HST-Railway station as Dynamic Nodes in Urban Network

05082621B.SOVHP/IHE137Version March 1st, 2006HST-Railway Stations as Dynamic Nodes in Urban NetworksPaper to be presented at the conference ‘Land Use and Transportation Planning in Urban China’, Ministry of Construction & China Planning Society, Beijing, June 14-16, 2006Hugo Priemus, Faculty of Technology, Policy and Management, Delft University of Technology The author is Scientific Director of the Habiforum program ‘Innovative Land Use’, cofunded by Dutch government.AbstractIn the European Union we observe increased shares of high-speed trains and light rail + metro lines in the modal split of passenger transport. This may suggest a revival of rail transport and of railway station areas. Particularly in the context of the dynamics of urban networks, HST railway stations areas are now being redeveloped into urban nodes. In a number of EU-countries this is stimulated by local governments, but also by national governments.The Dutch government is co-funding the redevelopment of HST railway stations. But in general developers and real estate investors have a decisive role: they determine whether private investments will be made in and around these stations. The development of HST railway station areas is a challenge for public-private partnerships, in which private initiatives and public values will be combined. Modern large urban networks are best served by a multinodal approach, with the development of not one but several big stations, each with its own functional programme and its own special transport features. This prevents overburdening one station where all transport networks (roads and railways) meet. Following the model of Japanese best practices, more synergy could be achieved between urban development, urban real estate development, and the development and operation of transport infrastructure networks.Keywords: Railway Station, High-Speed-Train, Network City, Urban Network, Public-Private Partnership.Acknowledgement: The author thanks Piet Rietveld and Bert van Wee for their comments on an earlier version of this paper.11. IntroductionIn most Western European countries railway companies are leading a troubled existence.Traditionally they are public monopolies which are being made independent with considerabledifficulty and in a few cases completely or partly privatized. The train has only a modest share in the modal split of passenger transport, while the passenger car and the aeroplane are still gainingground. For a long time the stations and the station areas shared in the malaise of the railwaysector. In the Netherlands the rail sector was demerged and privatized in the late twentieth century. At that time the operation of station buildings and the development of station areas was placed in the hands of separate limited liability companies. This marked the beginning of a new dynamic in station areas in the Netherlands. In many other European countries we see similar developments.A second factor which has reinforced this dynamic is the development of the European HSTnetwork. More and more countries, such as the Netherlands, are connected with this network. Itis expected that the stations where the high-speed train stops will have to cope with a larger number of passengers, partly of high socio-economic status. This will bring spending power into the station building and its vicinity, thus reinforcing both the urban and transport functions of the station (Bertolini, 1996; 1998; Bertolini & Spit, 1998; Serlie, 1998; Zweedijk & Serlie, 1998). There is a third factor which considerably reinforces the previously mentioned factors: the construction of light rail connections and metro systems, which improve public transport systems at the local and regional levels (Raad voor Verkeer en Waterstaat, 1996; Priemus & Konings, 1999; 2000; 2001; Priemus, 2004). The relatively old stations in urban areas serve as central nodes for these public transport systems which connect networks with each other at various scale levels. Not only developments in the transport sector mark a new dynamic phase in the existence of urban station areas; the recovery of the urban economy in advanced economies is also relevant (Boelhouwer et al., 1995; Priemus, 2001). This brings us to the fourth factor. The structural shift from employment in industry to employment in business and personal services is putting the cities back on the map. The rapid development of information and communication technology raises the productivity of production processes and supports consumer demand. This leads to more opportunities for creative activities and knowledge development (Florida, 2002). Cities provide a more attractive environment for activities of this kind than for large-scale