Anglian Water Services Ltd v Environment Agency [2020] EWHC 3544 (Admin)

[2020] EWHC 3544 (Admin)
IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
Venue Royal Courts of Justice Strand, London, WC2A 2LLDate 21 December 2020
Before
THE HON. MRS JUSTICE THORNTON DBE
Between
Anglian Water Services Limited; Environment AgencyClaimantEnvironment AgencyDefendantAnglian Water Services Limited Claimant- and -Environment Agency Defendant
Mr Meyric Lewis (instructed by Anglian Water) for ClaimantMs Jacqueline Lean (instructed by Environment Agency) for DefendantHearing Hearing dates: 18 - 19 November 2020
JUDGMENT(Approved by the court)Anglian Water Services v Environment AgencyThe Hon. Mrs Justice Thornton Introduction
[1]By this judicial review, Anglian Water Services Limited challenges the Environment Agency’s classification of the water quality at three popular beaches in Lincolnshire as ‘good’ in 2019. By contrast, the water quality was classed as ‘excellent’ in 2018.[2]The classifications were made pursuant to the Bathing Water Regulations 2013/1675, which implement Directive 2006/7/EC concerning the management of bathing water quality. The legal framework protects public health by a robust classification scheme for grading the water quality of bathing waters as ‘excellent’; ‘good’; ‘sufficient’ and ‘poor’. The gradings enable bathers to make informed personal choices about whether to swim; provide an assessment of regulatory compliance and assist in the management of water quality.[3]Anglian Water is the statutory water and sewerage undertaker for the Anglian region, including Lincolnshire. It carries out its functions pursuant to the Water Industry Act 1991.[4]The company contends that the 2019 classifications are legally flawed because the Agency did not discount water samples taken in June 2019 showing highly elevated levels of faecal contamination. The sampling coincided with periods of heavy rain in Lincolnshire and are said, by the company, to be unrepresentative of the water quality at the bathing water. There is provision in the Regulations for the Environment Agency to discount samples taken during short term or suspend sampling during unexpected pollution events which the Agency did not employ. The inclusion of the samples in question is said to have led directly to the drop in water quality classification from ‘excellent’ to ‘good’.[5]The Defendant is the Environment Agency. The Environment Agency is responsible for profiling, monitoring, assessing and classifying the quality of bathing waters in England and Wales under the regime in the Bathing Water Regulations. The Environment Agency contends that its decision not to discount the samples or suspend sampling was an entirely rational exercise of discretion in a complex regulatory area which the Court should be slow to interfere with.[6]The issue raised by this claim is how the Environment Agency responds to short term or unexpected pollution events in its sampling and grading of bathing waters. In particular:a. how the Environment Agency exercises its discretion to discount water quality samples taken during short term pollution events pursuant to Regulation 14(5) of the Bathing Water Regulations;b. the Agency’s use of its Pollution Risk Forecasting system to predict short term pollution events;c. whether the Environment Agency acted unlawfully in failing to declare the heavy rainfall of June 2019 to be an ‘Abnormal Situation’.

Background

[7]Bathing in contaminated water can cause gastrointestinal infections and other health impacts. The most common cause is faecal contamination, the main sources of which include; pollution from sewage, animal faeces and run off from agricultural land. The presence of human or animal faeces in water is commonly detected by monitoring several faecal index bacteria, including Intestinal Enterococci (IE) and Escherichia Coli (E Coli).[8]It used to be common practice across Europe to discharge raw or partially treated sewage into coastal waters, regardless of the danger to human or ecosystem health. Faced with a growing tourism industry, increased awareness of threats to human health and concerns about the environment, the EU adopted the first Bathing Water Directive (76/160/EEC) in 1976 with the ambition of improving bathing water quality within ten years. The 1976 Directive set limit values for contaminants, established monitoring standards, specified sampling frequencies and imposed reference methods for analysis. As a result, bathing water quality improved across Europe, particularly in coastal waters. The 1976 Directive was subsequently replaced by Directive 2006/7/EC, concerning the management of bathing water quality, which is transposed in the UK by the Bathing Water Regulations 2013/1675.[9]There are over 400 designated bathing waters in England including the three breaches under scrutiny in this claim: Cleethorpes, Humberston Fitties and Ingoldmells South. Cleethorpes is described by the Agency as a popular resort with a gently sloping sandy beach. The beach at Humberston Fitties is wide and sandy and backed by dunes, with caravan parks nearby. The local authority for both beaches is North East Lincolnshire Council. The beach at Ingoldmells South is located in a busy resort town on a fairly developed stretch of coastline. The local authority is East Lindsey District Council.

