Guidance

Site-specific quality numeric permit limits: discharges to surface water and groundwater

Updated 17 January 2019

The Environment Agency sets numeric discharge quality limits in permits to make sure discharges comply with water quality objectives.

These rules apply to water discharge activity and point source groundwater activity permits with numeric effluent composition conditions. There are additional requirements for how we assess compliance at waste water treatment works.

This guide mainly relates to samples collected under a routine pre-scheduled sampling programme. However non pre-scheduled samples, like samples collected by officers investigating a pollution incident, are also used to assess compliance against maximum limits.

When to take samples

The ‘sampling body’, which is either the operator or the Environment Agency, will take the samples. Your permit will provide details of any sampling you must carry out.

The sampling body should sample discharges evenly over a 12-month period in a regular but randomised programme. Regular and randomised means approximately equal intervals during the year and includes samples from different days of the week.

We only use samples that the sampling body has pre-scheduled for routine monitoring (or catch-up samples taken when the routine sample or a parameter has been missed) to check compliance with percentile or annual mean limits. A parameter is an individual characteristic of the water that can be tested.

If the sample is not taken, or there is no result for another reason, the sampling body must reschedule the sample or missing parameter.

If no discharge is occurring on the planned sample date, we normally treat that as a compliant sample.

If no discharge is occurring frequently at a discharge point, we may change the monitoring arrangements to avoid unnecessary visits.

If you’ve got operator self monitoring (OSM) permit conditions, you must follow the OSM guide for rescheduling and no-flow sampling events.

Unusual weather conditions

Unusual weather conditions include:

• low ambient temperatures, or the freezing of equipment in the works
• significant snow deposits preventing operation
• substantial tidal or river flooding of the site
• weather-related failures of the electricity distribution system causing loss of power that could not have reasonably been prevented by standby generation facilities

Heavy rainfall alone is not an unusual weather condition. Your treatment system should be designed to cope with this.

You must continue sampling during unusual weather, where possible.

If you cannot collect a pre-scheduled sample, for example because the monitoring point is flooded, you must record evidence of this and reschedule the sample.

If you think unusual weather conditions affected your treatment process at the time of a sample, provide the Environment Agency with evidence such as:

• photographs
• temperature records
• process monitoring records

The evidence must show that the treatment was operating correctly before the unusual weather. You should record all relevant details, including any mitigation measures, whenever unusual weather conditions could have adversely affected treatment.

To make a claim against a numeric limit failure or exceedance, you must submit your supporting evidence within 14 days of the date you were made aware of the failure or exceedance.

If we agree that the operation of the treatment process was adversely affected by unusual weather conditions outside of your control when a sample was taken, the sample is counted as compliant. This means the sample is included in the annual sample frequency, but is not considered in deciding if any emission limits are complied with even if the limits were exceeded.

You must take all practicable steps after an adverse operational effect to prevent or reduce the risk of it recurring. If you do not, we may not accept that a recurrence is outside of your control.

Compliance limits

When assessing compliance with a limit the Environment Agency does not make any allowance for the uncertainty of chemical analysis in the results of individual samples. We always take and use the sample results, including ‘less than’ and ‘greater than’ results. The compliance assessment methods used on multiple sample results allow for this uncertainty, and by considering the samples as a group the impact tends to be cancelled out.

We use 4 types of compliance limit:

• percentile
• mean
• maximum
• differential

Percentile compliance limit

There are concentration limits for some parameters that the discharge must be under at least 95% of the time. This is the 95-percentile limit.

To prevent short-term pollution during the remaining 5% of the time, the Environment Agency also normally sets a higher maximum concentration limit for that parameter. This maximum compliance limit must not be exceeded.

Percentile limit use

We normally only use the percentile limit for routine sampling of water company treated sewage discharges. It’s used for limits on these sanitary parameters:

• biochemical oxygen demand (BOD)
• suspended solids
• ammoniacal nitrogen
• colour
• chemical oxygen demand

It may also be used for limits of hazardous pollutants in water discharge activity permits.