industrial processes, which to an increasing extent are moving to rapidly-developing low-wage countries. Finally, it seems that in certain respects the city is gaining ground as a residential area. Surveys of households’ residential preferences in Western Europe show that the suburban residential environment is still the most popular. However, a significant minority of 25% to 30% prefer to live in a metropolitan environment, close to urban facilities and in a relatively central location (Spaans et al., 2004). This trend makes the station area attractive as a residential environment and supports the recent policy shift in the Netherlands from ‘compact city’ to ‘network city’ (Govers et al., 1999; Bontje, 2003; Van den Burg & Dieleman, 2004). The network city is polynuclear and has several nodes, with stations contributing significantly to the accessibility of the urban centres. At a slightly higher scale level, a few network cities together form an urban network, such as ‘Brabantstad’ and ‘Randstad Holland’ in the Netherlands.The new dynamic of stations and station areas means a considerable opportunity for investments for both the private and the public sector. Broadly speaking, government authorities are responsible for the improvement of transport infrastructure and the public space. There is also plenty to co-ordinate in the area of spatial economics. For the private sector there are potential2profits to be gained in various areas: land and property development in the station area and possibly the operation of some public transport connections.This paper analyses the growing dynamic of stations and station areas in the EU and the Netherlands.The central problem is: What can the redevelopment of HST railway station areas mean for the spatial and economic development of urban areas in EU-countries in general and in the Netherlands in particular?We refer in particular to the synergy between urban networks and HST station area redevelopment.The following research questions are dealt with:(1)How is high-speed rail transport in the EU developing since 1990 amidst a changingmodal split (Sections 2 and 3)?(2)How can the HST-railway station nodes be analysed, making a distinction between thecumulation of urban functions and transport relations (Section 4)?(3)What urban impact in different zones can be expected after the advent of the HST(Section 5)?(4)How is the Dutch program of New Key Projects defined, promoting the redevelopmentof HST railway station areas (Section 6)?(5)How can the synergy between HST-networks and urban structure be strengthened? Whatis – in this context – the relevance of the transformation from mononodal to multimodal urban structures (Section 7)?(6)How can the (re)development of HST station areas be defined as public-private challenge(Section 8)?The study is conducted, based on some EU statistics, a literature review and an analysis of Dutch policy documents.In Section 2 we analyse the changes in modal split in the EU and the Netherlands since 1970. Section 3 focuses on the development of high-speed rail transport in the European Union. After these two contextual sections, we analyse nodes as concentrations of urban functions and traffic flows in Section 4. We follow the approach of Bertolini (1996; 1999), who distinguishes the ‘node value’ and the ‘place value’ of railway station areas. Section 5 deals with the urban impact of the advent of the HST. The Dutch HST railway stations as so-called New Key Projects are presented in Section 6. The Dutch national government is taking at least part of the responsibility for developing HST railway station areas in order to stimulate urban economy. Section 7 shows that the Dutch New Key Projects policy does not take enough advantage of the urban design of the ‘network city’ and the ‘urban network’. In most big cities a transition can be observed from a mononodal approach to a multinodal one. In Section 8 we make clear in which way the development of HST railway station areas is a challenge for public-private partnerships. The contribution finishes with section 9 in which conclusions are formulated.2.Changes in Modal Split in the EU and the NetherlandsIn the Sections 2 and 3 we deal with the question: ‘How is high-speed rail transport in the EU developing since 1990 amidst a changing modal split? First we analyse the changing modal split since 1970.Between 1970 and 2002 the share of rail transport in the modal split of passenger transport in the “old” EU (15 member states) fell from 10.4% to 6.2%. During the same period the share of tram & metro fell from 1.6% to 1.0%. The bus also lost ground: from 12.7% to 8.3%. The winners in this period were the passenger car (from 73.8% to 78.8%) and the aeroplane (from 1.6% to 5.7%)34(see Table 