Shadow effect of the High Speed Lines in Spain.

Transcription

1 Shadow effect of the High Speed Lines in Spain. C. Zamorano 1 ; E. Romo 2, J. Sastre 3 1 Transport Research Center - Universidad Politécnica de Madrid; Madrid, Spain; 2 PROINTEC,S.A., Madrid, Spain;,3 Fundación Caminos de Hierro, Madrid, Spain Abstract The traditional assessment of the modifications induced by a new HSL in town and country planning is oriented towards pinpointing the positive side of high speed infrastructure, i.e. the growth of metropolitan areas having HSL stops. In most of the cases, key developments are not taken into account, e.g. the impact on areas near cities with HSL stops (in these areas, the cities may lose weight compared to metropolitan areas of cities with HSL stops) and especially what is called the shadow effect, which happens in the cities that are left without a HSL stop, i.e. left in the shadow of the new infrastructure but having to experience the disturbances due to HSL construction as well as suffering from the permanent barrier effect just created by the new transport mode although without reaping the benefits. This paper presents a study carried out jointly by the Centro de Investigación del Transporte, Transyt, de la Universidad Politécnica de Madrid and the Fundación Caminos de Hierro para la Investigación ferroviaria. The goal of this study is the analysis of the impact of HSL implementation in Spain as regards medium and small size cities along the high speed corridors but with no HSL stops. 1. Introduction Without any doubt, Railway has been the transport mode having undergone more changes in recent years. High Speed has indeed deeply impacted the transport industry and its implementation has changed the image of a service which was considered slow, uncomfortable and obsolete. This new offer, based upon speed, direct access to city centers, punctuality and comfort as well as an optimum Quality of Service has allowed railway to gain customers from other modes of transport and even generate a new demand. This worldwide trend is especially important in countries like Spain where medium-size cities are a few hundreds of kilometers away and high speed provides an effective and competitive mode of transport. In recent years the policy of many city corporations and regional bodies has focused on belonging to the high speed network as experience shows this to be a growth factor. The construction of a high speed railway network represents an important challenge for different reasons, among which the high costs of investment and operation as well as the huge impact of the line on the affected areas from an environmental and territorial point of view. However, two situations have to be considered: A new High Speed Line (hereinafter referred to as HSL ) within an area without prior railway service, thus providing a brand new service. A new line improving the existing railway service, either by replacing the old line or building a new HSL in parallel. Usually in European countries, the railway network is consolidated and the latter situation is the most usual, thus giving rise to the concept of Shadow effect or Tunnel effect. When the HSL is superimposed over an existing line, many stations will disappear, as high speed needs a large separation between stops (distances as long as hundreds of km) to optimize the maximum operating speed. In the case of the new HSL running in parallel with the existing line, it will mean the end of most of Long Distance trains scheduled for the conventional line, that shall then provide Regional or Freight service. The Shadow Effect will have different consequences depending on how the HSL will be implemented along the new corridor: Superimposing the HSL over an existing line to be dismantled Construction in parallel and near to the existing line

2 Construction of a brand new line away from the existing line Therefore the Shadow Effect will affect two types of cities and thus modify the mobility of its inhabitants: Cities left with no railroad service whatsoever Cities losing all or most of Long Distance service and getting regional services in return, thus depending on bigger cities nearby for long or medium distance travel. This effect is difficult to avoid because high speed railway can be considered as a spatially selective mode of transportation, as its main goal is the interconnection of major cities without considering the territory between HSL stations (Gutierrez, 2004), which merely becomes the support of the infrastructure without being connected. 2. Objectives and methodology The goal of this study is the analysis of the impact of HSL implementation in Spain as regards medium and small size cities along the high speed corridors but with no HSL stops. All three current Spanish HSL, see Figure 1: have been analyzed: Madrid-Sevilla (1992), Madrid- Barcelona (2004) and Madrid-Toledo (2005), assessing the effect of high speed on the relevant areas and establishing the extent and impact of the Shadow Effect for each of these areas. One of the interesting points of this research is the analysis of three different types of high speed line layouts: the first one is a HSL with a new layout which is somehow separated from the conventional line, the second one is a HSL running almost in parallel to the conventional line and the third is a high speed branch line replacing the existing line. Two more HSL have been put into service as of December 2007: Madrid-Valladolid and Córdoba-Málaga Figure 1. Layout of three Spanish HSL analyzed The research methodology is based upon: Analysis of line modification, considering several aspects: infrastructures, investment cost, status of services before and after HSL commissioning as well as current status Direct inquiries to mayors and politicians of cities where railway services have been greatly reduced, in order to get first-hand knowledge of the impressions of main responsible people and the general public in these areas

