«Patterns of sustainable mobility and the structure of modality in the Randstad city-region Jorge GIL, Stephen READ Department of Urbanism, Faculty of ...»
VOL: 11, NO: 2, 231-254, 2014-2
Patterns of sustainable mobility and the structure
of modality in the Randstad city-region
Jorge GIL, Stephen READ
Department of Urbanism, Faculty of Architecture, TU Delft, UCL, Delft,
Received: December 2013 Final Acceptance: April 2014
The sustainable mobility vision for city-regions proposes a more integrated and
‘seamless’ multi-modal public transport system around quality neighborhoods, shifting mobility to soft transportation modes and to public transport at various scales. Existing models of sustainable urban form address this challenge focusing on the location, density and diversity of activities, on the composition of the street layout, and on the presence of transport nodes and the quality of the public transport service. In order to better understand the relation between urban form and sustainable mobility patterns we propose to additionally measure the structure of mobility networks, including network proximity, density and accessibility, for different transport modes. The analysis of a multi-modal network model of the Randstad region in the Netherlands, integrating private and public transport infrastructure networks and land use information, reveals the structures of modality in the city-region. These structures are used to identify a typology of ‘modality environments’ that tested against travel survey data demonstrate support for specific patterns of mobility, i.e. walking, cycling, car use, local and regional transit.
This classification can contribute to a new urban form based method for evaluating the potential of neighborhoods for sustainable mobility.
Keywords: Network analysis, multi-modal networks, sustainability, mobility patterns.
1. Introduction The Randstad region in the Netherlands is one of the paradigmatic polycentric city-regions in Europe (Hall and Pain 2006), comprising the four largest cities in the country (Amsterdam, Rotterdam, The Hague and Utrecht) and a series of middle size cities (Amersfoort, Haarlem, Leiden, Dordrecht and Hilversum) that together constitute its Daily Urban Systems (DUS) against a background of suburban neighborhoods and a mostly preserved rural and natural area at the centre called the “Green Heart” (van Eck and Snellen, 2006).The Randstad urban centres and their suburbs are served by an established multi-modal mobility network of local walking and cycling infrastructure, comprehensive road and public transport networks, and connected by rail and motorway networks. The Randstad’s combination of mobility infrastructure networks with land use concentration and mix should offer the baseline conditions for sustainable mobility patterns within the local neighbourhoods and across the region (Figure 1).
Figure 1. Map of the Randstad city-region, showing its areas, main urban centres and main mobility network infrastructure.
The Randstad’s current configuration is the result of a long spatial planning tradition based on carefully planned neighborhood development since World War II (Wassenberg 2006) that over the decades has evolved from implicit to explicit sustainable urban development (Goedman et al. 2008), reflected in policy documents since the late 1980s (Buijs 1992, VROM 2001, VROM 2008). The Fourth Spatial Planning Framework Extra, also known as VINEX, introduced a program of urban expansion of new residential areas focusing on the core concepts of sustainable neighborhood development and sustainable mobility in particular. The Fifth Spatial Planning Framework, the latest spatial strategy for the Netherlands, sets as key objectives the reduction of traffic congestion, the intensification of land use and the development of the network for multi-modal transport provision (VROM, 2001; Snellen and Hilbers, 2007) with the aim of achieving a more sustainable mobility. Understanding the spatial conditions that support these policy objectives is a primary concern.
Some of the main VINEX objectives have in general not been achieved, i.e.
increase in walking and cycling in the neighborhood, use of public transport for commuting or reduction of car use. In particular, the locations in green field sites do not lead to more sustainable mobility patterns when compared to other parts of the country and continue to perform worse than new and old inner city locations (Hilbers and Snellen, 2005).
232 ITU A|Z 2014 - 11/ 2 - J. Gil, S. Read While this can in part be explained by differences in socio-economic profile between these different locations, for a particular type of location one might find a consistent trend of mobility pattern. With the aim of exploring this assumption we look at empirical evidence from a mobility survey and at network structure characteristics of the city-region within a framework of sustainable mobility indicators. This paper follows from previous research analyzing public transport networks using the space syntax configurational approach (Gil and Read 2012), which revealed the structure and hierarchy of each network and of their integrated effect, towards assessing the potential of different neighborhoods to support sustainable mobility patterns.
2. Sustainable mobility patterns in the Randstad city-region The general sustainable mobility vision for city-regions proposes a more integrated and ‘seamless’ multi-modal public transport system around quality neighborhoods and vibrant city centres, with land use distribution matching the needs of population, business and institutions, shifting mobility to soft transportation modes such as walking and cycling and to public transport for long distance travel (Banister 2005). These objectives can be monitored through the use of sustainable mobility indicators, like the ones found in numerous urban from and travel studies and policy documents,such as distance traveled per mode or per person, modal share and number of journeys (Cervero and Kockelman 1997; Newman and Kenworthy 1999; Banister 2008;
Bruun, E., Schiller, P.L.L. & Litman, T., 2012; Gilbert, R., Tanguay, H., 2000;
European Commission, 2001). Using empirical data from the Netherlands Mobility Survey from the years 2004 to 2009 (MON 2004-2009) containing 282,543individual home based journeys between the 4-digit postcodes of the Randstad city-region,one can identify the sustainable mobility patterns of the population according to a collection of sustainable mobility indicators (Table 1). In this table, the mean, minimum and maximum values for each indicator are given for the whole Randstad, providing baseline against which one can compare the performance of specific postcodes.
