New line Tirana – Elbasan

This is an update of an earlier proposal (2009), for a new rail line south-east from Tirana (Tiranë). The line would extend the proposed Shkodër – Tirana line, and could ultimately form part of a high-speed route toward Lake Ohrid and Thessaloníki. Even without further extension, however, proximity and combined population justify a new line Tirana – Elbasan. Since the first proposal in 2009, a motorway between the two cities has been largely completed. The new line would closely follow this A3 motorway.

Although Albania’s rail system has been de facto abandoned, there was a rail connection between Tirana and Elbasan. It was very indirect, because a tunnel under a mountain range was beyond Albania’s limited resources. Tirana is not on the coast, but it is on the edge of the central plain, where Albania’s population is concentrated. Elbasan is further inland, but located in the broad Shkumbin valley. The journey from Tirana to Elbasan started in the opposite direction, to the port city of Durrës. From there, the rail line runs south along the coast, and finally eastwards along the Shkumbin valley.

Click to enlarge: old route to Elbasan, new line in blue…


The communist-era railway was extended to the shore of Lake Ohrid, but never even reached its planed terminus, Pogradec, let alone Macedonia or Greece.

Tirana had 420 000 inhabitants at the 2011 census, although the agglomeration is now substantially larger. The Elbasan municipality had 142 000 inhabitants. The cities are separated by a mountains, with a maximum elevation of 930 m. The new motorway climbs to a relatively short tunnel, but a railway cannot climb steeply: it must enter tunnel at a lower elevation, and the tunnel will therefore be longer, about 10 km.


The proposed new line from Shkodër would end at a new underground terminal station in central Tirana, which is at about 110 m elevation. The design should allow for a south-eastern extension, also in tunnel, parallel to the boulevard Dëshmorët e Kombit. It would pass under the university zone and park, which are on higher ground at about 150 m elevation. The line would emerge alongside the motorway on relatively level terrain, at the edge of the built-up area.

The line would then enter tunnel again, under a ridge, emerging at the village of Mullet. It is now in the valley of the Erzen River. The main valley southeast of Tirana is a tectonic feature, and quite broad, but the valley floor of the Erzen itself is narrow. It is intensively farmed, and some demolition is inevitable. The new railway would run alongside the A3, which itself parallels the old main road.

Central Tirana, behind it the motorway climbing up the Erzen valley…


At the village of Ibë, the motorway leaves the Erzen, and follows a tributary valley toward the main Krrabë ridge. The motorway crosses this ridge in tunnel, into the Shkumbin valley. The tunnel portal is at about 500 m elevation, but the rail tunnel would start at about 225 m, and that means it would start closer to Ibë. The southern portal of the tunnel would also be at about 225 m, in the valley of the Kusha, a tributary of the Shkumbin. These constraints mean that the rail tunnel will probably pass under the motorway tunnel, but that is not a problem in itself.


If geological conditions are unfavourable, a longer tunnel might be needed, with more difficult approach lines. In the most favourable case, the tunnel would be 9 km long.

The southern portal would be near the Mamël-Dopaj exit on the motorway. The line can run very close to the motorway, but not directly alongside it because of the curves. Possibly several short flank tunnels would be needed here. About 3 km further on, the Kusha valley is wide enough for a straight motorway alignment, and if the gradient allows, the railway can exactly follow it.

The motorway descending toward Elbasan…


At Shijon, the line would turn away from the motorway, turning about 80 degrees to join the existing alignment. From there, it is only 3 km in a straight line to Elbasan station, which is itself about 1 km from the city centre.

Function in the network

The new line would be about 35 km long, station to station. It would carry passenger trains only, which allows for steeper gradients. Even in that case, the line will not allow very high speeds, and a journey time of 20 minutes would be reasonable.

The existing indirect route via Durrës would be upgraded. The route to Durrës would be improved in combination with the proposed new exit line, toward the airport and Shkodër. The line from Durrës to the Shkumbin is also the route to southern Albania, so there is no reason to abandon it. The section along the Shkumbin valley, from Rrogozhinë to Elbasan, would be retained for freight – most traffic will be to Durrës, not Tirana. This section might also carry a Durrës – Elbasan regional service.

New line Tirana – Elbasan

New rail line Shkodër – Podgorica

This is an updated proposal, first published in 2009, for a new rail line between Podgorica in Montenegro, and Shkodër in Albania. It would connect with the proposed new line Tiranë – Shkodër. Both would be built for high speeds, creating a fast route from Tiranë to Podgorica. There it would join the existing Belgrade – Bar line, which links the Danube basin to the Adriatic. As with all other posts here, the borders are ignored.