The EU Bathing Water Directive

[10]Directive 2006/7/EC concerns the management of bathing water quality in England and Wales. Recitals to the Directive provide: “(6) ... This Directive should use scientific evidence in implementing the most reliable indicator parameters for predicting microbiological health risk and to achieve a high level of protection… ... (8) ... The public should receive appropriate and timely information on the results of the monitoring of bathing water quality and risk management measures in order to prevent health hazards, especially in the context of predictable short-term pollution or abnormal situations. New technology that allows the public to be informed in an efficient and comparable way on bathing waters across the Community should be applied. (9) For the purpose of monitoring, harmonised methods and practices of analysis need to be applied. Observation and quality assessment over an extended period are necessary in order to achieve a realisticbathing water classification. (10) Compliance should be a matter of appropriate management measures and quality assurance, not merely of measuring and calculation. A system of bathing water profiles is therefore appropriate to provide a better understanding of risks as a basis for management measures....” (underlining is the Court’s emphasis).[11]The purpose and scope of the Directive is set out in Article 1: “(1) This Directive lays down provisions for:a. The monitoring and classification of bathing water quality;b. The management of bathing water quality; andc. The provision of information to the public on bathing water quality. (2) The purpose of this Directive is to ‘preserve, protect and improve the quality of the environment and to protect human health’.” a. The monitoring and classification of bathing water quality; b. The management of bathing water quality; c. The provision of information to the public on bathing water quality.[12]Article 2 contains relevant definitions, including two of central relevance to the arguments in this case:
“8. ‘short-term pollution’ means microbiological contamination as referred to in Annex I, column A, that has clearly identifiable causes, is not normally expected to affect bathing water quality for more than approximately 72 hours after the bathing water quality is first affected and for which the competent authority has established procedures to predict and deal with as set out in Annex II; 9. ‘abnormal situation’ means an event or combination of events impacting on bathing water quality at the location concerned and not expected to occur on average more than once every four years;”
[13]Article 3 sets out provisions on monitoring:
“Monitoring … (2) Member States shall ensure that monitoring of the parameters set out in Annex I, Column A, takes place in accordance with Annex IV. … (6) Samples taken during short-term pollution may be disregarded. They shall be replaced by samples taken in accordance with Annex IV. (7) During abnormal situations, the monitoring calendar referred to in paragraph 4 may be suspended. It shall be resumed as soon as possible after the end of the abnormal situation. New samples shall be taken as soon as possible after the end of the abnormal situation to replace samples that are missing because of the abnormal situation. 9. Member States shall ensure that the analysis of bathing water quality takes place in accordance with the reference methods specified in Annex I and the rules set out in Annex V…”
Annex I is set out below. Annex IV provides instructions for sampling. Annex V contains rules on the handling of samples for microbiological analysis.[14]Article 4 on Bathing water quality assessment provides:
“(1) Member States shall ensure that sets of bathing water quality data are compiled through the monitoring of the parameters set out in Annex I, column A. (2) Bathing water quality assessments shall be carried out: (a) in relation to each bathing water; (b) after the end of each bathing season (c) on the basis of the set of bathing water quality data compiled in relation to that bathing season and the three preceding bathing seasons; and (d) in accordance with the procedure set out in Annex II. …”
Annex I sets out the values for IE and E-Coli for each classification (see further below). Annex II sets out the criteria for each classification. Extracts are set out below. (a) in relation to each bathing water; (b) after the end of each bathing season (c) on the basis of the set of bathing water quality data compiled in relation to that bathing season and the three preceding bathing seasons; and (d) in accordance with the procedure set out in Annex II.[15]Article 5(1) on the classification and quality status of bathing waters provides:
“As a result of the bathing water quality assessment carried out in accordance with Article 4, Member States shall, in accordance with the criteria set out in Annex II, classify bathing water as: (a) ‘poor’; (b) ‘sufficient’; (c) ‘good’; or (d) ‘excellent’.”
[16]Article 7 on management measures in exceptional circumstances provides:
“Member States shall ensure that timely and adequate management measures are taken when they are aware of unexpected situations that have, or could reasonably be expected to have, an adverse impact on bathing water quality and on bathers’ health. Such measures shall include information to the public and, if necessary, a temporary bathing prohibition.”
[17]Article 12 on information to the public provides:
“(1) Member States shall ensure that the following information is actively disseminated and promptly made available during the bathing season in an easily accessible place in the near vicinity of each bathing water: (a) the current bathing water classification and any bathing prohibition or advice against bathing referred to in this Article by means of a clear and simple sign or symbol; … (c) in the case of bathing waters subject to shortterm pollution: - notification that the bathing water is subject to short-term pollution, - an indication of the number of days on which bathing was prohibited or advised against during the preceding bathing season because of such pollution, and - a warning whenever such pollution is predicted or present, (d) information on the nature and expected duration of abnormal situations during such events; (a) the current bathing water classification and any bathing prohibition or advice against bathing referred to in this Article by means of a clear and simple sign or symbol; … (c) in the case of bathing waters subject to shortterm pollution: - notification that the bathing water is subject to short-term pollution, - an indication of the number of days on which bathing was prohibited or advised against during the preceding bathing season because of such pollution, and - a warning whenever such pollution is predicted or present, (d) information on the nature and expected duration of abnormal situations during such events; …”
[18]Annex I sets out the values for each classification as well as the measurement criteria:[19]Annex II on bathing water assessment and classification sets out the criteria for each classification. The criteria for ‘good quality’ are set out below. The criteria for ‘sufficient’ and ‘excellent’ are essentially the same but varied so as to reflect the numerical parameters for each classification:
“… 3. Good quality Bathing waters are to be classified as ‘good’: 1. if, in the set of bathing water quality data for the last assessment period, the percentile values for microbiological enumerations are equal to or better than the ‘good quality’ values set out in Annex I, column C; and 2. if the bathing water is subject to short-term pollution, on condition that: (i) adequate management measures are being taken, including surveillance, early warning systems and monitoring, with a view to preventing bathers’ exposure, by means of a warning or, where necessary a bathing prohibition; (ii) adequate management measures are being taken to prevent, reduce or eliminate the causes of pollution; and (iii) the number of samples disregarded in accordance with Article 3(6) because of shortterm pollution during the last assessment period represented no more than 15% of the total number of samples provided for in the monitoring calendars established for that period, or no more than one sample per bathing season, whichever is greater.”
Bathing waters are to be classified as ‘good’: (i) adequate management measures are being taken, including surveillance, early warning systems and monitoring, with a view to preventing bathers’ exposure, by means of a warning or, where necessary a bathing prohibition; (ii) adequate management measures are being taken to prevent, reduce or eliminate the causes of pollution; and (iii) the number of samples disregarded in accordance with Article 3(6) because of shortterm pollution during the last assessment period represented no more than 15% of the total number of samples provided for in the monitoring calendars established for that period, or no more than one sample per bathing season, whichever is greater.”

The 2013 Bathing Water Regulations

[20]The Regulations follow the structure of the Directive in making provision for: monitoring and classification of bathing waters; their management; and provision of information to the public.

Part 1 General provisions

[23]Regulation 2 concerns interpretation. 24.“Short term pollution” means:
“contamination by intestinal enterococci or Escherichia coli where the appropriate agency— (a) has identified the causes, and (b) does not normally expect the contamination to affect bathing water quality for more than approximately 72 hours after the bathing water is first affected.”
[25]“Relevant procedures for short term pollution” means:
“(a) in relation to the appropriate agency, any measures, including appropriate agency management measures, to prevent reduce or eliminate the causes of pollution and may include surveillance, early warning systems or monitoring with a view to preventing bathers’ exposure to pollution by means of a warning, or, where necessary, advice against bathing; and (b) in relation to a local authority, any measures, including local authority management measures, to notify the public that the bathing water is affected by short-term pollution;”
[26]“Abnormal situation” means: “an event or combination of events impacting on bathing water quality which the appropriate agency would not expect to occur, on average, more than once every four years”.[27]By Regulation 3(1) the Secretary of State must identify and maintain a list of bathing waters at which a large number of people are expected to bathe. By Regulation 3(4) the Secretary of State is obliged to publish, annually, a complete list of all bathing waters in England stating their classifications.

General duties

[28]By Regulation 5(1) the Secretary of State and the Environment Agency must exercise their relevant functions in England and Wales—a. so as to ensure that, by the end of the bathing season in 2015, all bathing waters are classified under regulation 11 at least as “sufficient”;b. so as to take such realistic and proportionate measures as they each consider appropriate with a view to increasing the number of bathing waters classified under regulation 11 as “good” or “excellent”; andc. in all other respects, so as to ensure compliance with the requirements of the Bathing Water Directive.

Part 2: Bathing Water Profiles

[29]By Regulation 7 the Environment Agency must ensure that prior to the start of each bathing water season it has established a bathing water profile for every bathing water.

Part 3: Monitoring of bathing waters and public information

[30]Regulation 8(1) requires the Environment Agency to “establish a monitoring programme for every bathing water”. The details of the monitoring for E-Coli and IE is set out in Schedule 4 Part 1, including the location of the monitoring point; the monitoring calendar; storage and transport of samples before analysis; as well as methods of analysis.[31]Regulation 9 imposes a duty on local authorities to ensure that specified information about the bathing water is actively disseminated and made available to the public in the vicinity of the bathing water.

Part 4: Bathing Water Assessment and Classification

[32]By Regulation 10, the Environment Agency must, at the end of every bathing season (15 May to 30 September in the UK) in respect of every bathing water:a. Prepare a set of bathing water quality data for that season; andb. Carry out a bathing water quality assessment using the set of bathing water quality data compiled in relation to that season and the relevant assessment period.[33]The bathing water quality data is compiled from the results of samples taken under Part 1 of Schedule 4 (Regulation 2(1)). The “relevant assessment period” is the immediately preceding three bathing seasons (Regulation 10(2)).[34]Regulation 11(1) provides for the classification of bathing waters:
“On the basis of each assessment made under regulation 10, the appropriate agency must classify every bathing water as “poor”, “sufficient”, “good” or “excellent” in accordance with Schedule 5.”
[35]Schedule 5 sets out the standards for IE and E Coli for coastal bathing waters.

