High-Speed Rail

High-speed rail (HSR) is a class of railway passenger service that runs much faster than conventional trains, usually on purpose-built lines with gentle curves, few gradients, and no level crossings. Its defining feature is not one speed but a combination of design features that make sustained high speeds safe and economical. The first large-scale system, Japan's Tōkaidō Shinkansen, opened in 1964, and networks now operate across East Asia, Europe, and parts of the Middle East and North Africa, with more under construction. China has the largest network by a wide margin.

The subject matters because it connects several larger themes. Technically, it applies electric traction, signalling and control engineering, and aerodynamics to a single transport system. Economically, it raises questions about very large fixed infrastructure costs, the benefits of shorter travel times, and how a state-led project competes with air travel and cars. Environmentally and socially, it is often promoted as a lower-carbon alternative to flying, and it has been linked to urban growth, regional redistribution, and debates about who benefits.

This entry first defines HSR and explains the physical limits that shape it, including adhesion, aerodynamic drag, and the infrastructure needed to support high speeds. It then covers track engineering, electric traction, rolling stock, signalling, and the historical development of the technology. The second half surveys the global network and the main national programmes, including China, Europe, and the California project, followed by economic, environmental, and urban effects, competition with other transport modes, unresolved controversies, and future technical directions.

Definitions and speed thresholds

Official definitions

There is no single worldwide standard for high-speed rail. The International Union of Railways (UIC) is the most cited reference. Its principal criterion is a commercial speed of 250 km/h, which it treats as the jump that distinguishes high-speed service from conventional operation. It accepts a secondary criterion on average-distance corridors without air competition, where 220–230 km/h, or at least a speed clearly above 200 km/h, may capture most of the market. UIC also ties the label to the system as a whole: track, trainsets, signalling without trackside signals, long-range control centres, and separation of freight from passenger traffic.

European Union law uses a more formal three-part classification in the Interoperability Directive (EU) 2016/797. Specially built lines are generally equipped for 250 km/h or more. Specially upgraded lines are equipped for about 200 km/h. A third category covers upgraded lines where topography, relief or town-planning constraints require speeds to be adapted case by case. This third category is what allows high-speed trains to run into city-centre stations over conventional track.

National statistics do not always follow these rules. China's standards, for instance, separate high-speed railway (roughly 250–380 km/h), "rapid" railway (160–250 km/h) and conventional railway, and the Chinese design code uses 250 km/h as the design threshold for passenger-dedicated lines. A network total that counts upgraded 200 km/h lines in one country may therefore exclude them in another. Comparisons should always check which definition a figure uses.

Dedicated, mixed-traffic and "semi-high-speed" lines

Dedicated lines are new alignments built for passenger trains only, with large curve radii and full grade separation. Mixed-traffic upgraded lines keep freight, regional or slower passenger services on the same tracks, which lowers the maximum practical speed and reduces capacity. Many European upgrades fall into this group.

The term "semi-high-speed" is used for existing track upgraded to speeds that approach the high-speed threshold, typically around 200 to 250 km/h. Its status is contested. Critics argue that counting such upgrades as high-speed inflates network length without necessarily shortening journey times, because trains must slow on the unupgraded sections. Defenders argue that upgrades deliver most of the benefit at a fraction of the cost of new lines. Sources also disagree on the EU threshold for new track: some summaries present 300 km/h as the minimum for entirely new lines, while the directive's wording is usually summarised as 250 km/h "generally", with higher speeds in suitable conditions.

Design speed, operating speed and records

Design speed is the speed for which the alignment, superelevation, signalling and power supply are engineered. Operating speed is what timetables actually use, and it is often lower because of stops, urban sections, tunnels or mixed traffic. Italy's Rome–Florence Direttissima, opened in 1977, was designed for 250 km/h but was operated at about 200 km/h by locomotive-hauled trains.

Record speeds are a different category again. They are set on special test runs, often with modified trains and selected track sections, and they do not describe normal service. The current wheel-on-rail record of 574.8 km/h was set by a specially modified French TGV set on the LGV Est in 2007. The fastest speed ever recorded for a rail vehicle is the 603 km/h run by Japan's experimental L0 Series maglev on the Yamanashi test line on 21 April 2015, using a manned seven-car set. Its predecessor, the MLX01, had reached 581 km/h in 2003. Maglev trains are not wheel-on-rail systems, so they fall outside the UIC and EU line categories, even though the planned Chūō Shinkansen is designed to carry passengers commercially at about 500 km/h.

Brand names and technical categories

Brand names identify operators, manufacturers or marketing programmes, not technical classes. Shinkansen (Japan, since 1964), TGV (France, since 1981), ICE (Germany, since 1991), AVE (Spain), KTX (Korea) and the CR and Fuxing series (China) each cover trains that can operate on several kinds of line. A TGV, for example, runs at its highest speeds on dedicated LGVs but also on conventional track at lower speeds. Shinkansen services likewise combine dedicated lines with sections where speeds are limited by geography or urban areas. Reading a brand name as a guarantee of a particular speed or line type therefore leads to errors; the infrastructure, the trainset and the timetable each need to be checked separately.

Physical principles and speed limits

Speed on a railway is bounded by a few physical quantities, and the most basic is the grip between steel wheel and steel rail. Traction and braking forces pass through contact patches about the size of a coin, and the wheel–rail adhesion coefficient is roughly 0.25 in dry conditions. It falls with rain, leaves, or contamination, and it tends to decline as speed rises. The acceleration available from grip is therefore about $\mu g \approx 2.5$ m/s² on fully powered axles in good conditions, but practical acceleration at high speed is much lower because power, not grip, becomes the binding limit. Gradients work similarly. A 3.5 per cent grade requires about 35 N of tractive effort per kN of train weight just to hold speed against gravity, which is well within what the rails can supply but matters for heavy trains, restarting on a slope, and braking downhill. This is why high-speed lines are normally built with gradients of about 3.5 per cent or less, and why steeper sections are avoided where terrain allows.

Air resistance is the second constraint, and it grows with the square of speed. The drag force is $F = \tfrac{1}{2}\rho C_d A v^2$, so the power needed to overcome it scales with the cube of speed, $P = Fv \propto v^3$. Doubling speed multiplies aerodynamic power demand by roughly eight. At around 300 km/h, aerodynamic drag typically accounts for the largest share of running resistance on level track. A modest increase in speed therefore requires a disproportionate increase in installed traction power, which in turn raises the demands on motors, grid supply and substations. For a given power rating, this effect sets a practical speed ceiling for each class of train. The shape of the nose and the streamlining of the body are the main design responses, discussed elsewhere in this entry.

Curves impose a lateral constraint. A vehicle moving at speed $v$ around a curve of radius $R$ needs a centripetal acceleration of $a = v^2 / R$. To reduce the lateral load on passengers and the track, the outer rail is raised, a feature known as cant or superelevation. Part of the required sideways force is then supplied by the component of gravity along the inclined track plane. The equilibrium cant is $h = \dfrac{G v^2}{g R}$, where $G$ is the distance between rail centres (about 1.5 m for standard gauge) and $g$ is gravitational acceleration. For example, at 300 km/h (83.3 m/s) on a 7,000 m radius curve, the equilibrium cant is about 150 mm. Cant cannot be increased without limit, because slow or stopped trains then feel uncomfortably tilted. Designers therefore cap the residual lateral acceleration a passenger can experience, and the permissible values differ between national and European standards. Together these limits set the minimum radius of new alignments, which is why high-speed lines use curves of several kilometres in radius and why existing winding routes are often unsuitable without new construction or tilting trains.

Kinetic energy, given by $E = \tfrac{1}{2} m v^2$, grows with the square of speed, and so does the energy that brakes must dissipate. A 400-tonne trainset at 300 km/h carries about 1.4 GJ, or roughly 386 kWh, of kinetic energy. Braking distance follows the same square law. For a constant deceleration $a$, the distance is $s = v^2 / (2a)$. At 300 km/h, a deceleration of 0.5 m/s² implies a distance of about 6.9 km, and even 1.0 m/s² implies about 3.5 km. These distances are far longer than the spacing of stations and junctions, so the consequences extend well beyond the train. Signalling must supervise speed against continuous braking curves and provide block sections long enough to contain them. Alignments need long, clear sight lines and gentle transitions, and emergency braking must be engineered so that wheel slide does not further reduce the already limited adhesion. Regenerative or dynamic braking can recover part of the kinetic energy, but friction brakes remain essential for safe stopping.

Track and civil engineering

High-speed lines are built as dedicated alignments that carry passenger trains only and are fully grade-separated, so there are no level crossings. Geometry is the main constraint. The speed a curve allows depends on its radius, its cant (the outer-rail elevation) and the cant deficiency the passengers can tolerate. A simplified form of the relation is $$V = \sqrt{R\cdot\frac{I_{lim}+D_{lim}}{11.8}}$$ where V is speed in km/h, R is radius in metres, I and D are the permitted cant deficiency and cant in millimetres, and 11.8 is the constant for 1,435 mm standard gauge (The permissable speed for a standard-gauge railway alignment is governed by the following equation). Design standards therefore give curve radii as desirable, minimum and exceptional values, and a layout built from curves at their minimum values is generally regarded as a weak design (Alignments consisting of curves that are all at or approaching minimum values are an indication of a poor quality design.). On the Chongqing–Kunming line, designed for 350 km/h, the minimum radius is 7,000 m on normal terrain and 5,500 m on difficult terrain, with maximum gradients of 2% and 3% (Normal: 7,000 m (4.3 mi); difficult: 5,500 m (3.4 mi)). Gentle gradients matter because they limit the effort needed for acceleration and braking. Flatter alignments, however, need more earthworks, bridges and tunnels, which are extremely expensive.

The track structure is the other major choice. Traditional ballasted track places sleepers in a gravel bed, which gives some resilience and can be re-levelled by tamping. Slab track replaces this bed with a rigid concrete or asphalt base, and it is considered the standard for high-speed and heavy-haul railway lines. In many systems the rails are fixed to the slab through elastomeric pads. Practice varies by country. For example, France and Spain adopt a predominant ballasted network, while Japan's high-speed network is roughly 70% slab track, although its first line, opened in 1964, used ballast. Research comparisons generally find that ballasted tracks exhibits higher train-induced differential settlement compared to slab, and the gap widens at higher line speeds. The trade-off is construction precision and cost: slab must be cast to tight tolerances, and once the concrete sets, correcting its geometry is difficult and expensive, unlike ballasted track which can be adjusted through tamping. Cost comparisons also depend on lifecycle assessment. One Japanese comparison suggested that the lower maintenance costs associated with slab track can offset the higher construction costs after 9 years. The Cologne–Rhine/Main line, opened in 2002, used ballastless track on about 150 km of its length, with a minimum radius of 3,350 m and a maximum gradient of 40‰.

