Touring Israel's ancient water systems

85

CURA AQUARUM IN ISRAEL II, Schriften der DWhG, Band 21, Siegburg 2014, ISBN 978-3-86948-333-7

Touring Israel's ancient water systems Tsvika Tsuk There are hundreds of ancient water systems throughout Israel, dating from the 7th millennium BC to the 20th cen-tury AD. They range from lonely cis-terns to huge underground reservoirs; from simple wells to long aqueducts beginning at springs. The semi-arid cli-mate of the land, which means a lack of water, has compelled its inhabitants to find solutions to bring water to their communities. Thus some of the enormous water sys-tems are local inventions of this country, such as underground hewn tunnels to the water table like those at Tel Gezer, Me-giddo, Hazor, etc. Two years ago I published my book "Water at the End of the Tunnel" (Tsuk 2011), which recommended 40 tours to ancient water systems in Israel. The appendix to that book lists 80 more places and there are even more (Fig. 1). During our conference we visited 11 sites in the central and northwestern parts of Israel as well as a site in the Judean Desert. The sites are: Masada, Jerusalem, Tel Gezer, Ramla, the Eshkol Site (part of Israel’s National Water Carrier), Sepphoris, Acre, Megiddo, Mei Kedem, Nahal HaTaninim Dam and Caesarea. They range in date from the Middle Bronze Age to modern times, including the Iron Age, Roman, Byzantine and Early and Late Islamic periods. The sites are very diverse in terms of chronology, hydrology, technology and regions. This article briefly describes the sites as background to the tours. Fig. 1. Map of the main ancient water systems in Israel (Avigdor Orgad).

86 Tsvika Tsuk, Touring Israel's ancient water systems


Fig. 2: Masada National Park and the water systems (Drawing: Shalom Kweller; adaptation: Avigdor Orgad). Masada (fig. 2 and 3) is a rocky natural cliff in the Judean Desert near the Dead Sea, topped by

a plateau about 9 hectares in size. In the 2nd century BC, the Hasmo-neans settled at the site. In the 1st cen-tury BC, Herod built an impressive fortress palace there, where Jewish rebels entrenched themselves against the Romans in the 1st century AD until Masada fell in 73 AD. Masada’s main water systems date from the time of Herod and in-clude 17 huge cis-terns, 9 medium-size and small cis-terns and 2 aque-ducts for flood water that led to 12

(out of the 17) gigantic cisterns on the western side of the mountain; 5 bathhouses; and 21 ritual baths (Netzer 2002, 353-365). Some 48,000 cubic meters of water could be con-tained in these cisterns and other installations. To bring up the water from the cisterns in the mountainside a system of paths was created

plied by donkeys bearing the water up to the plateau. During the siege, the rebels had enough water to hold out. In fact, the Romans were the ones with a major water prob-lem – they had to bring water from the springs of Nahal Ze’elim and En Gedi. Masada is a national park un-der the aegis of the Israel Na-ture and Parks Authority and was declared a UNESCO World Heritage Site in 2001. Fig. 3: Masada from Mt. Elazar (photo: Haim Castelnouvo).

Tsvika Tsuk, Touring Israel's ancient water systems 87


Fig. 4: Jerusalem water systems – City of David National Park (Drawing: Shalom Kweller; adaptation: Avigdor Orgad). Jerusalem is Israel’s capital, and it was built as far back as biblical times on a long spur west of the Kidron Valley. The ancient city eventually expanded westward to the hills be-yond the Tyropoeon Valley. The Old City of today contains remains from the time of the First and the Second Temple (the 1st millenni-um BC until the 1st century AD). The earliest ancient water systems in the City of David date from the Middle Bronze Age II (about 1700 BC) and the Israelite period II (Iron Age II) (the 8th century BC) and were created around the Gihon Spring. This spring flows at a rate of 600,000 cubic meters of water per year, and was the main source of water for Jerusalem until the beginning of the 20th century. In re-cent years these water systems (fig. 4) have been uncovered and made accessible to the public (Reich 2011). Our tour took us first to Warren’s Shaft (fig. 5), which was hewn during the Middle Bronze Age and led to a pool surrounded by towers that acted as huge fortifications around the water source. Thus the ancient inhabitants could draw water during a siege. Also part of

this system is the natural portion of Warren’s Shaft, which was uncovered by accident during quarrying in antiquity. Here some of our trav-elers chose the “wet route” through the Siloam Tunnel (533 m long) and others the “dry route” along Channel II. The hewn Siloam Tunnel is an engineering wonder whose construction scholars have still not quite figured out. Con-troversies abound as to the date of the tunnel (most opinions attribute it to the time of King Hezekiah at the end of the 8th century BC), how it was hewn, whether it is partly natural or completely artificial, how long it would have taken to quarry and other issues. Even without consensus on most issues, going through the Siloam Tunnel (where the water is about 0.5 m deep) is very exciting and takes about 45 minutes. The purpose of the tunnel was appar-ently to bring the spring water into the city to an area where it would be easy for the inhabit-ants of the western city to obtain their water. The dry route, Channel II, dates from the Mid-dle Bronze Age II and is about 150 m long. This channel was partly hewn from the surface and partly as a tunnel. In the surface portion,



gigantic wedge-shaped stones were brought in to create roofing. The Siloam Tunnel ends at the Siloam Pool. This massive pool, which has been uncovered in recent years, dates to the Second Temple periods (1st century BC– 1st century AD). However, evidence exists that it dates back as far as the Israelite period II (8th century BC?). The pool measures 50 x 50 m and it is stepped on all sides. The steps allowed it to be utilized it for various purposes: as a large ritual bath, to fill water vessels at numer-ous points, crafts involving water, irrigation, etc. On the western side of the pool an impos-ing street was found dating to the Second Temple period, under which was a large and impressive drainage channel. The channel was uncovered along about 800 m and after clean-ing and preparation for visitors, people can ascend it. The channel is about 2 m high and about 0.8 m wide on average and can be trav-ersed easily up to the southwestern corner of the Temple Mount under Robinson’s Arch. Toward the end of the route, the channel cuts through two cisterns and passes above a large water reservoir; all three of these elements date from the Israelite period II (Iron Age II, 10th–6th centuries BC).

