Know what is buried before you dig

The most dangerous part of a Glasgow site is usually the part you cannot see. Above ground, the tenement blocks, the converted warehouses and the newer residential towers are legible and well documented. Below ground sits a different city: gas mains, electricity cables, water and sewers, abandoned services from buildings long demolished, and in an industrial city the occasional surprise that predates anyone's records entirely. The first time most managers think hard about any of it is when a trial pit, a new bin store footing or a drainage repair puts a machine bucket into something that was very much still live.

Striking a buried service is one of the few site events that is simultaneously expensive, dangerous and almost entirely preventable. The prevention is not heroics on the day. It is knowing what is down there before the dig, and holding that knowledge somewhere the next person can find it. A blind excavation is a gamble; a controlled one is a paperwork exercise done in the right order.

The ground keeps secrets that records should not

The reason buried services catch people out is that the knowledge of them is scattered and decays. The original drawings sit with a utility company or a long-gone contractor. The as-built for a service installed twenty years ago may never have made it onto any site record. A previous owner capped a main and never told anyone. Each gap is small; together they mean that "what is under this car park" is rarely a question anyone can answer with confidence on demand.

For an older Glasgow site with layers of industrial and residential history, the uncertainty is greater, not less. Made ground, filled cellars, redundant culverts and disused services are part of the inheritance. None of that is a reason to avoid groundworks. It is a reason to treat the subsurface as something to be surveyed and recorded rather than discovered.

Before any spade goes in

Safe digging follows a settled sequence, and the value of writing it down is that it forces the questions to be asked before the machine arrives rather than during the incident. A practical pre-dig record covers:

• Utility plans. Request and gather records from the relevant utilities for gas, electricity, water, telecoms and drainage covering the area of work.
• A site survey. Where records are thin or the area is sensitive, commission a detection survey to locate services on the ground rather than relying on drawings alone.
• Marking up. Transfer what is known onto a plan and onto the ground, so the excavation team can see the constraints.
• Safe-dig method. Define how the work will proceed near services, including hand-digging or vacuum excavation where appropriate.
• A permit. Authorise the excavation formally, naming who checked the services and who signed the work on.

The discipline is the same one that protects a building when contractors work on its systems. We have written about contractor permits and keeping work accountable; excavation is the case where the permit and the buried-services check are the same conversation, and skipping it has the sharpest consequences.

Drawings are not enough on their own

A trap worth naming: utility records are an indication of what should be there, not a guarantee of what is. Services get re-routed, depths vary, and drawings are drawn to be useful rather than precise. The records tell you what to expect; a detection survey tells you what is actually present in the ground you are about to open. For anything beyond the most trivial dig in known-clear ground, treat the two as complementary, not interchangeable. The record narrows the search; the survey confirms it.

Hold the subsurface on the site, not in a drawer

The reason this matters as a records problem, and not just a method-statement problem, is repetition. A site is not dug once. The bin store this year, the cycle shelter next year, the drainage repair the year after each reopen the same ground, and each time the knowledge gathered last time should be available rather than re-bought. When the utility plans, the detection survey and the marked-up locations live with the site record, the second excavation starts from what the first one learned. When they live in a folder on a contractor's laptop, every dig starts blind again.

This is the same argument we make about the land generally: the things that matter about a site, above and below ground, belong attached to its boundary so they survive the people who found them out. The subsurface is simply the most literal example, and the one where forgetting is most expensive. It connects directly to the drainage and SuDS record, since much of what you most want to avoid hitting is the drainage you are also obliged to inspect.

A short close on responsibility

Buried-service strikes are investigated as failures of planning, not bad luck, because they almost always are. The defensible position is simple to state and only a little work to hold: we requested the records, we surveyed where the records were thin, we marked it up, we permitted the dig, and here is the dated evidence of all four. Build that habit and groundworks stop being the part of site management you dread. In SAMRISK, the survey results, utility plans and excavation permits can sit on the site record, so the ground a building stands on is as well documented as the building above it.