Emotional mapping and its participatory potential: Opinions about cycling conditions in Reykjavík, Iceland

Cities 61 (2017) 65–73

Contents lists available at ScienceDirect

Cities journal homepage: www.elsevier.com/locate/cities

Emotional mapping and its participatory potential: Opinions about cycling conditions in Reykjavík, Iceland Jiří Pánek a,⁎, Karl Benediktsson b a b

Department of Development Studies, Faculty of Science, Palacký University Olomouc, Czech Republic Department of Geography and Tourism, Faculty of Life and Environmental Sciences, University of Iceland, Reykjavík, Iceland

a r t i c l e

i n f o

Article history: Received 1 June 2016 Received in revised form 15 November 2016 Accepted 20 November 2016 Available online xxxx Keywords: Urban planning Cycling infrastructure Crowdsourcing Participatory mapping

a b s t r a c t Many cities have prioritised the provision of bicycle infrastructure, as part of a transition to more sustainable transport. Information from the users of bicycle facilities is crucial for successful bicycle planning. The article presents a case study of Reykjavík, Iceland, where a simple ‘emotional mapping’ platform was used to enable cyclists to express their emotional reactions to routes and places. A sample of 100 users identified some 541 features lines and points - on a map of the city, associated them with either ‘good’ or ‘bad’ emotions and wrote textual comments to elaborate on the reasons for their judgement. The results indicate clearly the importance of the natural environment for cyclists, as well as the negative feeling engendered by cycling close to car traffic or in the street with the cars. These data support the emphases found in the present bicycling plan of Reykjavík city. In general, volunteered geographical information and crowdsourcing has much potential for increasing citizen participation in urban planning. A flexible software platform for participatory mapping, such as the one used in the study, can be a valuable addition to the planner‘s toolbox. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Sustainable urban transport has become one of the most important planning issues that city governments have to deal with (Banister, 2005). Cycling is seen as a central plank in sustainable transport policies. In recent years, and similar to many other cities around the world (Pucher, Dill, & Handy, 2010; Pucher & Buehler, 2012), the city of Reykjavík, Iceland, has aimed to increase the popularity of cycling as a mode of transport. A cycling policy was adopted in 2010 (Reykjavíkurborg, 2010) that envisaged a network of dedicated bicycle paths traversing the city region, and the adaptation of existing streets to accommodate bicycles. Design guidelines have been drawn up (Erlendsdóttir, 2012) and considerable resources have been allocated to building the infrastructure for cycling. Many different solutions for cycling have been tested in Reykjavík, including fully segregated bike paths, paths shared by cyclists and pedestrians, on-street bicycle lanes, and chevrons that indicate the presence of cyclists on the streets. The policy has been successful: the proportion of trips made by bicycle has grown steadily. According to a survey of travel behaviour in Reykjavík, carried out in October and November 2014, 5.5% of all trips were made by bicycle (Reykjavíkurborg, 2015), an increase from 4.7% in a similar survey in 2011 (Capacent Gallup, 2011). In 2015, a new policy was drafted for the period 2015–2020 (Reykjavíkurborg, 2015). A goal is set for increasing the modal share of bicycling to 6.5% and to ⁎ Corresponding author. E-mail addresses: [email protected] (J. Pánek), [email protected] (K. Benediktsson).

http://dx.doi.org/10.1016/j.cities.2016.11.005 0264-2751/© 2016 Elsevier Ltd. All rights reserved.

have some 8% of all bike routes completely separated from car traffic and pedestrians at the end of the period. The development of a network of separated bike paths is to be continued. However, little concerted effort has been made by the city’s planning authorities to collect information directly from those who make use of the cycling infrastructure – the cycling public. This paper reports the results of a pilot study of ‘emotional mapping’ as a participatory tool for urban planning, with cycling routes in Reykjavík as the empirical example. Apart from yielding valuable information to planners, we argue that such emotional mapping can also be seen as an important way of increasing the level of participation by specific stakeholders or interest groups in urban planning, i.e. as a procedural innovation as much as an instrumental one. The outcome of the pilot study can be seen as a version of a cycling map created by a community. ‘Official’ cycle maps usually only focus on infrastructure, and omit the community’s views and emotional responses (Perkins & Thomson, 2005). More participatory input from the cycling public is desirable. Numerous methods have been developed for increasing public participation in planning. Some of the most interesting relate to the emerging technologies of ‘crowdsourcing’ (Howe, 2006; Brabham, 2009; Seltzer & Mahmoudi, 2012), or ‘volunteered geographical information’ – VGI (Goodchild, 2007; Elwood, Goodchild, & Sui, 2012) coupled with GIS analysis and the presentation of data. Bicycle planning is an area that lends itself well to such methods, as active cycling depends on a certain spatial awareness, and in cycling cultures there is moreover a certain tradition of sharing information about route conditions (Kessler, 2011). Adaptations of conventional transport engineering

