Addressing stakeholder conflicts in rural South Africa using a water
supply model
Floortje d'Hont}, Jai Clifford-Holmes? and Jill Slinger!
1Faculty of Technology, Policy and M Delft University of Technology, Jaffalaan 5, 2628 BX Delft,
Netherlands (j.h.slinger@tudelft.nl )
2Institute for Water Research, Rhodes University, Old Geology Building, Artillery Road, Grahamstown,
South Africa
Abstract
A system dynamics modelling approach is adopted to deepen understanding of the effects of
operational management on the performance of the Greater Kirkwood water supply system in
South Africa. Currently, the interrupted operation of the system has led to perceptions of
systemic social injustice on the part of citizens and contention between the municipality
responsible for supplying water to the citizens and the Water User Association responsible for
delivering untreated water to the municipality. The model is used as a means of exploring the
technical constraints of the water supply system, and of supporting discussions between
stakeholders about contentious issues, and the ways they can address these issues. Research on
the utility of the model as a means of supporting strategic conversations between stakeholders
(cf. Howick & Eden, 2010) is ongoing
Key words: water management; socio-technical system; conflict resolution; strategic
conversations; soft systems methodology
1. Introduction
Following the apartheid era, the democratically-elected government of the Republic of South
Africa undertook to ensure that all citizens had access to basic water services. This was a
significant challenge given that the water resource and water services were no different from
the other resources and services in South Africa, which were inequitably distributed across
racial groups in a purposeful and designed manner. In particular, the technical supply of water
services was primarily designed for the white minority. From 1996 onwards numerous changes
at a national level resulted in a complete restructuring of water policy and legislative
frameworks. The changes relevant to water services include: the Constitution of 1996, which
positions access to water as a basic human right for all; the Water Services Act of 1997, which
separates provision and regulation in a decentralised manner; the National Water Act of 1998,
which legislates a basic human needs reserve that comes before all other allocations in the
country, and the policy of free basic water, which was formalised in 2001.The Constitution
differentiates three spheres of government, namely national, provincial and local government
(Juta’s Statutes Editors, 2010), with the responsibility for service delivery residing in the ‘local
government’, or municipal, sphere. One of the service delivery duties of the local government
is to ensure the provision of water and sanitation services to all users within a municipal
jurisdiction. The Sundays River Valley Municipality (SRVM), located within the Lower
Sundays River Valley sub-catchment, within the Eastern Cape province of the Republic of
South A frica (Figure 1), is struggling with this task.
Page 1 of 12
_Sao With no large urban settlement, and a
\ combination of multiple small towns and
- commercial farming, the SRVM is a primarily
i mural or prototypical ‘Category 3’
municipality. As of 2009, 111 of the 278
municipalities in South Africa are classified in
ch y ‘ this category (World Bank, 2009: 10-11). As
Soe such, the SRVM does not stand alone in South
\ : Africa in its struggle with the provision of water
and sanitation services to citizens living within
q the municipal boundaries. Nor is it alone in
a Sundaye Rver valley Municwaity having to deal with contentious issues related to
Af "e%5" Taam the historical design and operation of its water
supply system.
Figure 1: Location of the Sundays River Valley
Municipality within South Africa
In this paper, we follow Howick and Eden (2010) who argue that quantitative modelling can
serve to support ‘strategic conversations’ between actors, first developing a system dynamics
model of a water supply system run by the Sundays River Valley Municipality and then
exploring the utility of the model in such a context. We demonstrate its use as a neutral tool to
support strategic conversations in a multi-actor space, where actors are historically divided and
still contending with one another.
2. Method
The recognition of the variety and conflicting interests of the stakeholders involved and the
lack of clarity about the effect of operational procedures on the security of the water supply,
led to the decision to adopt a system dynamics modelling approach to the seemingly intractable
water supply problems of Greater Kirkwood. The system dynamics method was designed for
application in socio-technical systems (Lane 2000), with the aim of clarifying and quantifying
the interactions between elements of a physical system and the humans intervening in, or
operating the system. Stakeholders are able to locate themselves and their actions within the
“stocks” and “flows” used to characterize the system, and so experience the method as
enhancing communication (Stave 2003). An added advantage of system dynamics for this
particular situation is that detailed and accurate data on the water supply system (e.g. the
evolution in water demand over time, variations in the water treatment pumping capacity and
the storage reservoirs fluctuations over time) are not prerequisite to modelling. Instead a deep
knowledge of the design capacity of the system, and experienced behaviour over time is
required. Whereas detailed data were not always available, information on system behaviour
was readily obtained from the following sources:
— Technical reports of the Sundays River Valley Municipality, including the Water
Services Development Plans (Sundays River valley Municipality 2010a, 2010b, 2011)
— Information pamphlets outlining how the Water User Association water supply system
works and the annual reports of the WUA (WUA 2007, 2011, 2012).
