The objective of the Scenario Generator (SG) section is to introduce the potential benefits of developing a generic, hydrologically focussed SG for use by the DWAF SEA team in association with an experienced hydrologist (or someone with hydrological simulation expertise).

This section is laid out in the following format.

• The responsibilities and challenges faced by the DWAF SEA team are introduced;
• the generic hydrologically-focussed SG is introduced;
• the delineation of water use activities is discussed;
• considerations related to supply and demand scenarios are introduced; and
• the use of the SG and its potential limitations and opportunities are discussed.

The DWAF SEA team is tasked to strategically assess water use, giving consideration to the environment, the economy and to society. It is useful to carefully consider the words that constitute the SEA mandate:

• Strategic

The term strategic suggests the assessment to be done at a "high level", with information generated being used at a strategic level. However, the SEA process may be undertaken at various scales (Steyl et.al., 2000), including a National scale, the scale of a WMA, or local scale. The ability of the SEA to be undertaken strategically at the different scales appears to be paradoxical, in that one may question if information generated at local and WMA scale is in fact "strategic". It is however possible to relate WMA and local scale scenarios/assessments to strategic level indicators and/or criteria. For example, as part of a WMA scale SEA irrigation water use is curtailed by say 20%. The impact of this scenario may be expressed using indicators and/or criteria at the WMA scale, as well as the National scale respectively. An example of a WMA scale indicator could be the expected change in contribution by agriculture to the WMA product (this is currently not a recognised indicator), while a national scale indicator could be the change in the contribution of agriculture to gross national product (GNP). However, translating the impact of local or WMA scale SEA water use scenarios to national scale indicators and criteria (for the SEA to be considered of strategic importance) may be very complicated indeed, as vast sets of information at the national scale may be needed to undertake this translation.

Consequently, the term "strategic" in the context of this project has been interpreted as the nature of information related to water use that is deemed to be important (strategic) by the SEA at the respective scales. The identification of information deemed to be "strategic" at the various scales, and for different areas in the country, is the starting point of the SEA process, and will depend on whose interests the SEA is to serve. The SEA in effect, needs to answer the following questions at the scale that the SEA is undertaken at:

• Who will use the information generated by the SEA?
• What are the "strategic" economic, social and environmental water use related information requirements of the user/s?
• Where and how can this information be obtained?
• What water use related indicators and/or criteria need to be developed for economic, social and environmental considerations in order to provide a method of:

• Illustrating the water use conflicts and challenges associated with current water use, and to
• Comparing/assessing the outcome of potential future water use scenarios?

Notes:

• Although the client (i.e. the funder of the SEA process) usually is the user of information generated by the SEA process, this may not always necessarily be the case.
• There may be more than one user of the SEA generated information, which may complicate the SEA process, as the assessment may need to accommodate the requirements of multiple users.
• Strategic information in the context of an SEA would probably refer to information regarding both current and potential water use. In other words, the SEA is tasked to assess what water use scenarios (future water use) are potentially feasible.

The term "assessment" suggests that an appraisal (evaluation) is undertaken as part of the SEA process. The definition of assess reads as follows:

"Assess: to determine the importance, size or value of…" (Webster’s Ninth New Collegiate Dictionary).

In order for assessment to be possible, the SEA generated information should possess the ability to carry value, size or importance. One potential solution for the DWAF SEA team to meet this objective is to identify and/or develop suitable criteria and/or indicators of economic, environmental and societal use of water in a manner or format in which magnitude (or value) may be attributed to the indicators/criteria of water use. The importance (or value) of each selected and/or developed criteria and/or indicator will depend upon the importance of the indicator/criteria to the user/s of the information generated by the SEA.

• Water use

The term "water use" is central to the DWAF SEA mandate. The challenge is to define water use in a manner flexible enough to accommodate a multi-disciplinary use of water (i.e. the economic, societal and environmental use of water). A generic wording of water use, which is consistent with the NWA, is as follows: The SEA is responsible for quantifying the use of water by water using activities. These activities include economic activities and non-economic activities. Within the non-economic activities a differentiation may be made between the environmental use of water, and water used by people (society) for purposes other than commercial activities.

