MANUAL ON PUMPING TEST ANALYSIS IN FRACTURED-ROCK AQUIFERS
Report No 1116/1/02
March 2002

EXECUTIVE SUMMARY

1. Background and Motivation

This report presents the findings of research conducted by the Institute for Groundwater Studies for the Water Research Commission on the development of a manual to analyse pumping tests in fractured-rock aquifers. The manual provides a detailed summary of the different types of pumping tests and the methods that can be used to analyse data from these tests. The sustainable yield of a borehole can be defined as the discharge rate that will not cause the water level in a borehole to drop below a prescribed limit. Determining the sustainable yield of a borehole is important for the overall management of the aquifer and therefore a whole chapter has been devoted to discussing the term. Protection is an important consideration when managing a well field. For this purpose, computer software, BPZONE, has been developed and included in the FC_EXCEL software. The manual includes a step-by-step guide to assist the reader in planning and executing pumping tests. Finally, recommendations and conclusions are drawn.

Pumping tests are important tools that provide information on the hydraulic behaviour of a borehole, the reservoir and the reservoir boundaries. All this information is essential for efficient aquifer and well field management. In general, the objectives of a pumping test are:

• To obtain an understanding of the aquifer,
• To quantify the aquifer's hydraulic and physical properties and
• To determine the sustainable yield and efficiency of a borehole.

The interpretation of pumping test data is based on mathematical models that relate drawdown response to discharge in the abstraction borehole. The results obtained from these short duration tests can then be used to project the borehole's performance over a long period of time. In fractured-rock aquifers, the geometry and permeability of the system have a large influence on the drawdown. The scale of heterogeneity in a fractured rock system may be large in relation to the scale of the test. Therefore convention models developed for homogeneous porous aquifers might not be viable in fractured rock systems. This manual focuses on methods and software specifically developed to analyse pumping tests conducted in fractured-rock systems.

2. Statement of Objectives

The objectives of this research project are:

• To develop a scientifically sound manual to conduct and evaluate pumping tests in fractured rock aquifers. This manual will be a step-by-step guide for performing and analysing various types of pumping tests such as step draw down and constant rate tests.
• To determine the optimal rate for a constant rate test.
• To use pumping test data to delineate borehole protection zones.
• To enhance the FC_EXCEL software. On the completion of the project, this software will include the following:
1. Tools to delineate borehole protection zones, taking into account point source pollution such as pitlatrines.
2. Methods to analyse step drawdown and multirate pumping tests.
3. Methods to analyse slug tests.

3. Meeting the Objectives

All the objectives of this project were met. In fact, the project team did more than what was required:

• The project team also developed TPA. Test Pumping Analysis or TPA is a windows-based software program, developed in Delphi. The software is a curve fitting procedure for analysing fractured rock pumping test data. TP A includes the following methods to analyse fractured-rock pumping test data:
1. Moench method for double porosity aquifers.
2. Gringarten, Kazemi, Warren and Root and Stallman methods for single and vertical fractures. Included here are uniform flux, finite- and infinite conductive fractures, boundary conditions and a solution for a dyke aquifer.
3. Solutions for porous aquifers. iv. Diagnostic plots.
• One of the initial objectives of the project was to enhance the FC_EXCEL software to include protection zones and step drawdown, multirate tests, and slug tests. However, the project team went beyond the requirements and included:
1. Barker's generalised radial flow model, which uses non-integer flow dimensions to analyse pumping data.
2. A method to estimate the sustainable yield of a borehole by using the data collected during the step drawdown test (non-linear solution).
3. Comparing the suitability of water to the water quality classes developed by the Department of Water Affairs and Forestry.

4. Summary of Methods and Results

Theoretically, the best method to obtain fractured-rock aquifer parameters is a three- dimensional numerical flow model. However, the data required for these models are usually not available. The emphasis of the manual is therefore on using analytical procedures to analyse pumping test data.

The geometry of the fracture system is seldom known when performing a pumping test. Interpretation of test data may therefore have to include identification of the fracture system. Hydraulic conductivity and storativity values for these systems are scale dependent because the rocks are heterogeneous.

