Water Usage

ESTABLISHING A HOUSEHOLD
WATER CONSUMPTION BENCHMARK

FOR SOUTH AFRICA
Report
to the Water Research Commission
by
JL du Plessis, HE Jacobs, A Ilemobade and M Crouch
on behalf of
Bernoulli (Pty) Ltd
Report No. 2980/1/20

ISBN 978-0-6392-0165-8
June 2020
Obtainable from:
Water Research Commission
Private Bag X03
Gezina, 0031
South Africa
orders@wrc.org.za or download from www.wrc.org.za
DISCLAIMER
This report has been reviewed by the Water Research Commission (WRC) and approved
for publication. Approval does not signify that the contents necessarily reflect the views
and policies of the WRC, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
ii
Establishing a household water consumption benchmark for South Africa
EXECUTIVE SUMMARY
Per capita water use is commonly employed as a baseline for the estimation of water demand, and is
utilised in many countries. Per capita household water use benchmark studies based on South African
data are limited and outdated. International studies do not cater for typical South African levels of
service or dwelling types. The planning and management of water supply and distribution institutions
would benefit from this study that collates and reports on per capita water use, consistent for South
African conditions.
This study was commissioned to research the available household benchmark literature and data, and
to establish a per-capita water use guideline. Various South African and international studies were
reviewed to formulate a standardised approach that would satisfy local conditions and global trends.
Studies reviewed included domestic water use, categories of water use and level of service (LOS),
water loss and leakage, factors affecting water use, water end-uses and activities, minimum water use,
household size, water price, household income and property value, geography and climate.
Five main parameters were identified for inclusion in the methodology of this research: level of service,
usage scenario, the number of people per household, geographic region and property value. The
available parameter data was further analysed to express its effect on water use in households. A model
was then formulated to integrate the relationship that the parameters would have on overall water use
in households.
A Microsoft (MS) Excel-based tool was subsequently developed to ease the implementation of the
model for practitioners. The tool allows the user to change parameters and arrive at an estimated water
use per capita for a specific use scenario, as a function of selected inputs. As more data is analysed
and further research is conducted on the topic, further development of the tool would improve its
accuracy. However, the current version of the tool compares well with the available data and other
sources of research that depict water use in South Africa.
Experts from industry, government and academia were consulted in workshop format to source potential
information and test whether the available information was relevant to the outcomes of the study.
Workshops were held in Cape Town, Midrand and Durban, and the outcomes of the workshops were
recorded in the form of meeting records. The model was adjusted and verified, based on inputs obtained
from experts at the three workshops.
The tool is able to determine the low, intermediate and high water use for five LOS categories (standpipes,
communal ablution blocks, yard connections, low-cost housing, house connections with indoor use only
and full house connections), five climate regions (varying from arid to humid), a varying number of persons
per household (one to seven) and three property value levels. Table 0.1 illustrates the typical variability of
the outputs, showing a range of results obtained from the model for six different levels of service and four
different household sizes. The results in Table 0.1 should be used as an indication of expected model
results. However, the tool should be used to determine specific outputs for specific scenarios.
iii
Table 0.1: Typical results for per-capita water use from the litre per capita per day (ℓ/c/d) model
Estimated water use (ℓ/c/d)

Number of persons per household
Level of service 1 2 3 4
LOS 1 Standpipes 22 22 22 22
LOS 2 Communal ablution blocks 54 42 37 34
LOS 3 Yard connections 85 62 52 46
LOS 4 Low-cost housing – limited fixtures 163 111 89 76
LOS 5 Full house connections (indoor) 275 198 163 143
LOS 6 Full house connections (including outdoor) 407 300 251 221
Based on the analyses and scenario testing of the model and tool, the following findings were derived:
• A multiparameter tool was developed that can be used to estimate household water use based on
five key input parameters.
• The model is most sensitive to the number of persons per household, with a notable decline in per-
capita consumption from a single-person household to three persons per household. The reduction
in per-capita demand then flattens off from four to seven persons per household.
• Based on input obtained from the workshops, the climate regions in the model were linked to the
regions defined in the modified Köppen-Geiger climate classifications (CSIR, 2015).
• Property value was used as a proxy for household income, in line with earlier studies (e.g., Van Zyl
et al., 2008) and the effect of property value was included for LOS 3, LOS 4, LOS 5 and LOS 6.
• Considering the proposed input parameters, the tool produces a set of 270 different values (six
levels of service for three usage scenarios, five regions and three property value categories) for
each household size option. If one were to consider one to seven persons per household, the tool
would produce 1,890 different answers of per-capita water use, depending on the selected inputs.
The values presented in Table 0.1 show a small selection of only 24 results from the full result set.
• The model outputs compared well with other studies, such as the Department of Human
Settlements (DHS) (2019), although the range of per-capita consumption is larger, as could be
expected since additional parameters such as climate region and number of persons per
household were incorporated in this study.
iv
ABBREVIATIONS
AADD Annual average daily demand

CABs Communal ablution blocks
CSFWUES California Single Family Water Use Efficiency Study
CSIR Council for Scientific and Industrial Research
DHS Department of Human Settlements
DWA Department of Water Affairs
FBW Free Basic Water
HDLI High-density, low-income
HENH High Efficiency New Home
JAWARA Journal of the American Water Resources Association
ℓ/c/d Litre per capita per day
LCD Tool Litre per Capita per Day Tool
LOS Level of Service
MS Microsoft
NSFHS National Survey of Functional Health Status
PPH Persons per household
REUWS Residential End Uses of Water Study
SNHG Standard New Home Group
USEPA United States Environmental Protection Agency
VB Visual Basic
WHO World Health Organisation
WRC Water Research Commission
v
TABLE OF CONTENTS
EXECUTIVE SUMMARY ....................................................................................................................... iii

