Khirbat Umm el-‘Umdan (c. 12 dunams), located in Modi‘in, was first surveyed in 1873–1874 by Clermont-Ganneau (Clermont-Ganneau 1896:52–83), who documented architectural remains, including a church with five columns and a mosaic floor. According to Clermont-Ganneau, the name of the site originated from the numerous columns that were visible on the ground. Three seasons of excavations were conducted at the site in 2001–2002 (HA-ESI 114:64*–68*) and remains of a Jewish settlement, which dated to the Persian–Hellenistic periods and from the Hellenistic–Early Roman period to the time of the Bar Kokhba Revolt, were documented and exposed. The antiquities ascribed to the Byzantine period consisted mostly of rock-hewn and built agricultural installations and remains of roads. Building remains and retaining walls that dated to the Umayyad period were discovered and during the Early Islamic period, the region was used as a necropolis. The current excavation was located c. 100 m southeast of the site’s center.
In 2003, an excavation along the routes of Highways 11 and 12 in Modi‘in was conducted (HA-ESI 120); rock-hewn agricultural installations, including winepresses, cupmarks and bodedot, were exposed, as well as built agricultural installations, namely watchman’s huts, fences, farming terraces and roads. In addition, cisterns and burial complexes, dating to the Roman and Byzantine periods, a crushing stone of an olive press and quarries, were revealed.
In 2010, an excavation was conducted c. 100 m down the slope of the wadi to the west; a large dam that was built of large stones placed across the wadi was exposed (Permit No. A-5944). In April 2012, an excavation was carried out in a lot whose southwestern corner abutted the northeastern corner of the current excavation (HA-ESI 125). Farming terraces and an agricultural installation, consisting of a treading floor and a collecting vat, were exposed.
Ten farming terraces were exposed in the current lot, on a steep slope that descends toward a deep channel. The terraces were aligned east–west in accordance with the contour lines, from the high part of the slope to the bottom of the channel (W1; W3–W10; Fig. 2). Some of the farming terraces were preserved to a length of 1.0–1.5 m, while others survived several meters long. The farming terraces were built of boulders and various shaped rocks in dry construction. Some were built on top of natural bedrock terraces and their bottom layer was set directly on the bedrock. In those places where no bedrock terrace was evident or a depression was in the rock, the bottom layer of the terrace was placed on soil and fieldstone fill. Some of the terraces were preserved 1–2 m high, while others were 0.5–1.0 m high.
Farming Terraces 1–4 (Fig. 3)
A section of a farming terrace (W1; Fig. 4) that was built of large fieldstones placed on the bedrock was excavated. The terrace was preserved to a maximum of three courses high. Large fieldstones were sometimes set on a foundation of medium-sized fieldstones. The terrace was built along a contour line that descended gently eastward.
A farming terrace (W2) was excavated; it was built of large fieldstones, some of which were placed directly on the bedrock and some on medium-sized fieldstones that served as a base intended to stabilize the terrace. Three walls near the northern side of the terrace formed a small square structure (L101; Fig. 5) that probably served as a watchman’s hut. Gaps between the walls indicate that some of their stones had been robbed.
Two short farming terraces (W3, W4), built one slightly above the other, on a contour line that descends gently toward the channel at the bottom of the slope, were excavated. The terraces were built of different size fieldstones and were preserved to a maximum three courses high.
Farming Terraces 5–9 (Fig. 6)
Farming terraces (W5, W9) were excavated in the middle of the northern slope; they were built of large fieldstones with medium fieldstones between them and were set on the bedrock or on a foundation of small fieldstones. Two other farming terraces (W6, W8) were excavated on the lower part of the slope; these were built of small and medium fieldstones set on top of soil fill. The terraces were preserved for just several meters long. Near the channel, where the slope is moderate, large fieldstones were presumably not needed for building the terraces (Fig. 7).  
Farming Terrace 10 (Fig. 8)
A farming terrace that faced the wadi channel at an oblique angle was excavated at the bottom of the slope. This orientation might have been the result of a partial collapse of the terrace. The terrace was built of large fieldstones, set on directly on the bedrock, and was just several meters long.
No potsherds that could assist in dating the ten farming terraces and the watchman’s hut were found in the excavation. The terraces might have been built or used by the residents of the nearby village of Umm el-‘Umdan.

Clemont-Ganneau C.S. 1896. Archaeological Researches in Palestine during the Years 1873–1874. Vol. II. London.