66

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

approaches are commonly used in current practice, such as the Bicycle Level of Service (BLOS) index (Landis, Vattikuti, & Brannick, 1997), together with GIS-based analysis (Rybarczyk & Wu, 2010). Such approaches stress the technical parameters of bicycle facilities, but arguably they do not adequately take into account the subjective experience of those who use the facilities. Here, we consider that crowdsourcing GIS methods, together with the collection of subjective and qualitative data, hold much promise. 2. The concept of emotional mapping Emotional mapping enables the display of subjective, qualitative and bottom-up spatial information about the environment in highly hierarchical, quantitative and top-down GIS settings. Emotional maps are often a neglected part of cartography, yet they contain relevant information, especially for urban planners and city administrators. Participatory approaches in mapping and qualitative elements of GIS (sometimes also called GeoParticipation) allow city planners and decision makers to deploy new tools and methods that can collect both qualitative and quantitative data about cities, their dynamics and the people living in them (Kloeckl, Senn, Di Lorenzo, & Ratti, 2011). The basis of emotional mapping is the fact that emotions, spaces and places are very much connected. Emotions are one of the defining characteristics of every human being and yet their presence in maps and spatial data is uncommon (Griffin & Mcquoid, 2012). Several authors (e.g. Reeve, 2014; Russell, 1980; Barrett, 2006) have described emotions as a two-dimensional structure, with the axes being pleasantunpleasant and high arousal-low arousal. Geographers, on the other hand, have described emotions as subjective, relational flows between places and people (Smith, Bondi, & Davidson, 2012), adding a crucial spatial dimension. Every location can evoke an emotion (Mody, Willis, & Kerstein, 2009) and places can be felt to be attractive, boring, dangerous or scary, among other emotions (Korpela, 2002). Emotions provide a strong influence on how the environment is perceived and emotions have an effect on the spatial distribution of the perceptions (Zadra & Clore, 2011). The physical layout of the environment and the built structures affect the emotional perceptions of the place (Hille, 1999; Schmeidler, 2000). For example, this is especially evident when exploring fear of crime (Block & Block, 1995; Sherman, 1995). It would be possible to argue that ‘emotional mapping’ is not the correct term, as it is not exactly ‘emotions’ that are mapped, but merely perceptions or experiences from/with a place. Nevertheless, the authors have decided to continue using the term emotional mapping, mainly based on the argument of Perkins (2009, p. 130), who states that “emotional maps chart human feelings onto a cartographical landscape and allow users to devise and customise their own emotional landscape, choosing what kinds of thoughts or experiences, feelings or passions, to map”. Griffin and Mcquoid (2012) distinguished three categories when talking about maps and emotions. These categories are (1) maps of emotions, (2) using maps to collect emotional data, and (3) emotions while using maps. The case study described in this paper is a combination of the first two categories. Maps were used to collect the emotional information and to visualise the emotional data. Although historically, cartography was mainly focused on representing that which is objective, visible or measurable, and can be mapped (for example, air temperature or wind speed) (Wilson, 2011), critical cartographers have always advocated mapping a space as people experience it, with subjective emotions as well (Pearce, 2008). In the past ten years, several projects have dealt with georeferenced emotions and/or perceptions. Emotional maps have been produced in various fields, such as tourism (Mody et al., 2009), navigation (Huang, Gartner, Klettner, & Schmidt, 2014; Gartner, 2012), urban safety (Salesses, Schechtner, & Hidalgo, 2013; Pánek, Pászto, & Marek, 2017) and city planning (Raslan, Al-Hagla, & Bakr, 2014). Yet, emotional mapping has largely been an academic exercise, as Hauthal and Burghardt

(2016, p. 2) state: “mappers of georeferenced emotions are almost exclusively researchers”. This contrasts somewhat with various community mapping methods that have gained popularity, especially after the development of the Local Agenda 21 Planning Guide created during the United Nations Rio Conference on the Environment in 1992, where community mapping was identified as a best practice for locally-based sustainability planning (ICLEI & IRDC, 1996). Since the Rio conference, many scholars have been engaged in both the theory and the practice of community mapping (Chambers, 2003, 2006; Perkins, 2007; Glöckner, Mkanga, & Ndezi, 2004; Pánek & Vlok, 2013; Forrester & Cinderby, 2012; Elwood, 2002; Craig & Elwood, 1998; Parker, 2006). Nevertheless, it was only recently that a subjective layer (Huang et al., 2014) or the concept of qualitative GIS (Elwood & Cope, 2009) was introduced. In this case study, the authors perceive cyclists to be a specific community with whom to work, in order to understand their preferences and behaviour while using bicycles as a means of transport. The methods used to gather emotional data can be divided into three groups: (1) biometric measurements (Bergner, Zeile, & Papastefanou, 2011; Nold, 2009), (2) extraction from user-generated content such as Twitter, Flickr, Facebook, etc. (Biever, 2010; Bollen, Mao, & Zeng, 2011; Mislove, Lehmann, Ahn, Onnela, & Rosenquist, 2010), and (3) surveys (Huang et al., 2014; MacKerron & Mourato, 2010; Mody et al., 2009). The authors’ approach falls under this last category. More specifically, it is a version of a Computer-Assisted Web Interviewing (CAWI) method, which is in alignment with the concept of participatory planning support system (PPSS) as defined by Kahila and Kyttä (2009). Regarding cyclists specifically, the need to understand emotional responses has also been highlighted in several studies concerning the perception of safety (Møller & Hels, 2008; Lawson, Pakrashi, Ghosh, & Szeto, 2013), as well as studies focused on using cyclists as sensors (Reddy et al., 2009; Reddy et al., 2010).

3. The online mapping tool: technical aspects The data were collected via the crowdsourcing online tool PocitoveMapy.cz, which is designed as web-application based on a Leaflet library. Similar to other web-based tools for crowdsourced mapping, it allows users to collect spatial data on a slippery map background. Unlike Ushahidi, Umap, ArcGIS Online and many others, PocitoveMapy.cz does not require the registration or installation of any specific software, plug-in or virtual server. The application is created as a single-page web application using two main open-source JavaScript libraries; jQuery for basic user interactions and app control and Leaflet, the library for map interactions. The source code of the frontend part is divided into small modules. The crowdsourced data are saved in a MySQL database, which is usually available on every hosting, and therefore there is no need to have specialised hosting or own server with geodatabase installed. Each entry in the database contains a unique user-ID (randomly generated), question identifier, number of points/lines/marked and geometry. These entries are later merged together by GeoJSON PHP library script (Mikola, 2015) which allows GeoJSON to maintain data from multiple users. Furthermore, simple SQL queries can be used to filter data, based on the user ID, type of the question, etc. Up to now, the predominant methods for spatially-explicit preference mapping have been marking points for locations or sketching polygons annotated with expressions of preference (Jankowski, Czepkiewicz, Młodkowski, & Zwoliński, 2015). Brown and Pullar (2012) suggested that points instead of polygons be used in future PPGIS applications, but their study was focused on mapping largescale landscape values. This article presents mapping using two methods that are closely linked to the topic - lines for bicycle routes and points for events/places on the bicycle routes. The application by default allows users to also mark polygons by clicking or by free-hand drawing, but these features were not used in this case study.

