Yorkshire Water | Water Resources Management Plan 2014 |
Water Resources Management Plan Yorkshire Water Services Ltd August 2014
1
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
2
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Contents Contents ......................................................................................................................... 3 List of figures .................................................................................................................. 4 List of tables.................................................................................................................... 5 4
Demand Forecast ..................................................................................................... 7 4.1
Introduction ....................................................................................................... 7
4.2
The base year .................................................................................................... 9
4.3
Accounting for demand in the base year ...................................................... 11
4.3.1 Water delivered ........................................................................................... 11 4.3.2 Distribution losses ....................................................................................... 22 4.3.3 Distribution system operational use ............................................................ 23 4.3.4 Water resource zones split.......................................................................... 23 4.4
Background to forecast changes in demand to 2039/2040 ......................... 24
4.4.1 Water delivered ........................................................................................... 24 4.4.2 Distribution losses ....................................................................................... 47 4.4.3 Distribution system operational use ............................................................ 47 4.5
Forecast demand to 2039/40 .......................................................................... 48
4.5.1 Water delivered ........................................................................................... 48 4.5.2 Distribution Input ......................................................................................... 52 4.6
Weighted average demand............................................................................. 54
3
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
List of figures Figure 4.1 Summary of demand forecasting methodology ..........................................................7 Figure 4.2 Components of distibution input ...............................................................................11 Figure 4.3 Components of total water delivered ........................................................................12 Figure 4.4 Components of total leakage ...................................................................................23 Figure 4.5 Methodology for forecasting measured household water delivered ..........................25 Figure 4.6 Methodology for forecasting unmeasured household water delivered ......................26 Figure 4.7 Methodology for forecasting measured non-household water delivered ...................27 Figure 4.8 Methodology for forecasting unmeasured non-household water delivered ...............27 Figure 4.9 Domestic meter optants forecast ..............................................................................29 Figure 4.10: New build households forecast ..............................................................................32 Figure 4.11: Household property forecast .................................................................................32 Figure 4.12: New commercial connections forecast ..................................................................33 Figure 4.13: Non-household property forecast ..........................................................................34 Figure 4.14: Total population forecast .......................................................................................35 Figure 4.15: Measured household properties and population forecasting methodology ............36 Figure 4.16: Measured household properties and population forecasting methodology ............37 Figure 4.17: Methodology for forecasting measured household volume ....................................39 Figure 4.18: Methodology for forecasting unmeasured household volume ................................39 Figure 4.19: Percentage breakdown of measured household PCC ...........................................42 Figure 4.20: Percentage breakdown of unmeasured household PCC .......................................42
4
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.21: Impact of climate change on household demand...................................................45 Figure 4.22: Water delivered to households ..............................................................................48 Figure 4.23: Water delivered to measured non-households ......................................................49 Figure 4.24: Water delivered to unmeasured non-households ..................................................50 Figure 4.25: Total water delivered .............................................................................................51 Figure 4.26: Total distribution input ...........................................................................................53 Figure 4.27: Comparison of dry year annual average and critical period (peak week) scenarios for the East SWZ.......................................................................................................................54 Figure 4.28: Weighted average demand ...................................................................................56
List of tables Table 4.1: Dry year annual average distribution input .................................................................9 Table 4.2: Regional maximum likelihood estimation table .........................................................10 Table 4.3: Occupancy rates for the different property categories in the base year ....................14 Table 4.4: Comparison of ‘dry year’ and ‘normal year’ demand .................................................18 Table 4.5: Comparison of 2011/12 demand and average ‘normal year’ demand .......................18 Table 4.6: 2011/12 demand for non household SIC categories impacted by dry year ...............19 Table 4.7: 2011/12 total demand for households and non-households impacted by dry year effect .........................................................................................................................................19 Table 4.8: Summary of dry year uplift ........................................................................................20 Table 4.9: Measured and unmeasured household dry year uplift ..............................................21 Table 4.10: Cumulative water efficiency changes for AMP5 and AMP6 ....................................30
5
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Table 4.11: Summary of customer demand management built into the demand forecast ..........31 Table 4.12: Summary of population forecast by AMP period .....................................................36 Table 4.13: Breakdown of Grid SWZ PCC by microcomponent (dry year annual average scenario) ...................................................................................................................................41 Table 4.14: Breakdown of East SWZ PCC by microcomponent (dry year annual average scenario) ...................................................................................................................................41 Table 4.15: Water delivered to households ...............................................................................49 Table 4.16: Water delivered to measured non-households .......................................................50 Table 4.17: Water delivered to unmeasured non-households....................................................51 Table 4.18: Total water delivered ..............................................................................................52 Table 4.19: Summary of dry/normal years.................................................................................55 Table 4.20: Weighted average demand ....................................................................................55
6
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
4 Demand Forecast 1.
A 25 year demand forecast has been produced to cover the period 2011/12 to 2039/40. This forecast is based on assumptions about how the key factors influencing water demand will change over the plan period.
4.1
Introduction
2.
The demand forecast is produced for our two water resource zones, the Grid SWZ and the East SWZ, individually and also the Yorkshire Water region as a whole.
3.
A conservative approach is taken to forecasting demand (dry year annual average), which includes assumptions of climate change and dry year uplift as standard. In addition, critical period (peak week) analysis has been performed for the East SWZ as it is considered to be potentially more vulnerable to short term peaks in demand due to its reliance on a single source of supply.
4.
The demand forecast has been prepared in line with the best practice methodology set out in the Demand Forecasting Methodology (UKWIR/NRA, 1995) and the Water resources planning guideline (Environment Agency et al., 2013).
5.
The methodology can be simplified to the process summarised in Figure 4.1.
Figure 4.1 Summary of demand forecasting methodology
6.
The methodology includes the forecast of water use by our four customer groups, which are defined by property type. These categories are defined in Section 4.3.1 and comprise:
7.
Unmeasured households Measured households Unmeasured non-households Measured non-households.
Each of the property categories have their own set of demand drivers and assumptions for future growth rates. These include population projections, households switching to
7
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
paying by meter (domestic meter optants), new connections and the economic environment. 8.
In addition, there are also minor components of demand, which include: void properties, operational use and water taken unbilled.
9.
There are various terms used in relation to demand for water, which include different components. The main terms used in this report and the Environment Agency’s planning tables are defined as follows:
10.
11.
Consumption - the water used by a property. It includes the volume of water used and meter under registration, but excludes supply pipe leakage.
Water Delivered - comprises the volume of water supplied from treatment works, less the volume the water company uses (distribution system operational use), or is lost through the company’s pipes (leakage).
Distribution Input - the average amount of drinking water entering the distribution system to be supplied to consumers in an appointed water company’s area of supply. This is essentially total demand for water as it includes consumption, leakage, water taken unbilled and distribution system operational use.
The following key data sources and assumptions have been included in the forecast:
Yorkshire Water historical operational data
Policy based population forecast
The effect of climate change on demand
Household projections incorporating data from local planning strategies from Local Authorities
Assumed demand in new properties is based on the Code for Sustainable Homes (Communities and Local Government, 2007)
Micro-component based household demand forecasts.
The change in total demand (distribution input) for water throughout the planning period is shown in the dry year annual average scenario in Table 4.1.
8
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Distribution Input Ml/d Grid SWZ
9
2014/15 2019/20 2024/25 2029/30 2034/35 2039/40 (end AMP5) (end AMP6) (end AMP7) (end AMP8) (end AMP9) (end AMP10) 1,288.92
1,274.46
1,277.82
1,282.49
1,285.38
1,291.24
6.98
6.90
6.95
7.00
7.05
7.11
1,295.90
1,281.36
1,284.77
1,289.50
1,292.43
1,298.35
East SWZ
Total
Table 4.1: Dry year annual average distribution input
4.2
The base year
12.
The base year has been defined for the WRMP as 2011/12. Outturn data for 2012/13 and 2013/14 is now available, however, the Water resources planning guideline, (Environment Agency et al., 2013), advises that the base year does not need to be updated unless there is a very good reason to do so. In the absence of such a reason, we have decided to keep the base year as 2011/12.