ABBREVIATIONS .................................................................................................................................. v
LIST OF FIGURES ............................................................................................................................... vii
LIST OF TABLES ................................................................................................................................ viii
CHAPTER 1: INTRODUCTION .............................................................................................................. 1
BACKGROUND ....................................................................................................................... 1
OBJECTIVES .......................................................................................................................... 1
OUTPUT .................................................................................................................................. 1
METHODOLOGY .................................................................................................................... 1
CHAPTER 2: LITERATURE REVIEW ................................................................................................... 3
2.1 DOMESTIC WATER USE ....................................................................................................... 3
2.2 CATEGORIES OF WATER USE AND LEVELS OF SERVICE .............................................. 4
2.3 WATER LOSS AND LEAKAGE ............................................................................................... 4
2.4 FACTORS AFFECTING WATER USE .................................................................................... 4
2.5 WATER END-USES AND ACTIVITIES ................................................................................... 5
2.6 MINIMUM WATER USE .......................................................................................................... 6
2.7 HOUSEHOLD SIZE ................................................................................................................. 7
2.8 WATER PRICE, HOUSEHOLD INCOME AND PROPERTY VALUE ..................................... 9
2.9 GEOGRAPHY AND CLIMATE ................................................................................................ 9
2.10 LEVEL OF SERVICE ............................................................................................................. 10
CHAPTER 3: ANALYSIS OF PER-CAPITA CONSUMPTION ............................................................ 11
3.1 OVERVIEW ........................................................................................................................... 11
3.2 LEVEL OF SERVICE ............................................................................................................. 11
3.3 HOUSEHOLD SIZE ............................................................................................................... 11
3.4 CLIMATE REGION ................................................................................................................ 13
3.5 USAGE LEVEL ...................................................................................................................... 14
3.6 PROPERTY VALUE .............................................................................................................. 15
CHAPTER 4: CONCLUSION ............................................................................................................... 16
CHAPTER 5: FURTHER RESEARCH ................................................................................................. 18
BIBLIOGRAPHY ................................................................................................................................... 19
ANNEXURE A – PER CAPITA WATER USE STUDIES ..................................................................... 24
ANNEXURE B – LCD TOOL USER GUIDE ........................................................................................ 28
vi
LIST OF FIGURES
Figure 2.1: Water inputs and outputs of the international space station (Tobias et al., 2011) ............... 7
Figure 2.2: Summary of studies showing the effect of persons per household on per capita water
demand.................................................................................................................................................... 8
Figure 3.1: Measured South African per-capita consumption data for varying household sizes .......... 12
Figure 3.2: Per capita consumption for six levels of service and increasing household size ............... 12
Figure B-1: Input section of LCD Tool ................................................................................................... 29
Figure B-2: Results of LCD Tool for Case 1 (example case shown) .................................................... 30
Figure B-3: Map for selecting geographic regions (CSIR, 2015) .......................................................... 30
vii
LIST OF TABLES
Table 0.1: Typical results for per-capita water use from the litre per capita per day (ℓ/c/d) model ........ iv
Table 2.1: Residential annual average daily demand (per-capita water consumption guidelines for
South Africa) (DHS, 2019) ...................................................................................................................... 3
Table 3.1: Summary of multiplication factors for climate regions for each level of service .................. 14
Table 3.2: Multiplication factors for usage level .................................................................................... 15
Table 4.1: Water use per capita for different levels of service and number of persons per
household .............................................................................................................................................. 17
Table 4.2: Water use per capita for different levels of service vs climate regions ................................ 17
viii
CHAPTER 1: INTRODUCTION
BACKGROUND
Limited benchmarking information exists that relates to per-capita water use in South Africa. Other
countries rely on per-capita water use as a baseline for water use estimation, but in South Africa, the
available per-capita water use information is not readily available and is outdated. This information and
data are not available, and related information has not been collated in a standardised format that
pertains to South African conditions. This has an impact on planning and managing water supply in the
face of extreme weather events and against growing water scarcity. Growing settlement pressure and
the dichotomy faced by municipalities further adds to the complexity. Thus, there is a need to
consolidate and verify the information and provide water use estimation and benchmarks based in
South Africa.
Sustainable development solutions for water services provision can only be achieved if water use
estimation is relevant to South African conditions (level of service, socio-economic conditions,
demographics, regional climate, etc.). An international norm is often applied to a South African application,
which could be inappropriate for a largely hot and dry country. Policy and decision making can be informed
at local authority level by developing a South African benchmark of per-capita water usage.
It would be necessary to consider per-capita water use for different levels of service and from an
availability-of-water point of view. Once the information has been compiled, an up-to-date central
platform should be available from where estimators could access the per-capita water use.
OBJECTIVES
Benchmarking of per-capita water use in South Africa under South African conditions (climate, etc.)
does not exist. An international norm is continuously used to determine what would be appropriate for
a largely hot and dry country. The objective of this study is to determine a South African benchmark of
per-capita usage. The intent of the study is to provide benchmarking criteria of per-capita water use in
South Africa under South African conditions and an improved presentation of per-capita water use data.
OUTPUT
In support of the benchmarking exercise, a per-capita water use estimation tool was developed, called
the Litres per Capita per Day Tool (LCD Tool). It is available online at www.wrc.org.za/software/lcdtool.
The tool will assist municipalities, designers and regulators to estimate water use for existing and
proposed settlements. The LCD Tool will ultimately be a way for designers, planners and researchers
to compare conventional water use estimates with the tool’s estimates.
METHODOLOGY
The study involved applying research techniques to address the research problem, employing well-
known and accepted theories and principles, with immediate potential application. Several methods
were used to address the research objectives.
Formerly published per-capita water consumption was collected by means of a comprehensive desktop
review and targeted email requests from local experts. Where household water consumption is available
in combination with the corresponding household size, the per-capita consumption will be calculated
from the former values. The intention is not to collect new data, but rather to collate information that is
already available.
1
The desktop review focused primarily on formerly published journal articles, conference papers and
WRC research reports that were identified by the project team and key role players. Earlier studies were
identified, collated and classified as being either micro scale, where consumption is reported for
household use downstream of a consumer meter (e.g. Jacobs, 2007; Du Plessis and Jacobs, 2018), or
macro scale, where consumption is reported on a city-wide scale, also including all water loss in the
system and non-residential use. With macro-scale studies, the total consumption is typically divided by
the total population (e.g. Du Plessis, 2007) to provide a crude estimate of per-capita water use.
Guidelines for water use in South Africa have focused primarily on stand (plot) size as the independent
variable (Jacobs et al., 2004).
As part of this study, a few selected key factors that influence per-capita consumption were incorporated
into the LCD Tool. The following key input parameters were incorporated:
• Level of service: Six different levels of service were considered: standpipes, yard connections,
communal ablution blocks, low-cost housing with limited in-house connections (e.g. government-
subsidised housing; formerly RDP housing), full house connections: indoor only (including flats and
townhouses) and full house connections: including outdoor use (full title home on plot with garden).
The level of service will be used as a proxy for socio-economic factors such as income and property
value.
• Usage scenario: Three usage levels were incorporated, to allow for the average unrestricted use,
relatively comfortable unrestricted use and low-use scenarios that could be deemed a realistic
target during water restrictions. It would make sense to encourage water infrastructure planners to
provide for the comfortable unrestricted usage level to ensure adequate infrastructure capacity.
On the other hand, water usage targets during restrictions would consider the low-use scenario,
for example.
• Number of people per household: Various previous studies have confirmed a reduction in the
per-capita consumption with increased household size, both locally and abroad. The tool proposed
in this study allowed for this reduction by fitting a mathematical curve to all the available data, thus
allowing for the said reduction to be incorporated in the result.
• Geographic region: The regional mapping is likely to be more subjective and boundary setting
would rely on inputs from stakeholders and academic experts, and would be based on available
socio-economic, political, geophysical and vegetation maps of South Africa. For this study, five
water use regions were identified and mapped to describe water use in the different regions relative
to the other regions.
• Property value: Property value was incorporated as a proxy for income level, which is often
proportionate to willingness and ability to pay a specific price for water.
Considering the proposed input parameters, the tool produces a set of 270 different values (six levels
of service for three usage scenarios, five regions and three property value categories) for each
household size option. If one were to consider 10 options, for one to 10 persons per household, the tool
would produce 2,700 different answers of per-capita water use, depending on the selected inputs.
2
CHAPTER 2: LITERATURE REVIEW
2.1 DOMESTIC WATER USE
Water is required for many everyday activities. It is required for consumption, such as for drinking and
indirectly for cooking, and also for essential hygiene purposes, such as washing hands, preparing food,
showering or bathing. In some cases, water required for hygiene may also be linked to religious
activities. Water is also needed by those who aspire for more than just survival, in order to sustain a
certain desired lifestyle. For example, outdoor residential environments have been found to be
extremely important to homeowners (Blaine et al., 2012), also affecting residents’ sense of social status
and mental health. Gardening and outdoor water use have important positive effects on individuals, as
well as on the urban ecosystem. A poorly maintained garden has also been found to lower the potential
monetary value, not only of that property, but also neighbouring ones (Clayton, 2007). A lack of clean
water often leads to a lack of hygiene, which can lead to diseases such as diarrhoea or other faecal-
oral diseases, typhoid, and skin and eye diseases such as louse (Bradley, 1997). Esrey et al. (1985)
determined many years ago that the quantity of water used in a neighbourhood has a greater effect on
the frequency of diarrhoea events than the quality of the water.
Van Zyl et al. (2008) defined domestic water as water that is used for any household activity, both indoor
and outdoor, including water for drinking, cooking, laundry, cleaning, flushing toilets, garden use, pool
use, pet care, car cleaning, etc. Thompson et al. (2001) specifies four categories of domestic water use:
hygiene (including personal and household cleaning), consumption (including drinking and food
preparation), amenity use (including car washing, and garden and lawn irrigation) and productive use
(including water for livestock, small-scale horticulture and other household productions). The latter
category refers to developing countries or regions where people may cultivate backyard crops and keep
livestock, or run bartering or trading businesses from home.
Several water requirements are essential for living, with regard to health and hygiene. These include
water that is required for consumption, such as drinking and cooking water, as well as water needed for
hygiene purposes, such as cleaning the body, clothes and the household. Conversely, other water use
may not be vital to health and hygiene, but may be considered necessary for maintaining a relatively
higher standard of living. These include irrigation and water for car washing or pools. The Department
of Water Affairs (DWA) (2009) and later also the DHS (2019) provided guidelines for per-capita water
consumption for different settlement categories in South Africa. Consumption values range from
25 to 400 ℓ/c/d for a residential household. The DHS (2019) presented a useful table with per-capita
water use for different types of water supply, as per Table 2.1.
Table 2.1: Residential annual average daily demand (per-capita water consumption guidelines for South
Africa) (DHS, 2019)
Land use Persons Typical annual Annual average

per unit average daily daily demand
demand* (ℓ/c/d) range* (ℓ/c/d)
Standpipe 5 25 10-40
With dry sanitation 5 50 40-60
Yard connection With low-flow sanitation 5 60 50-70
With full-flush sanitation 5 70 60-80
Low-income housing 5 90 60-120
Residential 5 230 120-400
House connection
Group/cluster housing 3-5 120 130-120
Flats 1-4 150 250-110
* Per-capita calculated on persons per unit
3
2.2 CATEGORIES OF WATER USE AND LEVELS OF SERVICE
Table 2.1, as presented by the DHS (2019), outlines various levels of service from the South African
perspective. These levels of service could be linked to different categories of use described by others.
Two categories of water use were suggested by Willis et al. (2011) from an Australian perspective to
describe the essential versus non-essential requirements. A third category could be considered to
describe basic survival in line with the WHO (2003) and DHS (2019), which is equal to 20-25 ℓ/c/d. The
three categories of use considered for the purpose of this study include the following:
• Basic survival (minimum for survival, typically provided from communal standpipes)
• Non-discretionary (essential for sustainable urban living, which could be viewed in this study as
supply from yard taps provided on-site at each home)
• Discretionary (linked to improved standard of living), which could be further segregated to describe
indoor use only, versus indoor and outdoor water use
These categories link to the level of service. In recent years, the line between non-discretionary and
discretionary water use has become blurred as people tend to use non-discretionary end-uses as
discretionary end-uses. For example, a shower, which is typically a non-discretionary end-use, with the
purpose of cleaning the body for hygiene purposes, has become a discretionary end-use as people no
longer use showers simply for sanitation, but rather as a leisurely activity. Thus, Willis et al. (2011)
argue that there should be a set amount of water of approximately 40 to 70 ℓ/c/d that is a set requirement
for basic human needs. The value presented by Willis et al. (2011) is in line with the recent publication
of the DHS (2019) for yard tap supply, with a range of 40 to 80 ℓ/c/d. This level of use could be viewed
as non-discretionary water use and any water use above this value should be considered discretionary
water use, irrespective of what it is used for. The same approach is used in South Africa with supply of
Free Basic Water (FBW), where consumers in low-cost houses would qualify as indigents based on
their relatively low household income. The South African government initiated the FBW concept in 2001
(Smith, 2010). An indigent FBW allocation, normally set at 6 kℓ per month per household, is unique to
South African disadvantaged communities. The quantity of FBW would be provided each month,
regardless of what the FBW was used for.
In recent years, communal ablution blocks (CABs) have been introduced in areas where service
providers improved the level of service for communities that were previously not serviced or were
dependent on yard taps. Although the arrangement of CABs varies from different suppliers and projects,
they often consist of containerised showers, wash basins, laundry facilities, urinals and toilets (Crous
et al., 2013). Roma et al. (2010) reported that CABs have a water use of 35 to 40 ℓ/c/d.
2.3 WATER LOSS AND LEAKAGE
Water loss and leakage is common in residential homes. Leakages are neither non-discretionary nor
discretionary as they are not influenced by behaviour. Leakage and water losses are normally excluded
when reporting per-capita consumption. Water leakage and water losses were not defined as a water
use category in this study, because water leakage or loss would be estimated separately and would
then be added to the per-capita use.
2.4 FACTORS AFFECTING WATER USE
Water use is dependent on a variety of factors. As part of this study, a few selected key factors that
influence per-capita consumption were incorporated into the estimation tool. The following four key input
parameters were found to be notable: level of service, usage scenario (e.g. unrestricted versus water
restrictions), household size or number of people per household, and the geographic region, which
would mainly impact on outdoor use.
4
Per-household and per-capita consumption are affected by household size, (Beal et al., 2011; Gato et
al., 2011; Lee et al., 2012), the age of the occupants (Kenny et al., 2008; Makki et al., 2011; Willis et
al., 2009), household income (Beal et al., 2011; Kenny et al., 2008; Loh and Coghlan, 2003), the
efficiency of water use appliances (Beal et al., 2011; Gato et al., 2011; Heinrich, 2009) and the presence
of a pool and/or garden (Ferrara, 2008).
Per-capita water consumption can only be determined accurately if the number of occupants in the
household is known. Once the household size (people per household) is known, the daily household
consumption can be divided by the number of occupants to determine the per-capita consumption.
Logically, not all members of the household will use exactly the same amount of water, as each member
will have varying water use habits. However, this method provides a fairly accurate average per-capita
consumption for a household and is commonly employed in research studies (Willis et al., 2013;
Rathnayaka et al., 2015). A summary of earlier studies that reported on per-capita water consumption
is provided in Appendix A.
2.5 WATER END-USES AND ACTIVITIES
The following international studies were used in the development of the LCD Tool: Roberts (2005),
Blokker et al. (2010), Hussien et al. (2016) and Gleick (2003). Other less notable studies include Richter
(2010), Richter (2011), Hand et al. (2005), Rosenberg (2007) and Vinogradova et al. (2012). Some
notable South African studies include Jacobs and Haarhoff (2004), Van Zyl et al. (2008), Du Plessis
(2007) and Jacobs (2007).
An Australian residential end-use study conducted by Roberts (2005) placed high-resolution water
meters in 100 homes in the Yarra Valley. Water meter readings were recorded every five seconds for
a two-week period – repeated in winter and summer. Water end-uses were disaggregated from the
water meter readings using Trace Wizard software. Results from surveys conducted at the measured
households were compared to the measured water use data. This technology allows for disaggregation
accurate enough for the purpose of determining average water uses for different water end-uses.
Blokker et al. (2010) developed a stochastic end-use model to determine water demand patterns at
residential level for a one-second time scale. Statistical parameters found for frequency, intensity and
duration, as well as the penetration rate of different water end-uses, based on census data, were used
in the development of the model. Measured data was compared to the simulated results of the model,
which were found to be comparable. Hussien et al. (2016) conducted a survey of 407 households in
Duhok City, Kurdistan, Iraq, to help determine the water use consumption patterns of developing
countries. The survey consisted of 40 questions that covered household characteristics such as
household size, number of adults and children, and garden area, as well as questions pertaining to
water end-use behaviours. The results provided insight into household water use in low-, medium- and
high-income households in Duhok. Furthermore, information about the characteristics of different end-
uses was gathered. Per-capita water consumption of 241, 272 and 290 ℓ/c/d was determined for low-,
medium- and high-income houses, respectively.
Gleick (2003) provided insight into the possible effects of implementing water-reduction measures in a
Californian household. The conservation technologies that were investigated included low-flow toilets,
flow-reducing faucets and showerheads, efficient residential dishwashers and washing machines, and
drip- and precision irrigation sprinklers. The conservation policies investigated included water pricing
schemes, subsidy, rebate and financial incentive programmes, the implementation of new state and
national efficiency standards for appliances, education and public awareness programmes and water-
metering programmes. The results showed that a 30% reduction in water use could be achieved from
California’s water use in 2000.
5
Van Zyl et al. (2008) found that the guidelines published by the Council for Scientific and Industrial
Research (CSIR) (2005) prior to their revision in 2019 (DHS, 2019) only accounted for 53% of suburbs.
The study database of Van Zyl et al. (2008) comprised ~1.1 million consumption records from
48 municipalities, located in five water regions in South Africa. The proposed guidelines used only stand
area to give different confidence intervals of annual average daily demand (AADD). The study found
that households in coastal areas consume less water compared to their equally sized and valued
counterparts inland, which is consistent with the per-capita water needs suggested by the DWA (2009).
Du Plessis (2007) investigated the per-capita consumption of 57 communities in nine different

