CHARACTERISTICS OF TRAFFIC ACCIDENTS IN COLD, SNOWY HOKKAIDO, JAPAN

CHARACTERISTICS OF TRAFFIC ACCIDENTS IN COLD, SNOWY HOKKAIDO, JAPAN Motoki ASANO Director Traffic Engineering Division Civil Engineering Research Institute of 1-3 Hiragishi, Toyohira-ku, Sapporo, 062-8602, Japan Fax: +81-11-841-9747 E-mail: m-asano@ceri.go.jp Masayuki HIRASAWA Researcher Traffic Engineering Division Civil Engineering Research Institute of 1-3 Hiragishi, Toyohira-ku, Sapporo, 062-8602, Japan Fax: +81-11-841-9747 E-mail: hirasawa@ceri.go.jp ABSTRACT:, the northernmost prefecture of Japan, lies in the subarctic zone. For approximately the four-month period of late November to mid-march, the weather is cold and snowy, and the daily mean temperature is below freezing. In 1992, the law regulating studded tires was enacted to prevent air dust pollution generated. Since 1992, when the studded-tire-equipping rate fell to almost 0%, the skid resistance of road surfaces has fallen markedly and slippery road surfaces have emerged. Winter-specific accidents increased accordingly. We analyzed the characteristics of traffic accidents in and compared these with other regions of Japan. It was found that has more traffic fatalities per traffic accident and more fatal single- and multi-vehicle collisions in rural areas. The results also revealed that many winter-specific accidents occur in December, early winter. Key Words: traffic accident, skid accident, winter road, studded tire regulation 1. REGIONAL AND CLIMATIC CHARACTERISTICS OF HOKKAIDO 's location at the northernmost extreme of Japan has given this prefecture a very different climate from elsewhere in Japan. While most of the other regions lie in the temperate climatic zone, is in the subarctic zone, although has four distinct seasons just as do other regions of Japan. is at a latitude of roughly 43 ºN. European regions at this latitude include those in Southern France and Central Italy. Winter winds bring cold air from Siberia to, which gives it the severe winter climate. The lowest temperature recorded in Asahikawa, the city at the geographic center of, is -41 ºC (-41.8 ºF). The normal annual number of days with snowcover is 148.2 days, and days with snowfall is 143.8 days. The maximum recorded depth of snowcover is 138 cm. The normal monthly temperature of Sapporo, which is the political and economic center of, is about -4 ºC(24.8 ºF) during the coldest months, January and February. The normal annual number of days with snowcover and days with snowfall are 132.4 days and 124.7 days, respectively. The population of Sapporo exceeds 1.8 million. The modern full-scale development of dates back to the early Meiji era, about 130 years ago. This history is the shortest of anywhere in Japan. Therefore, is typified by low population density, long inter-city distance, and low traffic volume. Also, it lags behind the rest of the nation in the percentage of completion for

expressways projects. 2. STUDDED TIRE REGULATION AND ITS EFFECTS In Japan, studded tires began to gain widespread use from around 1970, soon becoming the standard winter tire. However, in the 1980s, they were found to cause pavement abrasion, rutting, and air pollution by dust. Movements by citizens, the enactment of ordinances, and the voluntary restriction of studded tire production culminated in the promulgation of "The Law on the Prevention of the Generation of Particulates from Studded Tires" in 1990. While this law contributed to dramatic improvements in air quality, very slippery road surfaces have emerged each winter since fiscal 1992. This has raised concerns about the increase in winter-specific traffic accidents and the environmental effects associated with increased application of anti-freezing agents (Figure 1). Figure 1 The Several Indices Related to The Rate of Studded Tires 3. CHANGES IN TRAFFIC FATALITIES The worst record for traffic fatalities in Japan was 16,765 in 1970. The number dropped to 8,466 by 1979, half the 1970 figure, before increasing again. It reached a second peak of 11,451 in 1992, after which the number started falling. As of 2001, the number was 8,747. has more traffic fatalities per capita than Japan overall. The peaks for were 889 and 715 in 1971 and 1990, respectively. The number of traffic fatalities fell dramatically to 573 in 1991, since which date it has repeatedly risen and fallen. As of 2001, it was 516. 4. CHANGES IN TRAFFIC ACCIDENTS

