Study Name:
Orange River Integrated Water Resources Management Plan
Report Title:
Current Analytical Methods and Technical Capacity of the four Orange Basin
States
Submitted By: WRP Consulting Engineers, Jeffares and Green, Sechaba Consulting, WCE Pty Ltd,
Water Surveys Botswana (Pty) Ltd
Authors:
R Mckenzie, C Seago
Date of Issue: August 2007
Distribution:
Botswana: DWA: 2 copies (Katai, Setloboko)
Lesotho: Commissioner of Water: 2 copies (Ramosoeu, Nthathakane)
Namibia: MAWRD: 2 copies (Amakali)
South Africa: DWAF: 2 copies (Pyke, van Niekerk)
GTZ: 2 copies (Vogel, Mpho)
Reports:
Review of Existing Infrastructure in the Orange River Catchment
Review of Surface Hydrology in the Orange River Catchment
Flood Management Evaluation of the Orange River
Review of Groundwater Resources in the Orange River Catchment
Environmental Considerations Pertaining to the Orange River
Summary of Water Requirements from the Orange River
Water Quality in the Orange River
Demographic and Economic Activity in the four Orange Basin States
Current Analytical Methods and Technical Capacity of the four Orange Basin
States
Institutional Structures in the four Orange Basin States
Legislation and Legal Issues Surrounding the Orange River Catchment
Summary Report
TABLE OF CONTENTS
1
INTRODUCTION ..................................................................................................................... 1
1.1 Background to Study .................................................................................................... 1
1.2 Background to Modeling............................................................................................... 1
1.3 Background to Technical Capacity............................................................................... 3
1.4 Task and report description.......................................................................................... 4
2
METHODOLOGY .................................................................................................................... 6
3
MODELS.................................................................................................................................. 8
4
TECHNICAL CAPACITY ....................................................................................................... 16
4.1 South Africa ................................................................................................................ 16
4.2 Namibia....................................................................................................................... 18
4.3 Lesotho ....................................................................................................................... 19
4.4 Botswana .................................................................................................................... 20
5
CONCLUSIONS AND RECOMMENDATIONS .................................................................... 22
6
REFERENCES ...................................................................................................................... 23
Orange IWRMP
Task 11: Analytical methods
1
INTRODUCTION
1.1
Background to Study
In view of the existing and possible future developments which will influence the availability
of water in the Orange River, a project has been initiated by ORASECOM and
commissioned and funded by GTZ involving all four basin states (Botswana, Lesotho,
Namibia and South Africa). The main objective of the project is to facilitate the
development of an Integrated Water Resources Management Plan for the Orange River
Basin. The plan will in turn facilitate the following specific objectives:
· Maximise benefits to be gained from Orange River water;
· Harmonise developments and operating rules;
· Foster peace in the region and prevention of conflict;
· Encourage proper and effective disaster management;
· Ensure that developments are sustainable and encourage the maintenance of
biodiversity in the basin; and
· Management of potential negative impacts of current and possible future
developments.
1.2
Background to Modeling
Because of their integrative power, predictive abilities and demonstration utility, computer-
based decision support models can play a major role in the operations of all Water
Management Areas. The objective of water resource system analysis is to provide
analytical decision-making tools for optimum utilization of available resources and to
facilitate development planning to satisfy the increase in water demand. In the Southern
African context the water resource systems analyses are based on generated stochastic
streamflow records in addition to the normal available historical streamflow records. This is
necessary due to the relatively short historical streamflow records (typically around 40
years) which tend to be too short to adequately cover long drought periods which can
exceed 20 years in some parts of the Orange River Basin. The advantage of stochastic
hydrology as opposed to historical hydrology is that the reliability of supply, expressed in
annual return periods or exceedance probability percentages, can be determined. In
addition the range of possible streamflow sequences generated by the models will
encompass even the most severe events resulting from possible climate changes which is
know to be an important factor in the river basin.
13/11/2007
Final
1
Orange IWRMP
Task 11: Analytical methods
Systems analysis models were first introduced to South Africa in 1984 following an
extensive worldwide investigation to identify the most robust and reliable modeling
techniques available. Following both an in-depth literature review and a fact finding
mission, it was agreed that the techniques developed in Canada by ACRES International
represented the world's best practice and the development platform selected was
therefore the ACRES Reservoir Simulation Program (ARSP). Further development of the
ARSP then took place over a number of years to modify and tailor the model to suit South
African circumstances. The most important developments included the introduction of the
probabilistic analysis techniques requiring the development of a highly sophisticated and
robust stochastic streamflow generator. This component of the development took more
then ten years to perfect and was sent to two of the Universities specializing in such
analysis techniques (Colorado State University: Prof Niel Grieg and Prof Darrel Fontaine
as well as Prof Pete Loucks and Prof Jerry Stedinger at Cornell University) for an
extensive peer review process. The techniques were acclaimed by several of the world's
recognized specialists as the most advanced and innovative techniques in the world.
The Vaal River System was the first system in South Africa where use was made of water
resources analytical methodologies. Following the original Vaal River System Analysis,
another detailed study, the second of its kind, was undertaken for the Orange River
entitled "The Orange River Systems Analysis (ORSA) Study" which commenced in 1987.
The ORSA Study was commissioned to evaluate the status of the water resources of the
Orange River Basin in order to facilitate the planning of future developments as well as to
undertake a comprehensive and holistic assessment of the river basin including possible
impacts of the Lesotho Highlands Water Project (LHWP). It should be noted that prior to
the development of the new system analyses techniques it had not been possible to
undertake an holistic assessment of the water resources of such a complex water
resource system since neither the software nor the required computing power were
available.