13.
Data for the base year is taken from the annual report to Ofwat 2012, which covers the year April 2011 to March 2012.
14.
Within the annual report to Ofwat data, the reported distribution input exceeds the water that can be accounted for. Therefore there is an adjustment for this surplus water. The maximum likelihood estimation (MLE) technique has been used to allocate this discrepancy across all components. This is done at a regional level.
15.
The demand forecast uses an adjusted MLE table which varies slightly from the one presented in the annual report to Ofwat. These variations result from amended assumptions, such as meter under registration (MUR) rates, which are explained where appropriate in this report. This affects the water accounted for and therefore the amount of surplus water that needs to be accounted for in the MLE adjustment. The amended assumptions will be incorporated into future annual reporting to Ofwat data.
16.
The annual report to Ofwat data was then rebased for the dry year annual average scenario as detailed in Section 4.3 below.
17.
The adjusted MLE table is presented in Table 4.2. The right hand column (post MLE) is the base year data used as the basis for the demand forecast.
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
2011/12 Measured Households Consumption SPL internally metered SPL externally metered Meter Underregistration
Base Ml/d 209.08 188.31 11.356 5.539 3.88
Accuracy Confidence + or Range
10
% of Total Variance
Adjustment
Final Figures Rounded
2% 5% 5% 50%
7.53 1.14 0.55 3.88
2.98% 0.45% 0.22% 1.53%
0.99 0.15 0.07 0.51
189.30 11.50 5.61 4.38
Post MLE Ml/d 210.79
Measured Non-Households Consumption SPL Meter Underregistration
269.76 253.25 4.63 11.89
2% 5% 50%
10.13 0.46 11.89
4.01% 0.18% 4.70%
1.33 0.06 1.56
254.57 4.69 13.44
Unmeasured Households Consumption SPL Meter Underregistration
499.41 442.91 47.641 8.86
10% 5% 50%
88.58 4.76 8.86
35.06% 1.89% 3.51%
11.60 0.62 1.16
454.51 48.26 10.02
Unmeasured Non-Household Consumption SPL
2.44 1.81 0.629
25% 5%
0.90 0.06
0.36% 0.02%
0.12 0.01
1.93 0.64
Water taken illegally
16.09
50%
16.09
6.37%
2.11
18.20
18.20
Water Taken legally Void SPL
14.60 5.51
50% 25%
14.60 2.75
5.78% 1.09%
1.91 0.36
16.51 5.87
22.38
2.72
50%
2.72
1.08%
0.36
3.07
3.07
1,016.89
1,039.43
168.42 29.53
168.42 29.53
1,240.46
1,240.46 1,240.45 -
Distribution Operational Use Total Water Delivered (excl DOU) Distribution Losses Losses in Distribution Management Areas Trunk Main and Service Reservoir Losses Distribution Input Water accounted for Difference as % of DI Total Supply Pipe Leakage Total Leakage
272.70
512.79
2.57
1,016.89
166.24 28.06
5% 20%
16.62 11.22
6.58% 4.44%
1,246.99 1,213.91 33.08 0.03
2%
49.88
19.74%
75.300 269.60
22%
Table 4.2: Regional maximum likelihood estimation table
2.18 1.47 (6.53)
4.92
76.57 274.52
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
4.3
Accounting for demand in the base year
18.
The base year demand for the dry year annual average demand forecast is based on the annual reporting to Ofwat data, with the MLE adjustment (set out in Section 4.2) and an uplift for dry year (to be discussed in Section 4.3.1).
19.
The total demand for water is termed distribution input and is made up of several elements of demand as shown in Figure 4.2.
Figure 4.2 Components of distibution input
20.
4.3.1
In this section of the report, the base year distribution input is split into its component parts, which then allows us to forecast future changes to each of these components and therefore future demand. Water delivered
21.
As stated earlier, water delivered is defined as the volume of water supplied from treatment works, less the volume we use for operational purposes (distribution system operational use), or is lost through our pipes (distribution losses).
22.
Total water delivered comprises the water delivered to each property category and water taken unbilled, as illustrated in Figure 4.3.
11
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.3 Components of total water delivered
Properties, population and occupancy rates Properties 23.
The amount of water delivered to each of the six property categories, laid out in Figure 4.3, depends on the number of properties in the category and, in the case of household properties, the population associated with those properties.
24.
The number of properties in each customer category is extracted from our billing file. This is provided by our Tariff Team, and is the data used for the annual report to Ofwat.
25.
The occupied properties are divided into four categories:
12
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
26.
Measured households - domestic properties with a meter (meter can be internal or external). This category includes existing measured households and new build households, both of which pay for water on a metered supply; and domestic meter optants (DMOs), which are those customers opting to have a water meter installed.
Unmeasured households - domestic properties without a meter that pay for water based on the rateable value of the property.
Measured non-households - commercial properties with a meter. This category includes new commercial connections and takes into account demolitions.
Unmeasured non-households - commercial properties without a meter. These tend to be small businesses/properties, e.g. family shop with flat above (mixed use properties), scout hut, animal troughs etc.
The other two property categories, household and non-household voids, are properties that are registered as empty on our billing files. Although they have no consumption, they do have some leakage from the supply pipes which connect the properties to the mains. Population and occupancy rates
27.
The population for the base year has been taken from the mid-year estimate (MYE) 2011 population data published by the Office of National Statistics (ONS).
28.
The ONS data is only available by Local Authority level. The population was split between the two resource zones by Edge Analytics based on digitised maps of the resource zone areas. Trend based population estimates were provided for the two resource zones, which factor in forecasts of birth, death, immigration etc.
29.
The measured non-household population is taken as the communal population (prisons, nursing homes, university halls of residence etc.) from the ONS data. We have excluded the population of a military garrison in North Yorkshire which has its own water supply. The measured non-household population is subtracted from the total population.
30.
The remaining base year population is then allocated to the other three property categories. This was done using average occupancy rates for measured and unmeasured households and unmeasured non-households.
31.
We have worked with Edge Analytics to align populations, properties and occupancy rates in the base year (and forecast years). We have used occupancy data obtained from a customer survey and from our Domestic Consumption Monitor (DCM). For new builds (households) we have used information from the WRc (2012) project Water Consumption of Homes Built to Part G and Code for Sustainable Homes Standards.
32.
Occupancy rates are only relevant to those properties which have domestic residents. Therefore they were only derived for the measured households, unmeasured households, and mixed use properties within the unmeasured non-households category. The latter are
13
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
14
primarily domestic use, e.g. small shops with flats over, and their occupancy rate is assumed to be the same as for unmeasured households. 33.
The occupancy rates used in the base year are detailed in Table 4.3 below. The same occupancy rates have been assumed for the Grid SWZ and East SWZ, though they vary between property types.
Measured Households
Unmeasured Households
Unmeasured NonHouseholds (mixed use)
Grid SWZ
2.18
2.64
2.64
East SWZ
2.18
2.64
2.64
Table 4.3: Occupancy rates for the different property categories in the base year
34.
The calculation of the measured households occupancy rate is complicated as there are three separate occupancy rates within this category. These were for existing measured households, new builds and DMOs.
35.
In addition to the population recorded in the census, there is a population known as ‘hidden and transient’, which would not show up on the census. This category includes illegal immigrants and second home owners not captured in the measured households; and tourists not captured in the measured non-households category. This population is ‘hidden’ within the unmeasured household category, and not included in the census data.
36.
The majority of visitors and tourists to the Yorkshire Water supply area will be captured through water meters at tourist sites, in hotels and in most other types of commercial accommodation, i.e. measured non-households. However, there will be a small proportion of tourists not captured in this category, including those staying with friends and family; in small bed and breakfasts where water meters are not in place; or day visitors to family and friends.
37.