Traffic accidents in Japan numbered 720,880 in 1969, which was then a record high. The number decreased thereafter, but started increasing again from 1977. In 1993, the 1969 record was broken. The number had increased to 947,169 as of 2001. Traffic accidents in peaked at 30,042 in 1970. Thereafter the number began to fall, before rising again from 1975, a year in which 15,971 traffic accidents occurred. The number increased to 30,531 as of 2001, almost equal to the number for 1970. The number of traffic accidents in has followed a trend similar to that for Japan overall, although the per-capita number for is smaller than for Japan overall. 5. CHANGES IN TRAFFIC FATALITY RATE Figure 2 shows changes in the number of fatalities per 100 traffic accidents (traffic fatality rate) in and Japan overall. The traffic fatality rate of Japan overall has been decreasing, except for the years 1987 to 1990. The rate of decrease has been particularly high since 1990. The reason is a decrease in fatalities accompanied by a sharp increase in accidents, which is the denominator. The fatality rate of is higher than that of Japan overall, i.e., there are fewer accidents and more fatalities than the national average. However, it started decreasing from 1997 and fell below 2.0 in 1998. One of the causes for 's high fatality rate is thought to be high travel speed in rural areas. is not densely populated, and its cities are widely distributed. The inter-city distance of is roughly twice that of Honshu, the main island of Japan. These factors have caused travel speeds to be high in rural areas. 4.00 Fatalities per 100 Traffic Accidents 3.50 3.00 2.50 2.00 1.50 1.00 2.94 2.80 2.26 2.33 3.28 3.26 3.28 3.30 3.37 3.37 3.21 3.02 3.05 3.052.91 2.88 2.89 2.95 2.80 2.76 2.60 2.38 2.30 2.23 2.20 2.32 2.34 2.33 2.38 2.29 2.19 2.32 2.412.48 2.33 2.28 1.89 1.78 2.07 1.69 1.94 1.891.80 1.84 1.801.811.811.79 1.81 1.68 1.61 1.58 1.681.681.75 1.68 1.65 1.51 1.46 1.40 1.291.24 1.15 1.060.970.92 Japan 0.50 0.00 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Figure 2 Fatalities per 100 Traffic Accidents in and Japan 6. TRAFFIC FATALITY RATE AND ACCIDENT RATE PER 100 MILLION

VEHICLE-KM OF TRAVEL Vehicle-km of travel is a total of the travel distance of vehicles based on road traffic census. This index (cumulative unit road length traffic volume of each unit) represents the total road traffic demand. By dividing the number of traffic accidents by the vehicle-km of travel and multiplying the quotient by 100 million, we obtain the traffic accident rate per 100 million vehicle-km of travel. By the same calculation, we obtain the traffic fatality rate per 100 million vehicle-km of travel from the number of traffic fatalities. Figures 3 and 4 show the changes in these rates on national highways and expressways. The traffic accident rate per 100 million vehicle-km of travel on national highways in has been almost constant at 60% of that for Japan overall. On expressways, the rate for is about the same as that for Japan overall. The traffic accident rates on expressways are very low compared with national highways both in and in nationwide. 100.00 93.31 94.76 90.00 80.00 85.97 82.65 84.38 76.31 80.97 82.04 86.22 81.15 81.81 84.78 86.04 70.00 Highway in Japan 60.00 50.00 52.20 52.15 51.45 43.83 44.39 47.21 49.14 45.10 40.68 Highway in 44.39 44.48 46.03 44.82 40.00 30.00 20.00 10.00 0.00 Expressway in Japan Expressway in 11.88 12.48 13.55 10.06 11.22 11.70 12.13 10.23 10.12 10.01 8.26 9.17 9.22 12.24 10.13 9.68 6.34 5.64 6.85 7.86 8.01 7.67 8.50 9.21 9.85 8.75 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Figure 3 Traffic Accidents per 1 billion vehicle kilometres in and Japan Year

2.50 2.00 1.50 1.00 0.50 0.00 2.05 2.11 1.99 1.82 2.00 2.01 1.91 1.73 1.22 1.11 0.94 0.84 0.84 0.63 0.56 0.31 Expressway in 1.78 1.85 1.74 1.60 1.74 1.71 Highway in Japan 0.75 0.57 0.57 0.63 0.50 0.00 Highway in 1.70 1.51 1.47 1.48 1.40 1.43 1.53 1.45 1.38 1.33 1.33 1.25 Expressway in Japan 0.52 0.50 0.42 0.41 0.32 0.46 0.36 0.41 0.35 0.29 0.24 0.18 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Year Figure 4 Fatalities per 1 billion vehicle kilometres in and Japan 7. CHARACTERISTICS OF TRAFFIC ACCIDENTS BY ACCIDENTS TYPE Figure 5 presents the number and percentage of traffic accidents and fatal traffic accidents by accident type. Compared with Japan overall, the proportion of rear-end collisions among total traffic accidents is high in. Among fatal traffic accidents, the proportions of head-on collisions and deviation accidents are high. Possible reasons include 1) the road structure (most roads in are two lanes without a median strip), 2) the road surface condition (slippery for most of the winter), and 3) the shortage of expressways.