Fig. 5: Warren’s Shaft (photo: Tsvika Tsuk).

Tel Gezer is a biblical city mentioned 12 times in Scripture in various contexts. The archaeo-logical excavations at the site uncovered 26 strata of settlement, one atop the other. The main finds are from the 2nd and 1st millennia BC. Tel Gezer’s water system was excavated 107 years ago by the Irish archaeologist R.A.S. Macalister (Macalister 1912, 256-265). Fortu-nately, he did not cover it up again as he did the rest of his excavations as required by Ot-toman law. It was also fortunate that he was unable to excavate the water basin, but instead covered it up with a layer of stones. The water system is divided into four parts: the unroofed entrance corridor, sloping tunnel (fig. 6), reservoir and cavern.

Fig. 6. The sloping tunnel at Tel Gezer (photo: Art Beaulieu) The maximum measurements of the sloping tunnel are: length 50 m; height 7.5 m; width 4 m. The hewn volume of the water system is ap-proximately 1,100 cubic meters. New research was launched at the site in 2010 by a joint expedition of the Israel Nature and Parks Authori-ty and the New Orleans Baptist The-ological Seminary. For three seasons the expedition cleaned the stone and earthen debris that had penetrated

the system since Macalister’s excavation.



Fig. 7: Tel Gezer National Park and the Ramla aqueduct (Avigdor Orgad). In the fourth season (2013) Macalister’s layer of stones was excavated, revealing archaeolog-ical finds beneath it dating to the Late Bronze Age II (around 1300 BC), apparently the last period of the system’s use. As for the begin-ning of its use, the expedition believes that it should be associated with Gezer’s other mon-umental construction projects, such as the gate, the city wall, the huge tower and the gigantic steles at the high place. All of these are dated to the Middle Bronze Age IIb (around 1600 BC). This is the largest Canaanite water system known anywhere in the world to date, based on the volume of material hewn from it. Its magni-

tude attests to a prosperous royal city whose rulers wisely built a water system within it. Ramla was the capital of the region known as Jund Filastin during the Early Islamic period under the Umayyads. It was founded in the early 8th century AD. During this time, many complexes were built there, among them the Dyers House, the White Mosque and water installations – the aqueduct and pools (fig. 7). The aqueduct channeled water to the city from springs and wells at the western foot of Tel Gezer.



During our tour we visited two sites that be-long to Ramla’s water system: 1. The aqueduct near Road 6 2. The Pool of Arches in the northern part of Ramla. 1. The aqueduct near Road 6 (cf. fig. 7) In excavations during construction of Road 6 in 2001 A. Gorzalczany uncovered a portion of an aqueduct 150 m long (see the article in this book, pp. XXX-XXX).

Fig. 8: The aqueduct near Road 6 (photo: Haim Cas-telnouvo). The aqueduct (fig. 8), which was built out of large, dressed limestone blocks, was 0.50 m wide and 1.20 m deep and was roofed by slabs that were 0.80 m long. It was coated with plas-ter inside and out. The inner plaster, which was reddish in color, consisted of one layer 2 cm thick containing pulverized potsherds and laid over gray mortar. Over the plaster was a thin

0.6 cm accumulation of sinter. The sinter on the sides of the channel only goes as high as 0.35 m, which shows the usual water level. In a number of places a second layer of mortar, gray-white in color, was found. The lower lay-er is dated to the Umayyad period and the up-per layer marks a repair during the Abbasid period. Manholes were discovered along the aqueduct that allowed entry for cleaning. The Israel Antiquities Authority and the Road 6 administration recognized the importance of this find and reconstructed it on either side of the road using stones taken from the portion where the road now passes. The reconstructed portion is about 50 m long. Most of it is origi-nal and only the upper part, which includes the roofing stones and the manhole, are recon-structed. The reconstruction level is marked with a continuous line of potsherds under the top course of the aqueduct. 2. The Pool of Arches in the northern part of Ramla (cf. No 2 in fig. 7)

Fig. 9: The Pool of Arches (photo: Haim Castelnouvo). The reservoir known as the Pool of Arches (fig. 9) was built toward the end of the 8th century AD and was added to the mosque reservoirs that had been built about 70 years earlier. The pool, which is also known as Bir el-Anziye and the Pool of St. Helen, is one of the hallmarks of Muslim architecture in the Land of Israel. It was built very meticulously using superior materials and well-cut stones. Its quality of