In addition to systems analysis tools, a number of supporting analytical methods were also
developed to facilitate the process of the water resources analyses. These include, for
example, modeling techniques for the classification and patching of rainfall data. Most
rainfall and streamflow records can not merely be used in there original formats as a result
of missing or incorrect data. Modeling tools assist hydrologists to produce records that are
useful.
13/11/2007
Final
2
Orange IWRMP
Task 11: Analytical methods
1.3
Background to Technical Capacity
As mentioned previously, water resources analysis using the analytical methodologies
based on the ACRES Reservoir Simulation Programme (ARSP) was first introduced to
South Africa in 1984. The techniques as they exist today were developed jointly by the
South African Department of Water Affairs and Forestry in association with ACRES of
Canada and first applied to the Vaal River System. The initial ACRES Model was used as
the starting point from which two more sophisticated models were developed namely the
Water Resource Yield Model (WRYM) and the Water Resource Planning Model (WRPM).
Both models are sophisticated network models which use a derivative of the well known
"out-of-kilter" algorithm to analyse complex water networks. The initial Canadian
techniques have been enhanced significantly over a 20 year period to such an extent that
the models used throughout South Africa, and to a lesser extent in Namibia and Lesotho
are now regarded by most specialists as the most potent water resource models available
worldwide.
Over the past 20 years, a great many individuals from numerous companies and
government organisations have been trained in the use and application of the two water
resource models. The models have been used to analyse most of the large and often
complex water resource systems throughout Southern Africa including systems in
Namibia, Lesotho as well as all major systems in South Africa. The models and modeling
techniques are specifically designed to handle water resource modeling in arid and semi-
arid areas which can have critical hydrological sequences of 15 years as opposed to
critical periods of less than one year which are often experienced in more temporate areas
such as most parts of Europe.
Since the models and modeling techniques are relatively complex, they are normally
employed by specialists with an engineering, hydrology or mathematical background. The
models can be run and operated by less technical personnel, however, the initial set-up
and testing of the model requires a sound understanding of the models and the modeling
techniques which in turn can take several years of "hands-on" operation to gain sufficient
experience to set up a system from scratch.
The number of civil engineers and technicians graduating from Universities and
Technikons throughout Southern Africa has been on a steady decline over the last ten
years and there has been a steady exodus of experienced personnel to other countries
which has exacerbated the problem. As a result, the number of suitable specialists in the
13/11/2007
Final
3
Orange IWRMP
Task 11: Analytical methods
water industry is small and appears to be in decline. The problem is particularly evident in
the government sector where many of the experienced engineers and hydrologists have
moved to the Private Sector or to another country. The whole issue of training and
capacity building is therefore not simply a case of providing training to specific individuals
within the various government organisations but a more complicated issue of providing
sustainable support without creating reliance on the Consultants. Before any training or
capacity building is considered, the issues must be clarified and the government
organisations concerned must understand what they want to achieve and if such ambitions
are realistic. There is little value in arranging and presenting "hands-on" technical training
to personnel who have no intention of using the models and in such cases a less technical
high level of training is more appropriate.
It is important to ensure that the different government organisations in each of the four
basin states are given the opportunity to create the capability of using the models in their
own water departments with their own personnel. Unless this form of capacity building can
be achieved, the personnel within the different organisations will gradually lose the ability
to run the models and will eventually become fully dependant on the services of the
consultants or specialists provided by the donor organisations in fact a form of de-
capacitation as opposed to capacity building. Unfortunately this tends to be the case in
many other countries throughout Africa, however, it should be noted that Namibia, South
Africa, Botswana and Lesotho currently have technical personnel who are either capable
of using the models or have in fact used them sometime during the past 20 years. In
some cases, the personnel involved require a refresher course to bring them into line with
the latest model developments, but they do possess sufficient technical capacity to retain
control of the models in their respective countries. It would therefore be very sad if such
capacity is lost.
1.4
Task and report description
The objective of this task and document is to compile a list and description of the major
hydrological and yield determination analytical methods that are currently in use in the four
basin states of the Orange River (Botswana, Lesotho, Namibia and South Africa). Both
surface and groundwater analytical methods are covered and in addition, the technical
capacity available within each basin state to apply the analytical methods has been
evaluated, albeit at a very preliminary level in line with the budget available to undertake
the work. The results of the assessment are presented in the form of specific details of the
13/11/2007
Final
4
Orange IWRMP
Task 11: Analytical methods
specialists, practitioners and trainees as well as organizations that are currently active in
the field of water resource management in each basin state.
This document includes a brief overview of the methodology used to gather the required
information in Section 2. Section 3 then provides a brief description of the analytical tools
used while Section 4 provides an overview of the technical capacity available to use these
tools and the associated analytical methods.
13/11/2007
Final
5
Orange IWRMP
Task 11: Analytical methods
2
METHODOLOGY
In order to establish the level of expertise and capabilities with regard to water resource
management in the four basin states, the various specialists and/or government
departments in each country were contacted. In the case of South Africa, which clearly
has the largest government department dealing with water resource management issues,
the authors of this report were able to call on their own personal contacts who provided
most of the necessary information. An assessment with very similar objectives was
undertaken in South Africa in 2002 to identify the key methods used throughout the
country to analyse and manage water resources. The results from this assessment
provided much valuable information which has been used as the basis for this current
report. The assessment undertaken by the South African Department of Water Affairs and
Forestry is presented in the report entitled "Guidelines for Models to be used for Water
Resources Evaluation" which was completed in June 2002.