municipalities in the Western Cape. Du Plessis (2007) used bulk water usage, after treatment, divided
by the population size of the communities to provide the daily per-capita consumption. The bulk usage
gives an indication of the overall water required for everyday living, including domestic use, non-
domestic use and water leakage. The study found 10 communities with unexplainably high or low water
consumption. Removing these communities, the average water consumption was 201 ℓ/c/d, with about
15% being non-revenue water.
Jacobs (2007) conducted a study that investigated the water usage of high-density, low-income (HDLI)
households in the Western Cape. Since most HDLI properties do not contain garden areas, the water
demand from these households is considered to represent indoor water demand. The study, which was
conducted with 113 respondents, found a range of 66 to 156 ℓ/c/d, depending on household size (five
to two persons per household, respectively).
2.6 MINIMUM WATER USE
The World Health Organisation (WHO) (2003) conducted a study to determine the minimum amount of
water needed to meet basic health-related needs. The study determined the water requirement for food
preparation, hydration and basic hygiene. The study found that people who had to collect water from a
communal facility used, on average, 20 ℓ/c/d, while those who had a single tap at their dwelling used,
on average, 50 ℓ/c/d. However, for the case where multiple taps are available in a household, a value
of greater than 100 ℓ/c/d was determined.
Gleick (1996) set out to determine an absolute minimum water requirement that should be provided to
all human beings in order to meet their basic human needs (Gleick, 1996). A minimum value of 50 ℓ/c/d
was recommended, with 5 ℓ/c/d for drinking water, 20 ℓ/c/d for sanitation, 15 ℓ/c/d for bathing and 10 ℓ/c/d
for cooking and kitchen use. Gleick (1996) noted that different levels of water use are expected for
different levels of service, as well as different climatic conditions, with higher water consumption
expected for dry regions and fully connected houses with gardens.
The amount of water needed for human health, as well as the social and economic development of a
country, was researched by Chenoweth (2007). The water requirements considered both domestic and
commercial water use. However, agricultural water was excluded. The minimum water requirement for
development was determined, firstly, by investigating water use by developed countries, allowing for
the interconnected nature of industries to be taken into consideration, and later verifying this by using
a first-principles approach. The first-principles approach included investigating the hypothetical
minimum water requirements for each economic sector. A minimum water requirement for social and
economic development was determined to be 135 ℓ/c/d. Some 10 to 15 ℓ/c/d of this water is attributed
to a ±10% water loss in the system, and the remaining 120 ℓ/c/d is for domestic and commercial use.
Chenoweth (2007) noted that even though it may be theoretically feasible to meet domestic and
commercial development needs with a water use of 135 ℓ/c/d, of all the currently developed countries,
only the United Arab Emirates and Kuwait have water use less than 135 ℓ/c/d, while most “low water
use” countries reported consumption values of between 270 and 430 ℓ/c/d.
6
The international space station could be considered to represent the lowest water consumption possible,
given current technology. Consumption of a mere 1 ℓ/c/d for crew and 0.45 ℓ/d for payloads is maintained
aboard the space station (Tobias et al., 2011). Such a low consumption value is possible through an
almost closed-loop system of recycling and producing water. Water is recycled from urine and
condensation from sweat and other sources of evaporated water by collecting and treating all moisture
and returning the fluid to a potable state. In a separate closed system, water is used to produce oxygen
and hydrogen. The remaining hydrogen is then combined with CO2 to produce water and methane gas.
A summary of the water inputs and outputs for a single crew member is given in Figure 2.1.
Figure 2.1: Water inputs and outputs of the international space station (Tobias et al., 2011)
2.7 HOUSEHOLD SIZE
Household size varies notably and contributes significantly to domestic water consumption, having been
found to be the most significant factor affecting household water consumption (Rathnayaka et al., 2014).
As household size increases, its water consumption increases due to the increased occupants in the
household each requiring water. Conversely, the per-capita water consumption of a household
decreases as the household size increases. The decrease in per-capita consumption is due to many
household water uses, such as washing machines, dishwashers, cooking and irrigation, being shared
among the occupants of the household. Domestic water consumption is also affected by the age and
occupation of the members of a household. The age of the occupants has an effect on water
consumption as the activities of individuals change with age and the associated lifestyle (Browne et al.,
2014). When showers are used, children and teenagers have been found to shower for longer than
adults, thus increasing household water consumption (Mayer and DeOreo, 1999). In Germany,
household water consumption has also been found to increase with age, often because the elderly, who
are retired, spend more time in the house, which implies an increased likelihood to use water (Schleich
and Hillenbrand, 2009).
Attempts have been made to study and model the effect of an increase in household size on per-capita
consumption. Schleich and Hillenbrand (2009) determined, in a study in Germany, that with a 50%
increase in the average number of household occupants, per-capita consumption will decrease by 22%.
7
Cavanagh et al. (2002) found comparable results, while Höglund (1999) found, in a Swedish study, that
the per-capita consumption decreases as much as 27-35% with a 50% increase in household size. Jacobs
(2004) determined an equation (Equation 1.1) to model the decrease in per-capita water demand with an
increase in household size, based on studies by Edwards and Martin (1995) and Morgan (1973). A
number of international empirical studies have been conducted that have measured water consumption
for households of varying sizes (Arbués et al., 2010; Lee et al., 2012; Sadr et al., 2016; Koketso and
Emmanuel, 2017; Smith, 2010). DeOreo and Mayer (2012) compiled a review of five different end-use
studies conducted in North America, specifically, the Residential End Uses of Water Study (REUWS)
(Mayer and DeOreo, 1999), the United States Environmental Protection Agency (USEPA) combined
retrofit report (Aquacraft, 2005), the California Single Family Water Use Efficiency Study (CSFWUES)
(DeOreo, 2011), the National Survey of Functional Health Status (NSFHS), which was split into the
Standard New Home Group (SNHG) and the High Efficiency New Home (HENH) Group (DeOreo, 2011).
A graphical representation of the decrease in per-capita water demand with an increase in household size
for each study is shown in Figure 2.2.
𝑷𝑷𝑷𝑷𝑷𝑷 𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪 𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾 𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫 = −𝟔𝟔𝟔𝟔 × 𝒍𝒍𝒍𝒍(𝒅𝒅) + 𝟐𝟐𝟐𝟐𝟐𝟐 Equation 1.1
where d = household size
Typical household size for Western countries generally ranges between two and three persons per
household (PPH) (House-Peters et al., 2010; Rathnayaka et al., 2017), while the household size in urban
areas in less-developed countries ranges between two and five PPH (Jacobs and Haarhoff, 2004).
Townships or communal living areas in less-developed countries, such as South Africa, have household
sizes ranging between five and 10 PPH (Emenike et al., 2017; Jacobs and Haarhoff, 2004; Mazvimavi
and Mmopelwa, 2006). Caution should be taken with studies conducted in countries that have a mix of
townships and urban areas, as data on the household size might be skewed by studies that include both
development types.
500 Jacobs (2004) Arbués (2010)

