Paper presented at Walk21-X More footprints, less carbon, The 10 th International Conference on Walking in the 21 st Century, October 7-9 2009, New York, USA A Study on the Effects of Urban Landscape on Citizen s Attitude Using Virtual Reality Simulation Model -A Case Study of Utsunomiya City- Akinori Morimoto, Utsunomiya University Hirotaka Koike, Utsunomiya Kyowa University Miki Saito, Utsunomiya University Contact details Department of Civil Engineering Utsunomiya University 7-1-2 Yoto, Utsunomiya, Tochigi, Japan 321-8585 Phone/FAX: +81-28-689-6224 Email: mt086418@cc.utsunomiya-u.ac.jp Abstract In recent years, motorization has accelerated the expansion of urban sprawl and the decline of central city areas in many cities in Japan. In order to reverse this trend, the Government of Japan enacted the Downtown Revitalization Act in June 2007. To revitalize the city centers, it is very important to increase the attractiveness of urban space as a whole, rather than only introducing attractive stores. Creating an attractive urban landscape is indispensable in order to revitalize a central city area. After the Landscape Act was enacted in June 2005, many local governments in Japan have developed urban landscape plans expecting to attract more visitors to their cities. Another important factor in attracting citizens to the central city areas is the necessity to provide accessible traffic environment. With growing concerns for environmental issues
and for rapidly progressing aging society in recent years, Light Rail Transit (LRT) draws strong attention as a barrier-free and environmentally desirable public transportation system. Many cities in the United States and Europe achieved the revitalization of city centers by introducing LRT. Discussions to introduce LRT have recently become very active in many cities in Japan. Utsunomiya City is one of the front runners to introduce a new LRT system aiming to create an attractive and sustainable city. Recently citizens' voices are increasingly reflected in the city planning process, and it is crucial to provide citizens with appropriate information about the future image of the city with LRT. However, it is difficult for ordinary people to imagine what the streets with LRT will look like because they have never seen or ridden such a system. Therefore, it is necessary to develop a tool that effectively helps people to understand the city scene with LRT. In this study, we developed a future urban landscape of downtown Utsunomiya by using virtual reality (VR) based on the animated three-dimensional computer graphics. Although the actual city consists of many landscape objects including dynamic objects like transportation systems and static objects like buildings, sidewalks, roadside trees, etc., VR can reproduce most of these objects in the urban space. The purpose of this study is to analyze effects of future urban landscape on citizens behaviors using VR. Animation images of future city were shown for three points in time, namely, short-term (about three years), mid-term (about five years), and long-term (about ten years). Then, a questionnaire survey was conducted. According to the results of the questionnaire, 64% of the citizens answered that VR was easy to understand and helped to evaluate the city landscape. Furthermore, far more people preferred the long-term city image with LRT and the transit mall (74% in favor) compared to the short-term and mid-term city images with autos and buses. In addition, people who want to visit or live in downtown tend to evaluate the attractiveness of city landscape by considering both dynamic and static objects.
Biography Akinori Morimoto Dr. of Engineering (Waseda University) He is an associate professor of civil engineering in Utsunomiya University. His main research area is integration between land-use and transportation. Other research areas include energy conservation, TOD and compact city. Hirotaka Koike Ph.D. (University of Washington) He is a professor of civil engineering in Utsunomiya Kyowa University. Prior to that, he worked in UBC in Vancouver, Canada, and Utsunomiya University, Japan. His research interests are pedestrian, bicycle, public transport and community development. Miki Saito She is a graduate student in Utsunomiya University. Her research area is virtual reality simulation of urban landscape.
