Report on water demand and deficiency by Partner 7, CSIR South Africa Modeling Mkomazi water demand and deficiency The activities for both these work tasks are reported collectively for the Mkomazi due to the integrated nature of the CSIR’s involvement in both the demand and deficiency modelling in this catchment, compared to only deficiency modelling for the individual Mbuluzi and Mpfure catchments. The collation of suitable input data for modelling water demand and the final production of spatially referenced demand data for the Mkomazi were completed in close collaboration with the partner 6 University of Natal), because of the associated overlap of technical expertise in terms of hydrological and spatial modelling. Suitable environmental, socio- economic and topographic datasets were sourced from a variety of organisational and published sources. Significant inputs and assistance was received from both Umgeni Water and the Department of Water Affairs and Forestry, who are also key potential end-users of the IWRMS project deliverables. Digital GIS coverages that spatially represent the various demand / deficiency scenario’s have been produced, and are available in the project database (in ArcInfo format). Three different approaches have been used to develop water “demand / deficiency” scenario’s for the Mkomazi (South Africa). The primary approach has been to model stream flow and associated demand / deficiency values for different present and future scenario’s using the ACRU hydrological model. These results, which constitute the definitive hydrological dataset for the catchment,are reported in Taylor et al (2000). Two alternative approaches are reported, which were developed as simplistic, spreadsheet-based alternatives to final demand / deficiency modelling within ACRU (although outputs from initial ACRU model runs are used as primary inputs in some of these spreadsheet models). These two procedures are presented for comparative purposes only. It is important to realise that in all three modelling approaches, the same fundamental spatial frameworks have been used to ensure a degree of standardisation. These are based on the same sub-quaternary catchment cells, and the land-use patterns described in the South African National Land-Cover database (Thompson, 1999; Fairbanks et al , 2000). The two alternative modelling scenario’s provide a more generalised overview of current and future demand / deficiency conditions in the Mkomazi catchment, based on very specific sectoral demand and abstraction parameters (not all of which are directly comparable to the primary ACRU model outputs). The first approach uses a simple balancing-model approach, based on the difference between ACRU simulated natural baseline (i.e. pristine vegetation conditions), and ACRU-modelled stream flow conditions for various sector abstraction conditions. The second
17
Embed
full report csir - Herzlich Willkommen auf der Webseite ... fileReport on water demand and deficiency by Partner 7, CSIR South Africa Modeling Mkomazi water demand and deficiency The
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Report on water demand and deficiency by Partner 7, CSIR South Africa
Modeling Mkomazi water demand and deficiency
The activities for both these work tasks are reported collectively for the Mkomazi due to the
integrated nature of the CSIR’s involvement in both the demand and deficiency modelling in this
catchment, compared to only deficiency modelling for the individual Mbuluzi and Mpfure
catchments. The collation of suitable input data for modelling water demand and the final
production of spatially referenced demand data for the Mkomazi were completed in close
collaboration with the partner 6 University of Natal), because of the associated overlap of
technical expertise in terms of hydrological and spatial modelling. Suitable environmental, socio-
economic and topographic datasets were sourced from a variety of organisational and published
sources. Significant inputs and assistance was received from both Umgeni Water and the
Department of Water Affairs and Forestry, who are also key potential end-users of the IWRMS
project deliverables. Digital GIS coverages that spatially represent the various demand /
deficiency scenario’s have been produced, and are available in the project database (in ArcInfo
format).
Three different approaches have been used to develop water “demand / deficiency” scenario’s for
the Mkomazi (South Africa). The primary approach has been to model stream flow and
associated demand / deficiency values for different present and future scenario’s using the ACRU
hydrological model. These results, which constitute the definitive hydrological dataset for the
catchment,are reported in Taylor et al (2000). Two alternative approaches are reported, which
were developed as simplistic, spreadsheet-based alternatives to final demand / deficiency
modelling within ACRU (although outputs from initial ACRU model runs are used as primary
inputs in some of these spreadsheet models). These two procedures are presented for
comparative purposes only. It is important to realise that in all three modelling approaches, the
same fundamental spatial frameworks have been used to ensure a degree of standardisation.