Part 5: Management of bathing water quality

[36]Part 5 of the Regulations makes provision for the management of bathing waters.[37]Regulation 12 applies where the Environment Agency, local authority or sewerage undertaker is aware of an incident of contamination by E-coli or IE other than an incident of short-term pollution other than an incident of short term pollution to which Regulation 15 applies. The Environment Agency must: consult the local authority and, if necessary, the local sewerage undertaker and actively and promptly disseminate information to the public. The sewerage undertaker must, if necessary, promptly consult the Environment Agency and the local authority that controls the bathing water. The local authority must consult and actively disseminate information.

Short term pollution

[38]Regulation 14 applies where, having consulted the local authority that controls a bathing water, the Environment Agency has established ‘relevant procedures for short-term pollution’ at the bathing water. Regulation 14(2) requires the Environment Agency to include information about anticipated short term pollution and proposed management measures in the bathing water profile; publish information on its website and notify the local authority when short term pollution is predicted.[39]Regulation 14(3)-(5) is a particular focus of the claim and is set out in full:
“(3) The local authority must ensure that the following information is actively disseminated and promptly made available to the public during the bathing season in an easily accessible place in the near vicinity of the bathing water – (a) an indication of the number of days for which advice against bathing was issued there during the immediately preceding bathing season because of short-term pollution; and (b) a warning whenever short-term pollution is predicted or present there. (4) In the event of short-term pollution, the appropriate agency must take one additional sample, as soon as possible after the pollution incident is presumed to have ended, to verify that it has in fact ended. (5) The appropriate agency may disregard samples taken during short-term pollution from the set of bathing water quality data for the bathing water if – a. as soon as possible after the end of a short-term pollution incident, the appropriate agency has taken the additional sample required by paragraph (4) in order to verify that the incident has ended; b. the appropriate agency has not included that sample in the set of bathing water quality data for the bathing water; and c. seven days after the end of a short-term pollution incident, the appropriate agency has, if necessary, taken an additional sample to ensure that it has the minimum number required for the bathing water for the bathing season.”
[40]Regulation 15 is headed ‘Relevant procedures for short term pollution’. It provides as follows:
“15. Where there is short-term pollution at a bathing water to which regulation 14 applies – (a) the appropriate agency must – (i) notify the local authority that controls the bathing water, and (ii) operate, or cause to be operated, any relevant procedures for short-term pollution which are not in operation and for which it is responsible; (b) the local authority must – (i) take the local authority management measures which form part of the relevant procedures for short-term pollution there; and (ii) ensure that notification that the bathing water is affected by short-term pollution is actively disseminated and promptly made available to the public during the bathing season in an easily accessible place in the near vicinity of the bathing water.” (a) the appropriate agency must – (i) notify the local authority that controls the bathing water, and (ii) operate, or cause to be operated, any relevant procedures for short-term pollution which are not in operation and for which it is responsible; (b) the local authority must – (i) take the local authority management measures which form part of the relevant procedures for short-term pollution there; and (ii) ensure that notification that the bathing water is affected by short-term pollution is actively disseminated and promptly made available to the public during the bathing season in an easily accessible place in the near vicinity of the bathing water.”

Schedule 4 Monitoring and Abnormal Situations

[41]Schedule 4 paragraph 2 provides that the EA— “Monitoring Calendar 2-(1) The appropriate agency must – (a) establish a monitoring calendar for every bathing water before the start of every bathing season; and (b) take samples at every bathing water no later than four days after the date specified in the monitoring calendar.(2) In relation to any abnormal situation, the appropriate agency - (a) may suspend the monitoring calendar for the duration of the situation; and (b) as soon as possible after the end of the situation, must take sufficient additional samples to replace those missing due to the suspension and to ensure that it has the minimum number required for the bathing water for the bathing season.”

Short term pollution - Pollution Risk Forecasting

[42]In 2012, the Department of the Environment, Food and Rural Affairs (Defra) commissioned the Environment Agency to put together a system capable of making predictions of bathing water quality that could be used to meet the relevant procedures for short term pollution required for Regulation 14 to apply. A pilot system in 2013 developed into the system known as Pollution Risk Forecasting (PRF), which was put in place in 2014. The purpose of the system is to forecast pollution risk. This means it predicts potential future events rather than determining short term pollution events retrospectively. It produces a daily prediction of the average water quality at a bathing water, based, in 2019, on rainfall data. The system is not necessary for all beaches. At some beaches the water quality is consistent so there is no need for a pollution warning system. Other beaches can show elevated levels of faecal contamination but there is no reliable relationship between rainfall and bathing water quality, so the system will not work.[43]The Agency has published a ‘method statement’ explaining the principles and procedures for short term pollution and pollution risk forecasting: “Background The Environment Agency makes a daily forecast of bathing water quality based on parameters known to increase the risk of reduced water quality. The main predictable factor affecting bathing water quality is rainfall, this mobilises faecal indicator organisms from the catchment to the sea thereby increasing the risk of the public swimming in these. When rainfall (or other factors) that are likely to increase the risk of reduced bathing water quality occur, a warning is issued; this process is known as Pollution Risk Forecasting (PRF). The Bathing Water Directive and … Regulations have provisions to disregard samples taken during periods of ‘Short Term Pollution’. ... from being included in classification. This is providing certain management measures are taken including warning the public. The Directive does not specify in detail how countries should implement these management measures however… We have put in place the PRF system to proactively warn people before they bathe of the increased risk as sampling will not allow proactive warnings before people bathe. This goes beyond the minimum requirements of the Directive… ... Information to the public We have interpreted this to mean that there is a requirement for a permanent sign together with an additional reactive sign whenever an increased risk is present. … The permanent sign needs the following information: 1. That the bathing water is subject to STP, or similar words. 2. A count of the number of warnings issued in the previous season. 3. A description of the PRF system including that warnings are likely to last for less than 72 hours. 4. A description of the sources of pollution likely to lead to a reduction in quality. 5. A link to where further information about PRF can be accessed. This information is made available to those beach managers participating in the PRF system through the beach profiles. The reactive sign needs the following information: 1. The sign must be visible to the public, 2. It must state there is an increased risk of pollution, 3. The sign must be appropriately dated. These requirements go beyond those specified as a minimum by the Directive. … Principles and procedures … requirements of the Directive… Information to the public The permanent sign needs the following information: The reactive sign needs the following information: Principles and procedures 5. STP events only take place on those days when a warning coincides with a compliance sample. Without sampling a Pollution Risk Warning is just for public information and protection, not STP 6. We will check the beach signage for the presence ofthe appropriate information warning that the site issubject to STP (or similar words) and if appropriate thenumber of warnings in the previous season. Without thisinformation samples will not be eligible to bedisregarded under STP. 7. If a warning is issued on a sampling day we willcheck for appropriate beach signage warning the publicof the likely reduction in water quality for that day. 8. If appropriate advisory/warning signage is notpresent on a sampled day when we have issued awarning we will not count this as STP, or disregard thissample.” (underlining is the Court’s emphasis)

