landscape design in clear zone - aggie horticulture article.pdftransportation industry to provide...

Transportation Research Record 1851■ 119Paper No. 03-3937

The relationship between landscape installations and safety or pedes-trian activity was evaluated in pilot studies. The sites selected for studywere those adjacent to or within the areas often referred to as the “clearzone” of the transportation corridor, an area shared by pedestrians anddriver perception. Case study research on the impact of environmentalmitigation on driver safety is summarized to identify the landscapeinstallations in the clear zone that appear to have an effect on safety.This is followed by case study research on the identification of variablesthat encourage walking for health purposes. Preliminary findings indi-cate that the improved definition of spatial edge resulting from typicalcurbside and median landscape treatment in the clear zone appears tosolicit positive behavioral responses by either attracting pedestrianactivity or improving driver safety. It is not possible to draw definitiveconclusions from the results because of the small sample sizes in the pilotstudies (a pedestrian survey included 52 responses), but indications arethat the landscape in the clear zone may be having a positive impact onsafety or pedestrian activity under certain circumstances. Both pilotstudies were conducted separately in Canada and the United States.

While the transportation industry reconsiders its standards and de-sign guidelines for the greening of the streets to respond to context-sensitive design, the Centers for Disease Control and Prevention andthe Surgeon General’s office are taking a close look at identifyingvariables that favor pedestrian and bicycle use (active living) overautomobile use (passive living) to improve the nation’s health andreduce obesity and related medical care costs (1, 2). The growing in-terest in sustainable soft surface boulevard treatment and greeninfrastructure within transportation corridors increasingly includesindividuals in the health field. This paper presents the results of re-search from partnerships among landscape architects, safety engi-neers, planners, and health industry researchers seeking to identifyenvironmental variables that have a positive correlation with driversafety and that encourage pedestrian activity within the clear zone.

The hypothesis of the first pilot study is that certain types of land-scape variables introduced in midblock clear zones in transportationcorridors may positively correlate with reductions in accident fre-quency and severity (3). In the second pilot study, it was hypothe-sized that these same landscape treatments, by virtue of being aes-thetically pleasing and safe, effect an increase in pedestrian activity(4–8). The methodology used to establish the design treatments intypical environmental mitigation plans associated with corridor land-scaping projects and findings on the effect of greening on driversafety are presented. Preliminary findings regarding the identification

of landscape variables that have an effect on pedestrians’ responsesto the quality of walking environments are then presented. Again,the methodology from an ongoing pedestrian health and safety studythat focuses on the impact of environmental variables preferred bywalkers is presented. The hypothesis is that the various types ofenvironmental features in the clear zone identified by subjects asbeing part of a “good” place to walk are the same types of landscapefeatures that result in reduced accident frequencies and severities.

This work is the result of the collaborative efforts of practitioners,researchers, and students from various institutions across the south-ern United States and Canada. This paper presents transdisciplinaryresearch on the potential role of the landscape architecture of theroadside as a pedestrian health and traffic safety strategy.

The purpose of this research paper is to present a case studyexamination of the effects of landscape architectural design and en-vironmental variables on both driver safety and pedestrian activity.While these issues may seem disparate and broad at first glance, itis the area along the roadside, often within the clear zone dimensionsof standard transportation policy, where both landscape enhance-ments and pedestrians are seeking accommodation. Clear zone di-mensions are recommended in the 1988 AASHTO Roadside DesignGuide (9). The clear zone is defined as an area continuously adja-cent to the paved edge of the road, maintained clear of obstacles.Obstacles include break in slope, signs, guardrails, lampposts, trees,planters, shrubs, architectural features, and walls. The width of theclear zone between the edge of the driving lane and the permissibleobstacle is related to the design speed of the road, the design aver-age daily traffic for the road, and the slope of the adjoining verge. Itis not calculated on the basis of contextual design considerationssuch as pedestrian activity or community adjacencies. As the demandfor evidence-based, manageable, context-sensitive design increases,it is important for researchers to understand the interrelationship ofa sustainable green streets program on safety and pedestrian access.Consideration of this interrelationship is presented in this paper tocontribute to integrated health- and safety-based standards relatedto landscape architecture in the road allowance.

STUDY I: ENVIRONMENTAL VARIABLES ANDSAFETY CORRELATION

This section presents the findings from a pilot study developed byRosenblatt and Bahar (3) to review and assess the impact of envi-ronmental variables on driver safety. The findings from the casestudy as they relate to the identification and cost of implementingenvironmental design treatments and landscape variables that appearto influence driver safety are presented.