Viaducts and tunnels account for a large share of many new lines, especially in mountainous or densely built areas. Route design tends to use a large proportion of tunnel and viaduct (in hilly areas) or elevated structure (in flatter areas), the latter to conserve farmland and minimize severance. The World Bank's review of early Chinese lines reported that the 835 km Guiyang–Guangzhou line had 54% of its length in tunnel and 29% on viaduct and bridges. Some lines are dominated by elevated structures; a section of Beijing–Guangzhou line is 69% of its length on viaduct and bridges. Individual structures can be very long. The Tianjin Grand Bridge on the Beijing–Shanghai line is 113.7 km (70.6 mi) long, and a tunnel under the Yangtze, linking Shanghai and Jiangsu, is 14.25 km, of which 11.18 km runs beneath the river. Tunnels create engineering problems beyond cost: a train entering at high speed compresses the air ahead of it, and the resulting pressure waves cause noise and discomfort at the exit, which is why tunnel portals and nose shapes are designed together with the track.

Keeping the track at high-speed tolerances requires continuous monitoring and maintenance. Inspection cars and trackside sensors measure gauge, cross-level, alignment and twist, and the results trigger localised tamping, rail grinding to remove corrugation and restore the rail profile, and planned renewal of rails, fastenings, sleepers or ballast. Ballasted track needs frequent tamping as the bed settles, while slab track needs fewer interventions but more targeted repair of cracks, bearing pads and joints. The required regime depends on the design speed, traffic density and the track type chosen at construction.

Electric traction and power supply

Overhead supply and the choice of voltage system

Most high-speed trains draw current from an overhead catenary through a roof-mounted pantograph. The main technical choice is the supply system. Most new high-speed lines in Europe use 25 kV at 50 Hz, and Japan's Shinkansen uses 25 kV 50/60 Hz AC systems. This choice is tied to grid frequency: the supply runs at the standard utility frequency, which simplifies traction substations. Older European networks rely on 3 kV direct current, and the Italian operator FS faced this directly. It electrified its new high-speed lines with 25 kV alternating current instead of the 3 kV direct current used on the classic network, because 3 kV is technically limited to 250 km/h operation, which was too slow for 300 km/h service. Some lines still use 3 kV, and dual-voltage trainsets such as the P (politensione, multi-voltage) variant can run on both systems. Cross-border operation adds further complexity. Multi-system locomotives such as the Vectron family are designed for cross-border and cross-system services between different countries and railway systems.

Distributed versus locomotive-hauled traction

Japan and Europe took different routes. The shinkansen is powered by electric motors mounted in multiple locations, while the TGV and ICE use a locomotive to haul unpowered cars. In a distributed design, traction equipment is spread along the train, which has several engineering consequences. This reduces the maximum axle load to 16 tonnes, which eases stress on track and bridges. Because the powered axles carry the train's weight, they have a higher power-to-weight ratio than a locomotive-hauled train, and this also gives them greater adhesion on gradients. The drawbacks are real. Distributed trains need very many more traction motors, which increases maintenance and spare-parts costs, and the noise in passenger saloons is higher than with concentrated power (locomotive) trains. Regulation also shapes the choice: in the United States, rules for services above 125 mph have been read as requiring locomotives, especially to meet the crash safety requirements. Even so, the Shinkansen and the ICE3 show that MUs are becoming an interesting option for high-speed service as well.

Traction substations and grid loads

Trains draw power from traction substations, which take power from the public grid and feed the catenary through feeder circuits. Substation size varies widely. One study of a Chinese high-speed line reports that the rated capacity of a TSS can reach up to 30 MW, with some substations on the Beijing-Shanghai high-speed railway handling up to 50 MVA. The loads are lumpy. In the same study, the peak traction power reaches 25 MW, and the lowest regenerative braking power reaches −18 MW, meaning that braking energy can be returned at levels comparable to the peak draw. Because a single-phase railway load sits on a three-phase grid, the transformers must be arranged carefully. Scott or similar balanced connections are used so that when energy flowing through the α and β port is equal on the single-phase railway side, the VU will be naturally eliminated in three-phase grid side. Utilities may also treat returned energy as a commercial matter. One analysis notes that power grid companies will charge for regenerative braking energy feedback without considering reverse transmission.

Regenerative braking and energy use

Electric motors can run in reverse as generators during braking, converting the train's regenerated energy into electricity. Where that energy goes depends on the system. The most efficient outcome is that it is absorbed by an adjacent accelerating train on the same electrical section, and the efficiency of that transfer depends on closely spaced substations, where the resistance is low and transfer efficiency is high. The alternative is storage or feedback to the grid. On AC 25 kV systems, modern four-quadrant converter technology allows regenerated energy to flow naturally back through the traction transformer to the grid, whereas DC systems need reversible substations with active front-end inverters. Storage devices are being tested as well. In one simulation of a Chinese traction substation, a supercapacitor device reduced energy consumption by 4.39% compared with the case without storage. Energy efficiency also depends on the train design rather than traction alone. One comparative database estimates an energy efficiency potential for single vehicle: 5 - 10% from multiple-unit designs, but only 1 - 2% across a whole fleet, which shows how sensitive these figures are to scope. Estimates of this kind should be read with the assumptions in mind, since electricity generation mix and traffic patterns also decide the final footprint of a line.

Rolling stock and vehicle design

The body of a high-speed trainset is shaped by three competing demands: aerodynamic drag, the loads the vehicle places on track and structures, and the need to keep passengers comfortable on curves. The nose is the most visible compromise. A short, blunt nose saves interior space, but when a train enters a tunnel it compresses the air ahead of it, and that pressure is released at the far portal as a sharp impulse known as a micro-pressure wave. Optimisation studies have found that shaping the nose's cross-sectional area along its length can reduce the intensity of the micro-pressure wave by 18–27% compared with a parabolic nose. Longer, more gradual noses cost driver-cab and passenger space at the front of the train, so designers trade aerodynamic performance against seating. On the infrastructure side, tunnel entrance hoods are required for long concrete slab tunnels to limit the same effect. The Shinkansen 500 Series, whose nose was widely reported to be modelled on a kingfisher's beak, is the best-known example of a nose designed for both drag and tunnel performance.

Mass is the second constraint, and it is measured per axle as much as per train. The axle load determines how much stress the rails, sleepers, and bridges absorb, and at high speed repeated axle loads can excite resonance in bridge decks. European design rules reflect this: the high-speed load model used for bridges is valid for constant axle loads below 170 kN each, which is roughly 17 tonnes. To stay within such limits, builders use aluminium or stainless-steel car bodies, lightweight bogies, and equipment spread along the train, while keeping the total mass low enough to limit energy use and wear. Weight also directly affects curve performance. The Italian ETR 610 was too heavy for the main Gotthard route from Zurich to Milan for which they had been ordered, and its weight forced a lower permitted cornering speed. By contrast, the ETR 460 maintains a low axle load of 14.5t/axle that allows the train to run 35% faster on curves compared to conventional trains.

Tilting trains address the curve problem from the vehicle side. Rather than building new straight alignments, a tilting train leans its car body into a bend, which reduces the lateral acceleration passengers feel. The Pendolino family, developed by Fiat and later built by Alstom, is the best-known case; its maximum tilt of 8 degrees allows the trains to reach a speed that is up to 35% higher than conventional trains. The system depends on on-board sensors and gyroscopes to detect curves, and a delay in detecting a curve entrance can cause discomfort, which is why later designs use more sensitive control. Tilt is not a substitute for dedicated lines at the top speed. The UK Class 390 has a design speed of 140 mph (225 km/h); however, limitations to track signalling systems restrict the trains to a maximum speed of 125 mph (200 km/h) in service. Likewise, the Finnish Pendolino cannot run at as high a speed as, for example, the TGV, due to the lines, although it can share track with conventional trains.

Noise and passenger systems form the final group. Noise sources include the pantograph, which collects current from the overhead wire, and aerodynamic and rolling noise from the train body and bogies; mitigation measures generally focus on pantograph shrouds, bogie covers, and smooth, well-maintained wheels and rails, together with noise barriers along exposed sections. Inside the train, operators treat information displays and connectivity as part of the product. During refurbishments, for example, operators have added power sockets and new customer information screens, and newer fleets increasingly include onboard Wi-Fi and digital passenger information. These features add weight and electrical load, which feeds back into the axle-load and power-supply constraints discussed above.

Signalling, control, and safety systems

ETCS Level 2 and cross-border interoperability

Modern high-speed signalling has shifted from lineside signals to continuous computer-based supervision. Under ETCS Level 2, the trackside sends movement authorities by radio from a Radio Block Centre (RBC), using GSM-R (with FRMCS as its planned successor). The onboard computer then calculates the permitted speed and braking curve from that data. Eurobalises along the track mainly provide position references, which distinguishes Level 2 from Level 1, where authority arrives at fixed points. Because the onboard and trackside functions follow a common European specification, a train fitted with ETCS can cross national borders without changing its signalling equipment. Level 2 can also be installed as an overlay on an older line that still has lineside signals and train detection, which makes it a migration path as much as an end state. Level 3, which would replace fixed block sections with moving blocks, remains at the conceptual or experimental stage.

National systems and the cost of mixed standards

Europe is not the only model. China's CTCS family is graded: CTCS-2 uses track circuits for continuous movement authority on 200–250 km/h lines, while CTCS-3 uses GSM-R and an RBC on lines of 250 km/h and above. CTCS-3 is broadly comparable to ETCS Level 2 and is normally backed by CTCS-2 as a fallback. Japan's system has a different origin. The Shinkansen's automatic train control (ATC) was designed in the 1960s because trains travelling at high speed leave a driver almost no time to react to trackside signals. Speed limits are transmitted to onboard equipment through the rails, and the brakes are applied automatically if the train is too fast. Taiwan's Shinkansen-derived trains illustrate how systems are adapted when a line is built under different standards: European safety requirements led to extra features beyond the Japanese original. Where trains must run on lines with different systems, operators have to fit multi-system onboard equipment, which adds cost, weight, and maintenance and complicates driver training.

Automatic protection, operation, and capacity

Automatic train protection (ATP) enforces speed limits whatever the driver does, while automatic train operation (ATO) controls acceleration, coasting, braking, and stopping at stations. ATO can reduce driver workload and energy consumption; China put Level 2 with ATO into service on the Guangdong intercity railway in 2016 at a maximum of 200 km/h. Capacity depends largely on braking distance. At 300 km/h a train covers about 83 metres every second, and an emergency stop can need several kilometres, so the length of each block section and the kinetic energy to be dissipated limit how closely trains can follow one another. Radio-based control with continuous updates allows shorter headways than fixed-block signalling, which is why the Tōkaidō Shinkansen could run peak headways of three to five minutes using tightly coordinated ATC and dedicated track. JR East's digital DS-ATC moves most speed-supervision logic onto the train, which reduces trackside equipment and allows smoother, one-stage braking commands.