From the mean values in Table 1 one can observe certain mobility trends in this city-region. The overall number of cycle journeys share is high at 25%, even higher than walking, but this depends on the distance traveled because more than half of the short local journeys are done by walking, followed by the bicycle at 30.66%.Transit share is on average very low, which is surprising considering the extensive public transport infrastructure, however many locations away for the larger urban centres are not served by a variety public transport modes, and in urban areas public transport share can be as high as 36% of the journeys. Despite the relatively high values of some sustainable mobility indicators, the car journeys share is the highest on average 44%, approaching a 75% share when it comes to total distance traveled. For that reason, there are policies in place to reinforce the positive change towards sustainable mobility, represented in Table 1 by the symbols in the ‘Sustainability direction’ column.
One aspect that can be found in the data set is the close relation between multi-modal journeys and overall public transport journeys. While the large majority of multi-modal journeys use public transport (86%) either in one or more legs of the journey, the other legs are mostly walking (54% at origin and 71% at destination), cycling (13%) and with the car (8,5% as driver and 5% as passenger).
Patterns of sustainable mobility and the structure of modality in the Randstad city-region Table 1. Selection of sustainable mobility indicators. The ‘Sustainability direction’ column shows the intended direction of the indicator in relation to general sustainable mobility objectives.
What is clear from the minimum and maximum values in Table 1 is that there is a large amount of variation for certain mobility indicators, which is suggestive of a local variation in conditions that support specific mobility patterns. We can map the sustainable mobility indicators in the region using scaled values centred on the Randstad’s mean value, with red showing indicator values below the baseline and green indicator values above the baseline (Figure 2).
Looking at the variation of indicator values on the maps, they present clear spatial patterns, further reinforcing the notion that urban form and configuration characteristics can be used as indicators of sustainable mobility especially in planning.
3. The configuration of multi-modal urban networks Existing models of sustainable urban form, such as transit-oriented development (TOD), and of sustainable accessibility, such as‘Multi-modal urban regional development’ (Bertolini and Clercq 2003), relate specific urban form characteristics to sustainable mobility patterns. In terms of urban form characteristics, these models focus on the presence of transport nodes, on the public transport’s network size and service quality, and on the location, density and diversity of activities. They use node, density and accessibility measures (Cheng et al. 2012) where the network provides the connection between opportunities (land use units or transportation nodes) and is used to measure the distance to them (accessibility) and their number or size (density) reachable from a given location.
Other urban form models focus on the characteristics of the street network itself, measuring the composition of the street layout (Marshall 2005), network reach (Peponis et al. 2008) and network centrality (Hillier and Hanson 1984),providing the network affordances of all locations assuming that the opportunities are the same everywhere in a general form of accessibility(Batty 2009). These street network models are used in the context of sustainable development to describe and measure the configuration of urban areas and can extend to cover entire cities and city-regions.
In order to better understand the complex relation between urban form and sustainable mobility patterns it isproposed that the city-region needs to be measured according to the configuration characteristics of its mobility infrastructure networks, and for that we need integrated urban network models. These models canaddress the organising role of the mobility infrastructure networks, where these whole,integrated structures define the relational condition of urban areas in a city-region (Read et al. 2007; Read and Gil 2012).
3.1 Multi-modal network models inspace syntax research The spatial network developed in space syntax theory most used in urban and regional studies is the ‘axial map’ (Hillier and Hanson 1984; Hillier 1996), and its derivatives that split the lines of the map into smaller segments producing the‘segment map’ (Turner 2001; Hillier and Iida 2005) or merging lines based on their angular connectivity producing the ‘continuity map’ (Figueiredo and Amorim 2005). The most conventional geographic representation of the Patterns of sustainable mobility and the structure of modality in the Randstad city-region street network in GIS is the road centre line, with linear segments drawn along the middle of the street or of the individual traffic lanes. The resolution of the ‘road centre line’ based models is at the level of the street segment and the crossing node. Inlarge-scale studies, and to allow the use of publicly available street databases, methods have been developed to apply space syntax centrality analysis to road centre line networks (Dalton, Peponis, and Conroy Dalton 2003; Turner 2007; Peponis, Bafna, and Zhang 2008; Chiradia et al. 2008; Jiang and Liu 2009). Both the road centre line and the axial map representations are used to describe the street networks used by private transport, i.e. pedestrian, bicycles and cars.
As for the public transport networks, their representation is a standard feature in transportation network models, where the public transport stops are represented as nodes on the network with the links connecting these stops along the service routes or tracks. There are some examples of adding public transport networks to the models based on the ‘axial map’ (Chiaradia, Moreau, and Raford 2005; Gil 2012; Law, Chiaradia, and Schwander 2012), most of the times opting for a simplified topological representation linking the stops and stations directly, and considering additional topological links for transfer between modes.
The power of these street and multi modal network models can be further increased by integrating the activity and land use information using the buildings or building plots and connecting these to the nearest street (Ståhle et al. 2005; Marcus 2005; Sevtsuk 2010).
Beyond aspects of network representation, the analysis of network models uses the concept of network distance, which can take different forms (Hillier et al. 2010). This can be physical distance based on the length of the street segment, topological distance where every change of direction counts as one topological step, or angular distance where the angle of direction change is taken into account and a 90-degree change of direction is equivalent to one topological step (Turner 2001; Dalton 2001; Hillier and Iida 2005).In the case of the public transport network, the focus is on the network structure and the impedance is simply topological, with network transfers representing additional topological steps. However, when one starts working with multimodal networks where flows happen at different speeds, one should also consider temporal distance where physical distance takes travel speed into account.
3.2 Measuring multi-modal network models Table 2 provides a summary of different network metrics that can be calculated to characterize the mobility conditions of local urban areas using a multi-modal network model.