1). Figure 1 (EU, 2004) shows how the passenger car has become more and more dominant in the EU since 1970, how the market share of the aeroplane gained ground in the modal split and how modest the shares of bus, rail and tram plus metro have remained.In 2002 the share of the passenger car in the Netherlands (81.5%) was above the EU average (78.8%), while the share of the aeroplane in the Netherlands (5.5%) was slightly lower than the EU-15 average (5.7%). The train had a higher market share in the Netherlands than in the EU (8.1% against 6.2). In the Netherlands the bus lags behind the EU average (4.1% against 8.3%); this is also true for the tram plus metro (0.8% against 1.0%) (Table 2).Table 1 Performance by Mode for Passenger Transport, EU-15, 1970-2002, selectedyears (pkm, horizontal %)Passenger CarsBus & Coach Railways Tram & Metro Air Total 1970 73.8 12.710.4 1.6 1.6 100.0 1980 76.1 11.88.4 1.2 2.5 100.0 1990 79.0 9.36.7 1.0 4.0 100.0 1991 78.9 9.26.8 1.1 4.1 100.0 1995 79.5 8.76.3 0.9 4.6 100.0 1999 78.8 8.46.2 0.9 5.5 100.0 2000 78.5 8.36.3 1.0 5.9 100.0 2001 78.5 8.36.3 1.0 5.9 100.0 2002 78.8 8.36.2 1.0 5.7 100.0[Figure 1 about here]Table 2 Modal Split for Passenger Transport in EU-15, the Netherlands and theUSA, 2002 (pkm, horizontal %)Passenger Cars Bus & Coach Railways Tram & MetroAir Total EU 15 78.8 8.3 6.2 1.0 5.7 100.0 The Netherlands 81.5 4.1 8.1 0.8 5.5 100.0 USA (2001) 85.7 3.1 0.3 0.3 10.6 100.0 Source: EU, 2004: 3.3.2; 3.4.17.In the United States the high share of air transport and passenger cars is striking; the market share of trains and trams is negligible here (Table 2).Table 3 Development of Passenger Transport in the Netherlands, 1990-2002,selected years (1000 mio pkm)1990 2000 2001 2002 Passenger Cars 137.30 141.10 141.60 144.20 Bus & Coach 13.00 7.50 7.60 7.20Tram & Metro 1.26 1.43 1.44 1.45Railways 11.06 14.76 14.29 13.85Air Transport 4.08 9.59 9.44 9.74 Source: EU, 2004: 3.3.7-3.3.12.Table 3 shows that the shares of the passenger car and air transport have also grown substantiallyin the Netherlands since 1990. The bus has lost market share. Since 1990 the shares of both tram and rail services have risen slightly. This is due mainly to the introduction of the free publictransport pass for students.While devoting attention in this paper to the redevelopment of railway station areas, we mustbear in mind that in the Netherlands and the EU both the railways and the rest of the publictransport services are losing ground. A couple of bright spots are the increase in light rail transport and – in particular – the rapid development of high-speed rail transport in Europe, which is discussed in the following section.3.Development of High-Speed Rail Transport in the EUIn the European Union high-speed rail transport has increased significantly since 1990. Between 1990 and 2003 passenger transport by high-speed train increased almost fivefold (Table 4).Table 4 High-Speed Rail Transport, EU-15, 1990-2003 (1000 mio pkm)Year X 1000 mio pkm1990 15.21991 20.41992 25.21993 27.61994 30.71995 32.81996 37.51997 43.41998 48.51999 52.72000 59.12001 65.52002 68.22003 70.5Table 5 High-Speed Rail Transport in EU countries, 2003 (1000 mio pkm)Country X 1000 mio pkmFrance 39.6Germany 17.5Italy 7.4Spain 2.5Sweden 2.3Belgium 0.9Netherlands 0.2Finland 0.1EU 15 70.5Table 5 shows how a number of countries participated in high-speed rail transport in 2003. France has a dominant position in European high-speed rail transport, followed by Germany, Italy and Spain (see also Table 6). In the Netherlands, high-speed rail transport began only recently, on existing tracks, which do not yet allow really high speeds. This will change in 20075when the new dedicated HST-line connecting Amsterdam and Belgium will be opened (Visser & Bentvelsen, 1991; TCI 2004a; 2004b). The United Kingdom was not yet represented in the 2003 data.Table 6 Length of dedicated high-speed rail network per country, EU-15, 2002 (km)France 1.395Germany 687Spain 377Italy 259Belgium 135Total EU15 2.853Source: EU, 2004: 3.5.3.Table 7 High-speed lines under construction in the EU (2004)km33 Belgium: Liege-Germanborderkm Belgium: Antwerp-Dutchborder 38 Amsterdam/Schiphol-Belgian border 120 km Netherlands:km88 Germany: Nuremberg-IngolstadtSpain: Madrid-Lerida 481 km France: TGV-Est 302 km Italy: Milan-Bologna 196 km Italy:km Bologna-Florence 77 Italy: Turin-Novara 92 km Italy: Rome-Naples 220 kmkm Sweden: Södertälje-Linköping140 Sweden: Nyland-Umeá 190 km United Kingdom: Ebbsfleet-London (St. Pancras) 38 kmA considerable number of high-speed lines are under construction in Europe (see Table 7). Step by step, the isolated high-speed tracks will be integrated to form a true European high-speed rail network. Several nodes in this network may develop