Safety record and how accidents shaped the rules

Japan's Shinkansen is built on a crash-avoidance principle: a dedicated track free of level crossings, combined with ATC that prevents train-on-train collisions and excessive speed. Operators report no passenger fatalities since the line opened in 1964 and no collisions between Shinkansen trains in the network's history. These figures come from operators and industry bodies, and they should be read as reported records rather than independent audits. The ATC was revised after accidents in 1973 and 1974, a pattern common to many safety regimes, where investigations feed back into equipment and procedures. Earthquake warning systems also stop trains automatically. The crash-avoidance philosophy has a notable consequence: most developed-world regulators require collision-safety standards for high-speed rolling stock, whereas Shinkansen rolling stock is not subject to equivalent collision standards, because prevention is treated as the primary defence. Radio-based systems add a further concern, cybersecurity, which must now be assessed alongside derailment and collision risks.

Historical development

The first dedicated high-speed railway was a response to congestion. By the late 1950s, the Tōkaidō corridor linking Tokyo, Nagoya and Osaka was close to its capacity limit, and planners wanted a separate route for fast passenger trains Postwar Japan faced an urgent transport challenge as the Tokaido corridor between Tokyo and Osaka reached near-capacity by the late 1950s. The idea was tied to the shortage of capacity on the existing main line, which carried traffic between Japan's three major metropolitan areas The idea for the Shinkansen came in response to the severe transportation capacity shortages on the existing Tokaido Main Line. The Tokaido Main Line was a critical route linking Japan's three major metropolitan areas—Tokyo, Nagoya, and Osaka.. Engineer Shinji Sogo promoted the concept in 1955, and Hideo Shima led the technical development In 1955, engineer Shinji Sogo promoted the idea of a dedicated high-speed railway. Chief engineer Hideo Shima led the technical development of the project. Construction began in 1959 on a new route of about 515 km, with tunnels, bridges and viaducts Construction began in 1959 on a completely new 515-kilometer line featuring tunnels, bridges, and purpose-built stations across difficult mountainous terrain.

The Tōkaidō Shinkansen opened on 1 October 1964, shortly before the Tokyo Olympics. It is often described as the first railway built for regular service above 200 km/h it was the first railway in the world built for regular operation at speeds above 200 km/h (120 mph) and became the model for later high-speed rail systems. Reported travel-time savings vary by source. One source gives roughly three hours and ten minutes, compared with about seven hours on the conventional line reducing the travel time between the two cities from 7 hours by existing railway to a mere 3 hours and 10 minutes. Another describes a cut to about four hours. Opening-day top speed is also reported inconsistently, as 210 km/h in some accounts and 220 km/h in others. Ridership grew quickly, from about 11 million passengers in fiscal 1964 to more than 100 million by fiscal 1973 Annual passenger numbers, which stood at approximately 11 million in fiscal year 1964, grew consistently. It surpassed 100 million in less than a decade (fiscal year 1973).

France followed with the TGV. The Paris–Lyon line was inaugurated on 22 September 1981 by President François Mitterrand, and commercial service began five days later. The first run averaged about 260 km/h and cut the trip from about 3 hours 40 minutes to 2 hours 40 minutes, but only two-thirds of the route had been upgraded for high speed at that point The first trip between Lyon and Paris took 2 h 40, at an average of 260 km/h (compared with 3 h 40 previously). It took only two-thirds of the route having been adapted for high speed. The line was finished in 1983 It would take until 1983 for the line to be completed, bringing the two cities to just two hours by train.. In its first ten months, the line carried about ten million passengers ten million passengers travelled on the LGV Sud-Est within its first ten months of operation while domestic flights between Paris and Lyon declined substantially.

Germany's ICE (1991), Spain's AVE (1992), Korea's KTX (2004), Taiwan's system (2007) and China's first intercity lines (2008) followed. These are listed here from general knowledge rather than from the sources gathered for this section. Several of these systems were based on foreign designs. The TGV design, for example, was adapted for later systems: Derivative versions operated by Eurostar and national companies in South Korea, Spain and the US. Japan's influence on Europe is also widely acknowledged; one account states that France's TGV, Germany's ICE, and Spain's AVE, all owe their existence to Japan's groundbreaking project.

Speed records need to be separated from operating speeds. The 1990 TGV record was set on 18 May 1990 at 515.3 km/h The culmination of these test programs was a new world speed record of 515.3 km/h (143.1 m/s or 320.3 mph), set on 18 May 1990.. The run used a heavily modified trainset, not a standard service vehicle Extensive modifications. Set formed of 5 cars.. A later wheel-on-rail record of 574.8 km/h, set in 2007, was also achieved with a specially configured train.

Magnetic levitation (maglev) has its own record history, which is kept separate from wheel-on-rail records. Japan's experimental vehicle reached 517 km/h in 1979 during testing, and in 2015 the L0 Series reached 603 km/h on a 42.8 km test track. Shanghai's Transrapid service recorded 501 km/h in 2003 in a test run before commercial operation began in 2004 501 km/h (311 mph) 2003-11-12 Shanghai Maglev Train China Transrapid SMT ... Recorded in 2003 at a test run before the commercial operation in 2004.

Across these national programmes, the early pattern was licensing, foreign supply and adaptation, followed by domestic design. Japan's early system was built with indigenous engineering, while later networks drew heavily on imported technology, and national manufacturers gradually developed their own trainsets. The exact path differed country by country, and the sources gathered here do not allow a complete comparison of licensing terms.

Global network scale and geography

Reported network lengths are hard to compare, because each figure depends on its definition, reference date and source. China is the clearest case. The national operator reported about 50,000 km at the end of 2025, a milestone reached after adding roughly 5,000 km of new lines in two years. Earlier figures differ by method: the Ministry of Transport gave 48,000 km for the end of 2024, while the UIC atlas, which uses a 31 December 2024 extract and counts only the "in operation" column, gives 47,803 km for China. The gap is small, but it shows how the reference date and data cut change the total.

Spain's network is usually cited as the largest high-speed rail network in Europe with 3,973 km and the second-largest in the world, trailing only China. Older or narrower figures circulate as well: an ADIF-maintained network of 3,762 km in December 2021, and a 3,402 km figure in an industry source. The spread reflects both the date of the count and whether upgraded lines are included. France shows a similar problem. One reference lists 2,734 km of dedicated high-speed lines and 929 km of upgraded lines, while another gives 1,876 km of LGV lines. These totals measure different things and should not be placed side by side.

For Japan and Germany, the most recent comparable figures I could verify are from the UIC's 2020 table, which lists 3,041 km in service for Japan and 1,571 km for Germany. These are dated and should be updated against the current UIC atlas before use. The 2025 atlas ranks China first, followed by Spain, Japan, France, and Germany, so the order is stable even where the exact lengths are not.

Global totals also vary by edition. UIC describes almost 56,000 km of lines in operation worldwide, but its 2022 atlas reported about 59,000 km, and the 2024 edition cited roughly 65,000 km of lines identified in all stages. The 2025 edition puts the total at over 67,700 km across all project stages, with over 16,000 km of new lines currently under construction.

Geographically, East Asia dominates. In the 2022 atlas, Asia Pacific had more than 44,400 km of lines, followed by Europe with almost 12,000 km and the Middle East with about 1,500 km, while North America and Africa had 735 km and 186 km. In Europe, the network runs predominantly in Western Europe, with comparatively few lines in Eastern Europe. Expansion programmes are now active in the Middle East, Africa, North America, and emerging markets in Asia and Latin America. Southeast Asia is a newer and smaller cluster, anchored by a few lines rather than a continental grid.

The distinction between operating, under-construction and planned lines matters because it can change a country's position quickly. In the UIC's 2020 table, Turkey had 724 km in service and 1,743 km under construction, so its ranking would look very different once those lines opened. Any comparison should therefore state which category is being counted.

Data sources also diverge. The UIC atlas is compiled from direct input from UIC members and is the most consistent cross-country reference, but national figures come from ministries and infrastructure managers such as ADIF, and encyclopedic compilations often repeat those national numbers without the same definitions. When sources disagree, the discrepancy should be reported with the reference year, the category (in operation, under construction or planned) and whether upgraded lines are included.

China's network and industrial model

China's high-speed network grew from a single short intercity line into the largest in the world in under two decades. Its starting point was the Beijing–Tianjin intercity link in 2008. By the end of 2025, the total railway length reached 165,000 km, of which 50,400 km were high-speed lines. The high-speed total grew by over 12,500 km between 2021 and 2025, which shows that most of the growth came in the 14th Five-Year Plan period. Official figures also differ by source and year. For 2024, the UIC reported 47,803 km for China, while the Ministry of Transport reported 48,000 km, so totals should be compared only on the same definition.

The next phase is set by the 15th Five-Year Plan, which covers the years from 2026 to 2030. China State Railway Group expects high-speed route length to reach 60,000 km by 2030, and the total national network to reach 180,000 route-km, with 64% double track and 78% electrified. Other outlets give different targets; one travel guide cites a broader modern-rail goal of about 70,000 km, including more than 45,000 km of high-speed line, so the figures should be checked against the operator's own documents. In 2026 the government plans for more than 2,000 kilometers of new national railway lines to be put into service. Emphasis is also shifting. Planners want to coordinate new line construction with the upgrading and capacity expansion of conventional railways, and commentators describe a move from pure expansion toward linking existing lines and removing bottlenecks.

Rolling stock development follows a similar pattern of domestic consolidation. CRRC's Fuxing series operates at 350 km/h on many lines. The CR450 is the next generation: it is designed for commercial speeds of up to 400 km/h. After two CR450 prototype trainsets rolled off the production line at the end of 2024, they underwent type tests on the Beijing Ring Railway test line and on the Chongqing–Xiamen and Shanghai–Chongqing–Chengdu lines. The prototype reached 453 km/h during trials. The project is also meant to reduce reliance on imported parts, with major components including bogies, pantographs, and high-voltage cables being domestically developed. Sources disagree on timing. One report expects that the two prototype trainsets will enter commercial service during 2026, while the official line emphasises that the project will complete operation assessments and design finalization in 2026, moving a step closer to commercial operation. Full testing requires a total required mileage of 600,000 km before series production.

Financing relies heavily on state-owned entities rather than on a single budget line. China State Railway Group, the national operator, works with provincial railway companies and local government investment vehicles that hold equity in individual lines. Policy and commercial banks supply much of the long-term debt, and joint ventures with local governments and other investors are used on some projects. Railway investment is large: in 2025 alone, investment in fixed assets in the railway sector reached 901.5 billion yuan, and in early 2026 it expanded 3.2 percent year-on-year over January–April. This model allows rapid build-out, but it raises debt questions. Analysts point to the gap between capital costs and fare revenue on many lines, and to borrowing by local vehicles that may sit outside formal government debt totals. How far the national system's liabilities are backed by future operating cash flows remains a contested question.