3 The determination of service modifications in areas affected by new HSL has been established with the help of matrixes for different time scenarios, which summary are presented in this paper: Socio-economic indicators of the affected cities Number and type of services for each stop and city Direct connections between each stop and others on the line. The analysis of these indicators will allow to assess the impact on mobility and interdependency of the different cities, as well as the assessment of the direct impact of the new infrastructure: The list of the Scheduled trains for the different corridors, for different dates, extracted from official RENFE schedules (after 1992 the Infotren program has been used). This research only considers direct trains. For each station, boundary destinations have been taken into account, i.e. those affected by the new HSL. Connections between stations. In order to find out the changes in mobility induced by the opening of a new HSL, an origin-destination matrix has been created where the daily direct trains between stations in one direction. The sum of all these possible direct destinations from each one of the stations is a simple and realistic indicator of the travel options of the inhabitants. The gain or loss of connections then indicates the time-varying mobility. By comparing schedules at different times it is possible to know about its evolution. The population and other socio-economics-data have been fetched from Instituto Nacional de Estadística (INE), and the regional administration 3. Contents of the research 3.1. The Spanish railway network The current Spanish railway network shows a star topology with a low level of arborescence, inherited from the XIX century, with track layouts allowing low level performance, exceptionally over 120 km/h. The level of electrification is higher than other European countries but the percentage of double track is lower. According to UIC data from 2004, in Spain only 32,1 % of all existing lines are double track while France reaches 55,25 %; Germany 51,72 % and Belgium 73,70 %. Only Italy, with 39,97 %, shows a percentage close to Spain s. Furthermore, discarding Spanish HSL, the percentage of double track in Iberian (wide) gauge lines decreases to 25,7 %. Even the main radial corridors departing from Madrid are not fully double track. Railway is a transportation mode that has undergone a significant market share decrease in Spain. In the last thirty years Spanish railways have gone from transporting 21% of goods and 28% of passengers to 6% of goods and 7% of passengers (CE, 2006) during a period where demand has been constantly growing. The first HSL in Spain was born out of the need to clear railway line sections of single track and obsolete layout (Zamorano, 2006). Furthermore the new HSL also tried to improve service by improving the system and not by simply saturating the existing lines which were mostly underused. Another milestone in the first Spanish HSL was the use of UIC gauge in order to fully interconnect the Spanish lines with the rest of Europe thus avoiding gauge adaptation systems and using common signaling and power systems. Consequently the Spanish network is unique as it implies the coexistence of two gauges and rolling stock can be used for one gauge only except for trains with gauge adaptation systems (Spanish technology from TALGO and CAF). The section between Tardienta and Huesca, put into service in 2005, is a branch from the HSL Madrid-Barcelona with dualtrack gauge although this is a complex technology that has not been technically fully assessed. The Spanish railroad network is being managed since 2005 by the Administrador de Infraestructuras Ferroviarias (ADIF) and includes two separate networks almost independent: Conventional network using the so-called Iberian gauge(1668 mm), with more than km High speed network using the UIC gauge (1435 mm) which is ready for interoperability with other European networks according to EC directives regarding power (25 kv) and signaling (ERMTS).