Podgorica and Shkodër are both located on the plains around Lake Skadar (Lake Scutari, Lake Shkodër, Lake Shkodra). Between the two cities is the Hotit arm of the lake (Humsko Blato, Hotski Zaliv). It is aligned to the north-east, with two parallel ridges alongside it. The existing rail line runs around this inlet, the proposed new line would cross it. The original version suggested a long bridge approached in tunnel, but that is unnecessary. The new line can simply cross the shallow inlet on viaduct, it is only a few metres deep. That alignment would be further south, and in fact is is the shortest between the two cities.

The region in 1904: Austro-Hungarian map from ELTE


Click to enlarge: lake panorama by Toma Koma under CC BY-SA 2.0 licence


The existing Podgorica – Shkodër railway, 64 km long, opened in 1986, as a freight link to the Yugoslav rail network. It was out of service for years due to war, sabotage, and disorder, and is only partially restored.

Podgorica had a population of 204,877 at the 2011 census. Most of the city was built after the Second World War. The municipality contains 10% of Montenegro’s territory, but 30% of its population. The main line through the city is the 476-km Belgrade – Bar railway, Serbia’s access to the sea. Podgorica station is about 1500 m from the central square: it is relatively large, with room for expansion.

The Shkodër line originally ran straight from Podgorica station towards Tuzi, but the junction was relocated further south. After Tuzi, the line climbs up the side of a ridge, and through a short tunnel, to run along the shore of the inlet. At the head of the inlet, the line makes a U-turn, and runs south through the plain. Approaching Koplik, it turns again, and runs parallel to the lake shore, into Shkodër. The station is on the east side of the city, close to the river Kir. Shkodër has a population of 136 000. The station is disused, awaiting restoration of the line from Durrës and Tirana.

New line

The new line does not need to cross the ridge near Tuzi, so it can diverge from the Bar line further south. It would simply cross the plain, running south of Tuzi. There would be no stations: the existing line would be electrified, for freight and limited local services. (There are only two settlements large enough to justify a station: Tuzi and Koplik.) The new line could take the most direct route towards Koplik, with a long viaduct across the marshes and lake. Depending on the geology, it might be better to locate the crossing at the tip of the ridges. In that case, the alignment would be closer to the existing line.


About 3-5 km north of Koplik, the new line would join the existing line, and approximately follow it across the plain for about 23 km into Shkodër. Approximately, because urban sprawl has obstructed the alignment. The line near Koplik was built alongside the main road, when Albania had almost no cars, and the line carried almost no trains anyway. Now it crosses numerous access roads for new houses. If there is no better alignment, a viaduct is the obvious solution. In Shkodër itself, there is barely enough space left for a single track: the alignment must be cleared by demolition.

Spot the railway: on the outskirts of Shkodër…


There is a sharp curve 4 km north of Shkodër station, which must be replaced by a new alignment. Apart from that, clearing the alignment will create sufficient space for upgrading. The railway is very close to the Kir river, and there is only a single line of houses on the eastern side. The station itself is well located, about 1500m from the city centre.

The new line would be 50 km long, in the shortest variant. Longer alignments would be up to 55 km long. With an entirely new line, that should allow a Podgorica – Shkodër journey time of 20 minutes. Journey time Podgorica – Tiranë, with one stop in Shkodër, should be around 65 minutes. That would be a dramatic improvement for inter-regional travel, but maximum impact of the HSL route Podgorica – Tiranë depends on further HSL connections, to the north and to the south-west.

The Belgrade – Bar line was a major project of the former Yugoslavia, but it was built for freight, and speeds were never high. After 20 years of war and neglect, it is no longer viable as a European passenger axis. A new parallel HSL would be a very difficult project, but it is a logical link to the rest of Europe, and there is no alternative route within 100 km. If this ‘trans-Montenegrin’ HSL was built, the line proposed here would complete the HSL route, from the Danube to Tiranë.

South of Tiranë, the a new rail line across the mountains near Krrabë, would dramatically shorten the route to Elbasan. That could form the start of a new high-speed line east to Thessaloniki, and possibly south-east to Athens. As with the Podgorica – Belgrade line, this crosses the main mountain axis of the Balkans, and it would be a very difficult project.

New rail line Shkodër – Podgorica

New rail line Tiranë – Shkodër

This is an updated version of a post first published in 2009, proposing a new rail line between Tiranë (Tirana) and Shkodër. The Albanian economy has grown since then, and new road infrastructure has been added. Tirana has also grown, although not as much as some thought it would. The central plain with the three cities of Tiranë, Durrës and Shkodër, has a population of about 1.3 million, enough to justify new rail lines. The flat plain favours a high-speed line: the version proposed here is for 200 km/h or more. It would be technically a high-speed line, even if it carried mixed traffic.