Sadr et al. (2016) Lee et al. (2012)
450 REUWS CSFWUES
SNHG USEPA
400 HENH Smith (2010)
Per Capita Water Demand (L/c/d)
Koketso & Emmanuel (2017) Theoretical Average

350
300
250
200
150
100
50
0
0 1 2 3 4 5 6 7
Household Size
Figure 2.2: Summary of studies showing the effect of persons per household on per capita water demand
8
2.8 WATER PRICE, HOUSEHOLD INCOME AND PROPERTY VALUE
Price of water influences the volume consumed and is arguably one of the first demand-related variables
to be investigated. Hanke and Davis (1971) first recognised price of water as a demand management
measure, although the first price elasticity value for water demand was published much earlier by Metcalf
(1926). Howe and Linaweaver (1967) presented the first detailed account of price elasticity for water
demand that could be traced in the literature. Price elasticity of water is relatively inelastic (the absolute
value is smaller than 1), implying that an increase in price would decrease water demand and at the same
time lead to increased revenue. The law of demand states that price and demand are inversely related,
all other factors held constant. However, at a low price, there is a limit to the amount of water anyone
would use, even if it were free. On the other hand, there is a certain minimum quantity of water that anyone
would require, even if it were very expensive. Unfortunately, water price is an inconvenient parameter for
inclusion in estimation models.
In South Africa, the matter is further complicated by FBW allocations, non-payment, water account arrears,
short-term price fluctuations brought about by seasonal water restrictions and relatively complicated block
tariff structures. Agthe and Billings (1980) presented a detailed comparison of three different price
variables and concluded that the use of average price alone produces less accurate results than a
marginal price (highest block rate) and a “difference value” combined. Howe (1982) concluded that the
exact interpretation of the “difference value” and the rationale for the magnitude of its estimated coefficient
remain something of a mystery. Subsequent researchers included household income and property value
instead of price. Water price was also not considered as a model input in this study.
The consumer “buying power” (financial ability to pay) in relation to the actual value paid for water
influences the volume consumed by a particular consumer. Common sense suggests that household
income could be linked to water use – with higher-income homes using more water than similar homes
with lower-income occupants. However, accurate household income data is not readily available. Property
value is often used as a proxy for household income, since property value is relatively easy to obtain.
Different levels of service, as described in this study, are indirectly linked to household income and property
value. However, the service level alone does not adequately account for the impacts of household income
on water use. Although some of the levels of service, such as yard taps or CABs, may be linked quite
closely to household income, others are not. Studies show that there is a correlation between household
income and water use (Ferrara, 2008; Van Zyl et al., 2008; Beal and Stewart, 2011). The increase in water
use with increased income may be due to landscaped gardens that require more water, more water-
intensive appliances being present in the household and a smaller regard for the price of water.
2.9 GEOGRAPHY AND CLIMATE
The location of a household affects water consumption as the climate, terrain and nature of the activities
performed in the area all affect the water use habits of the household’s occupants. Climate and
temperature have been found to affect water use. Typology and the presence of an irrigated garden and
a swimming pool have the greatest impact on water use in summer, while household size and appliance
efficiency have the greatest impact on water use in winter (Rathnayaka et al., 2014). The effect of
temperature and rainfall on water use is more significant in areas with warmer climates. The effect of
warmer climates is most prominent in households with large gardens and swimming pools as their outdoor
use, which is often weather dependent, is higher (Jorgensen et al., 2009). An increase in rainfall in a
season has also been found to decrease water use as there is usually a reduction in outdoor water use
(Rathnayaka et al., 2014). The reduction in water use due to the occurrence of rainfall is more
psychological. It is not necessarily the quantity of rain that reduces outdoor water use (Martinez-Espiñeira,
2002). However, there is a threshold beyond which rainfall, as well as temperature, no longer have a
significant effect on water use. There has, however, been minimal investigation into a rainfall and
temperature threshold for different regions (Arbués et al., 2003).
9
Holiday homes are used periodically, either by the owner or by tourists who occupy the home temporarily.
The water use for these houses varies according to the season with water use during peak season being
found to be higher than water use in residential areas without holiday homes (Hadjikakou et al., 2013).
Therefore, water use in areas with a high population of holiday homes is not a true indication of residential
water use.
2.10 LEVEL OF SERVICE
Asefa et al. (2015) defines the level of service as “an informal contract between a utility and its customers
for a certain degree of inconvenience”. In other words, the level of service with regard to water is the ease
of access to water provided by water utilities. The WHO divides level of service into four categories: no
access, basic access, intermediate access and optimal access. The levels of service are defined by the
travel distance or time to the access point of clean water, or by the number of access points for higher
levels of service (WHO, 2003). Furthermore, the level of service dictates the typical water demand for a
household.
The DHS (2019) divides levels of service into access from a standpipe, yard connections and house
connections for low-income housing, cluster homes, flats and residential houses. The expected
consumption is based on the development level of the dwelling.
10
CHAPTER 3: ANALYSIS OF PER-CAPITA CONSUMPTION
3.1 OVERVIEW
The per-capita consumption tool developed as part of this project is described in this report as the Litre
per Capita per Day Tool, or simply as the LCD Tool. In line with the initial project proposal, the LCD Tool
considered the following independent variables: level of service, household size, climate region and water
usage level. Six different levels of service were considered, varying from standpipes to fully serviced
houses with outdoor use. Household size varied from one to seven people, taking five different climate
regions into consideration. Three usage levels were considered in the development of the tool. The LCD
Tool was developed in MS Excel, incorporating some Visual Basic (VB) functions, allowing the analyst to
select input values for each parameter, with the LCD Tool presenting the subsequently derived per-capita
water consumption value (ℓ/c/d).
The CSIR (2005) presented guideline values for estimating water use, which were referred to in the initial
project proposal. However, the CSIR (2005) document was revised in the period 2017-2019 and the DHS
published an updated guideline in July 2019 as the result of a project that was conducted at the same
time as the development of the LCD Tool. The project team was aware of DHS’s pending new publication,
but only gained insight into the final publication in July 2019. The development of the LCD Tool was
subsequently adapted to align with values published by DHS (2019).
The development of the LCD Tool is described in this chapter, with specific reference to the four key input
parameters and how limits were set for each parameter. The tool development was also based on three
project workshops and subsequent stakeholder input. The description presented below does not
incorporate the workshop feedback, which is described separately in the three workshop summaries.
3.2 LEVEL OF SERVICE
In line with the proposal for this study, the DHS (2019) divided the levels of service into the following
categories (refer to Table 2.1): standpipe, yard connection and house connection. The levels for yard
connections and house connections were further subdivided. The levels of service presented by DHS
(2019) were aligned with the following levels of service considered as inputs for the LCD Model:
• Standpipe (LOS 1)
• Communal ablution blocks (LOS 2)
• Yard connection (LOS 3)
• Low-cost or subsidised housing (LOS 4)
• Full house connection: indoor only (LOS 5)
• Full house connection: including outdoor use (LOS 6)
Each level of service was given a sequential number by which it could be identified in the VB code, ranging
from LOS 1 for standpipes to LOS 6 for full service with outdoor use.
3.3 HOUSEHOLD SIZE
Per-capita water consumption decreases with increased household size. A baseline per-capita water
consumption value was determined for the LCD Tool, based on the number of people in the household
for each level of service. In order to determine the per-capita consumption for each level of service and
each household size, all available measured South African data was compiled. Attention was given to
include only values that relate directly to actual individual household consumption where the number of
occupants was also known. Per-capita consumption values based on generalised information (e.g. census
or population) were excluded. The results were compared to international publications and to generalised
values. A summary of the measured data is given in Figure 3.1.
11
Figure 3.1: Measured South African per-capita consumption data for varying household sizes
The measured data was a compilation of five different data sets, including a study of residences in a gated
housing estate in Johannesburg (Ilemobade et al., 2018), data from 17 University of Stellenbosch student
homes, data from upmarket homes in Hermanus, Western Cape, and from 20 low-income houses in
Kleinmond (Pretorius et al., 2019), as well as data from a few relatively low-income households in
Eastwood, Pietermaritzburg (Smith, 2010). Once all the data had been compiled, various curve fits were
considered. The data was filtered according to the level of service in each case. The Pretorius et al. (2019)
and Smith (2010) data was found to represent a low level of service, while the Johannesburg, Hermanus
and Stellenbosch data was found to represent a high level of service. Two trendlines were fitted to the
high and low level of service data, independently. The trendlines were used as a basis to derive water
consumption values as a function of household size for LOS 6 and LOS 4, respectively, as presented
in Figure 3.2.
Figure 3.2: Per capita consumption for six levels of service and increasing household size
12
The baseline water consumption values for LOS 5 were calculated as the average of LOS 4 and LOS 6,
because no data was available for LOS 5 specifically. The per-capita consumption for standpipes (LOS 1)
was assumed to be constant over the range of household sizes – this was considered appropriate due to
the fact that water is carried from the standpipe, and each additional person would typically carry another
container (of fixed size) from the standpipe. Research is currently under way to assess water use by South
African consumers depending on standpipes, but no measured data was available at the time of this study.
The baseline water use from standpipes (LOS 1) was set equal to 20 ℓ/c/d, although the project team
initially used 25 ℓ/c/d in the model for this purpose. The value of 20 ℓ/c/d was considered appropriate,
based on input from practitioners at the project workshops, where it was pointed out that the typical
container size used to carry water was 10 or 20 ℓ. The baseline value of 20 ℓ/c/d was such that the resultant
water use for an arid region, after application of the various multiplication factors in the model, would be
24.4 ℓ/c/d – almost 25 ℓ/c/d. The baseline water consumption values for LOS 3 (yard tap) were calculated
as the average of LOS 1 and LOS 4, while LOS 2 (CABs) was, in turn, calculated as the average of values
for LOS 1 and LOS 3.
3.4 CLIMATE REGION
The climatic region in which a property is located will affect the overall water use of the property (Van Zyl
et al., 2008; Jacobs, et al., 2004). Drier regions will require more frequent irrigation and pool filling than
wetter areas for the same garden layout. Climate will have the greatest effect on households with outdoor
water use (i.e. LOS 6), as irrigation and pool use are the most affected by climate. Households without
outdoor water use may still be affected by climatic regions as households may have small sections that
need irrigation and are affected by climate (Fransolet, 2015).
The Köppen-Geiger climate classification can be used to determine the climatic conditions of an area,
based on temperature and precipitation (CSIR, 2015). The Köppen-Geiger climate classifications can be
simplified in regions based on aridity (CSIR, 2015). The CSIR (2015) defines five aridity regions: humid,
moist sub-humid, dry sub-humid, semi-arid and arid. The five aridity regions specified by the CSIR (2015)
were used for the development of this tool. The user has to select a climate region from a map in order to
identify the region number, which is entered as a model input.
Earlier guidelines for estimating water use (Jacobs et al., 2004) were used to determine the factor by
which the climatic region would affect water use for the highest level of service, LOS 5 (i.e. a fully serviced
house with outdoor use). Factors for high water use (arid regions) and low water use (humid regions) were
determined in relation to the average values. Jacobs et al. (2004) determined AADD values for increasing
stand sizes for four different regions with varying climates: Cape Town, Ekurhuleni, Windhoek and
George. Even though Windhoek is not located in South Africa, the climatic factors were considered
representative of South Africa and are, therefore, considered relevant to South African studies. Cape
Town’s AADD was considered to be representative of high water use, correlating to water use in an arid
region – with dry and hot summers. Windhoek and Ekurhuleni were representative of medium water use,
thus correlating to water use in a dry sub-humid region. George was considered to represent low water
use, correlating to water use in a humid region.
A multiplication factor of 1 was set for dry sub-humid climates for all levels of service, with an increasing
multiplication factor for increasing aridity and a decreasing multiplication factor for decreasing aridity. To
determine the LOS 6 multiplication factor for arid regions, the ratio between the AADD for Windhoek and
Ekurhuleni and the AADD for Cape Town, as calculated by Jacobs et al. (2004), was determined. For a
stand size of 2,000 m2, the AADD for Windhoek and Ekurhuleni was approximately 1,400 ℓ/d and the AADD
for Cape Town was approximately 1,800 ℓ/d. The ratio of 1,800/1,400 is 1.29. Consequently, a multiplication
factor for arid regions, for LOS 5, was chosen at 1.3. The multiplication factor for humid regions was based
on the ratio between the AADD for Windhoek and Ekurhuleni and the AADD for George.
13
For a stand size of 2,000 m2, the AADD for George was approximately 1 000 ℓ/d. The ratio of 1,000/1,400
is 0.71. To be conservative, a multiplication factor for humid regions, for LOS 6, was chosen at 0.75. The
multiplication factor for the moist sub-humid region was calculated by averaging the multiplication factors
for the humid and dry sub-humid regions. The multiplication factor for the semi-arid region was calculated
by averaging the multiplication factors for the arid and dry sub-humid climate regions. Climate was
considered to have a minimal effect on water use for LOS 1. A 10% increase and decrease in water use
was assumed for LOS 1 for arid and humid climates, respectively. Therefore, multiplication factors for arid
and humid climates were 1.1 and 0.9, respectively. The multiplication factors for the other regions were
calculated in the same manner as those for LOS 6.
Water use between each level of service does not increase in a linear manner (DHS, 2019). Therefore, a
relationship between the water use for each level of service was determined in order to interpolate
between the multiplication factor for LOS 1 and LOS 6 for the humid and arid climatic regions. The
expected water use values, as set out by the DHS (2019), were used to determine the proportional
increase in water use per service level. In order to interpolate the climate multiplication factors from LOS 1
to LOS 6, the factors were scaled in the same proportions as water use increased from LOS 1 to LOS 6.
The interpolation was only performed for humid and arid regions. The multiplication factors for moist sub-
humid and semi-arid regions were calculated in the same manner as those for LOS 1 and LOS 6, by
averaging the humid and dry sub-humid, and the arid and dry sub-humid conditions. A summary of all
multiplication factors can be found in Table 3.1. For example, the unit value of 1 (for LOS 1 and a dry sub-
humid region) would be 20 ℓ/c/d.
Table 3.1: Summary of multiplication factors for climate regions for each level of service
Description LOS ratio 1 1 2 4 8 16

