2018 Storm Overflow Assessment Framework
2018 Storm Overflow Assessment Framework
2018 Storm Overflow Assessment Framework
1
Environment Agency June 2018 Version 1.6
Stage 2: where hydraulic capacity is identified as the cause of trigger exceedance during the
first stage of the investigation, the level of environmental impact will be quantified;
Stage 3: improvement options are assessed including analysis of the costs and benefits;
Stage 4: a decision is made based on the cost benefit results;
Stage 5: delivery of the most cost beneficial solution (subject to appropriate funding and
prioritisation) to reduce environmental impact and/or reduce the frequency of discharges.
2
Decision framework for assessing and addressing
Environment Agency high frequency discharges
June 2018from storm overflows Version 1.6
under the UWWTR
STAGE Pollution/Failed
STAGE 11 High frequency
Environmental
spillers identified
objectives
Is the
Exceptional Yes intermittent Intermittent
Investigate Other Source e.g.
weather discharge also a overflow identified
Source diffuse pollution
high frequency as cause
spiller?
Follow other e.g. Environment
environmental Agency’s reason for
No driver processes failure database
Why is it a high
frequency spiller?
Hydraulic capacity
Asset
maintenance STAGE
STAGE 22
Rectify
STAGE
STAGE 33
Do Nothing.
Cost are
WaSC led STAGE disproportionate
STAGE 44 Decision
investigation when compared
to environmental
benefits
Figure 1. Assessment framework for addressing high frequency discharges from storm
overflows under the UWWTR.
3
Environment Agency June 2018 Version 1.6
4
Environment Agency June 2018 Version 1.6
d) Infiltration assessments
Where existing long or short-term flow data is available, such as MCERTified data at STWs
or flow data used to verify sewer models of the catchment, this data could be reviewed
to see if the catchment has a strong seasonal flow response due to groundwater or rainfall
induced infiltration. The EA’s water situation reports include an assessment of
groundwater levels that may assist with investigations should infiltration be suspected.
The reports categorise groundwater levels according to 7 classes from ‘exceptionally low’
through to ‘exceptionally high’, based on historic datasets from observation boreholes.
An example of a groundwater assessment for Yorkshire is shown in Appendix C. In the
case that infiltration is suspected or already known to be an issue in the catchment,
infiltration studies should be carried out. Tools, such as the infiltration risk tool developed
by the UKWIR Strategic Infiltration project, could be used to prioritise surveys to parts of
the catchment at most risk based on groundwater availability and pipe integrity (UKWIR,
2012).
5
Environment Agency June 2018 Version 1.6
Where there is the potential for the foul/combined system to interact with surface water
drainage upstream of the overflow, investigations might assess whether misconnected
surface water is the cause of high spill frequencies. For example, surface water sewers in
poor service or structural condition on dual manhole systems may cause surface water to
enter the foul/combined.
If maintenance issues are identified as the likely cause of frequent discharges, then these
problems should be rectified. Timescales for resolving the issue will vary according to the
problem, local constraints and the proposed resolution. However, with the exception of
infiltration, it is expected that most issues will be resolved within the calendar year that the
investigation is triggered. Where infiltration is the issue, it is expected that an infiltration
reduction plan (IRP) will be developed to address the problem. It is recognised that
investigation and resolution of infiltration issues can be difficult, that solutions may be
iterative, and that IRPs may only succeed over the medium to long-term.
Once the issue has been rectified, the annual EDM dataset(s) that triggered the investigation
should be archived and not included in future assessments against the triggers. Only the
annual dataset(s) affected by the maintenance issue should be excluded from future
assessments. For example, if it is known based on level trends that a partial blockage only
caused an issue during one calendar year, then only this year’s EDM data should be excluded.
Evidence of the maintenance issue will be required to exclude datasets. This may include
photographic evidence of poor service condition, such as from CCTV surveys or manual
inspections, level trend data from EDM indicating transient silt, and records of maintenance
carried out (e.g. sewer jetting and pump repairs).
In the case that asset inspection does not identify reasons that are likely to be responsible for
the high spill frequency, the following investigation of the hydraulic performance of the
overflow is required:
Stage 1c – Hydraulic assessment
If a verified hydraulic model of the overflow is already available, this should be used to assess
whether the high spill frequency is a genuine reflection of the permitted hydraulic design of
the asset, and the amount of connected area contributing rainfall runoff. Alongside asset
inspections carried out under stage 1b (above), models may have already been used to
determine that the high spill frequency is not due to maintenance issues.