A high-speed line (HSL) between Tiranë and Shkodër could be extended north to Podgorica, and extended south-east toward Elbasan and Lake Ohrid. Further extensions would be much more difficult to build: from Podgorica toward Belgrade, parallel to the Lim valley line and south-east to Athina (Athens) and Thessaloniki. They would integrate an Albanian plain HSL with a European HSL network. All proposals here ignore current borders, and are made from a European perspective.

Current situation

Albania had a modest railway network, built entirely during the communist period. There was no direct line from Tiranë to Shkodër. The first line, from the port of Durrës to Tiranë, opened in 1949. It was later extended slowly northwards, starting from Vorë, which is about halfway between Durrës and Tiranë. The network was built to minimal standards: Albania was probably the poorest country in Europe. After the collapse of the communist regime, the network was almost entirely abandoned. In the last 15 years, the road network has been substantially upgraded, with new main roads between Tiranë and Shkodër.

Apart from the new roads, urban sprawl is the most visible sign of Albania’s transformation. Tiranë had 420 000 inhabitants at the 2011 census, and the agglomeration may now have twice that. The city has been expanding north-west into the plain, which complicates construction of new infrastructure. In fact, the northward expansion of Tirana might soon require the relocation of the international airport at Rinas (Mother Theresa Airport). The image shows the central plain, with exaggerated vertical axis (elevation)…

Click to enlarge…


Tiranë and Shkodër are about 85 km apart. Even at this scale, the sprawl around Tiranë, and ribbon development along main roads, are clearly visible. Much of the ribbon development is illegal: private developers simply cut their own access onto new highways, to build restaurants, petrol stations and shops.

Old and new alignment

The communist-era railway first runs north from Vorë, crosses the plain to Mamurras, and then runs at the foot of the mountains, as far as Lezhe. It follows the old main road through the villages, which avoided the marshy plain. The alignment is reasonable for a local line, with stations roughly 10 apart. North of Lezhe, the line does cross the plain, in straight sections joined by curves. The marshy coastal plain is now entirely reclaimed for agriculture, and the new main roads cross it in straight lines.

The line between plain and mountains, elevation exaggerated …


The first phase of any new line must be a new central station in Tiranë, with a new exit line. The dilapidated old station, north of the city centre, has already been demolished, The approach tracks were also removed, so there is currently no railway line into Tirana. Without a new central station, the line to Durrës can never be viable, even if it is modernised. The obvious location is the huge central Skënderbej square, and that would require a 5-km tunnel under the main Durrës road. Most of this street is very broad, allowing construction of a 4-track tunnel, with separate tracks for a regional metro service. The section nearest the square is narrower, and would require a bored tunnel – or an alternative alignment and station site. The exact location of the central station it is not considered further here, but it should allow for a south-eastern line toward Elbasan.

On the outskirts of the city, the new exit line would join the existing line toward Durrës. Near Bërxull, a new alignment would turn north-northeast toward the airport. This would avoid the worst of the sprawl, but some demolition is inevitable. At the airport, the line could run on viaduct along the access road, with a station directly connected to the terminal. The station would be 15 km from central Tiranë.

Line out to Tiranë, with curve from Durrës…


The junction at Bërxull should also allow trains from Durrës to access the airport creating an interchange with fast trains to/from Durrës. That route could be further improved by a cut-off line south of Vorë, but that requires at least 5 km of tunnel, and is not considered further here. Either way, the existing line north of Vorë could be retained for freight services.

From the airport, the second section of new line would cross the plain, to join the main highway, SH1, at Fushë-Krujë. There would be a station here, serving the town itself (18 000 inhabitants), and also the north end of the Tiranë agglomeration. (Fushë-Krujë means ‘Lower Krujë’, the historical town of Krujë is in the foothills to the east).

North of Fushë-Krujë, the third section of new line would run north to Lezhë, at first alongside the SH1 highway. At Thumane, the new line would have a connection to the old line, for regional trains to Mamurras and Laç. Obviously that requires the electrification and modernisation of the old line, as far as Lezhe. The new line would turn away from the SH1 at Thumane, and run west of the built-up area, rejoining the motorway A1 about 6 km further north.


From there, it would follow the A1 for about 13 km, and then diverge onto an alignment through the fields, to the Drin River bridge at Lezhe. About 700 m before the bridge, the new line would join the existing alignment into Lezhe station, which would then be 57 km from central Tiranë.