Landscape Design in Clear ZoneEffect of Landscape Variables on Pedestrian Health and Driver Safety

Jody Rosenblatt Naderi

Department of Landscape Architecture and Urban Planning, College of Architec-ture, Texas A&M University, 3135 TAMU, College Station, TX 77843-3135.

120 Paper No. 03-3937 Transportation Research Record 1851

Traditionally, the introduction of landscape features into the clearzone is discouraged in transportation policy (10). Communitiesacross the United States, however, put tremendous pressure on thetransportation industry to provide landscape and trees within theclear zone as part of community economic development, neighbor-hood beautification, or traffic-calming endeavors. See Figure 1 foran example of a landscaped clear zone. This has resulted in a seri-ous conflict that has become part of a national research strategy andthe focus of much research (11–13). At the same time, the context-sensitive design movement is beginning to provide the engineeringcommunity with skills and new standards and policies related to theincorporation of community preferences and needs into traditionaltransportation projects. Nevertheless, serious liability issues areattached to the management and maintenance of clear zones, andmany states have developed intricate policies to address the need ona broad basis (14–16 ).

As a performance indicator, safety may be considered a majorfactor in contributing to a feeling of community well-being. It is alsowell documented in safety research and community planning that asense of safety is a significant contributor to active living. Becauseof the correlations between trees, environmental aesthetics, andwalking, it has become important to understand the nature of thesafety effect of landscaping within the right-of-way clear zone. Themethodology presented below is one way to determine what kindof landscape treatment is beginning to show promise as a safetymeasure and how to ascertain the magnitude of the effect.

Methodology

A landscape architect and traffic safety engineer team evaluated theimpact of implementation of green streets infrastructure improve-ments on five arterial roads in Toronto, Ontario, Canada, between1992 and 1995 and found that a correlation between greening of theroad and a reduction in midblock accident frequency and severitypossibly exists. The thrust of the case study was to conduct an analy-sis of pre- and postconstruction accident statistics on five arterialcorridors that had been subjected to landscape aesthetic improve-ments. The review and assessment also included an evaluation of thecost–benefit of the landscape improvements in terms of the costsavings from reductions in the numbers of accidents.

Briefly, five sites on which various environmental and aestheticimprovements were installed in response to context-sensitive designissues were selected.

All of the road segments make up a part of the arterial road sys-tem, and all are classified as major arterials in the metropolitanToronto region (Table 1). The adjacent land uses included high-risecommercial and mixed-use developments, apartment towers, civiccenters, university campuses, low-rise institutional developments,green space, and shopping malls. All roads carried mass transit aswell as private vehicles. The posted speeds on the road segmentsranged between 50 and 70 km/h. The roads either were divided withan existing raised median or a raised median was added as part ofthe improvements. In all cases, the amounts of green landscape ele-ments on the center median as well as along the edge of the curb laneincreased. Pedestrian facilities were either widened or buffered withlandscape material. Sidewalks were present on all segments. Thesafety performance at these sites was not a criterion in the selectionof the site for landscape improvements. These sites were selected forthe case study because landscape improvements were implementedand funded as part of the transportation agency’s response to com-

munity concerns attached to major road reconstruction and com-munity development projects. Additional environmental featureswere placed behind the curb and within the central median areas andintensified the overall presence of landscape features in these edgeareas. These included

• Raised concrete planters,• Shrubs,• Decorative colored paving stones on the median and boulevard,• Decorative lights,• Decorative noise barriers,• Flowers,• Sculpture,• Entry markers and bollards,• Grass,• Wildflowers, and• Trees.

Accident frequencies and severities before and after constructionfor each road segment were examined. The total incidences of fatal-ities, injuries, and property damage attached to all accidents for aperiod of 3 years before construction and 3 years after construction(simple comparison) were evaluated. The data were reviewed byseparating the information for the midblock sections from that forthe intersections. The concentration was on the results for the mid-block sections because of the possibility that the accidents at theintersections may have been the result of cross-street behavior andresources were not available to differentiate the results. The ambientaccident frequency and severity for the region were increasing at12% annually during the study period.

Findings

The findings from this pilot study indicated that a positive correla-tion exists between landscape improvements to the roadside and areduction in midblock accident frequencies. While nearly all of thetree planting and landscape improvements occurred within the clearzone, midblock accident frequencies decreased between 5% and20%. This raises many questions regarding the clear zone policy thatwas theoretically based on improving the safety of the arterial roadallowance. Certainly, off-road accidents with objects in the clearzone are a major concern. The data generated from the case studyindicate that the presence of a well-defined edge may have a posi-tive effect and may result in an overall decrease in the frequency ofoff-road collisions with obstacles. A major impetus warranting fur-ther research related to clear zone conflicts stems from the desire ofcommunities to have trees and landscaping in the pedestrian realm.

