatic Data Center Storm Data reports. Review of reports on flash flood fatalities in general, where the death circumstances are provided, reveals that most fatalities are motor vehicle-related (80%). Moreover, data from the reports indicate that from 1959 to 2009 the number of motor vehicle-related flash flood fatalities in Texas exceeds by far the total number of flood fatalities in any other state. Demographic analysis of the flash flood fatalities indicates that, in Texas, all ages are at risk and that males are much more affected than females. Spatial analysis indicates that the highest numbers of fatalities occur in counties having major urban areas. The increase in the frequency and intensity of storms and floods that is projected to result from climate changes and variability, and the rapid urbanization in the region may considerably increase the impacts of weather on road safety in Texas. 1. Introduction Road safety is significantly affected by weather conditions. In fact, weather conditions are reported to be responsible for up to 25% of crashes on the United States highways (Pisano et al. 2008). The effects of weather on road safety include reduced visibility, reduced pavement friction, malfunctioning of traffic control devices, and flood water on the road (e.g. Khan et al. 2008). Rainfall and subsequent flood are reported to have more impact on road safety than other weather-related factors. For instance, Goodwin (2002) found out that rainfall accounted for 80% of the injuries and 69% of the fatalities on US highways that were caused by poor weather conditions. Other studies reported that rainfall increases the number of road crashes by 100% or more (e.g. Bertness 1980, Brodsky and Hakkert 1988, Andrey et al. 2003). An interesting finding of empirical studies is that rainfall leads to a stronger increase in the number of fatal accidents when it comes after a dry spell (e.g. Koetse and Rietveld 2009). Eisenberg and Warner (2005) also found that although *Corresponding author. Email: hatim.sharif@utsa.edu Geomatics, Natural Hazards and Risk ISSN 1947-5705 Print/ISSN 1947-5713 Online ª 2012 Taylor & Francis http://www.tandf.co.uk/journals http://dx.doi.org/10.1080/19475705.2011.615343 312 H.O. Sharif et al. precipitation, including rainfall and snow, significantly increases the number of accidents, it appears to reduce the severity of accidents. Flooding is the most damaging consequence of rainfall. Floods, in general, are naturally occurring events dependent not only on rainfall rates and durations, but also on other factors such as the topography, land use, soil types of the catchment area, and antecedent moisture conditions (Sharif et al. 2006). A flash flood is a short term flood event that occurs less than 6 hours after heavy rainfall has ended, and requires immediate action to protect life and property. (In contrast to a flash flood, a typical river flood usually occurs beyond 6 hours after heavy rainfall has ended.) Flood-related fatalities in the US have been the topic of several studies. The Federal Emergency Management Agency (FEMA) considers flooding ‘America’s Number One Natural Hazard’ (Brown 2005). Moreover, floods are the most common and widespread of all weather-related natural disasters. Between 1983 and 2003, flooding caused an average of 98 deaths and $4.5 billion in property damage per year in the United States (NWS 2005, Downton et al. 2005). Despite flood management efforts in many communities, US flood damages remain high, largely due to increasing population and property development in flood-prone areas (e.g. Changnon 1998, Pielke and Downton 2000, Burby 2001). Although the dominant characteristic of hurricanes is their strong winds, most fatalities and damage are caused by flooding due to high runoff produced by torrential rains and storm surge propelled by these winds. Research for the US East Coast and Gulf area shows that the effect on transport and transport infrastructure is substantial (Koetse and Rietveld 2009). The most serious and irreversible consequence of flooding is the number of flood- related fatalities. French et al. (1983), in one of the first published studies examining flood fatalities in detail, found that most of the flash floods during 1969–1981 occurred during the warm season spanning July–September, with September representing the peak fatality month. They also found over 90% of flash flood fatalities were due to drowning, and of those 42% were vehicle-related. Most flood- related deaths are due to drowning, which can be caused by several actions such as driving into the flood or trying to save a person from a flood. In the US, over half of these casualties involve people driving into flooded roads (figure 1), and then either drowning in their vehicles or escaping only to perish in the open water (Drobot et al. 2007). Ashley and Ashley (2008) reported 4586 flood fatalities in the US between 1959 and 2005. They found that the number of fatalities varied from year to year, with anomalously high years coinciding with either tropical system-produced floods or sudden flash floods, often associated with structural failures of dams or levees. Ruin et al. (2007) used cognitive mapping to identify several factors influencing motorists’ flash flood risk perception. Younger drivers underestimate flood risk, as do drivers of lower income and drivers with no children. They found that urban people underestimate the risk for a car to be swept away by running water and are relatively more threatened by walking and driving in flood conditions than are people living in rural areas. One of the most dangerous flash flood areas in the US is the urban corridor between Dallas and San Antonio in Central and South-central Texas called locally the ‘Flash Flood Alley’, figure 2. Actually, within the conterminous US, the greatest concentration of exceptional flood peaks is at the Balcones Escarpment of central Texas, where maximum US rainfall amounts apparently coincide with basin physiography to produce many of the largest measured US floods (O’Connor and Flash flood vehicle fatalities 313 Figure 1. Road inundation caused by flash floods in San Antonio. Courtesy of the City of San Antonio (used with permission). Available in colour online. Costa 2004). This paper discusses the results of a study of motor vehicle-related flash flood fatalities in Texas for the period between 1959 and 2009. 2. Methods Information on flood fatalities and damage can be obtained from health and safety authorities and other sources. However, the most comprehensive source for flood fatality and injury data in the US is the Storm Data publication, which is maintained by the National Climatic Data Center (NCDC) of the National Oceanic and 314 H.O. Sharif et al. Figure 2. The Flash Flood Alley. Obtained from the Flood Safety Education Project http:// www.floodsafety.com (used with permission). Available in colour online. Atmospheric Administration (NOAA). This monthly publication includes a chronological listing of state-by-state fatalities, injuries and property damage for all weather-related events. Event types include floods, flash floods, hurricanes, tornadoes and other weather-related events. Storm Data includes the location and time of fatality or injury by county for most of the events. NCDC Storm Data also provides detailed description of the weather event associated with fatalities. In addition, Storm Data includes narratives that provide detailed information on casualties, weather records and other anecdotal information for significant weather events. Information included in the Storm Data publication comes from the National Weather Service (NWS) offices and personnel, the media, law enforcement authorities, government agencies, private companies and individuals. The operation manual for Storm Data (Mandt 2002) describes the criteria for inclusion of weather- related fatalities and injuries in the