At least one additional track is needed through Lezhe station, because this section would be shared with freight trains, and with regional services from Mamurras and Laç. Lezhë town has a population of 16 000, the municipality 65 000. The station site is very close to the centre, and has enough space for complete reconstruction, which would certainly be necessary.

Immediately north of the station a new tunnel would be needed, under a small ridge. The line would then cross the Drin again, and continue through the plains to Shkodër, the fourth section of the new line. One option is to follow the existing alignment, which is mainly straight and level. The new line might also follow the new main road, SH1.

However a better alignment is possible, straight through the plain, avoiding the small hills and ridges. It would parallel the SH1 for only 3 km, near Gocaj. It would pass west of Blinisht, east of Dajç, and then cross the Drin at Stajkë, to join the existing alignment about 7 km from Shkodër. The existing line crosses the Drin on a secondary dam at Mjedë, but the new line would have its own bridge, about 1200 m downstream from the dam.

Click to enlarge…


Shkodër is the only city between Tiranë and Podgorica, with a population of 136 000. Using the existing alignment, the new line would cross the river Kir into the station, which is just north of the river, and east of the centre.

The Lezhe – Shkodër section would be 34 km long, giving a total length for the new line of 91-92 km. Until there are connecting high-speed lines, for instance to Podgorica, a fast inter-regional service is sufficient. With three intermediate stations, on a completely new line, a Tiranë – Shkodër journey time of 45 minutes is feasible.

New rail line Tiranë – Shkodër

High-speed bypass of Roermond

This high-speed line (HSL) is an alternative for the longer HSL Eindhoven – Sittard, proposed here earlier. It is part of a high-speed corridor from Amsterdam to Aachen, and was originally described in Dutch. This post completes the English-language description of the corridor, which includes the existing Amsterdam – Utrecht line, a new HSL Utrecht – Eindhoven, and a new HSL Sittard – Aachen.

Both alternatives have the same function – to bypass the existing route through Roermond. The longer HSL to Sittard is the shortest option, almost a straight line between Geldrop and Sittard. However, it must pass protected zones in Flanders, and requires a tunnel under Weert. The shorter HSL proposed here follows existing infrastructure, the A2 motorway via Maasbracht. There is a disadvantage in following the motorway: the line might curve too much as it crosses the Maas, so that trains would slow down.

The original proposal suggested the line could be built in two phases, first from Weert to Echt (in black on the map), and then a northern bypass of Weert (in blue). However, the second phase may not be necessary, since there is space for a new curve on the western edge of Weert. In that case additional fast tracks would be needed, through Weert station.

Original version …

Hogesnelheidslijn Eindhoven - Sittard, vanaf Weert langs snelweg A2.

Both variants require four-tracking and upgrading of the existing line from Eindhoven to Weert. That is not described further here.

Alignment from Weert to Echt

The HSL would diverge from the existing line east of Weert, before it crosses the Wessem – Nederweert canal. The new line would cut through fields, to join the route of the A2 motorway. In fact, the line follows the 1929 canal, because the motorway was built alongside it. The terrain is level: there are no locks on the canal here.

The new line would fit between the motorway and the canal. For the first 6 km of the shared route, the motorway is right next to the canal, and here the motorway must be relocated. The HSL would take over the current northbound carriageway. There are no road crossings here.

After junction 41, the crossing with the old Napoleonic road (Napoleonsbaan), there is enough space for a railway between motorway and road. The next section of the alignment, which crosses the Maas, has three curves: at junction 41, at Wessem just before the river bridge, and at Maasbracht just after the bridge. There is enough space at junction 41 to accommodate a high-speed curve.


As it crosses the Maas, however, the alignment must fit between the canal, gravel pits, and the two villages of Wessem and Maasbracht. Probably, the HSL must stay within 100 m of the motorway, and so the curves will not allow high speeds as the line pases Wessem and Maasbracht. That will be partly compensated because trains will slow anyway, to climb to the bridge, and then accelerate downhill away from it.

The alignment would allow a station at Maasbracht, only 400 m from the centre of the village. The line here would run directly alongside the motorway, as far as junction 44, and require almost no demolition. A station before the bridge is problematic. It cannot be built at Wessem, because the line climbs and turns. It might be possible at Panheel, further back from the river. To accommodate the station an additional curve would be needed, and in any case, the site (at a brick factory) is too far from the three villages.

If only one station is possible, then there seems little point. Since this line is intended for high-speed long-distance trains, the station would require a separate regional service, in addition to regional services on the existing line. The disadvantages outweigh any benefits for Maasbracht.

