Wales
Compliance
HDC – Penrhos
In July 2024, Hafren Dyfrdwy reported a coliform failure in a sample taken at Penrhos service reservoir. The asset was taken out of supply and ingress was observed through the hatch seals during a flood test as the sealant had deteriorated, the company therefore attributed the failure to this ingress point. An external inspection conducted in August 2024 noted an additional potential contamination route due to an absent flap valve on the environmental discharge point. This had not been noted during the external inspection, which was conducted immediately following the failure, therefore an earlier opportunity to identify a route of contamination into the asset was missed. The company subsequently replaced the hatch seals and fitted appropriate backflow protection on the discharge pipe.
HDC – Bongoch
Taste and odour failures (“musty”) occurred in the Bontgoch water quality zone in April and July 2024. This is a continuation of a pattern of such failures since 2020 and a definitive root cause has not been identified. Contributing factors featuring in investigations include property fittings infringements, low turn-over of the local network, elevations in local chlorine residuals and the occurrence of low concentrations of 2,3,4-trichloroanisole in sample results. The company have employed an independent body to investigate the root cause(s) of these breaches, and the results of this investigation are due in July 2025.
DWR – Plas Talgarth
Dŵr Cymru Welsh Water reported a coliform failure associated with a routine sample taken at Plas Talgarth service reservoir in September 2024. The subsequent company investigation included an internal inspection. During this inspection, a flood test was conducted which identified ingress into the asset through the single-skinned glass reinforced plastic (GRP) hatches. These were not compliant with the company security specification for hatches; bespoke replacement covers were subsequently installed at this site in January 2025.
The single-skinned GRP hatches were noted in the most recent external inspection of the service reservoir conducted prior to the coliform detection. However, they were not raised for replacement or remediation at that time. Instead, at the time of the failure, the relevant risk assessment line was assigned a DWI category of A ‘Target risk mitigation achieved, verified and maintained’ with ‘Technocovers’ listed as a control measure, even though they were not in place at this site.
The company proactively conducted a review into the presence of non-standard hatches at their service reservoirs; 15 sites were identified with non-standard hatches. The company has concluded that capital investment is required at all sites to rectify the identified security issues and work is ongoing. The Inspectorate recommended that the company review the risk assessment categorisation for the relevant risk assessment lines at sites with non-standard hatches and puts in place appropriate mitigation actions.
Events
DWR – Grosmont
In July 2024, a contractor reported to Dŵr Cymru Welsh Water an oily residue in the communications pipe excavation between two consumer properties. The company visited the area on the same day and sampled a single downstream property. The following day one of the properties beside the excavation was sampled, an odour was detected on the pre-flush sample, however no restriction of use was issued. Two days later the company returned to take additional investigatory samples. Six days after the initial report of hydrocarbon contamination, ‘do not drink’ advice was issued to the two properties adjacent to the contaminated excavation.
Following the event the company made changes to their customer contacts procedure to ensure that delays in taking appropriate samples would not result in a delay to the issuance of precautionary restriction of use advice. The procedure was updated to include boundary box sampling where access to the consumer property was not immediately possible, and to consider the presence of on-site hydrocarbon soil contamination as a trigger for issuing restriction of use. Companies are encouraged to widely consider information and data inputs into dynamic risk assessments, particularly when feeding into decisions regarding precautionary restrictions of use.
DWR – Neath – shared supply lead.
Dŵr Cymru Welsh Water reported an event in July 2024 following the failure of a lead sample taken at a consumer property. A subsequent sampling survey did not highlight any elevated concentrations of lead in neighbouring properties. The company found that its communications pipe was polyethylene, however, the consumer’s supply pipe was found to be lead. The company responded promptly, replacing the supply pipe and resampling at the property to confirm that lead concentrations were below the limit of detection as a result. The Inspectorate issued a recommendation regarding the issuance of restriction of use advice in the case of properties on a shared supply pipe. A ‘do not drink’ notice was issued to the original property following the initial lead failure, however, a second property on the shared lead supply was not issued with a restriction of use at the same time. The sample taken at this property showed lead less than the limit of detection at the time of sampling. The company has since updated its lead response procedure to require ‘do not drink’ advice to be extended to all properties that share a supply pipe, where such advice has been issued to one of the properties.