publication. The Storm Data was formerly called Climatological Data in 1950 and recorded only tornados. In 1955 thunderstorm and wind data were added. In 1959 this publication was officially named Storm Data and recording of all unusual weather phenomena data that are included in today’s publication began. The NCDC, the National Environmental Satellite, Data and Information Service and NOAA prepare and distribute Storm Data. In this study, we inventoried and reviewed flood fatality information for Texas included in the NCDC Storm Data publication for the period between 1959 and 2009. A total of 612 publications were reviewed. The main difference between this and other previous studies (e.g. Ashley and Ashley 2008) is that instead of reviewing summary statistics, we examined the original publications for the entire period to extract all the reported descriptions of each fatality. Descriptive statistics derived in this study include the date, time, location and weather conditions under which the flood fatality occurred. Demographic analysis of the flood fatalities was also performed. Spatial analysis was also used to identify locations where flood fatalities may be clustered (e.g. in and around urban centres). Flash flood vehicle fatalities 315 3. Results Floods, especially flash floods, are a serious hazard in Texas. Examination of the 1959–2009 NCDC Storm Data records indicates that Texas is the only state that reported flood-related fatalities every single year during that period. By far, Texas leads the nation in flood fatalities. From 1959 through 2009 there were three times more fatalities in Texas (854) than in the next leading state, Pennsylvania (265). Table 1 lists the five states with the most number of flood fatalities. Flood events are typically classified in Storm Data as floods, flash floods (defined earlier), or floods due to tropical systems. The distribution of fatalities by flood type is shown in table 2. About 68% (584 out of 854) of Texas flood fatalities are caused by flash floods. 3.1 Person Storm Data provides the gender of only 81% (286 out of 351) of motor vehicle- related fatalities caused by flash floods and the ages of only 58% (202 out of 351) of the victims. For cases when the gender of the victim is known, the ratio of males to females is 1.7 to 1 (180 to 106). The age distribution for victims with known age is shown in table 3. The age group with the most fatalities is young adults (39 known victims). Infants, children, teenagers and the elderly (less than 20 years of age and more than 59 years) represent 47% of the victims (94 out of 202). The number of victims who are infants or young children is significant (26). 3.2 Place The distribution of flash flood fatalities in Texas with known activity/location from NCDC Storm Data is shown in table 4. ‘In water’ describes the case when a person intentionally walked into the floodwater. Review of the data reveals that 80% (351 out of 440) of the fatalities with known circumstances are motor vehicle-related. Table 1. The five states with the highest numbers of total flood fatalities, 1959–2009. State Number of fatalities Texas 854 Pennsylvania 265 California 248 South Dakota 244 Virginia 237 Table 2. Flood fatalities classified by flood type. Flood type Fatalities Percentage Flash flood 584 68% Flood 221 26% Flood due to tropical storms 42 5% Coastal/tidal/river flooding 7 1% 316 H.O. Sharif et al. This percentage is higher than reported in other studies. For example, according to the Flood Safety Education Project (2010), most flood drownings are vehicle related, and in Texas, 76% of flood deaths are vehicle related. The circumstances of death (i.e. whether the death was motor vehicle-related or not) are not described in Storm Data for 25% (144 out of 584) of flash flood fatalities. Confirmed motor vehicle fatalities caused by flash floods were reported for all but three years of the record, as shown in figure 3. The 51-year average number of confirmed fatalities is 6.9 per year. The monthly distribution of motor vehicle-related flash flood fatalities and the average of the climatological mean rainfall in the major urban areas with highest flood fatalities are shown in figure 4. The two lines have similar distributions and peaks. The correlation coefficient between the two series is 0.90. Table 3. Number of motor vehicle-related flood fatalities classified by age. The age is not provided in Storm Data for 149 fatalities. Age group Fatalities Percentage 0–9 26 13% 10–19 26 13% 20–29 39 19% 30–39 17 8% 40–49 24 12% 50–59 28 14% 60–69 21 10% 70–79 14 7% 80–89 7 4% Table 4. Number of flash flood fatalities classified by activity/location. The activity/location is not provided in Storm Data for 144 fatalities. Activity/location Fatalities Percentage Boat 8 2% In water 65 15% Mobile home 6 1% Permanent home 10 2% Vehicle 351 80% Figure 3. Motor vehicle-related flash flood fatalities in Texas from 1959 to 2009. Flash flood vehicle fatalities 317 The Texas counties with the highest numbers of vehicle fatalities caused by flash floods are listed in table 5. These ten counties account for more than 50% of the fatalities. Figure 5 shows motor vehicle-related flood fatalities in Texas categorized by county. Eight counties had 11 or more motor vehicle-related flood deaths, 15 counties had 4 to 10 flood deaths, and 67 counties had 1 to 3 deaths. 3.3 Time Storm Data also associated motor vehicle fatalities caused by flash flood with time of occurrence. The time was reported for 306 (out of 351) of the fatalities. It is not clear whether the time reported in Storm Data is the time of the accident or the time of death. As seen in table 6, most of the 306 fatalities (55%) have taken place at night. 4. Discussion A worrying consequence of rainfall road safety is the motor vehicle-related flash flood fatalities. In Texas, the number of motor vehicle-related flash flood fatalities Figure 4. Motor vehicle-related flash flood fatalities in Texas and mean precipitation in the Flash Flood Alley region by month. Table 5. Texas counties with the highest numbers of vehicle fatalities caused by flash floods. County Fatalities Bexar 30 Tarrant 30 Travis 24 Dallas 22 Harris 16 Kerr 16 Edwards 12 Bell 11 Real 9 Burnet 8 318 H.O. Sharif et al. Figure 5. Map of motor vehicle-related flash flood fatalities by Texas county. Available in colour online. Table 6. Number of motor vehicle fatalities caused by flash floods classified by time of occurrence. The time is not provided in Storm Data for 45 fatalities. Time Fatalities Percentage Morning 84 35% Afternoon 57 10% Night 165 55% far exceeds the total number of flood fatalities in any other state between 1959 and 2009. There is an increasing trend in the number of annual fatalities. That is probably due to the increasing population density in Texas, figure 3. Years associated with major floods witnessed the highest numbers of fatalities (e.g. 1979, 1981, 2007). A review of the data indicates that 80% (351 out of 440) of the flash flood fatalities with known circumstances are motor vehicle-related. The number of those fatalities in Texas is expected to be actually higher than 351, as we expect that a significant number among the fatalities whose circumstances are not provided would be motor vehicle-related. More than 68% of floods that result in fatalities in Texas are flash floods, which typically arrive immediately after the storm with little or no warning. Tropical systems, although associated with very intense rainfall and ensuing floods, result in Flash flood vehicle fatalities 319 a very small number of fatalities (less than 5% of the fatalities). This small number may have several causes. First, tropical systems are typically forecasted with longer lead times (many hours or days), and the warnings are thus more widely broadcasted to the general public. Secondly, the longer lead times allow emergency personnel to develop a