9.1 Fencing at hotspots


What does this measure refer to and what is its objective?

Fences and other barriers at known high risk locations outside stations identified as hotspots in order to prevent access to the railway in open line. Access is usually restricted by fencing along the tracks.


Best practice and lessons learned

  • Determine where to install high security fencing, considering past incidents for fatalities, equipment debris strikes and trespass reports. You can take into account the presence of schools, parks, frequent standard fence breaches, presence of homeless, presence of graffiti, gathering points, trails, etc. Fencing is more frequently necessary when the rail line is located close to the city centre.
  • Fence at hotspots should be at least 500 meters both sides of the hotspot otherwise there could be a waterbed effect, i.e. displacement of the incidents in the neighboring area (see study results).
  • For guiding positioning of fences, look at the most logical and used illegal path and try to put up the fence along this line.
  • The effect on incidents depends on the physical dimensions, construction and maintenance of fences and barriers, and how easy it is to go round, over or under them. The effect of fencing will be better when you fence a whole line and combine with anti-trespass grids.
  • In the vicinity of level crossings, other measures can be added such as special U-turn fencing cameras, special anti-trespassing grid.
  • Combine with landscaping, in order to meet regulation concerning visibility road- track and track road.

Warning points

Expected difficulties and issues you should pay attention to

  • For practical grounds there needs to be a gate in the linear fence to provide access to the tracks in case this is needed. Give special attention to escape routes for train passengers in case of emergency and entry gates for railway workers and their materials and emergency routes for emergency services. Consider the systems for securing gates and their use.
  • The effect is stable assuming that control and maintenance is done. Trespassers might change locations over time, however. A person aiming at suicide or trespass for criminal purposes could easily find other routes around short sections of fencing.
  • Make sure that fencing on electrified lines (e.g. third rail, OHL electrification) is earthed.
  • There may be some cultural issues, associated with different tolerances of risk or standards of protection (e.g. the degree of fencing that is usual in railway environments). Some countries (e.g. UK) require all rail routes to be fenced.
  • At some locations, excessive vegetation needs to be removed in order to increase the visibility of the train driver and to reduce the opportunities for concealment for a potential victim. These locations might then require fencing.
  • Consider that even more robust fences can be damaged with newer tools.


Other points that you should not forget

  • The type of fencing needs to be considered - strand, chainlink, paling, security - height, etc.? Short fences (e.g. 100 m long) do not prevent wildlife from crossing the railway.
  • The total cost for more than 13 kilometers of fences, installation of 18 gates at service entrances and removal of obstructing vegetation at 5 locations in Belgium was €700,000.
  • An example of excessive barbed fence in a hotspot near Lille, France (2013). The goal was to prevent trespassing and cable theft near a gypsy camp. The fences are 3 meters high and 1,8 km long. Fencing is reinforced by several rolls of barbed wires and by security patrols which provide surveillance.
  • However, there is pilot evidence suggesting that even cheap fences can be effective in reducing the number of deaths and seriously injured due to suicide. A Swedish study (Fredin-Knutzén et al., 2021) showed that cheap fences (≈25% the costs of a standard fence) with a height of 2100 mm and a 400 mm long 90 degrees overhang were able to prevent up to 80% of people trying to climb over it in a small-scale, experimental setting (even more difficult to climb than standard fences). The authors recommended how this fence should be used to gain effects on suicide and argued that it is important to fence over wider areas than what’s done today, which could be exceptionally profitable from a socio- economic perspective if the effects were to be reproduced in a naturalistic context. However, this needs to be further investigated and tested in a real setting before any definitive conclusions can be made.

Study results

Data or other evidence supporting the measure's effectiveness

  • Evidence regarding effect of fencing alone on trespassing has been found in one study (Silla & Luoma, 2011) where it reduced trespassing by 94.6%, but for one specific location in Finland. No evidence has been reported about the effect on suicide.
  • Supported by 44.5% of respondents (Silla & Luoma, 2012b)
  • Although some studies have demonstrated that fencing and landscaping are the most effective ways to reduce trespasses (Silla & Luoma, 2011), focus groups with British as well as Spanish experts showed that fences are not necessarily effective, such as for the determined people who will get over or under a fence if they really want to, or for the people who will break it in order to cross.
  • Urban fencing would reduce trespassing fatalities by a maximum of 100 persons a year if well maintained (Savage, 2007).
  • Barriers can be costly, are permanent, need to reach six and nine feet to be effective, and therefore need to be placed where there is a risk of suicide (NIMH, 2006).
  • The literature suggests that this is one of very few approaches for which there is strong evidence of effectiveness. The theory behind restricting access to means (at hotspots) is that it may ‘buy time’ for the individual to reconsider his or her actions, particularly in situations where these actions are associated with impulsivity or ambivalence (Cox et al., 2013).
  • INFRABEL reported that in Belgium fences in open line decreased by 75% of the number of suicides/attempted suicides in some hotspots.
  • Analysis by PRORAIL of 5,000 suicide-events in the Netherlands looked at the distances between the location where a person was hit by a train and the nearest level crossing or station. The results in the graph below show that the blue line (level crossing) lies above the read line (station) until approximately 500 meters. The data suggests that suicidal persons wander away from the level crossing for about 500 meters before they are being hit by a train. The conclusion being that they do not want to be seen by train drivers and level crossing users as these people could intervene. Having objects close to the track where a suicidal person can hide behind (railway equipment, trees, bushes, etc.) should therefore be avoided. By installing fences with a minimum length of 500 meters and removing trees and bushes on the trackside alongside the fences make it possible for train drivers and level crossing users to see a suicidal person and potentially intervene. Likewise, railway equipment and buildings where people can hide behind should ideally not be inside those fences. As this is frequently not possible or very expensive, these objects should have fences around them to prevent the option of easily running to the track when the train approaches.

last update: 2022-12-01