A Study on the Effects of Urban Landscape on Citizen s Attitude Using Virtual Reality Simulation Model -A Case Study of Utsunomiya City- Akinori Morimoto, Utsunomiya University Hirotaka Koike, Utsunomiya Kyowa University Miki Saito, Utsunomiya University Introduction Background and Study Objectives In recent years, motorization has accelerated the expansion of urban sprawl and the decline of central city areas in many cities in Japan. In order to solve this problem, the Government of Japan enacted the Central Urban District Revitalization Act in June 2007. There are wide ranges of measures to revitalize city centers because a city has various functions. Some of them are provision and maintenance of urban facilities, promotion of the downtown living, and revitalization of commercial activities. In addition to these measures, creating an attractive urban landscape is expected to revitalize city centers. To bring people back to the central city areas again, it is very important to preserve the urban space and create an attractive urban landscape. Urban landscape consists of many landscape objects including dynamic objects with people's movements and activities created, and static objects like buildings, sidewalks, roadside trees, etc. Traditionally the focal points of landscape are on static objects such as color or design. There, lively views with dynamic objects had not been considered so much. However, dynamic view created by traffic functions increases the attractiveness of urban landscape. Therefore, it is important to create an overall future urban landscape by taking both static and dynamic objects in the city into consideration. Then, a communication tool is needed to give the general public information about the future urban landscape. Such a tool will assist citizens to understand the entire image of the city visually. In addition, it is better to use a communication tool with animation function because there are static objects and dynamic objects in the actual urban landscape.
In this study, in reference to the landscape planning in Utsunomiya City, we developed the future urban landscape in Utsunomiya City where a light rail transit system (LRT) will be introduced. This communication tool was developed by using virtual reality (VR) based on the animated three dimensional computer graphics. Then, the computer graphics animation obtained from VR was shown to the citizens and a questionnaire survey was conducted. VR can reproduce most of the space components of the urban landscape. Therefore it will be possible to evaluate the landscape of not only a part of area in the city but also the whole city. In addition, it is possible to create the situation that is close to the field of human vision in the city because it can make a viewpoint freely. In recent years, the technology of the CG simulation is used in various cases. It seems that visualization with the CG simulation makes information easier to understand for citizens. This study aims to analyze effects of future urban landscape on citizens behaviors using VR. We focused on the space components of the urban landscape in evaluating the city. Furthermore, we examined the effectiveness of the communication tool that offers both dynamic and static information of the city simultaneously. Review of Previous Researches and Study Design Recently, CG animation and VR are used in a wide field including consensus formation and landscape evaluation. Sakakibara et al. (1993) focused on the analysis of the role of computer graphics using dynamic views and concluded on the effectiveness and necessity of dynamic simulation and gave some suggestions on principles and guidelines for visual simulation. Doug Walker (2007) explained that visualization can serve as a common language for collaboration. However, though there are a lot of researches on the landscape evaluation using visual tools, few studies have been done that focus on the space components or the citizen s attitude. In this study, we developed step-by-step city images as time goes by, and classified the space components into dynamic and static objects. The citizens will easily understand the change of the present situation by providing a process in gradual change of the city. We improved the contents of visual objects, and expanded the reproduction area based on the three-dimensional simulation model that Osada et al. (2004) developed.
Development of VR and summary of the questionnaire survey The study area and description of VR The study area is the central urban area in Utsunomiya city. Utsunomiya city is a regional capital located 100 km north of Tokyo with population of half a million. Utsunomiya City is one of the front runners to introduce a new LRT system aiming to create an attractive and sustainable city. Recently citizens' voices are increasingly reflected in the city planning process, and it is crucial to provide citizens with appropriate information about the future vision of the city with LRT. However, it is difficult for ordinary people to imagine what the streets with LRT will look like because they have never seen or ridden such a system. Therefore, it is necessary to develop a tool that effectively helps people to understand the city scene with LRT. We used UC-win/Road supplied by the Forum 8 Co., Ltd to visualize the future urban landscape. UC-win/Road is a three-dimensional landscape simulator. The map of the study area is shown in Figure 1. The study area is the main street of 1.5 kilometers in the central business district of Utsunomiya city. Major stores and business buildings are concentrated along this street. We reproduced the roadside buildings and the area around. Figure 1 Study area Animation images of the future city were shown for three points in time, namely short-term (about three years), mid-term (about five years), and long-term (about ten years). The dynamic information of the city means moving images of transportation systems being created. Therefore, the dynamic view is defined as traffic of the main street. In the same way, static information in the city is defined as roadside landscape.