heightened awareness for greater involvement. Thirdly, tropical storm system floods are less frequent than flash floods in Texas. A review of ages of motor vehicle-related flash flood accidents indicates that all age groups are at risk of dying from floods, including significant numbers of victims of ages below 20 years or above 60 years. Significantly more males die in motor vehicle accidents during a flash flood than females. This agrees with studies that addressed vehicle-related fatalities in general. McKenna et al. (1998), for example, suggested that the presence of a female passenger significantly reduces the likelihood of an accident, and Chen et al. (2000) reported that the presence of a male passenger almost doubles the per capita death rate, regardless of the driver. Motor vehicle-related flash flood fatalities appear to be related to topography and climate. Very few fatalities occurred in counties with dry climate even when the population density is high. El Paso County is one such example. This agrees with the findings of Khan et al. (2008), who found significant spatial correlation between the number of accidents and weather patterns. Fatalities are clustered in counties located along the Balcones Escarpment and those that include major urban centres. The fatality numbers in Harris County are smaller than some counties along the Balcones Escarpment although it is one of the counties with the highest number of flood fatalities in general. The monthly distribution of motor vehicle-related flash flood fatalities closely follows the monthly distribution of rainfall in the flash flood region in Texas Flash Flood Alley. This indicates that the frequency of storms plays an important role in increasing the flash flood hazard. Other studies have found a similar correlation between road accidents and rainfall (e.g. Levine et al. 1995). Most motor vehicle-related flash flood deaths in Texas appear to have occurred at night (table 6). This increase in fatalities at night may be attributed to people not being able to see or estimate the depth and speed of the flood water in inadequate lighting. Some earlier studies reported that 75% of flash-flood deaths occurred during the hours of twilight and darkness (Mooney 1983). The general behavioural reaction to severe weather includes trip cancelling, trip shortening, route changing and more use of public transportation (De Dios Ortuzar and Willumsen 2001). The behavioural response of a driver to poor weather conditions on the road, however, is influenced by several factors. Kates (1971) reported that factors that may affect risk perception include the nature and features of the natural hazard (its magnitude, duration, frequency and temporal spacing), and the recency, frequency and intensity of personal experience with past events of similar nature. The dense road networks and numerous low water crossings throughout Texas may be contributing to the higher recurrence rates of floods that pose a danger to vehicles (figure 6). Several ephemeral streams in Texas, especially along the Balcones Escarpment, have steep slopes. The highly intermittent flow of these streams reduces the overall financial efficiency of such large, expensive structures at road-stream crossings. Therefore, many crossings are constructed as armoured sag vertical curves, often with a small corrugated metal culvert pipe as a relief structure to prevent long-term ponding of water upstream of the roadway embankment. In flood 320 H.O. Sharif et al. Figure 6. Map of some of the low water crossings in San Antonio. Courtesy of the City of San Antonio (used with permission). Available in colour online. conditions water has to flow over the road. These low-water crossings pose immediate danger to vehicles that try to cross during flooding conditions. It is at these crossings where most motor vehicle-related flash flood fatalities happen. What many drivers do not know is that, according to the Federal Emergency Administration (FEMA 2010), six inches (0.15 m) of water will reach the bottom of most passenger cars causing loss of control and possible stalling, and two feet (0.61 m) of rushing water can carry away most vehicles, including sport utility vehicles and pickups. Excerpts from the narratives in Storm Data that describe some the circumstances that lead to fatalities are shown in table 7. Driving into flood water appears to be the cause of the majority of motor vehicle-related flash flood fatalities in Texas. This behaviour is mostly intentional and controllable. Solutions such as installation of traffic barricades, alarm signs, and water depth gauges at hazardous locations may not be enough. There are many practical limitations to these installations. Depth gauges, for example, do not necessarily convey the water depth and the implication of that depth to the lay person. Depth gauges may be improved by providing colours to indicate hazardous water depths. Barricades, on the other hand, are often installed or uninstalled too late. The driver may see barricades and choose to drive around or think that his/her truck or sports utility vehicle can cross the low lying water. Some studies confirm that passengers play a major role in controlling the behaviour of the driver (e.g. Chen et al. 2000). Modifying this behaviour will involve educating the public about flood risks and the seriousness of flood warnings. Siegrist and Gutscher (2008) found that people could envisage what physical risk a flood poses, but they could not envisage the negative affect that could be associated with such an event. In order to boost mitigation motivation, they suggested that risk communication must also help people to envisage the negative emotional consequences of a flood. Flash flood vehicle fatalities 321 Table 7. Examples of Storm Data descriptions of the circumstances that led to vehicle fatalities caused by flash floods. Date County Storm Data narrative 8/16/2007 Bexar The area of extremely heavy rainfall associated with the remains of Tropical Storm Erin continued to spread northwestward across Bexar County, with a general 4 to 5 inches of rain over the county. Near midnight a young woman was driving with three friends and a baby near North Star Mall when she accidentally drove her sport utility vehicle into deeper water where it was slammed against a bridge and then was swept into a drainage ditch. The three other adults in the vehicle were able to get the baby out of the vehicle through the window and escape. But when the three looked back for the driver, she was gone. Her body was found later by emergency responders when the water receded. 9/05/2007 Bexar Up to 2 inches of rain fell over the northwestern part of the county. At around 1:00 am a man and a woman were washed away in a four door sedan as they attempted to cross the Tower View Bridge on Scenic Loop Road, which was in flood due to Helotes Creek. The man was later rescued with minor bruises to his arms and torso. The woman’s body was found less than a mile downstream at 3:19 pm. 3/30/2008 San Augustine Widespread flooding reported across the northern portion of the county. Flooding was severe on County Road 21 in town and at the circle and along County Road 103. Three foot of water was also reported over Highway 3079. One fatality was reported when two people driving on State Highway 21 ran into a ditch and became caught in swift moving water. One occupant in the vehicle was swept away in the vehicle. 4/12/2009 Harris Heavy rainfall persisted across the county as several thunderstorms produced rainfall totals of 8 to 10 inches across the county. Numerous roads were closed in areas of the county, including Jersey Village, Houston Heights, La Porte, Pasadena, Webster, and near the Houston Hobby airport. The highest one hour rainfall total was 6.90 inches recorded at Bay Area Blvd and Clear Creek. This exceeds the highest one hour rainfall rate measured during Tropical Storms. In the northern part of the county five children were killed after the car they were riding in inadvertently drove into a drainage ditch that was filled with nine feet of water from the heavy rainfall. Two adults and one child were able to escape the floodwaters, but the five younger children, all under 7 years old, were not able to escape. 