A study by Leeds University provided a range (low, medium and high) of estimates for the hidden and transient population in the Yorkshire Water supply area. We used this information in stochastic modelling to create a normal distribution. This resulted in a hidden and transient population estimate of 71,296.
38.
As the hidden and transient population is not included in the census data, and therefore the population forecast, they have been added into the demand forecast as a set volume of consumption (based on this population estimate), rather than as a population with an associated per capita consumption (PCC).
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Consumption 39.
Consumption is defined as the water used by a property. This includes the volume of water use and meter under registration, but excludes supply pipe leakage. Household volume
40.
The volume of water consumed by household properties is calculated from a per capita consumption (PCC) figure, and therefore depends on population.
41.
PCC for measured households in the base year was derived by dividing the known water used (consumption minus MUR), obtained from our billing file, by the population assigned to this category. This gave an average PCC of 105.87l/h/d.
42.
The unmeasured household PCC is derived from our DCM. This gave a base year PCC value of 145.14l/h/d.
43.
The DCM measures average unmeasured per capita consumption and average unmeasured household size, through analysis of a representative sample of unmeasured properties. The DCM consists of properties which have logged meters installed but which continue to pay on an unmeasured basis. The DCM is used to determine PCC at a regional and resource zone level.
44.
The loss of properties from the DCM in recent years has resulted in the survey containing less than required 1000 properties. In 2012 we carried out a recruitment campaign to increase the number of properties on the DCM, resulting in an additional 180 properties. In 2013 we added a further 120 properties to the survey and will continue to recruit additional properties to improve the sample size.
45.
We have completed a logger replacement project in which all manual loggers and cello loggers have been replaced with telemetered halma loggers. Consumption data from all DCM properties is now downloaded daily on to our reporting system Netbase.
46.
We include an element of sample bias in the PCC estimation, to account for the frequency of contact from Yorkshire Water and a deficit of high water users in the survey sample. Following a review by Leeds University, this bias was estimated to be 4%. This bias is added to the PCC estimated from the DCM. Hidden and transient volume
47.
The unmeasured household volume of water used also includes the water used by the hidden and transient population. As highlighted above, the amount of water used by this population is added into the demand forecast as a volume. For the base year this volume was estimated to be 8.8Ml/d, based on assumed PCC and occupancy rates for the different categories of hidden and transient population.
15
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Non-household volume 48.
The volume of water consumed by non-household properties is taken from the annual reporting to Ofwat data, with our MLE adjustments factored in (Table 4.2). Voids
49.
By definition, both household and non-household voids have no population associated with them and therefore they do not have any consumption. The only water delivered to these properties is therefore supply pipe leakage, which is not part of consumption. Metering
50.
The split between metered and unmetered properties for the base year was obtained from our billing file.
51.
Two adjustments are made to the water use recorded by meters: MUR and SPL. SPL is excluded from consumption as discussed later. Meter under registration
52.
Meter under registration (MUR) is the volume of water that is not recorded by the water meters due to an error in recording as meter age and wear. This applies to measured and unmeasured households and measured non-households.
53.
A MUR rate does not apply to unmeasured non-households.
54.
The MUR rates were calculated from annual report to Ofwat data, which gives a volume of MUR. The calculated rates are the MUR volume expressed as a percentage of consumption:
Measured households
2.0%
Unmeasured households
2.0%
Measured non-households
4.7%
55.
Previously in the base year the measured household MUR was 3.7% and unmeasured households 1.1%. These rates were based on testing of a sample of meters. For the WRMP, MUR has been adjusted to reflect aging DCM meters and the ongoing meter replacement policy for measured households. The MUR rates for both these categories have therefore been adjusted to 2%.
56.
The 4.7% MUR applied to measured non-households is the industry average as reported in the annual reporting to Ofwat. This figure is considered to be appropriate for the Grid SWZ as there are large commercial users with large meters. However, the East SWZ is a small zone with tourism accounting for the majority of measured non-households, e.g.
16
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
bed and breakfasts and small hotels. These supplies would be much smaller and therefore a lower MUR rate of 2.7%, closer to that of households, is assumed to be more appropriate for measured household properties in the East SWZ. Supply pipe leakage 57.
Supply pipe leakage (SPL) is defined as leakage from pipes location within property boundaries, i.e. between our main and the customers’ taps.
58.
Within the measured households category, the meters can be positioned either internal or external to the property. The position of the meter is recorded in our billing file.
59.
The location of the meter makes a difference in terms of SPL volume. If the meter is internal to the property then the volume of SPL is not recorded on the meter. However, if the meter is external to the property then the SPL is captured as part of the total volume recorded by the meter. It is assumed that a burst would be identified and repaired more quickly in an externally metered property, as it would show up as an increase in water use on customer bills. Therefore the amount of SPL added to externally metered properties in the demand forecast is lower that for the internally metered properties.
60.
The base year SPL rates are taken from the annual report to Ofwat. Water taken unbilled
61.
Water taken unbilled includes water taken legally and illegally that is not paid for by the customer. It is added into the demand forecast as a separate item rather than being assigned to a property category. As this water use is unbilled the volume has been estimated using best available information.
62.
Water taken legally includes water used for fire-fighting and training, standpipe use and water used at our own sites (which is not recorded on the billing file).
63.
Water taken illegally includes occupied voids, illegal hydrant use and illegal connections.
64.
The amount of water taken unbilled, legal and illegal, is required as part of the annual report to Ofwat. Our adjusted MLE values (Table 4.2) have been used for the base year volume. Dry year effect
65.
In accordance with the Water resources planning guideline (Environment Agency et al., 2013), the demand forecast has been prepared for a dry year annual average scenario. This is an uplift in demand due to a dry year alone.
66.
In historic data there are a number of years where the uplift in demand has occurred in winter due to increased leakage. This is not considered to be part of the dry year effect.
17
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Amount of uplift 67.
To calculate the uplift in demand due to the dry year effect, we analysed monthly average demand (distribution input) for the period January 1999 to March 2012. All winter months with high demand driven by high leakage were excluded from the analysis where appropriate.
68.
Each of the 13 years analysed were categorised as having ‘normal’ or ‘dry’ demand, as reported in the security of supply index.
69.
Average demand in the normal and dry years was calculated and is presented in Table 4.4. The demand in 2011/2012 was excluded from this analysis as this was considered to be an abnormally cold year, with supressed demand. The summer was cool and cloudy with 115% of long term average (LTA) rainfall in July, and 131% of LTA rainfall in August. From examination of Met Office data, 2011 had the third lowest recorded hours of sunshine and had the third highest average wind speed recorded in the past 20 years.
Average ‘dry year’ demand
1297Ml/d
Average ‘normal year’ demand
1282Ml/d
Difference – dry year uplift
15Ml/d
Table 4.4: Comparison of ‘dry year’ and ‘normal year’ demand
70.
Demand in 2011/12 was 1251Ml/d (Table 4.5), 31Ml/d lower that the average demand in all normal years since 1999, due to a prolonged cool and cloudy summer. Therefore it is inappropriate to consider 2011/12 to be a normal year in terms of water demand.
Average demand 2011/12
1251Ml/d
Average ‘normal year’ demand
1282Ml/d
Difference
31Ml/d
Table 4.5: Comparison of 2011/12 demand and average ‘normal year’ demand
71.
The 31Ml/d difference between the base year (2011/12) and average normal year demand has been added to the calculated dry year uplift to give a total dry year effect of 46Ml/d. This is a fixed uplift that is applied to all years in the plan period, including the base year, which has been uplifted accordingly.
18
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
19
Allocation of dry year uplift 72.
We analysed household and non-household demand in 2011/12 and identified the elements of customer demand that could be impacted by dry weather.
73.
Table 4.6 shows the non-household standard industrial classification (SIC) categories with potential for dry weather demand increase, and the measured demand for these categories in 2011/12.
74.
The total demand for these non-household properties, compared with household demand in 2011/12 is shown in Table 4.7. This has been used to allocate the 46Ml/d dry year uplift between household and non-household properties, as presented in Table 4.8 (note that the 11% and 89% split has been rounded to 10% and 90%).
