The Degirmendere Aqueduct at Ephesus

The most important result of the preliminary studies on water engineering in Ephesus is the fact that the immediate connection of this settlement`s history with managing its water supply is far more easily to comprehend in this city than in other places. However, in order to be able to study the relation of the settlement`s expansion and the development of springs further away in detail, it will be necessary to initiate a comprehensive research project focussing on water supply as this has hitherto not been taken into consideration very much on the contrary to the Ephesus` city history. The Degirmendere Aqueduct has a number of superlatives to offer. With 43 km it is the longest of five aqueduct lines leading to Ephesus. With its cross-section of ca. 80 x 230 cm, it also used to be able to bring most of the water to town by winding itself through the valleys following the gravity as an un-pressurised line thereby crossing at least 20 bridges and passing through two tunnels. Presumably, it was built no earlier than the middle of the 2nd century BC. It also is the youngest waterline with an own marked-out route. As it was used throughout many centuries - the first section supplied the Caravanserai of Scala Nova in the Ottoman period and also the tourist town of Kusadasi with water until about 20 years ago - many traces of repairs and rebuilding phases are detectable. Nevertheless, a mistake must have been made right at the beginning of the construction when difficulties arose in overcoming a saddle due to an initially false estimate of the gradient of a section, so that the water could not continue to flow easily towards the city. This is why a new stretch - some 8 km long - had to be built on to the unsuccessful channel on the side towards the hill, thereby re-using the abandoned construction partly as a retaining wall for the new channel as well as dismantling the larger part in order to be able to use that as building material for the new line. How and where was the Degirmendere Aqueduct constructed, repaired and rebuilt? Can any borders of building sections be ascertained by taking the gradients of the various sections into consideration? Are there any remains of the former infrastructure in building the aqueduct, such as: access roads, working ways or quarries still left? How much water did the spring yield? How much water actually did arrive in the city, and how much was lost through leaks? Was the Sultaniye Aqueduct an Augustan predecessor known through inscriptions as "Aqua Iulia"? Where does the supply line come from which leads into the main aqueduct at the Mercankuyu Bridge? These and many other questions should be scrutinized and clarified in the course of documenting the building in the framework of an interdisciplinary project, supported by new measuring methods and analysing the sinter- and mortar samples as well as archaeological digs. A comprehensive publication of the results will appear in a volume of "Forschungen in Ephesos". The new insights will form an important basis for the further investigation of the smaller and less laborious aqueducts of Ephesus which should supply further insight into the connection of the gradually extended water supply system and the settlement development mentioned above.

The De?irmendere Aqueduct is the longest and youngest waterline running to Ephesus. With a length of 37.5 km and a transverse section of 70 x 50 cm, it was erected during the Hadrianic period (between 124 and 129 A.D.) to transport 21.000 m3 of water across 24 bridges to town every day. This channel was only in use for 25 to 30 years. An earthquake which can be dated due to the recent discovery of an inscription of the year 159 A.D., caused the neogene clay stratum at the Içmetepe fault line to sink 3 m lower than the marble back, thereby interrupting the inlet. This is why a completely new channel had to be built in Antonine times, which was then able to collect the water from the Keltepe springs too. With about 50.000 m3 water per day, the larger, far more easily walkable vaulted transverse section with a diameter of 75 x 215 cm was able to supply the city with a far larger amount of water. After 120 years of service, this aqueduct was destroyed by a series of earthquakes in the second half of the 4th century and was renewed in its first section only for Scala Nova in Venetian-Genoese and later Osmanic times. It was used for the tourist metropolis of Ku?adas? until 20 years ago. The older cross-section only needed to be enlarged.in the first 18,5 km long part of the aqueduct starting at the source and ending at the fault where the channel lies mostly underground or has been destroyed by the construction work for the growing town of Ku?adas?. Between the disrupture line and the Bahçecikbo?az Bridge where a pressure pipeline made of clay pipes crosses the valley and runs inland far away from the coastline, the 3 m difference of the base of duct was adjusted within a length of only 3,5 km. As a result of this, an extremely gentle decline always was a problem for transporting water so that most of the newly discovered bypasses lie in this section where the water needed to be diverted several times. The section after the Bahçecikbo?az Bridge was built with a gentler decline than its Hadrianic predecessor so that it could reach the city at an 8m higher point in order to supply a larger area with more water. In the 11 km long mid-section, from the fault until the spot where the channel coming from the coast bends inland, both aqueducts are preserved entirely thanks to their position higher up on the slope. Here, all the bridges of the Antonine aqueduct were attached to the already existing Hadrianic bridges so that all the buildings in this area had to be executed twice. Exceptions are the three great bridges where the efforts for rebuilding would have been too elaborate and expensive, so that the new line was built over the older bridge.In the course of the FWF-Project, all the individual buildings such as bridges, tunnels, basins, bypasses, supporting pillars and passages could be recorded after thoroughly grubbing the area. In the course of documenting the monuments, the team discovered eight of the 24 bridges, and all the tunnels, basins and bypasses. The sections of the aqueduct channels were documented as far as they were visible, and a number of further details clarified after digging sondages. Analyses of sinter were able to determine how long the aqueduct was in use, and the reasons for the existence of double channels could be found due to the geological conditions along the fault. The final publication is in preparation. At the same time, a GIS-project will be available online, where all the information will be retrievable in order to follow the aqueduct along its entire length in further detail from source to town.