— Three workshops on the ‘turnaround of the SRVM water services’ (31 November to |
December 2011, 10 October 2012 and 28 February 2013), attended by representatives
from the national Department of Water Affairs, the SRVM and the WUA.
Accordingly, after first characterizing the situation of the Sundays River Valley Municipality
and its water supply at catchment scale, we chose to focus on the complex, real world problem
situation of the Greater Kirkwood water supply and represent this as a system diagram (Figure
Page 2 of 12
3, section 3). Next, the system diagram is specified as a system dynamics model in VenSim
PRO Version 6.0a-1. To corroborate the physical system understanding, three site visits to the
Greater Kirkwood area were undertaken by one of the authors in the period January to
November 2012. Presently, the model is in use as a means of supporting discussions between
stakeholders about contentious issues, and about the ways they can address these issues.
Research aimed at evaluating the utility of the model as a means of supporting strategic
conversations between stakeholders (cf. Howick & Eden, 2010) is ongoing.
3. Lower Sundays River Valley, its water supply situation and involved actors
The SRVM has a population of 54 500 people living in 14 700 households (Statistics South
Africa, 2012: 2; 27). Agriculture provides 48% of the employment in the area with tourism and
community services accounting for a significant portion of the remaining 52% of the
employment (SRVM WSDP, 2010, D: 4). Unemployment and dependency on social grants are,
however, endemic: an average of 47% of the SRVM population has a household income of less
than R800 ($85/£60) per month, with unemployment estimated to be as high as 44% (SRVM
WSDP, 2010, D: 10).
A schematic for the water supply system of the Lower Sundays River Valley is given in Figure
2. The three primary actors involved in the supply of water to the Lower Sundays River Valley
area are the Department of Water Affairs, the Lower Sundays River Water User Association
and the Sundays River Valley Municipality.
Page 3 of 12
Managing
Authorities
& Key:
Water transfers aiaciaaas
iment
from the Orange of Water
sf River Affairs
. Cos
Dam CO Dams
‘ ‘ Lower
Commercial farms
Korhaans Sundays
Weir LSRV Emerging farmers River
; Water User
rural users Rural commercial Association
Rural domestic —
Sunday,
Bridge ‘Warts
SRVM 3 SRVM2
Sundays
River Valley
Municipality
reservoirs (SRVM)
Domestic
Q emer
users
Scheme
(PBS)
Urban
domestic
> Potable
water
bu e —> = Raw water
— Pipelines
Figure 2: Schematic of the water supply system in the Lower Sundays River Valley area. The focus
for the system dynamics model is demarcated by a dashed black box.
‘WUA = Water User Association;
The Department of Water Affairs manages the inter-basin transfer scheme, which delivers
untreated water from the Orange River system some 300 kilometres away, via a series of
tunnels, pipelines, canals and interconnecting rivers (not shown). Downstream of the
Darlington Dam, the Lower Sundays River Water User Association (WUA) manages the bulk
water supply system, which delivers untreated water to a range of users (summarized in the
middle section of Figure 2). The canal system managed by the WUA is run in a network
fashion with no major storage points within the system: water orders are calculated on a
weekly basis, with releases from Darlington dam planned according to demand. Water is
released from Darlington dam, flows down the Sundays River, and is abstracted into the
WUA’s canal system at the Korhaans Weir. The water is then diverted through a network of
canals that spans the length of the irrigated farming region in the LSRV, with over 900 delivery
points in the form of sluices delivering to users in the Valley (Lower Sundays River Water
User Association, 2007). It is the responsibility of the users (primarily commercial farmers) to
Page 4 of 12
store adequate water ‘off-site’ (i.e. in dams) to last them for the periods when the WUA does
not supply — for example, over weekends and public holidays.
The SRVM’s system is depicted in the third section of Figure 2. The SRVM receives
untreated water from the WUA to three supply systems (Enon-Bersheba, Addo and Greater
Kirkwood). The water source for the fourth system (Paterson) has historically been
groundwater, but a new pipeline will connect Paterson to the Addo water treatment works in
the near future. This will make all four urban settlements in the SRVM increasingly reliant on
the WUA for the supply of untreated water.
