National Water Quantity Data

User License & Data Access

SANSA is granted for the contractual period a non-exclusive, multi-user license to use and distribute the supplied monthly national surface water extent and associated dam volume information in both spatial and non-spatial formats to all national government departments, provincial and municipal sub-departments, and any of their legally associated entities.

Both SANSA and the indicated government departments (and associated legal entities) may use this information internally, or provide access to independent 3rd parties specifically undertaking contractual work on behalf of SANSA or any of the indicated government departments.

Neither SANSA, the indicated government departments or any contracted 3rd party may use the supplied information in order to develop a competitive, comparable national, monthly water and volume modelling capability and service, either for South Africa or elsewhere internationally..

At the end of the contractual period the supplied monthly surface water and volume data can be retained by SANSA and the government departments and associated legal entities. All 3rd, non-government parties provided with access to the monthly surface water and volume data must delete all copies of the data from their computer systems and data archives, once the contracted work has been completed. No 3rd party may retain nor use the SANSA or department supplied data for their own use or purpose. It is the responsibility of SANSA to ensure that these conditions are applied to by both departments and any departmentally contracted 3rd parties.

The multi-user license does not allow public dissemination of the supplied spatial and non-spatial data.

Monthly National Dam Water Volumes

Deriving the information

The surface water area information generated by the process described in section 3 is used to generate an estimate of the monthly National Dam Water Volumes. This information is only generated for the man-made water bodies in a catchment. The reason for this as the main distinction is that the product is designed to provide an estimate of the usable water supply volume in a particular area, for this reason natural water bodies in South Africa have been excluded as they tend to be of a temporary nature and in most cases are not sustainably exploitable for water supply. In order to obtain this information the following elements have been undertaken.

Man-made versus natural water

A mask has been generated which encompasses all the water supply dams identified across South Africa. This mask distinguishes between natural and man made water bodies. Each water body has been identified and buffered according to a formula. This is to ensure if a dam is above the identified full supply level/area that we are still able to capture the volume. The mask has been separated into 4 different categories, namely:

  1. All large dams already monitored by DWS are Category 1 dams
  2. All dams that are on the Dam Safety Database are classified as Category 2 dams
  3. All dams remaining dams larger than 25 000 m3 are Category 3 dams
  4. All the remaining dams that are considered man made are Category 4 dams

Equations

A set of equations have been developed to determine the surface water volume from the surface water areas identified on the satellite. An equation has been developed for each dam identified as man-made across the country. The method of deriving these equations is different for the different classes of dams. This process is described in detail in the paper written for this purpose . The equations were determined using the following methodology for each of the different dam categories.

Derived volume estimates

The derived volume estimates have been generated from the process where for each individual dam the equation is used along with the monthly area estimates to determine the monthly volume estimates, this information has also being used to derive the full supply volume for these particular dams. Thus FLC information plus the present supply volume as a percentage of the FLC is provided for each of the dams.

Estimates of accuracy and uncertainty

There are several different elements which can impact the accuracy of the volume estimates, these include:

  1. Errors of commission (Water surface identified but does not exist)
  2. Errors of omission (Water surface identified as something else)
  3. Fractal or pixilation errors resulting from the pixel size where water is identified
  4. Bias in determining Volume Area equations
  5. Uncertainty in deriving single catchment equations
  6. Errors in curve fitting VA equations

Catchment based water resources availability

The combination of the information provided in this particular product are thus able to provide us with and assessment of the overall water availability in a catchment. The idea is that we are able to calculate the current volume of water against the full supply capacity of water resources in an area and thus obtain an estimate of the overall supply in an area. These assessments have been done at a quaternary catchment level. There are two sets of information provided here, namely:

  1. Category 1 dams: the percentage full and actual volumes are supplied
  2. Category 2,3 and 4 dams are all lumped together at a quaternary catchment level to provide and overview of the percentage of full supply capacity as well as the estimated volumes for those dams

References

M Thompson; J Hiestermann, B Eady and J Hallowes. (2018) Frankly my dear I give a dam! Or Using satellite observation to determine water resource availability in catchments. SANCIAH conference 2018 proceedings.