Water Resource Protection Policy Implementations

Resource Directed Measures for Protection of Water Resources: Groundwater Component

WATER RESOURCE PROTECTION POLICY IMPLEMENTATIONS

RESOURCE DIRECTED MEASURES FOR PROTECTION OF WATER RESOURCES

APPENDIX GW1:CASE STUDIES AND WORKED EXAMPLES OF INTERMEDIATE RDM DETERMINATION

Senior Author:Roger Parsons, Parsons and Associates

Editor:Lizette Guest, Guest Environmental Management

Guy Pegram, Pula Strategic Resource Management

Heather MacKay, Department of Water Affairs and Forestry

Version 1.0

Date:24 September 1999

M:\f_rdm_october\groundwater\version 1.0\app_gw1_version1.0.doc

CASE STUDY – CATCHMENT G21B

ATLANTIS

0.The town of Atlantis, 50 km north of Cape Town, has been supplied with water from the Atlantis Aquifers since 1976. Exploration work, wellfield development and ongoing monitoring and re-evaluation has resulted in the aquifer being reasonably well understood.

A good conceptual model of the site has been developed by the CSIR (Fleisher and Eskes, 1992; Wright, 1991; Fleisher, 1990; Tredoux et al., 1989). This in turn led to development of a numeric model. The area has also been investigated by DWAF (Bredenkamp and Vandoolaeghe, 1982; Vandoolaeghe and Bertram, 1982) and the IGS (Muller and Botha, 1986). The level of confidence at which the groundwater component of the IRD assessment can be defined, therefore, is ‘high’ to ‘very high’.

1.The total area of Catchment G21B is 304 km2 (see map[1]). The catchment was divided into three homogeneous response units (Unit 1, Unit 2 and Unit 3). Division was based on geological and geohydrological characteristics and the 1 : 500 000 scale general geohydrological map of the area (DWAF, 1995). Units 1 and 2 comprise primary aquifer systems where essentially unconsolidated sand deposits overly greywackes and shales of the Malmesbury Group. Aquifers in Unit 3 comprise weathered and fractured rocks of the Malmesbury Group.

2.Three distinct geohydrological region types were identified. A sea water geohydrological region extending 2 km in from the coast line was demarcated in Units 1 and 2 while the Witzand and Silwerstroom wellfields were demarcated as high volume abstraction geohydrological regions. Remaining areas were classified as terrestrial vegetation geohydrological regions. However, it is questionable whether the natural vegetation (coastal fynbos) obtains water from the aquifer. Natural fynbos and strandveld vegetation probably obtain water from unsaturated soil and upper sandy horizons (Scott, pers.comm., 1999).

3.Water levels have been monitored regularly since development of the groundwater abstraction scheme. Water level contour maps are available for each geohydrological region. Extensive groundwater quality monitoring has also been carried out. Current geohydrological conditions do not differ significantly from the reference conditions.

Groundwater quality in the catchment is essentially good with EC typically being in the order of 100 mS/m. The water displays a range of characters. Younger water tends towards a Ca HCO3 character while older water tends to a NaCl character. Water abstracted from the wellfields is hard and requires treatment before distribution.

4.The present status category of the High Volume Abstraction geohydrological region in Unit 1 was set at D as large volumes of groundwater are abstracted, but no signs of over abstraction are evident. Total present abstraction amounts to about 60% of the estimated groundwater allocation.

Groundwater levels in the vicinity of the two wellfields have dropped by 5 m to 7 m over the past 20 years, but have now stabilised (Tredoux, pers.comm., 1999). Affected areas measure 4.5 km2 (Witzand wellfield) and 3 km2 (Silwerstroom wellfield) respectively. This amounts to about 2.5% of the total catchment area.

No visible signs of any impact caused by groundwater abstraction are visible. It is pertinent to note the Silwerstroom spring is still flowing. This is in spite of continued groundwater abstraction from the Silwerstroom wellfield during the past 22 years.

The present status of the remainder of the catchment was classified either as ‘A’ or ‘B’ as limited groundwater abstraction takes place elsewhere in the catchment. Groundwater usage in Catchment G21B amounts to some 8.3 x 106 m3/a and was estimated as follows:

Table 1: Estimated groundwater usage

Area / HRU / Geohydrological
Region / Use / Estimated
Volume
(x 106 m3/a)
Witzand wellfield
Silwerstroom wellfield
Silwerstroom spring
Atlantis township
Mamre area
Koeberg wellfield / Unit 1
Unit 1
Unit 1
Unit 1
Unit 1
Unit 2 / HVA
HVA
SF
TV
TV
TV / urban supply
urban supply
urban supply
industry, irrigation
irrigation
industrial, irrigation / 3.7
2.1
0.5
1.0
0.5
0.5

In assessing the performance of the Atlantis aquifer and validity of this IRD assessment, artificial recharge activities in the catchment need to be considered. Approximately 2 x 10 6 m3/a good quality water is artificially recharged into the Atlantis aquifer at Pan 7, located 2 km east of the Witzand wellfield (Murray and Tredoux, 1998). Poorer quality water is recharged via a series of coastal recharge basins located in the sea water geohydrological region in Unit 1.