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

The original dataset included points and lines that were later merged by the spatial join (see Fig. 1) tool with a hexagonal grid, created by Repeating Shapes for ArcGIS toolbox (Jenness, 2006). Each hexagon has sides of 25 m and a 50 m diameter. The area of each hexagon was 1623.8 m2. The orientation offset of the hexagons was 30 degrees. The hexagonal grid was selected based on the research of Burian, Pászto, and Langrová (2014), where hexagons were identified as an optimal distribution grid. The visualisation of the results was done via a whitered colour ramp for ‘bad’ points and routes and a white-green colour ramp for ‘good’ points and routes. 4. The case study As stated above, the purpose of the survey was to elicit information from bicycle uses about their emotional responses to bike routes in Reykjavík. Several individuals from the bicycle advocacy community in Reykjavík took part in the preparation of the mapping and survey questions. In August 2015, when the online mapping tool was complete, the survey was opened. The interface was bilingual, in English and Icelandic (Fig. 2). A link to the survey was sent to various groups and mailing lists; those related to cycling and other groups. Bicycle clubs published the link on their websites and members were encouraged to respond. Thus the resulting sample can be considered a version of a snowball sample. It is not a random sample of the urban population, but has an obvious bias towards those who are active users of bicycles. Most answers were submitted during the month of August, although a handful of respondents took part during September and October. In total, 101 cyclists filled in the online map-based questionnaire. While it is not a large enough sample to be used for statistical testing for example, we nevertheless believe that it gives a good view of how members of the cycling community in Reykjavík experience conditions in the city. The survey itself was kept as simple as possible. First, the respondent was presented with a map of Reykjavík (OpenStreetMap background) on which he/she was asked to indicate good places and cycling routes, those that had positive emotional connotations. Having indicated each place (point) and route (line) on the map, the respondent could add verbal comments that explained the emotional responses more fully. This was followed by a similar process for bad places and routes. Finally, the respondent answered a few background questions about such things as gender, age, and the nature and frequency of their bicycle use. Respondents were asked to indicate their residence by postcode. Some 94 out of the 101 did so. Of these, 55% (n = 52) were residents of the central part of Reykjavík (west of the river Elliðaár), while 17%

67

(n = 16) lived in the suburbs. The remaining 28% (n = 26) were from municipalities adjacent to the city of Reykjavík, which are an integral part of the Greater Capital Region. The high proportion of answers from central Reykjavík reflects the fact that this is a relatively dense part of the otherwise rather sprawling city (Reynarsson, 1999), and commuter cycling is more practical than in the suburbs and outlying municipalities. This is also where much of the recent cycling infrastructure has been concentrated. Proportionally, many of those who answered the survey were dedicated commuter cyclists: 63% reported that they mainly used their bikes for commuting and three-fifths used the bicycle four or more days each week (Fig. 3). 17% were recreational cyclists and another 15% mainly used their bicycles for physical training. The majority of respondents were male (61%, n = 61), the most typical respondent was a male who commutes by bicycle four or more times a week (n = 29). The second largest group were females who commute four or more times a week (n = 15). Regarding the age distribution, the respondents were between 20 and 69 years old, while the average age was 42.5 and the median age was 43.5. These data reflect the method used to recruit participants. Those who did respond to our invitations were, on the whole, enthusiastic cyclists interested in providing useful data to further improve conditions for cycling. In total, the respondents identified 541 features, divided into 211 points and 330 lines. Many particular points and line segments were identified by several respondents. As can be seen in Table 1, on the whole, more positive features than negative ones were identified. However, there is a difference between these feature classes: the majority of all points identified were for negative features, whereas two-thirds of all lines were associated with positive characteristics. This gives an interesting insight into the cyclists’ perception of the urban environment: parts of routes that are perceived as bad for cycling are often quite localized (sudden discontinuities on an otherwise good route; blind corners; difficult street crossings etc.) whereas good environments consist of longer stretches of bikeways. Most respondents identified only one or two good or bad features, as indicated by the low median number (Table 1), but the most dedicated respondents located many more points and lines on the map. The line features in particular demanded some time from the participants, as they needed to be digitized individually. Often the digitization was not fully accurate, but close enough so there was no doubt as to the route indicated. Most of the features the respondents identified were in the peninsula where the older areas of Reykjavík are located – to the north and west of the two valleys that separate the outer suburbs and adjacent municipalities from the urban core. This was also where most of the respondents came from. Thus we will in the next section present graphical examples only from this part of the urban area and discuss them further. A subsequent section focuses on the qualitative comments. Having presented the main results, we will then proceed to look at how they relate to results of previous research on bicycling conditions. 5. Analysis of map data

Fig. 1. An example of point and lines spatially joined within the hexagonal grid. The darker the green colour, the more points and lines fall within a hexagon. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

On the generalised map of the respondents’ assessments (Fig. 4), one can see that several stretches are universally considered ‘good’. It is interesting to note that all these routes have been developed, in accordance with the cycling plan, to be two-way bike paths fully separated from both car traffic and pedestrians. The southernmost route, through the Fossvogsdalur valley and along the northern coast of Skerjafjörður, was completed some years ago. It is an environmentally attractive route, and also has a pedestrian path that is much used by joggers and leisure walkers (a on Figs. 4 and 5). For cyclists it offers not only a pleasant environment, but also a reasonably direct route between the southeastern suburbs and the city centre. A new ‘bicycle highway’ that has become an important artery for bicycle commuter traffic between the old city centre and the eastern parts receives positive reaction (b on

68

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

Fig. 2. The user interface of the emotional mapping tool.