It was originally anticipated that the assessment of the capabilities in each of the four basin
states would be established through the use of a questionnaire which would be compiled
and sent to relevant officials. From recent experience, however, it has been found that
questionnaires are often not the most efficient mechanism of obtaining the necessary
information since most of those asked to complete the forms are too busy and the
questionnaires are often neglected., especially if the person conducting the survey has not
introduced themselves to the recipient of the questionnaire.
Much of the information included in this report was therefore obtained from existing
documentation supported by personnel communication with various individuals in each
country required to confirm which models are currently being used in the various countries
as well as the technical capacity available in each country to operate the models.
Telephone conversations were undertaken with key officials in the various countries to
obtain the necessary information. A number of discussions were held with representatives
from each country in order to obtain various opinions from people in the fields of
Consultancies, Government and Parastatals. Although the level of effort involved in this
assessment was limited by the available budget to a desktop assessment, it has been
possible to establish the general capabilities and aspirations within each country. This in
turn will assist future teams in establishing the most appropriate approach for technical
assistance and capacity building in each country. Table 2-1 provides a summary of those
13/11/2007
Final
6
Orange IWRMP
Task 11: Analytical methods
contacted in the course of the assessment and includes their contact details for future
reference.
Table 2-1: Persons contacted
Country
Person
Organisation
Position
Contact no.
Department of Water
Lesotho
Mr Mojakifane
Director
00266 22 317 516
Affairs
Department of Water
Head of Surface
Lesotho
Ms Nteso
00266 22 322 734
Affairs
Water
Department of Water
Head of Water
Lesotho
Ms Motanya
00266 22 312 383
Affairs
Resources
Department of Water
Head of Ground
Lesotho
Mr Lebohang
00266 22 313 602
Affairs
Water
Lesotho
Mr Pepperell
J & G Consulting
Director
GWC Consulting
Lesotho
Mr Bakhaya
Engineers
Lesotho
Mr Ramasoei
Commissioner of Water
Engineer
00266 22 320 127
South
WRP Consulting
Dr Mckenzie
Managing Director
0027 12 346 3496
Africa
Engineers
South
WRP Consulting
Mr van Rooyen
Director
0027 12 346 3496
Africa
Engineers
Ex WCE Consulting
Namibia
Mr Crerar
00251 91 118 3880
Engineers
WCE Consulting
Namibia
Mr Muir
Engineer
00264 61 370 9000
Engineers
Water Sciences
Ground Water
Namibia
Ms Botha
00264 61 257 411
(consultancy)
Specialist
Mr van
Department of Water
Namibia
Surface Water
00264 61 208 7257
Lagenhoven
Affairs
Namibia
Mr Mostert
NamWater
Surface Water
00264 61 710 000
Water Surveys
Botswana
Mr Preston
Managing Director
00267 39 00 541
Botswana (Consultants)
13/11/2007
Final
7
Orange IWRMP
Task 11: Analytical methods
3
MODELS
Numerous hydrological and water resources system models are currently used in the four
basin states by various specialists as well as government personnel. A short description of
the different models is provided in the remainder of Section 3 to highlight the diverse level
of expertise existing throughout the four basin states. Some of the models are well known
throughout the world while others have been developed specifically for use by one or more
of the basin states and have been tailored to suit local conditions. The details provided are
relatively brief and more detailed information can be obtained from the model developers,
most of whom are based in either a University research group or one of the government
departments involved in water resource management.
ACRU Rainfall Runoff Model
ACRU is a multi-purpose and multi-level integrated physical-conceptual modeling system that can
simulate streamflow, total evaporation and land cover / management and abstraction impacts on
water resources at a daily time step (Schulze, 1995). Input to the menu is controlled by a
"menubuilder" program where the user enters parameter or catchment related values, or uses
defaults provided. The ACRU model uses multi-layer soil water budgeting. Streamflow is generated
as stormflow and baseflow dependent upon the magnitude of daily rainfall in relation to dynamic soil
water budgeting. Components of the soil water budget are integrated with modules in the ACRU
system to simulate many other catchment components including irrigation requirements and system
yield. The model treats groundwater dynamics through a non-linear reservoir and allows riparian
zones to be saturated from upland throughflow processes. ACRU requires a degree of calibration.
The model is continually being upgraded and has been used extensively throughout Southern Africa.
The ACRU Model tends to be used predominantly in the Academic field in South Africa while the
Department of Water Affairs and Forestry preferred to use the various Pitman based models which
utilised a monthly time step as opposed to the daily (or less) time step incorporated in the ACRU
Model. It should be noted that the ACRU Model has found more support in recent years in South
Africa in cases where a more detailed assessment for a small area is required. This level of detail is
often required for water licensing assessments and in such cases the ACRU Model has been of use
although the coarser monthly models are still recommended for larger scale rainfall-runoff studies.
SPATSIM Rainfall Runoff Model
The SPATSIM models are variations of the Pitman Model developed by research personnel at
Rhodes University in South Africa. This model is available in two versions namely: SPATSIM-VTI
and SPATSIM-PITMAN
SPATSIM-VTI
This fine-scaled model was developed for daily streamflow production from individual small
catchments and can therefore be considered as an alternative to the ACRU Model when modeling
smaller catchments in greater detail. The model is driven by daily rainfall and potential evaporation
and uses soil-moisture budgeting in more than one macro soil layers according to a spatial
discretisation based on soil texture classes and land use practices. A range of land-covers, be they
natural or human related, can be superimposed. The model allows simplistically for the development
of variable source areas for runoff generation and for physically based groundwater dynamics. The
model requires a degree of calibration which can be problematic if reliable daily flow data are not
13/11/2007
Final
8
Orange IWRMP
Task 11: Analytical methods
SPATSIM Rainfall Runoff Model
readily available.
