The Flood Risk TAN15 Doesn’t Map: Quarries and Mine Tips in Wales

On the afternoon of Thursday 30th July 2026, the village of Gwaelod-y-Garth, below Garth Hill on the north-western edge of Cardiff, flooded within about a minute of a blast at the quarry above it. Four days on, the operator has confirmed the flooding followed one of its own planned blasts, the Health and Safety Executive and South Wales Police are investigating jointly, and the First Minister has visited the village and promised answers. What nobody has yet said is which regulator was responsible for the water before it arrived on Heol Berry. That silence is the substance of quarry flood risk in Wales. The maps that govern planning decisions show rivers, the sea and surface water; the safety regime that governs quarries is written around rock, explosives and tips. Stored water beneath a working quarry falls between them.

This article was updated on 3rd August 2026. It reflects the position as at that date, in a story that is still moving.

What we now know, and what nobody has yet said

Fifteen properties were affected and thirteen were evacuated by South Wales Fire and Rescue Service, with water reaching 1.5 metres in the worst-affected homes on Heol Berry and Heol y Nant. Residents described brown, sediment-laden water arriving in under a minute; one told ITV News that the water "was just brown and rushing down the hill like a river". Dŵr Cymru Welsh Water has ruled out its own network, which removes the obvious alternative explanation of a burst main.

The decisive change came the following day. Breedon, which operates Taff's Well Quarry immediately above the village, moved from describing an unspecified "incident" to naming its own operation as the event that preceded the flood.

That is a sequence, not an admission of cause, and the distinction matters. But it narrows the field considerably. First Minister Rhun ap Iorwerth visited on Saturday 1 August, told residents "we have your back", and called for an in-depth investigation so that lessons are learned. Cardiff Council reopened Main Road that weekend but reported on 2 August that Heol Goch remains shut to vehicles, pedestrians and cyclists, that the road needs significant repairs, and that the embankment alongside it must be independently certified as safe before any reopening decision is taken.

Four days after the event, no Section 19 flood investigation has been announced, Natural Resources Wales has issued no news release, the Health and Safety Executive has said nothing publicly beyond confirming it is working with South Wales Police, and not one geologist or hydrogeologist has been quoted by name in any coverage.

The Welsh Government's response is worth reading closely for who gave it. The statement came from the Cabinet Minister for Local Government, Housing and Planning rather than from an environment or climate portfolio, and it drew a firm line under further comment.

Who is investigating, and under which law

Four bodies have a stake in what happened at Gwaelod-y-Garth, and each is looking at a different question under a different statute. None of them owns the whole event, which is precisely the problem.

The Quarries Regulations 1999 are detailed where you would expect them to be. Part V governs explosives and shotfiring, setting duties on the operator and the shotfirer and dealing with misfires. Part VI governs excavations and tips, and regulation 33 requires a geotechnical assessment whose required content, down to cross-sections and site investigations, is set out in Schedule 1. What the regulations do not contain is a part devoted to water held in or beneath the quarry. That risk is picked up, if at all, through the geotechnical assessment, the operator's health and safety document, and the general duty under section 3 of the 1974 Act to protect people who are not employees. It is a duty rather than a regime.

The reservoir rule that does not apply to quarries

If a body of water above a Welsh village is held behind a raised structure, there is a well-developed safety system for it. Under the Reservoirs Act 1975, Natural Resources Wales is the enforcement authority, and Wales lowered the registration threshold to 10,000 cubic metres above natural ground level. Registered reservoirs are inspected by panel engineers on a statutory cycle. NRW's most recent reporting records 405 large raised reservoirs in Wales as at 31 March 2025, of which 277, or 68 per cent, are designated high-risk.

Quarries sit outside that system by design. NRW's own registration guidance excludes quarry lagoons and tips regulated under the Quarries Regulations 1999, and mine lagoons regulated under the Mines Regulations 2014, from reservoir registration altogether.

Water held at a quarry is not a registered reservoir. There is no panel engineer, no entry on NRW's register and no statutory inspection cycle. It falls to the operator and to HSE instead.

The logic of that exclusion is sound enough on its own terms: you should not regulate the same asset twice, and HSE already has inspectors on quarry sites. The difficulty is that the two regimes are not equivalent. One produces a public register, an independent engineer and a defined inspection interval. The other produces an internal document. And neither addresses water that was never impounded by anyone in the first place, but was sitting in the rock. This is the same structural issue we set out in our explainer on reservoir flooding and, more closely still, on infrastructure flooding, where water escaping an engineered system requires site-specific review because no national mapping covers it.

Will there be a Section 19 flood investigation?