Tsvika Tsuk, Touring Israel's ancient water systems


91

construction, like that of the mosque pools, was proven when powerful earthquakes destroyed the city’s houses a number of times but left these reservoirs, which were constructed deep below ground, practically undamaged. In 1867 an inscription was found carved into the wall of the reservoir near the entrance and the steps. The inscription reads: “In the name of Allah…[which] the agent of the Emir of the faithful, may he live a long life, commanded to build, in the month of Hajj, in the year 172 [May 789].” The inscription reveals that the pool was built in the Abbasid period, during the reign of the last caliph, Harun al-Rashid of “Arabian Nights” fame. The pool, which has 20 pillars and 38 slightly pointed arches, is trapezoid in shape and measures 24 x 26 m. Its volume is approxi-mately 5,700 cubic meters. The roof originally had 20 openings for drawing water but only 16 can still be seen. The water may have reached the pool either from the aqueduct to Ramla, which was ex-tended during the Abbasid period also to fill this pool, or from a spring over which the pool was built. This is the only intact public build-ing from the Abbasid period found in Israel. Eshkol Site – Heart of the Israel National Water Carrier – Visit to Meko-rot Visitor Center (written by Hillel Glassman) Mekorot, Israel’s national water company, began as the “impossi-ble dream” of Israel’s pioneering leaders a decade before the estab-lishment of the State of Israel. Today it is one of the world’s most advanced water companies, with leadership in the areas of water-resource management, desalinization, wastewater treat-ment and effluent reuse, rain enhancement, water quality, wa-ter security and water-project engineering. Mekorot’s water supply system integrates most of the regional water plants, the National Water Carrier, and the Yarkon-Negev plant. The system combines

water from various sources: the Sea of Galilee, mountain and coastal aquifers, as well as drill-ings, desalinated sea water and brackish water. The National Water Carrier delivers fresh water from the north, with its fairly abundant rainfall and water sources, to the dry south. The National Water Carrier is a technical chal-lenge that traverses a wide range of terrains and comprises a network of aqueducts, tunnels, reservoirs and pumping stations to deliver about 400 million cubic meters of water a year. Water Quality and the Central Filtration Plant After implementing a sophisticated design and construction plan, in 2007, Mekorot inaugurat-ed the Central Filtration Plant at the Eshkol Site (fig. 10). Built at a cost of more than $100 million, and with an annual filtering capacity of more than 500 million cubic meters, the Central Filtration Plant is the largest plant of its kind in Israel, and one of the largest in the world. A global leader in terms of efficiency, water quality targets and the quantity of water it handles, it ensures that Israel’s residents will obtain even higher quality water via the Na-tional Water Carrier. Its water quality targets match the accepted targets in Western coun-tries.

Fig. 10: Eshkol Site (photo: Hillel Glassman).



In terms of water filtration, research and expe-rience around the world indicate that the most effective way to remove parasite-caused dis-eases and turbidity particles is advanced filtra-tion technology. In all advanced countries mandatory surface water filtration rules have been introduced for drinking water supply sys-tems to ensure the removal of the pathogenic parasites. The turbidities of filtered surface water were limited to below 1.0 Nephelometric Turbidity Unit (NTU; mostly <0.5 NTU). For the same reason, the increasingly stringent drinking-water standards introduced in Israel limited the turbidity to 1.0 NTU (desired 0.5

NTU). To meet this requirement it was decided to build the Central Filtration Plant at the Eshkol Site in the Beit Netofa Valley. The plant was built by Mekorot together with ex-pert Israeli and international companies. In order to ensure a reliable and safe filtering process, while keeping costs as low as possible, the Central Filtration Plant includes sophisti-cated operations combining the most advanced technologies with many engineering disci-plines. Most of the plant’s control system oper-ates automatically, requiring a smaller staff compared with similarly sized facilities.

Fig. 11: The aqueducts to Sepphoris (Avigdor Orgad).



Sepphoris (fig. 11) was an important ancient city in the Lower Galilee dating from the Hel-lenistic, Roman and Byzantine periods. It is the place where the Mishnah was codified by Rab-bi Yehuda Hanasi in the early 3rd century AD. Many impressive mosaic floors were uncov-ered here during archaeological excavations as well as a system to bring water from afar. The area’s largest spring is located near the city (only 2.5 km away) but it is lower than the city. The inhabitants therefore had to seek water sources in the mountains to the east, and to that end they created two aqueducts from the Nazareth Mountains to Sepphoris (Tsuk 1999, 161-175; 2002, 278-294). The older of the two, dated to the 1st century BC and which came from the area of Mashhad, has left meager, rock-hewn remains. The main aqueduct, created in the 1st century AD, came from e-Reine. It is 0.30 m wide and up to 0.45 m deep, with a flow volume of 36 cubic meters per hour. Only the e-Reine aqueduct can be seen near Sepphoris, where it diverges: The southern branch led to the giant reservoir and the northern branch led to the Mashhad pool and the Arches reservoirs. The spring’s rela-tively small quantity of water manifests itself both in the small dimensions of the aqueduct and the large size of reservoir whose purpose was to collect every drop and send it to the city in a controlled manner. The location of the reservoir, about 1.5 km from the city, was cho-sen because of a geological fault at this place, juxtaposing soft chalk rock with hard lime-stone, overlaid with hard nari rock that formed the reservoir’s ceiling. Thus only the soft chalk needed to be hewn, up to the seam between it and the hard rock. The reservoir is 250 m long, 4 m wide, 12 m high and its volume is approx-imately 4,300 cubic meters. It was excavated from 1993 to 1994 by this writer for the Tel Aviv University Institute of Archaeology with funding from the Israel Nature and Parks Au-thority and the Israel Government Tourist Cor-poration. The excavation included conservation and preparation for visitors, assisted by the Jewish National Fund, leading to the opening of the site to the public in 1995. The aqueduct channeled the water to a sedi-mentation basin at the entrance to the reservoir.