5.A management class ‘b’ for the high volume abstraction geohydrological regions requires that appropriate monitoring continue and the IRD assessment be reviewed within 3 years. Outside of these areas, the management class was set at ‘a’ and review is thus required within 5 years.

6.Average annual recharge for the entire catchment was calculated (see data sheet). Mean annual precipitation in catchment G21B is 420 mm/s. Recharge calculations excluded the area of the sea water geohydrological regions. A low maintenance baseflow of 0.5 x 106 m3/a was set to ensure continued spring flow. The Buffels River is ephemeral and only flows for sort periods after heavy rainfall.

Until recently, Atlantis was entirely reliant on groundwater as a water source. It was hence assumed all basic human needs are to be met from the groundwater allocation. Groundwater is abstracted from Unit 1 for this purpose. Based on a population of 120 000 requiring at least 25 L/p/d, the BHN component of the Reserve was set at 1.09 x 106 m3/a.

In light of the response of the aquifer to abstraction during the last 20 years, the calculated groundwater allocation for Unit 1 appears reasonable. Those calculated for Unit 2 and Unit 3 therefore are also assumed reasonable. In assessing the validity of the calculated groundwater allocations, the following information was considered:

• some 10 x 106 m3 of groundwater stored in Units 2 and 3 is brackish ( 1 500 mg/L) and is unlikely to be used for domestic or irrigation purposes without blending or treatment;

• saline intrusion and the ingress of poor quality water from Unit 2 and Unit 3 poses a potential threat to Unit 1 and hence needs to be managed water quality monitoring is required;

• approximately 5.8 x 106 m3/a is abstracted from the two wellfields in Unit 1. Abstraction is evenly distributed throughout the year. The gradual lowering of the water level by 5 m over a 20 year period in the vicinity of the wellfields has not resulted in any apparent negative ecological impacts.

• abstraction will be distributed throughout the year, with summer abstraction being only marginally higher than abstraction during the winter months.

Based on quality of data available and duration of monitoring aquifer response to abstraction, the level of confidence of these estimates are ‘high’ to ‘very high’.

The following Resource Quality Objectives and limitations have been set to ensure integrity the aquifer system remains in tact:

• no groundwater abstraction is allowed in the sea water geohydrological region

• the dynamic water level may not drop below 10 mamsl at any time

• static water level outside of the high volume abstraction geohydrological region may not decline over the long term

• static water level in the high volume abstraction geohydrological region may not drop by more than 5 m below the ambient static groundwater level over the long term.

• the dynamic water level may not drop by more than 5 m below static water level for a period longer than 7 days.

7.Example IRD Notice:

To ensure the ability of the groundwater component to satisfy the Reserve on a sustainable basis, the following groundwater allocations, resource quality objectives and drawdown limitations for the three homogeneous response units within Catchment G21B are set:

Unit 1: groundwater allocation is 9.09 x 106 m3/a

Resource quality objectives and drawdown limitations

• static water levels outside the high volume abstraction geohydrological region may not decline over the long-term

• static water levels in the high volume abstraction geohydrological region may not drop by more than 5 m below the ambient static groundwater level over the long term

• dynamic water levels in the high volume abstraction geohydrological region may not drop by more than 5 m below static water level for periods longer than 7 days

Unit 2:Groundwater allocation is 3.45 x 106 m3/a

Resource quality objectives and drawdown limitations

• static water levels may not decline over the long term

Unit 3:Groundwater allocation is 1.05 x 106 m3/a

Resource quality objectives and drawdown limitations

• static water levels may not decline over the long-term

Other:-No groundwater abstraction is allowed in the sea water geohydrological region

• Dynamic water levels may not drop below 10 mamsl

• Groundwater levels and quality must be monitored at least quarterly and a review report submitted to the Catchment Management Agency annually

• Should any of the above conditions not be met or the following changes detected within the catchment, then the IRD assessment must be reviewed by the Catchment Management Agency and the Minister of Water Affairs and Forestry informed thereof within 60 working days of the condition being detected:

• deterioration of groundwater quality; and

• impact on vegetation in the catchment.