4 HSL construction in Spain has shown sustained growth in recent years: in 2005, km out of km were high speed, growing to km at the end of 2006, with the opening in 2007 of two new HSL (Córdoba-Málaga and Madrid-Valladolid) with the expected opening in 2008 of the HSL Tarragona-Barcelona and a grand total of km presently under construction. The Plan Estratégico de Infraestructuras de Transporte (PEIT) has set ambitious goals for the development of the Spanish railroad network: In 2010 Spain shall be a world leader in kilometers of operational high speed lines even surpassing countries with a solid high speed background as Japan and France, with more than km in service, see Figure 2. In 2020, at the time of completion of PEIT, Spain shall have km of HSL, connecting all provincial capitals and 90% of the population living less than 50 km away from a HSL stop. Figure 2. Spanish Railroad Network in 2020 HSL 3.2. Shadow effet of HSL Madrid-Sevilla The Madrid-Sevilla HSL was opened April 20th The journey between these two cities took 5h52m before with the fastest train (TALGO) down to 2h55m with high speed trains operating at 250 km/h, thus saving almost 3 hours on the journey. Currently the AVE is operating at its maximum design speed of 300 km/h and the journey is again down to 2h20m, for a time savings of 3h32m. It was the first HSL as well as the first line using UIC gauge in Spain. The main motive of its construction was the bottleneck at Despeñaperros, which was the only access point to Andalusia from the rest of Spain. The old infrastructure had small curvature radii and did not allow speeds greater than 120 km/h, and the line was almost all single track, with excessive waiting times to allow crossings and overtakings, as the line was fully saturated. The most relevant features of the HSL are as follows: Overall final construction costs greater than million euros. Number of circulating trains from in 2003 to more than in Number of daily trains operating on the line from 12 to 130. Daily frequency of trains between Madrid and Sevilla from 12 to 42.

5 Now, how shows Figure 3, there are two lines between Madrid and Seville: the former one, with Iberian gauge (574 km), named conventional line, and the new HSL (471 km). Discussion Figure 3. Schematic Layout of Madrid-Sevilla HSL and Madrid-Toledo HSL The conventional Madrid-Sevilla line had 55 stations before the opening of the HSL, while the HSL has 5 stops of which only Madrid, Sevilla and Córdoba stations are shared with the conventional line. The other 2 stops of the HSL Ciudad Real and Puertollano belonged to the Madrid-Badajoz line that was dismantled due to the construction of the new HSL with the loss of railway service in 15 points: 8 stations and 7 sidings, which had between 1-7 trains every day to the Spanish West and to Portugal. 15 of the 55 stops of the former conventional line, i.e. 28%, serving a total population of people, have completely lost railways services and 12 of these 55 stops have been incorporated into the commuter train network of Madrid and Sevilla. For the remaining stops of the line the number of scheduled trains has remained constant or has even increased in some stops, precisely due to the increase of Regional trains after the opening of the HSL. But if we only regard the long distance services, as show in Table 1, there were a important reduction just after the opening of the HSL, but, now, in 2007, 8 of the 15 stations, more than the 50%, had lost the long distance services. ESTACIONES POPULATION (2006) NUMBER OF SCHEDULED TRAINS DIFFERENCE After/Before HSL After HSL/2007 Before HSL After HSL Aranjuez % % Alcázar de S Juán % -7-82% Manzanares % 0 0% Valdepeñas % 1-33% Sta Cruz de Mudela % % Almuradiel-V Mqués % % Vilches % 0 0% Linares - Baeza % -7-80% Espeluy % -3-60% Andujar % -5-78% Villa del Río % % Montoro % % Palma del Río % % Lora del Río % % Los Rosales % % Table 1. Evolution of long distance daily direct services. Conventional Madrid-Seville line.