After heavy debris sank into the basin, the wa-ter was channeled through an opening about 1 m above the bottom of the large reservoir. The reservoir’s walls were coated with two layers of plaster: The first coat was of a light pink color and was dated to the 2nd century AD, while the second coat was dark pink and was dated by the potsherds it contained to the 4th century AD. The latter plaster was apparently applied during renovations made after 363 AD, the year a powerful earthquake severely dam-aged the reservoir. The reservoir apparently continued in use until the 7th century AD. The reservoir was hewn by means of nine openings cut into the hard nari rock. After the openings were cut, hewing continued from the bottom of each opening under the hard rock ceiling and in opposite directions. Two of these openings – one in the center and one at the end of the reservoir – have steps by which laborers could reach the reservoir floor for cleaning and maintenance. During excavation of the reservoir archaeolo-gists searched for its outlet in the city. Quite a large quantity of stones and earth were re-moved from the western end, revealing a stair-case beneath which was a small opening in the floor level. This opening was the beginning of a narrow tunnel some 55 m long that had been sealed for about 1,300 years. Protrusions had been left along its walls on which oil lamps would have been placed to illuminate the dark passage. The tunnel tapered to a point where a wall was built. In the middle of the wall was a round, narrow opening – that is the end of the reservoir. A lead pipe 10.5 cm in diameter was inserted in this wall during construction. From the observation point at the end of the reservoir visitors can see the beginning of the tunnel that led to the lead pipe. At the other end of the pipe a large valve, which has now disap-peared, regulated the flow of water to the city. So far only one other such valve has been found anywhere else in the Middle East – at the Nabatean-Roman site of Humeimeh in south-ern Jordan. Calculations of flow rate and water pressure in the reservoir show that such a valve was essential to stop the continuous flow of the water that would otherwise have emptied it.



Further along the pipe a tunnel was hewn (fig. 12). The work began with the cutting of six shafts in the nari rock, and when a suitable level was reached the laborers began to quarry horizontally in both directions. Ac-cording to the marks on the rock, four points at which the laborers met were found. The Israel Nature and Parks Authori-ty prepared the tunnel for opening to the public in 2012. Visitors can de-scend from shaft no. 4 and walk through the tunnel to shaft no. 6 (92 m out of a total of 235 m).

Fig. 12: The six-shaft tunnel (photo: Tsvika Tsuk). Acre (fig. 13) is a city with 4,000 years of his-tory, at some points during which it was one of the most important cities in the Land of Israel. Twice in its history aqueducts were built to the city from the Kabri Springs, about 13 km away. The first time was in the Hellenistic period, when the city was called Ptolemais, and the second time was in the Ottoman period (Frankel 2002, 82-87). The Hellenistic aque-duct represents the beginning of aqueducts in this country, the technology for which was

brought from the West, from the Hellenistic world. The second aqueduct operated from 1815 to 1949; it marks the end of the use of aqueducts in this country.

Fig. 13: Map of the aqueducts to Acre (Avigdor Orgad). The second aqueduct, which was built by Jazzar Pasha, was destroyed during Napoleon’s siege on the city in the 18th century. Jazzar’s successor, Suleiman Pasha, built a new aque-duct along a route closer to the sea. It consists of a built channel (0.50 x 0.75 m) covered with stone slabs, raised on a high, vaulted bridge (13 m high, external width 3 m). The arches are highest where the bridge crosses three streams – Ga’aton, Bet Ha’Emeq and Yasaf. The aque-duct, which was built over a single year (1814–1815), continued to channel water until 1949. It begins at En Hashayara, one of the four Kabri Springs, which has a flow volume of 1.9 mil-



95

lion cubic meters of water per year and is 71 m above sea level. This large quantity of water easily supplied all of Acre’s needs. The villages along the aqueduct also received their drinking and irrigation water from it. Twice a year the inhabitants of the village of Mazra‘a (which belonged to Suleiman Pasha) would clean it over the period of a few days, and in exchange they received a franchise to use the water all year round. About 200,000 cubic meters of water were needed every year for irrigation. About 10,000 people were living in Acre at that time, con-suming some 20 cubic meters of water per person per year, for a total of about 200,000 cubic meters of water per year. Thus, the aque-duct supplied about 400,000 cubic meters of water a year, which is one fifth of the spring’s flow volume. We visited the following points on the tour along the Ottoman aqueduct (see the map fig. 13). No 3 in fig. 13: Mazra‘a At the point where the main road of the village of Mazra‘a crosses Nahal Bet HaEmeq the aqueduct is carried by arches some 13 m high. Some of the arches have collapsed, leaving only their piers. On the eastern side of the bridge, the Mandate government laid an iron pipe to replace a portion that had been de-stroyed in a flood in 1946. The pipe, which served as an inverted siphon, constituted a later repair and a change in technology. No 5 in fig. 13: Kibbutz Lohamei HaGeta’ot The Ghetto Fighters Museum has an excellent view of the aqueduct where it crosses Nahal Yasaf on an impressive, 13 m-high, vaulted bridge. South of the streambed, three phases can be discerned in the aqueduct during the time it was in operation: In the first phase, a high, open aqueduct was built; after it col-lapsed, an inverted siphon was built on the lower arches; in the third phase, the inverted siphon was blocked and was replaced once again by the original open, high aqueduct (fig. 14). From the observation point visitors can go down to the aqueduct and see the water channel at the point where the arches begin. The plaster

