In the archaeological season of 2008 a new study aiming to re-appraise the Butrint aqueduct’s source, route, urban distribution and date, was undertaken.
The aqueduct arcade is traceable as alignments of upstanding brick piers and pier bases across the Vrina Plain from the Vivari Channel for nearly 4 km towards the hill of Xarra village. Two springs are known at Xarra and from their elevation it is clear that both – perhaps in combination – could have provided the water source for the aqueduct. The main arcade line has a derivation to its west, to supply the suburban settlement where a cistern acted as a distribution node. Here, water supply is confirmed by the remains of a terracotta pipeline and the aqueduct must have supplied the Roman townhouse’s large ornamental fountain and baths.
The main aqueduct arcade terminated close to the south bank of the Vivari Channel in a large, waterproofed concrete header tank for a pipeline system. Most probably, the water was brought across the channel using the inverted siphon method in a pipeline laid either along the channel bed or along one side of the road bridge.
Possible foundations of a receiving tank for the pipeline and further aqueduct piers exist close to the north end of the road bridge by the medieval Water Gate.
In Butrint the aqueduct originally flowed into the large nymphaeum by the Great Basilica, which served as both public fountain and distribution centre for the urban water network via ground-level pressure pipelines. A sequence of piers to the north of the nymphaeum provides evidence of the later extension of the water supply system, undoubtedly as a response to the expansion of the city.
The new piers were constructed out towards the sanctuary area, passing close by the Tower Gate and the so-called Tower of Inscriptions and could now supply parts of the city situated at higher elevations including most, if not all, of the city’s six known bath-houses.
Ceramics from the excavation of one of the aqueduct piers on the Vrina Plain suggest an approximate construction date after the late 1st century BC. This is consistent with the commemoration of the aqueduct on Augustan and Neronian coin issues from the mint of Butrint.
These show a triple-arched structure carrying either a balustrade or smaller arches, and include one example with a possible depiction of water cascading from the structure, suggesting that it represents the aqueduct.
When and why the aqueduct went out of use is currently uncertain, but given the dramatic collapse of many of its components, seismic activity is a possibility.
A fallen, intact aqueduct pier exposed on the Vrina Plain sealed deposits, providing an approximate late 4th- to early 5th-century date for its collapse.
With the arcade interrupted, the aqueduct would have ceased to supply water to the fountains and bath-houses of Butrint.