SPATSIM-PITMAN
This is the monthly version of the model and is structured around relatively simple empirical
algorithms of the major rainfall-runoff processes that determine a large catchment's hydrological
response. The model produces "naturalized" flows and the the influences of human impacts have to
be explicitly added through support routines. The model parameters have to be determined by
calibration against observed flows, or transferred from comparable catchments for which parameters
are known.
SHELL Rainfall Runoff Model
The SHELL modeling package is a hybrid version of the original Pitman rainfall runoff model and
incorporates the PO8 rainfall aggregation model as well as the RESSIM reservoir simulation model.
It allows for configuration of multi-catchments and multi-reservoir flow systems and can be used with
natural or developed flow regimes. It is very similar to the WRSM90/2000 modelling system which in
turn is the updated version of the Pitman Model, The model includes a range of graphical user
interfaces and incorporates the basic monthly Pitman model (Pitman 1973) as its core rainfall-runoff
model. Various additional modules have been added to enable the model to and through its various
other routines can simulate, on a monthly basis: natural rainfall-runoff processes; reservoir and farm
dam balances; irrigation and other abstractions; land-use return flows; streamflow reductions due to
afforestation; streamflow reductions due to invasive alien plants; alluvial river-bed transmission
losses.
WRYM Reservoir Simulation Model
This model is used to optimize the allocation of water from reservoirs on a monthly basis throughout
a large multi-use river system, according to a penalty structure, for a given time horizon of water
demands and allowing stochastic variation of streamflows. It is used to calculate the long-term yield
from a reservoir or system of reservoirs and can be used to examine different operating rules or to
develop short-term or long-term yield-reliability curves. The model was the first of its type in the
world to utilise stochastic streamflow sequences and as such is recognised as one of the most potent
reservoir simulation models in the world. Many other countries are now developing similar models
based on the model including the UK, USA and Australia. The model is ideally suited for use in arid
and semi-arid areas which can have critical periods in excess of 15 years as opposed to the
European conditions where critical periods of a year or less are often encountered.
The model is based on the well known Acres Reservoir Simulation Program developed in Canada by
the former Acres International specifically for the analysis of hydro-power systems. Through the
continued development in Southern Africa (mainly Namibia and South Africa) the model has grown in
its ability to analyse increasingly complex water resource systems often involving up to 200
reservoirs as well as the conjunctive use of surface and groundwater resources.
The WRYM model has formed the core of the water resource analysis techniques used in Southern
Africa for the past 15 years and continues to be the model of choice by the key government
organisations in Namibia, Lesotho and South Africa. The model has been used by personnel from
the various Departments of Water Affairs in at least 3 of the 4 basin states and numerous personnel
from all 4 basin states have attended one or more of the many training courses presented on the
model on an annual basis.
13/11/2007
Final
9
Orange IWRMP
Task 11: Analytical methods
WRPM Reservoir Simulation Model
The WRPM is a hybrid version of the previously described WRYM and operates on a very similar
basis. The key difference between the two models is that the planning version is able to simulate
water quality constraints as well as a changing system configuration with time. This makes the
model an ideal tool to analyse the impacts of growing demands as well as the incorporation of new
reservoirs or other enhancements to the water resource infrastructure.
This model has the same structure as the WRYM and essentially uses the same algorithms to solve
the to optimize the allocation of water on a water on a monthly basis throughout a large multi-use
river system.
Model outputs can be used to indicate the required timing of augmentation measures and schemes
to maintain given assurance levels. The impact of management options on reliability of supply can be
examined. Since the model also includes certain water quality modeling capabilities, it can be used to
model and manage salinity in the system Where certain water quality constraints are modeled and
releases are made to ensure that the salinity does not breach certain user-defined limits. This is of
great importance especially in the Vaal River System where salinity is a critical issue in the vicinity of
the Vaal Dam and Vaal Barrage due to the high volumes of effluent return flows from the numerous
sewage treatment plants.
The WRPM is also used to provide a 5-year or 10 year projection of reservoir levels which in turn can
be used as an operational aid to the system managers. Using the results from the model, the ater
resource managers can take informed decisions regarding possible curtailments at an early stage in
any drought event rather than waiting until the situation becomes critical and it is too late to take
proper evasive action to avoid severe water restrictions.
WSAM
The WSAM was initially designed as a coarse-scaled model to assess the yields from any
catchments in South Africa at a quaternary level based on a simple cascading water balance model.
Unfortunately the original concept of a very simple model that can be used to derive a "quick and
dirty" assessment was not achieved and the final version of the model is considered by many to be
as complicated as the sophisticated yield model (WRYM). For this reason, there is relatively little
advantage , if any, to be gained from using WSAM as opposed to the WRYM.
Although WSAM is heavily promoted by the South African DWAF, and amny training courses are
presented to train personnel on its use, the model is largely ignored by most water resource
practicioners in South Africa who prefer to use the more powerful models based on the original
WRYM.
ISIS
This model is a one-dimensional hydraulic streamflow model based on a finite difference application
of the fully St Venant's flow equations to a series of cross-sections of the river channel and flood
plain and any hydraulic conduits that are built in the flow path. It is aimed at modeling the flow of
water in a river channel and can in theory be used in real-time mode although this has yet to be
proven outside a research environment. The ISIS model is regarded as one of the leading models of
its type in the world and originates from the UK as a fully commercial package. The costs associated
with the model can be prohibitive and it should only be used where the additional capabilities and
ease of use are sufficient to justify the costs over the freely available models such as HEC-RAS. The
13/11/2007
Final
10
Orange IWRMP
Task 11: Analytical methods
ISIS
model has been used in both South Africa and Namibia over the past 10 years.