These are based on the same sub-quaternary catchment cells, and the land-use patterns
described in the South African National Land-Cover database (Thompson, 1999; Fairbanks et al,
2000).
The two alternative modelling scenario’s provide a more generalised overview of current and
future demand / deficiency conditions in the Mkomazi catchment, based on very specific sectoral
demand and abstraction parameters (not all of which are directly comparable to the primary
ACRU model outputs). The first approach uses a simple balancing-model approach, based on the
difference between ACRU simulated natural baseline (i.e. pristine vegetation conditions), and
ACRU-modelled stream flow conditions for various sector abstraction conditions. The second
approach is based on the spatial disaggregation of pre-published sector abstraction demands,
that have been spatially re-distributed on the basis of associated land-use patterns that have
been derived from a ‘standard’ land-cover database. The objective being to compare the variation
in outputs from very different modelling approaches, with different levels of both technical and
scientific complexity, but with essentially the same objectives and spatial frameworks.
Results
1 GIS data collection and modeling : Mkomazi water demand (WT 720)
1.1 Results from the ACRU-based modelling
A series of demand / deficiency scenario’s were generated using data generated within the ACRU
hydrological model. This modelling process is documented in Taylor et al (2000). What is
described below are the key ACRU derived datasets that have been used to generate the GIS
spatial coverages that illustrate specific demand / deficiency scenario’s. The following ACRU
model outputs were used :
• Baseline flow : values represent simulated, accumulated stream flow the scenario of
for rural, municipal, livestock and irrigation sectors under both present and future
scenarios. 1 The same livestock demand levels have been used for the future scenario, since no significant change in livestock population is expected up to the year 2050, up to which future predictions were made (Meigh et al 1998)
2. total demand for all sectors combined = baseline runoff – integrated sector runoff (after
abstraction), for both present and future scenarios.
3. % natural baseline runoff remaining after total demand abstractions = integrated sector
runoff (after abstraction) expressed as a percentage of baseline runoff, for both present
and future scenarios. Since no environmental reserve has been calculated for the Mbuluzi
catchment this approach has been used to provide an estimate the ‘impact’ of the various
abstraction demands, and the amount of remaining runoff available to the environment.
(The results of the above calculations are listed in the excel spreadsheet “mbuluzi modelling data
for report.xls” which is supplied along with the digital GIS coverages ).
In some scenarios it appears that apparently erroneous data has been generated, for example,
where the simulated combined runoff after deduction of integrated sector abstraction exceeds the
simulated natural baseline runoff. In sub-catchment 5, baseline MAR = 74.53 Mm3, whereas MAR
after total combined abstraction = 75.37 Mm3.. This is explained however by an understanding of
the assumptions in the modelling process. Under baseline conditions, it is assumed that the
natural vegetation is in pristine condition, with no erosion or degradation. This infers high leaf or
canopy interception, infiltration, water up-take by roots and likewise, transpiration rates. Under
‘present’ (and future) conditions, the vegetation cover is in many localities degraded. This may
lead to a reversal of the hydrological response as there could be less (vegetative) interception,
and likewise, infiltration, and transpiration, with the end result that simulated runoff may actually
increase in comparison to baseline conditions.
The abstraction sector (per sub-catchment), with the greatest simulated demand is in most cases,
the same for both annual and driest month conditions. Only in a few sub-catchments does the
dominant abstraction sector change between an annualised and driest month basis. A single,
digital GIS coverage has been created which contains the various modelling scenarios within the
attribute parameters. Duplicate copies of this coverage have been generated in both ArcInfo and
ArcView format (geographic coordinates). Selected mean annual and mean monthly (August)
demand and runoff spatial modelling scenarios are illustrated below (in Mm3 or %) :