Asite investigation is only as good as its desk study and this in turn is only as good as the information it contains. Engineers carrying out a desk study must gather as much information as possible to ensure that investigations are properly focused once on site.
The problem, as ever, is time. With shorter and shorter lead-in times for work, engineers' time is increasingly precious. Often information is difficult to locate and many hours can be wasted looking for data rather than analysing it.
This was the challenge facing researchers at the University of Wales in Cardiff when they were commissioned by Railtrack Great Western to source historical information for two stretches of railway line in Wiltshire.
Work involved gathering maintenance data dating back to original construction for a section of the Great Western Railway line between London and Bristol, built by Isambard Kingdom Brunel, and part of the South Wales Direct railway.
Studies, which also involved site walkover surveys, revealed that 30% of embankments in the research area had developed shallow, deep rotational or translational failure. Historically, remedial work involved injection grouting or toe weighting. While grouting records exist, they are often difficult to locate. Records of toe weighting are few and far between, and although they are easily recognised by visual inspection, it is generally impossible to find out when work took place.
Part of the problem stems from the history of ownership and operation of the UK railway network. Before nationalisation in 1948, individual companies that owned and worked on lines kept their own records.
Following nationalisation, these formed the core of the national collection held at British Railways archivist's department. But in 1968, the Transport Act came into force and records maintained by the British Railways Board were transferred to the Public Record Office in Kew, London and the new National Railway Museum in York.
Further movement of documents occurred with privatisation of the rail network and the formation of Railtrack in 1994. Decentralisation meant that information was spread among the various regional offices.
Although the research by Jacquiline Pryor and Steve Bentley was for specific project areas, the procedure should be applicable for any UK railway investigation.
The first step for engineers is to contact one of the seven Railtrack zone offices. However there is no guarantee that all the historical maintenance information (originally held by the British Rail Infrastructure Services Group and now at Railtrack in Bristol) will be available, mainly because not all repair work was logged.
A vast amount of archive material is available at various institutions across the UK. The Public Record Office holds mainly historical information on rail companies, proposed lines and operations but some maintenance records are held.
County records offices should also hold information on maintenance and land purchases, including drawings and correspondence on proposed construction and remedial work.
The National Railway Museum in York principally holds periodicals, books, technical records and a large collection of historical photographs (about one million images of railways are stored).
Other archives are held at the Science Museum in London and at Bristol University, although the latter's collection is based largely around Brunel and the Great Western Railway.
The Permanent Way Institution was found to be a particularly useful source of information. Members are former railway employees, some of whom were directly involved in maintenance work and so can provide first hand accounts of the day to day working of the railways.
There are also many railway historians attached to various societies across the UK, although they are mainly concerned with locomotives and the general running of the railways themselves.
Researchers are keen to point out that while these are general sources of information, specific projects will have sources specific to the area in question. The information is meant purely as a guide and should help reduce the amount of time spent looking for the information rather than using it.
This article is based on extracts from a research paper written by Jacqueline Pryor and Steve Bentley of the University of Wales.