Where a verified hydraulic model is not already available, a new model will be required to
predict the performance of the overflow. A verified model is also likely to be required in order
to quantify the environmental impact of the overflow under stage 2. In order to have
confidence in model predictions, models should be verified in accordance with the CIWEM
Urban Drainage Group Code of Practice for the Hydraulic Modelling of Urban Drainage
Systems (CIWEM UDG, 2017). The EDM datasets will assist with verification.
6
Environment Agency June 2018 Version 1.6
7
Environment Agency June 2018 Version 1.6
8
Environment Agency June 2018 Version 1.6
9
Environment Agency June 2018 Version 1.6
10
Environment Agency June 2018 Version 1.6
11
Environment Agency June 2018 Version 1.6
During WFD assessments prediction and classification of invertebrate quality is carried out for
each of the individual spring and autumn samples. A mean EQR is then calculated for the two
seasons. Overall classification is based on the worst status class assigned for the multi –
season mean WHPT NTAXA and WHPT ASPT. The RICT uses Monte Carlo processes to simulate
uncertainty in observed and expected EQRs due to factors such as sampling variation, error
in measuring environmental variables, and laboratory processing errors (bias). The software
typically uses 10,000 ‘shots’ to build up a distribution of potential EQRs in order to estimate
confidence of status class. To assess the impact of high frequency spillers, the RICT Compare
Module will be used to compare the quality of the upstream and downstream sampling sites.
The ‘Compare – At a Glance’ report will be used. This shows the percentage number of
simulations where the downstream sample is in a different status class to the upstream
sample for both WHPT NTAXA and ASPT. The scoring system in Tables 5a and 5b below will
be used for both indices (WHPT NTAXA & ASPT):
Table 5a. Invertebrate impact scoring for WHPT NTAXA & ASPT.
% of simulations the
downstream sample is one or
Score Class Multiplier
more classes worse than
upstream
1–4 1
5–9 2
10 – 29 4 × No. of classes the downstream
sample is worse than upstream
30 – 49 6
50 – 70 8
71 – 90 10
>90 12
12
Environment Agency June 2018 Version 1.6
Table 5b. Invertebrate impact classification for WHPT NTAXA & ASPT.
Total score Overall classification
1 No impact
2–3 Very low
4–5 Low
6–7 Moderate
8–9 High
10 – 11 Very high
12 – 15 Severe
16 – 19 Very severe
20 or more Extremely severe
The worst score for WHPT NTAXA and ASPT should be used to assign impact. The scoring
process will be repeated for each of the individual spring and autumn samples, and the overall
mean of the seasons in order to produce a short – term and long – term impact assessment
(Table 5c). A worked example is shown in Appendix F.
Table 5c. Overall short and long – term invertebrate impact classification
Where available, existing biological monitoring data for fish and invertebrates used for WFD
classification may be used to provide additional evidence that the overflow is not causing an
environmental impact. For example, where representative sampling points are present
downstream of the overflow, in close proximity, or in locations likely to be sensitive to
discharges from the overflow, and these consistently record good or high status, then this
may be used as evidence to support no impact classifications.
13
Environment Agency June 2018 Version 1.6
14
Environment Agency June 2018 Version 1.6
Level 2
This is similar to level 1. However, instead of a stochastic approach to representing upstream
river flow, a river flow time series is used. This allows the flow and therefore dilution available
in the river at the time of a spill to be better represented. As in level 1, simplified river
hydraulics and water quality are still used to predict the time of travel for pollutants along the
reach, and the depth and velocity of flow used to predict re-aeration rates.
Level 3
In level 3 studies calibrated flow routing models are used to more accurately predict time of
travel along longer and more complex water bodies. This allows better representation of
advective pollutant transport. More complex water quality simulation can be used with the
model calibrated for the key parameters – BOD, ammonia and dissolved oxygen – using
observed event sampling and water quality sonde data. Storm sewage quality is represented
using observed sampling data or calibrated sewer quality models.
Level 4
This is the most complex form of impact model. Calibrated hydrodynamic river models used
to simulate the varying depth and velocity of flow within the watercourse. Advection and
dispersion is calibrated against observed data (e.g. dye tracing). Various levels of water quality
simulation are possible with calibration and verification against event sampling and water
quality sonde data.
For all levels, a long (minimum 10 year) historic or synthetic rainfall time series representative
of the catchment is required.
New models are not required in all cases. Where they are ‘fit for purpose’, existing sewer and
river impact models from recent drainage planning or UPM studies should be used.
Impact scoring
The worst water quality score from the two types of assessment (99 percentiles and FIS)
should be used as follows:
1) 99 percentile quality
Two approaches are available depending on the type of modelling tool used:
A) Estimate of 99 percentile
Select the relevant 99 percentile BOD and total ammonia standards for the receiving water
according to WFD water body typology. These standards can be obtained from the third
edition of the UPM manual (FWR, 2012). As an example, Table 6 below shows the 99
percentile classes for water body types 3, 5 and 7. Where there is a drop in 99 percentile
status class between the modelled upstream and downstream assessment points assign a
score of 45.
