Adopted: May 13, 1981
MOTOR VEHICLE COLLISIONS WITH TREES
ALONG HIGHWAYS, ROADS, AND STREETS: AN ASSESSMENT
Each year there are numerous accidents where a vehicle runs off the road and collides with a tree. The causes of the accident are varied, but often involve human factors such as fatigue and inattention or failure to perceive changes in roadway alignment. Regardless of the cause, this type of accident can produce devastating results. A tree represents a narrow, relatively unyielding, fixed object which can deeply penetrate the vehicle, and as a result individuals at the scene of the accident immediately assume a high speed was involved; this assumption may not always be correct. To eliminate or reduce the severity of these accidents, either the vehicle must stay on the road or the tree must be protected or removed. This study will discuss these alternatives.
The highway safety community does not have a standard definition of a "tree accident." Therefore, it is difficult to compare tree accidents from one State's data base to another. It is particularly difficult to differentiate between a tree and a shrub, and also between large, unyielding shrubs, and smaller shrubs which might actually slow down an errant vehicle beneficially. For example, the National Highway Traffic Safety Administration's (NHTSA) Fatal Accident Reporting System (FARS) data does not make a distinction between trees and shrubs, but instead has a category of "impact with tree or shrub." Various data sources were analyzed and cited in this study. In each case, we have footnoted the specific definition of tree and/or shrub, used in that data base. However, to simplify the report we have used the term "tree accident" throughout since this is sufficient for our purposes. Nevertheless, the reader should bear this in mind when analyzing specific data or conclusions in the report.
The National Highway Traffic Safety Administration's (NHTSA) Fatal Accident Reporting System (FARS) reported that an average of 2,900 fatal accidents 1/ resulting in 3,280 fatalities occurred annually during 1978 and 1979 involving highway vehicles in which the "first harmful event involves collision with tree/shrub." 2/ Fatal accidents with trees, the most common fixed object struck in a fatal accident, 3/ represent 6.65 percent of all fatal accidents. The number of fatal accidents with trees has risen from 2,731 in 1977, to 2,873 in 1978, to 2,940 in 1979.
A study (1) indicated that of the accidents in which a single vehicle strikes a fixed object, those with trees had the highest severity with 61 percent resulting in an injury or fatality. Data obtained from eight States 4/ also indicated that 61.2 percent of all vehicle impacts with trees result in a fatality or injury.
In addition, the data provided by the eight States indicated that for every fa tal accident that occurs with a tree, 23 injury accidents, and 15 vehicle-damage-only accidents also will occur that involve a tree. Based on these figures, the following is a nationwide estimate of accidents involving trees each year:
-2,900 fatal accidents resulting in 3,280 fatalities.
-66,700 injury accidents resulting in 94,400 injuries.
-44,200 vehicle-damage-only accidents.
Based on the 1975 NHTSA cost estimates for accidents (2), the annual projected cost of accidents involving trees would be about $1.25 billion.
In 1978, there were about 143,100,000 licensed drivers in the nation. Since about 114,000 accidents involve trees annually, 1 out of every 1,250 drivers may be involved in a tree accident annually. Assuming the average driving career with substantial driving exposure to be about 40 years (20 to 60 years old), an estimated 1 in every 95 drivers will be involved in an accident with a tree during their driving career.
The National Transportation Safety Board (NTSB) initiated this special study because of the number and severity of vehicle collisions with trees and because of a lack of tree removal or protection projects at the State level. Federal Highway Administration (FHWA) safety officials have expressed considerable concern about accidents involving trees. According to FHWA officials, removal projects of roadside obstacles have diminished at the State level in recent times, and tree removal projects have essentially been eliminated in some States because of extreme local opposition. Some States are only removing trees at highly hazardous accident locations; some States are not even removing those trees known to be hazardous. Because of the severity of the problem and the lack of action at the State level, countermeasures need to be established to reduce the severity and provide a safer roadway environment.
To that end, the Safety Board initiated this special study to highlight the severity of the problem, to determine the characteristics of these accidents, and to explore methods that could be implemented to reduce the number of fatalities and injuries that result from vehicle collisions with trees.
The Safety Board reviewed NHTSA's FARS data to determine the locations of accidents, the percentage of fatal vehicle accidents with trees to all fatal vehicle accidents, and significant characteristics relating to the accident and roadway conditions. The Safety Board reviewed National Crash Severity Study (NCSS) data to determine injury severity and vehicle deformation. In addition, the Safety Board investigated 19 accidents with trees that resulted In 24 fatalities and 19 Injuries. Seven States supplied information regarding accident data, roadside obstacle pro .grams, and tree removal or protection programs.
As a result of this special study, the National Transportation Safety Board made these recommendations:
--to the Federal Highway Administration:
Develop several county-wide demonstration projects to evaluate the potential of reducing the number and severity of accidents with trees, especially at curves on county roads, by improving signing and delineation in various combinations. (Class III, Longer Term Action) (H-81-20)
--to the National Highway Traffic Safety Administration:
Revise the FARS form and other nationwide reporting forms to include, as a minimum, the measurement of the distance from the edge of the road to a fixed object struck and measure of road curvature (if curve is present). (Class III, Longer Term Action) (H-81-21)
--to the National League of Cities, National Association of Towns and Townships, and the National Association of Counties:
Encourage the development of local programs and policies to reduce the number of accidents with trees. Programs should emphasize improvement of roadway curves and locations where trees have been struck previously through delineation, signing, and removal or shielding of trees. Policies should be developed to prevent the future placement of trees that grow large enough to become a hazard, 4 inches or more in diameter, within the warranted clear recovery areas. (Class III, Longer Term Action) (H-81-22, -23, -24)
--to the International Association of Chiefs of Police:
For all fixed-object accidents, encourage the recording of distance from the edge of the road to the fixed object struck and a measure of curvature of the road (when curve is present) on accident reports at the State and local level. (Class III, Longer Term Action) (H-81-25)
PATRICIA A. GOLDMAN, Member, filed the following concurring statement:
Although I concur with the findings of this study I have certain reservations regarding the recommendation to the International Association of Chiefs of Police. Historically, the Federal Government has repeatedly requested assistance from State and local police departments to support Federal safety programs. In fact, that support is critical to the existing FARS 'and NASS programs, which I support. Therefore, to the extent that this recommendation would encourage policy support for improvements in FARS and NASS, I support the recommendation. However, I am concerned that the intent of the recommendation as adopted by the majority of my colleagues is much broader.
From a Federal perspective, I am not convinced that additional data beyond that which would be part of FARS and NASS are justified. Furthermore, if additional State data are needed I believe that the decisions related to collecting it would be best left to the discretion of the State and local officials.