Figs. 4 and 5). The northernmost ‘good’ route is the new bicycle path along the coast, with views of the bay, islands and mountains to the north (c on Figs. 4 and 5). Interestingly though, many respondents indicated a ‘bad’ spot in the middle of this route, which is adjacent to a prominent seaside sculpture very popular with tourists. Here the segregation from pedestrians is interrupted and cyclists have to slow down and be aware of the numerous walkers. Turning to the ‘bad’ sections and places, some stretches also stand out. Many of those are in the western part of Reykjavík, where specific infrastructure for bicycles has not been put in place. For instance, the western part of Hringbraut, a four-lane street with dense and rather fast traffic and no special provisions for cyclists, is universally considered unpleasant for cycling (d on Figs. 4 and 5). The same can be said about Mýrargata by the harbour, which has heavy traffic and is a key

link between the city centre and the western part. Various other segments are perceived negatively, mostly those where bicycle routes are in close proximity to thoroughfares with heavy traffic, or where specific infrastructure for bikes is completely lacking. One street is particularly interesting because of the mixed emotions it engenders. This is Borgartún, a street east of the centre, where bicycle lanes were put in as part of a recent redevelopment of the road space. While there are some who think it is a good piece of infrastructure, there are rather more cyclists who come out on the negative side. One reason is that the bike lanes are merged with the car lanes at roundabouts (e on Figs. 4 and 5). This is confusing for many cyclists and is considered dangerous by some. Gender or age does not make any difference in the evaluation of this stretch, nor of others analysed within this pilot study.

Fig. 3. Main characteristics of bicycle use by respondents.

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73 Table 1 Features identified by respondents. Good bike route or spot

Bad bike route or spot

318 95 223 22

223 116 107 17

3.15

2.21

2

1

Number of features Number of points Number of lines Maximum number of features marked per person Average number of features marked per person Median number of features marked per person

6. Analysis of qualitative comments The comments from the respondents provide further useful clues as to what exactly engenders positive or negative emotions and opinions. These comments were read carefully and repeatedly and several common themes were identified. We will now discuss the major themes that emerged, with selected examples from the comments. The positive comments relate both to the cycling facility itself, its routing and surface, and to the environment beyond. First, it is clear that cyclists greatly appreciate being separated from both car and pedestrian traffic. A few examples of the comments that indicate this (sometimes along with other things that are deemed positive) are: • “Comfortable and safe route in the city centre with a separate bike path.” • “Separated bike path, beautiful environment.” • “Separation of bikers and walkers. Good overviews. Part of a connection that provides a shortcut, gives an advantage over cars. No crossings. Good view of the road. Nature.” The river valley that separates the peninsula from the outer suburbs received the following comment: • “The Elliðaár Valley is good for cycling if you are not in a hurry. The mixed paths are popular for walking so bikers need to be especially careful with regard to the walkers.”

69

This valley has been kept free of urban development. It has been extensively planted with trees and it is now much used for recreational purposes, but is part of a regular commuting route for some cyclists coming from the southeastern suburbs to the city centre. The green environment is much appreciated, while the drawbacks for commuters of sharing the paths with recreational walkers are acknowledged. Indeed, the positive emotions that originate from the diverse experiences of nature that a route may afford the biker are often evident. This includes visual quality – good views are much appreciated. The aforementioned coastal routes to the north and south of the peninsula are often commented upon favourably (a on Figs. 4 and 5): • “Magnificent view, invigorating breeze, and the separation of cyclists and walkers has started.” • “Very beautiful route – views of the sea, birds etc.” As the comments indicate, the appreciation of nature goes beyond the visual element in the strict sense to include various other aspects: • “Far from traffic, babbling brook, very calming environment and rather flat.” • “One of the most enjoyable bicycle routes in Reykjavík. Diversity, good paths, good connections to nature.” Some positive comments relate to the planning of the route itself, its surface and other qualities, in addition to the adjacent environment: • “Wonderfully gentle slope, shielded from traffic.” • “This is a great route. Rather flat and the asphalt on the paths is good.” Turning to the negatively evaluated spots or route sections, the comments associated with these also reveal several distinct themes, relating to perceived lack of safety under certain conditions as well as to the state of the paths. A prominent theme is the perceived incompatibility of cycling and motorised traffic: • “Too much traffic, noise and pollution.” • “Heavy traffic and dangerous in places.” • “Uncomfortably heavy traffic, considering the width of the road.”

Fig. 4. Generalised map of ‘good’ and ‘bad’ features (lines and points). The letters (a, b, c, d, e) refer to locations described in the text and shown in photographs on Fig. 5.

70

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

Fig. 5. Conditions at selected locations, shown on Fig. 4 and discussed in the text. Photographers: Páll Guðjónsson (a) and Karl Benediktsson (b, c, d, e).