Non-household SIC categories with potential for dry weather demand increase
2011/12 demand (Ml/d)
Animal and dairy products and vegetable processing
8.3
Manufacture of miscellaneous food and drink
21.7
Pubs, clubs, restaurants and hotels
31.0
Sport, recreation and other personal services
10.8
Agriculture, forestry and fishing
19.8
Total
92
Table 4.6: 2011/12 demand for non household SIC categories impacted by dry year
Demand component with potential for dry year uplift
2011/12 demand (Ml/d)
Percentage of total demand
Non-household
92
11%
Household (measured and unmeasured)
723
89%
Total
815
100%
Table 4.7: 2011/12 total demand for households and non-households impacted by dry year effect
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Demand component
20
Percentage of dry Volume of dry year year uplift uplift (Ml/d)
Non-household
10%
4.6
Household (measured and unmeasured)
90%
41.4
Total
100%
46
Table 4.8: Summary of dry year uplift Split between measured and unmeasured households 75.
We have analysed household demand data for 2010/11 and 2011/12 from our DCM and measured customer monitor.
76.
The DCM, which contains in the region of 900 representative unmeasured households, gives an estimate of unmeasured household demand.
77.
The measured customer monitor was established in November 2010 to provide a picture of actual measured monthly consumption. The monitor is required as this data is not accurately available from the billing file, due to the read-estimate household billing cycle and frequency of meter reads. The monitor is made up of 1000 metered properties selected to be representative of the measured household property base. It provides average monthly household consumption.
78.
The data shows that 2010/11 was considered to be a dry year due to a relatively warm, dry summer. Whereas 2011/12 was a wet year with lower customer demand.
79.
Data from the DCM and measured customer monitor was used to compare customer demand in the months April to September in 2010/11 and 2011/12. Table 4.9 compares the relative uplift in demand calculated for measured and unmeasured household customers in these two years.
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
2010/11 uplift (Ml/d) Measured households
Unmeasured households
November
0
0
December
0
0
January
0
0
February
0
0
March
0
0
April
16.6
54.6
May
5.3
10.3
June
11.8
32.8
July
8.1
33.5
August
1.9
13.0
September
3.0
32.9
October
0
0
Average
3.9
14.8
1
4
Ratio
Table 4.9: Measured and unmeasured household dry year uplift
80.
This data showed that the average uplift in demand in 2010/11 was split in the ratio 1:4 between measured and unmeasured household customers. This ratio has been used in the demand forecast to allocate the 41.4Ml/d household dry year uplift between measured and unmeasured households. Inclusion of dry year uplift in the demand forecast
81.
The dry year uplift to households (41.4Ml/d) is applied to the household PCCs as part of the Experian micro-component model. The model itself is discussed in Section 4.4.3. The components of demand on which the dry year is considered to have an impact are personal washing and garden watering.
82.
The PCC values for measured households are uplifted to 110.5 and 110.8litres per head per day (l/h/d) for the Grid SWZ and East SWZ, respectively.
83.
The corresponding reported PCCs in the planning tables are 113.6 and 114.0l/h/d, as this includes an estimated additional volume due to MUR. This additional volume is added
21
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
before the MLE adjustment is made. This means that the inferred MUR from comparison of these two PCC values includes both MUR and the MLE adjustment on this MUR. 84.
The PCC values for unmeasured households in the base year are uplifted to 156.0 and 155.6l/h/d for the Grid SWZ and the East SWZ, respectively. The corresponding reported PCC in the planning tables, which incorporates MUR and the hidden and transient population’s consumption, are 166.0 and 165.9l/h/d, respectively.
85.
The dry year uplift to non-households (4.6Ml/d) is added as a flat volume to the total consumption figures. The uplift is split between measured and unmeasured nonhouseholds based on the proportion of non-household water delivered to those categories.
4.3.2
Distribution losses
86.
Distribution losses are total leakage less the leakage from supply pipes (SPL).
87.
Total leakage for the base year, taken from our adjusted MLE table (Table 4.2) was 274.52Ml/d.
88.
Leakage is composed of several different components, as presented in Figure 4.4:
Service reservoir leakage Trunk main leakage Leakage in distribution management areas (DMAs).
22
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.4 Components of total leakage
89.
The leakage associated with service reservoirs and trunk mains is fixed at 28Ml/d (as reported for the base year).
90.
The leakage in DMAs makes up the remainder of total leakage, with SPL estimated as 31.2% of this figure. This is based on a survey of SPL at DCM properties carried out in 2007/08.
91.
Leakage is explained in more detail in Section Part 3, Section 6.
4.3.3
Distribution system operational use
92.
The final component of the base year distribution input is distribution system operational use (DSOU). This is water we use for activities such as mains flushing, air scouring and water quality testing.
93.
For the base year this is taken as our adjusted MLE volume reported in Table 4.2.
4.3.4
Water resource zones split
94.
The annual report to Ofwat data is provided at a company level. The demand forecast tables are completed at a water resource zone level. Therefore all of the elements of the demand forecast discussed above are split between the Grid SWZ and the East SWZ.
95.
The split of properties between the two water resource zones is provided as part of the Annual Review of the WRMP to the Environment Agency.
23
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
96.
The split of population between the two water resource zones is calculated based on the property numbers in each category and the relevant occupancy rate.
97.
For the other components (distribution losses, water taken unbilled, distribution system operational use etc.), zone ratios were used to split the total volume of water between the Grid SWZ and the East SWZ. This ratio is 0.45% to East SWZ and 99.55% to the Grid SWZ, based on the Annual Report of the WRMP to the Environment Agency.
98.
The household dry year effect (41.4Ml/d) is split between the two water resource zones based on the number of measured and unmeasured household properties in the two zones. This is done within the micro-component model.
99.
The allocation of the non-household dry year effect (4.6Ml/d) to the East SWZ and Grid SWZ has been done based on analysis of data from recent dry years. This data indicated that on average 0.64% of the total dry year uplift is assigned to the East SWZ, which is equivalent to 0.30Ml/d of the total 46Ml/d dry year uplift. The East SWZ dry year allocation was then split with 60% to households and 40% to non-households, in line with the peak week assumptions, which resulted in 0.12Ml/d being assigned to nonhouseholds in the East SWZ. The remainder of the non-household dry year uplift (4.48Ml/d) is therefore assigned to the Grid SWZ.
4.4
Background to forecast changes in demand to 2039/2040
100. The forecast demand to 2039/40 is built up from forecast changes in each component of distribution input. These components are set out in Figure 4.2. The predicted future changes to these components, which are described in this section, are all forecast from a starting point of the base year numbers. 4.4.1
Water delivered
101. The forecast water delivered is driven by our own policies; changes to property numbers, population and consumption; and climatic variations. 102. Before discussing the individual changes in each component of water delivered, Figure 4.5 to Figure 4.8 present the methodologies by which water delivered is forecast for household and non-household properties. The forecasts for the individual components within these diagrams are discussed in this section.
24
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.5 Methodology for forecasting measured household water delivered
25
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.6 Methodology for forecasting unmeasured household water delivered
26
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.7 Methodology for forecasting measured non-household water delivered
Figure 4.8 Methodology for forecasting unmeasured non-household water delivered
27
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Yorkshire Water policies 103. The first changes that affect the forecast water delivered, relate to our own policies. We have a policy of demand management, through both reduction in customers’ water use through water efficiency and metering, and also through reduction in leakage on our own distribution system. Leakage 104. The forecast changes to total leakage impact on SPL, which affects water delivered. The remainder of leakage comprises distribution losses. 105. Total leakage is included in the demand forecast and was 274.52Ml/d in the base year (2011/12), taken from our adjusted MLE calculations (Table 4.2). We have calculated the sustainable economic level of leakage (SELL) at 297.1Ml/d, which is used as the value of total leakage for the lead-in and forecast years. 106. Base year leakage and our baseline leakage forecast is covered in detail in Part 3, Section 6. Metering strategy 107. The volume of water consumed by measured and unmeasured households is affected by the number of properties who opt to switch to a metered supply, known as domestic meter optants (DMOs). 108. We operate a policy of meter optants with moderate promotion. As part of this policy a meter penetration forecast has been developed. The final assumptions behind this forecast are as follows:
Actual number of meter optants for 2011/12, 2012/13 and 2013/14
During AMP6 the target is to deliver 200,000 optants over a 5 year period. This is an increase above the average number of optants in the current AMP and will be achieved through promotion of metering to ‘vulnerable’ customers and customers with affordability issues.