15
Environment Agency June 2018 Version 1.6
Where the overflow does not cause a drop in status class but causes a degree of within class
deterioration, assign a score according to the percentage within class deterioration as shown
in Table 7 below. Use the worst score returned for the BOD and total ammonia assessments.
Table 7. 99th percentile within class deterioration scores.
Percentage within class deterioration Score
1 – 10 5
11 – 25 15
26 – 50 25
51 – 75 35
>75 45
B) Duration of exceedance
Where modelling tools are used which do not calculate a 99th percentile, but instead estimate
the duration for which a 99th percentile standard is exceeded, then use the following scoring
system in conjunction with the 99th percentile BOD and total ammonia standards for good
status (Table 8). The impact duration with the worst score should be used.
16
Environment Agency June 2018 Version 1.6
Use the following scoring system where the discharge causes a deterioration (increase) in the
frequency of allowable exceedances:
Table 10. Scoring system for increases in FIS exceedances for un-ionised ammonia and
dissolved oxygen.
The worst score obtained from the FIS and 99 percentile assessments should be used for the
water quality impact classification set out in Table 11 below.
Table 11. Water quality impact classification.
Water quality Score Water quality impact classification
0–5 No impact
6–9 Very low
10 – 19 Low
20 – 29 Moderate
30 – 39 High
40 or more Severe
17
Environment Agency June 2018 Version 1.6
ii) The overflow does not cause an environmental impact as assessed under SOAF
Stage 2, but the overflow is located within an agglomeration which has a PE of
2000 or more. The UWWTR require sewer networks for agglomerations with a
PE of 2000 or more to be designed, constructed and maintained according to
BTKNEEC. Consequently, where frequently spilling overflows in these drainage
areas do not cause environmental impacts, BTKNEEC still needs to be
considered through an assessment of the costs and benefits of reducing spill
frequency. The interpretation and definition of ‘agglomeration’ and
‘population equivalent’ is described in Appendix H.
The economic assessment of overflow improvement options involves an ecosystem services
approach, which identifies both the direct and indirect benefits of overflow improvement.
The ecosystem services considered include a range of environmental, social and economic
services, which have the potential to be impacted by storm discharges. A detailed
methodology and framework for carrying out the assessment is available in a separate report
and accompanying practitioner’s guide (Water UK, 2017). The aim of the process is to assess
the costs and benefits of improvement options to allow investment decisions to be made
under Stage 4 of the SOAF. The assessment framework involves six key steps. These are
summarised below and shown in Figure 3:
18
Environment Agency June 2018 Version 1.6
These include the timeframe for assessment and implementation of solutions, discount rates,
geographical scale, beneficiary groups to be considered, and approaches to uncertainty.
Figure 3. Summary of the overall framework for the assessment of costs and benefits (after
Water UK, 2017).
19
Environment Agency June 2018 Version 1.6
Step 3 – Screening
The options identified under step 2 are screened during step 3 to remove any options that
are unlikely to be cost – beneficial, and to ensure detailed CBA is focussed on the options
where the net benefits are greatest. A series of basic questions are used to establish whether
the benefits are a) likely to be significant (net present value >£100,000), and b) greater than
the likely costs. If the benefits are significant and potentially greater than the costs then these
options are considered for detailed assessment under step 4. Due to the inevitable
uncertainty associated with estimates of costs and benefits, options with a benefit cost ratio
(BCR) of 0.5 will also be taken forward for more detailed assessment under step 4.
benefits are explicitly incorporated within the analysis. Finally, due to the uncertainty
associated with CBA, sensitivity analysis is carried out on the results. This analysis will be
relatively simple, and involve looking at the effect of changes to key parameters such as
discount rates, assessment periods and cost estimates.
5. Stage 4 – Decision
Following the assessment of options and the cost – benefit analysis carried out during Stage
3, a final decision is made on whether to deliver an option to reduce the frequency of storm
discharges, or do nothing if no cost – beneficial solution can be found.
Delivery of the most cost beneficial solution is carried out under Stage 5 in order to reduce
environmental impact and / or the frequency of storm discharges. This will be subject to
appropriate funding and prioritisation.
21
Environment Agency June 2018 Version 1.6
22
Environment Agency June 2018 Version 1.6
References
Benyon, Richard. 2013. Letter to Water and Sewerage Company Chief Executive Officers, 18
July 2013.