5/03/2009 Upshur Numerous counties and parishes were flooded with rainfall amounts in excess of 6 inches in a 12 hour period common. A 78 year old man drowned when flood waters swept away his automobile on a rural northwestern Upshur County Road. The man was found in his back seat on Fox Road. 10/21/2009 Burnet A woman was killed while trying to cross a low water crossing. Karen Eichelberger, age 47, was driving home after working in Austin this Wednesday night. She tried to cross the flooded low water crossing on Deer Springs Drive in her neighbourhood when her 1994 Jeep was apparently forced off the road and into the creek. Her body was found one half mile downstream from where the Jeep was located. 322 H.O. Sharif et al. The consequences of rainfall and flooding on transport safety are well studied. However, motor vehicle-related flood fatalities have received little attention in the transportation literature. There is evidence that rainfall intensity and flooding frequency are increasing in many regions around the globe. In addition, studies predict that the intensity and frequency of storms and hurricanes in Texas will rise due to climate change and variability (IPCC 2007). Although rainfall and flooding are believed to reduce the severity of traffic accidents (e.g. Koetse and Rietveld 2009), they cause significant motor vehicle fatalities in Texas. Attention to this problem may help reduce these fatalities. In addition, physical and hydraulic field investigation is needed to better understand the factors that lead to the high frequency of these events in Texas and the impact on climate change of these factors. That the Flash Flood Alley region of Texas is experiencing unprecedented urbanization and population growth, leading to an increase in the impervious areas and reduction in infiltration, suggests that the flood fatality rate will become even higher if the issue is not given due attention. References ANDREY, J., MILLS, C. B., LEAHY, M. and SUGGETT, J., 2003, Weather as a chronic hazard for road transportation in Canadian cities. Natural Hazards, 28, pp. 319–343. ASHLEY, S.T. and ASHLEY, W.S., 2008, Flood fatalities in the United States. Journal of Applied Meteorology and Climatology, 47, pp. 805–818. BERTNESS, J., 1980, Rain-related impact on selected transportation activities and utility services in the Chicago area. Journal of Applied Meteorology, 19, pp. 545–556. BRODSKY, H. and HAKKERT, A.S., 1988, Risk of a road accident in rainy weather. Accident Analysis and Prevention, 10, pp. 161–176. BROWN, M.D., 2005, Being flood smart means being prepared. Available online at http:// www.fema.gov/media/archives/2005/050613.shtm (accessed 15 December 2010). BURBY, R.J., 2001, Flood insurance and floodplain management: the U.S. experience. Environmental Hazards, 3, pp. 111–122. CHANGNON, S.A., 1998, Comments on ‘Secular trends of precipitation amount, frequency, and intensity in the United States.’ Bulletin of the American Meteorological Society, 79, pp. 2550–2552. CHEN, L., BAKER, S.P., BRAVER, E.R. and LI, G., 2000, Carrying passengers as a risk factor for crashes fatal to 16 and 17 year old drivers. Journal of the American Medical Association, 283, pp. 1578–1582. DE DIOS ORTUZAR, J. and WILLUMSEN, L.G., 2001, Modelling Transport, 3rd edn (Chichester: John Wiley & Sons). DOWNTOWN, M.W., MILLER, J.Z.B. and PIELKE JR, R.A., 2005, Reanalysis of U.S. National Weather Service flood loss database. Natural Hazards Review, 6, pp. 13–22. DROBOT, S.D., BENIGHT, C. and GRUNTFEST, E.C., 2007, Risk factors associated with driving through flooded roads. Environmental Hazards, 7, pp. 227–234. EISENBERG, D. and WARNER, K., 2005, The effects of snowfalls on motor vehicle collisions, injuries, and fatalities. American Journal of Public Health, 95, pp. 120–124. FEMA (FEDERAL EMERGENCY ADMINISTRATION), 2010, Are You Ready Guide. Available online at: http://www.fema.gov/areyouready/preface.shtm (accessed 15 August 2010). FLOOD SAFETY EDUCATION PROJECT, 2010, Available online at: http://www.floodsafety.com (accessed 15 December 2010). FRENCH, J. G., ING, R., VON ALLMEN, S. and WOOD, R., 1983, Mortality from flash floods: a review of the National Weather Service reports, 1969–1981. Public Health Report, 98, pp. 584–588. Flash flood vehicle fatalities 323 GOODWIN, L., 2002, Analysis of weather-related crashes on U.S. Highways, Mitretek Systems, prepared for the FHWA Road Weather Management Program. Available online at: www.ops.fhwa.dot.gov/weather/resources/publications/fhwa/crashanalysis2001.pdf (accessed 15 December 2010). IPCC (INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE), 2007, Climate Change 2007: The physical science basis: summary for policy makers. IPCC Secretariat, Geneva. KATES, R.W., 1971, Natural hazard in human ecological perspective: hypotheses and models. Economic Geography, 47, pp. 438–451. KHAN, G., QIN, X. and NOYCE, D., 2008, Spatial analysis of weather crash patterns. ASCE Journal of Transportation Engineering, 134, pp. 191–202. KOETSE, M.J. and RIETVELD, P., 2009, The impact of climate change and weather on transport: an overview of empirical findings. Transportation Research Part D, 14, pp. 202–221. LEVINE, N., KIM, K.E. and NITZ, L. H., 1995, Daily fluctuations in Honolulu motor vehicle accidents. Accident Analysis and Prevention, 27, pp. 785–796. MANDT, G.A., 2002, Storm Data preparation. National Weather Service, National Oceanic and Atmospheric Administration, Department of Commerce, National Weather Service Instruction 10-1605, NWSFD 10-16, Silver Spring, Maryland, 79 pp. MCKENNA, A.P., WAYLEN, A.E. and BURKES, M.E., 1998, Male and female drivers: how different are they? The University of Reading, Foundation for Road Safety Research, Reading. MOONEY, L.E., 1983, Applications and implications of fatality statistics to the flash flood problem. Preprints, Fifth American Meteorological Society Conference on Hydrometeorology, Tulsa, OK (Boston, MA: American Meteorological Society), pp. 127–129. NWS (NATIONAL WEATHER SERVICE), 2005, Flood fatalities. Hydrologic Information Center. Available online at: www.nws.noaa.gov/oh/hic/flood_stats/recent_individual_deaths. shtml (accessed 15 July 2010). O’CONNOR, J.E. and COSTA, J.E., 2004, The world’s largest floods, past and present: their causes and magnitudes. US Geological Survey Circular, no. 1254, 13 pp. PIELKE Jr, R.A. and DOWNTON, M.W., 2000, Precipitation and damaging floods: trends in the United States, 1932–97. Journal of Climate, 13, pp. 3625–3637. PISANO, P.A., GOODWIN, L.C. and ROSSETTI, M.A., 2008, U.S. Highway crashes in adverse road weather conditions. Proceedings of the 88th Annual American Meteorological Society Meeting, 20–24 January, New Orleans, LA (Boston, MA: American Meteorological Society), Paper 8.1, CD-ROM. RUIN, I., GAILLARD, J. and LUTOFF, C., 2007, How to get there? Assessing motorists’ flash flood risk perception on daily itineraries. Environmental Hazards, 7, pp. 235–244. SHARIF, H. O., YATES, D., ROBERTS, R. and MUTOFF, C., 2006, The use of an automated now- casting system to forecast flash floods in an urban watershed. Journal of Hydrometeorology, 7, pp. 190–202. SIEGRIST, M. and GUTSCHER, H., 2008, Natural hazards and motivation for mitigation behavior: people cannot predict the affect evoked by a severe flood. Risk Analysis, 28, pp. 771–778. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png "Geomatics, Natural Hazards and Risk" Taylor & Francis

Person-place-time analysis of vehicle fatalities caused by flash floods in Texas

Download PDF
13 pages

Loading next page...