The aforementioned newly redesigned Borgartún street, where bicycle lanes were introduced but they do not provide full separation between bicycles and cars, engendered some quite negative comments (d on Figs. 4 and 5): • “A badly designed revamp where bikers are repeatedly forced off the bike path and into the car traffic in the roundabouts. New lamp posts only serve the cars on the street, not the bike path. Far too many crossings where the cars come out from the parking lots.” The aesthetic and other environmental characteristics of cardominated areas receive some negative comments: • “Ugly environment that is hard to negotiate, designed for cars.” What is clearly called for is the separation of bicycle paths from car traffic, to the extent possible. The general top speed for cars in Reykjavík is 50 km/h, but outside the main multi-lane arteries it is higher, up to 80 km/h in some instances. In many residential streets however, the speed limit has been further lowered to 30 km/h, with traffic-calming measures such as speed bumps. In the densest central parts of the city, traffic is generally slow. Interestingly however, this is not automatically deemed a suitable environment for cycling: • “The Þingholt area as a whole is difficult for cyclists, mainly because you are always really close to the cars. The streets are also uneven and full of patched potholes, so you often lose speed.”

Finally, and similar to the last remark, a number of negative comments relate to the state of the bike paths themselves: • “Awful paving, wreaks havoc on the bicycles.” • “Walking path and bicycle path badly indicated, painting has faded.” • “This section is completely hopeless, on an otherwise fine route. A messed-up sidewalk.” • “The pavement needs fixing and preferably the separation of cyclists and walkers. The present paving is showing its age and is not comfortable for cycling.” • “Too many blind turns.” All in all, the comments add considerably to the understanding of the subjective emotions attached to certain infrastructures or locations. As the examples show, brand-new infrastructure is no guarantee of a positive evaluation. There is ample capacity for urban planners and designers to learn from such focused comments on particular infrastructural solutions. 7. Lessons learned The pilot study has taught us several different lessons. They relate not only to the issue that was at the heart of the case study – the emotional responses of cyclists to the environments and infrastructures they encounter while cycling – but also to the technological aspects of the

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

project, and last but not least, to the wider potential of similar approaches and their limitations. With regard to the first lesson, two aspects stand out in our data: the positive contribution of the natural environment to the emotional experience of the cyclist, and the overwhelmingly negative feelings engendered by cycling close to motorised transport or sharing space with it. Having a good view of the sea and mountains is mentioned by many respondents as a positive factor, as well as cycling in a lush, forested environment. This is in line with other research that has stressed the importance attached to aesthetic experiences by cyclists (Stefánsdóttir, 2014). On the other hand, cyclists are very negative about using street-space designed for cars. This is interesting in the light of the long-standing dispute about the promotion of ‘vehicular cycling’ on the streets versus the installation of separate cycling infrastructures (for a review, see Furness, 2010). The lesson to be learned by city planners from our respondents is that the latter policy may provide better results in terms of an increase in the number of cyclists. Research elsewhere (see summaries by Heinen, van Wee, & Maat, 2010, and Pucher et al., 2010) has indeed come down in firm support of separating cyclists from motorised traffic to the extent possible. This is not least important for less experienced commuters and has played a crucial role in increasing the modal share of cycling (Pucher et al., 2010). Turning to the technological aspects, after hours of technological struggle and testing various possibilities, we learnt that the optimal approach is to combine various feature types into one dataset via a regular grid network using the spatial join function in ArcGIS 10.3. In previous tests we often worked with interpolation methods or the heat-map function. Nevertheless, the interpolation approach blurs the data and does not allow for the clear identification of hot-spots. Although Brown and Pullar (2012) suggested using points as basic input data features for creating and analysing perception maps, we have proven that in specific cases, such as cycling routes, lines can be a very valuable source of information that should not be ignored. Some of the users were confused by the classic GIS-method for creating a line; clicking on nodes and forming a line. With this approach, that is not natural for line drawing for non-GIS users, we found that some bike lanes seemed to cross the sea or go through buildings, because the survey respondents were not careful enough in their line creations. For further use with line features that should represent the movement of people or matter, we suggest using either a free-hand line drawing tool or some kind of line snapping (ideally to the road network), and that would prevent the creation of cycling lanes that cross the sea or go through buildings. Another technological learning point was that users preferred different feature types for different perceptions. Lines were dominantly used for good bikes routes, while points were more often used to mark bad places related to their cycling experience in Reykjavík. This finding will most likely not affect the variety of feature types offered for mapping in future deployments, but it definitely opens up another research question for further investigation – if and how the question asked affects the feature type used for mapping. 8. Conclusion The case study presented in this paper aimed to capture the reactions of cyclists in Reykjavík to the various routes and places that they encounter. As mentioned in the introduction, the city is currently carrying out an ambitious plan for increasing bicycling (Reykjavíkurborg, 2015). Crucially, the development of a network of separated bike paths is to be continued. Our emotional mapping exercise confirms that this is an important part of a successful cycling policy. Much of the recently installed biking infrastructure is of this kind, indicating that planners have already recognised the advantages of separation wherever this is possible. It is also noteworthy that the new plan includes improvements of some of the stretches that came out particularly badly in our pilot study. But, in some instances, recent infrastructural

