Section A1 was excavated near Olive Tree I, to its southeast, on both sides of an agricultural boundary wall that was founded on bedrock and preserved four courses high (W100; excavated length 3.5 m, width 1.1–1.2 m). The wall was built of two rows of large and medium-sized local fieldstones with a core of small stones between them. The wall runs parallel to the streambed for at least 12 m. Its north face is straight, whereas its south face is crooked and uneven, even though both faces are similarly built. No quarrying marks were discerned in the bedrock uncovered on either side of the wall. A layer of alluvium (c. 1 m deep) rich in small and medium-sized stones was documented to the northeast of the wall. A uniform yellowish layer of loess devoid of any stones (1.2–1.3 m deep; Fig. 2) was excavated to the southwest of the wall.
Section A2 was opened to the east of section A1, above the dam wall (W101; 2.5 m excavated length, width 1.5 m, length at least 5.5 m) and at the base of the dam, near Olive Tree II. The wall, built of large and medium-sized local stones (11–12 courses, height at least 2.3–2.4 m), spanned the streambed. Its lower outer face slopes slightly in the direction of the water flow. Two layers were documented: an upper layer (depth up to 0.4 m) rich in small and medium-sized stones; and a lower layer of loess soil with a uniformly yellowish hue. Due to technical limitations, the excavation did not extend down to bedrock through the loess layer. The upper part of the wall included a later course, which was built on a layer of loess (see Section A4, below).
Section A3 was excavated near and to the north of Olive Tree II. The excavation reached a maximum depth of 0.7 m, unearthing two layers similar to those described in Section A2, above.
Section A4 was excavated above Dam 101, at a place where an elongated pile of loess soil, which slopes from the cliff higher up the streambed, ends. As the excavation along W101 was deepened by 0.2 m, it became apparent that the upper dam wall was composed of a single stone course. A section in the pile of soil above it revealed a 0.6 m-deep yellowish loess sediment. The excavation also indicated that the top course of the wall was built above a similar layer of loess soil.
Section A5 was excavated to the north of Olive Tree II and c. 2 m east of W100, to the south of a large and wide natural basin that was probably carved out by floodwater. The section was dug at the top of the alluvial layer, which contains small and medium-sized stones and overlays a second layer of grayish brown soil that is not homogeneous. Extending 0.6 m down from the surface, the section cut through the second layer and down to bedrock.
The five sections excavated beside the two olive trees in Nahal Zetan provide a better understanding of the agricultural system that was established here (Kedar 1957; Tepper and Bar-Oz 2016). The excavation shows that the system was designed to collect homogenous, stone-free loess that formed soil layers on both sides of the dam. The runoff water collected on the slopes probably flowed to these locations together with the loess. In places where layers containing large quantities of stones were found inside the area enclosed by the boundary walls, the stones were apparently washed here by more recent flooding. The pile of loess above the dam was also deposited rather late and after the agricultural system went out of use, indicating a period of abandonment. The meager pottery finds from the site indicate that the dam was built in the Byzantine period and that there was also human presence at the site in the twentieth century CE. OSL samples taken from the loess sections and samples of organic material analyzed by radiocarbon dating will enable researchers to date the Nahal Zetan trees and the agricultural system with greater accuracy. Botanical samples collected during the excavation and from the trees will help identify the specific olive variety that can thrive and yield fruit in extremely arid desert conditions. The study of the agricultural system and of the trees planted in antiquity that survived to this day contributes to our understanding of the settlement sites that flourished nearby during the Byzantine period (Tepper, Weissbord, and Bar-Oz 2015; Tepper et al. 2018).
Ashkenazi E., Chen Y., Avni Y. and Lavee S. 2015. Fruit Trees’ Survival Ability in an Arid Desert Environment Without Irrigation in the Negev Highlands of Southern Israel. Israel Journal of Plant Science 62/1–2:1–12.
Baumgarten Y. 2004. Map of Shivta (166) (Archaeological Survey of Israel). Jerusalem.
Kedar Y. 1957. Ancient Agriculture at Shivta in the Negev. IEJ 7:178–189.
Rotem R. 1971. The Olive trees in Nahal Zetan. Teva Ve-Aretz 13/5:234 (Hebrew).
Tepper Y., Weissbord L. and Bar-Oz G. 2015. Behind Sealed Doors: Unravelling Abandonment Dynamics at the Byzantine site of Shivta in the Negev Desert. Antiquity Project Gallery 348.
Tepper Y., Erickson-Gini T., Farhi Y. and Bar-Oz G. 2018. Probing the Byzantine/Early Islamic Transition in the Negev: Shivta in Light of the Renewed Excavations, 2015–2016. Tel Aviv 45:55–87.
Woolley C.L. and Lawrence T.E. 1914. The Wilderness of Zin: Archaeological Report (PEF 3). London.