Continuation Project in Kurunegala

After 1985, a subsequent project of GTZ was carried out in Sri Lanka under new management, focusing on the dry areas of Kurunegala District. Besides several measures related to the improvement of the infra-structure, the development of rural water supply constituted one of the main tasks. As in the previous project, and apart from a few wells intended for small central water works, it consisted essentially in the construction of individual supplies, that means wells with a yield of 5 l/minute and equipped with hand pumps.

The hydrogeological situation is, in all aspects, comparable with the one in Vavuniya District. The water detection depends on finding fractures in hard rock which are not very pronounced in mostly gneiss and granite formations. A former study of the water supply possibilities only investigated the construction of long distribution pipelines to exploit shallow lying water [10] — by now a no longer acceptable procedure.

Making use of the experiences gathered in phase 1 of the project, dowsers were also appointed in the still ongoing subsequent project. Meanwhile, the following procedure has been found to work well: initially, the target groups (communities) propose well sites to their liking. Then, a hydrogeological team investigates the target area and determines, whenever possible, promising well sites; geo-electrical techniques represents the typical measuring aid. The study of aerial photographs did not play a role due to insufficient resolution and because the aim was to locate well sites of moderate yields in the middle of villages. Finally, two dowsers were appointed to pinpoint the actual drilling sites.

In following this procedure about 1200 boreholes were drilled by the end of 1990. The success rate was 80%, whereby every well with a yield below 5 l/minute was considered as "dry". Although this rate is somewhat lower than in the preceding GTZ project, it is altogether a good result. Both locally involved hydrogeologists, as well as experts who visited the area occasionally, certified that it was extremely difficult and sometimes impossible to find very narrow fractures—like those located in the previous project—by either "intelligent guessing" by means of theoretical trivial methods or pinpointing by means of geo-electrical or other classic measuring techniques.

It must be mentioned that Schröter was not one of the two appointed dowsers; the first was a German and member of the project drilling team, the second was a local person trained for that task by the exploration team since the beginning of the project. Although both dowsers are unable to provide such precise indications as Schröter does, their appointment was very positive. That was also reflected in the following fact: occasionally, when the hydrogeological procedure was out of order, the drilling points, driven by necessity, had to be selected exclusively by the dowsing technique; amazingly, the high success rate was not noticeably reduced. GTZ plans therefore to continue the well-controlled employment of these apparently appropriate dowsers within the described team work of the project.

Final Evaluation of the Sri Lanka Project

The information from previous projects encompasses quite a large framework. In the past, a program of this magnitude had never been performed, in which dowsing played an apparently decisive role, and which led to the accumulation of so many verifiable facts and data. In the following section, these results will be discussed further with regard to the possible significance of the dowsing technique. The quantitative success of the GTZ program in which Schröter determined all the definitively selected drilling points, is incontestable. Accordingly, the question must be asked if this success would have been also possible without the aid of a dowser. This alternative, however, seems to be improbable on account of the many arguments already described before. Above all, the only two conceivable counter arguments, which could be put forward (and will be treated in more detail in part 3.1) are probably unfounded and cannot be advanced:

• Are there so many fractures in the underground of the project area that there is a 90% chance to find ground water anywhere with a mean yield of about 30 l/minute?

• Can appropriate drilling points be consciously or unconsciously recognized through classic and passive techniques, namely by subtle observation of topography, morphology, flora and fauna?

The first argument can be refuted on both the basis of general as well as practical experiences gathered at the sites and on hydrogeological models which were derived from these data. Small fractures and low permeability have been repeatedly proven to prevail. We must, nevertheless, concede a limited relevance of the second argument; however, in crystalline underground with relatively small and locally very concentrated aquifers, the indications one may possibly gather at the surface are generally not sufficient to enable the location of a well site with the necessary accuracy. The three control tests described above, as well as the comparison with the results from other projects credibly support the rejection of both of these.