 
/lp/taylor-francis/person-place-time-analysis-of-vehicle-fatalities-caused-by-flash-W0AHvWS6Jk

References (32)

Publisher
Taylor & Francis
Copyright
Copyright Taylor & Francis Group, LLC
ISSN
1947-5713
eISSN
1947-5705
DOI
10.1080/19475705.2011.615343
Publisher site
See Article on Publisher Site

Abstract

Geomatics, Natural Hazards and Risk Vol. 3, No. 4, November 2012, 311–323 Person-place-time analysis of vehicle fatalities caused by flash floods in Texas HATIM O. SHARIF*, Md. MOAZZEM HOSSAIN, TERRANCE JACKSON and SAZZAD BIN-SHAFIQUE Department of Civil and Environmental Engineering, University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249, USA (Received 6 July 2011; in final form 15 August 2011) A significant number of crashes on roads are caused by adverse weather conditions. Among the most serious consequences of rainfall and flooding in regard to road safety are the motor vehicle-related flash flood fatalities. These fatalities are of particular concern in Texas. Information on motor vehicle fatalities caused by flash floods was extracted from the National Climatic Data Center Storm Data reports. Review of reports on flash flood fatalities in general, where the death circumstances are provided, reveals that most fatalities are motor vehicle-related (80%). Moreover, data from the reports indicate that from 1959 to 2009 the number of motor vehicle-related flash flood fatalities in Texas exceeds by far the total number of flood fatalities in any other state. Demographic analysis of the flash flood fatalities indicates that, in Texas, all ages are at risk and that males are much more affected than females. Spatial analysis indicates that the highest numbers of fatalities occur in counties having major urban areas. The increase in the frequency and intensity of storms and floods that is projected to result from climate changes and variability, and the rapid urbanization in the region may considerably increase the impacts of weather on road safety in Texas. 1. Introduction Road safety is significantly affected by weather conditions. In fact, weather conditions are reported to be responsible for up to 25% of crashes on the United States highways (Pisano et al. 2008). The effects of weather on road safety include reduced visibility, reduced pavement friction, malfunctioning of traffic control devices, and flood water on the road (e.g. Khan et al. 2008). Rainfall and subsequent flood are reported to have more impact on road safety than other weather-related factors. For instance, Goodwin (2002) found out that rainfall accounted for 80% of the injuries and 69% of the fatalities on US highways that were caused by poor weather conditions. Other studies reported that rainfall increases the number of road crashes by 100% or more (e.g. Bertness 1980, Brodsky and Hakkert 1988, Andrey et al. 2003). An interesting finding of empirical studies is that rainfall leads to a stronger increase in the number of fatal accidents when it comes after a dry spell (e.g. Koetse and Rietveld 2009). Eisenberg and Warner (2005) also found that although *Corresponding author. Email: hatim.sharif@utsa.edu Geomatics, Natural Hazards and Risk ISSN 1947-5705 Print/ISSN 1947-5713 Online ª 2012 Taylor & Francis http://www.tandf.co.uk/journals http://dx.doi.org/10.1080/19475705.2011.615343 312 H.O. Sharif et al. precipitation, including rainfall and snow, significantly increases the number of accidents, it appears to reduce the severity of accidents. Flooding is the most damaging consequence of rainfall. Floods, in general, are naturally occurring events dependent not only on rainfall rates and durations, but also on other factors such as the topography, land use, soil types of the catchment area, and antecedent moisture conditions (Sharif et al. 2006). A flash flood is a short term flood event that occurs less than 6 hours after heavy rainfall has ended, and requires immediate action to protect life and property. (In contrast to a flash flood, a typical river flood usually occurs beyond 6 hours after heavy rainfall has ended.) Flood-related fatalities in the US have been the topic of several studies. The Federal Emergency Management Agency (FEMA) considers flooding ‘America’s Number One Natural Hazard’ (Brown 2005). Moreover, floods are the most common and widespread of all weather-related natural disasters. Between 1983 and 2003, flooding caused an average of 98 deaths and $4.5 billion in property damage per year in the United States (NWS 2005, Downton et al. 2005). Despite flood management efforts in many communities, US flood damages remain high, largely due to increasing population and property development in flood-prone areas (e.g. Changnon 1998, Pielke and Downton 2000, Burby 2001). Although the dominant characteristic of hurricanes is their strong winds, most fatalities and damage are caused by flooding due to high runoff produced by torrential rains and storm surge propelled by these winds. Research for the US East Coast and Gulf area shows that the effect on transport and transport infrastructure is substantial (Koetse and Rietveld 2009). The most serious and irreversible consequence of flooding is the number of flood- related fatalities. French et al. (1983), in one of the first published studies examining flood fatalities in detail, found that most of the flash floods during 1969–1981 occurred during the warm season spanning July–September, with September representing the peak fatality month. They also found over 90% of flash flood fatalities were due to drowning, and of those 42% were vehicle-related. Most flood- related deaths are due to drowning, which can be caused by several actions such as driving into the flood or trying to save a person from a flood. In the US, over half of these casualties involve people driving into flooded roads (figure 1), and then either drowning in their vehicles or escaping only to perish in the open water (Drobot et al. 2007). Ashley and Ashley (2008) reported 4586 flood fatalities in the US between 1959 and 2005. They found that the number of fatalities varied from year to year, with anomalously high years coinciding with either tropical system-produced floods or sudden flash floods, often associated with structural failures of dams or levees. Ruin et al. (2007) used cognitive mapping to identify several factors influencing motorists’ flash flood risk perception. Younger drivers underestimate flood risk, as do drivers of lower income and drivers with no children. They found that urban people underestimate the risk for a car to be swept away by running water and are relatively more threatened by walking and driving in flood conditions than are people living in rural areas. One of the most dangerous flash flood areas in the US is the urban corridor between Dallas and San Antonio in Central and South-central Texas called locally the ‘Flash Flood Alley’, figure 2. Actually, within the conterminous US, the greatest concentration of exceptional flood peaks is at the Balcones Escarpment of central Texas, where maximum US rainfall amounts apparently coincide with basin physiography to produce many of the largest measured US floods (O’Connor and Flash flood vehicle fatalities 313 Figure 1. Road inundation caused by flash floods in San Antonio. Courtesy of the City of San Antonio (used with permission). Available in colour online. Costa 2004). This paper discusses the results of a study of motor vehicle-related flash flood fatalities in Texas for the period between 1959 and 2009. 