CIWEM UDG, 2017. Code of Practice for the Hydraulic Modelling of Urban Drainage Systems.
Published by the Chartered Institution of Water & Environmental Management’s Urban
Drainage Group.
Dempsey, P. 2005. Default and sensitivity values for use in simplified UPM modelling studies.
Water Research Centre, Report Ref. UC6835.
DETR. 1997. The Urban Waste Water Treatment (England and Wales) Regulations 1994.
Working document for Dischargers and Legislators. A Guidance Note issued by the
Department of the Environment, Transport and the Regions and the Welsh Office.
Environment Agency. 2013. Risk based approach to the monitoring of storm discharges, 30
September 2013.
Environment Agency, 2014. Freshwater macroinvertebrate analysis of riverine samples:
Operational Instruction 024_08. Issued 28/01/2014 Environment Agency, Bristol.
Environment Agency, 2017. Freshwater macroinvertebrate sampling in rivers: Operational
Instruction 018_08. Issued 01/03/2017 Environment Agency, Bristol.
FWR, 1994. Report No. FR0466 – User guide for assessing the impacts of combined sewer
overflows. Foundation for Water Research, 1994.
FWR, 2012. Urban Pollution Management Manual – Third Edition. Foundation for Water
Research.
UKTAG, 2014. UKTAG River Assessment Method Benthic Invertebrate Fauna: Invertebrates
(General Degradation): Whalley, Hawkes, Paisley & Trigg (WHPT) Metric in River Invertebrate
Classification Tool (RICT). Water Framework Directive – United Kingdom Technical Advisory
Group (WFD – UKTAG), July 2014.
UKWIR. 2012. Strategic infiltration. United Kingdom Water Industry Research Limited. Report
Ref. 12/SW/01/1.
Urban Waste Water Treatment (England and Wales) Regulations 1994. SI 1994/2841.
Water UK. 2017. Valuing the benefits of storm discharge improvements for use in cost –
benefit analysis. Report Ref. No. NL5946 Final Report.
Water UK. 2017. Valuing the benefits of storm discharge improvements for use in cost –
benefit analysis: practitioners’ guide. Report Ref. No. NL5946 Practitioners’ Guide.
23
Environment Agency June 2018 Version 1.6
Appendices
Appendix A – Monitoring frequencies and spill counting method
Monitoring frequency will be set at 2 minute intervals for high significance discharges (EDM1),
and 15 minute intervals for medium significance discharges (EDM2). Spills will be counted
using the 12/24 counting method, which is defined as follows: one or more overflow events
within a period of 12 hours or less will be considered to be one spill, one or more overflow
events extending over a period of greater than 12 hours up to 36 hours will be considered to
be 2 spills. Each subsequent 24 hour duration counts as one additional spill and the whole of
the 24 hour block is included. Three examples are provided below to illustrate this method:
Example 1:
Counting starts when the first discharge occurs. Any discharge(s) within the first 12 hour block
are counted as one spill. In this example there is a single continuous discharge over the whole
of this 12 hour block. This is counted as one spill. After the first 12 hour period, any further
discharge(s) in the next 24 hours are counted as one additional spill. In this example the first
discharge lasts for 13 hours and so there is an hour of discharge within the 24 hour block
between 12 and 36 hours after the start of the first discharge. This is again counted as one
spill. Thereafter, any further discharge(s) in the next and subsequent 24 hour blocks are each
counted as one additional spill per block. In this example, there is one additional spill.
24
Environment Agency June 2018 Version 1.6
Counting continues until there is a 24 hour block with no discharge, after which the 12 and 24
hour block spill counting sequence starts again. In this example the total spill count is 3.
Example 2:
In this example, there are 3 separate periods of discharge within the first 12 hour block, each
lasting a few hours or so. These are counted as one spill. There are further intermittent spills
in the next 24 hour period between 12 and 36 hours after the start of the first discharge.
These discharges are again counted as one spill as they all fall within the same 24 hour block.
This 24 hour block is then followed by a 24 hour period during which no discharges occur. At
this point the 12/24 hour counting sequence starts again when the next discharge occurs. In
this example the total spill count is 2.
Example 3:
In example 3, there are intermittent periods of discharge within the 12 hour period following
the start of the first spill. These are all counted as a single spill. The 12 hour block is then
followed by a period of 24 hours during which no discharges occur. Consequently, the 12/24
counting process starts again at the time of the next discharge. In this example, the next
discharge involves two periods of spill which cross the 12 hour block and continue into the
next 24 hour block. Consequently, these are counted as 2 spills.
