The 1333 km Warsaw – Vilnius – St. Petersburg line was built in 1862, entirely within the Russian Empire. (Poland was nominally a separate kingdom, but that meant little in practice). It was therefore built with Russian gauge, and operated as a unit. That all changed in the 20th century: the operation and function of the line were disrupted by two World Wars, and the subsequent boundary changes. At present it crosses four borders, crossing the EU external boundary three times, and it has a break of gauge. There are, not surprisingly, no through services, and the Belarus – Lithuania section is not in use.
Disused cross-border section: public domain image by Šarūnas Šimkus…
There are no technical or geographical obstacles to a major upgrade. The line runs through flat or rolling terrain, rarely above 200 m elevation. In Poland it runs through rural areas, and further north mainly through forest, with low population density. North of Białystok, most of the line is single-track. On flat terrain, the alignment consists of straight sections with short curves between them. In the rolling uplands, there is some continuously curving track.
Apart from a few bridges over the main rivers, no complex structures are required. The line speed after upgrading would simply be a function of the resources applied. In turn, that would be determined by the line’s function, and its place in the rail network. That is what this post considers.
The primary function of the line would not be travel from Warsaw to St. Petersburg. That was probably never the case – the line is too long for that. A theoretical upgrade for 180 km/h would allow a through journey of about 10 hours, but that is still very long, except for tourist traffic. Paradoxically, shorter journeys would benefit from the upgrade, and their relative share of traffic would probably increase.
With a limited role for end-to-end passenger traffic, then break of gauge is not a problem in itself. The issue is its location, and that is related to the general question of rail gauge in the Baltic States. In principle, north-south lines could be standard gauge and east-west lines Russian gauge. A general conversion to standard gauge would make sense in Lithuania, and in Kaliningrad Oblast. The lines in the former German territories (East Prussia) were all standard gauge originally. Between the two World Wars, much of the Lithuanian network was also standard gauge. And from 1918 to 1939, Poland extended to Vilnius and almost to Daugavpils, and the St. Petersburg line was standard gauge to Daugavpils Station. The line to Daugavpils apparently changed gauge four times during the 20th century, including a German wartime conversion in 1941, and ended as a Russian-gauge line in the Soviet Union.
Although the St. Petersburg line passes through Belarus, its route through Hrodna is isolated from the rest of that network. It could be converted to standard gauge, provided there is a short dual-gauge approach to Hrodna station from the north. The isolated branch to Druskininkai can also be converted without any problem.
In Lithuania, the line is also relatively separate from the rest of the network, except at Vilnius. There is a branch to Utena, at present freight-only, and the line to Pastavy in Belarus shares a station only (at Pabradė). However, the important section though the Vilnius agglomeration is shared with the Kaliningrad – Minsk route. For freight there is an easy solution – the existing southern by-pass of Vilnius could remain Russian gauge. For passenger traffic, however, parallel Russian and standard gauge lines through Vilnius seem unavoidable. The proposed high-speed line Kaliningrad – Vilnius would require a standard gauge line into the main station. With parallel routes through the station, the line from Lida in Belarus can remain at Russian gauge.
The route to Sankt-Peterburg
The urban region of Warsaw (Warszawa) has a population of around 3,5 million. For 60 km out of the city, the line would be used by the proposed high-speed line Warsaw – Kaunas – Riga. The St. Petersburg railway originally had its own station in Warsaw, the Vilnius Station or Warszawa Wileńska. That station is still in use for suburban trains, but rebuilt as a shopping centre. The two high-speed routes therefore require a good connection to the cross-city east-west line line and Warszawa Centralna, or possibly a new central station. They best option is a second tunnel Elsnerów tunnel – the other Elsnerów tunnel was proposed for a Warsaw urban-regional metro. (The planned Rail Baltica line would detour over a freight line).
Tunnel in red, official link plans in blue-green…
Via the tunnel trains would acces the original St. Petersburg line, now Polish line 6. The fastest trains would serve Białystok in Poland (population 295 000) and Hrodna (360 000) in Belarus. The line then enters Lithuania, and continues to Vilnius (population 540 000). Between Hrodna and Vilnius (155 km), it runs almost entirely through forest, including a national park. (Outside the two urban areas, only six settlements are large enough to justify a station).
The Warsaw- Vilnius route (630 km) is approximately comparable with the Warsaw – Berlin route. An upgrade for line speeds of 200 km/h is appropriate, comparable with the German Ausbaustrecke. The alignment is favourable for a high-speed reconstruction.
Beyond Vilnius, the 170-km line to Daugavpils has more curves, and might be upgraded for a lower speed, about 150 km/h. Again only a few settlements would justify a station, about six or seven. Daugavpils (population 96 000) is probably the best place for the break of gauge. At the station, the line shares a route with the Riga – Smolensk line, but only for about 1500 m. A new line into the station from the south, with a new bridge over the Daugava river, would shorten the route.
North of Daugavpils, the 530 km line to St. Petersburg serves fewer urban centres: the only other large city is Pskov, about half-way between the two. The 85 km section through the region of Latgale to Rēzekne (population 32 000) might justify a regional service. But even in Latgale, once ‘densely populated’ by Latvian standards, the current population could support only four intermediate stations. Rēzekne itself has two stations: to facilitate interchange with east-west services, all trains could pass through the northern station. They could then turn north on a new link, to rejoin the St Petersburg line.
Beyond Latgale is a classic ‘main line’, dominated by long-distance traffic. There are only a few towns, villages are small and far apart, and the line is single track and not electrified. Rural population density here is under 10/km2. On the 170 km of line between Rēzekne and Pskov, only three stations would be needed: Ostrov (population 21 000), Pytalovo (6000) and Kārsava (2500). Nevertheless upgrading must provide a double track electrified line, suitable for at least 150 km/h.
From Pskov (population 194 000), it is another 133 km to Luga (population 36 000), where the character of the line changes. From here on, the line is double-track and electrified, and settlements are closer together. From Luga it is another 95 km to Gatchina (population 96 000), at the edge of the St. Petersburg agglomeration. Here, the line from Tallinn joins the Warsaw line. A four-track upgraded route would be needed for the last 45 km from Gatchina, to separate freight and urban services.
The ‘Federal Subject Sankt-Peterburg’, Са́нкт-Петербу́рг, has five million inhabitants. It would be the main destination for passenger traffic originating north of Daugavpils. The original terminal of the Warsaw line has been converted to a museum, with long-distance services diverted to the Vitebsk Station, and suburban services to the Baltic Station. With major upgrading of the Warsaw line, the Vitebsk Station might be overloaded. The Baltic Station is better placed as a terminal for the Warsaw line, and has more than enough space for expansion, especially with new underground platforms.
Varshavsky (Warsaw) Station in Saint Petersburg, 2014: image by Alex (Florstein) Fedorov CC3.0 unported licence…
With break of gauge at Daugavpils, services north of that city would probably consist mainly of inter-regional trains for the five intermediate centres: Rēzekne, Ostrov, Pskov, Luga and Gatchina. (An intercity service, calling only at Gatchina and Pskov, would not be much faster). North of Luga there would be regional services into St. Petersburg, and from Gatchina a more intensive urban-regional service. Local services would continue for the few remaining village stations on the line between Rēzekne and Luga. However, most of the rural halts, which have no facilities and a minimal service, would simply close. If there are any through trains from Warsaw to St. Petersburg, they would be fast variable gauge trains, with gauge changing at Daugavpils station. Such a service might serve only the five largest cites: Białystok, Hrodna, Vilnius, Daugavpils, and Pskov.