It must be suspected that without the incorporation of the dowsing technique the GTZ project would have resulted in a success rate of no more than 30 - 50%. The necessary number of wells would have undoubtedly been reached by means of modern prospecting methods, but cost and time inputs would have been considerably higher than was the case for the actual, effectively carried out GTZ project.

Due to the increasing and credible evidence, one may come to the conclusion that the appointment of appropriate dowsers does notably increase the success rate in a well-planned hydrogeological project. Nevertheless, completeness of argumentation requires the admission that as of now the results described and the derived conclusions cannot represent a final scientific proof for the relevance of the dowsing technique. Such a proof can only be established when, independently of the evidence gathered by individual groups, sufficient repetition of the experiments is achieved due to the involvement of independent researchers: only then, and provided that further results confirm the initial observations, a lasting and reliable consensus between competent specialists might be attained.

Comparable Experience from a Pilot Project (Burkina Faso)

The difficulty of water detection in arid areas with a crystalline rock basement can be understood in the light of a more recent study carried out in Burkina Faso, where a maximum of hydrogeological and geophysical investigation techniques has been exemplarily implemented [11]. This research program has been set up in order to investigate whether one could improve the highly unsuccessful drilling results produced by means of the ground water investigation techniques used in the past. The area to be studied was characterized by essentially similar conditions as in Sri Lanka; the yearly precipitation varied between 600 and 1000 mm/yr.; dry seasons are long, and the water reserves are mainly stored in fracture zones in the crystalline bedrock.

The study made use especially of highly sensitive satellite images (MSS and TM) taken at different seasonal periods, and electromagnetic measuring procedures (VLF with movable transmitter and horizontal loop procedure with a penetration depth of up to 40 m), including measurements of the magnetic field and electrical vertical soundings. From the beginning, the combination of the numerous hydrogeological and geophysical procedures was considered to be necessary because otherwise drilling points could not have been located with sufficient reliability. The results were no real surprise to experts and, although obtained in a particular area, are considered as typical also for a variety of other arid zones.

Obviously, certain details of structures in the underground may be made visible by geophysical procedures, but not with aerial photographs—and vice versa. The link between EM anomalies and cracks in the rock is not always evident. Investigated fractures and weathered zones may present resistivity anomalies in cases where they are significantly extended, but statements on stored water reserves or waterflow are not possible; deep lying extensions cannot generally be detected through routine measurements, but, if ever possible, only with additional costly supplementary measurements. Fracture zones with narrow weathered spaces do not appear clearly or cannot be located by EM measurements. Besides the resolution limitation of satellite and aerial photographs, non-fracture-tectonic origins of EM anomalies, lithologic and tectonic anisotropies of the rocks and ambivalences of the interpretation of the tectonic structure are methodic weakness of the conventional procedures.

The combination of the numerous applied procedures allowed the location of intersections between tension zones and shear zones which are considered to be particularly promising; here, drillings should be successful because they encounter permeable cracks. The success of that extremely expensive project could be demonstrated in practice. Out of the total of 16 (test-) drillings with depths to 80 m only 2 were dry, whereas the yield of 14 wells was between 10 and 100 l/minute. Nevertheless, it is important to remark that the applied prospecting input of tremendous proportion was only possible within a research project and cannot be considered as a standard for routine investigations. Unfortunately, the possibility of dowsing investigations had not been considered. Excellent tests of this unconventional procedure would have been possible and particularly useful because of the availability of unusually detailed data of the underground conditions.

These findings support the previously mentioned conclusion that the few technical prospecting methods used in Sri Lanka could not warrant a high success rate; the interpretation of aerial photographs and geoelectrical techniques alone has to be considered as being insufficient, as its significance with respect to the ground water flow in the underground is largely undetermined. As the success rate of the first project phase in Sri Lanka was still higher than the one achieved in Burkina Faso, there is a well-grounded argument in favor of the combined technique which, provided that certain conditions are met, makes use of both classical geophysical techniques and dowsing procedures.