a strategic function. Within a decreasing share of the train in the modal split the role of the high speed train in Europe will increase in the next decades.4.HST-stations at the crossroads of urban functions and transport networksIn this section the research question is: ‘How can the HST-railway station nodes be analysed, making a distinction between the cumulation of urban functions and transport relations?’Pol (2002; 2005) points out in his dissertation that there has been little or no investment in many European urban station areas for decades. With the arrival of the high-speed trains and the expansion of the European HST network this is now changing significantly. Pol (2005: 23) (in translation): ‘The development of a European HST network can be seen as a revival of rail transport. This means not only the improvement of the stations and the redevelopment of the station areas, but in particular the creation of a stimulus for urban development in the wider sense.’6In Section 5 this statement will be made conditional: only in certain situations and in certain circumstances can it be expected that the redevelopment of a station area will lead to an impulse for urban development.Over the past few years a large number of studies of station areas have appeared in which a relative degree of consensus can be observed as to the definition of a node (see Bertolini, 1996, 1999; VHP/Goudappel Coffeng, 1999; Meijers, 2000). The essence of this definition is that a node entails both a concentration of transport and infrastructure and a concentration of urban functions (work, facilities, residential function). A readily applicable definition of a node in which this essence is clearly expressed is that of the Dutch National Spatial Planning Agency (in VHP/Goudappel Coffeng, 1999) (in translation): ‘A node is a multi-modal transfer point in the network of collective and individual transport at which a spatial concentration of functions and activities is also organized).’Bertolini (1996) distinguishes two components of station areas: ‘Station areas are (or may become) important “nodes” in emerging, heterogeneous transportation networks. On the other hand, they identify a “place”, both a temporarily and a permanently inhabited portion of the city, an often dense and diverse assemblage of uses and forms, that may or may not share in the life of the node’. (See also Bertolini & Spit, 1997; 1998: 9). In this context, Bertolini makes a distinction between property-led and transport-led developments. A property-led station development is stimulated primarily by changes in the place, and a transport-led development mostly by changes in the node (or related infrastructure). Pol (2002: 31) adds two more components to the picture: spatial quality and image.The extent to which transport streams interconnect at a node is referred to by Bertolini (1996; 1999) as the ‘node value’ and the degree of concentration of urban functions as ‘place value’.[Figure 2 about here]With the help of the ‘rugby ball model’ (see Figure 2), Bertolini distinguishes five types of station location:Accessible station locations: locations where the node and place values are equal and the two functions support each other. In Figure 2 these locations are clustered around the dotted line. Dependent station locations: at these locations there are very few transport facilities and also very few cafés and restaurants, shops, businesses and houses. These stations are probably not very viable and the question is why station locations of this kind are maintained.Tense situation: tension will appear if there are too many transport facilities and other functions present at a station location. The various facilities compete for the space and get in each other’s way, which may lead to conflict or chaos.More node than place: at this kind of location the transport facilities are very well developed, but there are few other functions. These locations are very interesting from the development point of view, because there are probably possibilities for creating new functions.More place than node: these locations are characterized by a large number of functions and relatively speaking few transport facilities. Locations of this kind are interesting from the transport point of view, because there are probably possibilities for developing new transport facilities.Bertolini argues that the node value and place value of a node should be more or less in balance. The node-place model, which ranks the two values against each other, is an instrument by which nodes can be characterized on the basis of the extent to which there is a balance between node and place value. Ideally, a node should be located inside the ‘rugby ball’ in such a way that there is no dependence or