25
Environment Agency June 2018 Version 1.6
Appendix B – Example water situation report for rainfall (Yorkshire, September 2016)
26
Environment Agency June 2018 Version 1.6
Appendix C – Example water situation report for groundwater (Yorkshire, September 2016)
27
Environment Agency June 2018 Version 1.6
28
Environment Agency June 2018 Version 1.6
The monitoring points currently used to classify the water body for the second cycle river
basin management plans are a long distance from Manthorpe Mill CSO. The exception is the
physico – chemical monitoring point at Barkston Bridge (site ref. WITH5) which is
approximately 4.4km downstream of the overflow. Historically, an invertebrate sampling
29
Environment Agency June 2018 Version 1.6
point at Barkston Bridge was also used in first cycle classifications (site ref. 55422). Fish
monitoring has also been carried out in the past at the A607 Road Bridge (site ref. 5897), and
upstream of Syston Weir (site ref. 5867). These monitoring locations are shown in Figure 10.
The amenity value of the reach of the River Witham immediately downstream of the overflow
is high. The first 1.5km runs through the grounds and parkland of Belton House (National
Trust). There are footpaths along the river, fishing, and a children’s adventure playground.
The separate components of the investigation are set out below, including the modelled
hydraulic performance of the overflow, and the environmental components of the impact
assessment (aesthetics, biology and water quality). The impact scores are summarised at the
end.
30
Environment Agency June 2018 Version 1.6
31
Environment Agency June 2018 Version 1.6
1. Hydraulic performance
The verified hydraulic model compiled for the 2010 Grantham Drainage Area Plan (DAP) was
used to assess the hydraulic performance. The long – term performance of the high level
overflow was simulated using 10 years of synthetic rainfall events derived using StormPac.
Predicted spill frequencies and volumes for each year in the series are shown in Table 2 below,
along with the annual average predicted spill frequency and volume.
Table 2. Simulated spill frequency and volume.
Year Spill frequency (no/year) Spill volume (m3/year)
2020 20 47631
2021 19 71431
2022 23 37374
2023 20 42100
2024 23 72435
2025 17 30544
2026 27 80348
2027 19 47802
2028 26 57640
2029 24 86281
Average 22 57359
At the time of the invertebrate surveys (February 2014), the high level overflow had a pass
forward flow of 442l/s, and a bar screen with 6mm apertures. The overflow and its permit
have since been modified (September 2015) to include 900m3 of storage and screening to
6mm in two dimensions. This example, including the modelled performance and biology
surveys, relates to the performance of the old overflow prior to the improvement works of
2015.
32
Environment Agency June 2018 Version 1.6
2a
2b
Figure 3. Photo 2a – the Running Furrows Dyke downstream of Manthorpe Mill CSO. Photo
2b – sewage litter is visible stranded on overhanging branches.
33
Environment Agency June 2018 Version 1.6
Figure 4. Photo 3 – River Witham immediately downstream of the Running Furrows Dyke and
Manthorpe Mill CSO.
Figure 5. Photo 4 – River Witham immediately upstream of the Running Furrows Dyke and
Manthorpe Mill CSO. The garden of The Lodge (residential property) is on the left bank.
34
Environment Agency June 2018 Version 1.6
Figure 6. Photo 5 – River Witham downstream of Manthorpe Mill CSO in Belton Park.
Figure 7. Photo 6 – River Witham downstream of Manthorpe Mill CSO in Belton Park closer
to Belton House.
35
Environment Agency June 2018 Version 1.6
36
Environment Agency June 2018 Version 1.6
Figure 10. River Witham between Manthorpe Mill CSO and Barkston Bridge showing WFD monitoring points.
37
Environment Agency June 2018 Version 1.6
2. Aesthetics assessment
The results of the aesthetics assessment are summarised in Tables 3 and 4 below. Overall, the
overflow scored 50 giving a classification of moderate impact. Significant amounts of sewage
litter were present in the Running Furrows Dyke immediately downstream of the outfall (see
photos in Figure 3). Small amounts (3 pieces) of sewage litter were also found caught on
nettles along the banks of the River Witham within Belton Park. However, there was no
evidence of sewage fungus, either within the Running Furrows Dyke or the main river. There
are no recorded pollution incidents due to storm sewage on the National Incident Reporting
System (NIRS), but Anglian Water indicate there have been twelve customer complaints
recorded over the last 10 years (2006 – 2016).
Table 3. Aesthetics impact assessment.