Percentile limit compliance assessment

If a parameter sample result is higher than its 95-percentile numeric limit it’s an individual look-up table (LUT) exceedance.

When a LUT exceedance happens, we compare the number of exceedances for that parameter with the number of samples taken in the 12-month period. The 12-month period ends on the last day of the month in which the latest exceedance occurred.

We use the LUT to see how many exceedances are allowed for the number of samples taken in the 12-month period. If the number of exceedances is greater than the maximum number allowed, it’s a LUT failure. The failure is recorded as happening on the date of the most recent exceedance.

Look-up table

Number of samples taken in any period of 12 months	Maximum number of samples permitted to exceed for given parameter
4 - 7	1
8 - 16	2
17 - 28	3
29 - 40	4
41 - 53	5
54 - 67	6
68 - 81	7
82 - 95	8
96 - 110	9
111 - 125	10
126 - 140	11
141 - 155	12
156 - 171	13
172 - 187	14
188 - 203	15
204 - 219	16
220 - 235	17
236 - 251	18
252 - 268	19
269 - 284	20
285 - 300	21
301 - 317	22
318 - 334	23
335 - 350	24
351 - 365	25

Mean compliance limit

The mean compliance limit is used to regulate parameters with low acute toxicity. It aims to limit the overall load of the substance discharged to the environment.

This limit is set as an annual (12 months) mean.

Mean limit use

The Environment Agency only uses the mean limit for discharges that are routinely sampled several times each year for effluent compliance. It’s used for limits on plant nutrients – that is compounds of phosphorus and nitrogen. It may be used on other parameters if appropriate.

Mean limit compliance assessment

We assess compliance with mean limits using the concentrations of the pre-scheduled samples taken in any period of 12 consecutive months unless a fixed 12-month period is specified by your permit.

The mean is the sum of the concentrations of the pre-scheduled samples taken in the 12-month period divided by the number of samples.

Any results recorded as ‘greater than’ values are included in the calculation as the recorded numeric value. Any results recorded as ‘less than’ values are included in the calculation as zero. This means ‘less than’ values are not included in the sum of the concentrations, but are included in the total number of samples in the divisor.

The mean calculation excludes samples:

• where there is no result because there was no discharge on a planned routine sampling occasion
• affected by unusual weather conditions

We calculate the mean and standard deviation of the annual data and then calculate the 90% confidence interval around the mean.

The formula is: lower confidence interval = mean – (t x standard error of mean).

Where:

• t is derived from the values of t table for n-1 degrees of freedom, where n is the number of samples
• the standard error of the mean is the standard deviation of the dataset ÷ √n.

If the lower (optimistic) confidence interval exceeds the permit limit, then we are 95% confident that the limit has been exceeded.

In this case, a failure is recorded as happening on the last date of the 12-month period.

You only have to calculate the lower confidence interval if the face-value of the mean exceeds the permit limit for the mean.

In some circumstances, an unusually high result may lead to a wide confidence interval because the approach assumes the data has a normal distribution (bell-shaped). This means that with the unusually high result included, the test does not show that the annual mean limit was breached with more than 95% confidence, but with the high result excluded, it does. If this happens you should contact the Environment Agency for advice.

Mean limit compliance assessment for phosphorus limits

For phosphorus limits we use a log-normal distribution to calculate the relevant summary statistics and to assess compliance.

The concentrations used in this assessment come from the pre-scheduled samples taken in any period of 12 consecutive months unless a fixed 12-month period is specified by your permit.

1. Take the log10 value of all P concentrations in the data set. For a zero value, take the log10 value of the minimum detection limit of the analytical method used. Treat less thans or greater thans as face value.
2. Calculate the mean of the log10 values.
3. Calculate the standard deviation of the log10 values.
4. Calculate the lower confidence interval using the formula lower confidence interval = mean – (t x standard error of mean).
5. Antilog the lower confidence interval.

If the lower confidence interval is less than or equal to the mean standard in the permit it passes.