By comparing the costs of these landscape elements (context-sensitive design landscape features) with the cost savings realizedfrom reduced accident frequencies and by using a “willingness-to-pay” formula from the Ontario Ministry of Transportation, it wasconcluded that, for these case study sites, a savings of more than$1,000,000 (in 1995 Canadian dollars, $1 Canada = US$0.71685,August 2003) was accrued within 3 years after the decision to pay$2,500,000 for landscape construction. There is a financial implica-tion between the shared budgets of the safety and health industries:greening of the streets may provide the basis for a cost-effectivehealth and transportation safety policy overall. If savings from reduc-tions in accident frequencies are at the apparent level of significanceindicated by the results of this preliminary investigation, investment

Naderi Paper No. 03-3937 121

(a) (b) (c)

(d) (e) (f)

(g)

(h)

FIGURE 1 Hort Garden adjacent to Hensel Drive, College Station, Texas: (a through f ) photos oflandscaping; (g) section; and (h) site plan.

in landscape is realized in calculable increased safety within rea-sonable budget and political cycles. This may indicate that some ofthe landscaping treatments used for environmental mitigation canfunction as a cost-effective safety measure. After the initial invest-ment is returned, the profits accrue year after year as the accidentreduction benefits continue to occur. This finding indicates that acertain level of importance needs to be ascribed to the developmentof capitalization and life-cycle management strategies for roadsidelandscaping and that this level of importance need to be commen-surate with that ascribed to the management of other aspects of thetransportation corridor. The question in this paper now turns to theissues related to whether the same environmental variables havesimilar effects on improving and encouraging pedestrian activity.

STUDY II: ENVIRONMENTAL VARIABLES ANDPEDESTRIAN ACTIVITY CORRELATION

This section presents a methodology developed to assess the impactof environmental variables on encouraging pedestrian activity. Theresearch in this area is ongoing. It is readily acknowledged that anexpanded database is necessary to support any definitive conclu-sions regarding the impact of environmental variables on walkingbehavior. The intention here is to determine whether the landscapeimprovements associated with traditional roadside environmentalmitigation or context-sensitive design issues are potentially increas-ing pedestrian activity. It is hypothesized that there are environ-mental variables that are more strongly associated with the decisionto walk and that the motivation for the walk may be dependent onthese variables. It is hypothesized that people walking for health andseeking positive health outcomes from their walk may require anaugmented pedestrian landscape to ensure that the desired healthoutcome is achieved. The author is seeking evidence to prove thishypothesis. The methodology and findings from the research to datein the context of the health industry’s economic interest in promotingphysical activity are presented below.

At the community level, researchers in the public health and plan-ning fields have been investigating whether different types of urbanmorphologies are having an impact on decisions related to walkingor car use (4–8, 17–19). For many years, medical researchers havedefined the positive benefits of physical activity on long-term qual-ity of health (20, 21). In addition, medical research is indicating thatthe majority of adults are not engaging in physical activity. This isperceived as the cause of a national health crisis and is promoting ahigh level of research interest at the National Institutes of Health,the Centers for Disease Control and Prevention, and elsewhere todetermine what would stimulate more physical activity (1). Thevariables being investigated include proximity, psychological andsocial issues, as well as physical attributes. At the site level, specific


variables related to level of service for the design of pedestrian facil-ities have emerged from many years of transportation research(22–29). Mathematicians, urban designers, and transportation re-searchers have reviewed the variables related to pedestrian activityand preferences for certain types of walking environments (27,30–35). This work, combined with research from the medical field,is yielding a body of information suggesting an ecological palette ofphysical, social, and psychological characteristics of the walkingenvironment that might encourage walking, particularly alongstreets. It is realized that the result of linking health and safetyresearch evidence to greening initiatives might warrant review ofand necessitate changes to policies and guidelines for landscapingin the transportation corridor’s clear zone. This would require signif-icant numbers of case studies and a robust review of a database muchlarger than that used for the present study. In this regard, it is criticalto first understand the possibility that a link exists between evidence-based environmental variables that encourage walking and that alsohave a positive effect on traffic safety. Given concerns among prac-titioners regarding the sustainable aspect of greening in transporta-tion corridors, sound business practices demand an understandingof the health and safety implications of introducing evidence-basedgreen corridors.