6 The municipal corporations most severely affected by the opening of the Madrid Sevilla HSL are those that were previously used as a railroad hub, i.e. Aranjuez and Alcázar de San Juan, which have experienced a drastic reduction of 48% of scheduled trains and more than 60% of connections. The remaining stations have lost between 40% al 60% of scheduled trains, comparing the figures in 2004 with those prior to HSL opening, but the impact on mobility has been worse if taking into account the loss of direct connections with other stations as show Table 2. POPULATION (2006) 1992 Before HSL 1992 After HSL 2004 Befote/Alter HSL Alter HSL/2004 Madrid % -42% Aranjuez % -69% Castillejo - Añover Aranjuez % -78% Villasequila % -63% Tembleque % -83% El Romeral % -83% Villacañas % -58% Quero % -71% Alcázar de San Juan % -60% Cinco Casas (Tomelloso) % -89% Manzanares % -64% Daimiel % -77% Almagro % -76% Ciudad Real % -41% Puertollano % -39% Valdepeñas % -49% Santa Cruz de Mudela % -51% Almuradiel - Viso Mqués % -52% Vilches % -41% Linares - Baeza % -56% Espeluy % -61% Mengibar - Anchuela % -36% Andújar % -55% Villa del Río % -64% Montoro % -65% Córdoba % -47% Posadas % -67% Palma del Río % -75% Lora del Río % -75% Sevilla % -4% Table 2. Journey Direct Connection s with other stops. All railways services. Influence area of Madrid- Seville lines. The number of scheduled trains has shown a significant increase for all HSL stations, especially Sevilla, Puertollano and Ciudad Real, where this number has doubled. As these trains have only a few stops, the number of connections with other stations decreased in 2004 with respect to the situation prior to HSL opening, especially in Ciudad Real and Puertollano but mainly in Madrid, with a reduction of more than 40%. The journey from Madrid to other stops of the HSL was thus considerably reduced although in contrast almost half of the possible destinations in direct trains from Madrid were withdrawn.

7 3.3. Shadow effect of HSL Madrid-Zaragoza-Barcelona The HSL Madrid-Zaragoza-Barcelona starts from Estación de Atocha in Madrid, where there is a connection with the HSL Madrid-Sevilla, and heads Northwest passing through the cities of Guadalajara, Calatayud and Zaragoza, then going to Lleida. From this city it goes to Tarragona, Barcelona and Gerona, and reaches the French border close to La Jonquera. It has been designed for a maximum speed of 350 km/h. The journey between Madrid and Zaragoza is expected to last one hour and a quarter in good operating conditions instead of three hours on the conventional line and Madrid-Lérida shall last one hour and three quarters instead of four and a half hours, while the journey from Madrid to Barcelona will take two hours and a half, for a distance of 621 km, instead of the current six and a half with a total time savings of four hours. At the time of HSL planning, the provinces around this HSL accounted for 30 % of total Spain s population and 39 % of Spanish GDP. The whole line shall be fully commissioned in the course of the year This research has focused on the first section opened in 2003 i.e. Madrid-Zaragoza-Lleida. In this section, the HSL runs parallel to the conventional line and both share two stations located at significant cities, i.e. Calatayud (population ) and Guadalajara were the new HSL station is far away from the old station. In 2006 the HSL service was extended until Tarragona. Discussion Madrid and Barcelona are currently linked through two railroad lines, a HSL and a conventional Iberian gauge line. Both lines have almost parallel layouts. In the conventional Iberian gauge line, Long Distance trains have been withdrawn or reduced to a minimum (a single train). The corridor has been subdivided in several sections with regional service o direct regional trains with less stops, as the new HSL has taken over all Long Distance service. Regarding all kind of services, unlike for the HSL Madrid-Sevilla, there has not been a drastic reduction in the number of scheduled trains servicing the stations of the line, because adjustments were made before the opening of the new line. When considering only long distance trains and the provincial capitals around the line, Pamplona, San Sebastián-Hendaya, Huesca, Barcelona, Tarragona and Gerona, the corresponding data indicated in Table 3 show the relevance of the effect in Guadalajara, although balanced by the new station Guadalajara-Yebes, and in Siguenza and Arcos del Jalon, and even one town, Jadraque, has lost all long distance service. Estación Direct destinations (number of regional capital s) Number of jurney services 2003 Before HSL 2003 After HSL Alcalá Henares 3 3 0% % Guadalajara % % Guadalajara Yebes (new HSL) Jadraque % % Sigüenza % % Arcos de Jalón % % Calatayud % % Table 3. Evolution of long distance daily direct services. Conventional Madrid-Zaragoza line Regarding the cities in the influence area on the section, and if we consider all railways services in the line, which appear in Table 4, we find with that after the opening of the LAV, the towns that had lost direct connections in the years before HSL have recovered except for smaller towns, e.g. Sigüenza, and Caspe.