From the end of AMP6, a gradual decline in the number of DMOs to 15,000/year in 2030/31 (AMP9), which is then a fixed rate for the rest of the planning period. This decline reflects the decreasing number of unmeasured households available to opt and with a financial benefit of opting.
The forecast for the remainder of AMP5 (i.e. 2014/15) is taken from the business plan
109. Figure 4.9 presents the forecast DMOs for the 25 year planning period.
28
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.9 Domestic meter optants forecast
110. This long term trend means that by 2039/40 approximately 60% of existing households will be metered. 111. The total annual DMOs are split between the Grid SWZ and East SWZ based on the distribution observed in 2010/2011 and 2011/2012, which was 99.35% to the Grid SWZ and 0.65% to the East SWZ. 112. In addition to the DMOs, all new build properties are fitted with a water meter. As a result of DMOs and new connections we forecast that 80% of all household properties will be metered by 2039/40. 113. The predictions for the split in external to internal meters are based on current fitting rates provided by our metering team:
Existing measured properties split 44%:56% external to internal New builds split 91%:9% external to internal DMOs split 48%:52% external to internal.
Water efficiency savings 114. Our water efficiency strategy is set out in Part3, Section 5. 115. The water efficiency savings are built into the demand forecast as a fixed rate of 2Ml/d each year, in line with the current water efficiency target. This is taken off the water delivered to household and non-household properties. 116. The water efficiency savings are cumulative and subject to one off implementation. This assumes that a water efficiency measure is only introduced once and that the water
29
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
30
savings drop off over a set amount of time. This is defined as the half-life, which is the year from time of peak savings until only half of the peak saving are achievable. Waterwise (2008 and 2010) Evidence base for large-scale water efficiency in homes assumes a typical asset life of 10 years, with a half-life of 5 years. 117. The cumulative water efficiency savings for AMP5 and AMP6 are illustrated in Table 4.10 below.
Year 2011-12
2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19 2019/20 0
2012-13
0
0
0
0
0
0
0
0
2
1.8
1.6
1.4
1.2
1
0.9
0.8
2
1.8
1.6
1.4
1.2
1
0.9
2
1.8
1.6
1.4
1.2
1
2
1.8
1.6
1.4
1.2
2
1.8
1.6
1.4
2
1.8
1.6
2
1.8
2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20 Cumulative Water Efficiency Savings
2 0.00
2.00
3.80
5.40
6.80
8.00
9.00
9.90
10.70
Table 4.10: Cumulative water efficiency changes for AMP5 and AMP6
118. This water efficiency savings are divided between household and non-household properties on a ratio of 3:1 based on current activity levels. The split between the property categories and water resource zones are based on the percentage of total properties in each category. Customer demand management 119. Customer demand management is built into our demand forecast. This comprises the metering strategy and water efficiency savings outlined above. These demand savings are summarised in Table 4.11.
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Customer Demand Management (Ml/d) Metering Strategy Water Efficiency Savings
31
2011/12 (Base Year)
2014/15 (end of AMP5)
2019/20 (end of AMP6)
2024/25 (end of AMP7)
2029/30 (end of AMP8)
2034/35 (end of AMP9)
2039/40 (end of AMP10)
3.84
17.48
41.39
60.14
73.00
83.68
94.94
-
5.40
10.70
13.35
14.68
15.34
15.69
3.84
22.88
52.08
73.47
87.65
98.99
110.60
Total
Table 4.11: Summary of customer demand management built into the demand forecast
Properties, population and occupancy rates 120. The amount of water delivered to each property category depends on the number of properties in that category and, in the case of household properties, the population associated with those properties. 121. Property data for the base year was extracted from the billing file, and is used as the basis for the forecast of properties in future years. Measured household properties 122. The forecast for measured households includes existing measured households, meter optants and the predicted numbers of new build households each year. 123. The DMO forecast is discussed above. 124. For 2012/13 the actual number of new build households was used. The new build household forecast (Figure 4.10) is based on housing projections from local strategies provided by Local Authorities. These strategies provided the total number of new households (320,700) to be built by 2028. The forecast is for a gradual increase in the number of new builds each year to 2021/22, followed by a gradual decline to 2027/28. This trajectory is based on information provided by City of Bradford Metropolitan District Council. Currently this is the only council that has provided an estimate of new builds per year to 2028, rather than just a total of new properties predicted to be built by 2028. 125. Beyond 2028 the number of new households per year is forecast to plateau at around 15,000 per year. 126. A total of 509,335 new household properties are forecast to be built in the planning period. This is an average of 18,191 per year.
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.10: New build households forecast
127. Figure 4.11 shows the property forecast for household properties over the planning period, divided into measured households, unmeasured households and voids.
Figure 4.11: Household property forecast
32
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Unmeasured household properties 128. The unmeasured household property forecast shows a decrease in the total number of properties over the plan period. This is a result of DMOs, which move existing unmeasured households into the measured household category. This is the only change to this category. Measured Non-households 129. The measured non-household property forecast includes the forecast number of new commercial connections per year (Figure 4.12) and demolitions/ change of use properties. The total number of measured non-household properties decreases over the plan period as the number of demolitions/ change of use properties exceeds the number of new commercial connections. 130. The forecast number of new connections (Figure 4.12) was provided by Loop (our customer services and billing department).
Figure 4.12: New commercial connections forecast
131. The forecast number of demolitions/ change of use properties was established from analysis of historical trends in the annual reporting to Ofwat data. The average number of demolitions/ change of use properties was inferred from total measured non-households less new commercial connections. 132. The average number of demolitions/change of use properties for AMP5 (1,275 properties per year) was then used, along with the forecast for new commercial connections, to forecast measured non-households.
33
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
133. The non-households property forecast is presented in Figure 4.13 below.
Figure 4.13: Non-household property forecast Unmeasured Non-households 134. The unmeasured non-households property forecast (Figure 4.13) is based on the assumption that there will be a decline in the number of properties during the planning period. 135. The historical trend in property numbers was established from annual reporting to Ofwat data. This data showed an average decrease of 274 properties per year in AMP5. We considered that applying this annual decrease throughout the whole plan period would result in too many properties being lost by 2039/40. Therefore the most appropriate forecast was assumed to be annual decreases of:
274 properties for the remainder of AMP5
200 properties for AMP6
100 properties from AMP7 onwards.
34
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Void properties 136. It is assumed that the number of void properties (household and non-households) is maintained at 131,001 over the planning period. This forecast is provided by the income team in our customer services and billing department. Populations and occupancy rates 137. We have followed the guidance on population projections from Household and Population Estimation and Projections: Methodological Guide (Experian, 2007). 138. Trend based population forecasts for the 25 year planning horizon (2011 to 2040) have been produced by the market and information analysts Edge Analytics. We have used information from Local Authority housing strategies to develop a policy based forecast, which is used in the demand forecast. 139. The basis of the Edge Analytics trend based population forecast is the ONS mid-year estimate (MYE) 2011. The projections are linked to the census output area boundaries and take into account trends in fertility, mortality and migration. These projections are available at local authority level, from which a population estimate is then produced for each of the water resource zones. 140. The population forecast for the plan period is provided in Figure 4.14 and Table 4.12. These indicate that both zones will experience a similar percentage increase in population over the forecasting period. The Grid SWZ shows an increase in population of 17% between the base year and 2040, and the East SWZ shows a percentage increase in population of 20% over the same period.