71

experiments have not been judged positively by the users. We argue that the tool presented here, by enabling a quick and easy gauging of emotional reactions, could be a valuable practical addition to the transport planners’ toolbox. Much research has considered transport, including bicycling, from a narrowly instrumental point of view. Our results confirm what has been amply documented in bicycling research (Forsyth & Krizek, 2011; Stefánsdóttir, 2014); that cyclists appreciate their environment on many different levels, and not least the aesthetic level. They are not only attentive to the instrumental qualities of the route taken, but also aware of the diversity of emotional experiences that the route may afford. Thus bicycling in itself becomes a meaningful activity (Spinney, 2009), the emotional and affective dimensions of which need to be better understood for the successful planning of infrastructure (Gatersleben & Uzzell, 2007). New bicycling routes and on-street facilities should be located so that they afford the maximum environmental quality to their users (Stefánsdóttir, 2014): transportation planning is about much more than maximum efficiency in getting people from A to B. The authors wanted to demonstrate both the technical aspects of the tool as well as a general approach to ‘emotional mapping’ that can be modified and deployed in various situations as well as locations. As Perkins (2009, p. 126) asserts, “[m]aps are now mostly mirror-like representations of the real, a factual tool to help us to navigate, plan, and control the world out there. The scientific maps produced by cartographers were the maps that mattered, and these were taken for granted; all other mapping was ignored or demonized”. Emotions and perceptions are nevertheless, a crucial part of modern cartography and a relevant source of information for city planners and decision-making stakeholders. Participatory planning support systems, such as the one presented in this case study, can create a link between the administration and the users of the services of the administration – the public. As almost every place can evoke an emotional response, it is only logical that maps are used to collect these emotional reactions. Maps also help to communicate the perceptions and preferences in a space and with the tools of ‘neocartography’ and web-mapping, they are could become a new platform for participatory planning. With the democratisation of cartography as well as the Internet and the rise of public participation, maps are becoming essential tools of communication between citizens and local authorities. Geographical crowdsourcing, we argue, has tremendous potential to be of practical value to urban planners. Participatory planning support systems, designed ad-hoc for a particular situation, group or issue such as the one presented in this paper, can bridge the gap between planners and the public; those whom the planning is intended to serve. Acknowledgements Sesselja Traustadóttir, Morten Lange and Harpa Stefánsdóttir are thanked for valuable input during the planning of the survey. Páll Guðjónsson receives thanks for allowing us to use one of his photos. Jiři Pánek thanks for financial support of his mobility to the University of Iceland through the EEA Grants scheme. This research did not receive any specific grant from funding agencies in the public, commercial, or non-profit sectors. References Banister, D. (2005). Unsustainable transport: City transport in the new century. London: Routledge. Barrett, L. F. (2006). Solving the emotion paradox: Categorization and the experience of emotion. Personality and Social Psychology Review, 10(1), 20–46. http://dx.doi.org/ 10.1207/s15327957pspr1001_2. Bergner, B. S., Zeile, P., & Papastefanou, G. (2011). Emotional barrier-GIS – A new approach to integrate barrier-free planning in urban planning processes. Proceedings REAL CORP (pp. 247–257) (http://realcorp.at/archive/CORP2011_27.pdf). Biever, C. (2010). Twitter mood maps reveal emotional states of America. New Scientist, 207(2771), 14. http://dx.doi.org/10.1016/S0262-4079(10)61833-7.