Climate region Code LOS 1 LOS 2 LOS 3 LOS 4 LOS 5 LOS 6
Humid A 0.900 0.891 0.881 0.863 0.825 0.750
Moist sub-humid B 0.950 0.945 0.941 0.931 0.913 0.875
Dry sub-humid C 1.000 1.000 1.000 1.000 1.000 1.000
Semi-arid D 1.050 1.056 1.063 1.075 1.100 1.150
Arid E 1.100 1.113 1.125 1.150 1.200 1.300
Elasticity
0.00 0.00 0.00 0.10 0.20 0.30
property value
3.5 USAGE LEVEL
Usage level was used to consider the water conservation efforts of the residents of the households and/or
over-/under-use relative to that which could be considered normal – keeping in mind that normal values were
those presented in the earlier graphs for household size. Water use may also be higher in houses that have
high outdoor use due to landscaped gardens with exotic plants or with numerous fishponds or fountains. A
larger pool may also contribute to a higher water use. Therefore, a usage level multiplication factor was
considered, allowing for conservative water use and higher-than-normal water use to be considered.
It is was considered appropriate to assume constant use for standpipes. The volume of water that is
collected from a standpipe only meets basic human hygiene needs. Therefore, water use cannot be
reduced. Furthermore, since water has to be carried from the standpipe, often over long distances,
excessive water use will not be relevant to standpipe users. Therefore, the usage level multiplication factor
was only applied to LOS 2 to LOS 6. High and low water use will have the greatest effect on households
with larger stand sizes (Jacobs, 2007), which is often representative of level of service. In order to
determine the multiplication factors, for usage level, for LOS 6, the theoretical household size graphs
shown in Figure 2.2 were used. The water use for a single-person household was used to determine the
proportion between high, average and low water use. The project team tested this concept at the
workshops, appreciating that a factor derived in this manner would not be independent of other inputs.
14
This approach was considered to be the best available option for determining factors for high and low
water use scenarios.
The average theoretical water use value of 240 ℓ/c/d was set as the average use. The lowest theoretical
value of 179 ℓ/c/d was used as low water use and the highest theoretical value of 331 ℓ/c/d was used as
high water use. Therefore, the low water use ratio for LOS 6 was calculated by dividing the low water use
by the average, resulting in a multiplication factor of 0.75 (compared to the average). The high water use
ratio for LOS 6 was calculated by dividing the high water use by the average, resulting in a multiplication
factor of 1.40 (compared to the average). The multiplication factor for average water use remained
constant at 1, while the multiplication factors for high and low water use were interpolated between LOS 1
and LOS 6 in the same manner as the multiplication factors for the humid and arid climatic regions in
Section 3.4. The resultant usage level multiplication factors are summarised in Table 3.2.
Table 3.2: Multiplication factors for usage level
Usage level LOS 1 LOS 2 LOS 3 LOS 4 LOS 5 LOS 6

Low use 1.000 0.984 0.969 0.938 0.875 0.750
Average use 1.000 1.000 1.000 1.000 1.000 1.000
High use 1.000 1.025 1.050 1.100 1.200 1.400
3.6 PROPERTY VALUE
Property value was incorporated into the LCD Model. Property value has been found to be positively
correlated to water use (Van Zyl et al., 2008; Husselmann and Van Zyl, 2005). Specific value categories
were defined so that the LCD Model would find practical application in South Africa. Three stand value
categories were chosen: low income, middle income and high income houses. Specific property value
ranges were linked to each category and were based on available information in terms of house prices.
Lemanski (2010) suggested a maximum value of R300,000 for a government-subsidised house in 2008.
Government-subsidised housing was considered to represent the low-income portion of the population. The
average housing inflation over the last 10 years has been 4.5% (Lightstone Property, 2019). Considering
inflation, a maximum property value of R500,000 was determined as an upper limit for low-income
properties. Middle-class houses were classified as having a stand value ranging from R500,000 to
R1,500,000, while high-income households were classified as having a stand value greater than
R1,500,000.
The effect of stand value on water use was modelled by incorporating an elasticity value, informed by
earlier work (Husselmann and Van Zyl, 2005). The elasticity of water demand with respect to stand value
ranged between 0 and 0.5, with no clear trends reported by Husselmann and Van Zyl (2005). Considering
the most basic types of service, no impact of property value on demand was assumed for standpipes,
CABs and yard taps. The model was constructed with a property value elasticity that increases with the
higher levels of service, as described below:
• Standpipe = 0
• Communal ablution blocks = 0
• Yard connection = 0
• Low cost housing: limited inhouse connections = 0.1
• Full house connection: indoor only (e.g. flats) = 0.2
• Full house connection: including outdoor = 0.3.
15
CHAPTER 4: CONCLUSION
Limited research into per-capita water use in South Africa has been published to date. Information is
also not available in a standard format that is easily referenceable. International references are often
used as a proxy for South African conditions. As part of this study, a tool was developed that would be
more relevant to local conditions. Per-capita water use was selected as a norm for this study, as it is
the most notable driver of water use and relates to internationally applied standards. Some of the levels
of service employed in this study, and used as model inputs, are unique to South African conditions.
The study involved applied research to address the research problem by reviewing formerly published
per-capita water consumption collected by means of a comprehensive desktop review and targeted
email requests from local experts. Expert input was made possible by hosting three project workshops
in different regions of the country. Where measured household water consumption was available in
combination with the corresponding household size, the per-capita consumption was calculated from
the former values. New data was not collected, but data was rather collated from information that was
already available among the project team members and stakeholders.
An extensive literature review was completed, including ~105 references that were not specifically cited
in this document. The project team obtained 101 data points of the per-capita water use of specific
homes from four regions in the country. The actual data was used to develop the first version of the
LCD model. The MS Excel model, or LCD Tool, was developed and amended with inputs from three
workshops held in Gauteng, the Western Cape and KwaZulu-Natal. The tool was verified by comparing
results to current DHS (2019) guidelines for per-capita consumption, and results were within limits
deemed to be realistic – also based on workshop inputs.
The per-capita consumption for single-person households was relatively higher than intuitively thought, but
based on the best available data (also local data), it was accepted to be the case for single-person
households. It should be kept in mind that the average household size in South Africa varies between three
and four persons per household. Household size has a notable impact on the resultant per-capita
consumption.
The following key input parameters were included in the LCD Model: level of service, property value,
geographic region, the number of people per household and water usage level. The tool was developed
in such a manner that it is relatively easy to operate.
The tool does not account for on-site plumbing leaks (on a consumer’s property), and also ignores water
network losses. Allowance could be made for plumbing leakage, say by adding 10 to 25% to the
estimated per-capita consumption in line with some earlier studies on consumer leakage. Consideration
could also be given to water network leakage and losses by allowing for typical values (as reported
elsewhere) in addition to the per-capita consumption.
The LCD Tool is MS Excel-based and is provided separately with this report in e-format. The LCD Tool
presents a large result set, based on the various inputs. A set of 270 different result values would be
possible for each household size option (e.g. six levels of service with three usage scenarios, five regions
and three property value categories). Considering seven household size options, for one to seven persons
per household, the tool would produce 1,890 different results for per-capita water use, depending on the
selected inputs. In order to summarise the results, two tables were developed to portray the effect of
adjusting certain input values. A few example results for typical input values are given in Table 4.1.
Table 4.1 presents results for a household with average water use, average property value and located
in the dry sub-humid region, while considering the different levels of service and household sizes. The
number of persons per household was varied to portray the variability of water use per capita for the
different levels of service.
16
Table 4.1: Water use per capita for different levels of service and number of persons per household