Each city image consists of traffic of the main street and roadside landscape and the combination of these two factors are selected based on the feasibility of the transportation system over time. Items that have a high probability of realization were classified as short-term. Each item is chosen based on the landscape planning in Utsunomiya city and plan to introduce LRT. Description of the city images is shown in Table 1. Table 1 Description of the city images We set the city images into four phases over time, because we considered that the future city images are different depending on citizens perceptions. Citizens were provided several choices along the time axis for easy understanding the images of the future city. Development of VR The town management organization (TMO) in Utsunomiya city helped us in the development of VR. TMO is an organization that manages the community development in downtown. Their opinions about the progress of VR were obtained once a month and we reflected their opinions in VR. Present situation To reproduce present situation in 2007, many photographs taken on site are used. For example, the photographs include the side views of buildings and surface patterns of pavements. The traffic volume and signal cycle period are determined from the data of actual traffic survey. The present situation reproduction was a reference basis to evaluate the city images in the future, and to evaluate the level of understanding depending on the degree of VR reproduction.
The future city images The future city images were made for short-term, mid-term and long-term by adding items of the city image shown in Table 2 based on the present image reproduction. As shown in Table 1, the dynamic information means traffic of the main street, and the static information means the roadside landscape. Table 2 Items of each city images Classification of information Dynamic information Items of the future city images Bus lane, Bus Short-term Mid-term Long-term Static information Dynamic information Static information Dynamic information Static information Avoidance of buildings with extreme design or color Removal of arcade Common parking space for delivery of goods Signboard that harmonizes with surrounding Illumination at night Semi transit mall, LRT Bicycle parking facility Urban redevelopment project in Baba Street West area Space for goods delivery along roadway Urban redevelopment project in Senju-Miyajima area Design of LRT station Application of the Banba Citizen Square Turf grass rail track for LRT Full transit mall, LRT Urban redevelopment project in Banba A area Urban redevelopment project in Oote area Barrier-free public spaces Plant design Street café About the long-term city image As an example of the city image, we focus on the long-term city image. CG images and descriptions of the long-term city images are shown in Table 3.
Table 3 CG images and descriptions of the long-term city image Dynamic informationtransportation of the main street Full transit mall LRT is the only transportation in the main street. The other part of the street is used for pedestrian. Static informationroadside landscape Urban redevelopment project in Banba A area Skyline and design upper the second floor of the building are unified. Medium high building blocks with their continuity. Redevelopment based on the landscape planning. Urban redevelopment project in Oote area This building is one big shopping center not as the same as the building group in Banba A area. Barrier-free public spaces The main street area consists of turf grass rail track for LRT and the sidewalk that are unified with barrier-free Plant design Trees and flower planters would be put on the empty spaces of the pavement due to its large pavement space. Street café Setting street café in the Banba citizen space as an application of the open space. Description of questionnaire survey The questionnaire survey was conducted when Utsunomiya Gyoza Festival was held in the city center in 20007. Citizens responded to the questionnaire after watching CG animation video displayed for three minutes on the large-scale screen TV. Three minutes CG animation video was edited from VR output. At first, we showed present image reproduction for 30 seconds. Next, we showed short-term, mid-term, and long-term for 50 seconds each. We made the video in such a way that people can easily distinguish the city images in different time. For instance, in the video editing work, when moving from short-term to mid-term a blank screen image was inserted between