In effect, the SEA may need to undertake the following steps:

• Clearly identify the different categories of water using activities (i.e. categories of water using activities that relate to the economic, social or environmental use of water) (cf. Section 6.4).
• Quantify the water use of the current and potential water using activities, giving consideration to:

• A suitable indicator of water use for the SEA decision making (e.g. mean annual streamflow reduction (MASR) by each activity, or the average streamflow reduction during dry periods, or the impact of each water use on catchment and/or systems yield)
• The spatial location of the activities, and
• The temporal variation of water use by the activities (e.g. water use during dry periods vs wet periods).

• Relate or link the water use of the categories of activities to indicators/criteria that reflect hydro-economic, hydro-socio and hydro-environmental considerations (and potentially combinations of these, e.g. hydro-socio-economic water use indicators).

Notes:

• The NWA may require categories of water use, such as SFRAs, to be further subdivided in order to provide the information at a level that decisions may be taken at. For example, the SFRAs may divided into a number of sub-categories, including, amongst others:

• The type of SFRA (e.g. forestry, vs say sugarcane if it is declared as an SFRA),
• The area of SFRA ownership (small growers may potentially not be subject to water use charges, hence must be discerned for scenario generation)
• The nature of SFRA ownership (e.g. previously disadvantaged persons may own the SFRAs)

• The quantification of water use by the various categories and sub-categories of water using activities is challenging for the following reasons:

• Current and potential water use must be quantified. A hydrological model will be required in order to simulate (estimate) the water use of current and future scenario water uses,
• The hydrological model selected should ideally be able to accommodate the different scales at which the SEA may operate, as well as for different areas in South Africa. A process based hydrological model, which operates at a fine time step (e.g. daily) is required to meet these objectives. We have selected the ACRU hydrological model for this purpose.

• There are two broad approaches to water conflict situations, including:

• Change the demand for water so that conflict is reduced, and/or
• Increase the supply of water so that conflict is reduced.

• Relating the water use of each activity to an indicator or criteria is challenging for the following reasons:

• Suitable indicators/criteria may need to be developed which requires a deep understanding of the requirements of the NWA, and
• The temporal variation of water use (e.g. water use during dry vs wet periods) may be difficult to accommodate in one criteria/indicator. Thus a number of criteria / indicators may be required, each of which may carry different values of importance.

• Relating the water use of each activity to indicators/criteria is outside the scope of this project. It is recommended that the undertaking that follows this project is to identify suitable indicators/criteria. It is recommended that these indicators/criteria be developed/selected in consultation with multiple-criteria decision analysis experts (MCDA), water resource managers, stakeholders, and economic, social and environmental experts.

The following concluding remarks can be made regarding the responsibilities and challenges faced by the DWAF SEA:

• The SEA may be required to provide information regarding water use at various scales (National, WMA or local scale),
• The nature of the water use information required by the SEA may vary for the different scales,
• Criteria and indicators of economic, social and environmental water use may need to be developed in order for a process of assessment to be undertaken,

• Assessing water use conflicts and challenges often requires a clear understanding of the conflicts and challenges from a number of varied perspectives that may be faced by various stakeholders and water resource managers. Thus the SEA is required to assess conflicts and challenges in a holistic, yet multi-faceted approach. This often requires a broad level of skill and strong participatory qualities by the SEA, which is a challenge in its own right.
• The nature of the information required from the SEA process largely depends on whom will use the information, and how the information is to be used,
• Water use conflicts and challenges may vary considerably from one spatial location to the next,
• Water use indicators and criteria may need to be temporally sensitive or specific (e.g. water use during dry vs wet periods),
• A fine-time step physical processed based hydrological model is ideally required to simulate (estimate) the water use by current and potential water using activities. This affords a generic method of estimating water use, both at different spatial scales (e.g. local vs WMA vs National), as well as for different spatial locations. The ACRU model has been selected as a suitable hydrological model to be used for the development of a scenario generator, and
• A number of indicators of water use are possible. It is important to select the appropriate water use indicator/s that address the needs of the SEA client (and/or information user/s), and
• There are two broad approaches to water conflict, which need to be included in the scenario generator, including

• changes to the demand for water (i.e. change how, when or where water is used), and
• changes to the supply of water.

From the discussion above, the following important statements were made:

• The focus of the DWAF SEA relates to water use.
• There are two broad categories of water related scenarios that can be run, including

• changes to the demand for water (i.e. changes to the use of water), and
• changes to the supply of water.

The main objective of developing a hydrologically focussed scenario generator is to have a tool that can easily be used to generate water related scenarios, which broadly include scenarios influencing the demand for water and/or changes to the supply of water. The reason for developing this tool is relatively straightforward. The SEA is tasked to assess the water use of current and potential water use and supply conditions. The value of the scenario generator is to assist in the generation of accurate, meaningful water use and supply scenarios.