Number of persons per household
Level of service 1 2 3 4
LOS 1 Standpipe 22 22 22 22
LOS 2 Communal ablution blocks 54 42 37 34
LOS 3 Yard connection 85 62 52 46
LOS 4 Low-cost housing: limited fixtures 163 111 89 76
LOS 5 Full house connection (indoor) 275 198 163 143
LOS 6 Full house connection (including outdoor) 407 300 251 221
For Table 4.2, a three-person household with average water use and an average property value was
selected for each level of service. The climate region was varied in this case to portray the variability of
water use per capita for the different levels of service.
Table 4.2: Water use per capita for different levels of service vs climate regions

Climate regions
Moist Dry
Semi-
Level of service Humid sub- sub- Arid
arid
humid humid
LOS 1 Standpipe 20 21 22 23 24
LOS 2 Communal ablution blocks 33 35 37 39 41
LOS 3 Yard connection 46 49 52 55 58
LOS 4 Low-cost housing: limited fixtures 77 83 89 96 102
LOS 5 Full house connection (indoor) 135 149 163 180 196
LOS 6 Full house connection (including outdoor) 188 220 251 289 326
The results are most sensitive to the level of service and number of people per household. It is therefore
essential that users ensure that these parameters are populated with insight into actual conditions. The
findings presented in the following chapter were derived based on the analyses and scenario testing of
the model and tool.
17
CHAPTER 5: FURTHER RESEARCH
This study presented a novel way of dealing with per-capita water use estimates. Further research and
development, with subsequent knowledge dissemination, would help to establish a sound approach to
per-capita use in South Africa. The tool presented in this study was based on the best available data
and previous knowledge. The limited data and available research at the time of study suggests that
future improvements could improve the model.
It would therefore be beneficial to investigate the following aspects further:
• Install and monitor more water meters at households across the country, and record the number
of people per individual home.
• Develop a method where household per-capita use can be accurately derived from census and
Municipal Treasury data.
• Expand the LCD Tool to incorporate water loss and peak water use so that actual peak flows in water
reticulation pipes could be estimated in a similar manner to the per-capita consumption in this case.
• Further develop a geospatial data set specific for water use that could be applied to the model.
• Develop the tool into a cellular phone application for ease of use.
• Describe model inputs as stochastic parameters and subsequently determine the sensitivity of
specific parameters to model outputs, and express results in terms of confidence intervals.
The further development of the tool would improve the model accuracy and would allow for increased
acceptance and uptake of the approach in industry.
18
BIBLIOGRAPHY
AGTHE and BILLINGS (1980)
AQUACRAFT (2005) USEPA combined retrofit report. Water and Energy Savings, Aquacraft, Boulder,
CO and US Environmental Protection Agency, Washington, DC, USA.
ARBUÉS F, GARCIA-VALIÑAS MA and MARTINEZ-ESPIÑEIRA R (2003) Estimation of residential

water demand: A state-of-the-art review. Journal of Socio-Economics 32 81‒102.
ARBUÉS F, VILLANUA I and BARBERAN R (2010) Household size and residential water demand: An
empirical approach. The Australian Journal of Agricultural and Resource Economics 54 61‒80.
ASEFA T, ADAMS A and WANAKULE N (2015) A level‐of‐service concept for planning future water
supply projects under probabilistic demand and supply framework. Journal of the American Water
Resources Association 51 (5) 1272‒1285.
ATHURALIYA et al. (2012)
BEAL C and STEWART A (2011) South east Queensland residential end use study: Final report. Urban
Water Security Research Alliance, Queensland, Australia.
BEAL C, STEWART R, HUANG T and REY E (2011) Applying smart metering technology to
disaggregate residential water end uses in South-East Queensland. Water: Journal of the Australian
Water Association 38 (1) 80‒84.
BLAINE TW, CLAYTON S, ROBBINS P and GREWAL PS (2012) Homeowner attitudes and practices
towards residential landscape management in Ohio, USA. Environmental Management 50 257–271.
https://doi.org/10.1007/s00267-012-9874-x
BLOKKER EJM, VREEBURG JHG and VAN DIJK JC (2010) Simulating residential water demand with a
stochastic end-use model. Journal of Water Resources Planning and Management 136 (1) 19‒26.
BRADLEY D (1997) Heath aspects of water supplies in tropical countries. In: CHICHESTER U, Water,
wastes and health in hot climates, Wiley.
BROWNE AL, PULLINGER M, MEDD W and ANDERSON B (2014) Patterns of practice: A reflection
on the development of quantitative/mixed methodologies capturing everyday life related to water
consumption in the UK. International Journal of Social Research Methodology 17 (1) 27‒43.
CAVANAGH SM, HANEMANN WM and STAVINS RN (2002) Muffled price signals: Household water
demand under increasing-block prices. FEEM Working Paper No. 40.
CHENOWETH J (2007) Minimum water requirement for social and economic development.
Desalination 229 (1‒3) 245‒256.
CLAYTON S (2007) Domesticated nature: Motivations for gardening and perceptions of environmental
impact. Journal of Environmental Psychology 27 (3) 215–224.
https://doi.org/10.1016/j.jenvp.2007.06.001
CONRADIE DCU (2013) Appropriate passive design approaches for the various climatic regions in
South Africa. The green building handbook, the essential guide 5 101‒117.
19
CONRADIE DCU (2012) Designing for South African climate and weather. The green building
handbook, the essential guide 4 181‒195.
COUNCIL FOR SCIENTIFIC AND INDUSTRIAL RESEARCH (CSIR) (2015) Climate indicators ‒
aridity. http://stepsa.org/climate_aridity.html.
COUNCIL FOR SCIENTIFIC AND INDUSTRIAL RESEARCH (CSIR) (2005) Guidelines for human
settlement planning and design (2nd ed.), CSIR, Pretoria, South Africa.
CROUS et al. (2013)
DEOREO WB (2011) Analysis of water use in new single family homes. Aquacraft, Boulder, CO, USA.
DEOREO WB and MAYER PW (2012) Insights into declining single-family residential water demands.
American Water Works Association 104 (6).
DEPARTMENT OF WATER AFFAIRS (DWA) (2009) Development of reconciliation strategies for all
towns in the southern planning region: Inception report, Pretoria, South Africa. Prepared by Umvoto
Africa in association with Aurecon on behalf of the Directorate: National Water Resource Planning,
Pretoria, South Africa.
DEPARTMENT OF HUMAN SETTLEMENTS (DHS) (2019) The neighbourhood planning and design
guide, creating sustainable human settlements. DHS, Pretoria, South Africa.
DIAS et al. (2018)
DU PLESSIS JA (2007) Benchmarking water use and infrastructure based on water services
development plans for nine municipalities in the Western Cape. Journal of the South African Institution
of Civil Engineering 49 (4) 18‒27.
DU PLESSIS JL and JACOBS HE (2018) Analysis of water use by gated communities in South Africa.
Water SA 44 (1) 130‒135. http://dx.doi.org/10.4314/wsa.v44i1.15.
EDWARDS K and MARTIN L (1995) A methodology for surveying domestic water consumption. Water
and Environmental Management 9 447‒488.
EMENIKE CP, TENEBE IT, OMOLE DO, NGENE BU, ONIEMAYIN BI, MAXWELL O and ONOKA BI
(2017) Accessing safe drinking water in sub-Saharan Africa: Issues and challenges in south-west
Nigeria. Sustainable Cities and Society 30 263‒272.
ESREY SA, FEACHEM RF and HUGHES JM (1985) Interventions for the control of diarrhoeal diseases
among young children: Improving water supplies and excreta disposal facilities. Bulletin of the World
Health Organization 63 (4) 757‒772.
FERRARA I (2008) Residential water use. OECD Journal: General Papers 2.
FRANSOLET GCG (2015) Universal design for low-cost housing in South Africa: An exploratory study
of emerging socio-technical issues. Cape Peninsula University of Technology, Cape Town, South
Africa.
GATO S, JAYASURIYA N and ROBERTS P (2011) Understanding urban residential end uses of water.
Water Science and Technology 64 (1) 36‒42.
GLEICK (1996)
20
GLEICK PH (2003) Waste not, want not: The potential for urban water conservation in California. Pacific
Institute for Studies in Development, Environment, and Security, Oakland, CA, USA.
GURAGAI et al. (2018)
HADJIKAKOU M, CHENOWETH J and MILLER G (2013) Estimating the direct and indirect water use of
tourism in the eastern Mediterranean. Journal of Environmental Management 114 548‒556.
HAND M, SHOVE E and SOUTHERTON D (2005) Explaining showering: A discussion of the material,
conventional, and temporal dimensions of practice. Sociological Research Online 10 (2).
HANKE and DAVIS (1971)
HAY et al. (2012)
HEINRICH M (2009) Auckland water use study – monitoring of water end uses. SB10, New Zealand.
HÖGLUND L (1999) Household demand for water in Sweden with implications of a potential tax on
water use. Water Resources Research 35 3853‒3863.
HOUSE-PETERS L, PRATT B and CHANG H (2010) Effects of urban spatial structure,