Methodology

This case study review acknowledges that pedestrian and environ-mental design interventions differ depending on the nature of thereason for the walk. The five sites selected included both on-roadand off-road paths with an intention of ensuring that the widest pos-sible range of walking environments was included in the pilot study.The assumption is that certain site-specific variables in the environ-ment encourage certain types of walking behaviors. The questionwas, What is the effect of landscape edge treatments on pedestrianpreferences? Therefore, the initial pilot study first questioned themotivation for the walk and then evaluated reactions to variousattributes of the environment.

Three types of walking behaviors were identified: walking forcommuting purposes, walking for health (curative and preventive),and walking for spiritual or relaxation purposes (36). The case studymethodology facilitates identification of the variables that wouldsupport active living, encouraging walking for pedestrian mobilityas well as for health. If a pedestrian identified a walk as a “good”walk, the attributes of a good walk might be consistent across a num-ber of samples. In this pilot study attributes from a literature review,previous on-site pedestrian surveys, and current standards consid-ered significant in transportation and health community researchwere used. Given the position that people respond to the environ-ment differently, depending on their purpose for walking, the authorconsidered whether the reaction to the variables could be groupedaccording to the purpose for walking. Starting with the environ-mental variables that were consistent with typical aesthetic improve-ments to roads, an expanded list of environmental attributes wasdeveloped as the basis for the field survey of the case study sites. Onthe basis of an extensive literature review of variables identified byboth transportation and health practitioners, a pilot questionnairewas developed and field tested. Subjects were asked to respond tothe questionnaire while in the walking environment. Videos of thesame environments were shown to the same subjects off site in alaboratory setting, and the same questionnaire was administered.Interestingly, the same sites were rated as “good” places in the field

Corridor Length (km)

ROW width (meter)

No. of lanes (bw)

Sidewalk/central median

Average weekday traffic volume (v/day)

Steeles Ave. 1.38 45 3+3 Yes/Yes 43,000 Markham Rd. 2.14 36+ 3+3 Yes/Yes 36,000 Ellesmere Rd. 0.83 42 2+2 Yes/Yes 30,000 Overlea Blvd. 1.80 33.6 2+2 Yes/Yes 24,000 Yonge Street 0.40 36 3+3 Yes/Yes 55,000

NOTE: ROW = right-of-way; bw = both ways; v/day = number of vehicles per day.

TABLE 1 Case Study Sites in Toronto Used for Evaluation ofEffect of Landscape on Traffic Accidents

but as “not good” places when the sites were evaluated by video.This step in the methodology yielded an identification of variablesthat were captured in the next iteration of the questionnaire. In par-ticular, the amorphous quality of “context” or “setting” was identi-fied as an important part of the pedestrian experience lacking in thevideo.

A second questionnaire was presented and asked the subjects toidentify the difference between the walking environment depictedin the video and their experience in the same walking environment inthe field. Certain variables were identified as critical to the experi-ence of “good” walks when the motive for walking was health orspiritual renewal. In this way, certain environmental variables, suchas smell or level of maintenance, were identified as variables thataffected some pedestrians and possibly significant determinants inthe attractiveness of a particular walk. In other words, because theabsence of a factor was noted as a negative factor in the video rep-


resentation, it was determined that the factor should be included asa variable in the on-site questionnaire. These preliminary studieshelped to further develop the variable list and to refine the surveytool; this was the subject of a previous presentation (36 ). A pedes-trian survey was conducted with the revised list of variables. Fifty-two responses have been tabulated at this time. The locations aredepicted in Figures 1 to 6. Descriptions of various aspects of thewalking environments appear in Table 2. People who walk for healthand people who walk for spiritual renewal were of particular inter-est to the research team, so the interviewers concentrated on reach-ing that group. The respondents were approached in the field duringtheir walk and were asked to volunteer to participate in the survey.The survey was then administered in the field or was taken homeby the subject and mailed in later. The respondents were directed tofocus on assessing whether a particular variable experienced on sitewas having a positive or a negative influence on their perception;

(a) (b) (c)

(d) (e) (f)

(g) (h)

FIGURE 2 Tanglewood Park at Carter Creek Parkway, Bryan, Texas: (a through f ) photos of landscaping; (g) site plan; and (h) section.

they also provided an overall rating of the walking environment as“good” or “bad.”

Two of these walking environments were completely withinarterial or collector road rights-of-way: Brothers Pond Park andUniversity Avenue (not shown in the figures). Two of the environ-ments adjoined or shared the rights-of-way within the clear zone,Tanglewood Park and Brothers Pond Park, and two of the facilitiesfunctioned as pedestrian-only zones with limited service vehicleaccess only, West Campus and Northgate.