Figure 4.14: Total population forecast
35
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
36
Population 2011/12 2014/15 2019/20 2024/25 2029/30 2034/35 2039/40 Forecast (Base Year) (end AMP5) (end AMP6) (end AMP7) (end AMP8) (end AMP9) (end AMP10) Grid SWZ
4,895,016
4,918,426
5,052,485
5,267,710
5,442,753
5,589,008
5,746,693
East SWZ
30,749
30,938
31,900
33,430
34,703
35,768
36,890
4,925,766
4,949,364
5,084,385
5,301,140
5,477,456
5,624,777
5,783,583
Total
Table 4.12: Summary of population forecast by AMP period 141. The total population is split between the different property categories based on occupancy rates. The occupancy rates are assumed to be the same for the Grid SWZ and the East SWZ. The split of population between the two resource zones is therefore based on property numbers. 142. For future years, the measured non-household population is the communal population, which was forecast by Edge Analytics. The Edge forecast was rebased to the mid-year estimate (MYE) 2011. The communal population of a military garrison in North Yorkshire was removed from the communal population as it has its own water supply. 143. The measured non-household population was then subtracted from the total zone populations to give total household and unmeasured non-household population. 144. The processes for forecasting measured and unmeasured household properties, population and occupancy rates, are summarised in Figure 4.15 and Figure 4.16 below.
Figure 4.15: Measured household properties and population forecasting methodology
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.16: Measured household properties and population forecasting methodology
145. Trends in the average household size (i.e. year on year percentage change) are then used as the basis for forecasting future occupancy rates for unmeasured households and mixed use properties (from the unmeasured non-household category). The occupancy rate increases gradually through the plan period. This is based on the assumption that DMOs will be those households with smaller occupancy rates because it is more economical for them to switch to a meter. As these properties leave the category, the properties with larger occupancies will be those that remain. Thereby increasing the average occupancy rate. 146. The measured household population is the residual population. The average occupancy rate for measured households is calculated by dividing the measured household population by the number of measured households for each year over the plan period. 147. The occupancy rates for measured households are more complex as this category comprises three different property types: existing measured, new builds and DMOs. The planning tables require the average measured household, new build households and DMO occupancy rates to be reported separately. 148. The new build household occupancy rate is based on findings from the WRc (2012) report Water Consumption of Homes Built to Part G and Code for Sustainable Homes Standards. This report gave an average nationwide occupancy rate for new builds of 2.37. Our forecast is for a gradual decrease in occupancy of new builds to 2.15 in 2027/28, where it plateaus for the remainder of the plan period. This is based on an assumed need to provide low/single occupancy properties for an increasing elderly
37
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
population who still live in their own homes, and a continued rise in single occupancy living. 149. We consulted with Edge Analytics, who confirmed that there will be an increasing elderly population choosing independent living rather than care homes. The assumption is that in many cases one spouse will therefore wish to downgrade to smaller properties. 150. The DMO occupancy rates show a gradual increase. This is based on the assumption that it is the unmeasured properties with the lowest occupancy rates that will chose to switch to a meter. Year on year the occupancy rate of the opting properties will increase as the properties with lower occupancy will already have switched to a meter in previous years. 151. As discussed in Section 4.3.1, the hidden and transient population, which is not included in the census, and therefore total population, is added into the demand forecast as a volume of water consumed rather than a population. It has been assumed that this population, and therefore the volume of water consumed by this population, is fixed throughout the plan period. Consumption 152. As stated in Section 4.3.1, consumption is the water used by a property, which includes the volume used by the property and meter under registration, but excludes supply pipe leakage. Household Volume 153. The volume of water used by household properties is calculated from per capita consumption (PCC) and population, as demonstrated in Figure 4.17 and Figure 4.18.
38
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.17: Methodology for forecasting measured household volume
Figure 4.18: Methodology for forecasting unmeasured household volume 154. Future PCC values have been forecast using a micro-component approach, which is in line with the Water resources planning guideline (Environment Agency et al., 2013).
39
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
155. A Yorkshire specific micro-component model (MCM) was produced for the demand forecast by Experian Business Strategies (Experian). The model produces an unmeasured household and measured household PCC forecast. The forecasts incorporate a range of water using appliances and project future use through assumptions around the effects of changes in water efficiency, regulations and social trends on PCC. The possible emergence of more water efficient appliances in the future has also been taken into account. 156. The majority of information on appliance ownership within the model was determined by a Yorkshire Water survey, undertaken in May to June 2012. Where data was unavailable, Experian used data from previous MCMs and national research sources, including the Environment Agency and Target Group Index (TGI) market research data. 157. Results from the survey on appliance ownership were coded by the socio-economic ACORN classification. This grouping was then used to project appliance ownership in the future using techniques developed by Experian, based on forecasts of employment, occupation and demographic change. 158. Appliance ownership and PCC are affected by the occupancy rate and property numbers. Therefore the same populations and property forecasts have been used in the MCM and the demand forecast. 159. The MCM provides the PCC in the following categories:
Toilet use Personal washing Clothes washing Dish washing Garden watering Other use (includes plumbing losses, swimming pools, drinking water etc.).
160. Table 4.13 to Table 4.14 and Figure 4.19 to Figure 4.20 summarise the breakdown of PCC by micro-component for the dry year annual average scenario. 161. Total PCC is summed from the various micro-components and applied to the population forecasts to derive the household volume of water used within the demand forecast. 162. Separate PCC estimates are provided for measured households and unmeasured households within the MCM. It is assumed that household consumption generally decreases when a property switches to a meter. Therefore it is assumed that the DMOs PCC is the same as the measured households PCC.
40
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Unmeasured Household
Measured Household
Household Type
41
2011/12 (Base Year)
2014/15 (end of AMP5)
2019/20 (end of AMP6)
2024/25 (end of AMP7)
2029/30 (end of AMP8)
2034/35 (end of AMP9)
2039/40 (end of AMP10)
Toilet use Personal washing Clothes washing Dish washing Garden watering
27.42
26.56
25.55
24.91
24.65
24.62
24.75
41.74 15.25 10.13 9.95
42.93 15.32 10.12 9.46
43.74 15.40 10.05 8.87
43.92 15.46 10.06 8.57
43.75 15.48 10.07 8.44
43.41 15.49 10.07 8.36
43.17 15.50 10.07 8.30
Other PCC PCC (Adjusted for New Builds)
6.04 110.53
6.06 110.46
6.09 109.71
6.11 109.03
6.12 108.51
6.13 108.08
6.14 107.92
110.53
110.92
111.10
111.26
111.27
111.23
111.40
Toilet use Personal washing
27.92
27.14
26.17
25.55
25.22
25.10
25.14
67.99
69.25
70.45
71.92
72.86
73.05
73.51
Clothes washing Dish washing
15.20 9.17
15.18 9.05
15.14 8.86
15.08 8.76
15.03 8.69
14.95 8.59
14.87 8.49
Garden watering Other PCC
27.19 8.52 155.99
28.34 8.53 157.49
30.90 8.54 160.06
33.61 8.54 163.48
35.97 8.55 166.32
38.27 8.54 168.50
41.31 8.54 171.85
Demand (l/h/d)
Table 4.13: Breakdown of Grid SWZ PCC by microcomponent (dry year annual average scenario)
Unmeasured Household
Measured Household
Household Type
2011/12 (Base Year)
2014/15 (end of AMP5)
2019/20 (end of AMP6)
2024/25 (end of AMP7)
2029/30 (end of AMP8)
2034/35 (end of AMP9)
2039/40 (end of AMP10)
Toilet use Personal washing Clothes washing Dish washing Garden watering
27.37
26.49
25.46
24.82
24.57
24.56
24.69
42.12 15.25 10.21 9.89
43.44 15.33 10.21 9.55
44.42 15.41 10.14 8.84
44.66 15.46 10.15 8.54
44.52 15.48 10.16 8.42
44.17 15.49 10.16 8.34
43.92 15.50 10.16 8.28
Other PCC PCC (Adjusted for New Builds)
5.94 110.80
5.96 110.84
5.99 110.26
6.01 109.64
6.02 109.16
6.03 108.74
6.03 108.58
110.80
111.40
111.83
112.12
112.18
112.18
112.35
Toilet use Personal washing
27.94
27.15
26.18
25.56
25.23
25.11
25.14
67.97
69.24
70.46
71.98
73.01
73.20
73.68
Clothes washing Dish washing
15.26 9.19
15.25 9.07
15.21 8.88
15.15 8.78
15.09 8.71
15.01 8.61
14.93 8.51
Garden watering Other PCC
27.00 8.32 155.67
28.15 8.32 157.17
30.72 8.33 159.78
33.46 8.33 163.27
35.84 8.34 166.21
38.14 8.33 168.40
41.18 8.33 171.78
Demand (l/h/d)
Table 4.14: Breakdown of East SWZ PCC by microcomponent (dry year annual average scenario)
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.19: Percentage breakdown of measured household PCC
Figure 4.20: Percentage breakdown of unmeasured household PCC
163. The PCC for new build households, which fall into the measured households category, is forecast to align with the Code for Sustainable Homes (CLG, 2007), with a pre-MLE PCC of 120l/h/d. This has been built into the average PCC for measured households.
