Balthazar Th. Verhagen

Schonland Research Centre, University of the Witwatersrand, Johannesburg, South Africa

Thirty years ago, ground water in the Kalahari thirstland, along with many other semi-arid to arid, sand-covered regions of the world, was regarded to be fossil, the remnant of earlier, "pluvial" recharge events. Evapotranspiration an order of magnitude higher than rainfall was reasonably seen to quantitatively account for any infiltrating rain water. This view was supported by observed deep ground water levels, the lack of obvious level responses to rainfall events as well as water quality considerations.

Early isotope surveys (Mazor et al. 1975, Verhagen et al. 1974) of ground water in the Kalahari of Botswana, challenged this concept. The demonstrable effect of ongoing rain recharge in the present semi-arid climatic phase started changing the geohydrological mindset from the pure ground water mining approach to the possibility of managed exploitation. This has both positive and negative connotations. If before, the rural subsistence and even commercial cattle ranching activities made insignificant and diffuse demands on ground water resources, the rapid development of the minerals industry and urbanisation required major and often localised water supplies in an area largely bereft of surface water sources. Conflicts can and increasingly do arise around access to resources and it is expected that water will dominate these conflicts. This places a heavy burden of responsible resource assessment, management, and conservation strategy development on the disciplines of geohydrology, and of isotope hydrology, which is increasingly seen as an indispensible tool in these endeavours.

A few highlights are presented of a selection of studies conducted over the past 30 years which will give a sense of the contribution of isotope hydrology to the development of rational resource assessment and management. These should be seen in terms of earlier concepts - some still prevalent - for the Kalahari: that recharge is possible only through very thin or absent sand covers; the existence of "recharge areas" along basin margins and ephemeral drainages and regional ground water flow towards deeper basin sections; and the association between ground water "age" and mineralisation.

Gordonia

Bank infiltration during periodic floods in an ephemeral river in the southern Kalahari of South Africa was conventionally seen as an important input for sustainable regional ground water exploitation from a sand-filled palaeochannel. An investigation of regional isotope and hydrochemical signatures of ground water (Verhagen 1984) showed that the effect of bank infiltration is limited to the immediate surroundings of the river. Ground water at distances greater than a few kilometres from the river is rather derived from diffuse rain recharge only. The palaeochannel therefore could not be regarded as a preferentially recharged zone. A possible inverse relationship between "pluvial" conditions and recharge is postulated on the basis of stable isotope relationships.

Toteng/Sehitwa

In an isotope study of the ground water supply potential of this cattle ranching project area even fairly high values of mineralisation were found in ground water with rather low residence times. Widespread, ongoing diffuse rain recharge could be demonstrated. Stable isotope results suggested surface evaporation before infiltration and ongoing evapotranspirative ground water losses as the cause for the sometimes high mineralisation. This model suggests further that exploitation should decrease mineralisation. This conclusion was confirmed by historical data: some 75 % of 15 production wells showed significant improvement in ground water quality over periods of up to 10 years.

Jwaneng

As opposed to Toteng/Sehitwa, the deltaic aquifer being exploited to supply Jwaneng diamond mine and town produces surprisingly fresh water, with Ca,Mg-HCO3 dominant chemistry. The apparently unusual radiocarbon distribution in the aquifer is interpreted (Verhagen 1993) as reflecting an actively flowing ground water system with localised recharge. The model requires that recharge input must be balanced by "leakage" into the overlying aquitard. Exploitation of the well field should therefore reverse this flow and produce a leakage input which was postulated in early models. This reverse leakage, induced by exploitation and land use changes in the area might well have longer-term water quality consequences.

Recharge assessment

One of the important hydrological aspects of environmental isotopes lies in their integrative power; the historical information which can be obtained from an "isotope snapshot". Since the first isotope studies some 30 years ago, water level, abstraction and rainfall figures spanning up to several decades are available for several areas of the Kalahari. Recharge assessments on the basis of equal volume analysis of hydrographs and cumulative rainfall departure simulations correspond well with isotope-based recharge rates, the relevant parameter in which was shown to be the total aquifer porosity - an important concept in future recharge assessments (Verhagen et al.1999). Corroboration of the underlying recharge was provided by the GRES project (Beekman et al.1996) which showed that preferential or by-pass unsaturated flow occurs through thick, apparently uniform sand covers.

Conclusions

A number of studies conducted in the Kalahari over the past 30 years have established that isotopes are indispensible in dryland hydrology and can provide important resource management information. They have shown that:

• recharge can occur through thick and vegetated sand covers and can be quantified

• recharge which occurs at rock outcrops and drainages has rather localised influence

• regional ground water movement is at best a second-order phenomenon

• evapotranspiration is an important loss mechanism, even from deep ground water levels

• mineralisation is a function of the ratio of lateral to vertical water flux and which is controlled by aquifer structure