In all cases, five of the sites mentioned above had the followingcommon characteristics:

1. All were accommodating walking environments within oradjacent to the street.

2. All were being used by people walking for curative orpreventive health care purposes.

3. All included landscape elements typical of the palette of tra-ditional transportation landscapes used in context-sensitive design.

(g)

(a) (b) (c)

(d) (e) (f)

FIGURE 3 Pedestrian district at Northgate, College Station, Texas: (a through f ) photos oflandscaping and (g) site plan.

4. All sites were within the same bioclimatic region of southTexas.

The variables used in the survey were broad and included variablesdescribing proximity as well as social, psychological, motivational,physical, ecological, demographic, and community-based charac-teristics. The respondent could select from a range of responses foreach variable. A typical variable would be “traffic.” The range ofresponses would be “too close,” “acceptable,” and “too far.” Edgecondition could be evaluated by using “well-defined,” “ill-defined,”“natural,” “urban,” or “combination.” Respondents were asked foran overall evaluation of the site at both the beginning and the end ofthe survey to determine whether the detailed examination changedtheir opinion about the place. They were also asked to offer three of their own variables that contributed to a positive quality of


the space as well as three variables that contributed to a negativequality of the space.

The PedLearner program was used to generate a decision tree (37)on the basis of the 52 survey forms. Each survey form had 50 fea-tures. The survey results were as follows: 46.3% reported a “good”environment, 50% reported a “not good” environment, and 3.7%reported “no conclusion.”

The entire data generated were used to build the model illustratedin Figure 7. To interpret the model, the treetop is parsed down indepth in a first-search fashion. For example, if edge of space is con-sidered “well defined” and the availability of seating space is con-sidered “sufficient,” then the environment is “good.” If edge ofspace is considered “natural” and motive is classified as “health,”then the environment is “good.” If edge of space is classified as “notconclusive,” then the goodness of the environment is based on the

(a) (b) (c)

(d) (e) (f)

(g)

(h)

FIGURE 4 Brothers Pond Park at corner of Rio Grande Drive and Brothers Boulevard, College Station,Texas: (a through f ) photos of landscaping; (g) section; and (h) site plan.

initial bias of the walker. The initial bias is what the pedestrianperceived the environment to be before walking.

Findings

Although 50 variables were ultimately used in the survey activity,certain variables were significant in pedestrian preference and deci-sion making regarding the quality of a pedestrian environment. Aninitial review of the data collected, while not conclusive, indicates,


first, that edge of space was the single environmental variable thatsplit the “good” from the “not good” sites for walking. “Edge ofspace” is defined as the sense of enclosure, environmental identity,or genius loci of the spatial sequence. In other words, variables suchas walking with a friend, the presence or absence of a sidewalk, thepresence of trees, or proximity were not as strong as determinants inthis pilot study as edge of space. Second, at the next level of deci-sion, the presence or absence of a well-defined edge combined withadequate seating was the single determining factor in the perceptionof a good walking site. A third finding was that if the edge is

(a) (b) (c)

(d) (e) (f)

(h)

(g)

FIGURE 5 Pedestrian path at Old Main Drive and Olsen Boulevard (West Campus), Texas A&M University,College Station, Texas: (a through f ) photos of landscaping; (g) section; and (h) site plan.

“natural” as opposed to “urban,” the respondents associated thisattribute with a “good” walk if they were specifically motivated towalk for health purposes (Figure 7). On the basis of this pilot studyof environmental variables that affect pedestrian activity, the edgecondition created for a pedestrian environment strongly influencespeople’s conclusion as to whether the space for walking is a “good”


space or not. A well-defined edge is a key decision in the determi-nation of a good pedestrian spatial sequence according to therespondents. Some interesting patterns emerged at a more detailedlevel of data analysis as well. For instance, in a pretrial test regard-ing the distance of the facility from traffic noise, several respondentsrated the walking site along noisy streets as “quiet,” while others

(b)(a) (c)

(e)(d) (f)

(g)

(h)

FIGURE 6 Rio Grande Drive, east side, College Station, Texas: (a through f ) photos of landscaping; (g) section; and (h) site plan.