We consider the statistical confidence of failure and the magnitude of the failure when we decide what enforcement action to take.

When we assess compliance with Urban Waste Water Treatment Regulations phosphorus limits we do not include the use of confidence intervals. Read the guide for compliance limits at waste water treatment works.

Values of t

Degrees of freedom	t for 90% confidence interval
2	2.920
3	2.353
4	2.132
5	2.015
6	1.943
7	1.895
8	1.860
9	1.833
10	1.813
11	1.796
12	1.782
13	1.771
14	1.761
15	1.753
16	1.746
17	1.740
18	1.734
19	1.729
20	1.725
25	1.708
30	1.697
35	1.690
40	1.684
50	1.676
60	1.671
70	1.667
80	1.664
90	1.662
100	1.660
1,000	1.646

Maximum compliance limit

A maximum limit is a concentration that no sample result must exceed.

Maximum limits are often called upper tier or absolute limits.

Maximum limit use

The Environment Agency often uses the maximum limit with the percentile limit. This is often called an upper tier limit.

We also use this limit for discharges that are not regularly sampled. For these discharges, it can demonstrate non-compliance with a single sample if the treatment of the discharge malfunctions and causes environmental pollution.

Maximum limit compliance assessment

The sample result is compared with the maximum limit. If the sample result exceeds the limit, then it’s recorded as a failure.

Any discharge results recorded as ‘greater than’ values are included in the assessment as the recorded numeric value. Any results recorded as ‘less than’ values are included in the assessment as zero.

Limits for pH are expressed as maximum and minimum limit values. Sample results above the higher pH value or below the lower pH value are recorded as a failure.

All sample results, including non-routine samples such as those we take during pollution incident investigations, are assessed against this limit.

We will consider the significance of the failure when deciding what enforcement to take. For this assessment we may consider:

• analytical uncertainty
• the previous compliance record to see if the fail is a one-off or part of a pattern of failures
• if the proportion of failing samples is increasing
• if the discharge quality is significantly deteriorating
• how much in excess of the limit are the failing results
• what impact the failure had on the receiving water
• the sensitivity of the receiving environment
• if the operator has taken action to mitigate the problem and prevent a recurrence

Differential compliance limit

A differential limit sets a limit on the increase that the operator may cause between the abstraction inlet and the discharge point.

It lets the Environment Agency distinguish between increases caused by the operator and natural environmental variations.

Differential limit use

We use the differential limit for operators abstracting water from groundwater or surface water for use before discharging it. The abstracted water must contain the parameters at concentrations similar to the required discharge limit.

It sets a limit on the increase the operator may cause between the inlet and the discharge point.

This limit is used for:

• temperature in cooling and heating water discharges
• BOD, suspended solids, ammonia and phosphorus compounds in aquaculture discharges
• phosphorus compounds in watercress farm discharges

It may be used on other parameters if appropriate.

Differential limit compliance assessment

Samples are taken at the inlet and the discharge sampling points on each sampling occasion.

For each relevant parameter, the result of the inlet sample is subtracted from the result of the discharge sample to calculate the difference.

If either sample is not taken, or a result for a parameter cannot be reported for any reason, the sampling body must reschedule both samples fully or for the missing parameter.

For inlet samples, ‘less than’ values are recorded at face value. For outlet samples, ‘less than’ results are recorded as zero.

Because in most cases there is no single ‘time of travel’ through a site, it’s impossible to sample the same water entering and leaving. The permit will specify that the 2 samples are collected as close to the same time as possible.

The differential limit may be set as a maximum limit or an annual mean limit. In this case, we will assess compliance using the dataset of differences and will apply the relevant assessment method.

Contact the Environment Agency

General enquiries

National Customer Contact Centre
PO Box 544
Rotherham
S60 1BY

Email enquiries@environment-agency.gov.uk

Telephone 03708 506 506

Telephone from outside the UK (Monday to Friday, 8am to 6pm GMT) +44 (0) 114 282 5312

Monday to Friday, 8am to 6pm.