4, System description and model development
4.1. System elements and system boundaries
The choice was made to focus on the Greater Kirkwood water supply system in this modelling
effort (dashed black line in Figure 2), primarily because the SRVM and the WUA indicated
that they were experiencing conflicts in this regard. The real world problem situation of
Greater Kirkwood’s water supply is represented in the system diagram of Figure 3. Three
elements interact with each other, and together form the Greater Kirkwood water supply
system. The Water User Association supplies the water according to an order placement
procedure. The water goes through the Sundays River Valley Municipality bulk water supply
system, where it is stored, treated and distributed to five separate zones in the Sundays River
Valley. Ideally, the water supply is based on the water demand of the domestic population in
Sundays River Valley Municipality. The system boundaries are depicted by the inner black
line, while the arrows impinging on the system represent the effects of external factors and the
means of stakeholders to intervene in the system. The blue arrows leaving the system represent
the ‘outcomes of interest’.
The brown horizontal arrows on the left depict the effects of external factors. They affect the
system behaviour and system performance, but cannot be influenced directly by the
stakeholders or the factors within the system (Enserink et al., 2010). The four identified
external factors are 1) seasonal variation in climate, including summer heat, droughts, varying
rainfall and winters; 2) policy changes regarding water provision as a basic human need, for
instance the free basic water policy from 2001; 3) the amount and time period of water supply
from the Orange River; and 4) aggregate lifestyle expectations.
The means of stakeholders to intervene in the system are depicted by red vertical arrows.
(Enserink et al., 2010). The Water User Association can adjust the water delivery schedule
currently in place. The Sundays River Valley Municipality can increase the capacity of the
water supply system by building extra storage or placing water pumping systems. Additionally,
the operational management of the water treatment works can be intensified by the SRVM.
Finally, the Department of Water Affairs and the South African National Treasury can award
grants to improve the system performance.
Page 5 of 12
Means:
SRVM :
increase SRVM : DWA and
WUA: capacity of | adjust operational National Outcomes
change delivery water supply | management of Treasury: of
schedule system WTWw award grants interest:
External
factors: al J a al Determine
max. capacity of
Seasonal 7
= > technical
variations —> asta
in climate WUA
Facilitate a
' supply
Policy changes =>, common
> understanding
SRVM of system
Water supply bulk water Reduce supply
from => system E> disruptions
Orange River
Demonstrate
L Demand > affects of
Aggregate operational
lifestyle => management
expectations
? => Identify limiting
factors
Figure 3: Systems diagram of the water supply system
The reasons to build a model in the first place, that is the purpose of the model, are described in
terms of the five outcomes of interest to the stakeholders. First, the maximum capacity of the
technical water supply system of Greater Kirkwood is currently unknown, needs to be
determined, and is of deep interest to the SRVM. Second, a common understanding of the
complex technical water supply system and its behaviour can serve as a basis for finding an
operational mode that is supported by all stakeholders. A third outcome of interest is the
identification of the weakest links in the water supply system serving the inhabitants of Greater
Kirkwood. Identifying these limiting factors within the technical water supply system allows
contingency plans to be made and enables the system operators to improve the system. Fourth,
it is of interest to all concerned that supply disruptions are reduced, both in frequency and
duration. Indeed, this forms the focus of an incentive driven performance evaluation approach
for municipalities in South Africa, known as the Blue Drop programme. The fifth outcome of
interest is deepening the understanding of the effects of operational management on the
performance of the water supply system. The interrupted operation of the system owing to
employee’s working hours or pumps cutting at low storage levels affects the water supply
service and the perception of systemic social injustice.
4.2 The model
4.2.1. The water supply system
Details of the technical water supply system of Greater Kirkwood are provided in Figure 4.
The figure represents the Greater Kirkwood area enclosed by the demarcated, dashed black box
of Figure 2. The Greater Kirkwood area itself comprises five zones: zone 1 is Kirkwood town,
the largest urban settlement in the area; while zones 2 to 5 represent the smaller urban areas of
Bergsig, Aquapark, Moses Mabida or Emsengeni. The Sundays River Valley Municipality is
the owner of the technical water supply system, and responsible for maintenance and
operations.