Site Location Walking Environment; Level of Landscape in Clear Zone

St. Andrews Church, 26th Street, Bryan, Texas

Sidewalk in historic small town CBD along 2-lane local with curb parking both sides, urban cross section; nearly all of the walking environment was within the clear zone; utility poles and overhead wires, fire hydrants, decorative concrete, some sod and trees, decorative lighting and steps

Hort Garden adjacent to Hensel Drive, College Station, Texas

Within 20 meters of and adjacent to 2-lane collector road with minimum land use development, heavily landscaped along edge and within clear zone; granular path, trees, shrubs, wildflowers, benches, winding path, berms

Tanglewood Park at Carter Creek Parkway, Bryan, Texas

Walking circuit facility shared within the clear zone of the adjoining right-of-way; 2-lane local road, urban section; concrete walking path on south side of park along road, mostly treed; formal sidewalk is included in circuit; lots of sodded areas

Pedestrian District at Northgate, College Station, Texas

Urban Plaza accessible from parking lots off local road network; overhead gateways, bollards, decorative concrete, flowers, decorative lighting; no clear zone areas

Brothers Pond Park at corner of Rio Grande Drive and Brothers Blvd., College Station, Texas

Formal fitness track within community park abutting 4- lane collector and 2-lane local road on two sides, some wooded areas, some open on track

Pedestrian path at Old Main Drive and Olsen Blvd., Texas A&M University, College Station, Texas

Pedestrian-only commuter link, leading to and from campus roadway; formalized tree planting with planters, lots of hard surface; no clear zone issues except at entrance area

Rio Grande Drive, east side, College Station, Texas

Standard sidewalk along 4-lane arterial, urban cross section, oversized lanes, open drainage ditch between sidewalk and walled high-end residential property, shrubs, ground covers

Labyrinth Plaza at corner of 4th Street and Market Street, Galveston, Texas

At corner of local bus routes with day care/community center and multifamily housing on two sides, parking adjacent to site; sod, decorative concrete, few trees, signage, sculptural art surface, standard city sidewalk on Market Street, dirt trail/no sidewalk along 4th Street

NOTE: CBD = central business district.

TABLE 2 Characteristics of Walking Environments

Well-defined Ill-defined Natural Urban Combination

EDGE OF SPACE

Availability of Seating Motive for Walk NOT GOOD

ENVIRONMENT

Health

GOOD ENVIRONMENT

All other motivations

NOT GOOD ENVIRONMENT


Not enough OR

Too much

Sufficient

GOOD ENVIRONMENT


FIGURE 7 PedLearner program results for pedestrian spatial survey data from pilot study.

ranked the site as having loud mixed traffic sounds. The site is on anambulance route to the nearby University of Texas Medical campusand is adjacent to a bus stop and traffic light. These people, how-ever, were walking for spiritual renewal and contemplation. In thisinstance it was observed that the motive for the walk appears to havea significant effect on whether the walking environment is consideredgood or not.

CONCLUSIONS

Policy makers, planners, and designers in transportation and land-scape architecture are exploring a variety of innovative solutions totraffic safety and various types of pedestrian mobility. Evidence-based policies and guidelines will require further rigorous researchinto design variables that are difficult to quantify. A much larger setof data from case studies is required. Through several iterations ofattributes and the definition of variables at the site-specific scale, thebeginning of a comprehensive list of variables generated frompedestrian user groups as well as a literature review, research, fieldinterviews, and laboratory work are now available. Some of thesevariables may ultimately turn out to be insignificant; it is hoped thatthese can be determined in the next phase of the work with a muchbroader data set. It is anticipated that the standard landscape ele-ments used in the green infrastructure and in the shaping of thepedestrian environment as well as the edges of travel ways will con-sistently appear to have a more specific relationship with both healthand safety. Ultimately, it will be possible to define the type of edgedevelopment between the travel way and the pedestrian realm thatwill contribute to safety while encouraging physical activity andpositive health outcomes from walking. The findings regarding theimportance of a well-defined edge are encouraging. These are con-sistent with the findings of urban design and architectural theoristsas well as environmental psychologists.

Interestingly, people motivated to walk for health purposes pre-ferred to walk in environments with a natural edge over all otheredge conditions. Preliminary findings indicate that work is neededto examine how various natural treatments on both sides of thepedestrian environment encourage people to engage in health walksboth at the larger community scale and at the smaller site scale. Atthe moment the data from the ongoing research and the resultingpreliminary studies are not sufficient to be used to draw definitiveconclusions regarding site-specific environmental design guide-lines. Further research with simulation and the PedLearner programmay provide an expert learning tool that can be used to design envi-ronments that pedestrians prefer and that can predict more specificpositive health and safety outcomes.

The issues related to transdisciplinary research among landscapearchitects, transportation engineers, and health practitioners hingeon the development of a common language between researchers.