China has also tried to export its high-speed package, which typically combines design, rolling stock, signalling, construction, and financing. The Jakarta–Bandung high-speed line in Indonesia, opened in 2023 through a consortium with Chinese majority participation, became a prominent example, and cost overruns and debt restructuring later drew scrutiny. In Southeast Asia, the Laos–China railway, opened in 2021, follows a similar model. In Africa, Chinese-financed and Chinese-built standard-gauge lines in Kenya and Ethiopia, the latter linking Addis Ababa and Djibouti, are often cited as examples of the same approach, which spreads Chinese technical standards alongside equipment and finance. Whether these projects bring lasting technology transfer or mainly serve export markets is a question on which evidence is still mixed.

European networks and cross-border links

European high-speed rail grew up as a set of national networks, each organised around radial trunk lines. France's TGV, Germany's ICE and Spain's AVE all follow a hub-and-spoke pattern, with lines fanning out from Paris, Frankfurt and Madrid and feeding regional and commuter services. Cross-border operation was added later and remains narrower in scope. Eurostar runs through the Channel Tunnel between London and Paris and Brussels. Thalys once linked Paris with Brussels, the Netherlands and western Germany, and its brand was later absorbed into the wider Eurostar offer. Because each country has its own signalling, power supply and operating rules, a train crossing a border often has to carry several onboard systems, which adds cost and complexity.

The EU's Trans-European Transport Network (TEN-T) is the main policy framework for closing these gaps, and the Connecting Europe Facility (CEF) is its principal grant instrument. For 2021–2027 the CEF allocates €25.8 billion to transport, and roughly €7 billion of that is earmarked for modernising the TEN-T, including ERTMS. In the current multiannual financial period (2021-2027) the new CEF instrument dedicates €25.8 billion to the transport sector. Within it, around €7 billion are allocated to actions relating the modernisation of the TEN-T network such as smart and interoperable mobility including ERTMS. ERTMS, the European Rail Traffic Management System, is a common signalling and train-control standard and is treated as a horizontal priority across the programme. Cohesion funds such as the ERDF and the Cohesion Fund can also co-finance ERTMS. The regulation sets deadlines of 2030 for equipping the Core Network Corridors and 2050 for the entire network. Progress lags behind: an analysis by Transport & Environment found that for most member states less than 20% of their TEN-T lines are equipped with ERTMS. The Commission estimates that EUR 24 billion is needed for deploying ERTMS trackside on the entire TEN-T Network, plus about €5 billion for rolling stock.

The Brenner Base Tunnel, between Austria and Italy, is the best-documented example of a cross-border bottleneck being addressed. The old Brenner line has gradients of up to 2.6% and tight curves, whereas the new line is planned with a maximum gradient of 0.4–0.7% Tight curve radii and gradients of up to 2.6% complicate rail operations. The new Brenner line would have a maximum gradient of 0.4%–0.7%. In July 2026 the two main tubes were connected beneath the Alps, leaving the project with all three cross-border tunnels complete. The 64km Brenner Base Tunnel, a core element of the Scandinavian-Mediterranean TEN-T corridor, is scheduled to be completed in 2031 and enter service at the end of 2032. Schedule and cost estimates have shifted: one source puts the project about 16 years behind its original timetable at roughly €8.5 billion, while another gives a total cost of €10.535 billion, so the figures should be checked against the same definition.

The Lyon–Turin line is a second major project, with its 57.5 km Mont Cenis base tunnel forming the central cross-border section. Estimated to cost $9.1 billion, the cross-border portion is 40% financed by the European Union, 35% by Italy and 25% by France. Official completion dates vary by source. One project description gives 2030, a construction report cites 2032, and a European agency page gives 2033. All contracts related to the civil works (for a total €4 billion) were awarded, and ten construction sites are currently in operation. The passenger design speed is modest by high-speed standards, with the base tunnel's passenger trains planned at about 220 km/h.

The Fehmarn Belt Fixed Link, a planned combined road and rail tunnel between Lolland in Denmark and Fehmarn in Germany, is expected to open in the early 2030s and will close a gap in the Scandinavian–Mediterranean corridor. Its completion date has been revised more than once, so current schedules should be checked directly against project documentation.

Rail Baltica, which will connect Estonia, Latvia and Lithuania with Poland and the wider network, involves converting the Baltic states' Russian broad gauge (1,520 mm) to standard gauge. One CEF allocation round granted The Rail Baltica project to build a standard-gauge railway connecting Estonia, Latvia and Lithuania will receive €352.1m. That grant is far smaller than the EU funding of more than €4 billion often cited for the project, so the two figures refer to different funding instruments or periods. Published route lengths also conflict, with figures of about 870 km and about 1,230 km in circulation. The gap probably reflects differences in whether connecting and regional sections are included, and it needs to be reconciled against the project's own documentation before either figure is used.

California High-Speed Rail project

California's project was approved by voters in 2008 as a San Francisco–Los Angeles line, but its schedule and budget have moved far from the original promises. Proposition 1A, a $9.95 billion bond measure, funded construction of a system meant to run at up to 220 mph. The ballot measure promised a $33 billion total cost, a journey time of 2 hours 40 minutes between San Francisco and Los Angeles, and completion in 2028. Some reports give the approved bond total as $33.5 billion, so the figure varies by source. The 2028 target has passed. The current plan puts the first operating segment, Merced to Bakersfield, in 2032 or 2033. The Authority's own timeline depends on lawmakers passing its legislative priorities quickly, which the state Office of Inspector General called "overly optimistic".

Construction is under way in the Central Valley, but the reported lengths differ depending on what is counted. Construction so far has been limited to three construction contracts running from just north of Madera to the Kern County community of Shafter, a span of 119 miles. Officials describe a larger 171 miles currently under design and construction from Merced to Bakersfield. The difference reflects the extension of the early operating segment into downtown Merced and Bakersfield, which earlier plans described as adding about 52 miles. More than 80 miles of guideway are complete, along with 61 major structures fully completed and 30 more underway, according to the Authority. In June 2026 the board approved a request for qualifications for a Merced-to-Madera civil works contract, which the Authority estimates at approximately $2.4 billion, with major construction on this new segment anticipated from late 2027 through 2030. The sources I reviewed describe track installation as starting later in 2026, but I could not confirm a separate 2026 approval of a dedicated track and systems installation team. Meanwhile, most of the Authority's proposed 2026 legislation failed to pass, and the existing law caps spending outside the Merced–Bakersfield portion at $500 million from the state climate fund.

Cost estimates in 2026 are not directly comparable because they cover different scopes and assumptions. The draft 2026 business plan put cumulative spending for the Merced–Bakersfield segment at an estimated $34.76 billion, while another report gave $35.7 billion for the same segment, with completion in 2032 and about $15.3 billion already spent. For the full Phase 1 line between San Francisco and Los Angeles/Anaheim, the plan gives $126.20 billion total, and other outlets cite a broader range of $89–$128 billion. Some reporting goes higher: the total cost of completing Phase I's 494-mile high-speed line is estimated at $231.3 billion, and a spokesperson described the $231 billion figure as a high-end, un-optimized scenario based on legacy design and delivery assumptions. The Authority also claims that a bottom-up review identified about $105 billion in avoided costs through right-sizing. The segment's financial case is also weak on the Authority's own numbers. The Legislative Analyst's Office calculated that, after federal grant losses, funding would fall about $2 billion short of the segment's cost once borrowing costs are included, and the plan projects annual ridership of just 1.4 to 1.9 million for the Merced–Bakersfield segment.

Federal support has largely ended. In July 2025 the Federal Railroad Administration terminated the funding effective immediately, removing about $4 billion in grants. California sued, but on Dec. 23, 2025, California dropped its lawsuit, and the state said federal sources had supplied only 18% of program expenditures. The state's main ongoing source is a cap-and-invest allocation: lawmakers committed $1 billion a year to the project through 2045, but state leaders and the High-Speed Rail Authority have said that level of funding will not be enough to finish the system. To fill the gap, the Authority has issued a solicitation for a clean-energy public-private partnership, under which a selected partner would spend six months identifying potential clean energy and commercialization concepts at no cost to the Authority before any further development. Critics remain skeptical: a long-serving member of the state's peer review group has called the draft plan a project at a dead end, and the Authority is expected to continue its search for private capital, though skepticism from both the public and private sectors remains high.

Other national programmes and projects

Outside East Asia and Western Europe, several countries are building or planning high-speed lines. Most follow a similar sequence: a first corridor is delivered with foreign or joint-venture technology, and the wider network follows in later phases. The schedules and cost figures below come from recent reporting and official statements, and they change often.

Morocco: extending an existing line

Morocco's network starts with a newly built line between Tangier and Kenitra. The 186 km section, which has a top speed of 320 km/h, is a newly built rail line between Tangier and Kenitra. The Kenitra–Marrakesh project would extend this corridor by about 430 km. In May 2026 the transport minister told parliament that the project had passed 30% completion and remained on track for delivery in September 2029. About 70% of the planned works consist of bridges and tunnels, because the route crosses the High Atlas. The estimated cost is MAD 55 billion ($5.9 billion). Financing is not yet fully settled: Morocco is expected to rely on international funding partners, and the timing of the extension depends on the availability of external financial support.

Sources disagree on the operating speed. Some describe the new line at 320 km/h, while others state 350 km/h, so the design and operating figures should be checked against the operator's documentation. The 2029 date is also tied to the 2030 football World Cup, which Morocco will co-host with Spain and Portugal.

Egypt: a multi-line plan built in segments

Egypt has no operating high-speed line, but it has a plan for about 2,000 km of lines. The first corridor, from Ain Sokhna on the Red Sea to Marsa Matrouh on the Mediterranean, is about 660 km long. In late September 2026 the transport minister said passenger services were targeted for July 2027, with high-speed trains travelling at up to 250 kilometres per hour, while regional trains stopping at intermediate stations reach up to 160 kilometres per hour. The government's stated reason for the new lines is capacity: the main line linking the Delta, Cairo, and Upper Egypt is already operating at full capacity, which leaves no room for extra tracks or services.

The schedule has already moved. An early projection for the first segment gave a completion date of 2023, and the Siemens-led consortium that won the main contract in 2021 was scheduled to finish in 2027. The later lines serve Abu Simbel and Hurghada, and the broader plan includes freight alongside passenger services.

India: the Mumbai–Ahmedabad corridor

India's Mumbai–Ahmedabad corridor is a 508-km dedicated high-speed rail line connecting Maharashtra's Mumbai with Ahmedabad in Gujarat. It is designed for 350 km/h, with a planned maximum commercial speed of 320 km/h. Its technology mix is distinctive. Japanese E10 trains are expected to run on the corridor in the early 2030s, while India is developing its own B28 train. Officials have said the Japanese trains are compatible with the ETCS signalling system chosen for the corridor, a point that links directly to the interoperability issues discussed elsewhere in this article.