Page 6 of 12
Zone 2
Bergsig
Zone 3:
Aquapark wes 3 Q) Qwes2 zone ae
Moses Mabida —_Emsengeni
ae Kirkwood
WTW
(@5
ML /d) a
Zone 1 yi
Kirkwood = Sete
Pe Sie
=) Q rnp
e =
sc
ater
Raw water reservoirs & =
Figure 3: Simplified schematic of bulk water supply system for the Greater Kirkwood area of the
Sundays River Valley Municipality (Eastern Cape of South Africa).
The intake of untreated water from the WUA main canal occurs via three sluices; one sluice to
the old storage canal, and two sluices to the water reservoirs of varying sizes. The four raw
water reservoirs and the old storage canal together represent the full off-site storage capacity
for untreated water. The exact volume of the old storage canal is unknown, but it is estimated
at 6,84 MI. In the model, the raw water reservoirs and the old storage canal are combined into
one stock variable ‘off-site storage’ with a given maximum capacity (Figure 5).
From the raw water reservoirs, the water is pumped through the water treatment works (WTW)
to reservoir 1. The potable water from reservoir 1 is either pumped through to the water
reservoirs storing the water for the urban areas of Bergsig, Aquapark, Moses Mabida or
Emsengeni, or it stays in reservoir 1 and 2 to supply Kirkwood town. Because of its location,
Kirkwood town is supplied by gravity feed. The water can only reach the other reservoirs by
pumping it using water pumping systems (WPS) 2 and 3. To avoid overheating and seizure, the
diesel engines of the WPSs cut out when the water volume in the reservoirs falls to 20% of the
reservoir capacity or below. Once the pumps have cut out, the reservoirs need to fill to 50%
capacity before the WPSs can start running again. From the time that the pumps first cut out,
the people living in Zones 2 to 5 are without water, while Kirkwood town receives the
remaining 20%. In short, because of its geographical location and the water supply system
design, Kirkwood town is the first to receive water and the last to run out of water.
The discharge rates of the pumps and the design capacity of the water treatment works together
determine the maximum feasible performance of the modelled water supply system (Figure 5).
Presently, available information suggests that the pumps of the water supply system are being
run over their maximum discharge rate in an attempt to meet the supply the demands for
domestic water in the Lower Sundays River Valley.
Page 7 of 12
iad flow
Zone 1: Kirkwood
Town
+
Zone 4&5: Moses
Mabida & Emsengeni
az I
Zone 2&3: Bergsig
and Aquapark
Figure 4: Water treatment, storage and supply sector of the model
|
vows Lamesa |
RK
Page 8 of 12
4.2.2 Water supply sub-system
The SRVM orders for water are placed with the WUA on a weekly basis. The weekly order
placement is designed to cope with demand fluctuations on the part of the WUA clients, who
are supposed to predict the water they need. Currently, the SRVM always request the
maximum possible allocation, in an attempt to prevent failure in the supply of water to the
people of the Lower Sundays River Valley. This often causes wastage of water when more is
supplied than can be stored. The order placement occurs on Thursday afternoon, and on
Monday morning the first water for the week arrives in Kirkwood, depending on the flow level.
The model of the water supply sub-system is depicted in the stock-flow diagram of Figure 6.
week time
<TIME STEP> <transit time>
| Waterin | y >
| ates,
Allocated water [allocation | receiving
allocation
Requested
water
g
Allocation requests
G. Kirkwood
/ Allocation
pulse train weekly rediaiegter
Thursday }
WUA allocation Percentage
perweek allocation Kirkwood
Figure 5: Water User Association Supply sector of the model
3.3.3 Water demand sub-system
As the allocation requests from SRVM are to be based on actual water demand from the towns
in Greater Kirkwood in the future, the growth in water demand is an influential variable. There
are little data available on this issue. Policy changes in 2001 were reflected in a dramatic
increase in demand, accompanied by a responsibility to supply water to all population groups.
Given the existing problems with the supply of water for domestic use to the inhabitants of
Greater Kirkwood, the growth in water demand is expected to pose significant and potentially
conflict-enhancing demands on the already failing water supply system. The model of the
water demand sub-system is depicted in the stock-flow diagram of Figure 7.