tension.7Although there are huge problems of quantification, the Bertolini model has been followed by policymakers since it also clearly reflects the potentials for node development. We prefer to usethe terms ‘transport value’ instead of node value and ‘functional value’ instead of place value. Following Bertolini (1996; 1999), the prevailing view in the literature and among policymakers isthat node development should be focused on balancing the transport value of a node with its functional value. The reason is that the two are thought to reinforce each other. If a location hasgood accessibility it will attract business, facilities and households. Vice versa, a high functionalvalue generates a great deal of transportation and thus provides support for transport systems (Serlie, 1998; Zweedijk & Serlie, 1998; AVV, 2000a; 2000b).At a node different sorts of infrastructure and transport systems, both collective and individual,meet. Efficient and convenient transfers between the different modes of transport at nodes is ofvital importance if they are to function properly. Access transport which serves only the nodemust be connected at the node to transport services which connect the nodes. In this sense anode forms a link between various scale levels of transport. Although attention is often focusedon the highest level of the transport system (it is often this level which determines the transportvalue; for example an HST station, an intercity station), the underlying transport systems, withtheir feeder function, are also important to the operation of the highest level. Nodes should be‘transfer machines’, in which the passenger can make a choice from various transport alternatives. Achieving complete integration of the passenger car with other, collective modes of transport atthe node is an important part of this, but until recently this has often been ignored by public authorities, in particular Dutch national government.Station locations can be distinguished on the basis of the number of passengers getting on andoff trains per day. Table 8 shows how the Dutch Railways (NS) categorize stations according tothis criterion (Janssen & Braun, 2005).Table 8 Station categories according to the Dutch Railways (NS)Category No. of passengers getting on and off trains per dayVery small < 2,500Small 2,500 – 10,00015,000 Medium 10,000–50,000 Big 15,000–50,000Mega >In thinking about the development of new urban patterns, an important role is attributed to thebig and the mega stations.As regards the urban functions of a node (which determine its functional value), it is important toplace complementary urban functions close to each other. This principle of complementaritymeans that a mix of urban functions (houses, workplaces, shops, restaurants, facilities) are concentrated at a node in such a way that they have added value for each other: for example, notonly workplaces, but also a child day care centre, a supermarket, a fitness centre and shoppingfacilities (Janssen & Braun, 2005). The idea is that people should be able to undertake different activities in one trip chain. Careful planning of a concentrated mix of functions may increasewelfare and save mobility.8There are two factors which determine the influence area of a node:a) the scale level at which the node functionsb) the quality of the underlying network.a) Govers et al., 1999: 290 argue that, to strengthen the position of public transport structurally, integrated long-term planning of space and infrastructure is needed. A good choice of spatial scale level is a crucial element in this. The higher the scale level, the higher the influence area of a node will be. The scale level of the node function and the scale level of the place function are both determining factors. The scale level of the node function is determined by the infrastructure and transport systems available at the node and the scale level they serve. Often attention is focused on public transport systems, which can be clearly attributed to a certain scale level, although compared with a few decades ago the systems overlap much more (see Figure 3).