2. Methods Information on flood fatalities and damage can be obtained from health and safety authorities and other sources. However, the most comprehensive source for flood fatality and injury data in the US is the Storm Data publication, which is maintained by the National Climatic Data Center (NCDC) of the National Oceanic and 314 H.O. Sharif et al. Figure 2. The Flash Flood Alley. Obtained from the Flood Safety Education Project http:// www.floodsafety.com (used with permission). Available in colour online. Atmospheric Administration (NOAA). This monthly publication includes a chronological listing of state-by-state fatalities, injuries and property damage for all weather-related events. Event types include floods, flash floods, hurricanes, tornadoes and other weather-related events. Storm Data includes the location and time of fatality or injury by county for most of the events. NCDC Storm Data also provides detailed description of the weather event associated with fatalities. In addition, Storm Data includes narratives that provide detailed information on casualties, weather records and other anecdotal information for significant weather events. Information included in the Storm Data publication comes from the National Weather Service (NWS) offices and personnel, the media, law enforcement authorities, government agencies, private companies and individuals. The operation manual for Storm Data (Mandt 2002) describes the criteria for inclusion of weather- related fatalities and injuries in the publication. The Storm Data was formerly called Climatological Data in 1950 and recorded only tornados. In 1955 thunderstorm and wind data were added. In 1959 this publication was officially named Storm Data and recording of all unusual weather phenomena data that are included in today’s publication began. The NCDC, the National Environmental Satellite, Data and Information Service and NOAA prepare and distribute Storm Data. In this study, we inventoried and reviewed flood fatality information for Texas included in the NCDC Storm Data publication for the period between 1959 and 2009. A total of 612 publications were reviewed. The main difference between this and other previous studies (e.g. Ashley and Ashley 2008) is that instead of reviewing summary statistics, we examined the original publications for the entire period to extract all the reported descriptions of each fatality. Descriptive statistics derived in this study include the date, time, location and weather conditions under which the flood fatality occurred. Demographic analysis of the flood fatalities was also performed. Spatial analysis was also used to identify locations where flood fatalities may be clustered (e.g. in and around urban centres). Flash flood vehicle fatalities 315 3. Results Floods, especially flash floods, are a serious hazard in Texas. Examination of the 1959–2009 NCDC Storm Data records indicates that Texas is the only state that reported flood-related fatalities every single year during that period. By far, Texas leads the nation in flood fatalities. From 1959 through 2009 there were three times more fatalities in Texas (854) than in the next leading state, Pennsylvania (265). Table 1 lists the five states with the most number of flood fatalities. Flood events are typically classified in Storm Data as floods, flash floods (defined earlier), or floods due to tropical systems. The distribution of fatalities by flood type is shown in table 2. About 68% (584 out of 854) of Texas flood fatalities are caused by flash floods. 3.1 Person Storm Data provides the gender of only 81% (286 out of 351) of motor vehicle- related fatalities caused by flash floods and the ages of only 58% (202 out of 351) of the victims. For cases when the gender of the victim is known, the ratio of males to females is 1.7 to 1 (180 to 106). The age distribution for victims with known age is shown in table 3. The age group with the most fatalities is young adults (39 known victims). Infants, children, teenagers and the elderly (less than 20 years of age and more than 59 years) represent 47% of the victims (94 out of 202). The number of victims who are infants or young children is significant (26). 3.2 Place The distribution of flash flood fatalities in Texas with known activity/location from NCDC Storm Data is shown in table 4. ‘In water’ describes the case when a person intentionally walked into the floodwater. Review of the data reveals that 80% (351 out of 440) of the fatalities with known circumstances are motor vehicle-related. Table 1. The five states with the highest numbers of total flood fatalities, 1959–2009. State Number of fatalities Texas 854 Pennsylvania 265 California 248 South Dakota 244 Virginia 237 Table 2. Flood fatalities classified by flood type. Flood type Fatalities Percentage Flash flood 584 68% Flood 221 26% Flood due to tropical storms 42 5% Coastal/tidal/river flooding 7 1% 316 H.O. Sharif et al. This percentage is higher than reported in other studies. For example, according to the Flood Safety Education Project (2010), most flood drownings are vehicle related, and in Texas, 76% of flood deaths are vehicle related. The circumstances of death (i.e. whether the death was motor vehicle-related or not) are not described in Storm Data for 25% (144 out of 584) of flash flood fatalities. Confirmed motor vehicle fatalities caused by flash floods were reported for all but three years of the record, as shown in figure 3. The 51-year average number of confirmed fatalities is 6.9 per year. The monthly distribution of motor vehicle-related flash flood fatalities and the average of the climatological mean rainfall in the major urban areas with highest flood fatalities are shown in figure 4. The two lines have similar distributions and peaks. The correlation coefficient between the two series is 0.90. Table 3. Number of motor vehicle-related flood fatalities classified by age. The age is not provided in Storm Data for 149 fatalities. Age group Fatalities Percentage 0–9 26 13% 10–19 26 13% 20–29 39 19% 30–39 17 8% 40–49 24 12% 50–59 28 14% 60–69 21 10% 70–79 14 7% 80–89 7 4% Table 4. Number of flash flood fatalities classified by activity/location. The activity/location is not provided in Storm Data for 144 fatalities. Activity/location Fatalities Percentage Boat 8 2% In water 65 15% Mobile home 6 1% Permanent home 10 2% Vehicle 351 80% Figure 3. Motor vehicle-related flash flood fatalities in Texas from 1959 to 2009. Flash flood vehicle fatalities 317 The Texas counties with the highest numbers of vehicle fatalities caused by flash floods are listed in table 5. These ten counties account for more than 50% of the fatalities. Figure 5 shows motor vehicle-related flood fatalities in Texas categorized by county. Eight counties had 11 or more motor vehicle-related flood deaths, 15 counties had 4 to 10 flood deaths, and 67 counties had 1 to 3 deaths. 3.3 Time Storm Data also associated motor vehicle fatalities caused by flash flood with time of occurrence. The time was reported for 306 (out of 351) of the fatalities. It is not clear whether the time reported in Storm Data is the time of the accident or the time of death. As seen in table 6, most of the 306 fatalities (55%) have taken place at night. 4. Discussion A worrying consequence of rainfall road safety is the motor vehicle-related flash flood fatalities. In Texas, the number of motor vehicle-related flash flood fatalities Figure 4. Motor vehicle-related flash flood fatalities in Texas and mean precipitation in the Flash Flood Alley region by month. Table 5. Texas counties with the highest numbers of vehicle fatalities caused by flash floods. County Fatalities Bexar 30 Tarrant 30 Travis 24 Dallas 22 Harris 16 Kerr 16 Edwards 12 Bell 11 Real 9 Burnet 8 318 H.O. Sharif et al. Figure 5. Map of motor vehicle-related flash flood fatalities by Texas county. Available in colour online. Table 6. Number of motor vehicle fatalities caused by flash floods classified by time of occurrence. The time is not provided in Storm Data for 45 fatalities. Time Fatalities Percentage Morning 84 35% Afternoon 57 10% Night 165 55% far exceeds the total number of flood fatalities in any other state between 1959 and 2009. There is an increasing trend in the number of annual fatalities. That is probably due to the increasing population density in Texas, figure 3. Years associated with major floods witnessed the highest numbers of fatalities (e.g. 1979, 1981, 2007). A review of the data indicates that 80% (351 out of 440) of the flash flood fatalities with known circumstances are motor vehicle-related. The number of those fatalities in Texas is expected to be actually higher than 351, as we expect that a significant number among the fatalities whose circumstances are not provided would be motor vehicle-related. More than 68% of floods that result