Aesthetics Score
0 0
1 – 10 5
Sewage derived litter (no. of items) downstream 11 – 25 10
26 – 50 15
>50 20
N 0
Sewage fungus on outfall (present / absent)
Y 5
0 0
<2 5
2 – 10 10
Sewage fungus in downstream mixing zone (% cover)
11 – 25 15
26 – 50 20
>50 25
Amenity Score
Low or none amenity 0
Moderate amenity 5
High amenity 10
Public complaints Score
0 0
1–4 10
No. of validated public complaints related to wet weather discharges
5–9 20
from the overflow
10 – 14 30
15 or more 40
Incident
Pollution incidents due to storm sewage Score
category
Cat 3 20
No. of NIRS incidents due to storm sewage attributed to the
Cat 2 60
overflow
Cat 1 100
Total score 50
38
Environment Agency June 2018 Version 1.6
3. Invertebrate assessment
Invertebrate samples were collected from the River Witham immediately upstream and
downstream of the confluence with the Running Furrows Dyke, and the discharge from
Manthorpe Mill CSO. The sampling point locations are shown in Figure 1. The samples were
collected 24 February 2014 and the results of the surveys are summarised in Table 5. Although
the invertebrate data available does not allow for the full WFD assessment set out in the
methodology (SOAF Section 2b), it does provide an indication of impact.
Table 5. Whalley Hawkes Paisley Trigg (WHPT) scores for the number of invertebrate taxa
(NTAXA) and the average score per taxon (ASPT).
Site WHPT NTAXA WHPT ASPT
Upstream 19 4.34
Downstream 19 4.62
The number of WHPT taxa recorded upstream and downstream of the overflow were the
same (19), while the average sensitivity score for the taxa found downstream of the overflow
was slightly higher than at the upstream site. Consequently, it is not thought that the overflow
is impacting the invertebrate community and the overflow receives a score of zero (Table 6).
It was not possible to predict environmental quality ratios for the sites as environmental
variable data was not available for use with the river invertebrate classification tool (RICT).
Historic invertebrate sampling approximately 4.4km downstream of the overflow at Barkston
Bridge (site ID 55422), has tended to show either good or high status for macroinvertebrates.
White clawed crayfish are also known to be present in the Witham through Belton Park to
Barkston Bridge. Populations tend to occur in good quality waters high in dissolved oxygen
and low in organic pollution, and their presence suggests the overflow is not having a serious
impact on water quality.
39
Environment Agency June 2018 Version 1.6
40
Environment Agency June 2018 Version 1.6
Figure 11. Flow duration curve for the River Witham upstream of Manthorpe Mill.
Water quality statistics for the mixed river and spill are shown in Table 8. Dissolved oxygen
for the mixed water was set high at 8mg/l based on high upstream oxygen readings at
Saltersford Footbridge (range 8 – 14.7mg/l, average 11mg/l) and the presence of weirs and
sluices upstream of Manthorpe Mill likely to increase aeration. Spill pollutant concentrations
were assumed to be 125mg/l for BOD, and 8mg/l for ammonia based on default guidance
values (Dempsey, 2005).
A single reach analysis was carried out for the Witham between Manthorpe Mill CSO and the
sluices at Belton House (see Figure 10). The reach characteristics are summarised in Table 9
below, and are based on the Environment Agency’s ISIS model of the River Witham.
41
Environment Agency June 2018 Version 1.6
Table 9. Reach characteristics for simplified channel between Manthorpe Mill and Belton
House.
Reach parameter Value
Length (m) 1511
Slope (m/m) 0.00091
Width (m) 6.84
Side slope (m/m) 0.7
Manning’s N 0.070
The parameters used in the simplified water quality impact model are shown in Table 10
above. Default values were used for the re-aeration coefficients (Dempsey, 2005). A BOD
decay rate of 0.35/day and an ammonia decay rate of 2.0/day was selected as there are
sewage treatment works upstream (e.g. Little Ponton), and the river has lots of macrophytes.
The rate of ammonia generation through BOD decay was set at 0.3.
Additional reaches of the River Witham downstream of Belton House to Marston sewage
works were not considered for FIS impact analysis due to the presence of various weirs and
sluices.
The results of the impact assessment are summarised below for A) 99 percentiles and B) FIS.
A) 99 percentiles – duration of exceedance
The scoring system for increases in the duration of 99 percentile exceedance is shown below
in Table 11. Water quality impact classes depending on the score are summarised in Table 15.
The predicted increase in the duration of 99 percentile exceedance for BOD and total
ammonia caused by the overflow is shown in Table 12 along with the score. Based on the
increase in the number of exceedances for BOD, the overflow scores a severe impact on water
quality, and is predicted to narrowly fail the 99 percentile standard for the 24 hour impact
scenario. For total ammonia the impact is low.
