Porth do not drink event

In May, Dŵr Cymru Welsh Water received a report from a consumer who said they had noticed a pink colour to their water supply. A company distribution inspector attended site the same evening to investigate and confirmed the presence of pink water at this property and a neighbouring property.

Pink water coming out of a tap

Image 1 – pink water coming out of a tap

Inside of a toilet cistern showing pick water

Image 2 – pink water within a toilet cistern

A survey of the area identified eight properties on the same terrace who shared a common supply pipe and therefore a ‘do not drink’ restriction was given to all eight properties. A fast response by the company ensured these consumers were protected, given appropriate precautionary advice and provided with an alternative bottled water supply the same evening as the initial report from the consumer.

A water fittings inspection was carried out the next day which identified, in one of the properties, a flushing toilet cistern with a pink toilet block used within it. The water fittings inspector identified that the flushing cistern contained a float operated valve which did not provide the required air gap and therefore this was thought to be the cause of the problems observed.

One sample from the survey of the affected properties returned a result of three coliforms per 100mL. A resample collected later was satisfactory. Following rectification of the plumbing infringements and satisfactory samples, the properties were given the all-clear to use their water as normal.

Following the conclusion of the company’s investigation and actions, the company reviewed its handling of the event, to identify any learning. It was highlighted that the samples collected initially were not analysed for odour or taste as this was not in the associated procedure. Whilst samples collected from properties given ‘do not drink’ notices were understandably not tasted, an odour test, and taste and odour from up and down stream properties would have contributed to a suitable investigation. The company has subsequently reviewed and updated this procedure to ensure that odour and taste (where appropriate) analysis is undertaken for similar events.

Furthermore, a water fittings inspection at one of the affected properties identified a similar toilet cistern set-up to that of the original property, thought to have been the root cause of the event. Although there were no toilet blocks within the cistern, the pipework set-up meant it was highly likely that back siphonage could occur in the event of a depressurisation. An infringement notice was therefore also issued to this property, as well as the original property, to ensure backflow protection was in place. This event highlights the potential effects that common plumbing infringements can have on consumers and their neighbours.

Nantyglo Boil water notice event

An investigation into a private supply pipe leak on 21 January 2025 highlighted that the leaking pipework was located within a drainage chamber on the property. This was escalated to the Water Quality team and precautionary boil water advice was issued whilst further investigation and mitigation was undertaken.

All samples taken in response were satisfactory, the leak remained under positive pressure and was resolved with further samples taken before the advice was rescinded. The event highlights the need to assess potential ingress risks arising from supply pipe routing decisions across consumer properties, particularly where foul water drainage channels and chambers have been chosen by installers for convenience and/or cost-efficiency.

Bolton Hill Audit

An audit of raw water deterioration and process breakthrough was conducted at Bolton Hill works in West Wales in January 2025. The site was selected based on historic breakthrough of the taste and odour precursor 2- methyl isoborneol (MIB) with elevated concentrations in the wash water supernatant return playing a key role in the treatment challenge in May 2023. 

Overall, the site was tidy and well run with evidence provided to demonstrate that company process control procedures were being followed. Raw water monitoring and alarm systems were in place for the discrete (remote) and combined sources.

Aids had been developed to inform operational responses to potential raw water contaminant breakthrough such as a local decision flow chart for Geosmin/MIB sample result responses relating to the control of the supernatant return and raw water systems. Of the 12 Granular Actived Carbon reactors (GACs) on site, 10 had spent carbon iodine numbers that were below 400 mg/g of carbon with the lowest at 257 mg/g of carbon. The company have since accelerated its GAC media regeneration programme for AMP8 to ensure that the new company iodine number target of 450 mg/g of carbon is achieved and sufficient GAC adsorbance capacity is maintained across its asset base.

An absence of routine functional testing was noted for some lower priority alarms on site, and a recommendation was made in relation to the enhancement of a risk-based programme of comprehensive, routine, end to end testing of all the critical site alarms and associated actions and controls to assure functionality and therefore compliance with regulation 26(4). 

Companies are encouraged to perform regular functional tests of asset operational systems to validate the functionality of their designed control philosophies. Observations were made on the day of the audit that the individual readings of some of the dual and triple validated dosing system control monitors were not visible on SCADA and that monitor deviation alarms were also not displayed in some cases. Companies are encouraged to ensure that all parameters and associated functions used in the control of such systems are clearly displayed on site control systems.

Discolouration complaint 2025/1954

A water quality complaint was received by the Inspectorate in February 2025 in relation to ongoing discolouration being experienced by a consumer in Monmouth. Reactive samples taken at the property and immediate neighbours contained concentrations of iron above regulatory limits and on investigation an oversized 77m length of cast iron main directly feeding the properties was identified as the root cause. The material of this main was originally recorded on the company GIS system as MDPE but was found to be cast iron on investigation with records subsequently corrected. Accurate records are critical aids to investigations and companies are encouraged to focus on the robustness and feedback systems involved in maintaining this accuracy.       

Flushing of this section of main was undertaken to remove accumulated sediment and it was then slip-lined to mitigate recurrence of the issue. During initial flushing activities follow-up sample results continued to fail for iron despite post flush colour grading being deemed satisfactory. Companies are encouraged to employ empirical flushing philosophies and practice to ensure the intended water quality outcomes are achieved and validated.