A range of conservative and non-conservative water quality routines are incorporated in ISIS. The
basic requirements for applying the model are regular cross-sections of the river channel and its flood
plains, boundary conditions in the form of upstream and tributary inflow series (including water
quality), and certain meteorological time series. Friction loss factors and water quality parameters are
derived by calibration. This means that reasonable flow and water quality records of in-channel
conditions are required. The model is useful to assess short-term downstream water levels and
discharges as well as water quality impacts to upstream operations, or to examine management
options related to localized flow and water quality issues. Full backwater effects are simulated. It also
comes with powerful graphical interfaces.
DUFLOW
This model is based on a finite difference application of the full St Venant's flow equations to a series
of cross-sections of the river channel and flood plain and any hydraulic conduits that are built in the
flow path. A range of conservative and non-conservative water quality routines are incorporated in
this model. The basic requirements for applying the model are regular cross-sections of the river
channel and its flood plains, boundary conditions in the form of upstream and tributary inflow series
(including water quality), and certain meteorological time series. Friction loss factors and water
quality parameter are derived by calibration. This means that reasonable flow and water quality
records of in-channel conditions are required. The model is useful to assess short-term down-stream
water level and discharges as well as water quality impacts of upstream operations, or to examine
management options related to localized flow and water quality issues. Full backwater effects are
simulated. It also comes with powerful graphical interfaces. DUFLOW has also been imbedded in a
South African-made information system with additional graphics support (Tukker, 2000).
MIKE 11
This model is also a commercial model originating in Europe and is based on a finite difference
application of the full St Venant's flow equations to a series of cross-sections of the river channel and
flood plain and any hydraulic conduits that are built in the flow path. A range of conservative and non-
conservative water quality routines are also incorporate in this model. The water quality module is
separate module and not included in the basic module. The basic requirements for applying the
model are regular cross-sections of the river channel and its flood plains, boundary conditions in the
form of upstream and tributary inflow series (including water quality), and certain meteorological time
series. Friction loss factors and water quality parameters are derived by calibration. This means that
reasonable flow and water quality records of in-channel conditions are required. The model is useful
to assess short-term and long-term downstream water levels and discharges as well as water quality
impacts of upstream operations, or to examine management options related to localised flow and
water quality issues. Unsteady uniform flows are simulated under a fully hydrodynamic flow
description. It also comes with powerful graphical interfaces.
13/11/2007
Final
11
Orange IWRMP
Task 11: Analytical methods
WQT MODEL
This a coarse-to-medium-scaled model for salinity (total salt load) production and transport in large
multi-user, multi-reservoir catchments, specially designed to be driven by the same natural flows that
drive the WRYM and WRPM system analysis models outlined above. The free-standing version of
WQT is used to determine salinity model parameters by calibration. These parameters are the input
to the WRPM model, which incorporates an integrated version of WQT, to generate
salt/concentrations during multiple stochastic optimization runs in large river systems.
NACL MODEL
This model was developed in South Africa around the around the daily Pitman rainfall-runoff model
and is designed to simulate the flow of salts through a multi-reservoir system. It allows urban washoff
as well as operation of reservoirs, wetlands, and coarse irrigation activities. Its parameters are
determined by calibration. It is envisaged that this model will be used in applications where certain
parts of a multi-user catchment require more detailed assessment than can be provided by the
monthly WQT model as mentioned previously. The NACL model is rarely used in South Africa and in
most cases the WQT or WRPM models are preferred.
DISA MODEL
This is a fine-scaled model for salinity production and transport through formalized irrigation schemes
and allows operation of supply reservoirs, river channel transport, diversion devices, primary and
secondary canals, balancing dams, artificial drainage , groundwater variability and wide range of
irrigation practices. It is driven by daily rainfall and uses soil texture classes, location on the
landscape, agricultural practices. It is packaged in a user-friendly modeling environment with strong
graphical support. It is recommended as support for any of the other models to assess, at finer
scales/resolutions, irrigation impacts of large or multi-off-take irrigation schemes, or to examine
operating rules and other management options for salinity control.
Relatively little use of the DISA Model is made in South Africa outside the academic institutions.
IMPAQ
This is medium-to-fine-scaled modeling system for salinity, sediment and phosphate production and
transport in large multi-user catchments, specially designed to be driven by the same natural flows
that drive the WRYM and WRPM system analysis models outlined above. It has a washoff routine
which uses SCS Curve Number to allow any mix of land-uses to affect sediment and phosphate
production, which are derived from combination of loading functions, potency factors and the so
called USLE approach. Non-conservative processes such as sedimentation and re-suspension are
allowed to play a role in a channel transport module and in a mixed reactor reservoir module. IMPAQ
allows more diverse land-use variability that WQT and its parameters need to be determined by
calibration. It is used in conjunction with WRYM to generate very long sequences of monthly
loads/concentrations of selected constituents in large river systems. The modeling system is object-
oriented coded and imbedded in a powerful graphical environment. As for the DISA Model discussed
previously, relatively little use of the IMPAQ model is made in Southern Africa outside the academic
institutions.
13/11/2007
Final
12
Orange IWRMP
Task 11: Analytical methods
FLOSAL MODEL
These are medium-to large-scale models for salinity production and transport in large multi-user,
multi-reservoir catchments. They are structured around the monthly/daily Pitman rainfall-runoff
models. FLOSAL allows irrigation activities, urban washoff, operation of reservoirs, wetlands, and its
daily version includes in-channel routing. Its parameters are determined by calibration. It is operated
in parallel to the system analysis models described above. The use of the model is currently limited
to a few researchers in various academic institutions in South Africa
VISUAL MODFLOW
Visual MODFLOW is a pre and post-processing package for the USGS MODFLOW is a 3D Finite-
Difference Groundwater Flow Model and is the most widely used groundwater model in the world.