From the Maas to Sittard

After Maasbracht, the HSL must cross the A73 motorway, and also pass under three power lines. The alignment must also turn southward, toward the line to Sittard, the Maas Valley line. The upgrading of this line was proposed here earlier, as part of a high-speed corridor along the Maas valley (Venlo – Sittard, in Dutch). The upgrade requires a realignment of the existing curve north of Echt, and that would also allow a junction with a high-speed line from Weert. It keeps the new line clear of the recently built logistics centre, north of Echt.


The new alignment via Maasbracht would end at this junction. From there to Sittard (13 km), the existing line must be upgraded, with four tracks. That should be relatively easy: it consists of two straight sections, with a curve between them. All trains would stop in Sittard, already a railway junction. South of Sittard, trains could continue on the HSL Sittard – Aachen, proposed here earlier. That HSL is not a precondition for any bypass of Roermond, but more capacity on the Sittard – Maastricht line probably is.

All Intercity services from Eindhoven south to Limburg, including future services to Aachen, would use the new cut-off line. The new section of line would be 19 km long. Including the upgraded existing lines, the new route between Weert and Echt would be 25 km long, as against 38 km for the historical route via Roermond. The reduction in journey time will be more significant: trains will be faster than current rolling stock, and the route avoids two stops in Weert and Roermond.

Through Weert or around Weert?

The original Dutch version of this post included a bypass of Weert, to avoid the sharp curve west of Weert station. The bypass would run close to the A2 motorway, but cannot follow it exactly, because it too has relatively sharp curves. It would probably require a tunnel, at the eastern edge of Weert.

If the curve was not so sharp, the new HSL could follow the existing alignment through Weert. Of course trains could also slow down at the curve, but that defeats the purpose of a high speed line. There is sufficient space in Weert for extra tracks. The line here was built as the ‘Iron Rhine’, a transit freight route from Belgium to Germany, which once carried heavy traffic. (The line Eindhoven – Weert was added later: the sharp connecting curve was not a problem at the time.)

It the curve can be improved, that would be easier than bypassing Weert entirely. That might be possible, if the new line diverges from the existing alignment outside Weert, and also at Weert Station.

Click to enlarge…


The new alignment could start about 4 km before Weert Station. It would first turn west, then back east, crossing the existing line twice. It would pass Weert Station on the site of the former goods yard, south of the present platform. It would rejoin the existing alignment about 1 km east of the station. A viaduct is the easiest option, but to reduce noise a tunnel might be preferable.

An alignment through Weert will make little difference to the total length of the route Eindhoven – Sittard. Journey time is saved because trains do not stop at Weert, and run at high speed through the built-up area. (Obviously there is no point in building such a high-speed route, if trains continue to stop at Weert).

The present Eindhoven – Sittard route, via Weert and Roermond, is 78 km long. The new bypass would cut that to 65 km. With no stops, a new middle section built for over 200 km/h, and the existing lines upgraded for 200 km/h, a journey time of 25 minutes is feasible. The present Intercity trains take 46 minutes, with two stops.

High-speed bypass of Roermond

Light metro in the Schiphol airport zone

This is an English language version of a 2009 proposal for a light metro network in the Haarlemmermeer region, around Amsterdam’s Schiphol Airport. It is intended to serve the region, rather than connect Amsterdam to the airport. To begin with, a short description of the airport location.

Schiphol Airport is located in a reclaimed lake: it is below sea level. The Haarlemmermeer, or Haarlem Lake was a large freshwater lake in the triangle between Amsterdam, Haarlem and Leiden. From the Middle Ages the lake grew constantly, as its banks were eroded in winter storms.

The lake around 1640…


By the 19th century all three cities were threatened by flooding from the lake. In 1848 it was enclosed by a ring dike and a ring canal, and pumped dry by 1852. At that time the cities were much smaller, and the reclaimed polder was used entirely for agriculture. There was one village, simply called Hoofddorp, ‘main village’. From the 1960’s on, airport development and suburban expansion of the adjoining cities and towns transformed the area. Hoofddorp itself now has 70 00 inhabitants.

Intensive use of the area, and the exceptional traffic flows, justify a metro here – even though it is not a city. The Municipality of Haarlemmermeer, which corresponds to the 1848 polder, has 145 000 inhabitants, and total employment is about 125 000.

Haarlemmermeer: map by Jan-Willem van Aalst, CC4.0 licence , click to download full-size image (32 MB)…


The region had its own transport network in the past: a local railway was built in the first decades of the 20th century (in Dutch: Haarlemmermeer-spoorlijnen). It had its own stations in Amsterdam, Haarlem and Leiden, and served the main villages in the polder. It was gradually abandoned, and the last passenger trains ran in 1950. Some sections survived until the 1980’s, as freight lines.