42
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
164. We provided data to the WRc (2012) project Water Consumption of Homes Built to Part G and Code for Sustainable Homes Standards, which concluded an average PCC for new build households of 113l/h/d. These findings support the use of the Code for Sustainable Homes PCC for the demand forecast. 165. The climate change impact and dry year effect are both built into the MCM, thereby allowing the different scenarios to be modelled and incorporated into PCC. Hidden and transient volume 166. The volume of water consumed by the hidden and transient population is estimated to be a fixed volume of 8.8Ml/d for the plan period. This figure was based on assumed PCC and occupancy rates for the different categories of hidden and transient population. The hidden and transient volume is added to the unmeasured households volume. Non-household volume 167. A forecast for measured non-household consumption was provided by Experian (October 2012). This has been used in the demand forecast as the base volume (pre-MLE, SPL and MUR). 168. This regional base volume consumption is forecast to decrease over the plan period from 253.68Ml/d in the base year to 225.13Ml/d in 2039/40. 169. The Experian forecast includes an econometric analysis to identify the historical relationship between non-household water demand and explanatory factors, such as industrial output and employment. The results of this statistical analysis were combined with Experian forecasts of output and employment by industry to provide forecasts for the Yorkshire Water supply area as a whole, and the two resource zones within the supply area. 170. The volume of water used by unmeasured non-households has been calculated from a consumption rate and the forecast property numbers. The consumption rate is assumed to be fixed over the plan period at 118.28 litres/property/day (l/p/d), which is the base year value from the annual report to Ofwat data. This rate was determined from a reassessment of the different components of unmeasured non-household use in 2008/09. The review identified four different types of unmeasured non-household properties:
Fixed charge properties Mixed use properties (i.e. small shops with a residential flat above) Lock ups or garages Others (e.g. animal troughs, council properties now domestic and mixed properties with a joint supply).
171. The total use for all unmeasured non-households was estimated to be 1.98Ml/d. This gives an average use of 118.28l/p/d.
43
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Meter under registration 172. MUR is assumed to remain at the base year rate. The MUR percentages for measured households, unmeasured households and measured non-households are given in Section 4.3.1. MUR does not apply to unmeasured non-households. MLE adjustments 173. The MLE adjustments are applied to the volume of water used, SPL and MUR using the rates derived for the base year (refer to Section 4.2). Supply pipe leakage 174. The forecast total volume of supply pipe leakage is directly linked to total leakage which for the forecast years is the SELL value of 297.1Ml/d. Total SPL amounts to 31.2% of leakage in DMAs, which is 83.9Ml/d (post MLE). 175. We have two different SPL rates for properties dependant on whether the property is metered and the location of the meter. The assumption is that SPL for externally metered household properties are less than for unmeasured properties or internally metered household properties. This is because it is assumed that the burst would be identified and repaired more quickly in an externally metered property due to an increase in customer bills. 176. SPL is added into the demand forecast as a flow rate applied to each property. It is assumed that SPL for externally metered households is a quarter of that of other properties. This is based on previous modelling and assessment of the DCM. 177. As the total number of properties in each category varies each year, the SPL rates vary subtly each year also. Water taken unbilled 178. The amount of water taken unbilled is assumed to be fixed at the 2012/13 volume (30.99Ml/d) for the remainder of the plan period. This figure has been used instead of the base year number (34.06Ml/d) as it represents a significant decrease in water taken unbilled. 179. This decrease in water taken unbilled was based on a reassessment of water taken legally unbilled. Properties included in this category are Yorkshire Water sites, measured customers not billed and Yorkshire Water unmeasured households. It was found that a number of these properties that were not historically included in our billing file are now included.
44
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Climatic influences on demand Climate change 180. The reported demand forecast includes an uplift due to climate change. 181. Following guidance from the Environment Agency, we have used the climate change scenarios presented in the Impact of Climate Change on Demand (UKWIR, 2012) to determine the potential impact of climate change on customer demand. 182. We have assumed that the Severn Trent scenarios are more appropriate to Yorkshire Water than the Thames Water scenarios, due to geographical and climatic similarities. We selected the Household Annual Average for the Humber North region as the most appropriate climate change scenario for the Yorkshire Water supply area. We also selected the mid-range P50 percentile scenario within the Humber North region as there is no evidence to justify use of the higher or lower ranges. 183. The Defra commissioned report Climate Change and Demand for Water (CCDeW, 2003) states that the major impact of climate change in north east England is likely to be on garden use and personal washing. Climate change has been applied to household demand through the MCM and this assumption has therefore been built into the model. The result is a forecast growth in household demand due to climate change, of 0 - 0.26% for the Grid SWZ and 0 - 0.31% for the East SWZ, over the planning period. 184. Figure 4.21 demonstrates the forecast increase in household demand as a result of climate change.
Figure 4.21: Impact of climate change on household demand
45
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
185. Climate change has not been included in the demand forecast for non-household properties. This is because there is no evidence of an impact on industrial demand. Equally, there is little potable water supplied for irrigation purposes. Therefore we are assuming no impact on agricultural demand. Dry year effect 186. The reported demand forecast is for a dry year scenario in accordance with the Water resources planning guideline (Environment Agency et al., 2013). 187. The dry year uplift of 46Ml/d is applied to every year in the plan period. This is incorporated into the demand forecast as described in Section 4.3.1. Peak week 188. Critical period (peak week) analysis has been carried out for the East SWZ only. The critical period is the average daily demand in the peak week. 189. For the Grid SWZ we plan for a dry year annual average scenario only. Network flexibility in this zone means it is not susceptible to supply demand issues at times of unusually high demand. 190. The East SWZ has only a run-of river source (where water is abstracted directly from the river and not via storage reservoirs or lagoons) and a small spring with limited storage. In addition, the area is prone to increases in summer population through tourism. It is therefore susceptible to peak summer demands and a possible temporary shortfall in supply. 191. To establish the level of uplift experienced in the East SWZ due to peak demand, distribution input data for the last 10 years was analysed. Data for the summer months was normalised to remove any dry year effects and then the maximum uplift (maximum weekly demand compared to normalised annual average demand) was established. The maximum uplift established using this method was 2.44Ml/d, which is the uplift observed in 2003. 192. The demand forecast already includes a 46Ml/d dry year effect, of which 0.30Ml/d is apportioned to the East SWZ. To calculate the total peak week uplift, this value has been removed from the maximum uplift to give a value of 2.14Ml/d. 193. The dry year uplift split between households and non-households was based on a 90% to 10% split (Section 4.3.1). This was not considered appropriate to the peak week because of the larger proportion of non-household properties that would be influenced by peak demand in this zone (e.g. hotels, restaurants, public facilities) related to tourism. It was therefore assumed that 60% of the uplift would be to households and 40% to nonhouseholds. Therefore in a peak week these properties would use a higher proportion of the additional demand.