72

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73

Block, R. L., & Block, C. R. (1995). Space, place and crime: Hot spot areas and hot places of liquor-related crime. In J. Eck, & D. Weisburd (Eds.), Crime and place (pp. 145–183). New York: Criminal Justice Press. Bollen, J., Mao, H., & Zeng, X. (2011). Twitter mood predicts the stock market. Journal of Computational Science, 2(1), 1–8. http://dx.doi.org/10.1016/j.jocs.2010.12.007. Brabham, D. C. (2009). Crowdsourcing the public participation process for planning projects. Planning Theory, 8(3), 242–262. http://dx.doi.org/10.1177/1473095209104824. Brown, G. G., & Pullar, D. V. (2012). An evaluation of the use of points versus polygons in public participation geographic information systems using quasi-experimental design and Monte Carlo simulation. International Journal of Geographical Information Science, 26(2), 231–246. Burian, J., Pászto, V., & Langrová, B. (2014). Possibilities of the definition of city boundaries in GIS – The case study of a medium-sized city. 14th SGEM GeoConference on Informatics, Geoinformatics and Remote Sensing. Vol. 3. (pp. 777–784). http://dx.doi.org/10. 5593/SGEM2014/B23/S11.099. Capacent Gallup (2011). Ferðir íbúa höfuðborgarsvæðisins [Personal travel in the capital area]. Reykjavík: Capacent Gallup. Chambers, R. (2003). Whose reality counts? Putting the first last. London: ITDG Publications. Chambers, R. (2006). Participatory mapping and geographic information systems: Whose map? Who is empowered and who disempowered? Who gains and who loses? The Electronic Journal of Information Systems in Developing Countries, 25(2), 1–11. Craig, W. J., & Elwood, S. (1998). How and why community groups use maps and geographic information. Cartography and Geographic Information Science, 25(2), 95–104. http://dx.doi.org/10.1559/152304098782594616. Elwood, S. (2002). GIS use in community planning: A multidimensional analysis of empowerment. Environment & Planning A, 34(5), 905–922. http://dx.doi.org/10.1068/ a34117. Elwood, S., & Cope, M. (2009). Qualitative GIS: A mixed methods approach. London: SAGE. Elwood, S., Goodchild, M. F., & Sui, D. Z. (2012). Researching volunteered geographic information: Spatial data, geographic research, and new social practice. Annals of the Association of American Geographers, 102(3), 571–590. http://dx.doi.org/10.1080/ 00045608.2011.595657. Erlendsdóttir, G. (2012). Hönnun fyrir reiðhjól – Leiðbeiningar [Design for bicycles – Guidelines]. Reykjavík: Efla. http://eldri.reykjavik.is/Portaldata/1/Resources/umhverfissvid/ myndir/skyrlsur/Leidbeiningar_honnun_hjol.pdf (Accessed 29 May 2016) Forrester, J., & Cinderby, S. (2012). A guide to using community mapping and participatory GIS. http://www.tweedforum.org/research/Borderlands_Community_Mapping_Guide_. pdf Forsyth, A., & Krizek, K. (2011). Urban design: Is there a distinctive view from the bicycle? Journal of Urban Design, 16(4), 531–549. http://dx.doi.org/10.1080/13574809.2011. 586239. Furness, Z. (2010). One less car: Bicycling and the politics of automobility. Philadelphia: Temple University Press. Gartner, G. (2012). Putting emotions in maps – The wayfinding example. Mapping mountain dynamics: From glaciers to volcanoes (pp. 61–65)Proceedings of the 8th Mountain Cartography Workshop in Taurewa, New Zealand. http://www. mountaincartography.org/publications/papers/papers_taurewa_12/papers/ mcw2012_sec3_ch08_p061-065_gartner.pdf Gatersleben, B., & Uzzell, D. (2007). Affective appraisals of the daily commute: Comparing perceptions of drivers, cyclists, walkers, and users of public transport. Environment and Behavior, 39(3), 416–431. http://dx.doi.org/10.1177/0013916506294032. Glöckner, H., Mkanga, M., & Ndezi, T. (2004). Local empowerment through community mapping for water and sanitation in Dar Es Salaam. Environment and Urbanization, 16(1), 185–198. http://dx.doi.org/10.1177/095624780401600115. Goodchild, M. (2007). Citizens as sensors: The world of volunteered geography. GeoJournal, 69, 211–221. http://dx.doi.org/10.1007/s10708-007-9111-y. Griffin, A. L., & Mcquoid, J. (2012). At the intersection of maps and emotion: The challenge of spatially representing experience. Kartographische Nachrichten, 6, 291–299. Hauthal, E., & Burghardt, D. (2016). Mapping space-related emotions out of usergenerated photo metadata considering grammatical issues. The Cartographic Journal (published online 16 February) http://dx.doi.org/10.1179/1743277414Y.0000000094 Heinen, E., van Wee, B., & Maat, K. (2010). Commuting by bicycle: An overview of the literature. Transport Reviews, 30(1), 59–96. Hille, K. (1999). “Gendered Exclusions”: women’s fear of violence and changing relations to space. Geografiska Annaler, Series B: Human Geography, 81(2), 111–124. http://dx. doi.org/10.1111/j.0435-3684.1999.00052.x. Howe, J. (2006). The rise of crowdsourcing. Wired Magazine, 14(6), 1–4. Huang, H., Gartner, G., Klettner, S., & Schmidt, M. (2014). Considering affective responses towards environments for enhancing location based services. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 40(4), 93–96. http://dx.doi.org/10.5194/isprsarchives-XL-4-93-2014. ICLEI & IRDC (1996). The local agenda 21 planning guide: An introduction to sustainable development planning. Toronto: International Council for Local Environmental Initiatives (ICLEI), International Development Research Centre (IDRC), United Nations Environment Programme (UNEP). Jankowski, P., Czepkiewicz, M., Młodkowski, M., & Zwoliński, Z. (2015). Geo-questionnaire: A method and tool for public preference elicitation in land use planning. Transactions in GIS. http://dx.doi.org/10.1111/tgis.12191. Jenness, J. (2006). Repeating shapes for ArcGIS. Jenness Enterprises. http://www. jennessent.com/arcgis/repeat_shapes.htm (Accessed 30 May 2016) Kahila, M., & Kyttä, M. (2009). SoftGIS as a bridge-builder in collaborative urban planning. Planning support systems: Best practice and new methods (pp. 389–411). Netherlands: Springer. Kloeckl, K., Senn, O., Di Lorenzo, G., & Ratti, C. (2011). LIVE Singapore! – An urban platform for real-time data to program the city12th International Conference on Computers in