that coated the channel is visible as is the sinter that accumulated on the sides of the channel due to the powerful flow of the water and the overabundance of dissolved lime it carried. No 7 in fig. 13: The towers Water flowed from Kfar Manof to the walls of Acre by means of a pressure pipe in a series of inverted siphons in four towers topped by open tanks. The towers were apparently built at the points were the pipeline made a sharp turn, to avoid lateral pressure on the pipe that might have caused it to burst. The open tanks at the top of the towers allowed the air pressure to escape. From the tanks the water could be let out from the main line to a secondary line that served the pasha’s private estate. These towers were called in Arabic tala‘a and in Turkish – suterazi. This type of system originated in Is-tanbul, where a similar water system can be found.The names of the towers, from north to south, are: Al-Awaj, Al-Imam, David Noy and Bab Arra (right opposite the walls). 7a in fig. 13: Al- Awaj (“the crooked”) Tower deviates to the southwest. One segment of a stone pipe can clearly be seen in the tower wall, at the point where the water enters the tower, which is about 1 m higher than its out-let. In the southern wall of the tower is a sealed door and window with pillow-arch reliefs, a rosette and dentils. 7b in fig. 13: Al-Imam Tower (fig. 15) rises to a height of 13 m and is 3.5 m long and about 3.4 m wide. A pipe brought the water up to the tank, entering it from one side. The water exit-ed to a descending pipe on the other side of the tank. A similar, lower tank was connected to another pipe. One pipe was made of clay and the other of stone. An overflow pipe was in-serted in the western side of the tower wall. 7c in fig. 13: David Noy Tower measures 2.3 x 3.2 m and is about 10 m tall. The doorway in the eastern side is blocked. A light-colored, broad strip of stones can be seen diagonally crossing Pinkas Street, which is parallel to the wall from the north. This strip was installed by the Israel Antiquities Authority to mark the route of the inverted siphon under the road, which connects this tower with the fourth tow-er, Bab Arra (7d in fig. 13).



No 8 in fig. 13: Old Acre The aqueduct reached the underground reser-voirs in Acre and supplied the installations in the hamam (the bathhouse), 12 sabilim (water fountains) and cisterns in the courtyards of dwellings. We ascended to the walls of Acre, turned left, and at the northeastern corner we looked across the moat (fig. 16) and saw the fourth tower, Bab Arra. On the inner side, at the base of the wall, at the place known as the Pasha’s Gardens, is a segment of stone pipe about 50 m long. At some points the concrete can still be seen that covered the pipe to insure

that internal pressure would not burst it. The pipeline reaches the remains of the wall from the time of Daher al-Omar and ascends to a height of about 12 m. Here, right inside the wall, as well as on its other side, are three plas-tered pools. The aqueduct was uncovered dur-ing construction of the Acre Knights youth hostel and guest house, and it can be seen in the lobby, the courtyard and the dining hall. The aqueduct emerges on the southern side of the guest house and disappears in the market.

Fig. 14: The vaulted bridge near Kibbutz Lohamei HaGeta’ot (photo: Tsvika Tsuk).

◄ Fig. 15: Al Imam Tower (photo: Tsvika Tsuk).

► Fig. 16: The stone pipe in the Acre moat, against the background of the city wall (photo: Tsvika Tsuk).



97

Fig. 17: Touring route at Tel Megiddo (Drawing: Shalom Kweller; adaptation: Avigdor Orgad). Megiddo (fig. 17) was of the most important cities in the time of the Bible and was in con-tinuous existence from the fifth through the first millennia BC. It is mentioned in Egyptian, biblical and Assyrian sources. Tel Megiddo has 25 strata of settlement one atop the other. It was first excavated by Gott-lieb Schumacher from 1903–1905. From 1925 to 1939 it was extensively excavated by an American expedition, which removed some of the strata of the mound, some down to bedrock, and exposed others to various depths. Today’s visitors can still see the results of those excava-tions: the gate system of the Bronze Age and the Iron Age, as well as the complex of temples from the Early Bronze Age III (fig. 18), the north-ern and southern Iron Age stables, the Assyrian palaces, the silo and the water systems (Lamon 1935). From 1960 to 1971 five seasons of excavation were carried out at Me-giddo by Yigael Yadin, during which ‘Solomon’s Palace’ was un-covered and probes were undertaken to further clarify the dates of the water system and other strata. Beginning in 1992, an expedition has been excavating the mound under the di-rection of David Ussishkin and Israel Finkel-stein of Tel Aviv University. These excava-tions unearthed a large temple from the Early Bronze Age I, an urban complex of private

dwellings from the end of the Late Bronze Age, cult steles near the Mycenaean tomb and other finds. For most of its existence, the city at Megiddo was surrounded by a wall. The cities of the various periods at Megiddo also contained emergency water systems that enabled them to withstand a prolonged siege. Tel Megiddo was declared a UNESCO World Heritage Site in 2005, together with the tells of Hazor and Beersheba, representing biblical tells in Israel.

Fig. 18: The large archaeological trench at Tel Megid-do with the temples and the cultic platform from the Early Bronze Age, looking toward the Valley of Jezreel (photo: Tsvika Tsuk).



