Forum Archaeologiae - Zeitschrift für klassische Archäologie 5 / XII / 1997


Goals of Project:
To examine the entire water system of an ancient city, including supply, distribution, use, and drainage. Until now, individual water system elements have been studied as archaeological objects, but infrequently subjected to analysis by such disciplines as fluids engineering and urban history. This work is part of two larger long-term studies, the first concerning the relationship between physical setting and urban growth in the Greco-Roman world, and the second a comparative study of Old and New World water management techniques in the pre-modern period.

Methods and Personnel:
Principal investigators are Dora P. Crouch, Ph.D., an urban historian specializing in Greco-Roman cities, who wrote Water Management in Ancient Greek Cities (Oxford University Press, 1993), and Charles Ortloff, a fluids engineer with twenty years experience assisting archaeologists in Peru and Bolivia to study canal networks and raised bed agriculture from before the Spanish conquest of the area. Ortloff has published extensively in the Journal of Field Archaeology and most recently as a contributor to Tiwanaku and Its Hinterland: Archaeology and Paleoecology of an Andean Civilization, (A.L. Kolata, ed., Smithsonian Institute Press, 1997).
Our methods in 1997 included intensive survey of all known water system elements within the site of Ephesus, which were recorded on a topographical map of the site. We also took photographs and made sketches, as well as consulting with archaeologists who have worked at the site for years. During the year between June 1997 and May 1998, Crouch will do library research on what has been published about water at the site, and will visit Vienna to consult the archives of photos and plans at the Austrian Archaeological Institute. Ortloff will utilize the super computers at his disposal to make a topographical map of the site with all known water system elements. This will enable him to begin analysis of the amount and speed of flow, with hypothetical perimeters for waste water drainage.

Abb 1.: Ephesus - Pollio aqueduct. Drawing G. Niemann (1903)
(aus: FiE 3 (1923) Abb. 264)

Our investigation has been facilitated by the collaboration of Prof. Dr. Ing. Unal Ozis, of the Civil Engineering Dept., Faculty of Engineering and Architecture, Dokuz Eylul University, Izmir, who with his students has for 25 years been studying the long-distance water lines of Ephesus. They have supplied to us present-day flow figures for the four sets of springs that supplied the city, and taken us to see the extant remains of these lines. We are also most grateful to Dr. Stefan Karwiese for his encouragement of this investigation, and others of the excavation house in Selçuk for the information they have so generously shared with us.

Findings of First Year:
Although it had been known that four aqueducts supplied water to Ephesus, the intra-city distribution system and drainage network of this water - the subject of the present investigation - is presently imperfectly known. Three sets of urban reservoirs and fountains houses have been identified in the 1997 field work, as well as numerous segmented ceramic pipelines and large open channels. We have tracked the networks of piping, and located them and the reservoirs on a topographical map of the city. Particularly, we have studied an elaborate hydraulic junction consisting of a 2-4th century CE fountain house (the water tower), with its related piping, adjacent to the upper agora. A highly efficient system is evident: Water pressure to individual water lines appears to be adjustable by varying the hydraulic head in (4) feed pools supplied from a common upper reservoir. Pipe networks appear to spread out from this fountain house, supplying water at different flow rates to baths, fountains, and residences according to the requirements of each. Individual water lines were designed to match water requirements of major structures en route, with residual supplies to second-level users at the terminus of each line.

Redundancy seems to be an important basic feature, in that certain areas might be supplied by overlapping piping networks, to facilitate cleaning and allow for variation of supply from the different sources. Another basic feature appears to be low flow rates sufficient to maintain needed levels, with excess supply carried on for downstream use or shunted to drains.
Abb. 2: Aqueduct, Augustean period (aus: W. Oberleitner, Ephesos 1978)

The original supply of water from natural springs and from wells was later supplemented by piped water, e,g., for the baths in the upper agora, and for the Hang Houses.

Chronological sorting of water system elements according to usage phases is a necessary next step, to group contemporary features; however, the geographical constraints of the city's location on mountain slopes and in the narrow valleys between the mountains dictated similar locations of water system elements in every period. Hence for Ephesus we expect to find fewer chronological differences in water system elements than at other cities.

Use of software for piping system flow analysis is anticipated in further study, leading to refinement of our understanding of Roman hydraulic engineering practice. The map we have generated in the 1997 field survey is a necessary first step for both the computer work and the projected development of the study.

Future Work:
With the first phase of computer analysis completed, and with the information derived from library research, we will be ready in the spring of 1998 to return to the field to excavate the water tower at the edge of the upper agora (No. 17). As the pivot of the water system, this structure should provide useful information about Roman hydraulic engineering at the site and during the imperial period.

©D. P. Crouch and Ch. Ortloff