• The groundwater allocations of Catchment G21B must be reviewed within three years from the date of publication of this notice.

Should abstraction in the catchment exceed the groundwater allocation or any of the stipulated conditions not be met, the Minister may request a comprehensive assessment be conducted.

Notes:

1. Conservative estimates were used. Some workers estimated recharge could be as high as 40% MAP in places. Fleisher and Eskes (1992) estimated recharge to be 16.3 x 106 m3/a in an area of 141 km2.

1. No IFR assessment was done, based on estimated flow over the last 20 years.

GW1/1

Department of Water Affairs and Forestry, South Africa

Version 1.0: 24 September 1999

Resource Directed Measures for Protection of Water Resources: Groundwater Component

References

Bredenkamp, D.B. and Vandoolaeghe, M.A.C., 1992: Die ontginbare potensiaal van die Atlantisgebied; Gh report 3227, Directorate of Geohydrology, Department of Water Affairs and Forestry.

Department of Water Affairs and Forestry, 1995: General geohydrological map - Cape Town 3317; draft plot, Directorate of Geohydrology, Department of Water Affairs and Forestry.

Fleisher, J.N.E., 1990: The geohydrology of the Witzand wellfield; Report No. 2/90, Groundwater Programme, Division of Water Technology, CSIR.

Fleisher, J.N.E. and Eskes, S.J.T., 1992: Optimization and management of the Atlantis groundwater resource; Report 33/92, Groundwater Programme, CSIR, Stellenbosch.

Murray, E.C. and Tredoux, G., 1998: Enhancing water resources - factors controlling the viability of artificial groundwater recharge; Int.Conf.Proc. WISA Biennial Conference, Cape Town, May 1998, Paper 1C-5 pp1-8.

Muller, J.L. and Botha, J.F., 1986: A preliminary investigation of modelling the Atlantis Aquifer; IGS Bulletin 14, University of the Orange Free State, Bloemfontein.

Scott, D., 1999: Personal communications.

Tredoux, G., Hon, A.J. and Engelbrecht, J.F.P., 1989: A groundwater model for the Atlantis Aquifer; Project No. 670/2603/9, Programme, Division of Water Technology, CSIR.

Tredoux, G., 1999: Personal communications.

Vandoolaeghe, M.A.C. and Bertram, W.E., 1982: Atlantis grondwatersisteem - herevaluasie van versekerde lewering; Gh report 3222, Directorate of Geohydrology, Department of Water Affairs and Forestry.

Wright, A.H., 1991: The artificial recharge of urban stormwater runoff in the Atlantis Coastal Aquifer; unpubl. M.Sc. Thesis, Rhodes University, Grahamstown.

GW1/1

Department of Water Affairs and Forestry, South Africa

Version 1.0: 24 September 1999

Resource Directed Measures for Protection of Water Resources: Groundwater Component

Case Study - Catchment G21B

DATA SHEET

Atlantis

BOUNDARIES AND TYPING / MANAGEMENT CLASS / RECHARGE / ADJUSTMENTS / ALLOCATION
Homogeneous
Response
Unit / Geohydrological
Region
Type / Present status / Management Class / Total
Area
(km2) / Effective
Area
(km2) / Recharge Method / Annual Recharge
(106 m3) / Low Maintenance
Baseflow Adjustment
(106 m3) / BHN Adjustment
(106 m3) / GRU
Groundwater
Allocation
(106 m3) / Confidence
Unit 1 / Terrestrial Vegetation / A / a / 34 / 34 / 20% MAP / 2.86 / 0.50 2 / 0.00 / 9.92 / very high
Sea Water / A / a / 22 / 0 / exclusion / 0.00 / 0.00
High Volume Abstraction / D / b / 103 / 103 / 20% MAP / 8.65 / 0.00 / 1.09
Unit 2 / Terrestrial Vegetation / A / a / 59 / 59 / 15% MAP / 3.72 / 0.00 / 0.00 / 3.72 / high
Sea Water / A / a / 32 / 0 / exclusion / 0.00 / 0.00
Unit 3 / Terrestrial Vegetation / A / a / 54 / 54 / 5 % MAP / 1.13 / 0.00 / 0.00 / 1.13 / high
Total for Significant Water Resource / 16.36 / 0.50 / 1.09 / 14.77

Case Study 1: Catchment G21B including the Atlantis Aquifer

GW1/1

Department of Water Affairs and Forestry, South Africa

Version 1.0: 24 September 1999

Resource Directed Measures for Protection of Water Resources: Groundwater Component

CASE STUDY - CATCHMENT N13A

GRAAFF -REINET

0.The Moordenaars groundwater unit, located some 20 km west of Graaff-Reinet has been used to supplement the town’s water supply for the last 10 years. Exploration work, well field development and ongoing monitoring and re-evaluation has resulted in the response of the aquifer to abstraction being reasonably well understood.