Touring Route 1. The Plastered Reservoir (fig. 19)

Fig. 19: Steps to the plastered reservoir (photo: Tsvika Tsuk). The Yadin excavations uncovered part of the northern water system of Tel Megiddo from the time of King Ahab. This system included an impressive staircase located outside the outer gate, which descends steeply northward to a plastered cistern. In 2006 the Tel Aviv Univer-sity expedition uncovered part of the floor of this plastered installation, which was apparent-ly a pool for the collection of surface runoff. The staircase was surrounded by thick walls and thus was part of the fortifications – as a secret passageway descending to a concealed reservoir. 3. The ‘Gallery’ (fig. 20) The ‘gallery’ is a narrow secret passage that led under and outside of the Megiddo city wall, apparently toward the opening of what is known as the well cave. The ‘gallery’ was orig-inally roofed and hidden. It is now 15 m long, but it was once longer; it is more than 1 m wide and about 2 m high. Its western, lower portion, near the entrance to the well cave and apparent-ly directly connected to the opening of that cave, was destroyed. It belongs to Phase 1 of the Israelite water system (see below).

Fig. 20: The ‘gallery,’ looking east (photo: Tsvika Tsuk). 4. The Large Israelite Water System Three phases were discerned in the Israelite water system: Phase 1 – the cave and the well (and the ‘gallery’); Phase 2 – the main phase: The well was ‘brought into the city’ by digging the shaft, the stepped tunnel and the horizontal tunnel and constructing a blocking wall; Phase 3 – the stepped tunnel was cut away and a slightly sloping horizontal tunnel (fig. 21) was hewn that led water from the spring to the base of the shaft. On our tour we descended the modern staircase – with a view of the ancient stairs on our right. In ancient times the water-drawers would stand at the top of the shaft and let down their vessels to the bottom of the shaft. We entered the hewn shaft. The modern steps were built over the remains of the ancient hewn steps. The shaft ends after two flights of steps in the hewn part and reaches a metal landing. This was the loca-



99

tion of the stepped tunnel that was cut away in Phase 3. The rock-cut steps have now been replaced by a metal staircase down to the hori-zontal tunnel. A. The shaft – Cut into the rock at a depth of 36 m, the shaft is square shaped and measures 5.0 x 6.5 m. The steps abutting the wall of the shaft reduce the amount of space in the shaft, which tapers downward to a height of 10 m above the level of the tunnel. Originally there was a staircase here, which has now been re-placed by a metal staircase. The shaft’s current form is the way it looked in its last phase of use (Phase 3), in which it was no longer possible to descend to the bottom of the shaft, but rather only to draw water from the bottom using a vessel and a rope. B. The purpose of the horizontal tunnel was to connect the bottom of the shaft to the well, to do away with the need for an outer entrance and to bring the water into the city. The tunnel is 50 m long and slopes slightly downward toward the well. Eventually the sloping tunnel was done away with and the vertical shaft was continued. The bottom of the horizontal tunnel was deepened and lengthened another 20 m to bring the well water to the bottom of the shaft, from which it was drawn upward (Phase 3). C–D. The cave and the well – The well has a diameter of about 3 m and a depth of about 0.5 m (fig. 22). In an especially rainy winter, such as we had in 1992 and 2003, the water level sometimes rises by as much as 10 m. In Phase A the well reached the aquifer outside the city of Megiddo and at its foot and had to be reached by exiting the walls. The loca-tion of the water source outside the city was a weak point during war because the enemy could easily prevent the inhabit-ants from reaching it during a siege. The ‘gallery’ (3) may have served as a secret passage to the water source. Major fluctuations in the water level made it necessary to deepen the well, thus creating a sloped cavern about 14 m below the surface.

Fig. 21: The tunnel (photo: Tsvika Tsuk).

End of use in Phase A: When it was realized that the secret passage (the ‘gallery’) did notfulfill its function, a new, much more suc-cessful idea was conceived and carefully planned: to block the well from the outside and bring the water into the city by means of a shaft and tunnel.

Fig. 22: The well, Tel Megiddo’s water source (photo: Tsvika Tsuk).



After this system was hewn, the old entrance was closed off by a thick wall (Phase B), which was camouflaged so that the entrance to the well could not be seen from outside the mound. Excavation at the site proved that the wall was built directly on the hewn steps, which obviat-ed them. Dating of the Water System According to Robert Lamon, Megiddo had a water system as early as the Late Bronze Age II, which was also the time that the well with the outer entrance came into use. Excavators, however, have information about the water supply to the city only beginning in the 12th century BC. After the earlier water system proved unsuccessful, the ‘gallery’ was built instead, probably during the reign of Ramses III, during the first half of the 12th century BC. According to the archaeological findings and a statue of Ramses VI found in the water system, the shaft, the tunnel and the blocking wall should be dated to the second half of the 12th century BC. With the collapse of Egyptian rule in the Land of Israel, Megiddo became an unwalled agricultural settlement and the water system silted up. During the time of King Sol-omon, in the mid-10th century BC, the system was cleaned and was continually used from then on until the 9th century BC. when it was destroyed together with the city. During the time of King Josiah, in the 7th century BC, the water system was repaired. The city continued in existence until around 30 BC when the shaft began to fill with mud and debris that flowed in with surface runoff. Ruth Amiran proposed dating the water system to the end of the Late Bronze Age (1350–1150 BC). Her reasons: The pottery found beneath the floor of the ‘guardroom’ and the latest stra-tum that was damaged by the digging of the shaft are from that period, and therefore the shaft and the tunnel must have been dug at this time. Yigael Yadin demonstrated that there was a clear link between the city walls and the water systems. The American expedition dated the city wall to the time of King Solomon and the ‘gallery’ that preceded it to the first half of the 12th century BC Yadin claimed that the city