fade-out and fade-in, and the word of "mid-term" was shown in the upper-right corner of the screen while a mid-term city image was being displayed. We thought it helped people to understand the gradual change of the city by providing such a process. Then, the still photo of CG of the city images was prepared, and investigators explained it to the people who did not understand the content of city images through the CG animation only. Still photo of CG image is a supplementary tool in this survey, so actually most people answered the questionnaire without seeing the still photos. CG animation doesn't have voice information, but consists of three viewpoint movements of walking, running, and flying simulation. Each item of the questionnaire did not contain visual information such as CG images, and people are expected to judge only from the sentences in the questionnaire form. Outline of the questionnaire survey is shown in Table 4. Table 4 Outline of the questionnaire survey Subjects Visitors to the event in downtown of Utsunomiya city Survey date November 3, 2007 Number of samples 311 personsincluding 158 citizens in Utsunomiya city Questionnaire items Before showing CG animation Current landscape evaluation (Five level from present landscape is very good to very bad ) Notions of the landscape formation (Five level from "I want to change it" to "I do not want to change it".) Number of times to go to downtown After showing CG animation Evaluation about the information (understandability of CG simulation) Evaluation about the city images (Five level from very good to very bad ) Reasons of evaluation about the city images (multiple selection from the table 2) Residence notions (Where do you want to live now?) Town-visiting notions after showing CG animation (If the city becomes like the future image shown, number of times to visit downtown increases or doesn't increase.) Residence notions after showing CG animation (Four levels from "I want to live" to "I do not want to live" in the future city image.)
Evaluation results of the information and city images Evaluation of the information Evaluation of understandability of the CG animation is shown in Figure 2. According to the results of the questionnaire, 64% of the citizens answered that CG animation was easy to understand, as compared with 8% of the citizens answered that CG animation was hard to understand. Therefore, it is concluded that the CG animation is easy to understand for the citizens, and citizen could understand intuitively and easy to evaluate the city image. Figure 2 Evaluation of understandability of the CG animation Evaluation of the city images Five levels for evaluation of the future city images are shown in Figure 3. According to Figure 3, people with good evaluation ( very good and good ) were more than 65% in all city images. Number of people with neutral answer in mid-term and long-term is smaller compared with short-term. The number of people with bad evaluation ( very bad and bad ) is small. Therefore, it can be said that mid-term and long-term evaluation are more clear than short-term. This result shows that people s evaluation changes by showing the step-by-step city images. The number of people with good evaluations increased gradually as time goes by. Especially, "very good" increased by about 1.6 times from mid-term to long-term. Therefore, it is assumed that there may be a factor that has influence on the improvement of long-term evaluation. Figure 3 Five level evaluation of the city images Numbers in the figure is number of samples
In this survey, the citizens chose from the items of the city image shown in Table 2 by multiple selections as reasons for the city evaluation. We searched for the items that the citizens were paying attention in evaluation of the long-term city image. The top three items listed as the reasons for good evaluation of the long-term are shown in Figure 4. According to Figure 4, about 70% selected full transit mall or barrier-free public space. It is assumed that these items were factors that have influence on the improvement of long-term evaluation. Figure 4 the top three items as reason of good evaluation for the city images (long-term) Relation between citizen s considerations and evaluation reasons for city images In this section, we focused on the space component in the evaluation of the city image and make an analysis about the influence of the future city landscape on the citizen's town-visiting notions and residence notions. In addition, we examine the effectiveness of the communication tool by showing dynamic and static information of the city scene simultaneously. Relation between town-visiting notions and evaluation reason for city image 68% of people answered increase, and 32% answered do not increase for the question that Do you think the number of times visiting town will increase if creating the city central area with the long-term future images shown by CG animation?. Moreover, the difference among these two answered groups in the evaluation reason of the city image was examined and the result was shown in Figure 5. Here, if we assume that people who answered increase are willing to visit town, we found that these people have tendency to choose both dynamic and static information showing in Table 2 as reasons of the city evaluation.