8 Population (2006) 2000 Before HSL 2003 After HSL / /2004 Madrid % -23% Alcalá de Henares % 7% Guadalajara Yebes (new HSL) Guadalajara % -40% Jadraque % 17% Sigüenza % -7% Torralba % 20% Arcos de Jalón % 23% Calatayud % 48% Zaragoza % 54% Pamplona % 60% Huesca % 9% Caspe % -19% Lérida % 105% Reús % 36% Tarragona % 27% St Vicent Çalders % 3% Barcelona % 22% Table 4. Journey Direct Connection s with other stops. All railways services. Influence area of Madrid-Zaragoza line Shadow effect of HSL Madrid-Toledo The old line connecting Madrid and Toledo was opened in It was a single track line electrified in In 1988 to construct the HSL Madrid-Seville, the section between Parla and Castillejo of the Madrid-Badajoz line, was closed down, after which all trains between Madrid and Toledo went through Aranjuez. The HSL Madrid Toledo was opened November 15th The new infrastructure built had a length of 20,5 km with double track. The trains going to Sevilla and Toledo share the first section (4,5 km) of the corridor as we can see in Figure 3. At P.K an underpass allows the left track to pass over the HSL Madrid - Sevilla and merge into the right track from which point begins the independent line to Toledo. After crossing the river Tajo, the layout coincides for 7.8 km with the former conventional line between Castillejos andtoledo that had to be dismantled. The adaptation of track and premises at La Sagra allows the branching of trains towards Toledo from the HSL Madrid - Sevilla at a speed of 220 km/h. The maximum speed on the line is 270 km/h. Construction costs have been 215 million euros. Discussion The opening of the HSL Madrid-Toledo represented the closure of part of the old line where 4 stations and 1 halt lost all railroad service, although the population serviced by these 5 stops was quite small. In fact, as seen in Table 5, these have gradually lost railroad service. The number of daily trains remained almost constant during the 1990s, although the progressive closing-down of stations reduced the number of connections with other stations. Regarding the evolution of services during the fifteen years prior to HSL opening, there was a progressive reduction of the number of stops: Infantas stop close down in 1987, Villamejor stop closed down in 1999, the Toledo Industrial stop was also closed down three years later. In November 2005 the Castillejo Añover station was closed down and at the beginning of HSL services the Aranjuez station ( inhabitants), who had already lost railroad services with Andalucia and Portugal because of HSL Sevilla-Madrid, lost its railroad link with the city of Toledo.

9 Population Before HSL (2005) After HSL (2005) Madrid Aranjuez Las Infantas (Aranjuez) Castillejo-Añover (Aranjuez) Villamejor (Aranjuez) Algodor (Aranjuez) Toledo Toledo Industrial (Toledo) Table 5. Evolution of services. Madrid/Toledo HSL line 2007 The construction of this high speed branch has brought about many changes for passengers. The intermediate stops between Madrid and Toledo have been withdrawn and journey time has been halved although, the number of journey services is the same than before, the commuter service has been fully eliminated and thus Toledo passengers must now pay for parking which was free and a long connection must be made in Madrid to go into the Subway or Commuter network. The Transport Card of the Comunidad de Madrid cannot be used although special cards have been offered to regular users of this service. In the future there can be a problem of scarcity of tracks, as the first part of the line is shared with the HSL Madrid-Sevilla Surveys for the HSL Madrid-Barcelona Seven mayors and transport councilors were surveyed in the main municipal corporations of the section between Madrid and Zaragoza of the conventional line Madrid-Barcelona, which runs almost parallel to the new HSL. Two of these corporations Guadalajara and Calatayud have stations for both lines although not interconnected and in the case of Guadalajara they are 12 km away. The survey was exhaustive and conducted through personal interviews, fetching interesting data: Towns left without a HSL station have resigned to the new situation. It can be surprising that Alcalá de Henares, with a population exceeding people, do not demand long distance service and only a direct connection with Barajas Airport. Only the mayor of Arcos de Jalón, possibly influenced by the tradition of this former railroad hub, demanded service improvements with higher train frequency and more direct destinations. All other towns with less than inhabitants only ask for a regional train service connecting directly to the provincial capital, with proper frequencies and schedules adapted to the main motives of traveling (i.e. shopping, medical care and to a lesser extent, work). Regarding towns with a conventional and HSL station, Calatayud and Guadalajara have sensibly improved although some aspects could be still improved such as service frequencies at different times of day and the connections through future HSL with Soria and Teruel. 4. Conclusions The decrease in number of scheduled trains and direct connections has been demonstrated and as the main conclusion of this research it confirms the significant impact on mobility that is induced by the construction of a new HSL in towns without a HSL stop. The three HSL analysed for this research represent three different types of impact on the territory around a HSL: when the new HSL layout requires dismantling an existing line, many towns lose railroad service, in this case 15 towns along the Madrid-Puertollano section of the HSL Madrid-Sevilla, i.e. 1 every 14 km of the new line and 4 towns along the HSL Madrid-Toledo, 1 every 5 km, plus 12 more along the HSL that become part of the commuter network. when the new HSL layout is close and parallel to the existing line, the latter loses almost all Long Distance trains, in many cases regional service is boosted and thus the mobility trend of