Metro system and other infrastructure

The proposal is a light automated metro, operationally separate from other infrastructure. That would simplify construction, and allow the vehicles to match the traffic levels. There would be no through trains onto the Amsterdam metro, which at present does not serve the airport anyway. Most of the new metro would be in cut-and-cover tunnel, some sections could run on the surface, or on viaduct.

The region is served by three rail lines. The oldest is the Oude Lijn or Old Line, from Amsterdam to The Hague and Rotterdam. It was built in 1847, before the lake was reclaimed, and therefore has an L-shaped route via Haarlem. A new cut-off line was built in the 1970’s, to shorten the route to Leiden, and to serve Schiphol Airport. It is now four-track, as far as Hoofddorp. In 2009 the HSL Zuid was opened, the start of the high-speed route to Paris. It diverges from the Schiphol line at Hoofddorp, but has no intermediate stations between the airport and Rotterdam.

A number of new lines were proposed here earlier, and some affect the alignment of the proposed metro. The most significant is a new high-speed bypass of Schiphol Airport, for which there are several route options. A bypass line is officially planned to serve a future second terminal, but it has so many advantages that it should be built sooner. Avoiding the congested tunnel under the airport, it would have a new station on the airport periphery, which could be served by the light metro. The map below shows the route options: HSL in blue, existing Schiphol and Haarlem line in solid grey.

Bypass spoorlijn naar Leiden en HSL, langs Schiphol.

The alignment of the proposed bypass also allows an additional link from the airport to Haarlem, avoiding the existing interchange station at Sloterdijk. The present public transport link from the airport to Haarlem is an express busway, the Zuidtangent, which has an indirect route through Hoofddorp. The Haarlem branch can be combined with a new line, from the south of Amsterdam to Haarlem, along the A9 motorway.

New diagonal line…

Spoorlijn langs A9 van Amsterdam-Zuid naar Haarlem.

On the western side, the light metro depends on a proposed relocation of the Old Line between Haarlem and Leiden (in Dutch, Verschuiving van de spoorlijn door de Bollenstreek). Although the railway would not move far, relocation would allow it to serve the villages of the flower-growing region, Bollenstreek. The light metro would connect with the relocated line at a new station in Lisse, or possibly at Hillegom. (That would serve as an alternative for a planned rail link between Schiphol and Lisse).

The light metro could also connect to a new regional line from Utrecht (in Dutch Regionale metro Utrecht – Aalsmeer). A regional metro around Utrecht is under construction, Randstadspoor. It uses existing rail lines, with extra tracks in some places. The proposed Aalsmeer line would simply extend one of those lines, using the alignment of a former local railway.

The proposed lines would partly follow the original drainage canals of the Haarlemmermeer polder. Present infrastructure often follows the original drainage canals: the long axis is aligned south-west to north-east, at 41 degrees. Here the metro lines are designated ‘blue’ (long axis) and ‘red’ (transverse axis). Like the drainage canals, they cross in the middle of Hoofddorp, originally the most accessible point in the polder. The lines are described in four sections, around that interchange station.

Axial metro: blue line

The blue line south of Hoofddorp would start at a new Lisse station, west of the centre, on the relocated line. The station would allow interchange with regional trains from The Hague and Leiden. The metro line would turn east, with a station of the north side of Lisse, and then cross the ring canal in tunnel. All roads and railways into the polder must cross this canal: the land in the polder is also 5 m lower. The line would have a third station at Lisserbroek, which is the part of Lisse inside the polder.

The line would now turn about 90 degrees, to follow a drainage canal (Nieuwerkerker Tocht). This alignment is not original: it was used by the former local railway. At the time, it was at the edge of a small village, Nieuw-Vennep. It has now grown to 30 000 inhabitants, and the metro alignment passes through the new suburban developments: it is in use as a busway. Nieuw-Vennep would have three metro stations, one at the western shopping centre.

Lisse Vennep tracé

In the fields north of Nieuw-Vennep, the line would turn toward the southern corner of Hoofddorp, and then turn again to follow the main drainage canal (Hoofdvaart). The central station would have interchange with the red line, another with the Zuidtangent busway.

North of Hoofddorp, the blue line would turn toward Schiphol Airport, possibly serving a fourth station at the edge of the town. The interchange stations at Schiphol determine where exactly the line turns east. It must cross the A5 motorway, a runway, and the A4 motorway, so a bored tunnel may be needed. A new high-speed line through the airport would probably be on the eastern side of the runway, with the station opposite Junction 2 of the A4. If so, then the metro can cross the A4 and terminate alongside the existing airport railway station.