[Figure 3 about here]The road system is also of great importance. This system also has hierarchic categories (international, national, provincial, but it is much less marked by hierarchy because of its convenient and frequent opportunities to transfer from one category to another, without requiring anything comparable with changing trains. Cars can also be used for transport over practically any distance. In short, Govers et al. (1999) conclude that car accessibility is a less distinguishing feature of nodes than accessibility through the public transport system. However, the more the capacity and costs of parking space are taken into account, the more – in my opinion - this statement must be modified.b) The quality and the spatial structure of the underlying network determine the size of a node area because of the fact that this network influences the time needed to reach the actual transport node from the node area. It is in fact not distance but travel time which determines the size of the node area. A good underlying transport system, for instance a bus network or an efficiently organized park-and-ride centre or bicycle path system, may mean that locations which are quite a distance from the actual transport node are still within its influence area. The notion of representing a node area by a circle with a radius of a certain length is too simplistic (Govers et al., 1999).The importance of the time needed to get to and from the actual transport node deserves discussion in greater detail. Some other factors which partly determine the influence area of a node are (VHP/Goudappel Coffeng, 1999): (a) Is the node the origin or the destination of the journey, and (b) the distance/time travelled in the main mode of transport.a) Is the node the beginning of the journey (origin) or the end of the journey (destination)?The average time accepted to reach a transport node for the beginning of a journey is longer than the average time accepted at the end of the journey to reach the final destination. Krygsman & Dijst (2001) have shown that in multimodal transport the average distance from home to transport node is 3.8 km, whereas the average distance to the final destination at the end of the journey is 2.6 km. Partly this has to do with the availability of transport. For example, in the Netherlands a person may have a bicycle available to get from home to the station, but not at the other end of the journey, and having to go on foot limits the radius of action.9b) The distance/time travelled in the main mode of transportThe average time accepted for getting to and from the railway station depends on the duration of the main journey and the scale of the network. See Table 9.Table 9 Average time accepted for travel to and from a railway station by scale levelDestination areaScale level OriginareaNorth West Europe 60 min 30 minPart of continent 30 min 15 minPart of country 20 min 10 min7.5minminRegion 15Metropolitan area 10 min 5 minCity 5 min 5 minSource: VHP/Goudappel Coffeng, 1999.The fact that the origin area is bigger than the destination area has consequences for the organization of the area. ‘Destination activities’, for example work or a facility, should be situated closer to the actual transport node than ‘origin activities’ – specifically, the residential function. In other words, houses can be further away from the actual transport node than facilities or workplaces and still be within its influence area. This opens up prospects for the realization of substantial housing programmes in and near nodes.If possible, the functional value of a node should also include a residential function. The Province of South Holland (Provincie Zuid-Holland, 2002) gives several arguments for this:-having a residential function increases the liveliness and livability of a node;-having a residential function and local service facilities embeds the node in the local context;-nodes offer room for the increasing demand for metropolitan living;-if a residential function is added, the location will develop into a more complete part of the city.Govers et al., (1999: 290) also focus on the nodes, because it is there that networks and spatial functions come together: ‘The desired connection between various networks and between networks and space is crystallized in the nodes. By focusing on the nodes in planning, an optimal framework for public transport can be attained’. The transport value and functional value of a station can reinforce each other if public policy focuses on this happening. The transformation of an ‘ordinary’ railway station into an HST railway station may be a reason for it to do so. However, we should take care not to have unrealistic expectations. It is by no means certain that the transformation of every ordinary railway station into an HST railway will have the major consequences for the economic development of a city or a region, which Pol (2002; 2005) is expecting.5. Urban impact of the advent of the HSTThis section deals with the following research question: What urban impact (in different zones) can be expected after the advent of the HST?Pol (2002: 18) distinguishes three options in this context: on the one hand, the HST may reinforce the existing hierarchical position of cities, while on the other hand it may promote the formation of a network of cities, given the existing hierarchical relationships between cities. He also suggests a third10。