The growing body of knowledge regarding the impact of envi-ronmental design and variables suggests a need for further transdis-ciplinary collaboration. Further inquiry into the development of theexpert learner is required to assist in the development of pedestrianenvironments that help people achieve particular health outcomes.Additional investigation into determining the nature of sequentialpedestrian edge treatments and how to sustain them cost-effectivelyis required before any sustainable consideration of improving thefunctionality of the roadside landscape can be proposed. The trans-portation profession needs to be provided with research evidencethat facilitates a growing sensitivity to the impact of the landscape


on both health and safety paradigms, especially as policies andguidelines for the retrofitting of today’s urban street settings intohealthy communities for aging and increasing urban populations arerethought and reshaped.

ACKNOWLEDGMENTS

Funding for the pedestrian health and safety research portion of thisstudy was provided in part by the Southwest University Trans-portation Center through the Texas Transportation Institute, TexasA&M University, and the Texas A&M University Department ofLandscape Architecture and Urban Planning. The author thanksBarani Raman; Geni Bahar of ItransConsulting, Toronto, Ontario,Canada; Chris Ellis of Texas A&M University; Memory Grober;Ron Hughes of the Highway Safety Research Center, University ofNorth Carolina; Sam Karff of the University of Texas School ofPublic Health, Houston; Byoung-suk Kweon of the Physiology Lab-oratory, Texas A&M University; Harlow Landphair and MichaelManser of the Texas Transportation Institute; Kay Sandor of theSchool of Nursing, University of Texas Medical Branch, Galveston;Vic Sierpina of the Family Practice Unit, University of TexasMedical Branch; and Rui Zhang.

REFERENCES

1. Williamson, D. F. The Prevention of Obesity. New England Journal ofMedicine, Vol. 341, No. 15, Oct. 7, 1999, pp. 1140–1141.

2. Satcher, D. The Virtual Office of the Surgeon General: Surgeon Gen-eral Launches Effort to Develop Action Plan to Combat Overweight,Obesity. www.surgeongeneral.gov/todo/pressreleases/obesitypressre-lease.htm. Accessed Dec. 19, 2001.

3. Rosenblatt, J., and G. Bahar. An Integrated Approach to EnvironmentalImpact Mitigation and Safety Management: Case Studies in the Munic-ipality of Metropolitan Toronto. Proc., XIIIth World Meeting of theInternational Road Federation, Roads/Transportation and the Envi-ronment Session, Toronto, Ontario, Canada, 1998.

4. Sallis, J. F., A. Bauman, and M. Pratt. Environmental and Policy Inter-ventions to Promote Physical Activity. American Journal of PreventiveMedicine, Vol. 15, No. 4, 1998, pp. 379–397.

5. Ball, K., A. Bauman, E. Leslie, and N. Owen. Perceived EnvironmentalAesthetics and Convenience and Company Are Associated with Walk-ing for Exercise Among Australian Adults. Preventative Medicine,Vol. 33, 2001, pp. 434–440.

6. Giles-Corti, B., and R. J. Donovan. The Relative Influence of Individ-ual, Social, and Physical Environment Determinants of Physical Activ-ity. Social Science and Medicine, Vol. 54, 2002, pp. 1793–1812.

7. Handy, S., M. Boarnet, R. Ewing, and R. Killingsworth. How the BuiltEnvironment Affects Physical Activity: Views from Urban Planning.American Journal of Preventative Medicine, Vol. 23 (Suppl. 2), 2002,pp. 64–73.

8. Pikora, T., B. Giles-Corti, F. Bull, K. Jamrozik, and R. Donovan. Devel-oping a Framework for Assessment of the Environmental Determinantsof Walking and Cycling. Social Science & Medicine, Vol. 56, No. 8,2002, pp. 1693–1703.

9. Roadside Design Guide. AASHTO, Washington, D.C., 1988.10. Roadside Design Guide. Task Force for Roadside Safety of the Stand-

ing Committee on Highways Subcommittee on Design, AASHTO,Washington, D.C., 1996.

11. Implementation of the AASHTO Strategic Highway Plan, Initial Draftof a Compendium of Strategies. Project 17-18(3). TRB, NationalResearch Council, Washington, D.C., 2000.

12. Zeigler, A. J. Risk of Vehicle–Tree Accidents and Management ofRoadside Trees. In Transportation Research Record 1127, TRB,National Research Council, Washington, D.C., 1987, pp. 37–43.

13. Zeigler, A. J. Guide to Management of Roadside Trees. FHWA, U.S.Department of Transportation, 1986.

14. Vegetation Control for Safety: A Guide for Street and Highway Main-tenance Personnel. Publication FHWA-RD-90-003. FHWA, U.S.Department of Transportation, 1990.