Urban Planning and Urban Management, Lake Louise, Canada, July 5–8, 2011 http://senseable.mit.edu/papers/pdf/20110708_Kloeckl_etal_LiveSingapore_ ComputersUrban.pdf (Accessed 23 April 2016) Kessler, F. (2011). Volunteered geographic information: A bicycling enthusiast perspective. Cartography and Geographic Information Science, 38(3), 258–268. http://dx.doi. org/10.1559/15230406382258. Korpela, K. (2002). Children's environment. In R. B. Bechtel, & A. Churchman (Eds.), Handbook of environmental psychology (pp. 363–373). New York: Wiley. Landis, B., Vattikuti, V., & Brannick, M. (1997). Real-time human perceptions: Toward a bicycle level of service. Transportation Research Record: Journal of the Transportation Research Board, 1578, 119–126. Lawson, A. R., Pakrashi, V., Ghosh, B., & Szeto, W. Y. (2013). Perception of safety of cyclists in Dublin city. Accident Analysis and Prevention, 50, 499–511. http://dx.doi.org/10. 1016/j.aap.2012.05.029. MacKerron, G., & Mourato, S. (2010). Mappiness. http://www.mappiness.org.uk/ (Accessed 23 April 2016) Mikola, J. (2015). GeoJSON implementation for PHP. https://github.com/jmikola/geojson (Accessed 2 September 2016) Mislove, A., Lehmann, S., Ahn, Y. -Y., Onnela, J. -P., & Rosenquist, J. N. (2010). Pulse of the nation: U.S. mood throughout the day inferred from Twitter. http://www.ccs.neu. edu/home/amislove/twittermood/ (Accessed 23 April 2016) Mody, R. N., Willis, K. S., & Kerstein, R. (2009). WiMo: Location-based emotion tagging. Proceedings of the 8th International Conference on Mobile and Ubiquitous Multimedia. 14. Møller, M., & Hels, T. (2008). Cyclists' perception of risk in roundabouts. Accident Analysis and Prevention, 40(3), 1055–1062. http://dx.doi.org/10.1016/j.aap.2007.10.013. Nold, C. (2009). Emotional cartography: Technologies of the self. (WWW.EMOTIONALCARTO GRAPHY.NET) http://emotionalcartography.net/EmotionalCartography.pdf (Accessed 10 April 2016) Pánek, J., Pászto, V., & Marek, L. (2017). Mapping emotions: Spatial distribution of safety perception in the city of Olomouc. In I. Ivan, A. Singleton, J. Horák, & T. Inspektor (Eds.), Lecture notes in geoinformation and cartography: The rise of big spatial data (pp. 211–224). Ostrava, Czech Republic: Springer International Publishing. http:// dx.doi.org/10.1007/978-3-319-45123-7. Pánek, J., & Vlok, C. (2013). Participatory mapping as a tool for community empowerment – A case study of community engagement in Koffiekraal, South Africa. In M. F. Buchroithner (Ed.), 26th International Cartographic Conference (Dresden) http:// icaci.org/files/documents/ICC_proceedings/ICC2013/_extendedAbstract/969_ abstract.pdf Parker, B. (2006). Constructing community through maps? Power and praxis in community mapping. The Professional Geographer, 58(4), 470–484. http://dx.doi.org/10. 1111/j.1467-9272.2006.00583.x. Pearce, M. W. (2008). Framing the days: Place and narrative in cartography. Cartography and Geographic Information Science, 35(1), 17–32. http://dx.doi.org/10.1559/ 152304008783475661. Perkins, C. (2007). Community mapping. The Cartographic Journal, 44(2), 127–137. http:// dx.doi.org/10.1179/000870407X213440. Perkins, C. (2009). Performative and embodied mapping. International Encyclopedia of human geography (pp. 126–132). Oxford: Elsevier. Perkins, C., & Thomson, A. Z. (2005). Mapping for health: Cycling and walking maps of the city. North West Geography, 5(1), 16–23. Pucher, J., & Buehler, R. (Eds.). (2012). City cycling. Cambridge MA: MIT Press. Pucher, J., Dill, J., & Handy, S. (2010). Infrastructure, programs, and policies to increase bicycling: An international review. Preventive Medicine, 50, S106–S125. http://dx.doi. org/10.1016/j.ypmed.2009.07.028. Raslan, R., Al-Hagla, K., & Bakr, A. (2014). Integration of emotional behavioural layer ‘EmoBeL’ in city planning. REAL CORP 2014. 8. (pp. 309–317). Reddy, S., Shilton, K., Burke, J., Estrin, D., Hansen, M., & Srivastava, M. (2009). Using context annotated mobility profiles to recruit data collectors in participatory sensing. In J. Hightower, B. Schiele, & T. Strang (Eds.), Location- and context-awareness (pp. 52–69). Berlin & Heidelberg: Springer. Reddy, S., Shilton, K., Denisov, G., Cenizal, C., Estrin, D., & Srivastava, M. (2010). Biketastic: Sensing and mapping for better biking. Proceedings of the SIGCHI conference on human factors in computing systems (pp. 1817–1820). New York: Association for Computing Machinery. Reeve, J. (2014). Understanding motivation and emotion. Hoboken: John Wiley & Sons. Reykjavíkurborg (2010). Hjólaborgin Reykjavík [Reykjavík Bicycle City]. http://reykjavik. is/sites/default/files/skjol_thjonustulysingar/hjolreidaa__tlun_LOW_OK.pdf (Accessed 29 May 2016) Reykjavíkurborg (2015). Hjólreiðaáætlun Reykjavíkur 2015–2020 [Bicycling Plan for Reykjavík 2015–2020]. http://reykjavik.is/sites/default/files/skjol_thjonustulysingar/ hjolreidaaaetlun_reykjavikur_2015-2020.pdf (Accessed 29 May 2016) Reynarsson, B. (1999). The planning of Reykjavik, Iceland: Three ideological waves–A historical overview. Planning Perspectives, 14(1), 49–67. http://dx.doi.org/10.1080/ 026654399364346. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. http://dx.doi.org/10.1037/h0077714. Rybarczyk, G., & Wu, C. (2010). Bicycle facility planning using GIS and multi-criteria decision analysis. Applied Geography, 30(2), 282–293. http://dx.doi.org/10.1016/j.apgeog. 2009.08.005. Salesses, P., Schechtner, K., & Hidalgo, C. A. (2013). The collaborative image of the city: Mapping the inequality of urban perception. PloS One, 8(7), e68400. http://dx.doi. org/10.1371/journal.pone.0068400. Seltzer, E., & Mahmoudi, D. (2012). Citizen participation, open innovation, and crowdsourcing: Challenges and opportunities for planning. Journal of Planning Literature, 28(1), 3–18. http://dx.doi.org/10.1177/0885412212469112.

J. Pánek, K. Benediktsson / Cities 61 (2017) 65–73 Sherman, L. W. (1995). Hot spots of crime and criminal careers of places. In J. Eck, & D. Weisburd (Eds.), Crime and place (pp. 35–52). New York: Criminal Justice Press. Schmeidler, K. (2000). Prostředí města, urbanistická tvorba a sociální patologie. Kriminalistika, 33, 316–325. Smith, M., Bondi, L., & Davidson, J. (2012). Emotional geographies. Farnham: Ashgate. Spinney, J. (2009). Cycling the city: Movement, meaning and method. Geography Compass, 3(2), 817–835. http://dx.doi.org/10.1111/j.1749-8198.2008.00211.x. Stefánsdóttir, H. (2014). Pleasurable cycling to work. Urban spaces and the aesthetic experiences of commuting cyclists. Norwegian University of Life Sciences (PhD thesis).

73

Wilson, M. W. (2011). ‘Training the eye’: Formation of the geocoding subject. Social & Cultural Geography, 12(4), 357–376. http://dx.doi.org/10.1080/14649365.2010. 521856. Zadra, J. R., & Clore, G. L. (2011). Emotion and perception: The role of affective information. Wiley Interdisciplinary Reviews: Cognitive Science, 2(6), 676–685.