None has been announced. Cardiff Council is the Lead Local Flood Authority, and section 19 of the Flood and Water Management Act 2010 requires an LLFA, on becoming aware of a flood in its area, to investigate which risk management authorities have relevant functions and whether they exercised them, and to publish the results. The duty is real but qualified: it applies only "to the extent that it considers it necessary or appropriate". As we explain in our guide to what a Lead Local Flood Authority does, there is no national definition of a significant flood, so each authority sets its own trigger, and Cardiff's Local Flood Risk Management Strategy sets no numerical threshold.

Fifteen properties flooded and thirteen households evacuated would meet most authorities' tests. But there is an awkwardness in the statute here that is worth naming. Section 19 is framed around the conduct of risk management authorities. A quarry operator is not one. An investigation that faithfully answered the statutory question could conclude that every risk management authority did what it was supposed to and still tell residents nothing about why their homes flooded.

• The precedent exists. After the Skewen mine water blowout in January 2021, Neath Port Talbot's section 19 report treated the Coal Authority as a risk management authority and made recommendations to it, including the design of a mine water remediation scheme.
• The timescales are long. Rhondda Cynon Taf's section 19 report into the Storm Dennis flooding of February 2020 at Taff's Well was published in January 2022, roughly 23 months after the event.
• The duty itself is contested. A Welsh Government-commissioned review published in August 2023 found a lack of clarity among local authorities about the purpose of a section 19 investigation, describing the duty as subjective and open to interpretation.
• Cardiff's baseline is unhelpful. The council's Local Flood Risk Management Strategy records that "there is little documented evidence of groundwater flooding in the area and therefore the risk of flooding from this source is considered to be small".

The wider statutory framework, including how section 19 sits alongside the asset register duty and the Welsh reservoir provisions, is set out in our article on the Flood and Water Management Act 2010.

Why the water probably came from underground

No official body has confirmed a mechanism, and no named geologist has been quoted on this incident. What can be said is that the geology at Taff's Well makes a stored-water release a plausible explanation rather than a fanciful one, and that the site has a documented history of quarrying breaking into natural voids.

The quarry works dolomitised limestones of Lower Carboniferous age, considerably folded and faulted, and produces on the order of a million tonnes a year. The South Wales Carboniferous Limestone is heavily karstified: the British Geological Survey's hydrogeology of Wales records around 80,000 dolines on the north crop alone, dye-tracing connections between swallow holes and resurgences of up to seven kilometres, and, pointedly, that mine dewatering and quarrying activities have reactivated karst systems, requiring remedial measures during infrastructure works.

Beneath Garth Hill there are caves, and there are old iron workings that followed the same solution features. Ogof Ffynnon Taf was uncovered by quarrying at this site in March 1986 and later connected to Lesser Garth Cave; it lies close to the quarry face. Taff's Well is also the site of Wales's only thermal spring, whose waters have circulated to depths of several hundred metres over thousands of years along the fault that crosses the Taff gorge. This is not a settled, well-drained aquifer. It is a network of conduits, historic workings and fractures, some of them water-filled, in rock that is being blasted.

Groundwater in karst does not behave like groundwater in a gravel aquifer. It moves through discrete conduits at speeds measured in kilometres, and where a conduit is intercepted, the response is immediate rather than gradual.

That is why a groundwater flood risk assessment in limestone country is a different exercise from one on a river terrace, and why, as we set out in our review of groundwater flood risk maps in the UK, there is no national map that would have flagged this. It is also a familiar principle from a different setting: in the Upper Tribunal case we examined in groundwater, heritage and the cost of getting flood risk wrong, engineering works altering subsurface flow did not remove water so much as redistribute it. The same is true of a hillside with holes in it.

What TAN15's Flood Map for Planning shows, and what it misses

Technical Advice Note 15, republished on 31 March 2025 and now replacing both the 2004 TAN15 and TAN14, tests development against the Flood Map for Planning produced by Natural Resources Wales. Groundwater appears in TAN15 essentially once, in the list of sources a Strategic Flood Consequences Assessment should consider. There is no category for mine water, quarry-void water or karst release, and nothing in the map would have identified the hazard above Gwaelod-y-Garth.

Natural Resources Wales, which responded to around 2,100 planning consultations in 2024–25 where flood risk was a factor, reviews flood zones and defences as statutory consultee. It does not, through the Flood Map for Planning, tell an applicant whether a site sits below a working quarry or above old workings. It is the same blind spot that allowed a slow, unseen leak to collapse a canal embankment at Whitchurch in Shropshire: the structure had been inspected, but nobody was looking underneath it.

Wales already treats extraction ground as a hazard in its own right

Wales does not leave the ground itself unregulated. It regulates it outside the planning system, through dedicated tip-safety law rather than TAN15 — a lineage that runs directly from Aberfan. On 21 October 1966 a colliery spoil tip built over a natural spring liquefied and slid into Pantglas Junior School, killing 144 people, 116 of them children. The Tribunal of Inquiry found the disaster had been preventable, and it led to the Mines and Quarries (Tips) Act 1969, which brought quarry tips as well as colliery waste under statutory safety oversight for the first time.