42
Environment Agency June 2018 Version 1.6
Allowable
Impact duration exceedances Score
(no./year)
+ 0.5 points for every 1.0/yr increase in
1 hour 87.6
exceedances
+ 3.0 points for every 1.0/yr increase in
6 hours 14.6
exceedances
43
Environment Agency June 2018 Version 1.6
Table 13. Scoring system for increases in FIS exceedances for un-ionised ammonia and
dissolved oxygen.
Allowable
Frequency
exceedances (no. / Score
(return period)
year)
+ 1.5 point for every 0.5/yr increase in
1 month 12
exceedances
The reach is also predicted to narrowly fail the 6 hour 1 year standard for dissolved oxygen,
and the impact of the overflow is mainly predicted to be on BOD/dissolved oxygen levels.
44
Environment Agency June 2018 Version 1.6
Performance against the 3 month return period scenarios was not reported due to a problem
with the software.
The predicted impact on water quality conflicts with the invertebrate assessment, which
suggested there was no impact. The FIS and 99 percentile standards are intended for design
purposes, and are considered to allow for a margin of error. Consequently, although, the
standards are predicted to be failed, the standards are not heavily breached, especially for
ammonia, and the impact assessment may be pessimistic. Further sensitivity testing may
eliminate the predicted failure of standards given the simplification used. For example,
sampling of the overflow might justify using reduced BOD & total ammonia concentrations
for spill quality.
Table 15. Water quality impact classification.
Water quality Score Water quality impact classification
0–5 No impact
6–9 Very low
10 – 19 Low
20 – 29 Moderate
30 – 39 High
40 or more Severe
Summary
The classification for the aesthetic, biology, and invertebrate components of the assessment
are summarised in Table 16 below. Aesthetic impact was moderate, with the score driven by
public complaint, and the presence of sewage litter within the Running Furrows Dyke. For
biology, invertebrate sampling upstream and downstream of the discharge revealed no
impact. This was supported by good or high status being recorded at biological monitoring
points up to 4.4km way for both fish and invertebrates. The presence of white clawed crayfish
along the reach to Belton Park and downstream toward Barkston Bridge also suggests the
overflow is not having a significant impact. Due to the availability of good biology data which
indicates no impact, a water quality assessment would not normally be carried out. However,
in order to test the process a modelled water quality assessment was undertaken. In contrast
to the invertebrate assessment, water quality modelling predicted a severe impact due to
increases in the number of exceedances of the FIS to the extent that the annual 1 hour
standards for dissolved oxygen and un-ionised ammonia were failed.
Aesthetics Moderate
Invertebrates No impact
45
Environment Agency June 2018 Version 1.6
References
Dempsey, P. 2005. Default and sensitivity values for use in simplified UPM modelling studies.
Water Research Centre, Report Ref. UC6835.
46
Environment Agency June 2018 Version 1.6
47
Environment Agency June 2018 Version 1.6
Tables 4 and 5 show the results of the WHPT indices and environmental variables respectively
for the downstream sample site. The downstream sample point is approximately 100m
downstream of the upstream site, is narrower, deeper, and has a higher proportion of sand,
silt & clay. Lower values of WHPT NTAXA and ASPT were recorded for both the spring and
autumn samples compared to upstream. The average environmental quality ratios (EQRs)
simulated by the RICT for the downstream samples, along with their quality class and
confidence of class for the spring, autumn and combined seasons are shown in Table 6.
Table 4. WHPT NTAXA & ASPT results for downstream spring and autumn samples.
Season WHPT NTAXA WHPT ASPT
Spring 14 4.82
Autumn 13 4.77
Tables 7, 8 and 9 show the results of the comparison between the upstream and downstream
samples for the individual spring and autumn seasons, and for the overall combined spring &
48
Environment Agency June 2018 Version 1.6
autumn classification. The tables show the percentage number of the 10,000 simulations
where the downstream sample was in the same or a different class to the upstream sample.
For example, in Table 9 which shows the comparison for the overall classification, 47.55% of
the simulations for downstream WHPT ASPT were one status class worse (-1) than the
upstream site.
Table 7. RICT compare module – at a glance results for comparison of upstream and
downstream samples collected in spring.
Table 8. RICT compare module – at a glance results for comparison of upstream and
downstream samples collected in autumn.
Table 9. RICT compare module – at a glance results for comparison of upstream and
downstream following multi – season classification (spring & autumn).
% of simulations where the downstream sample is in the same or a different WFD
WHPT status class compared to upstream
Index
-5 -4 -3 -2 -1 Even +1 +2 +3 +4 +5
NTAXA 0 0 0 0 79.51 20 0 0 0 0 0
ASPT 0 0 0 0 47.55 52 0 0 0 0 0
The scoring method for estimating impact is summarised in Tables 10a – 10c. The method
involves a ‘worst of’ approach for WHPT NTAXA and WHPT ASPT, and is repeated for the
individual spring and autumn season samples, as well as the overall multi – season
classification in order to estimate both short – term (single season) as well as longer – term
(overall) impacts.