The long-planned second terminal would be built about 2 km further north, and it would have its own station on any bypass line. It is probably easier for the metro to access this station site, but it would then need a U-turn to reach the existing terminal. That is not a logical route, especially since the express busway runs directly from Hoofddorp to the terminal. It would make more sense to extend the light metro toward Amsterdam, or alternatively to extend the Amsterdam metro, to the second terminal station. However, the second terminal project is uncertain anyway: government strategy has shifted to expansion of regional airports.

Second terminal

The blue line, from Lisse to the existing terminal , would be about 18 km long. The average station spacing is about 1600 m, and that is long for a metro: the vehicles should take account of that. Nevertheless, inside the built-up areas, the station spacing is comparable with urban metros.

A second branch is possible west of Hoofddorp, along the southern edge of Hoofddorp to Bennebroek and Hillegom. Like the line in Lisse, it would start at a relocated station in Hillegom, then run parallel to the relocated railway, and then turn east to the Haarlemmermeer polder. It would have two stations in Hillegom, one in Bennebroek, and one in Zwaanshoek after crossing the ring canal into the polder.

Bennebroek Hillegom metro

The line would then follow the Bennebroekerweg, the old road to Bennebroek, which is now the southern edge of Hoofddorp. Two stations would serve new housing which is relatively far from the centre of Hoofddorp. The branch would then turn 90 degrees, to join the line coming from Lisse, on the main canal axis through Hoofddorp. The additional line would be about 10 km long.

Axial metro: red line

The red line west of Hoofddorp would start at Haarlem Station. It would first run eastwards, and then turn south through the Schalkwijk housing development. The section within Haarlem would have 5-6 stations, and function as an urban metro connecting to the main station.

A bored tunnel will be necessary here. Two possible alignments are shown, one under an industrial zone, and another under low-rise housing. In Schalkwijk, the line would follow the Prins Bernhardlaan and a park strip, to exit the built-up area. It must then cross a lake or gravel pit, and the polder ring canal – probably still in tunnel.

Haarlem metro: map by Jan-Willem van Aalst, CC4.0 licence

Metro Schalkwijk

In the polder, the line would turn to join the Kruisweg, the polder’s transverse axis. At the regional hospital (Spaarneziekenhuis) it would join the route of the express busway Zuidtangent.

The metro would follow the Kruisweg axis through Hoofddorp, with four stations. At the old centre of the village it would cross the blue line, with an interchange station. The new shopping centre is further south, but still within walking distance.

As Hoofddorp

East of Hoofddorp, the red line would still follow the original polder axis, now a main road (N201). It would pass north of the existing Hoofddorp Station, but would have interchange with the Zuidtangent busway, which crosses the main road on viaduct. It would pass south of Schiphol Airport, but would serve new development along the N201 (Schiphol-Rijk). More development is planned south of the N201.

The metro line would cross the ring canal again, and follow the N201 through Aalsmeer. There would be two stations on this section: in the centre, and at Ophelialaan. Construction under this road will not be easy, but most traffic has already shifted to a new ring road. At the edge of Aalsmeer, the line would turn to follow the Legmeerdijk, possibly on viaduct. The next station would be at the main entrance to the Aalsmeer flower auction, the largest employer in the region after the airport. This is a suitable terminal station for the red line, which wold be about 21 km long.

Aalsmeer metro: map by Jan-Willem van Aalst, CC4.0 licence


As with the blue line, the light metro could be extended toward Amsterdam, or the Amsterdam metro could be extended to make an end-to-end junction. (The Amsterdam light-rail Line 51 terminates about 4 km away, but that line is behind downgraded to a tram). The red line could also be extended to Uithoorn, but that is not considered further here.

An additional branch in the Haarlem area could start at Heemstede-Aerdenhout Station, on the Haarlem – Leiden line. From there it would follow the roads as much as possible (Lanckhorstlaan, Heemsteedse Dreef, Cruquiusweg). Nevertheless it will pass under suburban housing, and a bored tunnel in Heemstede is needed, with two for three stations. The branch would cross the ring canal into the polder at Cruquius, where there would be a station, and then join the main route, at the regional hospital Spaarneziekenhuis. The branch would be about 6 km long.

Heemstede branch: map by Jan-Willem van Aalst, CC4.0 licence

Heemstede metro

The total length of this metro network is about 55 km. It connects the surrounding suburbanised villages with the airport and with Hoofddorp, and that is a great improvement on the existing infrastructure. (At present only Nieuw-Vennep has a rail link to both Hoofddorp and the airport.) The metro network would replace most of the existing bus lines, and bus services in the region would be restructured as feeder lines. This is a region with specific transport demand, and a light metro is the best infrastructural option.