HDC – Newtown do not drink
In July 2024 Hafren Dyfrdwy reported a suspected ingress event to the Inspectorate. The company was alerted to a burst on a three-inch main in the Newtown zone. Whilst the company were repairing the initial burst a secondary burst occurred which drained a tanker which was infusing into the network, the subsequent depressurization allowed ingress into the main. A single property on the affected section was issued with ‘do not drink’ advice and the company elected to install a temporary overland main to maintain a “sanitary supply” to this property. The company acted quickly to repair, disinfect, and flush the main. The ‘do not drink’ advice to the single property was lifted following two sets of satisfactory sample results. However, the company did not follow its own procedure, as taste analysis was not included in these sampling suites. Whilst the company does not carry out taste analysis following the issuing of ‘do not drink’ advice, once the remedial works are complete and chemical analysis suggests it is safe to do so, taste analysis should be completed on the appropriate samples before a restriction of use notice is rescinded.
HDC – Newtown T&O
In September 2024 a consumer from a community centre in Llanidloes contacted Hafren Dyfrdwy to report a chemical/chlorine odour. After some delay and following confirmation of “solvent” and “river” odours by the laboratory taste and odour panel, do not drink advice was issued to the consumer and the heating system cross contamination source was identified during a reactive fittings inspection. There were initial delays in reactive sampling associated with deficiencies in contact classification and escalation processes during the investigation. An opportunity to issue the do not drink advice at an earlier juncture when a quality technician (sampler) also detected a chemical odour at the time of sampling was also missed. Companies are reminded of the importance of clear, unambiguous classification of water quality descriptors during initial contacts and of opportunities to protect consumers when sufficient evidence presents itself prior to the receipt of reactive sample results.
Audits
HDC – tanks and reservoirs
In August 2024, the Inspectorate audited four of Hafren Dyfrdwy’s service reservoir sites in the Wrexham area.
Secondary chlorination using sodium hypochlorite is in use at two of the sites, Sugn Y Pwll and Bronwylfa. At both sites, the sodium hypochlorite is stored in a single bulk tank which is replenished using stock from other company sites, with new product added onto existing. However, there was no provision or procedure for draining and cleaning the bulk tanks. The dosing equipment and bulk tanks are situated in outbuildings at each site, there was no temperature monitoring of these buildings or the product itself, and there was no robust method of temperature control or cooling. Sodium hypochlorite solution gradually decomposes over time and this decomposition is greatly affected by heat, a by-product of this decomposition is sodium chlorate. The Inspectorate made a recommendation for the company to revise the sodium hypochlorite management at the two sites audited to meet the requirements of the relevant standard (BS EN 901:2013), with particular reference to the risk of chlorate formation. Chlorate monitoring was scheduled to take place four times per year, which meant that only a single sample was likely to be taken during the summer and was therefore unlikely to capture any spikes in chlorate concentration. In response, the company increased the chlorate monitoring at service reservoirs with secondary chlorination to eight samples per year from January 2025.
The sampling facilities at each site were inspected during the audit. At two of the sites, Gronwen and Higher Berse, the tapping points for the regulatory sampling points are located on the inlet/outlet main which is used for both filling and discharging the respective reservoirs, it is therefore not possible to ascertain whether samples representative of the reservoirs are being taken at any given time. A recommendation was made to review the sampling arrangements at all company ebb/flow reservoirs to ensure that representative regulatory samples could be taken; the company review is ongoing.
At Gronwen service reservoir, a further recommendation was made regarding the sampling facilities, which were found to be in poor condition with inadequate drainage and a redundant line in the cabinet. Critical facilities such as this should not be found in poor condition and not meeting company asset standards. The Inspectorate understands that programmes to deliver improvements to small but common infrastructure will be spread over time and prioritised according to risk, however deficiencies such as these found at Gronwen service reservoir, should have been resolved in response to historic inspections.