Construction has advanced unevenly. By mid-2026 land acquisition was reported as complete, and the project has completed civil works on the 97 km Surat–Vapi section. Official statements still separate an expected first service in 2027 from a final completion date and final cost for the full corridor. The cost picture has also shifted: the original estimate was ₹1.08 lakh crore excluding taxes, and in August 2026 reports said a revised estimate of about ₹2.0–2.1 lakh crore had been proposed for approval.

Other programmes

The Haramain line in Saudi Arabia links Makkah and Madinah, a route of roughly 450 km that opened in 2018. It uses Spanish-built Talgo trainsets and carries heavy pilgrim traffic. Spain's Murcia–Almería project is a planned link in the south of the country; I could not confirm its current status, length, or design speed from recent sources, so those details should be checked against the infrastructure manager's current plans.

Common patterns

These programmes share three features. First, they depend heavily on foreign technology: Alstom and Siemens supply trains and systems in Morocco and Egypt, Talgo built the Saudi trains, and India relies on Japanese equipment for its first trains. Second, delivery is organised around operator contracts, such as the ONCF's engineering and project-management appointments in Morocco and the national high-speed rail corporation in India. Third, most projects are phased. Morocco opened its first corridor before extending it, Egypt is building in segments, and India plans to open a 97 km section before completing the full 508 km corridor.

Phasing has clear benefits. It brings early revenue, lets operators learn from the first line, and spreads financing risk. Its costs are that corridors may be fragmented for years, that integration with the rest of the network is delayed, and that later segments can be more expensive once the early sections have set design and procurement choices. Full-network delivery avoids these problems but concentrates financing and delivery risk in a single, larger programme.

Economic costs and benefits

High-speed rail is capital-intensive, and its economic case rests on how its costs are counted and which benefits are included. Per-kilometre construction costs vary so widely that headline figures are hard to compare. A UIC study on financing models reports a global average of about EUR 45.5 million per kilometre, while noting that costs vary significantly depending on the country, the type of project, terrain conditions, the urban environment, or the maturity of public procurement systems. In difficult terrain or urban areas, costs can be considerably higher. A separate benchmark gives a much lower European range of EUR 12 and 45 million per kilometre. Chinese projects supported by the World Bank cost roughly $17–21 million per kilometre, even though it has a high ratio of big-ticket viaducts and tunnels. Cost structure also matters: one breakdown attributes about 5–10% of investment to planning and land, 10–25% to civil infrastructure, and a further 5–10% to superstructure, with hilly areas these may double the costs. Comparisons therefore need to state whether land, rolling stock and financing costs are included. In Chinese projects, civil works contribute about 50% of the cost while signalling, telecommunications and electrification each account for around 5% of the total. The Jakarta–Bandung line, at $52 million per kilometre, shows how high unit costs can become when terrain and population density are unfavourable.

Benefits and how they are measured

Cost-benefit analysis (see cost-benefit analysis) is the standard tool for both ex-ante appraisal and ex-post evaluation. In a UK appraisal of proposed services based on market prices, revenue may form 30% of gross benefits and rail user time savings (including from improved reliability) may be around 50% of gross benefits. Reduced rail overcrowding and wider economic benefits each represent roughly 10% of the total in that case. The wider economic benefits are the least certain component, particularly as a result of improved business-to-business connectivity, and it is unclear whether they can be added to the direct benefits without double counting. Safety is another benefit: the main social gain is the likely reduction in accidents as a result of the transfer of traffic to HSR.

Ridership forecasts are the most sensitive input. Analysts often rely on gravity model formulations to estimate demand. Forecasting errors matter because ex-post reviews of Spanish lines found that the Madrid–Seville and Madrid–Barcelona routes, which each started with initial usage levels of 5 million passengers per annum or less, showed negative social returns which are unlikely to be offset by wider economic benefits. The discount rate also shapes the result. One engineering study of a California line assumed a opportunity cost of capital of 7.5%, which produces a large annual capital charge, and shifting the rate by a few points can change the verdict on a marginal project.

Operating finances and subsidies

Few lines recover their full costs from fares. A US Congressional Research Service review found that few if any HSR lines anywhere in the world have earned enough revenue to cover both their construction and operating costs, and governments have typically funded construction and often operations as well. Outside Japan, critics note that few HSR systems consistently generate operating profits. Farebox recovery, the share of operating cost covered by fares, is a common metric, but it can mislead. Because the denominator excludes capital costs, this ratio exaggerates the degree to which a service pays for itself. An operating-positive line can still be a net drain on public finances once capital is included.

Italy is a frequently cited counter-example. There, Italy represents the first case of competitive HSR market without public subsidies. Competition between HSR operators brought a significant decrease in ticket prices (-40%), and passenger numbers rose sharply. Even so, the claimed economic effects are modest and contested. The same study estimates an extra growth of per capita GDP of + 2.6–5.6% in 10 years, which is a regional estimate rather than a national fiscal result.

Ex-post evidence

Ex-post studies in Europe and East Asia give mixed results. A dynamic model of China's network for 2002–2013 found that HSR infrastructure development in China has generated a positive regional economic impact. Growth effects were particularly substantial in the southwest, but relatively small in the developed eastern regions. A cross-national review of 491 studies from 2008 to 2023 found that effects differ across governance models. By contrast, a short-term analysis of a Swedish line found no significant effect on labour earnings. Spillovers to tax revenue have also been estimated: an ADBI study of France's LGV Est found that approximately €300 million in spillover tax revenues could be directed to the infrastructure operator. Such estimates depend on attribution methods and are not a substitute for a full appraisal.

Environmental and energy effects

Energy use per passenger-kilometre is the most common basis for comparing high-speed rail with other modes. A 2011 review by the UIC, the international railway union, found that under comparable travel times and if typical occupancy levels are assumed, HSR is on average 2 - 5 times more energy efficient than airlines or cars. Other estimates are lower still. IEEE Spectrum reports that older energy-intensity figures for Japan's Shinkansen were around 0.35 MJ/pkm, while newer TGV and ICE designs typically need just 0.2 MJ/pkm, which the same article describes as roughly an order of magnitude below airplanes. Results vary with occupancy: aircraft and cars become much less efficient per passenger when seats or seats-in-car are empty, so the same train can look very different at high and low load factors.

Carbon comparisons follow a similar pattern. One figure cited in a 2023 study puts aviation at about 122 g of CO2 per passenger-kilometre, against 23 g for HSR. Yet the carbon result depends heavily on the electricity supply. An ICCT analysis of US flights under 2,000 km estimated that electric trains would have cut emissions on those routes by 23%, but that smaller emission reductions—of 8% and 13%, respectively—would have resulted if the trains had operated with today's diesel and electricity mix. A Swedish life-cycle study reached the same conclusion from the other direction: key variables are modal shifts and carbon intensity of electricity. Operational advantages therefore rise with low-carbon generation and fall on a fossil-heavy electrical-grid.

Construction is the largest and most contested part of the carbon account. A life-cycle-assessment (LCA) typically covers infrastructure, rolling stock, operation and end-of-life. For California's San Francisco–Anaheim line, Chang and Kendall estimated 2.4 million metric tons of CO2 with material production comprising 80% of emissions, and found that while tunneling and aerial structures account for only 15% of the route's length, they are responsible for 60% of emissions. The payback period is equally disputed. The same study suggested that construction emissions would be recuperated in about two years and their global warming effect in about six after opening, but an earlier Berkeley study by Chester and Horvath, summarised in a Reason Foundation commentary, estimated that it would take high-speed rail 71 years of operation at medium occupancy to offset its own construction-related greenhouse-gas emissions. A Madrid–Toledo study found that infrastructure construction accounts for an average 75% of emissions, with payback between 19 and 23 years after operation begins. Tunnels, bridges and the carbon-intensive materials behind them thus dominate early-stage emissions. A tunnel study found that concrete, diesel and steel are responsible for between 89.3% and 99.9% in each environmental impact category.

Whether HSR reduces total emissions depends on modal shift. The research indicates that it works best if it operates on medium-distance, high-demand corridors, with high load factors and a decarbonising electricity mix. Traffic volume is decisive: one study found that traffic volumes of more than 10 million annual one-way trips are usually required to balance annualised construction emissions, and that most diverted passengers must come from aviation. Where HSR mainly draws riders from buses or rail, the gain is smaller. Research on Chinese lines found that highly demanded HSR lines currently served short routes and mostly caused modal shift from bus services, which were also efficient compared to car. Freight can matter too: in the Swedish Europabanan study, almost 60% of this coming from a shift from truck to rail freight and 40% from a shift from air and road travel to high-speed rail travel. Earlier European analyses were more pessimistic, with one concluding that the emissions reduction from new high-speed rail would be small and that it could take decades to compensate for the emissions caused by construction.

Land take and local ecological effects are less quantified in the sources available for this section, but several issues are widely recognised. A new line generally needs a dedicated, largely continuous corridor, and fences, embankments and noise barriers can divide habitats and disrupt animal movement, which is the core of the habitat-fragmentation concern. Viaducts and tunnels reduce surface land take and avoid some severance, but they concentrate construction materials and energy use. Noise and vibration near homes are managed through route alignment, barriers, track maintenance, and vehicle design, while crossings and underpasses are used to limit ecological damage. The net effect for any given project depends on the corridor, the traffic it carries and the electricity it draws.

Social and urban impacts

The main channel through which high-speed rail (HSR) reshapes cities is travel-time saving. A shorter trip makes a longer daily commute feasible, so workers can live in cheaper or more attractive towns while working in a large city. A Chinese panel study of 285 cities from 2008 to 2016 linked the network to a wider working catchment for labour, which in turn pushed up housing prices along the lines. A one-standard-deviation rise in network accessibility was associated with about a 10.3% increase in average housing prices, although the effect differed by region and housing type. The results are not uniform. A Jiangsu study found that out-of-city stations were followed by a significant relative decline in housing prices in affected areas, despite better accessibility. An Italian quasi-experiment around the Reggio Emilia station found price increases ranging between 2 and 6% for residential properties in areas 20–30 km or 15–30 minutes away, while commercial property showed little change.

Station-area development is the most visible urban effect, and it is also where planned outcomes most often diverge from realised ones. Chinese local governments have widely planned "HSR new towns" around stations. A planning document for Nanjing South, for instance, projected a 184 km² catchment with 1.6 million residents, a target rather than an observed result. Empirical work has been more cautious. One study of prefecture-level cities found that the station area has not formed into a new subcenter; instead, it might have contributed to urban sprawl, and found no significant land price appreciation within the station area. Research on remote stations warns that they can raise resource misallocation and ghost town risk in population-shrinking cities. Land-market data point the same way in mixed directions: average land prices rose in urban districts with HSR access, while residential land supply fell in urban districts and rose in suburban counties of medium and small cities.