in fatalities in Texas are flash floods, which typically arrive immediately after the storm with little or no warning. Tropical systems, although associated with very intense rainfall and ensuing floods, result in Flash flood vehicle fatalities 319 a very small number of fatalities (less than 5% of the fatalities). This small number may have several causes. First, tropical systems are typically forecasted with longer lead times (many hours or days), and the warnings are thus more widely broadcasted to the general public. Secondly, the longer lead times allow emergency personnel to develop a heightened awareness for greater involvement. Thirdly, tropical storm system floods are less frequent than flash floods in Texas. A review of ages of motor vehicle-related flash flood accidents indicates that all age groups are at risk of dying from floods, including significant numbers of victims of ages below 20 years or above 60 years. Significantly more males die in motor vehicle accidents during a flash flood than females. This agrees with studies that addressed vehicle-related fatalities in general. McKenna et al. (1998), for example, suggested that the presence of a female passenger significantly reduces the likelihood of an accident, and Chen et al. (2000) reported that the presence of a male passenger almost doubles the per capita death rate, regardless of the driver. Motor vehicle-related flash flood fatalities appear to be related to topography and climate. Very few fatalities occurred in counties with dry climate even when the population density is high. El Paso County is one such example. This agrees with the findings of Khan et al. (2008), who found significant spatial correlation between the number of accidents and weather patterns. Fatalities are clustered in counties located along the Balcones Escarpment and those that include major urban centres. The fatality numbers in Harris County are smaller than some counties along the Balcones Escarpment although it is one of the counties with the highest number of flood fatalities in general. The monthly distribution of motor vehicle-related flash flood fatalities closely follows the monthly distribution of rainfall in the flash flood region in Texas Flash Flood Alley. This indicates that the frequency of storms plays an important role in increasing the flash flood hazard. Other studies have found a similar correlation between road accidents and rainfall (e.g. Levine et al. 1995). Most motor vehicle-related flash flood deaths in Texas appear to have occurred at night (table 6). This increase in fatalities at night may be attributed to people not being able to see or estimate the depth and speed of the flood water in inadequate lighting. Some earlier studies reported that 75% of flash-flood deaths occurred during the hours of twilight and darkness (Mooney 1983). The general behavioural reaction to severe weather includes trip cancelling, trip shortening, route changing and more use of public transportation (De Dios Ortuzar and Willumsen 2001). The behavioural response of a driver to poor weather conditions on the road, however, is influenced by several factors. Kates (1971) reported that factors that may affect risk perception include the nature and features of the natural hazard (its magnitude, duration, frequency and temporal spacing), and the recency, frequency and intensity of personal experience with past events of similar nature. The dense road networks and numerous low water crossings throughout Texas may be contributing to the higher recurrence rates of floods that pose a danger to vehicles (figure 6). Several ephemeral streams in Texas, especially along the Balcones Escarpment, have steep slopes. The highly intermittent flow of these streams reduces the overall financial efficiency of such large, expensive structures at road-stream crossings. Therefore, many crossings are constructed as armoured sag vertical curves, often with a small corrugated metal culvert pipe as a relief structure to prevent long-term ponding of water upstream of the roadway embankment. In flood 320 H.O. Sharif et al. Figure 6. Map of some of the low water crossings in San Antonio. Courtesy of the City of San Antonio (used with permission). Available in colour online. conditions water has to flow over the road. These low-water crossings pose immediate danger to vehicles that try to cross during flooding conditions. It is at these crossings where most motor vehicle-related flash flood fatalities happen. What many drivers do not know is that, according to the Federal Emergency Administration (FEMA 2010), six inches (0.15 m) of water will reach the bottom of most passenger cars causing loss of control and possible stalling, and two feet (0.61 m) of rushing water can carry away most vehicles, including sport utility vehicles and pickups. Excerpts from the narratives in Storm Data that describe some the circumstances that lead to fatalities are shown in table 7. Driving into flood water appears to be the cause of the majority of motor vehicle-related flash flood fatalities in Texas. This behaviour is mostly intentional and controllable. Solutions such as installation of traffic barricades, alarm signs, and water depth gauges at hazardous locations may not be enough. There are many practical limitations to these installations. Depth gauges, for example, do not necessarily convey the water depth and the implication of that depth to the lay person. Depth gauges may be improved by providing colours to indicate hazardous water depths. Barricades, on the other hand, are often installed or uninstalled too late. The driver may see barricades and choose to drive around or think that his/her truck or sports utility vehicle can cross the low lying water. Some studies confirm that passengers play a major role in controlling the behaviour of the driver (e.g. Chen et al. 2000). Modifying this behaviour will involve educating the public about flood risks and the seriousness of flood warnings. Siegrist and Gutscher (2008) found that people could envisage what physical risk a flood poses, but they could not envisage the negative affect that could be associated with such an event. In order to boost mitigation motivation, they suggested that risk communication must also help people to envisage the negative emotional consequences of a flood. Flash flood vehicle fatalities 321 Table 7. Examples of Storm Data descriptions of the circumstances that led to vehicle fatalities caused by flash floods. Date County Storm Data narrative 8/16/2007 Bexar The area of extremely heavy rainfall associated with the remains of Tropical Storm Erin continued to spread northwestward across Bexar County, with a general 4 to 5 inches of rain over the county. Near midnight a young woman was driving with three friends and a baby near North Star Mall when she accidentally drove her sport utility vehicle into deeper water where it was slammed against a bridge and then was swept into a drainage ditch. The three other adults in the vehicle were able to get the baby out of the vehicle through the window and escape. But when the three looked back for the driver, she was gone. Her body was found later by emergency responders when the water receded. 9/05/2007 Bexar Up to 2 inches of rain fell over the northwestern part of the county. At around 1:00 am a man and a woman were washed away in a four door sedan as they attempted to cross the Tower View Bridge on Scenic Loop Road, which was in flood due to Helotes Creek. The man was later rescued with minor bruises to his arms and torso. The woman’s body was found less than a mile downstream at 3:19 pm. 3/30/2008 San Augustine Widespread flooding reported across the northern portion of the county. Flooding was severe on County Road 21 in town and at the circle and along County Road 103. Three foot of water was also reported over Highway 3079. One fatality was reported when two people driving on State Highway 21 ran into a ditch and became caught in swift moving water. One occupant in the vehicle was swept away in the vehicle. 