Environment Agency guidance on Abnormal Situations

[44]Technical guidance issued by the Environment Agency titled ‘Pollution Events affecting Designed Bathing Waters’ (latest version dated 23/04/19) provides guidance on the Environment Agency’s response to pollution incidents that can impact on bathing waters, including Abnormal Situations. The responding Agency officer will need to decide whether the pollution incident is likely to impact on a bathing water and in addition:
“In order to qualify as an Abnormal Situation the pollution incident must meet certain criteria: a. There must be an impact on the water environment (i.e. pollution must be occurring) with the potential to affect designated bathing waters; b. The pollution source is known and it is not likely to occur on average, more than once every four years (for that particular bathing water); and c. It must be within the bathing water season (1 May to 30 September).” a. There must be an impact on the water environment (i.e. pollution must be occurring) with the potential to affect designated bathing waters; b. The pollution source is known and it is not likely to occur on average, more than once every four years (for that particular bathing water); and c. It must be within the bathing water season (1 May to 30 September).”
[45]More guidance is given in Appendix 2:
“Some pollution incidents are really unusual events and where this is the case we can suspend monitoring at the affected bathing waters. If you are unsure whether an event is an Abnormal Situation do not declare this. You should defer the decision until you have further information.”
Chronology of events 10 – 12 June 2019: Heavy rainfall in the Lincolnshire region

Chronology of events

[46]Approximately 2.5 times the average monthly rainfall fell from 10 to 12June 2019 in parts of Lincolnshire. The rainfall caused extensive flooding in Lincolnshire. Around 600 homes in Wainfleet were evacuated and nearly 130 properties flooded when the River Steeping burst its banks. The Met Office described the rainfall in the Lincolnshire region between 10-12 June 2019 as “one of the most significant rainfall June rainfall events across Lincolnshire of the last 50+ years”. 10 – 13 June 2019: Sampling at the three bathing waters[47]The monitoring programme for the three bathing waters in 2019 provided for a water sample to be taken each week during the bathing season at Cleethorpes and Humberston Fitties, making a total of twenty samples for each beach. One sample was to be taken every two weeks at Ingoldmells South, making a total of 10 samples for the season.[48]On 10 June 2019, in line with the monitoring programme, the Environment Agency took samples from Cleethorpes, Humberston Fitties and Ingoldmells South. However, those samples had to be rejected as the courier’s van delivering them to the laboratory broke down and was replaced with a non-refrigerated van, in contravention of the storage and transportation requirements in the Regulations.[49]On 11 June, the Agency’s Pollution Risk Forecast model (PRF) warned of a risk of reduced water quality at Cleethorpes due to heavy rain.[50]On 12 June, samples to replace those taken on 10 June were taken at all three beaches. The results were: Ingoldmells South E Coli 2700 IE 5200 Cleethorpes E Coli 4000 IE 3100 Humberston Fitties E Coli 700 IE 360[51]It is common ground that the samples showed highly elevated levels of E-coli and IE which were sufficiently high so as to be in the ‘poor’ classification. The Agency’s Pollution Risk Forecast model did not issue a warning of short-term pollution at Cleethorpes on this date. Whilst taking samples at Cleethorpes beach, the Environment Agency’s sampler observed 1 bather and 3 beach users.[52]On 13 June, the Environment Agency’s PRF system predicted a short-term pollution event at Cleethorpes. Further pre-scheduled samples were taken at Cleethorpes and Humberston Fitties in line with the monitoring calendar. The Agency’s sampler observed a sign at Cleethorpes beach warning about the risk of pollution but the date on the sign was only partially completed. A photograph of the sign shows the date as “06/19” Accordingly, it is not clear whether the sign was current or had been put up some time previously or whether it was a permanent fixture. The sampler observed 50 beach users either side of the bathing water sample point. 20 June 2019: Skegness Abnormal Situation[53]On 20 June, the Environment Agency declared an Abnormal Situation at Skegness bathing water in light of pumping operations to reduce the Wainfleet flood response. The Abnormal Situation was declared closed on 26June 2019.

June – September 2019

[54]On 21 June Anglian Water and the Environment Agency discussed the sampling during 10-13 June at a quarterly meeting. The minutes of the meeting acknowledged that the failure of the Agency’s refrigeration resulted:
“…in 2 samples in 2 days, during high rainfall. This could result in 4 bathing waters losing their excellent status with consequent impacts on the local economy.”
[55]On 10 September 2019 the Environment Agency set out its position on the sampling in a letter to Anglian Water:
“There were elevated bacteriological counts at all seven local beaches, with the following beaches experiencing poorer quality samples on consecutive days (12th and 13th June): Cleethorpes, Humberston–Fitties, Mablethorpe, Skegness and Chapel St Leonards (Ingoldmells and Sutton on sea only had one sample scheduled). Only Cleethorpes is covered by the Pollution Risk Forecasting tool. Unfortunately at Cleethorpes the criteria for being able to discount the sample was not met because not only was the Local Authority warning signage on the day undated, but also the static signage did not meet Defra requirements. Our approach to bathing water sampling is set out in Annex 1. I confirm we followed the procedures correctly in this instance. Moving forward we will encourage Local Authorities to rigorously implement Pollution Risk Forecasting signage procedures, especially as it is in their interest to do so…”

October – November 2019: Classification

[56]On 28 October 2019 the Agency provided Anglian Water with draft classifications for the three beaches as ‘good’ Anglian Water responded by criticising the ‘back-to-back’ sampling on 12 and 13 June.[57]On 13 November 2019, the Environment Agency published its classification, in which it classified the bathing waters at Cleethorpes, Humberston Fitties and Ingoldmells South as being “Good”.

Grounds of challenge

[58]There are three grounds of challenge to the Environment Agency’s classifications: 1) There is no legally defensible basis for the Agency’s conclusion that the Local Authority’s failure to put out the “correct signage” on 13 June 2019 at Cleethorpes beach meant that the Environment Agency could not determine that a “short term pollution” event had occurred, permitting it to disregard samples taken during short term pollution from the set of bathing water quality data as required by reg. 14(5) of the Regulations. 2) The Agency’s classification of the quality of the three bathing waters was legally invalid as a result of the Agency’s failure to have in place an effective means of identifying a “short term pollution” event which would fall to be dealt with as provided for in regs. 14(4) and 14(5) of the Regulations. 3) Further, or alternatively, the Agency’s failure to exercise the power under paragraph 2(2) of Part 1 of Schedule 4 to the Regulations to declare an “abnormal situation” renders their classifications legally invalid.