Because of its ability to simulate a wide variety of systems, it's extensive publicly available
documentation, and its rigorous USGS peer review, MODFLOW has become the worldwide standard
groundwater flow model. MODFLOW is used to simulate systems for water supply, containment
remediation and mine dewatering. When properly applied, MODFLOW is the recognized standard
model used in litigation, regulatory agencies, universities, consultants and industry. A number of pre-
and postprocessors exists for MODFLOW, e.g. PMWIN (Chiang and Kinzelbach, 2000), Visual
Modflow, GMS, Modime and ARGUS 1 of which PMWIN is the most used in the world and also in
South Africa. It comprises a professional graphical pre- and postprocessor and the 3-D finite-
difference groundwater models MODFLOW-96 and MODFLOW-2000. Visual MODFLOW supports
the simulation of the effects of wells, rivers, reservoirs, drains, head-dependent boundaries, time-
dependent fixed-head boundaries, cut-off walls, compaction and subsidence, recharge and
evapotranspiration.
FEFLOW
FEFLOW is a 3D Finite Element groundwater modeling software package that combines powerful
graphical features with sophisticated analysis tools and robust numerical algorithms for transient or
steady-state flow, saturated and unsaturated flow, density-dependent flow (saltwater intrusion), multi
free surfaces (perched water table), and mass and heat transport. FEFLOW allows the user to
graphically create finite element meshes for simple or complex geological formations, import and link
data from external sources via FEFLOW's GIS/ DATA Coupling exchange system, assign all
necessary flow and transport parameters, run complex model simulations and visualize the
simulation results in two-or-three-dimensions.
PM WIN and MODFLOW
MODFLOW is a 3D Finite-Difference Groundwater Flow Model and is the most widely used
groundwater model in the world. Because of its ability to simulate a wide variety of systems, its
extensive publicly available documentation, and its rigorous USGS peer review, MODFLOW has
become the worldwide standard groundwater flow model. MODFLOW is used to simulate systems for
water supply, containment remediation and mine dewatering. When properly applied, MODFLOW is
the recognized standard model used in litigation, regulatory agencies, universities, consultants and
industry. A number of pre-and postprocessors exists for MODFLOW, e.g. PMWIN (Chiang and
Kinzelbach, 2000), Visual Modflow, GMS, Modime and ARGUS 1 of which PMWIN is the most used
in the world and also in South Africa. It comprises a professional graphical pre-and postprocessors
and the 3-D finite-Difference groundwater models MODFLOW-96 and MODFLOW-2000. PMWIN
supports the simulation of the effects of wells, rivers, reservoirs, drains, head-dependent boundaries,
time-dependent fixed-head boundaries cut-off walls, compaction and subsidence, recharge and
13/11/2007
Final
13
Orange IWRMP
Task 11: Analytical methods
PM WIN and MODFLOW
evapotranspiration. In addition to the standard packages of MODFLOW-96 and 2000, PMWIN
includes the following modules:
PMPATH 99 allows for particle tracking. Both forward and backward particle-tracking
schemes are allowed for steady state and transient flow fields. PMPATH calculates and
animates the particle tracking processes simultaneously and provides various on-screen
graphical options including head contours, drawdown contours and velocity vectors.
MT3D/ MT3D96 can be used to simulate changes in concentration of single species miscible
contaminates in groundwater considering advection, dispersion and simple chemical
reaction.
MT3DMS is the next generation of MT3D, MS stands for Multi-Species. PMWIN provides full
support for MT3DDMS and can take advantage of its capabilities by using the new solution
schemes. Up to 30 different species can be simulated with PMWIN
MOC3D features advection, dispersion and simple chemical reaction.
Two reactive models PHT3D and RT3D are available as separate Add-On modules to
PMWIN. These modules allow simulating multi-species transport and reactions, such as
BTEX degradation or sequential decay reactions of PCE-TCE-DCE-VC.
PEST and UCODE allow the following model parameters to be automatically calibrated: (1)
horizontal hydraulic conductivity or transmissivity, (2) vertical leakance, (3) specific yield or
confined storage coefficient, (4) pumping rate of wells; (5) conductance of drain, river, stream
or head-dependent cells, (6) recharge flux, (7) maximum evapotranspiration rate, and (8)
inelastic storage factor.
3D Groundwater Explorer provides three-dimensional visualization and animation of data
from groundwater flow and transport models.
GMS
GMS is a pre and post-processing package for the MODFLOW suite of models: MODFLOW,
MODPATH, MT3D, RT3D. It also contains a finite element model, FEMWATER, and 2D Finite
difference seepage model SEEP2D. MODFLOW is a 3D Finite-Difference Groundwater Flow Model
and is the most widely used groundwater model in the world. Because of its ability to simulate a wide
variety of systems, its extensive publicly available documentation, and its rigorous USGS peer
review, MODFLOW has become the worldwide standard groundwater flow model. MODFLOW is
used to simulate systems for water supply, containment remediation and mine dewatering. When
properly applied, MODFLOW is the recognized standard model used in litigation, regulatory
agencies, universities, consultants and industry.
AQUAWIN
AQUAWIN for Windows is a public domain two-dimensional finite element groundwater model and
includes a mesh generator, flow simulation, transport simulation, risk analysis, inverse modeling and
animation packages. The code is the result of a Water Research Commission project and has been
applied by many professionals in South Africa.