It is suggested that the scenario generator be GIS based, preferably in ArcView (as ArcView is easy to use, and has large functionality). Developing the scenario generator in ArcView has the potential to allow easy-to-understand, and realistic scenarios to be generated in a transparent manner. The technical challenge is to seamlessly integrate the ArcView SG with the ACRU hydrological model. Both ArcView and ACRU will also need to be seamlessly integrated to a carefully designed database. The seamless integration of the SG to ACRU may require that when certain water demand and supply scenarios are invoked, the user of the scenario generator is prompted for information that may be required by the ACRU hydrological model. For example, if a scenario to build a dam is invoked, the user may be prompted to supply information regarding, amongst other details, the capacity of the dam, the surface area of the dam and the release capacity of the dam.

A few of the requirements of the scenario generator are discussed in more detail.

• Generic

The term generic refers to the requirement that the SG should be able to accommodate water supply and demand scenarios at different scales.

• Temporal considerations

The ACRU model is a daily time step model. The outputs of the model (e.g. streamflow) may easily be aggregated to coarser time periods (e.g. monthly or annual streamflow).

• Spatial considerations: size

Although the ACRU model was developed with application to small-scale catchments (<30km²), the model has been successfully applied to Quaternary Catchment scales (Schulze, 1995). The Quaternary Catchment is currently the smallest recognised hydrological response unit. The implication is that ACRU can in fact be configured to units smaller that Quaternary Catchment scale, which is useful as this level of functionality may be required in the future. The ACRU model hydrological simulations will be undertaken at Quaternary Catchment scale or finer, which can then be aggregated into Tertiary, Secondary or Primary Catchment scale.

• Spatial considerations: location

As the ACRU model is a physical process based hydrological model, the model can be applied generically throughout South Africa. The model has already been extensively used and verified in numerous locations in South Africa (Schulze, 1995).

Thus the scenario generator, seamlessly integrated with the ACRU model, may be regarded as being generically applicable with respect to both temporal and spatial considerations.

• Water demand and supply scenarios

A requirement of the scenario generator is to generate water demand and supply scenarios that are consistent with the NWA. Water supply and demand scenarios are discussed in more detail in Section 8.4.

The table below (Table 2) illustrates some of the information types required from a SEA as identified in the Mhlathuze SEA document (Steyl et.al., 2000). In the same table comments have been made in the right hand column regarding the advantage of developing the scenario generator in ArcView.

Water use activities can be delineated based on

• hydrological considerations;
• the location of the water use;
• environmental considerations;
• economic considerations; and
• social considerations.

More detail on each of these activities is provided in the following sections.

The availability of water in a system is dependent on how and where water is used within the system. In other words the scenario generator should be sensitive to the location of water using activities, as well as the nature of the water using activity.

The manner in which water is allocated and used by water uses in a system varies, and the ability of water resource managers to control the use by these users also varies. The manner in which the water is used by the various activities may be divided into three broad categories, including:

Table 3 below illustrates the 3 broad categories of water using activities, and a few examples of each category, as well as important considerations associated to each category.

Table 3 Examples of water using activities and some important considerations

Note:

The delineation of the categories above is based water quantity considerations. Further delineation may be required if water quality is considered.

The scenario generator needs to be sensitive to the location of water using activities with respect to:

The sensitivity of the scenario-generator to this level of spatial detail may allow very specific scenarios to be run. For example if it is found that a given QC is stressed, one may wish to assess reducing say SFRAs in the QC. The importance of the relation of a water use in relation to engineering structures is that these structures may have a significant impact on the availability of water in the system. An SFRA located above a dam may have a significantly different impact on the availability of water to the system that the same SFRA located below a dam.

The delineation required here is very similar to considerations related to the spatial location of a water use. The following delineations are important with respect to spatial location of water uses in relation to environmental considerations:

The scenario generator must be sensitive to the spatial location of water uses in relation to

Table 4 shows an example of the division of water use into categories necessary for the scenario generator to permit specific scenarios to be generated which are consistent with the NWA (1998).

Notes:

• Consideration may need to be given to the location of the above-mentioned activities with respect to Quaternary Catchments, important dams, lakes, canals etc.
• More detailed delineation may be required for water quality considerations.

The SG should be able to accommodate scenarios relating to changes to the demand and supply of water.