wall dated from the time of Ahab and the ‘gal-lery’ was from the time of Solomon, that is, the 10th century BC and since the shaft and the tunnel post-date the ‘gallery’ they should be dated to the beginning of the 9th century BC, to the time of the Omride dynasty. Yigal Shiloh and Dan Cole concurred with Yadin. The water system at Megiddo is one of the most impressive in the ancient world. Its vari-ous phases attest to the development of water systems in the Israelite period II. Scholars dis-agree over the dating of its various phases; however, today most tend to attribute all three phases to the Israelite period II (Iron Age II, 10th–6th centuries BC). Mei Kedem, Nahal HaTaninim and Caesa-rea (fig. 23) Caesarea, along with Jerusalem, was one of the two large and important cities in the Land of Israel during the Roman and Byzantine peri-ods. Caesarea was built by King Herod be-tween 22 and 10 BC and it contained the coun-try’s main port, which was the largest in the ancient world at the time. A major aqueduct was built to bring spring water from afar to within the city. The city reached its zenith dur-ing the Byzantine period, with an area of about 150 hectares and a population of some 50,000. Caesarea’s water system is one of the most complex ever found in the Land of Israel and consisted of seven long aqueducts. The earliest of these is the High Level Aqueduct I, which was built in the 1st century BC by Herod or in the 1st century AD during the time of the proc-urators. It channeled water from the Shuni Springs to Caesarea. The second aqueduct, known as the High Level Aqueduct II, was built alongside the High Level Aqueduct I in the 2nd century AD by the Roman Legion. The third aqueduct is the Shaft Tunnel of upper Nahal HaTaninim. The fourth aqueduct came from En Tzur, and the fifth was the Low Level Aqueduct, which began at a huge dam in Nahal HaTaninim and apparently dates to the 4th century AD. This aqueduct channeled only saline water, which was used for various pur-poses other than drinking. The sixth aqueduct is the High Level Aqueduct III, which was



101

built during the Byzantine period into High Level Aqueduct II, obviating it. The seventh is the clay-pipe aqueduct from the Byzantine period, which channeled spring water from

Nahal Hadera to the southern part of Caesarea to irrigate gardens and villas. This aqueduct cannot be seen today (Porath 2002, 104-129).

Fig. 23: Map of the aqueducts to Caesarea (Avigdor Orgad). Touring Route 1. Upper Nahal HaTaninim – the Shaft Tunnel This tunnel was apparently part of the High Level Aqueduct to Caesarea. It is 8 km long, about 0.75 m wide and 1.20–1.70 m high. It was created by hewing diagonal shafts to a depth of 3–7 m. From each shaft a segment of tunnel was then hewn in the direction of the nearby shafts. Niches were cut in the walls to hold oil lamps to illuminate the tunnel. The tunnel is cut through chalk rock to the water table, turning the tunnel itself into a kind of spring. Most scholars believe that this system

was intended to augment the water system to Caesarea, but they are divided over its dating. 2. Mei Kedem – Alona Park (fig. 24) This site features a striking segment of a shaft tunnel through which visitors can walk some 220 m out of 294 m. The tunnel is 0.75–1.50 m wide, 1.50–2.50 m high and 0.40–0.70 m deep. Lines along the wall indicate past water levels, up to about 1.30 m. ‘Windows’ can be seen in two places – round openings through which water from Nahal HaTaninim apparently en-tered the tunnel. The laborers’ chisel marks can be seen along the wall, as well as niches for oil lamps at many points.



Fig. 24: The tunnel at Mei Kedem (photo: Tsvika Tsuk).

Fig. 25: The High Level Aqueduct near Bet Hananya. At left, Aqueduct I and at right, Aqueduct III with its pipes, which was built into Aqueduct II (photo: Tsvika Tsuk).

Fig. 26: Tenth Legion inscription near Bet Hananya (photo: Tsvika Tsuk). 6. Bet Hananya (fig. 25) This is one of the most fascinating portions of the High Level Aqueduct to Caesarea, with all three of its phases well preserved here. The northern side, Aqueduct II, was built during the time of the Roman Emperor Hadrian. In its wall are two inscriptions, found in 1973 on limestone blocks that stood out among the brown kurkar stones. The inscriptions note (fig. 26) that the aqueduct was built by the

soldiers of the Tenth Legion; one features the legion’s symbols in relief – an eagle (whose head is now broken off), and below it a wreath. Under the wreath is a relief of the goddess Nike, missing her head and feet. Visible to the west along the aqueduct’s arches is the southern side of Aqueduct I, whose arches are of a different width. The seam between the two aque-ducts can be seen between the arches. Aqueduct II in-corporated the protruding cornice of Aqueduct I. On top

of the aqueduct, the southern Aqueduct II and the northern Aqueduct I can be seen. Aqueduct III, which was built into Aqueduct II, consists of three parallel clay pipes. Farther to the west the split can be seen: One aqueduct continues straight while a new aqueduct turns left and makes a large detour, apparently because of the



103

Kebara Swamp and the sinking of the aqueduct wall. The construction style, the form of the arches, the space between them, the square cornice on both sides of the aqueduct and the composition of the plaster shows that the de-tour was built at the end of the 3rd century AD. Aqueducts I and II originally continued straight. Aqueduct I was dismantled intention-ally to use the stones to construct the detour. An inscription in Greek found in 1959 next to the detour aqueduct notes that it was renovated during the time of Flavius Florentius, procon-sul of Palestine, in 385 AD. 7. Nahal HaTaninim and the Low Level Aqueduct (figs. 27 & 28) This is where the Low Level Aqueduct to Caesarea start-ed (Peleg 2002, 141-147). The source of the water is the Ein Hananya Springs and the Timsah Springs, which pool behind a large dam. The dam is 194 m long, about 4 m wide and about 10 m high from the base. The outflow stoppers and the valve can be seen on the southern side. The openings leading the water into the Ottoman flour mills can also be seen. In the past, the wa-ter collected behind the dam to a height of 6 m above sea level, thus creating a very broad lake (600 hectares). The lake is now only about 0.5 hectares and the water level is about 4.5 m above sea level. The aqueduct that emerges from the southern side of the dam is rock-hewn, and is the largest in the country. Fig. 27: Map of Nahal HaTani-nim Nature Reserve (Avigdor Orgad).