Whether HSR concentrates growth in large metropolitan centres or also helps smaller places is contested, and the answer varies by country and by spatial scale. In Japan, a study of Shinkansen stations found that most stations had a modest agglomeration effect on their immediate surroundings, but the Shinkansen stations have a siphoning effect on population changes, as residents migrate from small, low-level local stations to major cities. A Chinese study across cities reported that HSR is conducive to the further rise of big cities, but not to the development of small and mediumsized cities. Other work points the opposite way. Research on the Shanghai–Kunming line found that the railway can significantly increase the economic aggregate and urbanization rates along the route, with the greatest impact on regional economic development and spatial patterns in economically backward areas, although its effect on growth rates was limited. A Taiwan study found that HSR-based towns on average gained population and employment, while HSR-based counties experienced relatively less growth of employment in the tertiary sector. European evidence suggests that HSR mainly strengthens integration within regions, over distances of up to about 200 km, rather than linking distant metropolises.

Tourism is one of the channels through which regional effects operate. One Chinese study found that tourism development mediates the relationship between HSR and regional growth, which makes destinations along a line, rather than only commuter towns, important beneficiaries. For inter-regional travel more broadly, the rail link mainly changes which trips are practical as same-day or weekend journeys, rather than changing where people live.

Accessibility for older and disabled travellers depends mainly on operational and design rules rather than on urban outcomes. Step-free routes from street to platform, level or near-level boarding, and advance assistance services determine whether the network is usable for these passengers, and these standards are set by operators and national or EU regulation.

Evidence on all of these effects faces a common methodological problem: infrastructure tends to be built where growth is already expected, which makes causal attribution difficult. Studies that use quasi-experimental designs, such as station openings, are therefore more persuasive than simple before-and-after comparisons.

Competition and integration with other transport modes

Competition between high-speed rail and air travel is sharpest on short routes, and France has used regulation to shift that contest. The French ban works on a simple test: if a high-speed rail alternative of 2.5 hours or less is available and sufficiently frequent, flights on that route are to be suspended. The rule was narrower than first proposed. The Commission scaled back the initial plan to ban eight short-haul flights, and it kept only three Paris-Orly routes, to Bordeaux, Nantes and Lyon, because the other candidates either lacked a proper 2.5 hour rail alternative or the rail options were unavailable early in the morning or late into the evening. Connecting flights are not affected, and Paris Roissy-Charles de Gaulle has been excluded. The policy's expected effect is small. The DGAC estimated that the measure could cut up to 55,000 tonnes of emissions - just 2.6 per cent of total emissions from domestic flights in France per year. Critics have called it largely symbolic, and the ban was part of a 2020 agreement, mid-pandemic, to bail out the national carrier Air France. Spain has planned to introduce a similar ban on short-haul flights where there are rail alternatives under 2.5 hours.

Price and time trade-offs with cars and buses

Rail's advantage over flying depends on the journey. City-centre to city-centre travel can favour trains once airport access, check-in and security are counted. A 2025 analysis found that taking the train is faster than flying for some short journeys within France. Advocacy groups make stronger claims: the consumer group UFC-Que Choisir argued that plane journeys emit 77 times more carbon dioxide per passenger than train journeys on routes less than 2 1/2 hours and are, on average, only 40 minutes faster. Such figures depend on the assumptions used, so they should be read as one side of the argument. Against private cars, the comparison turns on door-to-door time, parking, tolls and whether the traveller already owns a car, and the results vary by corridor. Against intercity buses, which are usually slower but cheaper, high-speed rail competes mainly on journey time and comfort, while buses compete on fare. Price competition therefore tends to be strongest where the rail operator faces rivals on the same corridor.

Integration with airports and urban transit

Many networks attempt to reduce the penalty of transfers by locating stations at airports or on urban transit hubs. Examples include the TGV station at Paris Charles de Gaulle, the long-distance station at Frankfurt Airport and the multimodal Hongqiao hub in Shanghai, where rail, metro and airport are connected. These arrangements matter because the total journey is the competitive measure: a train that is fast between city centres can lose passengers if the last-mile connection is poor. Integration also raises design questions, such as the walking distance between platforms, shared check-in or baggage arrangements, and whether an airport station is well served by a city-centre station. Stations placed far from the centre can reduce the time advantage that trains otherwise hold over flights. In France, the exclusion of Paris-CDG from the flight ban illustrates how policy and hub geography interact.

Market structure: state operators and open access

Most high-speed networks were built and first operated by state-owned incumbents. The EU approach has moved toward competition on long-distance services: for freight and long-distance trains, competition among the railway undertakings is the rule and public subsidies are admitted only in specific cases. Italy is the leading example. Italo, since April 2012, was the world's first private open-access operator of high-speed rail to compete with a state-owned monopoly. Evidence on its effects is largely positive. Researchers describe a triple competitive effect: more capacity, frequency and connections, lower prices and better services. A 2025 presentation from the FS group's research centre reports that Trenitalia kept market leadership, outperforming Italo in all metrics, while keeping a price premium of roughly 10–20%. Much of the cost-saving data comes from Italo itself, which estimated savings of around 800 million of euro between 2012 and 2016; such figures come from an interested party and should be treated cautiously.

Spain has added open-access competition since 2021. Ouigo launched on Madrid–Barcelona in May 2021, and Iryo started service in late 2022. A 2026 study finds that entry produced industry price reductions by 18.6% in Madrid–Barcelona and by 21.5% in Madrid–Valencia. The incumbent's response was uneven: the study notes that Renfe's low-cost brand Avlo increases Ave prices, whereas Ouigo and Iryo's entry decrease them. New entrants also face access problems. Iryo and Ouigo have complained to the national competition authority that Spain's dual-gauge network, which combines standard and Iberian gauge, adds significant technical complexity to operations and favours Renfe, whose trains can run on both. A separate dispute concerns maintenance: Renfe is fighting a Spanish order to open up its La Sagra depot to high speed rival Iryo despite an order from Spain's competition regulator to do so. These cases show that fare outcomes depend not only on the number of operators but also on access to tracks, rolling stock and maintenance capacity.

Controversies and open questions

Cost growth is the most persistent criticism of high-speed rail. Aggregate figures from megaproject research are frequently cited, but they cover rail as a whole rather than HSR alone. One widely quoted database reports that rail projects are among the worst offenders, with an average cost overrun of 44.7% and ridership shortfalls averaging 51.4%. A 2018 audit by the European Court of Auditors examined EU-supported lines and found aggregate cost overruns of €25.1 billion, a 78% overrun at the line level, and construction delays of more than a decade affecting half the lines studied. Whether these patterns are systematic is disputed. Flyvbjerg attributes them to "optimism bias," the tendency for project forecasts to underestimate costs and overestimate benefits. Other analysts argue that large projects differ in risk, geology and governance, and that averages hide lines that came in close to budget. The evidence supports a warning about forecasting, but it does not by itself settle how far any single project will overrun.

California's project shows how sharply estimates can diverge. Voters in 2008 were presented with a total cost of $33 billion and a completion date of 2028. The cost estimate has grown to $89–$128 billion in various 2026 reports. The state rail authority's 2026 draft business plan puts Phase 1 at about $126.2 billion, and the same document says the Merced–Bakersfield segment would recover only 35 to 49 percent of its operating and maintenance costs from fares and ancillary revenue. Press reports have circulated a figure of $231 billion, but the authority has said that figure represents a high-end, un-optimized scenario based on legacy design and delivery assumptions that have since been reevaluated. Readers should check which scope, price base and phasing each document uses before comparing numbers.

Low-ridership lines and the "white elephant" critique follow from the same arithmetic. Critics argue that proponents routinely understate costs and overstate demand, so that lines built for intercity travel carry far fewer passengers than projected and rely on public money to operate. Defenders reply that many lines were designed to relieve congested corridors, that ridership grows once a network is connected, and that cost-benefit studies should count time savings and wider effects. The appropriate network density is contested for the same reason: a dense network may serve many trips but cost more per passenger, while a sparse network concentrates demand on a few corridors and leaves other regions with weaker service.

Capacity conflicts arise where passenger and freight trains share track. Fast passenger services need long, uninterrupted paths, while freight trains are slower, heavier and often run at night or on the same corridors. Mixed use can force operators to limit either passenger frequency or freight volume, and upgrading shared lines can be expensive. Some networks have responded by building dedicated passenger lines or separate freight routes, but these solutions increase capital costs and shift the debate back to the economics of network design.

Another open question is whether HSR mainly strengthens large metropolitan centres or also benefits smaller cities and regions. Supporters point to new station-area development and improved access to labour markets outside the largest cities. Critics contend that faster links tend to draw jobs, offices and population toward the biggest hubs, leaving peripheral towns with fewer benefits. The empirical evidence varies by country and by corridor, and the answer may depend on local planning, station location and the availability of complementary transport.

The debate between wheel-on-rail systems and maglev is also unresolved. Japan's Chūō Shinkansen is the main test case. Its initial estimate was ¥5.52 trillion, the figure was later raised to ¥11 trillion, and the opening date has moved to the mid-2030s, nearly a decade later than the previous target of 2027. Much of the cost reflects tunnelling, since the line is designed with about 85% tunnels, and the route requires a dedicated guideway. Maglev proponents emphasise higher design speeds and the potential to relieve congested corridors, while wheel-on-rail advocates stress that conventional HSR is cheaper to build, can run on existing interoperable networks, and has a long operating record. The choice therefore depends on cost, geology, route density and the value placed on top speed.

Future technical directions

Faster wheel-on-rail trains

The nearest technical step is a move from the current generation of roughly 350 km/h trains to vehicles designed for 400 km/h on steel wheels. China's CR450 programme is the most advanced example. Two prototypes were completed at the end of 2024 and then went through type tests on the Beijing ring railway and on operating lines in the Chongqing–Xiamen and Shanghai–Chongqing–Chengdu corridors. A single trainset reached 453 km/h in testing, and two trains passing each other produced a combined speed reported as 891 or 896 km/h, depending on the source. The operator expects to complete operational assessment and finalise the design in 2026. At 400 km/h the reported figures include emergency braking distances below 6,500 metres and energy use below 22 kWh per km, though these are manufacturer-side test claims rather than independently verified service data. Higher speeds also need more than new trains: the existing track geometry, bridges, overhead wires and signalling must tolerate the extra aerodynamic loads, lateral forces and braking energy, and many lines are currently limited to 350 km/h for infrastructure reasons, so a 400 km/h service depends on upgrades to the corridor as well as on the rolling stock.

Maglev and the cost of new alignments

The main alternative to wheel-on-rail is the maglev. Japan's Chūō Shinkansen is a superconducting maglev line designed for commercial speeds around 500 km/h, and the record of 603 km/h set by the same technology in 2015 is a test figure rather than a service speed. Its schedule has moved repeatedly. The Tokyo–Nagoya section was originally due in 2027, but construction stalled in Shizuoka Prefecture over concerns about water resources. After the prefecture approved the works in 2026 and boring agreements were reached in July 2026, the central tunnel section is now expected to take around a decade to complete. Estimates for the first section vary between 2034 and 2036, and reports in 2026 still describe the mid-2030s as the earliest realistic date. The cost structure is the central trade-off. The Tokyo–Nagoya section has been estimated at about ¥5.5 trillion, with roughly 85% of its length in tunnels, because a maglev needs its own purpose-built guideway and cannot share the existing network. Wheel-on-rail trains, by contrast, can use upgraded conventional lines, as the CR450 plans illustrate. Whether maglev's speed advantage justifies the much larger civil-works bill remains an open question, and this is why many planners see wheel-on-rail as the more practical option for most corridors.