• Changing the demand for water

The demand for water may be influenced in the following broad ways:

• Direct control

• Water use licenses and license conditions may be used to directly control how, where and sometime when water is used. This form of water demand management is to be included in the scenario generator, and is discussed in more detail below.

• Indirect control

• Water use charges may be levied on registered or licensed water users (Perkins, 2000). The charges do not directly influence the use of water. This type of demand management will not be included in the development of a generic hydrologically focussed scenario generator, as it is very difficult to predict how water users will respond to water use charges, and water use trading.

• Suasion

• Demand for water may be influenced by water resource managers making a plea for certain water users to change their demand for water. The effectiveness of suasion to influence demand is very dependent on the moral fibre of the water users. It is not possible to include suasion into the scenario generator, as there can be no guarantee that society will change demand in response to suasion.

Certain categories of water use require water use licenses to use water legally (e.g. SFRAs, irrigation, industrial water use). Water use licenses legally permit a given category of water use (e.g. irrigation) to be undertaken by the holder of the water use license. There are however generally conditions attached to water use licenses, which further stipulate when, how and how much water a given user may use, and the circumstances under which the water may be used. The scenario generator should therefore be able to accommodate scenarios, including amongst others:

Water engineering structures are usually constructed in order to increase the water yield in the system. The scenario generator needs to accommodate:

Water engineering structures are usually financed by one or a number of water users. The amount of water a given water user has call to, is determined by the category of water use licenses held by water users having a call on the engineering structures, and the conditions of water use associated with a given water use license (as discussed above). However, there may be operating rules associated to a given water use structure, or combination of water use structures, which may be undertaken to achieve two broad conflicting goals, which include:

• The objective to maximise water yield, and
• The objective to minimise risk associated with flooding.

The system operating rules thus govern how water is moved within or released from a system, over and above the water that is released to meet legal water demand.

Notes:

• A distinction is made between a model parameter and a model variable,
• A model parameter refers to information related to the input of the model (e.g. dam size, landuse areas, etc), and
• Model variables refer to the output of the model, which are dependent on the configuration of the model, and the values of the input parameters.

Within the two main subdivisions of scenario categories discussed above (i.e. scenarios related to water use licenses, and scenarios related to water engineering structures and system operating rules), a differentiation can be made between "stand-alone" and "linked" scenarios. These sub-categories of scenarios are discussed below in more detail.

Stand-alone scenarios refer to scenarios that have model parameters that do not influence other model parameters. This sub-category of scenarios is probably only representative of dryland crops, of which SFRAs are the most important to water resource managers. The term "independent scenario" may be explained by use of an example:

• If a scenario is run in which say a piece of land currently planted to a non-SFRA is planted to forestry (an SFRA the change to the ACRU parameters (in this case landuse areas) will not influence, or be influenced by other parameters (however will in most likelihood influence model output – or variables). In other words, should a dam be located downstream of the landuse swap, no changes to the dam will be induced by the change in landuse. The scenario is thus independent, in that only model parameters directly related to the scenario are influenced, with no impact on other model parameters.

Stand-alone scenarios should be relatively easy to accommodate, i.e. model configuration and parameterisation should be relatively straightforward.

• Linked scenarios

Linked scenarios refer to scenarios that may require further scenarios to be run (or for model parameters not directly associated to the scenario to be changed). For example, if a "build a dam" scenario is generated, a number of other scenarios will/may need to be considered, such as:

• Who has ownership to the water in the dam?
• When and how must the water be released from the dam (system operating rules)?
• Do the license conditions of the new proposed owners of the dam need to be updated to reflect the presence of the dam?
• Does the dam influence the operating rules of other engineering structures in the system?

Linked scenarios may be quite complicated, and thus setting up a linked scenario may require careful adjustments to water use license conditions and system operating rules respectively. The type of scenarios associated with a linked-scenario may be location specific. In other words, the type of scenarios that may be linked to "build a dam" in say catchment A may be vastly different to those in say catchment B. This will require the modelling system to be very flexible, and will also require that linked scenarios are set up and simulated by experienced personnel.

Although the SG should be designed in a manner that is easy to use, it is recommended that the scenarios generated using the SG are either actively done by, or verified by an experienced hydrologist or simulation modeller. The purpose of the SG is not to replace a hydrologist with computer coding, but rather to allow the scenarios to be generated and run quickly and transparently.

Potential opportunities with respect to the SG include that