13/11/2007
Final
14
Orange IWRMP
Task 11: Analytical methods
NAMROM
NAMROM is an empirical rainfall-runoff model designed specifically by the Namibian Department of
Water Affairs for use in the very arid Namibian conditions. One of the key factors included in
NAMROM is the inverse runoff characteristics following a wet year in which the runoff in subsequent
years tends to reduce rather than increase. The explanation stems from the fact that after many very
dry years, there is little if any vegetation to reduce the runoff in which case a heavy rainfall event will
produce a large runoff. Following a wet year, however, a thick blanket of vegetation will grow which if
the same rainfall occurs will tend to reduce the runoff considerably. Virtually all other rainfall-runoff
models will not be able to handle this apparent anomaly and it was for this reason that NAMROM
was developed. The model is a simple program with few graphical enhancements and is essentially
a FORTRAN program which was converted to DOS from its original Hewlett Packard roots in the
early 1990's when the Namibian Central Area Water Master Plan was developed through GTZ.
The model is currently operational although it has not been used in many years due to an absence of
suitable projects and personnel available to use the model. It is due for a face lift and some training
will be required.
13/11/2007
Final
15
Orange IWRMP
Task 11: Analytical methods
4
TECHNICAL CAPACITY
In the remainder of this section, the following items will be used:
· Specialists: Individuals who are familiar with model inputs and can understand
the outputs but would not be able to configure and use a model as such.
· Practitioners: Individuals who would be able to configure a model and use it to
obtain the necessary information required.
· Trainees: Individuals who are presently being trained or may be trained in future
in the use of modeling techniques.
4.1
South Africa
4.1.1
Consultants
There are currently approximately 20 consultancies throughout South Africa which have
the ability to undertake water resources planning, surface and/or ground water modeling.
Within these consultancies there are approximately five specialists, 60 practitioners and 20
trainees. Some of the larger, more established consultancies include WRP Consulting
Engineers (Pty) Ltd, Ninham Shand, BKS (Pty) Ltd, Stewart Scott and WSM Leshika
(Ground water).
Most of the water resources work undertaken by South African personnel on the Orange
River Basin has been completed by individuals currently working for WRP (Pty) Ltd,
although much of the work was undertaken between 1989 and 1998 when the individuals
were employed by BKS Pty Ltd.
In the past five years, several studies have been
undertaken in which some other consultants have become involved with various aspects of
the Orange River and these include personnel from Ninham Shand, BKS and Jeffares and
Green.
In most of the studies, some component of the work was undertaken by the
personnel currently working through WRP Pty Ltd.
4.1.2
Government
The Department of Water Affairs and Forestry (DWAF) is the governmental organization
that handles all aspects of water and is the custodian of all water in South Africa. The
Department is divided up into many directorates. The Chief Directorate: Integrated Water
13/11/2007
Final
16
Orange IWRMP
Task 11: Analytical methods
Resources Planning contains three directorates, namely National Water Resources
Planning, Water Resources Planning Systems and Options Analysis. It is within these
three directorates that the expertise in terms of surface and ground water modeling fall.
The technical capacity consists of about 30 individuals, of which the majority would be
considered specialists. The DWAF carries out many training courses on the main models
used in South Africa, however, many of the attendees of these courses do not utilize the
models afterwards and therefore do not possess the "hands-on" skills necessary to set up
and run the models from scratch although they do have the skills necessary to understand
the operation of the models and the associated results. As with any software, the users
must work on the models regularly to be able to operate and run them properly. With most
DWAF personnel, such work is not part of their job description and it is normally sufficient
to understand how the models operate without actually having to set up and run them.
4.1.3
Parastatals
The main parastatal organisations within South Africa which have expertise in modeling
consist of Universities, Water Service Providers and Eskom. The University of KwaZulu-
Natal's School of Bioresources Engineering and Environmental Hydrology has a strong
history of modeling focusing mainly on surface water. The University of the Free State was
traditionally the institute with significant groundwater knowledge. Other universities with
modeling experience include The University of Johannesburg, Rhodes and Stellenbosch
Universities. Umgeni Water, the major water service provider in the province of KwaZulu-
Natal, carries out electronic systems analysis internally and contains engineering and
hydrological skills to do so. Rand Water, the largest Bulk Water Service Provider in the
country, does not do water resources planning and such work is left to the personnel within
DWAF. Eskom which is responsible for supplying power to the nation, did establish a
section to undertake the water resources analysis using the key yield and planning models
developed by DWAF. It appears that this capability has been lost to some extent over the
past ten years and the key personnel who were trained in the use of the models appear to
have moved to other companies or countries. It is estimated that there are approximately
20 practitioners in parastatals throughout South Africa whose knowledge is divided
between surface and ground water resources.
13/11/2007
Final
17
Orange IWRMP
Task 11: Analytical methods
4.2
Namibia
4.2.1
Consultants
There are two main consultancies within Namibia (WCE and Lund) that specialise in the
analysis and modeling of surface water resources. Within these consultancies there are
approximately four practitioners who are able to undertake basic water resource planning
work using the yield and planning models developed in South Africa. Namibia was one of
the first countries outside South Africa to recognise the value of the yield and planning
models which were first used to develop the Central Area Water Master Plan in the early
1990's. .One consultancy (Water Sciences) specialises in groundwater modeling, and
there is only one practitioner who is able to carry out the work. In general, there is a
serious threat that the existing capacity to undertake detailed water resource planning and
analysis in Namibia may be lost in future if steps to address the problem are not taken.
Should Namibia lose its capacity, it will be in a situation where it relies heavily on foreign
expertise to tackle water resource management issues which would be a great loss for a
country that has had such capable capacity in the past.
4.2.2
Government
The Department of Water Affairs of Namibia contains a surface and a ground water
section. The surface water section contains three specialists who have undergone training
in surface water modeling and can do the very basics.