49
Environment Agency June 2018 Version 1.6
Table 10a. Invertebrate impact scoring for WHPT NTAXA & ASPT.
Table 10b. Invertebrate impact classification for WHPT NTAXA & ASPT.
Total score Overall classification
1 No impact
2–3 Very low
4–5 Low
6–7 Moderate
8–9 High
10 – 11 Very high
12 – 15 Severe
16 – 19 Very severe
20 or more Extremely severe
Table 10c. Overall short and long – term invertebrate impact classification.
Type Description Value
Worst single season classification result for
Short – term No impact – extremely severe
WHPT NTAXA and ASPT
Tables 11a – 11d summarise the results of the SOAF scoring assessment for this hypothetical
example. For the spring scoring assessment the worst result was for NTAXA – 64.7% of the
simulations gave downstream NTAXA values one WFD status class worse than upstream.
From table 10a this gives a score of 8 which is classified as ‘High’ impact (Table 10b). For the
autumn assessment, the worst result was seen again for NTAXA. In this case the percentage
of simulations where the downstream sample was one class worse than upstream was slightly
higher (87.13%). From Tables 10a and 10b this gives a score of 10 and an impact classification
50
Environment Agency June 2018 Version 1.6
of ‘Very high’. The overall multi – season (spring & autumn) WFD assessment also gave a ‘Very
high’ impact classification based on NTAXA, which was again worse than ASPT. Since the worst
single season result was ‘Very high’ impact for NTAXA in autumn, this gives a SOAF short –
term impact classification of ‘Very high’. Impact was also ‘Very high’ for the SOAF long – term
classification due to ‘Very high’ impact for NTAXA for the overall spring & autumn comparison.
One class worse than upstream Two classes worse than upstream
WHPT Overall
Impact
Index % × no. of Total % × no. of Total score
Score Score
sims classes score sims classes score
NTAXA 64.7 8 1 8 0 0 2 0 8 High
One class worse than upstream Two classes worse than upstream
WHPT Overall
Impact
Index % × no. of Total % × no. of Total score
Score Score
sims classes score sims classes score
One class worse than upstream Two classes worse than upstream
WHPT Overall
Impact
Index % × no. of Total % × no. of Total score
Score Score
sims classes score sims classes score
Table 11d. Short (single season) and long – term (spring & autumn) impact classification.
51
Environment Agency June 2018 Version 1.6
10 year historic
Event mean Predicted Statistical Statistical Simplified WQ 10 year
flow time series Simplified
Verified concentrations using flow time distribution distribution processes & re- representative
from EA gauging channel,
2 sewer default values (e.g. series from from sampled from EA aeration using historic or
station or steady &
model Dempsey, 2005) or verified effluent routine default values for synthetic time
calibrated rainfall uniform
sampled values sewer model quality samples rate coefficients series
runoff model
10 year historic Advective pollutant
Predicted Statistical Statistical 10 year
flow time series transport, WQ
Verified Sampled values or flow time distribution distribution Calibrated representative
from EA gauging simulation
3 sewer calibrated sewer series from from sampled from EA flow routing historic or
station or calibrated from
model quality model verified effluent routine model synthetic time
calibrated rainfall event sampling &
sewer model quality samples series
runoff model sonde data
Calibrated
10 year historic
Predicted Statistical Statistical advection – 10 year
flow time series
Verified Sampled values or flow time distribution distribution Calibrated dispersion model, representative
from EA gauging
4 sewer calibrated sewer series from from sampled from EA hydrodynamic WQ simulation historic or
station or
model quality model verified effluent routine model calibrated from synthetic time
calibrated rainfall
model quality samples event sampling & series
runoff model
sonde data
52
Environment Agency June 2018 Version 1.6
Appendix H
The interpretation and definition of ‘agglomeration’ and ‘population equivalent’ is set out in
the UWWT Regulations and supporting Guidance Note (DETR, 1997). Agglomeration is used
to describe an area where the combined population’s sewage is collected and treated. An
agglomeration population is the total population connected to the sewerage network
upstream of the STW. The sewerage catchment is not subdivided further into smaller
agglomeration populations upstream of individual storm overflows. The population
equivalent is a measurement of organic biodegradable load. A population equivalent of 1 (1
PE) is the organic biodegradable load having a five-day biochemical oxygen demand (BOD5)
of 60g of oxygen per day (the load shall be calculated on the basis of the maximum average
weekly load entering the treatment plant during the year, excluding unusual situations such
as those due to heavy rain).
53