The Disused Mine and Quarry Tips (Wales) Act received Royal Assent on 11 September 2025, replacing the 1969 framework and creating the Disused Tips Authority for Wales, operational from April 2027. Wales has 12,566 disused coal tips on the Welsh Government's figures, inspected twice yearly and categorised A to D; Rhondda Cynon Taf holds 3,427 of them. More than £220m of combined UK and Welsh Government funding has gone into coal tip safety so far.

Taff's Well is an active, licensed quarry, not a disused tip, so none of that Act applies to it. But the pattern is instructive. Where Wales has decided that extraction-industry ground is a distinct hazard, it has built a register, an authority and an inspection cycle. Where the ground is still being worked, and the hazard is water rather than spoil, it has built none of those things.

The complaints that preceded the flood

One further piece of context belongs on the record, because it goes to what was being monitored and what was not. In March 2026, five months before the flood, The Cardiffian reported that residents of Pentyrch and Gwaelod-y-Garth were organising collectively over blast damage to their homes, some of them on Heol Berry. Blasting was reported as taking place weekly. Residents described cracking to walls and ceilings; Breedon said it had carried out numerous tests over the years and there was no link between the cracks and the blasts; and Cardiff Council confirmed that the quarry complied with its approved Vibration and Noise Management Plans and directed complainants to planning enforcement.

Blast vibration at Taff's Well was monitored under an approved management plan and found compliant. Groundwater was not part of that conversation, because no planning condition or map required it to be.

That contrast is the point, and it is a policy observation rather than an allegation. Nothing in the March reporting suggests anyone foresaw a flood, and Breedon's position on vibration and property damage was and remains that no link has been established. But a monitoring regime aimed entirely at ground vibration will not detect a hydrogeological hazard, however diligently it is run.

What a site-specific assessment should actually check

Quarry flood risk in Wales is not really a mapping problem. It is an information problem. The data that would answer the question — blast records, dewatering volumes, void surveys, the operator's own geotechnical assessment — exists, but it sits with the operator and with HSE, and an applicant has no route to it. In our experience the analysis is rarely the hard part on sites like this; getting hold of the records is. So the practical response is to treat the ground as a named flood source and be explicit about what you could not find out.

1. Check historic mapping and mineral records. Old Ordnance Survey editions, the Mining Remediation Authority's interactive viewer and, from April 2027, the Disused Tips Authority for Wales register establish what has been worked or tipped on or near a site.
2. Commission a geotechnical desk study. Made ground, backfill and landslip are exactly what Unda's free ground-conditions checker flags before infiltration or foundation design is finalised. Old quarry backfill is one of the named examples of made ground it identifies.
3. Treat karst and worked limestone as a groundwater question, not a geotechnical footnote. Where the bedrock is Carboniferous Limestone, conduit flow, swallow holes and historic workings need a design groundwater position derived from records and monitoring, not an assumption.
4. Ask what the neighbour does with its water. Where an active quarry, lagoon or dewatering operation sits upslope, its water management is a material fact about your site. It will not appear on any map, and the operator is not a statutory consultee, so it has to be asked for.
5. Build hidden-hazard scenarios into the Flood Consequence Assessment. Alongside the breach and overtopping scenarios TAN15 already requires for Defended Zones, describe what would happen if ground beneath or beside the site failed.
6. Engage Natural Resources Wales early. NRW reviews flood risk as statutory consultee, and confirming how it views an adjacent extractive site is a conversation worth having before submission rather than after.

What this means for developers and planners near quarry or mining legacy land

None of this changes what TAN15 asks of an application in the way that matters most. The burden sits with the applicant to demonstrate that flood consequences are acceptable, and the Flood Map for Planning is the starting point for that evidence, not the end of it. Wales' record £85m flood and coastal risk management programme for 2026–27 funds coastal schemes, river alleviation and natural flood management. None of it maps or manages the water beneath a hillside.

There is also a live planning dimension at this particular site. Cardiff Council's Replacement Local Development Plan candidate-site register includes a 4.6-hectare extension to Taff's Well Quarry, promoted by the operator. Whatever the investigation concludes, that proposal will now be considered by a council that is simultaneously the mineral planning authority, the lead local flood authority and the highway authority repairing Heol Goch.

Public flood-defence funding pays for the risk you can see. Nothing in the 2026–27 Welsh programme identifies the water sitting inside the hill above a development site.

If your site sits near current or former quarrying or mining activity, Unda prepares TAN15-compliant Flood Consequence Assessments that look beyond the Flood Map for Planning, drawing on our groundwater and ground-conditions expertise to test what the mapped data cannot see. For sites where the geology is the governing factor, our groundwater flood risk assessments are led by a hydrogeologist.