Human and Vehicle Collision Head-on Collision Rear-end Collision Car to Car Collision Construction Collision Deviation Accidents Independent Collision and Others Vehicle and Train Collision 98 140 17 100 89 40 9 1 Fatal Accidents 292 Japan 2,468 1,203 525 2,365 1,467 333 54 722 346 2611 1740 10983 14081 2 Traffic Accidents 33,854 27,038 321 20,525 Japan 86,094 276,911 482,081 128 5,303 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Figure 5 The Number and Percentage of Traffic Accidents by Accident Type 8. CHARACTERISTICS OF TRAFFIC ACCIDENTS IN WINTER Figure 6 shows the monthly changes in the number of traffic accidents in. The number for December started to increase from fiscal 1992. It exceeds those of the other months by more than 500. Introduction of studless tires has led to an increase in traffic accidents in early winter (i.e., December). One possible reason for this is that drivers have not yet adjusted to winter road surfaces at the early stage of winter. 4000 3500 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Number of Traffic Accidents 3000 2500 2000 1500 1000 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Year Figure 6 Annual Trend of The Number of Traffic Accidents by Each Month in and Japan

Accidents during winter (November through March) that are caused directly or indirectly by winter phenomena, such as snow accumulation, ice, and snowstorm, are defined as winter-specific accidents. Figure 7 shows changes in the number of traffic fatalities and accidents by winter-specific accidents. Winter-specific accidents include skid accidents caused by compacted snow and icy road surfaces, accidents caused by snowstorm-induced poor visibility, and accidents caused by ruts. The figure shows winter-specific accidents have been increasing. Winter-specific accidents greatly increased from fiscal 1989 to fiscal 1993, during which period the studded-tire-equipping rate was decreasing every year. Pervasion of studded tires is likely to have been responsible for this sudden increase. The number of winter-specific accidents stabilized thereafter, but it started increasing again from 1998. Conversely, the number of fatalities for accidents in winter and winter-specific accidents have been largely stable. 5,500 4,500 3,500 2,500 1,500 500 Accidents Fatalities 130 Winter-Type Accidents Skidding Accidents Fatalities Fatalities 5,015 4,535 110 4,129 3,941 3,813 4,315 3,423 3,547 3,561 3,464 3,226 3,445 3,313 90 3,067 3,001 3,216 2,918 2,563 2,752 2,090 64 2,162 66 56 55 56 1,593 54 58 52 61 52 49 52 51 51 41 48 48 48 46 42 44 70 50 37 30 89/90 90/91 91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99 99/00 Figure 7 Trend of winter-specific(=winter-type) accidents and fatalities

Figure 8 shows percentage of skid accidents by accident type. Rear-end collisions account for approximately 50% of all accidents. 3% 3% 2% 1% Rear-end Collision Interspection Collision Car to Car Collision 20% Head-on Collision Human and Vehicle Collision Construction Collision Deviation Accidents 50% Independent Collision and Others 12% Figure 8 9% Percentage of skid accidents by accident type 9. OCCURRENCES OF SKID ACCIDENTS, TEMPERATURE, AND SNOWFALL Figure 9 shows daily mean temperatures and snowfall, as well as the average number of skid accidents in Sapporo for the 1989-99 winter (November through March). The figure excludes days without skid accidents. Since penalties began to be imposed under the studded tire regulation law that took effect in fiscal 1992, the studded-tire-equipping rate has fallen to almost 0%. Therefore, one graph shows the fiscal years up to 1991, and another the years after. The number of skid accidents per day in fiscal 1992 and thereafter exceeded that in fiscal years up to 1991. The average number of skid accidents per day showed a peak when the daily average temperature was around -4 ºC(24.8 ºF) (Figure 9). At this temperature, very slippery surfaces are likely to form.

16 Daily Mean Number of Winter-type(skidding) Accidents 14 12 10 8 6 4 2 6.7 12.2 10 13 7.9 13.4 8.2 14.1 14.2 9.4 8.2 12.8 8.6 12.2 6.6 9.6 6 8.4 1989.3-1992.3 1992.12-2000.3 4.9 4.1 3.8 3 2.4 2.5 0 <-7-7--6-6--5-5--4-4--3-3--2-2--1-1-0 0-1 1-2 2-3 >3 Daily Mean Temperature Figure 9 The Daily Mean Number of Skid Accidents VS. Daily Mean Temperature 10. POSTSCRIPT Every year, traffic accident fatalities exceed 500 people in, and they number approximately 9,000 people in the entire nation. These figures are larger than that of non-traffic accidents or disasters. Societal losses are also large. Further efforts are required for improvement in the future. a) Journal Papers REFFERENCES Hirasawa, M., and Asano, M. (2001) Characteristics of Traffic Accidents in During Last Decade. Monthly Report of Civil Engineering Research Institute of, Vol.576, May. 2001, 13-21 b) Papers presented to conferences Asano, M. Hara, F., Tanabe, S., and Yokoyama, S.(2002) Evaluation of The Ten Years Since Studded Tires Were Banned in and Future Issues. II 156. XIth International Winter Road Congress 2002 Sapporo