sociodemographics, and climate on residential water consumption in Hillsboro, Oregon. Journal of the
American Water Resources Association 46 (3) 461‒472.
HOWE (1982)
HOWE and LINAWEAVER (1967)
HUSSIEN WA, MEMON FA and SAVIC DA (2016) Assessing and modelling the influence of household
characteristics on per capita water consumption. Water Resources Management 30 (9) 2931‒2955.
HUSSELMANN and VAN ZYL (2005)
ILEMOBADE AA, BOTHA BE and JACOBS HE (2018). Sorting high temporal resolution end-use data
– a pleasurable headache. Paper 128. 1st International WDSA/CCWI 2018 Joint Conference,
23‒25 July 2018, Kingston, Ontario, Canada.
JORDÁN-CUEBAS et al. (2018)
JACOBS (2004)
JACOBS HE (2007) The first reported correlation between end-use estimates of residential water
demand and measured use in South Africa. Water SA 33 (4) 549‒558.
JACOBS HE, GEUSTYN LC, LOUBSER BF and VAN DER MERWE B (2004) Estimating residential water
demand in Southern Africa. Journal of the South African Institution of Civil Engineering 46 (4) 2‒13
JACOBS HE and HAARHOFF J (2004) Application of a residential end-use model for estimating cold
water and hot water demand, wastewater flow and salinity. Water SA 30 (3) 305‒316.
JORGENSEN B, GRAYMORE M and O'TOOLE K (2009) Household water use behaviour: An

integrated model. Journal of Environmental Management 91 227‒236.
KOKETSO KN and EMMANUEL MN (2017) Water demand modelling and end-use study on
households within Johannesburg South Africa. University of Witswaterstrand, Johannesburg, South
Africa.
21
LEE D, PARK N and JEONG W (2012) End-use analysis of household water by metering: The case
study in Korea. Water and Environment Journal 26 455‒464.
LEMANSKI (2010)
LIGHTSTONE PROPERTY (2019)
LOH and COGHLAN (2003)
KENNY D, GOEMANS C, KLEIN R, LOWREY J and REIDY K (2008) Residential water demand
management: Lessons from Aurora, Colorado. Journal of the American Water Resources Association
44 (1) 192‒207.
MAKKI et al. (2011)
MANZUNGA and MACHIRIDZA (2005)
MARTINEZ-ESPIÑEIRA R (2002) Residential water demand in the northwest of Spain. Environmental

and Resource Economics 21 (2) 161‒187.
MAYER PW and DEOREO WB (1999) Residential end uses of water. AWWA Research Foundation
and American Water Works Association, Denver, CO, USA.
MAZVIMAVI D and MMOPELWA G (2006) Access to water in gazetted and ungazetted rural
settlements in Ngamiland, Botswana. Physics and Chemistry of the Earth 31 713‒722.
MEAD (2008)
METCALF (1926)
MORGAN WD (1973) Residential water demand: The case from micro data. Water Resources
Research 9 (4) 1065‒1067.
PARKER and WILBY (2013)
PRETORIUS D, CROUCH ML and JACOBS HE (2019) Diurnal water use patterns for low-cost houses:
a South African case study. Journal of Water, Sanitation and Hygiene for Development (in press).
RATHNAYAKA K, MAHEEPALA S, NAWARATHNA B, GEORGE B, MALANO H, ARORA M and

ROBERTS P (2014) Factors affecting the variability of household water use in Melbourne, Australia.
Resources, Conservation and Recycling 92 85‒94.
RATHNAYAKA K et al. (2015)
RATHNAYAKA K, MALANO H, ARORA M, GOERGE B, MAHEEPALA and NAWARATHNA, B (2017)

Prediction of urban residential end-use water demands by integrating known and unknown water
demand drivers at multiple scales II: Model application and validation. Resources, Conservation and
Recycling 118 1‒12.
RICHTER CP (2010) Automatic dishwashers: Efficient machines or less efficient consumer habits.
International Journal of Consumer Studies 34 228‒234.
RICHTER CP (2011) Usage of dishwashers: Observation of consumer habits in the domestic

environment. International Journal of Consumer Studies 35 180‒186.
22
ROBERTS P (2005) Yarra Valley Water 2004 residential end use measurement study. Demand
Forecasting Manager, Yarra Valley Water, Melbourne, Australia.
ROMA E, BUCKLEY C, JEFFERSON B and JEFFREY P (2010) Assessing users' experience of shared
sanitation facilities: A case study of community ablution blocks in Durban, South Africa. Water SA 36 (5)
589‒594.
ROSENBERG DE (2007) Probabilistic estimation of water conservation effectiveness. Journal of Water

Resources Planning and Management 133 (1) 39‒49.
SADR SMK, MEMON FA, JAIN A, GULATI S, DUNCAN A, HUSSEIN WA, SAVIC D and BUTLER D
(2016) An analysis of domestic water consumption in Jaipur, India. British Journal of Environment and
Climate Change 6 (2) 97‒115.
SCHLEICH J and HILLENBRAND T (2009) Determinants of residential water demand in Germany.

Ecological Economics 68 1756‒1769.
SHABAN and SHARMA (2007)
SIVAKUMARAN and ARAMAKI (2010)
SMITH J A (2010) How much water is enough? Domestic metered water consumption and free basic
water volumes: The case of Eastwood, Pietermaritzburg. Water SA 36 (5) 595‒606.
THIEL (2014)
THOMPSON J, PORRAS IT, TUMWINE JK, MUJWAHHUZI MR, KATUI-KATUI M, JOHNSTONE N

and WOOD L (2001) Drawers of water II: 30 years of change in domestic water use and environmental
health in East Africa. International Institute for Environment and Development (IIED), London, UK.
TOBIAS B, GARR II JD and ERNE M (2011) International space station water balance operations. 41st
International Conference on Environmental Systems, Portland, OR, USA.
VAN ZYL HJ, ILEMOBADE AA and VAN ZYL JE (2008) An improved area-based guideline for domestic
water demand estimation in South Africa. Water SA 34 (3) 381‒391.
VINOGRADOVA M, RAMSSON T and HAWKES A (2012) Technology brief R08: Dishwashers. Energy
Technology Network.
WILLIS R, STEWART RA, PANUWATWANICH K, CAPATI B and GIURCO D (2009) Gold coast
domestic water end use study. Water: Journal of the Australian Water Association 36 79–85.
WILLIS RM, STEWART RA, PANUWATWANICH K, WILLIAMS PR and HOLLINGSWORTH AL (2011)

Quantifying the influence of environmental and water conservation attitudes on household end use
water consumption. Journal of Environmental Management 92 (8) 1996–2009.
https://doi.org/10.1016/j.jenvman.2011.03.023
WILLIS et al. (2013)
WORLD HEALTH ORGANISATION (WHO) (2003) Domestic water quantity, service level and health.
World Health Organisation, Geneva, Switzerland.
23
ANNEXURE A – PER CAPITA WATER USE STUDIES
Data Micro/ Reported water Number of

Citation Country Location Level of service PPH Type of study Restrictions (Yes/No)
date Macro use (ℓ/c/d) households
South Africa Johannesburg 326
South Africa Cape Town Full house 352
Jacobs and Haarhoff
2004 Macro connection 3,00 Model No
(2004) South Africa George 246
(indoor and outdoor)
Namibia Windhoek 425
Kliprand 25 Empirical study Yes – permanent restrictions
Vredenburg 97
Hermon 98
Riebeek West 98
Raithby 103
Rietpoort 112
Riebeek Kasteel 113
Nuwerus 132
Bitterfontein 135
Paternoster 144
Pniel 146
Hopefield 147
Eendekuil 157
Aurora 160
Kylemore 170
Abbotsdale 173
2002- Chatsworth 173
Du Plessis (2007) South Africa Macro Mixed
2003 Gouda 175 Empirical study No
Koringberg 177
Piketberg 178
Kalbaskraal 178
Darling 181
Riverland 182
Elands Bay 194
Lutzville 195
Moorreesburg 195
Clanwilliam 200
Saldanha 201
Koekenaap 206
Klapmuts 213
Ebenhaezer 218
Velddrif 220
Tulbagh 222
Vanrhynsdorp 226
24

Calvinia 244
Strandfontein 249
Goedverwacht 251
Doringbaai 252
Graafwater 262
Wittewater 264
Citrusdal 266
Wellington 275
Porterville 277
Redelinghuys 284
Jamestown 287
Malmesbury 295
Saron 300
Franschoek 303
Paarl 321
Dwarskerbos 404
Klawer 407
Lamberts Bay 409
St Helena Bay 429
Langebaan 442
Stellenbosch 445
Vredendal 497
Yzerfontein 952
Leipoldtville 1479
Delhi 78
Mumbai 90
Kolkata Mixed: 116
Shaban and Sharma Supply based
2001 India Hyderabad Macro low cost to full house 96 No
(2007) estimate
Kanpur connections 77
Ahmedabad 95
Madurai 88
Full house
Rathnayaka et al. Yarra Valley, Empirical
2003 Australia Micro connection 238 3,02 Yes, Stage 1
(2014) Melboune study
Full house
Rathnayaka et al. Yarra Valley, Empirical
2011 Australia Micro connection 124 3,16 Yes, Stage 1
(2014 Melboune study
Full house
183 Survey and
connection: indoor
Sadr et al. (2016) 2015 India Jaipur Macro empirical No
Full house study
215
connection: outdoor
2002- Full house
Lee et al. (2012) Korea Nationwide Micro 159 Empirical study 146 No
2006 connection: indoor
25

