The authors thank Mr Puri for his comments on their paper. His point about the effect of anisotropy is a valid one. From our knowledge of the study area it is reasonable to assume that within the typical dry valley situation there exists (in the horizontal plane) both anisotropy (probably with the major axis coincident with the valley axis and the minor axis at right angles to it) and heterogeneity, as indicated in Fig. 2. As Mr Puri will be aware, the radial flow model can only deal with heterogeneity (fig. 7) it cannot, by definition, deal with anisotropy. Thus the basic question to be answered here is, if by some means or other the effects of anisotropy could also be taken into account by the model would the results at the pumped well be significantly different from those already obtained? I think the answer to this lies in knowing two things:
(i) the ratio between the maximum and minimum transmissivity tensors (←pha ) in the field; and
(ii) the critical values of ←pha which would produce a significant deviation from the isotropic case (←pha = 1) when incorporated into our vertically and laterally heterogeneous aquifer.
The former would involve some quite elaborate field tests over a statistically meaningful number of pumped wells. The latter could be investigated using regional modelling techniques, although modelling non-linear changes in aquifer parameters by such methods may prove difficult. The work carried out by Vandenberg (1977) may be a useful starting point for this
- © The Geological Society, London 1978