46
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
194. The peak week uplift was applied to the total water delivered to each of the four property categories. 195. The peak week demand is water use driven by a short period of hot dry weather. In a peak week it is considered that measured households are more likely to use a higher proportion of water for garden watering to look after their plants than in the dry year annual average scenario. 196. In general people who are on a water meter are more conscious of the amount of the water that they are using and therefore use less water than unmeasured properties. For example, the micro-component breakdown of household water use (Grid SWZ 2011/12) indicates that measured households use 9.95l/h/d for garden watering, whereas unmeasured properties use 27.19l/h/d. 197. However, in a peak week, which is a very short period, we assume that people in measured properties will temporarily change their habits to save their plants from drought. They will therefore use a higher proportion of water for garden watering. In addition, it is assumed that unmeasured households already use a lot of water for the garden and that there consumption will not change as much during a peak week. 198. The split between measured and unmeasured households has therefore been assigned as 60% to 40% respectively. 199. The split between measured and unmeasured non-households is 96% to 4%, based on water delivered in these categories, which corresponds to the split used for the dry year uplift. 200. The peak week uplift was applied to the total water delivered to each of the four occupied property categories. 4.4.2
Distribution losses
201. As discussed in Section 4.3.2, distribution losses comprises leakage from service reservoirs and trunk mains, plus the losses in DMAs which are not supply pipe leakage. 202. The leakage from service reservoirs and trunk mains is a fixed volume of 28Ml/d, which is applied throughout the plan period. 203. Leakage in DMAs is also fixed throughout the plan period at 185.2Ml/d, in line with SELL (Section 4.3.2). It is calculated as SELL (297.1Ml/d) minus leakage from service reservoirs and trunk mains (28Ml/d), and customer supply pipes (83.9Ml/d). 4.4.3
Distribution system operational use
204. The amount of water used for distribution system operation (e.g. mains flushing, air scouring, water quality testing etc.) is assumed to be fixed during the plan period at the 2012/13 volume (1.90Ml/d). 205. This number has been used instead of the base year number (3.07Ml/d) as it represents a significant decrease in our operational water use. This decrease has resulted from a
47
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
change in the way we clean water mains, from swabbing to air pigging which uses significantly less water.
4.5 4.5.1
Forecast demand to 2039/40 Water delivered Household water delivered
206. The forecast regional water delivered to household properties over the planning period is presented in Figure 4.22 and Table 4.15. Note that this includes the dry year uplift and climate change. 207. The measured household water delivered is forecast to increase from 217.9Ml/d in the base year to 545.7Ml/d by 2039/40. This increase is a combination of increased property numbers (new build households and DMOs) and climate change. 208. The unmeasured household water delivered is forecast to decrease over the planning period, from 551.2Ml/d in the base year to 248.8Ml/d in 2039/40. This is due to the number of customers continuing to opt for a metered supply.
Figure 4.22: Water delivered to households
48
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Measured Households
49
Unmeasured Households
Dry Year Forecast Water Delivered (Ml/d) 2011/12
2039/40
2011/12
2039/40
Grid SWZ
217.94
545.67
551.21
248.83
East SWZ
1.18
3.40
3.67
1.72
219.12
548.93
554.88
250.49
Total
Table 4.15: Water delivered to households
Measured non-household water delivered 209. The regional measured non-household water delivered (Figure 4.23 and Table 4.16) is forecast to decrease from 277.5Ml/d in the base year to 244.4Ml/d in 2039/40. This is primarily due to increased water efficiency, particularly within the non-service sector.
Figure 4.23: Water delivered to measured non-households
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Measured Non-household Water Delivered (Ml/d) Grid SWZ
East SWZ
Total
2011/12
2039/40
276.48
243.60
1.02
0.82
277.49
244.42
Table 4.16: Water delivered to measured non-households
Unmeasured non-household water delivered 210. The unmeasured non-household water delivered (Figure 4.24 and Table 4.17) decreases during the plan period, from 2.61Ml/d in the base year to 2.07Ml/d by 2039/40. This is due to the decline in property numbers over the planning period.
Figure 4.24: Water delivered to unmeasured non-households
50
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Unmeasured Non-household Water Delivered (Ml/d)
51
2011/12
2039/40
2.57
2.04
Grid SWZ
East SWZ
Total
0.04
2.61
0.03
2.07
Table 4.17: Water delivered to unmeasured non-households Total water delivered 211. The total water delivered is the sum of water delivered to properties (including voids) and unbilled water. 212. The water delivered to void properties (households and non-households) varies slightly over the planning period, from 5.9Ml/d in the base year to 6.2Ml/d in 2039/40. This is based on SPL volumes and total property numbers. 213. Water taken unbilled is fixed during the plan period. 214. The total water delivered is forecast to increase slightly over the planning period from 1,094.7Ml/d in the base year to 1,083.3Ml/d in 2039/40, as presented in Figure 4.25 and Table 4.18.
Figure 4.25: Total water delivered
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Total Water Delivered (Ml/d)
2011/12
2039/40
Grid SWZ
1,088.59
1,077.14
East SWZ
6.10
6.14
1,094.69
1,083.28
Total
Table 4.18: Total water delivered 4.5.2
Distribution Input Dry year annual average scenario
215. Forecast total demand (distribution input) increases over the planning period from 1,295.71Ml/d in the base year to 1,298.35Ml/d. 216. Figure 4.26 shows the build-up of the individual components of demand to produce distribution input.
52
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Figure 4.26: Total distribution input
53
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
Dry year critical period (peak week) scenario 217. Figure 4.27 is a plot of distribution input for the East SWZ, which shows the figures for the dry year annual average and peak week demand forecast. 218. Under the dry year annual average scenario, the distribution input for the East SWZ rises from 7.00Ml/d in the base year to 7.11Ml/d by the end of the planning period. In contrast, under the dry year critical period scenario, the distribution input for the East SWZ is 9.25Ml/d in the base year, rising to 9.35Ml/d by the end of the planning period.
Figure 4.27: Comparison of dry year annual average and critical period (peak week) scenarios for the East SWZ
4.6
Weighted average demand
219. The weighted average demand is a sense check on the dry year (annual average) demand forecast, required by Ofwat. It represents our view of the demand that we are most likely to face over the planning period on average. The weighted average demand will therefore be used to determine the company’s revenue forecast for setting price limits. It is not used for any other water resource planning purpose. 220. Calculation of the weighted average demand is based on an assumption of the number of years that will be dry and the number of years that will be normal over a given period. For the Yorkshire Water region it has been assumed that 40% of years will be dry years, this
54
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
55
is based on data from the past 10 years (Table 4.19). This produces a final planning weighted average demand forecast.
Year
Dry/Normal
2003
Dry
2004
Normal
2005
Dry
2006
Dry
2007
Normal
2008
Normal
2009
Normal
2010
Dry
2011
Normal
2012
Normal
Table 4.19: Summary of dry/normal years
221. The weighted average demand distribution input is provided in Table 4.20 and Figure 4.28 below.
Weighted average demand (Ml/d)
2011/12
2039/40
Grid SWZ
1,261.31
1,262.62
East SWZ
6.77
6.86
1,268.08
1,269.48
Total
Table 4.20: Weighted average demand
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
56
1,500.00 AMP5
1,450.00
AMP6
AMP7
AMP8
Distribution Input (Ml/d)
1,400.00 1,350.00 1,300.00 1,250.00 1,200.00
1,150.00 1,100.00 1,050.00 1,000.00
Normal Year
Figure 4.28: Weighted average demand
Dry Year
Weighted Average
AMP9
AMP10
Kelda Group | Water Resources Management Plan | Yorkshire Water Services Ltd
57
Western House, Halifax Road, Bradford BD6 2SZ
yorkshirewater.com