15. Hunter, W., F. Council, and A. Dutt. Project Selection for RoadsideHazards Elimination. University of North Carolina Highway SafetyResearch Center, Chapel Hill, 1977.

16. Turner, D. S., J. V. Walters, J. M. Hutt, and J. E. Patrick. Survey of StateUtility Manual Clear Zone Provisions. In Transportation ResearchRecord 1233, TRB, National Research Council, Washington, D.C.,1989, pp. 164–177.

17. Cervero, R., and C. Radisch. Travel Choices in Pedestrian Versus Auto-mobile Oriented Neighborhoods. Transport Policy, Vol. 3, No. 3, 1996,pp. 127–141.

18. Southworth, M. Walkable Suburbs? Journal of the American PlanningAssociation, Vol. 63, No. 1, 1997, pp. 28–44.

19. Peponis, J., C. Ross, and M. Rashid. The Structure of Urban Space,Movement and Copresence: The Case of Atlanta. Geoforum, Vol. 28,No. 3–4, 1997, pp. 341–358.

20. A Little Walking Goes a Long Way. Harvard Heart Letter, Vol. 10,No. 4, Dec. 1999, pp. 1–4.

21. Exercise: A Good Health Prescription. Harvard Women’s HealthWatch, Vol. 8, No. 11, July 2001, pp. 6–11.

22. Khisty, C. J. Evaluation of Pedestrian Facilities: Beyond the Level-of-Service Concept. In Transportation Research Record 1438, TRB,National Research Council, Washington, D.C., 1994, pp. 45–50.

23. Mori, M., and H. Tsukaguchi. A New Method for Evaluation of Levelof Service in Pedestrian Facilities. Transportation Research A, Vol. 21,No. 3, 1987, pp. 223–234.

24. Sarkar, S. Determination of Service Levels for Pedestrians, with Euro-pean Examples. In Transportation Research Record 1405, TRB,National Research Council, Washington, D.C., 1993, pp. 35–42.

25. Landis, B. W., V. R. Vattikuti, R. M. Ottenberg, D. S. McLeod, and M.Guttenplan. Modeling the Roadside Walking Environment: PedestrianLevel of Service. In Transportation Research Record: Journal of theTransportation Research Board, No. 1773, National Research Council,Washington, D.C., 2001, pp. 82–88.


26. Seneviratne, P. N., and J. F. Morrall. Level of Service on PedestrianFacilities. Transportation Quarterly, Vol. 39. No. 1, 1985, pp. 109–123.

27. Fruin, J. J. Pedestrian Planning and Design. Metropolitan Associationof Urban Designers and Environmental Planners, Inc., 1971.

28. Fruin, J. J. Designing for Pedestrians: A Level-of-Service Concept. InHighway Research Record 355, HRB, National Research Council,Washington, D.C., 1971, pp. 1–15.

29. Dixon, L. B. Bicycle and Pedestrian Level-of-Service PerformanceMeasures and Standards for Congestion Management Systems. InTransportation Research Record 1538, TRB, National Research Coun-cil, Washington, D.C., 1996, pp. 1–9.

30. Helbing, D., P. Molnar, I. Farkas, and K. Bolay. Self Organizing Pedes-trian Movement. Environment and Planning B: Planning and Design,Vol. 28, 2001, pp. 361–383.

31. Isaacs, R. The Urban Picturesque: An Aesthetic Experience of UrbanPedestrian Places. Journal of Urban Design, Vol. 5, No. 2, 2000, pp. 145–181.

32. Appleyard, D. Motion, Sequence and the City. In The Nature and Art ofMotion (G. Kepes, ed.), George Braziller, Inc., New York, 1965.

33. Bishop, I., J. Wherrett, and D. Miller. Assessment of Path Choices on aCountry Walk Using a Virtual Environment. Landscape and UrbanPlanning, Vol. 52, No. 4, 2001, pp. 225–237.

34. Jacobs, A. Great Streets. MIT Press, Cambridge, Mass., 1993.35. Kaplan, R., S. Kaplan, and R. Ryan. With People in Mind. Island Press,

Washington, D.C., 1998.36. Naderi, J. R., and B. Raman. Design Considerations in Simulating

Pedestrian Environments. Presented at 81st Annual Meeting of theTransportation Research Board, Washington, D.C., 2002.

37. Quinlan, J. R. C4.5. Programs for Machine Learning, San Mateo, Calif.,1993.

Publication of this paper sponsored by Committee on Landscape and Environ-mental Design.

landscape design in clear zone - aggie horticulture article.pdftransportation industry to provide...

Documents