Full house
141
connection: outdoor
United States and Full house Empirical
Mayer et al. (1999) 1996 Nationwide Micro 236 3,00 1,188 No
Canada connection: outdoor study
Full house Empirical
DeOreo (2011) 2007 United States California Micro 204 3,00 780 No
connection: outdoor study
Arizona
California
Colorado
2006- Full house Empirical
DeOreo (2011) United States Floria Micro 167 3,00 240 No
2008 connection: outdoor study
Nevada
Oregon
Utah
Loh and Coghlan 1998- Full house Empirical
Australia Perth Micro 335 120 No
(2003) 2001 connection: outdoor study
Mead (2008) 2008 Australia Toowoomba Micro 112 10
connection: indoor study No
Yarra Valley, Full house Empirical
Roberts (2005) 2004 Australia Micro 226 100
Melboune connection: outdoor study No
Willis et al. (2013) 2008 Australia Gold coast Micro 157 151
connection: outdoor study
Sivakumaran & Limited household Consumer
2004 Sri Lanka Trincomalee Micro 139 4,70 285
Aramaki (2010) connection survey No
Full house
131 194
2005- Eastwood, connection: outdoor Empirical No
Smith (2010) South Africa Micro
2007 Pietermaritzburg Limited household study
89 4,10 34 No
connection
Thiel (2014) 2013 The Netherlands Amsterdam Micro 119 1,349 No
connection: indoor study
Athuraliya et al. Yarra Valley, Full house Empirical
2012 Australia Micro 93 2,60 100 No
(2012) Melboune connection: indoor study
Athuraliya et al. Yarra Valley, Full house Empirical
2010 Australia Micro 105 2,60 100 No
(2012) Melboune connection: indoor study
Jordán-Cuebas et al. 2011- Full house Empirical
United states New York Micro 222 30 No
(2018) 2013 connection: indoor study
Adelaide, Western Full house
190
Cape connection: outdoor
Bedford, Full house
109
Western Cape connection: outdoor
Alice, Full house
Hay et al. (2012) 2008 South Africa Macro 193 Meter data No
Eastern Cape connection: outdoor
Stutterheim, Eastern Full house
245
Cape connection: outdoor
Mthatha, Full house
263
Eastern Cape connection: outdoor
26

Tulbagh, Full house
309
Beaufort West, Full house
206
Ashton, Full house
432
Plettenberg Bay, Full house
281
Bitterfontein, Full house
123
Limited household
Mabelreign 226 6,00 8
connection
Limited household
Mt. Pleasant 353 5,00 6
connection
Limited household
Marlborough 167 8,00 4
connection
Limited household Metered data
Kuwadzana 58 24,00 4
Manzunga and connection collected
2003 Zimbabwe Micro No
Machiridza (2005) Limited household through
Glen Norah 105 8,00 questionnaires 12
connection
Limited household
Budiriro 108 9,00 10
connection
Limited household
Tafara 69 10,00 6
connection
Limited household
Mabvuku 29 10,00 4
connection
East England
1992- Full house
Parker & Wilby (2013) UK Lincoln Micro 169 Model 100 No
2006 connection: indoor
Ruthamford
Rathnayaka et al. Yarra Valley Full house Empirical
2010 Australia Micro 113 3,10 117 Yes – Stage 1
(2015) City connection: outdoor study
Rathnayaka et al. Yarra Valley Full house Empirical
2012 Australia Micro 115 3,10 117 Yes – Stage 2
(2015) City connection: outdoor study
2015- Full house Empirical
Dias et al. (2018) Brazil Jointville 102 3,171 No
2016 Micro connection: indoor study
Mixed: delivered tank
Empirical
Guragai et al. (2018) 2016 Nepal Kathmandu Valley water to full indoor 56 6,40 28 Intermittent water supply
study
Micro house connections
Full house
274
connection: outdoor 7,04
Hussien et al. (2016) 2015 Iraqi Kurdistan Duhok Micro Survey 407 No
Full house
247,0
connection: indoor
27
ANNEXURE B – LCD TOOL USER GUIDE
Overview
The LCD Tool operates in the MS Office environment in MS Excel. All five input values are entered via
drop-down boxes and the result is displayed on the same sheet as the inputs, as discussed in this user
guide.
At least one input value –the level of service – is required to obtain a crude estimate of water use. In
order to obtain the most accurate estimate, all five input values are required: level of service, property
value, geographic region, number of persons per household and a relative indication of water usage
level (low, average or high use; where low use could represent consumers who are subjected to water
restrictions, for example).
The following cases are dealt with in this user guide:
• Case 1: The most basic case, where only the level of service is known.
• Case 2: All inputs are known.
Case 1: One input known – level of service
The LCD Tool is able to provide an estimate of per-capita water use based on only one input, using
typical (average) values for the other inputs. Refer to Figure B-1 and follow the steps outlined below to
obtain a crude estimate based only on the level of service:
• In the drop-down box marked “A”, select the level of service: standpipe, communal ablution block,
yard connection, low-cost housing with limited indoor use (no outdoor use), full house connection
with indoor use only (e.g. flats), or full house connection including outdoor use.
• In the drop-down box marked “B”, select “Increased spending capacity (R500k-MR1), which
represents the average value used in the model.
• In drop-down box “C”, select “Dry sub-humid”, which represents the average value used in the
model.
• In drop-down box “D”, select “3” for the number of persons in the household, which represents a
typical average value for South African homes.
• In drop-down box “E”, select “Average use”.
28
A
B
C
D
E
Figure B-1: Input section of LCD Tool
Case 2: All inputs are known
The LCD Tool is able to provide improved estimates of per-capita water use if all the inputs are known.
Refer to Figure B-1 and follow the steps outlined below to obtain the estimate based only on all known
inputs:
• In the drop-down box marked “A”, select the level of service: standpipe, communal ablution block,
yard connection, low-cost housing with limited indoor use (no outdoor use), full house connection
with indoor use only (e.g. flats), or full house connection including outdoor use.
• In the drop-down box marked “B”, select the appropriate property value, selecting one of the three
available choices. Note that property value has no impact on water use from standpipes.
• In drop-down box “C”, select the geographic region, using the map in Figure B-3 (CSIR, 2015) to
find the approximate region type, based on location.
• In drop-down box “D”, select the typical number of persons in the household, between one and seven.
Notes:
Household size has no impact on water use from standpipes.
For all other levels of service, the resulting water use decreases sharply from one to three persons
per household, which is supported by data presented in this report and by international studies.
• In drop-down box “E”, use best judgement to select the appropriate usage level by considering low
use, average use or high use. If in doubt, use the average use scenario.
Results
Results are presented on the same sheet as the inputs. Refer to Figure B-2, showing the results for Case 1
and a full consumer connection with indoor use only. Note that the result was adjusted for property value,
because the baseline value for property value is the lowest value category (and for Case 1, the middle
property value category was selected).
29
Figure B-2: Results of LCD Tool for Case 1 (example case shown)
Figure B-3: Map for selecting geographic regions (CSIR, 2015)
30

Water Usage

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Water Usage

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Water Usage

Uploaded by

Copyright:

Available Formats

ESTABLISHING A HOUSEHOLD

WATER CONSUMPTION BENCHMARK

to the Water Research Commission

JL du Plessis, HE Jacobs, A Ilemobade and M Crouch

Bernoulli (Pty) Ltd

Report No. 2980/1/20

orders@wrc.org.za or download from www.wrc.org.za

Estimated water use (ℓ/c/d)

AADD Annual average daily demand

EXECUTIVE SUMMARY ....................................................................................................................... iii

CHAPTER 2: LITERATURE REVIEW

2.1 DOMESTIC WATER USE

Land use Persons Typical annual Annual average

2.2 CATEGORIES OF WATER USE AND LEVELS OF SERVICE

2.3 WATER LOSS AND LEAKAGE

2.4 FACTORS AFFECTING WATER USE

2.5 WATER END-USES AND ACTIVITIES

Du Plessis (2007) investigated the per-capita consumption of 57 communities in nine different

2.6 MINIMUM WATER USE

2.7 HOUSEHOLD SIZE

𝑷𝑷𝑷𝑷𝑷𝑷 𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪𝑪 𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾𝑾 𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫 = −𝟔𝟔𝟔𝟔 × 𝒍𝒍𝒍𝒍(𝒅𝒅) + 𝟐𝟐𝟐𝟐𝟐𝟐 Equation 1.1

where d = household size

500 Jacobs (2004) Arbués (2010)

Koketso & Emmanuel (2017) Theoretical Average

2.8 WATER PRICE, HOUSEHOLD INCOME AND PROPERTY VALUE

2.9 GEOGRAPHY AND CLIMATE

2.10 LEVEL OF SERVICE

CHAPTER 3: ANALYSIS OF PER-CAPITA CONSUMPTION

3.2 LEVEL OF SERVICE

3.3 HOUSEHOLD SIZE

3.4 CLIMATE REGION

Description LOS ratio 1 1 2 4 8 16

3.5 USAGE LEVEL

Table 3.2: Multiplication factors for usage level

Usage level LOS 1 LOS 2 LOS 3 LOS 4 LOS 5 LOS 6

3.6 PROPERTY VALUE

Estimated water use (ℓ/c/d)

Estimated water use (ℓ/c/d)

CHAPTER 5: FURTHER RESEARCH

It would therefore be beneficial to investigate the following aspects further:

AGTHE and BILLINGS (1980)

ARBUÉS F, GARCIA-VALIÑAS MA and MARTINEZ-ESPIÑEIRA R (2003) Estimation of residential

ATHURALIYA et al. (2012)

CROUS et al. (2013)

DIAS et al. (2018)

FERRARA I (2008) Residential water use. OECD Journal: General Papers 2.

GURAGAI et al. (2018)

HANKE and DAVIS (1971)

HAY et al. (2012)

HOUSE-PETERS L, PRATT B and CHANG H (2010) Effects of urban spatial structure,

HOWE and LINAWEAVER (1967)

HUSSELMANN and VAN ZYL (2005)

JORDÁN-CUEBAS et al. (2018)

JORGENSEN B, GRAYMORE M and O'TOOLE K (2009) Household water use behaviour: An

LIGHTSTONE PROPERTY (2019)

LOH and COGHLAN (2003)

MAKKI et al. (2011)

MANZUNGA and MACHIRIDZA (2005)