Connected ro)
Greater
Kirkwood Net growth
connected
households
Apparent Greater ___Brtrater Kirkwood
Kirkwood water water demand
demand
Average water use
per household
Percentage loss due
to major leaks
Figure 6: Water demand sub-system
Page 9 of 12
5. Preliminary results and discussion
The system dynamics model was developed and is in use to support discussions between
members of the Sundays River Valley Municipality, the Water User Association and other
interested stakeholders. Indeed preliminary results crucial to system performance, and coherent
with the lived reality of users and stakeholders in Greater Kirkwood, have already been
obtained.
Presently, the municipality is running the pumps above their maximum discharge rate. This is
not sustainable in the long term, but is an indication that the technical capacity of the system
may in itself be insufficient to meet the demands placed upon it. Whereas the stakeholders
have been blaming each other in the past, there is a growing realisation that the supply system
may not be adequate. The Water User Association has long held the opinion that the
municipality does not order wisely — witness water wastage through overflow, nor does it
operate the system effectively — witness the failure to supply. The SRVM has long held the
opinion that the WuA is to blame for not supplying on weekends, but has itself not grasped
why the supply fails.
Through the system dynamics model, we have been able to establish that the off-site storage
capacity of the municipality is insufficient to meet the demand for water over a weekend in the
hot, dry summer. So even if the SRVM runs the water supply system extremely efficiently, the
supply of water will fail. This fact had not previously been understood by the municipality,
despite attempts by the WUA to communicate this from 1997 onwards.
It is not clear, however, whether the capacity of the water supply system would still be
insufficient were the WuA to continuously supply water i.e. through weekends as well. This is
related to the uncertainty about the volume that can be stored in the old storage canal. Clearly,
then the off-site storage capacity is a limiting factor in the supply system. Similarly, we have
been able to identify the pump capacity as a further limiting factor should the off-site storage
capacity be increased.
An additional insight that is working to enhance social cohesion between groups is the fact that
the safety mechanisms in the technical design of the pumps, together with the gravity feed to
Kirkwood town, are the reason that Kirkwood is the last to be cut-off and the first to receive
water. The situation is such that all areas of the municipality do not run dry at the same time.
Instead some areas run dry more quickly than others, notably the lower income areas. The zone
with the highest socio-economic status is Zone | (Kirkwood town), which runs out of water
last, and receives water first; whilst the opposite applies to the zones with the lowest socio-
economic status. This is perceived as systemic injustice, particularly as even twenty years after
apartheid ended Zone | is primarily white residents, unlike the populations of the poorer zones
2 to 5. The SRVM is dealing here with an historical artefact from the apartheid era when the
technical systems were designed to service “white areas”. The location of the storage tank to
gravity feed Kirkwood town was chosen for this in mind.
The demand for water has grown rapidly in the last decade with more houses being added to
the reticulation network. As more and more houses are added to the network, there is greater
expectation on the municipality to supply water to these houses. Whilst there is political
pressure to supply running water and sanitation to households, there is less pressure to
maintain and develop the water supply infrastructure required to continuously services to the
households. The anticipation of a strong growth in the demand for water means that the SRVM
now understands that the capacity of the technical system has to be improved if they are to
Page 10 of 12
meet the demand, and that they will have to plan for and acquire funding for such
infrastructural improvement.
Further, the SRVM is now confident that their system performance can be enhanced if they
were to receive a continuous supply of water. They will continue to argue for a change in the
WUA operating procedures, especially given the national priority placed on basic human needs
for water. It is worth noting that the WUA’s delivery system was first and foremost designed
for commercial agriculture, which has water demands different from domestic users. For
example, the open design of the canal system paired with the hot climatic conditions requires
that the canals are cleared of algae annually. The WUA’s chosen method for clearing algae is
to let the canals run dry. Such system maintenance is scheduled during the winter period, when
the farmers’ demands for water are lowest. This means that the canal system only operates for
fewer days per week during the winter, despite the fact that the SRVM, as a supplier of water
for domestic use, requires water all year round. However, the SRVM cannot only demand
changes of the WUA. They are also obliged to rethink their policy of only running the
treatment works on working days, as this is a potential further constraint on water supply that
lies directly under their control.
6. Conclusions
In developing and using a systems model of the Greater Kirkwood water supply system, we
have been able to identify areas of contention between stakeholders, where understanding or
information is currently unavailable, fuzzy or inconsistent. These ‘areas of contention’ form
the focus of the system dynamics modelling endeavour, which is opening up spaces for, and
supporting ongoing interactions between stakeholders. The further use of the model to support
changes in the operation of the water supply system of Greater Kirkwood will occur within the
context of a series of larger transdisciplinary project aimed at breaching barriers to water
policy implementation in South Africa.