Light metro in the Schiphol airport zone

Rail gauge in the Baltic States

Population density in the Baltic States is low, so local traffic alone would not justify the present rail network. In fact, these railways were built for long-distance traffic, from the start. Most of the region was part of the Russian Empire, and from the Russian perspective, the Baltic railways connected ice-free ports to the imperial hinterland, and to Moscow and Saint Petersburg. The railways typically ran east-west, and the networks in Estonia, Latvia and Lithuania have retained that structure. In German East Prussia there were some standard-gauge local railways, but the main lines were primarily a link to the rest of Germany. They typically ran north-east to south-west – towards Berlin.

The Russian main lines were Russian gauge, and the German lines were standard gauge. (There were some narrow-gauge local lines). In both World Wars, railway lines were converted and reconverted, as the ‘Eastern Front’ shifted. After the Soviet victory in 1945, all main lines were converted to Russian gauge, including those in northern East Prussia, now Kaliningrad Oblast.

New and upgraded rail lines in the Baltic States were proposed here earlier. A new standard-gauge railway through the Baltic States is officially planned, the Rail Baltica project. By definition it runs north-south, and the official planning isolates it from the ‘Russian’ network. Better integration with the existing network was also proposed here. That raises the question of the ideal arrangement of Russian and standard-gauge lines in the region. This post looks at the logic of gauge conversion. As with all posts here, political considerations are ignored.

The individual rail lines can be grouped by function and geography. The ‘east-west’ lines connect to specific ports: Tallinn, Ventspils, Liepāja, Klaipėda and Kaliningrad. All except Kaliningrad were originally Russian, but now Kaliningrad is the only one in Russia. Nevertheless all have traffic from Russia and Belarus.

A standard-gauge high-speed line Kaliningrad – Kaunas – Vilnius was proposed here earlier. The existing Kaliningrad – Vilnius line could remain at Russian gauge, or also be converted with parallel freight tracks. South of that main line are two lines connecting to the Polish network, which should logically be standard gauge.

HSL Kaliningrad – Vilnius: southern boundary of Russian gauge…

HSR Baltic - Nemunas

A 40-km section through Šiauliai is shared by two east-west rail routes: Klaipėda – Daugavpils and Liepāja – Vilnius. Logically, they would share the same gauge, and because of their relative isolation from other east-west lines, that can be standard gauge. The port of Klaipėda would then lose its Russian-gauge connection to the interior.

The Liepāja – Jelgava – Daugavpils line should be kept at Russian gauge, as a main freight link. With about 8 km of parallel tracks for the Vilnius line, Liepāja would also have a standard gauge exit from the port. The route Liepāja – Jelgava – Riga would also be Russian gauge. With separate tracks Jelgava – Riga, it can be combined with standard gauge on the existing north-south line Sovetsk – Šiauliai – Riga. The parallel high-speed line to Riga would also be standard gauge. Riga would be therefore be accessible by standard-gauge, for both freight and passenger trains.

The Ventspils – Riga – Daugavpils line should logically have the same gauge as the Liepāja – Jelgava – Daugavpils line. The east-west routes through Jelgava and Riga would therefore retain Russian gauge. The north-south high speed route through Riga to Tallinn would be built for standard gauge.

Russian gauge north of Ventspils – Daugavpils…

Daugava line

With this arrangement, all existing lines north of the Ventspils – Riga – Daugavpils line would remain at Russian gauge, with the high-speed line to Tallinn at standard gauge.


Three north-south high speed routes would pass through Kaliningrad Oblast. Two would converge at Sovetsk: Berlin – Riga and Poznań – Sovetsk – Riga. At either Riga or Jelgava, the combined HSL would converge with the proposed HSL Warsaw – Kaunas – Riga. (A possible Kaunas – Daugavpils standard-gauge line would be an addition to this pattern).

North of Riga there would be one standard-gauge route, the high-speed line to Tallinn. There might also be a high-speed cut-off line on the Riga – St Petersburg route, but that would use Russian gauge. The Warsaw – Vilnius – St. Petersburg line would remain at Russian gauge north of Daugavpils. (With the relative isolation of the Vilnius – Daugavpils section, it might be possible to have the break of gauge at Vilnius instead).

This pattern requires conversion of existing lines, and some new lines would be built at a different gauge. However, it does not need much additional infrastructure, just some sections of parallel track on the approach to cites. The longest section of mixed gauge route would be Jelgava – Riga, about 50 km.

Rail gauge in the Baltic States

Revised: Baltic States infrastructure

Updated with new maps:

Revised: Baltic States infrastructure