Submissions of the Parties

[59]On behalf of Anglian Water, Mr Lewis submitted that the Environment Agency had misinterpreted Regulation 14(5), permitting it to disregard samples taken during short term pollution from the set of bathing water quality data, as not applying where the local authority had failed to put up the correct signage. This he said involved reading words into the Regulation which are not there. There is nothing in the Regulations in relation to local authority signage, which could dictate whether or not a “short term pollution” incident had actually occurred within the meaning of the Regulations and such an interpretation was demonstrably unreasonable. Further, Anglian Water rejected the Environment Agency’s suggestion that they somehow exercised a “discretion”. The obligations on the various organisations; the Environment Agency, water company and local authority were freestanding and did not feed back into Regulation 14(5) as an additional requirement for its operation. The Agency had failed to take relevant considerations into account. There was he said a clear failure by them to observe the requirements of Regulation 14(4) and 14(5) (Ground 1).[60]Mr Lewis submitted that the PRF system was the sole means the Agency used to determine if there was a short-term pollution event so there was no mechanism to determine otherwise where the PRF system does not operate. Logically the other two beaches (Humberston Fitties and Ingoldmells South) should have been subject to the same warning system (and had been incorporated into the system in 2020). The provisions of Regulations 14(4) and 14(5) come into play ‘in the event of short-term pollution’. Regulation 14(4) cannot be excluded simply because the Agency does not have adequate procedures in place (Ground 2).[61]As to Ground 3, it is abundantly clear that an Abnormal Situation should have been declared by the Environment Agency in June 2019 as it was inevitable that the three bathing waters were adversely affected by the extreme rainfall conditions between 10 and 12 June 2019. The Environment Agency’s decision not to do so was said to be because it could not base a declaration on heavy rainfall alone. However, there was nothing in the Directive or Regulations to rule out such an incident being based on rainfall alone. The Regulations require the Environment Agency to form a judgement as to whether the situation before them on a given day was one which “they would not expect to occur, on average, more than once every four years”. It was, he said, no excuse to fail to have an existing system in place to enable them to form such a judgement as is required under the terms of the Regulations. In the circumstances, the Agency should manifestly have excluded the disputed samples whether by suspension of the monitoring calendar or otherwise (Ground 3).[62]On behalf of the Environment Agency, Ms Lean submitted that Anglian Water had misunderstood the Agency’s position on the absence of compliant signage at Cleethorpes beach in this regard. Regulation 14(5) granted the Environment Agency a discretion to disregard samples taken during short term pollution. The Environment Agency has set out how it exercises its discretion in its published method statement and it further explained matters in Mr Dunhill’s statement. The Environment Agency’s exercise of its discretion was entirely rational and logical and in accordance with World Health Organisation guidance (Ground 1).[63]Ms Lean submitted that Mr Lewis’ arguments in relation to Ground 2 were confused. Regulation 14(5) applies only where relevant procedures are in place (Regulation 14(1)) so there was no provision to disregard samples at Humberston Fitties and Ingoldmells South on grounds of short-term pollution. Nor was there any legal obligation on the Environment Agency to have ‘relevant procedures’ in place. If it was being said that the PRF system was inadequate, then this took the Court into R(Mott) v Environment Agency territory. It could not be said there was anything irrational about the Environment Agency’s system so as to justify the Court intervening in an area of technical and specialist judgment (Ground 2).[64]On Ground 3, Ms Lean pointed to the language of paragraph 2(2) of the Regulation (‘may suspend the monitoring calendar for the duration of the Abnormal Situation). This was to be read as a contemporaneous response to a current or ongoing situation. It was not easy to read the language as referring to a retrospective judgment after the event. Accordingly, it followed from this that the abnormal situation had to be identifiable at the time of its occurrence and could not rest on a difficult judgment call which would take time to be arrived at. Thus, the Agency did not declare an abnormal situation based on rainfall because this required on assessment after the event. In contrast, the Skegness abnormal situation could be identified and declared because it was based on over pumping to reduce flood waters.

Discussion

[65]Grounds 1 and 2 concern the Agency’s response to predictable short-term pollution at bathing waters, so it is convenient to consider them together. Ground 3 raises the separate issue of unexpected pollution at a bathing water. Short term pollution of bathing waters Short term pollution in the regulatory scheme

Short term pollution of bathing waters

[66]‘Short term pollution’ is defined in the Regulations as: “Contamination by intestinal enterococci or Escherichia coli where the appropriate agency— (a) has identified the causes, and (b) does not normally expect the contamination to affect bathing water quality for more than approximately 72 hours after the bathing water is first affected.” (Regulation 2(1))[67]Regulation 14 makes provision for short-term pollution. It applies where the Environment Agency has established relevant procedures for short-term pollution at a bathing water (Regulation 14(1)). Regulation 14(2) requires the Agency to include information about short term pollution in the bathing water profile and on its website and to notify the local authority when short term pollution is predicted. Regulation 14(3) requires the local authority to disseminate relevant information near the bathing water. Regulation 14(4) makes provision for additional sampling by the Agency at the end of the event. Regulation 14(5) allows the Agency to disregard samples taken during the pollution event from the set of bathing water quality data for the bathing water. In the event of short-term pollution occurring, the Agency must alert the local authority and activate relevant management procedures (Regulation 15(a)). The local authority must ensure the public is aware of the event (Regulation 15(b)).[68]These provisions reflect provisions in the Directive. Article 3(6) provides that “samples taken during short-term pollution may be disregarded”. Article 12 provides that the information about short term pollution must be actively and promptly made available to the public.Annex II links the implementation of adequate management measures for short term pollution and the discounting of samples taken during the event to the classification of water quality.

Short term pollution and water quality classification

[69]The role that short term pollution plays in the classification of bathing water quality is explained in the World Health Organisation (WHO) Guidelines for Safe recreational environments: “In some instances microbial water quality may be strongly influenced by factors such as rainfall leading to relatively short periods of elevated faecal pollution. Experience in some areas has shown the possibility ofadvising against use at such times of increased risk andfurthermore in some circumstances that individualsrespond to such messages. Where it is possible toprevent human exposure to pollution hazards in this waythis can be taken into account in both grading andadvice. Combining classification (based on sanitary inspection and microbial quality assessment) with prevention of exposure at times of increased risk leads to a framework for assessing recreational water quality as outlined in Figure 4.2. The resulting classification both supports activities in pollution prevention (e.g. reducing stormwater overflows) and provides a means to recognise and account for local cost-effective actions to protect public health (e.g., advisory signage about rain impacts).” (underlining is the Court’s emphasis)[70]Figure 4.2 referred to above sets out a simplified framework for assessing recreational water environments:[71]Thus, as the diagram above shows, a bathing water subject to short term pollution will not see its classification downgraded as long as bathers are effectively discouraging from swimming during any such pollution events. In effect, as the relevant boxes in the diagram indicate, the relevant water quality is ‘very good’, ‘good’ or ‘fair’, as the case may be, but ‘unsuitable for several days after rain’.

Decision not to discount samples in the absence of beach signage at Cleethorpes

[72]I accept Ms Lean’s submission that Regulation 14(5) grants the Agency a discretion to disregard samples taken during short term pollution. The Agency has published the criteria by which its discretion is exercised. “8. If appropriate advisory/warning signage is not present on a sampled day when we have issued a warning we will not count this as STP, or disregard this sample.”[73]Mr Dunhill provided further explanation in his witness statement:
“25. Disregarding samples allows classifications to be made without samples disregarded under STP, so in exercising discretion to disregard samples the Environment Agency decided to make sure the public were warned about predicted variations in water quality before doing so. This is in line with the WHO guidance detailed above, where users can be shown to be effectively discouraged from entering the water following occasional and predictable water quality deteriorations (linked to, for example, rainfall), the area may be upgraded to reflect the water quality that users are exposed to, but only with the accompanying explanatory material. 26. The Environment Agency expects bathers to take account of warnings and make an informed decision before bathing. The Environment Agency therefore wants to be satisfied that the appropriate signage was in was in place before deciding to disregard a sample, as the system is for the protection of the health of bathers rather than to inflate a bathing water quality classification. To ensure this a requirement is put in place that the warning sign is dated. This ensures bathers know if the warning is current, and that the signage is in place in time for bathers to pass the sign so they have the information before they enter the water. 27. The Environment Agency have made these decisions to enable PRF to be implemented based on the balance of protecting the public, providing the widest coverage of PRF and making the system workable and transparent. This approach was taken to give bathers confidence in the system and therefore take notice of warnings. The decisions on the criteria required to disregard samples from classification therefore needed to ensure it could be unequivocally demonstrated that bathers were warned before disregarding these samples. This follows the principles and spirit of the WHO guidelines and the Directive.”
[74]The exercise of the Environment Agency’s discretion not to disregard the sample at Cleethorpes beach on 13 June because of the failings in signage (which were common ground) was in accordance with its published criteria. The criteria are rational and logical given the context of the WHO guidance, the Directive and the structure of the Regulations. Beaches affected by predictable short-term pollution will not be ‘marked down’ on water quality providing public health is protected through effective provision of information to bathers about the pollution. This, in turn, is accordance with the purpose of the Directive and Regulations which is to protect public health.[75]Ground 1 fails.