A good conceptual model of the site was developed by Parsons (1987) while Steffen Robertson and Kirsten (1988) installed a series of production boreholes and established the groundwater production scheme. Routine monitoring of the response of the aquifer to monitoring was also carried out by SRK (1992). Due to declining groundwater levels, the yield of production boreholes were re-assessed (Woodford, pers.comm., 1999).

1.The total area of Catchment N13A is 430 km2. However, the catchment was divided into three homogeneous response units (Unit 1, Unit 2, Unit 3). Division was based principally on topographical, geomorphological and geohydrological characteristics.

2.Three geohydrological region types were also recognised, namely a high volume abstraction geohydrological region (only Unit 2), a riparian vegetation geohydrological region and a terrestrial vegetation geohydrological region. Division was based on the hydrological function of each geohydrological region type in the catchment.

3.An extensive hydrocensus was carried out by Parsons (1987) and static piezometric levels in the catchment were defined. These conditions essentially represent ambient groundwater levels (reference conditions) as groundwater abstraction in the catchment was limited at that stage. Groundwater flow mimics surface water drainage patterns. Parsons (1987) referred to a groundwater flow convergence zone which follows a path similar to that of the Moordenaars River. Groundwater levels fluctuate by about 3 m in the central low-lying parts of the aquifer system.

Spring flow at Corndale was estimated at 50 000 m3/a. Borehole yields are highly variable, but generally are ‘low’ to ‘moderate’ (92% of boreholes have yields less than 5 L/s). High yields (+20 L/s) are associates with specific geological features. Hydraulic conductivity ranges between 0.5 m/d and 10 m/d while storativity of the rock is in the order of 0.001. The hydraulic gradient at Corndale was measured at 0.0043.

Groundwater quality in the catchment is good. The 25 th, 50 th and 75 th percentile of electrical conductivity are 80 mS/m, 120 mS/m and 180 mS/m respectively. The water has a NaCl character and is thus classified as a ‘B’ type water. No groundwater contamination has been identified and groundwater quality is considered representative of pristine conditions.

Groundwater abstraction for agricultural purposes (domestic, stock watering) amounts to about 400 000 m3/a while abstraction from the Mimmosadale wellfield amounts to some 240 000 m3/a.

4.The wellfield used to supplement Graaff-Reinet’s water supply has been over-pumped (Woodford, pers.comm., 1999). This resulted in a steady decline of groundwater levels. However, the impact remain localised (ie. within 100 m of the production boreholes). Management actions were implemented and rate of abstraction was reduced from 665 m3/d to 180 m3/d. This resulted in a rise in groundwater levels. The current rate of abstraction appears to be sustainable over the long term. Because of over-abstraction, present status category of the high volume abstraction area in Unit 2 was set at ‘E’.

Outside of this region limited groundwater abstraction takes place. The groundwater is used for farm domestic supply, stock watering and small scale irrigation. Current conditions therefore are similar to the reference conditions described above.

5.As a result of the over-abstraction in the high volume abstraction geohydrological region, a ‘c’ management class was set. This requires that a comprehensive Reserve determination be initiated. Until such time as the more detailed determination can be completed, only 75% of the groundwater allocation determined by the IRD method should be abstracted.

6.Using the formulae presented by Kirchner et al. (1991), the average annual recharge in each geohydrological region type was determined (see data sheet). A MAP of 380 mm/a was used in Unit 1 while the MAP of Units 2 and 3 was taken as 295 mm/a.

In setting the groundwater allocation, no corrections were made for baseflow as the Moordenaars River is ephemeral. Further, the population in the catchment is limited (less than 200) and all basic human needs are met from groundwater resources.

In assessing the validity of the groundwater allocations presented in the data sheet, the following information was considered:

• some 0.4 x 106 m3/a is abstracted throughout the catchment for agricultural purposes while approximately 0.24 x 106 m3/a is abstracted from the Mimmosadale well field for town supply.

• a localised lowering of the water level in the vicinity of the well field has occurred, but no detrimental ecological impacts have been reported.

Based on the quantity and quality of data available for the IRD assessment, the level of confidence of the assessment was set as ‘medium’.

As the calculated long-term average annual recharge of Unit 2 and the volume of groundwater abstracted (mostly from the high volume abstraction geohydrological region) are similar, the IRD assessment is considered inappropriate and a more detailed comprehensive assessment is required.