Digitalisation, ETCS and automatic train operation

Future high-speed networks are expected to rely more on automation and digital signalling. The European Train Control System (ETCS) Level 2 provides continuous speed supervision and movement authority by radio, and automatic train operation (ATO) is designed to run on top of it. The European standard for ATO over ETCS is being developed under the Control-Command and Signalling specifications, which set common interfaces so that onboard and trackside equipment from different suppliers can work together. The grades of automation range from driver-operated systems to fully unattended operation (GoA4). In 2026, Finland demonstrated ETCS-based automated driving at GoA2 over a 19 km stretch, with a driver still in the cab, and FS Group in Italy completed GoA4 validation trials on an ETCS test ring. Driverless operation is common on metros, but on heavy-rail corridors automation is still at the trial stage, and high-speed lines are regarded as the easiest case for higher grades because of their fencing and controlled access. The expected benefits are shorter headways, more consistent running times and lower energy use, which are the reasons ETCS deployment and ATO are linked to capacity planning on existing lines.

Predictive maintenance is the second digital frontier. Trains and track increasingly carry sensors that record vibration, temperature, bogie and bearing behaviour, pantograph wear and rail geometry. Machine-learning models can use these streams to detect faults before they cause failures, shifting maintenance from fixed intervals toward condition-based schedules. This matters most for high-speed lines, where small defects in rail or wheel profiles can quickly affect ride quality and safety.

Energy transition and non-electrified routes

The carbon footprint of electric high-speed rail depends on the electricity supplying the power grid. Operators are therefore pursuing renewable power supply, including long-term purchase agreements, and are using regenerative braking to return energy to the network when trains slow down. Grid storage can smooth the large simultaneous loads created by several trains accelerating at once. Where a line is not fully electrified, or where a branch cannot justify overhead wires, onboard batteries or hydrogen fuel-cell units are being tested as alternatives to diesel traction. These options are generally better suited to regional or shorter routes, since their energy density and recharging times make them less competitive for long, fast intercity journeys, which remain the domain of electrified lines.