Pollution Risk forecasting

[76]As explained above, Pollution Risk Forecasting has been developed by the Agency to predict short term pollution at bathing waters. In 2019 only Cleethorpes was included within the model. The model has since been refined for the 2020 bathing water season and Humberston Fitties and Ingoldmells South are now included.

R (BACI Bedfordshire Limited) v Environment Agency

[77]Ian Dunhill, an Environment Agency advisor in the water quality planning team explains the development of the model in his witness statement:
“From the outset it was recognised that bathing water quality is inherently variable and it would not be possible to make absolute predictions of quality given the known uncertainty and variability of the bacteria intestinal enterococci and Escherichia coli which are often associated with faeces… After considering how this was best delivered, the approach decided on was to use a statistical assessment of the bathing water compliance data compared to antecedent rainfall. The end point of the system would be to produce a daily assessment of the average quality expressed as ‘warning’ or ‘no-warning’ when the risk of illness from bathing exceeded a given criteria. The criteria used was a derivative of the standards in WHO’s Guidelines for safe recreational water environments above which “there may be significant risk of high levels minor illness transmission”
. The threshold is a midpoint of a distribution of quality expected on a given day from the antecedent predictive factors which recognises a range of quality will be above, below and around this. For this reason the system was called the pollution risk forecasting system rather than short term pollution system as the purpose of the system is to forecast pollution risk. … The PRF “system” used in 2019 comprises a core forecasting engine based on a bespoke adaptation of the Delfit FEWS system used for flood forecasting purposes. Every morning between May-September this automatically gathers data feeds from rain radar averaged over pre-defined catchments (areas which water courses drain) or nearby rain gauges. These antecedent rainfall totals are compared to pre-defined rainfall thresholds based on an analysis of the average effect of rainfall on bathing water quality results. When a rainfall total is exceeded a pollution risk warning issued… Bathing water quality is complicated and affected by many different factors including, but not limited to, rainfall. Pollution risk forecasting does not claim to be definitive but is based on a statistical analysis of predictive factors against measured bathing water quality seeking to predict the typical response of bathing water to this predictive factor. Considerable time and research was spent to arrive at the most appropriate method to make predictions of bathing water quality. Statistical methods of bathing water quality prediction are widely used worldwide including by the United States Environmental Protection Agency who have produced extensive guidance and documentation for their use of statistical models for bathing water quality prediction… The conclusion of this research was that statistical methods of prediction were the most appropriate to use by the Environment Agency as these are at least as good as other predictive methods given the available data and very significant uncertainty…”[78]The Environment Agency has continued to refine the model since its introduction in 2013. The refinements are explained in Ian Dunhill’s witness statement:
“Forthcoming changes to the PRF system 28. While pollution risk forecasting has been taking place, research has continued and a new method to make pollution risk forecasts more accurate has been developed which is now in use for the first time in 2020. This new method continues to use rainfall, but also takes account of additional factors such as wind, tide, sunlight and seasonality and how these factors interact to affect bathing water quality. These interactions are relevant as rainfall alone while often being a good predictor of bathing water quality can be made more or less important depending on these other factors. 29. The new method is also able to simultaneously take account of rainfall over multiple time periods and catchments rather than simply selecting just one assessment period that best fits the bathing water samples as was used previously. The new method is based on a more sophisticated statistical assessment of bathing water quality than used previously and the factors that can be demonstrated to affect bathing water quality. 30. The format of the PRFs in 2020 will still remain as a fully automated daily forecast delivered by 9am with warnings as appropriate. 31. Using the new system in 2020 the number of beaches subject to PRF has increased. PRFs are now also made at Ingoldmells South and Humberston Fitties.”
[79]I accept Ms Lean’s submission that the PRF system is a classic example of scientific, technical, predictive judgements by a specialist regulator, with which, the Court should be slow to interfere:
“It was common ground that in principle the court should afford a decision-maker an enhanced margin of appreciation in cases…. involving scientific, technical and predictive assessments. Mott v Environment Agency [2016] 1 WLR 4338, para 69 … I see no need to enlarge or refine the basic principle itself, which – as Beatson LJ put it – was that “the Court should afford a decision maker an enhanced margin of appreciation in cases….involving scientific, technical and predictive assessments” at [69]. This was, as he showed…a principle already reflected in several decisions of this Court. He cites the observations of May LJ in R(on the application of British Union for theAbolition of Vivisection) v Secretary of State for theHome Department [2008] EWCA Civ 417 at [1] that although scientific analysis is “not immune from lawyers’ analysis”, a reviewing court must be “careful not to substitute its own inexpert view of the science for a tenable expert opinion”
. The court “should be very slow to conclude that the expert and experienced decision-maker assigned the task by statute has reached a perverse scientific conclusion” (at [77]). I agree. The court must always be astute not to step into the statutory remit of a regulator, or to engage in its own exercise of quasi-scientific judgment. Agency [2016] 1 WLR 4338, para 69 R (BACI Bedfordshire Limited) v Environment Agency 77. [2019] EWCA Civ 1962 at [99])[80]The PRF model has been cited by the EU as an example of a good practice for modelling short term pollution in a report for the European Commission assessing Member State compliance with the Bathing Water Directive entitled “Support to the assessment of Member States’ compliance with the Bathing Water Directive 2006/7/EC (BWD)” (Milieu Consulting SPRL March 2019).[81]I was not provided with any expert reports or other technical analysis to support an argument that the model was flawed. Mr Lewis conceded that he did not seek to challenge the model as so flawed as to be incapable of constituting ‘relevant procedures’ under Regulation 14. In his words, ‘the model is what the model is’ and ‘it is the best we have’.[82]The highest Mr Lewis’ submissions appeared to reach on this ground was a complaint that the PRF system had been extended to Humberston Fitties and Ingoldmells South for the 2020 bathing season, but not the 2019 season. Development and refinement of a complex modelling system is not however unlawful. It would be a perverse disincentive to regulators if a Court were to find it were. Mr Dunhill has explained the reasons for the refinement in question in his witness statement.[83]On the information before the Court, Mr Lewis came nowhere near persuading me that the Court should trespass into scientific, technical and predictive assessments underlying pollution risk forecasting and the refinements to the model between 2019 and 2020.[84]There is a short answer to any argument that the Agency should have disregarded samples at Humberston Fitties and Ingoldmells under Regulation 14(5). Regulation 14 only applies where ‘relevant procedures; (i.e., the PRF) are in place. There is no legal obligation on the Agency to have relevant procedures in place and Mr Lewis could not point to one. Thus, the Environment Agency had no power to disregard samples for Humberston Fitties and Ingoldmells South. In oral submissions, Mr Lewis said he ‘grudgingly’ accepted the position in this regard.[85]Accordingly, Grounds 2 fails.