The approach presented in this paper is a step away from group model building, which focuses
primarily on the model building component of the modelling cycle. Instead, here modellers
build the model with stakeholders, and then focus on its use by stakeholders. This unusual
application of system dynamics modelling aligns with the cybernetics tradition and ‘soft
systems methodology’ (Checkland, 2000). An analysis of this stakeholder-based use of
modelling and the contribution it makes to the field of soft systems methodology is being
prepared for publication at present (Clifford-Holmes & Slinger, 2013). However, even at this
stage, the utility of the system dynamics method in supporting strategic conversations between
dissenting actors appears promising.
References:
Checkland, P. (2000). Soft Systems Methodology: A Thirty Year Retrospective. Systems
Research and Behavioural Science, 17, 11-58.
Clifford-Holmes, J., & Slinger, J. (2013). A transdisciplinary modelling approach to a
contentious water supply issue in South Africa. Forthcoming.
Enserink, B., Hermans, L., Kwakkel, J., Thissen, W., Koppenjan, J., & Bots, P. (2010).
Systems Analysis. Policy Analysis of Multi-Actor Systems (pp. 51-77). The Hague:
Lemma.
Page 11 of 12
Howick, S., & Eden, C. (2010). Supporting strategic conversations: the significance of a
quantitative model building process. Journal of the Operational Research Society, 62(5),
868-878. doi:10.1057/jors.2010.103
Juta’s Statutes Editors (Ed.). (2010). The Constitution of the Republic of South Africa (Ninth
edit.). Cape Town: JUTA Law.
Lane, D. C. (2000). Should system dynamics be described as a ‘hard' or 'deterministic' systems
approach? Systems Research and Behavioral Science 17 pp 3-22.
Lower Sundays River Water User Association. (2007). Information Pamphlet (p. 13). Sunland.
Statistics South Africa. (2012). Census 2011 Municipal fact sheet (p. 51). Pretoria.
Stave, K. (2002). A system dynamics model to facilitate public understanding of water
management options in Las Vegas, Nevada. Journal of Environmental Management 67
(2003) pp 303-313.
Sundays River Valley Municipality. (2010a). ‘Chapter 5: Water Services Infrastructure
Profile’, Water Services Development Plan: January 2010 Review. Kirkwood, South
Africa
Sundays River Valley Municipality. (2010b). ‘Chapter 2: Service Level Profile’, Water
Services Development Plan: January 2010 Review. Kirkwood, South Africa
Sundays River Valley Municipality. (2011). Integrated Development Plan (IDP): 2011-2016
(5 Year Plan). Available at:
http://www.dwa.gov.za/Dir_WS/WSDP/WSDP/UserControls/WSDPDownloadAttachme
nt.aspx?id=1041 [Accessed: 01/12/2012].
World Bank. (2009). THE SOUTH AFRICAN URBAN AGENDA: Municipal Infrastructure
Finance Synthesis Report (p. 77).
WuA . (2007). Information pamphlet: Lower Sundays River Water User Association.
WuA . (2011). Annual report — 2010-2011: Lower Sundays River Water User Association.
WuA . (2012). Annual report — 2011-2012: Lower Sundays River Water User Association.
Acknowledgements:
Funding gratefully acknowledged from:
South Africa Netherlands Research Programme for Alternatives in Development (SANPAD), the
Mandela Rhodes Foundation, the SKILL programme of the South Africa Vrije Universiteit Amsterdam Strategic
Alliance (SAVUSA), the National Research Foundation (NRF) and the Water Research Commission (WRC) of
South Africa.
The assistance of the following organizations and institutions are key:
Sundays River Valley Municipality, the Lower Sundays River Water User Association (WUA), Amatola
Water, members of the Eastern Cape Rapid Response Unit (RRU) of the Department of Water Affairs; the
Unilever Centre for Environmental Water Quality (UCEWQ) at the Institute for Water Research, Rhodes
University; the Policy Analysis section and Multi-actor Systems Research Programme of the Faculty of
Technology, Policy and M: at Delft University of Technology.
Assistance from the ing indivi is gratefully acknowledged:
Professor Tally Palmer — Director UCEQW and international project leader of the SANPAD project;
Juan Gonzalez (TU Delft) for providing valuable modelling assistance in August 2012.
Page 12 of 12