In this area it is 1.8 m wide and 1.4 m deep. At this point, the aqueduct branches off to a road channel that led water to six mills in the Byz-antine period. The upper part of the aqueduct was apparently a rounded vault of which only remnants remain but whose form can be recon-structed. Visitors can walk southwest along a surviving 400 m segment of the aqueduct.



Fig. 28: Nahal HaTaninim dam (photo: Tsvika Tsuk). 9. Aqueduct Beach This 500 m-long portion of the High Level Aqueduct, which runs parallel to the Mediter-

ranean beach, features approxi-mately 80 arches (fig. 29). The base of the aqueduct is about 8 m above sea level. The three aque-ducts are revealed here very prominently and as we walked along them, we could clearly see numerous repairs. Between the arches the ‘seam’ between Aque-duct I and II can be seen: Aque-duct II, which was built during the reign of Emperor Hadrian, was built abutting Aqueduct I. Two inscriptions on white lime-

stone were uncovered on the western side of the aqueduct, facing the sea, one by the Second Legion and the other by the Tenth Legion. The Low Level Aqueduct can be seen opposite the northern part of the High Level Aqueduct and parallel to it some 80 m to the east.

Fig. 29: The High Level Aqueduct at Aqueduct Beach (photo: Tsvika Tsuk).

10. The Low Level Aqueduct This is the largest aqueduct in the country. It is about 5 kilometers long, 1.4 m wide and 1.8 m deep, and a maximum of 2,500 cubic meters of water could flow through it per hour. The sinter covers the sides of its channel only to a height of 0.85 m, which means that water usually only

filled it to 60% of its capacity. The aqueduct is roofed by a round barrel-vault 0.35 m thick, the only one of its type in Israel. The aqueduct begins with the dam at Nahal HaTaninim, some 6 m above sea level, and continues to Caesarea, most of whose dwellings were built 5 m above sea level. The bottom of the aque-



105

duct is rock-hewn and only the top is stone built. It passes west of the village of Jisr a-Zarqa, beneath the High Level Aqueduct, at which point it is completely built. The most impressing portion has been preserved at Aq-ueduct Beach. The springs are very saline and therefore they were not intended for drinking, but rather were channeled to bathhouses, pools and gardens and to wash the drainage channels out into the sea. According to Yeshu Dray, the aqueduct was intended to wash sand and solid debris from the port to the sea. The incline of the Low Level Aqueduct is 0 and its base is 5.5 m above sea level. The depth of the water and the pressure at the end point are what kept it flowing to Caesarea. The aqueduct was appar-ently built in the early 4th century AD and remained in operation until the end of the Byz-antine period (the 7th century AD).

16. The Theater The Roman theater (fig. 30) was in use in the 3rd and 4th centuries as a venue for water games, called naumachia. Two decorated piec-es of marble in secondary use were placed in the two entrance corridors to the orchestra (the semicircular space between the stage and the tiers of seats). Grooves were carved into these elements to insert wooden boards to keep the water in the orchestra, creating a pool about 1 m deep. Mock naval battles were held in this pool. Former generals seated in the audience gave orders to direct the ships, which were moved accordingly by stage hands in the water. North of the theater a plastered pool was found measuring 10x13m and 1 m deep, to which water from the aqueduct was channeled and from which the water was then channeled by pipe to the orchestra.

Fig. 30: The theater (photo: Tsvika Tsuk).



Bibliography Amit, D. Patrich, Y. & Hirschfeld, Z. (eds.) 2002, The Aqueducts of Israel, Portsmouth. Frankel, R. 2002, The Hellenstic aqueduct of Ak-ko-Ptolemais. In: Amit, Patrich & Hirschfeld 2002, 82-87. Lamon, R.S. 1935, The Megiddo Water System, Chicago. Macalister, R.A.S. 1912, The Excavations of Gezer I, London. Meyers, E.M. (ed.) 1999, Galilee through the Cen-turies Confluence of Culture, Winona Lake. Netzer, E. 2002, The aqueducts and water-supply of Masada. In: Amit, Patrich & Hirschfeld 2002, 353-365. Peleg, Y. 2002, The dams of Caesarea's Low-level aqueduct, In: Amit, Patrich & Hirschfeld 2002, 141-147. Porath, Y. 2002, The water-supply to Caesarea: a re-assessment, In: Amit, Patrich & Hirschfeld 2002, 104-129. Reich, R. 2011, Excavations the City of David where Jerusalem's History Began, Jerusalem. Tsuk, Ts. 1999, The Aqueducts of Sepphoris. In: Meyers 1999, 161-175. Tsuk, Ts. 2002, The Aqueducts to Sepphoris. In: Amit, Patrich & Hirschfeld 2002, 278-294. Tsuk, Ts. 2011, Water at the End of the Tunnel, Jerusalem (Hebrew).

Dr. Tsvika Tsuk Chief Archeologist Israel Nature and Parks Authority Yosef Haglili 18a Ramat Gan 52416 Israel E-Mail: [email protected] [email protected]

Touring Israel's ancient water systems

Documents