10 the population changes drastically: in the case of the Madrid-Barcelona corridor after the opening of the new HSL only a single train connects both cities on the conventional line. when the new HSL layout is away from the existing line, the opening of the new HSL creates a so-called mirror effect on the conventional line where regional Express service is boosted, stops are withdrawn and even stations are closed, maybe in an attempt to imitate HSL service. This has happened on the old Madrid-Sevilla line, where 22 stations have been closed after the opening of the new HSL. The construction of HSLs greatly modifies the management and operation of the whole railroad network. The new HSLs take over almost all Long Distance services. The journey times must be reduced until they can compete with flight times and thus the number of stops has to be optimized. When the conventional line, more or less parallel to the new HSL, is not dismantled, passenger traffic trends change abruptly as regional trains do not service anymore the destinations that were previously Long Distance. On the other hand this will benefit freight traffic as more tracks are available on conventional lines. Undoubtedly less people are affected than benefited from the new situation although this is a factor to be considered as it can worsen the territorial imbalance and thus hinder the desired economic convergence of regions. Therefore this effect shall be assessed on the political decision-making side because HSL construction represents a huge economic effort at national level. The European Commission has established that the policy of territorial development should ensure that high performance transport infrastructures be completed with secondary networks for all regions to share the benefits (CE, 2008). Consequently, this effect shall be anticipated and fully characterized for the different towns along new HSLs, implementing corrective and palliative measures to minimize the consequences of the shadow effect. Therefore a methodology is needed for the process, and the results of this research suggest the following outline: First of all, the effects of the HSL in each of the affected towns must be assessed depending on whether railroad service will be withdrawn or modified. If railroad service is fully withdrawn an alternative bus shuttle service to the nearest railway stations shall be provided with schedules adapted to railroad travel. If railroad service is preserved, an optimum service offer shall be established for each town, either by keeping some long distance service or by implementing a regional railroad shuttle service with HSL stations, with schedules adapted to the optimization of intermodality between the two types of service. References Comisión Europea. COM (2006) 314 final. Por una Europa en movimiento. Movilidad sostenible para nuestro continente. Revisión intermedia del Libro Blanco del transporte de la Comisión Europea de Comisión Europea. (2008) Working for the regions. EU Regional Policy 2007/2013 Gutierrez Puebla, J, (2006). El tren de alta velocidad y sus efectos espaciales. Investigaciones regionales. N. 5. Asociación Española de ciencia regional. Alcala de Henares. España. Zamorano, C. Romo, E (2006). Desarrollo de la red ferroviaria española: pensar en el ferrocarril, pensar en red. Revista: Ingeniería Civil. Número 141. Abril España. Acknowledges Authors acknowledge to the students: Samuel Santos and Adolfo Rincon for their participation in the accomplishment of this work.