Abnormal Situation (Ground 3)

[86]“Abnormal situation” is defined in the Regulations as: “an event or combination of events impacting on bathing water quality which the appropriate agency would not expect to occur, on average, more than once every four years” (Regulation 2(1))[87]This reflects the definition in the Directive: “‘abnormal situation’ means an event or combination of events impacting on bathing water quality at the location concerned and not expected to occur on average more than once every four years;” (Article 2)[88]It is apparent from the above that ‘Abnormal Situation’ is broadly defined. It requires the Environment Agency to exercise its judgment to determine whether an event or combination of events is 1) impacting on bathing waters, and 2) an event which the Agency would not expect to occur, on average, more than once every four years. In short, an Abnormal Situation means what it says.[89]The extract below from a 2019 EU commissioned assessment of Member State compliance with the Directive is informative: “The ‘abnormal situation’ exception is not often used by Member States. ...The main problems seem to be identification (i.e., the circumstances that fall under this concept) and reporting (i.e. providing justifications to the EEA/Commission) of abnormal situations. The BWD definition is broad and has not been further specified or refined at national level. An analysis of Member States’ practices shows that abnormal situations encompass a wide variety of events. Despite the definition in the Directive which refers to events impacting on bathing water quality, in practice, ‘abnormal situations’ are also understood as events when bathing and monitoring is not possible. ...Member States should be reminded that the use of this exception is reserved for events specifically related to bathing water quality.” (EU Overview Report/ 66 Milieu Consulting SPRL Brussels, March 2019. Support to the assessment of Member States’ compliance with the Bathing Water Directive 2006/7/EC (BWD) (B358))[90]The Environment Agency has issued guidance on the criteria for an event to qualify as an Abnormal Situation (Environment Agency Technical Guidance): “4) There has to be an impact on the water environment (i.e., pollution must be occurring) with the potential to affect designated bathing waters; 5) The pollution source is known and it is not likely to occur on average, more than once every four years; and 6) It must be within the bathing water season (1 May to 30 September).” (underlining is Court’s own emphasis)[91]The requirement that a pollution source be known narrows the broad definition in the Regulations and Directive. In this case it meant that the Agency did not exercise its judgment to determine whether the rainfall of June 2019 was an event which it would not expect to occur, on average, more than once every four years andwhich was impacting on bathing water quality at the three beaches under scrutiny. In correspondence with Anglian Water, the Agency explained that “Abnormal situations are not applied based on heavy rainfall alone” (correspondence with Anglian Water dated 15 November 2019).[92]The difficulties with the Agency’s approach appear to be starkly evident in the facts of this case. The Agency accepts that the main predictable factor affecting bathing water quality is rainfall because it mobilises faecal indicator organisms from the catchment to the sea thereby increasing the risk of the public swimming in them (Pollution Risk Forecasting and Short Term Pollution Method statement). It is common ground that the Met Office described the rainfall between 10 – 12 June 2019 as “one of the most significant June rainfall events across Lincolnshire for the last 50+ years”. The Agency has plotted the water quality sample results for E-coli and IE on graphs for the years 2016 – 2019. Despite Ms Lean’s efforts to suggest otherwise, the elevated levels in the June 2019 sample results provide a striking contrast with the other results. As Mr Lewis submitted, the events of June 2019 appear, on their face, to satisfy the definition of an Abnormal Situation in the Directive and Regulations. Yet it appears that the Agency may still not have a view as to whether they do. Ms Lean was unable to assist the Court with its query in this regard.[93]In contrast, the Environment Agency declared an Abnormal Situation at Skegness on 20th June. The Agency explained its reason for doing so in preaction correspondence:
“The abnormal situation that was applied at Skegness between 20 and 26 June was in response to exceptional flooding related issues (following the Wainfleet flooding) when floodwater was purposely over-pumped, not directly into the bathing water, but further down the coast.”
[94]In his witness statement Mr Dunhill explained that:
“The reason for declaring the Abnormal Situation was the likely impact on the water environment of the contaminated flood water being over-pumped close to the bathing water, not the heavy rainfall that caused the flood water.”
[95]Mr Lewis submitted that the difference in approach between Skegness and the three beaches under scrutiny in this claim appeared illogical because the Wainfleet flooding was caused by the same heavy rain which occurred on 10 – 12 June. I am inclined to agree.[96]The reason for the Agency’s requirement that the ‘pollution source must be known’ in order to qualify as an Abnormal Situation is explained in its letter of 10 September 2019 to Anglian Water: “We must sample unless an Abnormal Situation is declared or it is unsafe to do so. The decision to declare an Abnormal Situation (AS) sits with our incident management staff. If an AS is declared, the sampling is suspended until after the AS has ended. The Regulations state that an AS cannot be retrospectively declared.” and in pre-action correspondence: “Abnormal situations allow for the suspension of the bathing water sampling, in advance of the samples being taken. The suspension of the calendar during an abnormal situation is discretionary rather than automatic. Once a decision has been made not to suspend the calendar, there is no mechanism for removing samples retrospectively under the abnormal situation provisions in the Regulations. I confirm that we have followed the requirements of the Bathing Water Directive, Regulations, policy and best practice, consequently we cannot now remove any of the sample results as there is no mechanism to do so” (letter to Anglian Water dated 15th November 2019)

and in pre-action correspondence:

[97]In his third witness statement Ian Dunhill explains that:
“The Environment Agency’s position statement on Abnormal Situations… explains that while heavy rainfall could be an Abnormal Situation, this cannot be confirmed as 1 in 4-year event until after the event. As the Environment Agency cannot pre-determine if rainfall is abnormal in time to adequately warn bathers, we do not use rainfall alone as a reason to suspend the sampling calendar.”
[98]In her submissions, Ms Lean took me to the language of Schedule 4 paragraph 2(2):
“In relation to any abnormal situation, the appropriate agency – (a) May suspend the monitoring calendar for the duration of the situation and (b) As soon as possible after the end of the situation must take sufficient additional samples to replace those missing due to the suspension and to ensure that it has the minimum number required for the bathing water for the bathing season.” (a) May suspend the monitoring calendar for the duration of the situation and (b) As soon as possible after the end of the situation must take sufficient additional samples to replace those missing due to the suspension and to ensure that it has the minimum number required for the bathing water for the bathing season.”
[99]The language of ‘suspend the monitoring calendar’ for the ‘duration’ of the situation required, she said, a contemporaneous decision on sampling by the Agency to meet a current or ongoing situation. It could not be read as enabling a retrospective judgment after the event. Accordingly, it followed from this that the abnormal situation had to be identifiable at the time of its occurrence and its qualification could not rely on a complex judgment call which would take the Agency time to arrive at.[100]I accept Ms Lean’s submissions as to the language of the paragraph in question. However, the paragraph is concerned with sampling. It grants the Agency a discretion to suspend sampling for the duration of an ‘Abnormal Situation’. Yet the effect of the Agency’s approach is to use the paragraph to redefine and narrow the meaning of Abnormal Situation so as to close the door to unexpected pollution events with more complex causes. In doing so the Agency has impermissibly narrowed the breadth of the definition in the Regulations. In addition, its approach fails to recognise the distinct regulatory strands in the legal framework. Anglian Water’s complaint is not with the Agency’s decision to continue sampling (or the provision of information to the public which Mr Dunhill focussed on in his explanation of the Agency’s decision making in his witness statement). The logic of the Agency’s approach to sampling is logical. For more complex causes of pollution, it may not be possible for the Agency to form a contemporaneous assessment as to whether the event is one that was not expected to occur on average more than every 4 years.[101]Anglian Water’s complaint lies with the Agency’s treatment of the samples gathered during the pollution event in its subsequent assessment of the bathing water data. Assessment of the water quality data to arrive at a classification of the bathing water is a separate exercise of discretion under the Regulation/Directive. The Agency is given a broad discretion in this regard. Regulation 10 requires the Agency to:
“a. Prepare a set of bathing water quality data for that season; and b. Carry out a bathing water quality assessment using the set of bathing water quality data compiled in relation to that season and the relevant assessment period.”
[102]This reflects the discretion available under the Directive (Article 10).[103]The purpose of assessment is to arrive at realistic classifications of water quality. This is apparent from the recitals to the Directive:
“(9) For the purpose of monitoring, harmonised methods and practices of analysis need to be applied. Observation and quality assessment over an extended period are necessary in order to achieve a realistic bathing water classification.”
[104]It is also apparent in the prescriptive instructions for sampling laid down in the Regulations/Directive. An assessment cannot be realistic if the water quality data on which it is based is flawed.[105]I accept that the Directive and Regulations make no mention of the treatment of Abnormal Situations in the assessment of bathing water quality. This contrasts with the position on short term pollution. It may explain why the Agency has run into difficulties. It may also help to explain why other Member States do not often make use of the exception. Nonetheless, the Agency is granted a broad discretion under the Regulations to assess bathing water quality. Moreover, it is subject to a general duty under Regulation 5 to ensure compliance with the Bathing Water Directive. The Directive lays down a framework to ensure robust and realistic bathing water classifications based on the most reliable parameters. There is ample authority for the Agency to treat sampling as having been suspended during an Abnormal Situation, when it assesses the reliability of relevant samples, where it is necessary to do so to, arrive at a realistic classification for the bathing water in question.[106]The Environment Agency’s concerns about the practical difficulties of declaring an Abnormal Situation at the point of its occurrence in more complex cases recedes once the distinction between the sampling and assessment is understood. For events with an easily identifiable pollution source, as at Skegness on 20th June, there are no difficulties. Sampling may be suspended, as happened at Skegness. For events that look as if they might be an abnormal situation but where the causes are too complex to allow for an immediate judgment call, the Agency will need to continue to sample (and take an additional sample at the end of the event in case it is needed). The assessment process can consider at more leisure whether the event(s) qualified as an Abnormal Situation and thus whether the sampling should be treated as suspended for the duration of that period. In fact, this appears to be good practice at present, as the technical guidance indicates: “1. The responding duty officer (e.g., EMDO) must inform the Sampling and Collection Team Leader an Abnormal Situation is in place and the monitoring is suspended for the impact bathing water(s). 2. It is, however, best practice to attend the site and takewater samples at the bathing water, where safe to do so.If there is a planned sampling run this should go ahead,subject to health and safety consideration, with thesamples coded as Planned Investigation (PI) to ensurethey are not included in the formal assessment ofbathing water quality. If there is no schedule samplingthen the Environment Management team, or EMDO/ABC out of hours, should arrange for an officerto attend the sit to take water samples and collectevidence if necessary, to support future enforcement.” (underlining is the Court’s emphasis)[107]In addition, the witness statement from Ian Dunhill suggest a process of retrospective assessment is already in place: “Since the assessment of the merits of disregarding samples requires judgement, at the end of each bathing season the Environment Agency holds a Bathing Water Panel with members drawn from its bathing water practitioners and experts. One of the tasks of this panel is to assess whether eligible samples should be disregarded and only those samples that improve the classification are put forward to be disregarded.” (Ian Dunhill second witness statement)[108]Accordingly, the Agency misdirected itself in law, in its statements, in pre action correspondence to Anglian Water that: ‘The Regulations state that an AS cannot be retrospectively declared.” … ‘Once a decision has been made not to suspend the calendar, there is no mechanism for remove samples retrospectively under the abnormal situation provisions in the Regulations’ … Consequently, we cannot now remove any of the sample results as there is no mechanism to do so”.[109]Ground 3 succeeds. Summary[110]Drawing together the analysis above:a. The exercise of the Environmental Agency’s discretion not to disregard the water quality sample taken at Cleethorpes beach on 12th June was lawful. It accorded with published criteria which accord with the purpose of the Bathing Water Directive and Regulations which is to protect public health. Beaches affected by predictable short-term pollution will not be ‘marked down’ on water quality providing public health is protected through effective provision of information to bathers about the pollution so as to discourage bathing during the pollution event. Ground 1 fails.b. The Pollution Risk Forecasting (PRF) system used by the Environment Agency to predict short term pollution falls within the category of scientific, technical and predictive assessments by a specialist regulator with which the Court will be slow to interfere. The submissions on behalf of Anglian Water came nowhere close to persuading the Court that it should interfere with the Agency’s judgment in these respects. Ground 2 fails.c. The Agency’s requirement that a pollution source must be known in order to qualify as an Abnormal Situation impermissibly narrows the broad definition of Abnormal Situation in the Bathing Water Directive and Regulations so as to close the door to unexpected pollution events with more complex causes.d. The Agency has misdirected itself in law in considering it has no discretion to take account of an Abnormal Situation in its assessment and classification of bathing water quality in circumstances where the pollution source is not known, and sampling was not therefore suspended. Its approach fails to recognise the distinct regulatory strands in the legal framework. The Agency has ample authority under the legal framework to treat sampling as having been suspended during an Abnormal Situation, when it assesses the reliability of relevant samples, where it is necessary to do so to arrive at a realistic classification for the bathing water in question. Ground 3 succeeds.

Conclusion

[111]For the reasons set out above, the Claim fails on Grounds 1 and 2 but succeeds on Ground 3.[112]Anglian Water seeks an order quashing the relevant bathing water quality data and/or an order requiring the Agency to exclude the relevant bathing water samples and to reclassify and republish the quality of those bathing waters accordingly. During the hearing Ms Lean explained the position on classification at Cleethorpes and Humberston Fitties would vary depending on the precise samples to be disregarded. Discounting samples at Ingoldmells South would lead to a non-compliant sample set and she queried whether an unclassified beach would be a worse outcome for Anglian Water than the present classification. In response, Mr Lewis expressed concern that the consequences of classification in one year affect classifications in subsequent years because of the rolling programme of assessment.[113]In light of these complexities, which Counsel addressed me on in general terms, without knowledge of the outcome of the claim, I consider it appropriate to direct the parties to seek to agree appropriate relief. The matter will be listed for a hearing on relief and other consequential matters, if necessary.