4/12/2009 Harris Heavy rainfall persisted across the county as several thunderstorms produced rainfall totals of 8 to 10 inches across the county. Numerous roads were closed in areas of the county, including Jersey Village, Houston Heights, La Porte, Pasadena, Webster, and near the Houston Hobby airport. The highest one hour rainfall total was 6.90 inches recorded at Bay Area Blvd and Clear Creek. This exceeds the highest one hour rainfall rate measured during Tropical Storms. In the northern part of the county five children were killed after the car they were riding in inadvertently drove into a drainage ditch that was filled with nine feet of water from the heavy rainfall. Two adults and one child were able to escape the floodwaters, but the five younger children, all under 7 years old, were not able to escape. 5/03/2009 Upshur Numerous counties and parishes were flooded with rainfall amounts in excess of 6 inches in a 12 hour period common. A 78 year old man drowned when flood waters swept away his automobile on a rural northwestern Upshur County Road. The man was found in his back seat on Fox Road. 10/21/2009 Burnet A woman was killed while trying to cross a low water crossing. Karen Eichelberger, age 47, was driving home after working in Austin this Wednesday night. She tried to cross the flooded low water crossing on Deer Springs Drive in her neighbourhood when her 1994 Jeep was apparently forced off the road and into the creek. Her body was found one half mile downstream from where the Jeep was located. 322 H.O. Sharif et al. The consequences of rainfall and flooding on transport safety are well studied. However, motor vehicle-related flood fatalities have received little attention in the transportation literature. There is evidence that rainfall intensity and flooding frequency are increasing in many regions around the globe. In addition, studies predict that the intensity and frequency of storms and hurricanes in Texas will rise due to climate change and variability (IPCC 2007). Although rainfall and flooding are believed to reduce the severity of traffic accidents (e.g. Koetse and Rietveld 2009), they cause significant motor vehicle fatalities in Texas. Attention to this problem may help reduce these fatalities. In addition, physical and hydraulic field investigation is needed to better understand the factors that lead to the high frequency of these events in Texas and the impact on climate change of these factors. That the Flash Flood Alley region of Texas is experiencing unprecedented urbanization and population growth, leading to an increase in the impervious areas and reduction in infiltration, suggests that the flood fatality rate will become even higher if the issue is not given due attention. References ANDREY, J., MILLS, C. B., LEAHY, M. and SUGGETT, J., 2003, Weather as a chronic hazard for road transportation in Canadian cities. Natural Hazards, 28, pp. 319–343. ASHLEY, S.T. and ASHLEY, W.S., 2008, Flood fatalities in the United States. Journal of Applied Meteorology and Climatology, 47, pp. 805–818. BERTNESS, J., 1980, Rain-related impact on selected transportation activities and utility services in the Chicago area. Journal of Applied Meteorology, 19, pp. 545–556. BRODSKY, H. and HAKKERT, A.S., 1988, Risk of a road accident in rainy weather. Accident Analysis and Prevention, 10, pp. 161–176. BROWN, M.D., 2005, Being flood smart means being prepared. Available online at http:// www.fema.gov/media/archives/2005/050613.shtm (accessed 15 December 2010). BURBY, R.J., 2001, Flood insurance and floodplain management: the U.S. experience. Environmental Hazards, 3, pp. 111–122. CHANGNON, S.A., 1998, Comments on ‘Secular trends of precipitation amount, frequency, and intensity in the United States.’ Bulletin of the American Meteorological Society, 79, pp. 2550–2552. CHEN, L., BAKER, S.P., BRAVER, E.R. and LI, G., 2000, Carrying passengers as a risk factor for crashes fatal to 16 and 17 year old drivers. Journal of the American Medical Association, 283, pp. 1578–1582. DE DIOS ORTUZAR, J. and WILLUMSEN, L.G., 2001, Modelling Transport, 3rd edn (Chichester: John Wiley & Sons). DOWNTOWN, M.W., MILLER, J.Z.B. and PIELKE JR, R.A., 2005, Reanalysis of U.S. National Weather Service flood loss database. Natural Hazards Review, 6, pp. 13–22. DROBOT, S.D., BENIGHT, C. and GRUNTFEST, E.C., 2007, Risk factors associated with driving through flooded roads. Environmental Hazards, 7, pp. 227–234. EISENBERG, D. and WARNER, K., 2005, The effects of snowfalls on motor vehicle collisions, injuries, and fatalities. American Journal of Public Health, 95, pp. 120–124. FEMA (FEDERAL EMERGENCY ADMINISTRATION), 2010, Are You Ready Guide. Available online at: http://www.fema.gov/areyouready/preface.shtm (accessed 15 August 2010). FLOOD SAFETY EDUCATION PROJECT, 2010, Available online at: http://www.floodsafety.com (accessed 15 December 2010). FRENCH, J. G., ING, R., VON ALLMEN, S. and WOOD, R., 1983, Mortality from flash floods: a review of the National Weather Service reports, 1969–1981. Public Health Report, 98, pp. 584–588. Flash flood vehicle fatalities 323 GOODWIN, L., 2002, Analysis of weather-related crashes on U.S. Highways, Mitretek Systems, prepared for the FHWA Road Weather Management Program. Available online at: www.ops.fhwa.dot.gov/weather/resources/publications/fhwa/crashanalysis2001.pdf (accessed 15 December 2010). IPCC (INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE), 2007, Climate Change 2007: The physical science basis: summary for policy makers. IPCC Secretariat, Geneva. KATES, R.W., 1971, Natural hazard in human ecological perspective: hypotheses and models. Economic Geography, 47, pp. 438–451. KHAN, G., QIN, X. and NOYCE, D., 2008, Spatial analysis of weather crash patterns. ASCE Journal of Transportation Engineering, 134, pp. 191–202. KOETSE, M.J. and RIETVELD, P., 2009, The impact of climate change and weather on transport: an overview of empirical findings. Transportation Research Part D, 14, pp. 202–221. LEVINE, N., KIM, K.E. and NITZ, L. H., 1995, Daily fluctuations in Honolulu motor vehicle accidents. Accident Analysis and Prevention, 27, pp. 785–796. MANDT, G.A., 2002, Storm Data preparation. National Weather Service, National Oceanic and Atmospheric Administration, Department of Commerce, National Weather Service Instruction 10-1605, NWSFD 10-16, Silver Spring, Maryland, 79 pp. MCKENNA, A.P., WAYLEN, A.E. and BURKES, M.E., 1998, Male and female drivers: how different are they? The University of Reading, Foundation for Road Safety Research, Reading. MOONEY, L.E., 1983, Applications and implications of fatality statistics to the flash flood problem. Preprints, Fifth American Meteorological Society Conference on Hydrometeorology, Tulsa, OK (Boston, MA: American Meteorological Society), pp. 127–129. NWS (NATIONAL WEATHER SERVICE), 2005, Flood fatalities. Hydrologic Information Center. Available online at: www.nws.noaa.gov/oh/hic/flood_stats/recent_individual_deaths. shtml (accessed 15 July 2010). O’CONNOR, J.E. and COSTA, J.E., 2004, The world’s largest floods, past and present: their causes and magnitudes. US Geological Survey Circular, no. 1254, 13 pp. PIELKE Jr, R.A. and DOWNTON, M.W., 2000, Precipitation and damaging floods: trends in the United States, 1932–97. Journal of Climate, 13, pp. 3625–3637. PISANO, P.A., GOODWIN, L.C. and ROSSETTI, M.A., 2008, U.S. Highway crashes in adverse road weather conditions. Proceedings of the 88th Annual American Meteorological Society Meeting, 20–24 January, New Orleans, LA (Boston, MA: American Meteorological Society), Paper 8.1, CD-ROM. RUIN, I., GAILLARD, J. and LUTOFF, C., 2007, How to get there? Assessing motorists’ flash flood risk perception on daily itineraries. Environmental Hazards, 7, pp. 235–244. SHARIF, H. O., YATES, D., ROBERTS, R. and MUTOFF, C., 2006, The use of an automated now- casting system to forecast flash floods in an urban watershed. Journal of Hydrometeorology, 7, pp. 190–202. SIEGRIST, M. and GUTSCHER, H., 2008, Natural hazards and motivation for mitigation behavior: people cannot predict the affect evoked by a severe flood. Risk Analysis, 28, pp. 771–778.

Journal

"Geomatics, Natural Hazards and Risk" – Taylor & Francis

Published: Nov 1, 2012

There are no references for this article.