A characteristic of fractured aquifers is that most of the water flows along fractures. These fractures are usually embedded in porous matrix blocks (sandstone) or micro fissured blocks (quartzite), which are low permeable compared to the fracture conductivity but are able to store water. If fractures are densely interconnected, they form a 'fracture network continuum' characterised by a large storage capacity that contributes substantially to the yield of a borehole. Whether a fracture network can be considered as continuum or not is determined by the following three properties: the representative elementary volume, fracture connectivity and the difference in conductivity between the fracture and rock matrix.

The spatial distribution of fracture flow towards a borehole depends on the properties of the fracture network, such as fracture permeability, fracture density and connectivity. The characteristics of the fracture network determine the flow geometry, which is also known as flow dimension. The flow towards a fully penetrating borehole may be radial (two- dimensional) throughout a hard rock aquifer consisting of a highly connected fracture system with an isotropic distribution, or may be spherical (three-dimensional) if the borehole is only partially penetrating. Therefore, for a correct assessment of the hydraulic parameters of a fractured hard rock aquifer, it is necessary to know the flow geometry, or flow dimension.

Two kinds of models are applied in fractured-rock aquifers, namely the single fracture model and the double porosity model. However, the methods for porous media (i.e. Theis and Cooper-Jacob) are still applied for parameter estimation in fractured aquifers. Some of these models have inherent problems. The methods of Theis and Cooper-Jacob are only valid for radial flow.

Numerous field examples have been included in the manual. These demonstrate how to conduct and analyse pumping tests. They also highlight the limitations of the various tests and methods. The results from these tests indicate that monitoring must always be an integral part of aquifer management.

When determining the sustainable yield of a borehole, field examples indicate the yield is a non-unique number, dependent on the abstraction rate.

5. Conclusions

The development of this manual was driven by the need to establish a scientifically sound and documented approach to performing and analysing pumping tests in fractured rock systems. The manual has resulted from extensive consultations with pumping test experts and hours of extensive research on methods to analyse these pumping tests.

It is concluded in this manual that in order to analyse a pumping test correctly, the following important characteristics need to be identified:

• The correct geological conceptual model.
• Inner boundary conditions such as well bore storage effects.
• Outer boundary conditions, for example no-flow boundaries.
• Characteristic flow regimes.

In South Africa, pumping tests are performed for mainly two reasons: to determine the long-term sustainable yield of a borehole and to estimate aquifer parameters. The manual highlights the non-linear relationship between the abstraction rate and the drawdown, which is common in most South African aquifers. Fracture dewatering is dependent on the abstraction rate, which, in turn, will have an effect on the water level in the borehole. This feature should be treated with caution when assigning sustainable yields.

It will always be a challenge to perform and analyse pumping tests in a fractured-rock environment. This manual therefore highlights many of the limitations involved with each of the methods discussed. Hopefully the reader will be aware of many of the mistakes that can so easily be made when conducting these valuable tests.

6. Recommendations

When the objective of a pumping test is to determine aquifer parameters, the following is recommended:

• Perform a minimum one-hour step drawdown test to determine the position of the fracture and abstraction rate for the constant rate test.
• The recommended minimum duration of the constant rate pumping test is 8 hours. RPTSOL V is an excellent two-dimensional model which analyses pumping test data and calculates an accurate storativity value.

When calculating the sustainable yield, it is recommended that on the completion of the constant rate pumping test, the FC_EXCEL software is used to analyse the results.

Sustainable development of groundwater resources refers to a holistic approach to development, conservation and management of these resources. Good sustainable yield estimates are the result of scientifically sound pumping test analysis. It must be noted that these quantities vary with time and location and can only be estimated, and thus may carry a degree of uncertainty. All sources of uncertainty need to be recognised and their impact on water quality or the sustainability of the aquifer must be evaluated.

An essential consideration in tile development of an aquifer is the chemical quality of water produced, as the quality can limit its use. Groundwater movement induced by pumping may change the groundwater chemistry. Currently, the FC_EXCEL software includes methods to delineate borehole protection zones from certain microbial pollution sources. This section can be developed further to include chemical contaminants.

A large number of pumping tests that were performed on boreholes situated in fractured rocks in South Africa indicate that a non-linear relationship between drawdown and abstraction rate usually exists. Such non-linearities must be treated with caution when assigning sustainable yields for boreholes or when applying numerical flow models.