Once the different phases of the main retaining wall were understood (cf. above chap. II), a partial restoration of the wall became realisable. The results of our 2001 season showed that, as a matter of fact, the collapse of that wall is not due to water pressure during seasonal rainfalls, but rather to the misled ancient extension of the wall.
Fig. 38: Lower terrace, detail of main retaining wall from NW before restoration process (photo: S. G. Schmid)Fig. 39: Lower terrace, detail of main retaining wall from NW during restoration process (photo: S. G. Schmid)
Therefore, the parts that were already repaired in antiquity and broke out since were taken away (figs. 38. 39), the remaining stones were cleaned and prepared for restoration, and the wall was again closed and repaired until the level it showed before the restoration (Fig. 40).
Fig. 40: Lower terrace, detail of main retaining wall from N after restoration (photo: S. G. Schmid)
For the rebuilding an elastic mortar was used that also contains small pieces of broken pottery – as does Nabataean mortar – in order to improve its hydraulic and elastic qualities. Further, in addition to the water channel established during the later phases of the main retaining wall (cf. above chap. II and fig. 14) a second opening in the wall was constructed in order to allow seasonal flash floods to pass through that area without destroying the wall.
Fig. 41: Lower terrace, trench 1. Main entrance with floor slabs prepared for restoration (photo: S. G. Schmid)
The main entrance hall initially was covered with huge limestone slabs measuring 40 cm x 80 cm (cf. above chap. II). As some of them were found in the destruction debris of the area, they were put back in their initial position, covered with plastic and sand in order to be ready for restoration, once the area will be completely excavated (Fig. 41).