References

1. High-speed railen.wikipedia.org
2. The definition of High Speed Railuic.org
3. Trans-European high-speed rail networken.wikipedia.org
4. High-speed rail (HSR) - by Dan Jensen - The Rail Agendatherailagenda.com
5. Q&A - High Speed Railhighspeedrail.nl
6. Europe Bullet Train: High-Speed Rail Guide for 2026everyrail.com
7. High-Speed Railway_Baiduwikibaike.baidu.com
8. List of speed records in rail transporten.wikipedia.org
9. Fastest maglev trainguinnessworldrecords.com
10. L0 Series Maglevhigh-speed-trains.fandom.com
11. Japan Builds the World’s Fastest Train as Maglev Project Targets 603 km/hmyelectricsparks.com
12. High-speed rail in Germanyen.wikipedia.org
13. Track geometryhotrails.net
14. Alignment Design Standardshsr.ca.gov
15. Chongqing%E2%80%93Kunming high-speed railwayen.wikipedia.org
16. Ballastless tracken.wikipedia.org
17. Railway slab vs ballasted track: A comparison of track geometry degradation - ScienceDirectsciencedirect.com
18. Understanding High-Speed Rail Track: From Ballast to Aerodynamic Innovationhighspeedrailcanada.com
19. Five Key Take-Aways of a Ballastless Track Systemtum-asia.edu.sg
20. High-Speed Raildocuments1.worldbank.org
21. Tianjin Grand Bridgeen.wikipedia.org
22. China completes construction of high-speed railway tunnel beneath Yangtze Rivertvbrics.com
23. Hefei%E2%80%93Fuzhou high-speed railwayen.wikipedia.org
24. List of high-speed railway linesen.wikipedia.org
25. 25 kV AC railway electrificationen.wikipedia.org
26. FS Class ETR 500en.wikipedia.org
27. Vectron (locomotive)en.wikipedia.org
28. JNR's Choice of Traction System—From Loco-hauled to Multiple-unit Trainsejrcf.or.jp
29. High-Speed Railwayencyclopedia.pub
30. Multiple uniten.wikipedia.org
31. Loco-Hauling vs Multiple Units for InterCity Operations.railforums.co.uk
32. EMUs Versus Locomotivespedestrianobservations.com
33. Technologies - Multiple units (MUs) vs. loco-hauled trainsrailway-energy.org
34. Robust energy management of electrified railway traction substation providing secondary frequency regulation - ScienceDirectsciencedirect.com
35. Recent research progress and application of energy storage system in electrified railway - ScienceDirectsciencedirect.com
36. What is Regenerative Braking? Energy Recovery in Trains - Railway News - Rolling Stock, High Speed Rail & Infrastructurerailwaynews.net
37. Control strategy and capacity optimization of energy-storage-based railway power conditionerstet-review.org
38. List of high-speed trainsen.wikipedia.org
39. Electric multiple uniten.wikipedia.org
40. Optimal cross-sectional area distribution of a high-speed train nose to minimize the tunnel micro-pressure wavelink.springer.com
41. (PDF) Optimization of Train Nose Shape for Reducing Micro-Pressure Wave Radiated from Tunnel Exitresearchgate.net
42. Track–bridge interaction effects in the acceleration and displacement response of high-speed railway bridges: Simplified vs refined modelling - ScienceDirectsciencedirect.com
43. New Pendolinoen.wikipedia.org
44. Pendolino Tilting Train, Italy - Railway Technologyrailway-technology.com
45. British Rail Class 390en.wikipedia.org
46. VR Class Sm3en.wikipedia.org
47. PENDOLINO: Tilting for speedalstom.com
48. ETCS: Strategies, Architectures and Migration Pathways for Modern Rail Networks - Siemens Mobility Globalmobility.siemens.com
49. ERTMS/ETCS: The European Train Control System - Voie Librevoie-libre.com
50. ERTMS - The railway dictionary of Mediarail.beledicoferroviaire.mediarail.be
51. China’s next-generation signalling system targets automatic operation - International Railway Journalrailjournal.com
52. China to equip high-speed trains with CTCS - International Railway Journalrailjournal.com
53. Automatic train controlen.wikipedia.org
54. THSR 700Ten.wikipedia.org
55. Digital automatic train control system for the Shinkansen ...witpress.com
56. The Shinkansen’s legendary operation and safety recordhsrail.org
57. Safety and Convenience: How the Shinkansen Changed Japannippon.com
58. What is Shinkansen System?ihra-hsr.org
59. Progress of Safety in Japanese Railwaysscispace.com
60. Systematic Cybersecurity Risk Analysis of European Rail Traffic Management Systemarxiv.org
61. History of the Shinkansen: How Japan Built the Bullet Trainjrpass.com
62. The Tokaido Shinkansen Celebrates 60 Years of Awesome - Unseen Japanunseen-japan.com
63. Tokaido Shinkansenen.wikipedia.org
64. The Shinkansen: How high-speed rail transformed a nationhsrail.org
65. 1981 : la révolution du premier train à grande vitesseina.fr
66. LGV Sud-Esten.wikipedia.org
67. TGVen.wikipedia.org
68. TGV world speed recorden.wikipedia.org
69. The Shinkansen Turns 50: The History and Future of Japan’s High-Speed Trainnippon.com
70. SNCF TGV Sud-Esten.wikipedia.org
71. China High-Speed Rail Completes 50,000 km by 2025 - Railway News - Rolling Stock, High Speed Rail & Infrastructurerailwaynews.net
72. China High-Speed Rail 2025: 50,000 kmchinainfigures.com
73. High-speed rail in Europeen.wikipedia.org
74. Rail transport in Franceen.wikipedia.org
75. High-speed trains have the wind in their sailsvoyages-d-affaires.com
76. Publication of the UIC High-Speed Rail Atlasuic.org
77. Intercity and High-Speeduic.org
78. World’s high-speed railway network is evolvingrailwaypro.com
79. Publication of the UIC High-Speed Rail Atlas 2024 Editionuic.org
80. China is expanding its rail network: four high-speed lines opened and plans for a further 180,000 kmrailwaypro.com
81. High-speed rail in Europeen.wikipedia.org
82. The Spanish High Speed Rail Network: A Success Storythalesgroup.com
83. Fifteenth five-year planen.wikipedia.org
84. China: ‘Eight by eight’ network nears completion with latest wave of openings - Railway Gazette Internationalrailwaygazette.com
85. China's railway construction gains steady momentum in Q1english.www.gov.cn
86. China’s railway construction gains groundmanilatimes.net
87. The world's fastest conventional train is nearing launch in Chinarailwaypro.com
88. China's CR450 high-speed train moves a step closer to commercial operation - Chinadaily.com.cnglobal.chinadaily.com.cn
89. China's CR450 hits 453 km/h in testing — the world's fastest train is almost readyabit.ee
90. China CR450 Completes 453 km/h Tests for 400 km/h 2026 Design - Railway News - Rolling Stock, High Speed Rail & Infrastructurerailwaynews.net
91. China's railway investment up 3.2% in first four months - Chinadaily.com.cnchinadaily.com.cn
92. 15th Five-Year Planen.wikipedia.org
93. China targets 400km/h high-speed network - International Railway Journalrailjournal.com
94. China's Railway Spending Is on Track to Hit New High of USD126 Billion This Yearyicaiglobal.com
95. China High-Speed Rail Map: All Routes Visualized (2026)mychina.guide
96. Interactive factsheet - UNIFEunife.org
97. Connecting European railwaystransportenvironment.org
98. eu funding ertms entransport.ec.europa.eu
99. Brenner Base Tunnelen.wikipedia.org
100. Brenner Base Tunnel project achieves cross-border connection - International Railway Journalrailjournal.com
101. Excavation Begins on French End of 35-Mile Lyons-Turin Rail Tunnelenr.com
102. A new Alpine rail connection between Lyon and Turin - European Climate, Infrastructure and Environment Executive Agencycinea.ec.europa.eu
103. EU to provide €5.4bn for TEN-T projects - International Railway Journalrailjournal.com
104. Major Breakthrough for Brenner Base Tunnel Through the Alpsenr.com
105. Brenner Base Tunnel (BBT)international.strabag.com
106. Brenner Base Tunnel (Innsbruck/Fortezza, 2028)structurae.net
107. The Lyon-Turin tunnel, key element in the trans-European rail networkegis-group.com
108. For a complete TEN-T, an increased EU budget is neededunife.org
109. California High-Speed Rail 2026: SF-LA Timeline, Cost & Route Map - Railway Newsrailwaynews.net
110. California high-speed rail fails to accomplish most of its legislative goals for 2026kpbs.org
111. California High-Speed Rail board approves business plan that suggests moving downtown Merced stationkvpr.org
112. California High-Speed Rail Advances Merced–Madera Construction Procurementmetro-magazine.com
113. NEWS RELEASE: California High-Speed Rail Advances Major Construction Procurement for Merced to Madera Extension - California High Speed Railhsr.ca.gov
114. California’s High-Speed Rail: The 2026 Draft Business Plan Is Ambitious, Expensive, and Still Legally Problematiccaliforniapolicycenter.org
115. CHSRA Issues Draft 2026 Business Plan - Railway Agerailwayage.com
116. California High-Speed Rail Price Tag Explodes - Newsweeknewsweek.com
117. California sues to stop Trump administration from pulling $4 billion from high-speed rail projectcbsnews.com
118. California withdraws lawsuit challenging Trump’s $4 billion high-speed rail funding cutspokesman.com
119. California drops lawsuit over Trump decision to pull $4 billion in high-speed rail funds By Reutersinvesting.com
120. What comes next for California high-speed rail after federal lawsuit droppedabc10.com
121. California high-speed rail costs soar to $231 billionfoxla.com
122. news release california high speed rail board approves design contracts to prepare for construction into merced and bakersfieldhsr.ca.gov
123. New high-speed rail plan keeps pushing toward Merced, Bakersfield for interim operationsnews.yahoo.com
124. Cost of California’s High-Speed Rail Goes Up Again - Pacific Research Institutepacificresearch.org
125. California abandons lawsuit over $4 billion in federal high-speed rail fundssfchronicle.com
126. High-speed rail in Moroccoen.wikipedia.org
127. Kenitra-Marrakech LGV Project Hits 30% Progress, Set for 2029 Deliverymoroccoworldnews.com
128. Egypt Targets July 2027 for Ain Sokhna-Matrouh High-Speed Rail Serviceegyptianstreets.com
129. Egypt to launch Ain Sokhna-Marsa Matrouh high-speed rail in July 2027 - Urban & Transport - Egypt - Ahram Onlineenglish.ahram.org.eg
130. Mumbai Ahmedabad bullet train project hits milestone as first steel bridge launched in Maharashtra - BusinessTodaybusinesstoday.in
131. Mumbai–Ahmedabad Bullet Train: 2027 First Service Expected, but Final Completion Date and Cost Remain Openrailway.supply
132. Mumbai Ahmedabad Bullet Train (MAHSR) Route, Map, Stations, Cost, Progress & Latest Updates (2026)infracapitalist.com
133. Rail transport in Moroccoen.wikipedia.org
134. Egypt’s High-Speed Rail Network Targets July 2027 Passenger Launch in Broader Rail Overhaul - Ecofin Agencyecofinagency.com
135. High-speed rail in Egypten.wikipedia.org
136. High-speed rail in Egyptnorth-africa.com
137. Mumbai-Ahmedabad Bullet Train Project 2026, High-Speed Railvajiramandravi.com
138. Mumbai–Ahmedabad Bullet Train Progress 2026: Construction Update, Stations & What to Expectroadrailinfra.com
139. Study: High-speed rail lines cost an average of EUR 45.5 million per kilometerrailwaypro.com
140. Cost benchmarking of railway projects in Europeorbit.dtu.dk
141. Why China can build high-speed rail so cheaply - Global Construction Reviewglobalconstructionreview.com
142. How China builds high-speed rail for less - International Railway Journalrailjournal.com
143. The Economics of Investment in High Speed Rail Summary and Conclusions 30itf-oecd.org
144. 155 The Economics of Investment in High-Speed Rail Roundtable Reportoecd.org
145. (PDF) The full cost of high-speed rail: An engineering approachresearchgate.net
146. The Development of High Speed Rail in the United States: Issues and Recent Eventscongress.gov
147. Re-evaluating Economic Approaches to High-Speed Rail Investment in Asia - Asia Pathwaysasiapathways-adbi.org
148. Subsidizing Transportcato.org
149. Economic growth, transport accessibility and regional equity impacts of high-speed railways in Italy: ten years ex post evaluation and future perspectives - PMCpmc.ncbi.nlm.nih.gov
150. Measuring the regional economic impacts of high-speed rail using a dynamic SCGE model: the case of China: European Planning Studies: Vol 27, No 3tandfonline.com
151. Full article: The socio-economic impacts of high-speed rail: a cross-national comparative reviewtandfonline.com
152. Indonesia: The high cost of high-speed raillowyinstitute.org
153. A profile of high-speed railways Executive Summary bitre.gov.au
154. The Economics of Investment in High Speed Rail Summary and Conclusions 30oecd.org
155. High Speed Rail and Sustainabilityapta.com
156. Fast Trains Are Energy Efficient (And Fast) - IEEE Spectrumspectrum.ieee.org
157. The bullet train to lower-carbon travel - International Council on Clean Transportationtheicct.org
158. The role of high-speed rail in mitigating climate changesciencedirect.com
159. Life cycle greenhouse gas assessment of infrastructure construction for California’s high-speed rail system - ScienceDirectsciencedirect.com
160. High-Speed Rail Is Unlikely to Play a Major Role In Achieving Climate Goals - Reason Foundationreason.org
161. Life cycle greenhouse gases emissions from high-speed rail in Spain: The case of the Madridsciencedirect.com
162. Environmental impact assessment of high-speed railway tunnel construction: A case study for five different rock mass rating classes - ScienceDirectsciencedirect.com
163. Fast Rail in the Era of Modal Shift: Global High-Speed Networks and Their Environmental and Socio-Economic Impactsdoi.org
164. Can high speed rail offset its embedded emissions? - ScienceDirectsciencedirect.com
165. Evaluating the substitutability of short-haul air transport by high-speed rail using a simulation-based approach - ScienceDirectsciencedirect.com
166. The impact of high-speed railway on urban housing prices in China: A network accessibility perspective - ScienceDirectsciencedirect.com
167. High-speed rail expansion and housing prices in the "forgotten corners": Sub-City Level Evidence from Jiangsu Province in China - ScienceDirectsciencedirect.com
168. Does the Opening of a High-Speed-Rail Station Impact House Prices? Evidence from a Quasi-Natural Experiment: Reggio Emilia Mediopadanalink.springer.com
169. High-speed rail and distribution of economic activities: Evidence from prefecture-level cities in China - ScienceDirectsciencedirect.com
170. Remote high-speed rail stations, urban land supply, and the emergence of new economic activities - ScienceDirectsciencedirect.com
171. Do high-speed railways have an impact on population change? evidence from Japan - ScienceDirectsciencedirect.com
172. High-Speed Railway and Regional Economic Growth: An Empirical Study Based on Market Potentialscirp.org
173. Impact of high-speed railway construction on spatial patterns of regional economic development along the route: A case study of the Shanghai–Kunming high-speed railway - ScienceDirectsciencedirect.com
174. Abstract: Understanding the impact of high-speed rail (HSR) servicesjtlu.org
175. Impact of transportation infrastructure upgrading on the rapid development of regional economy: A quasi-natural experiment based on the opening of high-speed railway - ScienceDirectsciencedirect.com
176. Development of High-Speed Rail in the People's Republic ...adb.org
177. High-speed rail, new town development, and the spatial mismatch of land leases in Chinaideas.repec.org
178. (PDF) The Impact of High-Speed Rail on Housing Prices: Evidence from China’s Prefecture-Level Citiesresearchgate.net
179. The impact of high-speed railway construction on regional economic development in Hubei Provincelink.springer.com
180. France’s Short-Haul Flight Ban: A Bold Step Toward Climate-Resilient Mobilityasuene.com
181. TriplePundit • France Starts Banning Short-Haul Flights and Urges a Return to Travel by Railtriplepundit.com
182. France Bans Domestic Short Haul Flightsblog.tugo.com
183. France’s short-haul flight ban one year on: Has it encouraged more people to take the train?euronews.com
184. France Legally Bans Short-Haul Flights—Environmentalists Want Moreforbes.com
185. Newsletter Newsletters Events Events Podcasts Videos Africanewseuronews.com
186. france short haul flight banthepointsguy.com
187. 11 Discussion Paper 2016 • 11 Christian Desmaris Laboratoire Aménagementoecd.org
188. High-speed rail in Italyen.wikipedia.org
189. 5th INTERNATIONAL SYMPOSIUM ON HIGH-SPEED RAIL SOCIOECONOMIC IMPACTSfsitaliane.it
190. Benefits of the competition in the High speed rail and the limits ...italospa.italotreno.it
191. Sequential entry in Spanish HSR markets and its effect on prices - ScienceDirectsciencedirect.com
192. Ouigo, Iryo say Spain’s dual-gauge network gives Renfe unfair edge on liberalised routes - Railway Gazette Internationalrailwaygazette.com
193. ‘An outrage’: Renfe’s fight to keep Iryo out of its workshops tests Spain’s rail liberalisation - Railway Gazette Internationalrailwaygazette.com
194. france is set to ban short haul domestic flights 120522timeout.com
195. France Is Banning Short-Haul Domestic Flights in a Bid to Reduce Carbon Emissionsnews.yahoo.com
196. Op-ed: Italy's liberalization of high-speed rail offers a benchmark for Europetheparliamentmagazine.eu
197. Ouigo Espa%C3%B1aen.wikipedia.org
198. Costs of High-Speed Railcitizenresearch.ca
199. What High-Speed Rail Really Costscitizenresearch.ca
200. Ottawa's high-speed rail dream risks becoming a fiscal nightmaretroymedia.com
201. Japan’s maglev bullet train delayed to 2035 or later as costs rise to ¥11 trilliontripbytrip.org
202. Tunnelling agreement revives stalled Chuo maglev - Railway Gazette Internationalrailwaygazette.com
203. 1/37 Cost Overruns and Demand Shortfalls in Urban Rail andarxiv.org
204. Japan Maglev Train: Check Speed, Route, and Updatesjrailpass.com
205. China's CR450 high-speed train moves a step closer to commercial operationchinadailyhk.com
206. China Railway CR450AFen.wikipedia.org
207. L0 Seriesen.wikipedia.org
208. Shizuoka set to approve maglev construction - The Japan Timesjapantimes.co.jp
209. List of maglev train proposalsen.wikipedia.org
210. The Chūō Shinkansen Maglev: Japan's 505 km/h Future Trainjrpass.com
211. Fintraffic and Siemens Mobility ran Finland’s first automated train drive with ETCS and ATO - Railway Gazette Internationalrailwaygazette.com
212. FS Group and Europe’s Rail showcase GoA4 automation progress in Bologna - Railway Gazetterailwaygazette.com
213. Automatic train operation (ATO) - by Dan Jensentherailagenda.com
214. Trends and developments in the automation of heavy rail operationsglobalrailwayreview.com