Figure 5 relations between town-visiting notions and evaluation reason for city image Numbers in the figure are number of samples Relation between residence notions and evaluation reason for city image 76% of people answered I want to live, and 24% answered I do not want to live for the question that Do you want to live in the future city like the one shown in CG animation? Moreover, the difference among these two answered groups in the evaluation reason of the city image was examined and the result was shown in Figure 6. Here, if we assume that people who answered I want to live have residence notion, we found that these people have tendency to choose both dynamic and static information showing in Table 2 as reasons of the city evaluation. Figure 6 Relations between residence notions and evaluation reason for city image Numbers in the figure are number of samples People who selected both dynamic and static objects in evaluation of the city image are able to imagine the whole city image. Accordingly, they can assess the city image comprehensively by visualization of the city image. The result was that most of people who want to visit or live in downtown tend to evaluate the attractiveness of city landscape by considering both dynamic and static objects. Furthermore, it is also necessary to consider that providing traffic facilities as dynamic views should be attractive in order to attract people to downtown.
Evaluation reason for city images according to present residence notions Residence notions after showing CG animation We examined residence notions after showing CG animation according to present residence notions. The results were shown in Figure 7. The left side in Figure 7 is the result of people whose present residence notions are downtown. 92% of them want to live in downtown even if the city like the future images (with both I want to live and I would rather live ). It seems that they have a good impression for the image of the future. Also, the right side viewed in Figure 7 is the result of people whose present residence notions are suburbs. The result shows that 52% of them want to live in downtown with the city images like the future ones shown. If the downtown was the city like the future images, it had a possibility that the residence notions will change from the suburb to the urban. Therefore, the use of the CG animation for promotion of the urban residence can be expected. Figure 7 Residence notions after showing CG animation according to present residence notions Relation between residence notions and evaluation reason for city image We focused on the people whose present residence notions are suburb. After showing CG animation, 45 people changed their residence notions from suburb to downtown and 41 people still kept their residence notions without changing. People with the urban residence notions tend to select both dynamic information and static information. The results were shown in Figure 8.
Figure 8 Relation between residence notions and evaluation reason for city image Numbers in the figure are number of samples From the result of analysis above, it is important to show both dynamic and static information together because dynamic information only or static information only will have less impact on town-visiting notions and residence notions. It seems that people usually pay attention to not only appearance of the city but also convenience of life when they think about town-visiting and residence. Therefore, we thought that the city that will be chosen by the citizen should be a place having attractive landscape and comfortable to live in. However, this study is only a specific case but not general one because this CG animation is the one using only the city images of a specific area in the city. So, if the study area is different, the urban landscape will be also different and the result can be changed. It is necessary to make the survey repeatedly through the reproduction of other areas in order to obtain a general theory. In addition, the examination of measures is more necessary for the grasp of residence notions because we did not submit the information about necessary facilities for life. Conclusion In this research, we analyzed effects of future urban landscape using CG animation on citizens behaviors. There are three findings in this study. 1. The CG Animation was evaluated as an easy communication tool to understand. 2. We were able to know changes of the evaluations among people by showing the step-by-step city images over time. 3. Most of people who want to visit or live in downtown tend to evaluate the attractiveness of city landscape by considering both dynamic and static objects. In this research, we found it is better to use a communication tool that has both dynamic and static objects in order to promote citizens` understanding of the future city image.
Moreover, there were also people who changed their residence notions from suburban to urban residence after seeing the CG animation. Therefore, the use of the CG animation for promotion of the urban residence can be expected. Then, it will be necessary to make the CG animation of the future city images better by exchanging opinions among urban planners and citizens. It is concluded that the CG animation is a useful tool to communicate with various types of citizens. References Kazuhiro Sakakibara, Okinori Taniguchi, A Role of Computer Graphics Animation in Dynamic View Design, City Planning Review, No.183, pp.75-83, 1993. Doug Walker, Visualization as a common language for planning, TR NEWS 252 SEPTEMBER-OCTOBER, pp.7-10, 2007 Teppei Osada, Hirotaka Koike, Akinori Morimoto, Development of a consensus building tool to introduce a transit mall,walk21-v Cities for People, 2007.