The groundwater section out
sources all modeling work and is more involved with the policies surrounding ground
water. One individual from the ground water section would be considered a specialist. It
should be noted that virtually all of the personnel within the Department of Water Affairs
who were trained in the use of the key water resource models in the early 1990's have
since either left the Department or the country.
4.2.3
Parastatals
NamWater is the Water Service Provider in Namibia. Approximately three hydrologists are
employed by NamWater who are able to undertake water resource modeling although by
their own admission, they may require a refresher course to re-establish their capacity. As
mentioned previously, it is necessary to use the models on a continuous basis to retain the
13/11/2007
Final
18
Orange IWRMP
Task 11: Analytical methods
skills necessary to undertake any detailed assessments.
Fortunately the individuals
remain enthusiastic and capable of picking up the latest modeling techniques and have
expressed a wish to set up and run the latest system models within NamWater.
This
should be considered as a priority and any developments on the water resource modeling
should involve the re-establishment of a modeling section within NamWater and if possible
also the training of new personnel to ensure that the existing expertise is not lost. The
ground water section contains two specialists in ground water although all modeling work
is currently outsourced to external specialists and consultants.
4.3
Lesotho
4.3.1
Consultants
Most water resource work undertaken in Lesotho is outsourced to external specialists and
Consultants, many of whom reside in South Africa while others are contracted from
Europe through various funding agencies. Since many water resource developments are
associated with the transfer of water to South Africa, the majority of studies tend to involve
specialists from both South Africa and some overseas country. This has been the general
trend for many years and explains why no truly Lesotho based consultancy has evolved
over the years. Most of the projects are of such a magnitude that no small company could
possibly be responsible for the assessments and the water resource analysis is generally
left to the two government departments and their designated consultants. The main local
consultants who have been involved to any degree with water resources related work are
GWC and Sechaba Consulting Engineers although neither is skilled in the use of the main
yield and planning models which have been used to analyse the major water resource
developments in Lesotho.
4.3.2
Government
The Lesotho Government contains three organisations, namely, the Commissioner of
Water, the Department of Water Affairs and the Lesotho Highlands Development Authority.
The Commissioner of Water deals mainly with aspects of policy and utilises the
Department of Water Affairs for operational and assistance of a technical nature. It has no
water resource expertise within the department and does not expect to require such skills
which are outside its mandate. The Department of Water Affairs contains three divisions
13/11/2007
Final
19
Orange IWRMP
Task 11: Analytical methods
with expertise on water resources issues namely; a groundwater division, a surface water
division and a water resources division. There are approximately 18 professionals who
work in these divisions, made up of hydrologists and engineers. None of them currently
carry out any detailed modeling work although many have attended various courses and
training workshops on the use of the yield and planning models. Many of the individuals
have also attended various overseas sponsored training courses to highlight the water
resource analyses techniques used in Europe and elsewhere in the world. There is
considerable scope for training of the local Lesotho personnel who are often enthusiastic
to learn how the various models are used with particular emphasis on the Lesotho
Highlands Water Project and now also the Lesotho Lowlands Water Project. It is
anticipated that if the yield and planning models are established for the whole of the
Orange River System as suggested in the other reports, there will be sufficient capacity
and interest in Lesotho to set up and use the models using local personnel. Obviously this
will require some support in the form of "hands-on" training, however, the key resources
are in place and it is therefore a viable proposition to create such capacity.
4.3.3
Parastatals
No information on technical capacity was obtained from any parastatals in Lesotho.
4.4
Botswana
4.4.1
Consultants
In many regards, the availability of water resources expertise in Botswana mirrors that of
Lesotho in that the Government Department of Water Affairs is relatively strong, however,
there are few water resource specialists in the private sector. Due to its proximity with
South Africa, Botswana can call on any specialist expertise that may be required from
South Africa. The scope for private water resources personnel is limited by the relatively
sparse surface water resources making it uneconomical for companies to specialize in this
area since the scope for work is so small.
While the scope for surface water specialists is limited, there is possibly more scope for
groundwater specialists since much of the country relies heavily on groundwater
13/11/2007
Final
20
Orange IWRMP
Task 11: Analytical methods
resources. There are several groundwater specialists operating in Botswana and there
are currently two main Consultancies, one dealing with surface water (Ehes) and one with
groundwater (WSB). Two practitioners are able to carry out groundwater modeling within
the consultancies and significant expertise is available locally from the companies.
4.4.2
Government
The Department of Water Affairs in Botswana has considerable expertise in the fields of
surface water and groundwater management which is due largely to the country's policy of
developing a skills base within the country as opposed to relying totally on external
specialists. It is always interesting to note that any water resource planning courses held
in South Africa are attended by one or more personnel from the Botswana Department of
Water Affairs and it is also evident that they are well qualified and understand the various
modeling concepts being employed in South Africa, Lesotho and Namibia despite the fact
that the models have yet to be used in a major study in Botswana.
From the various discussions with the personnel from the Department of Water Affairs in
Botswana, it is clear that there are several highly qualified personnel who have the skills
and expertise necessary to undertake water resource planning for either surface or
groundwater. As is the case in all four basin states, these individuals would require some
"hands-on" training to allow them to gain sufficient experience and confidence to use the
models properly. The key issue is that local personnel are available and enthusiastic which
is usually the limiting factor.
4.4.3
Parastatals
No information on technical capacity was obtained from and parastatals in Botswana.
13/11/2007
Final
21
Orange IWRMP
Task 11: Analytical methods
5
CONCLUSIONS AND RECOMMENDATIONS
This report has summarised the analytical methods used and