Our Sites
The Rania Archaeological Project has investigated numerous sites across the Rania Plain since 2012. Our work combines excavation at key sites with systematic survey of the flood-risk zone around Lake Dokan, where dozens of ancient settlements face ongoing destruction from cyclical inundation.
This section presents the principal sites where our project has conducted fieldwork, from major multi-period tells like Shemshara and Basmusian to smaller, single-period sites that offer crucial evidence for specific moments in the region’s history. Together they document a settlement history spanning from the Neolithic to the Sassanian period, and illustrate both the extraordinary archaeological richness of the Rania Plain and the urgent threats it faces.
Navigate with us (video from the survey of December 2023).
Tell Shemshara
Excavations at Shemshara with a view north to the pass at Darband.
Tell Shemshara is the best-documented site on the Rania Plain and the focus of the Pisa Archaeological Project. The site is a complex of natural hills covering approximately 7 hectares, strategically positioned between the Lower Zab river and Wadi Boskin. Excavations by Danish (1957), Iraqi (1958–1959), and more recent teams (NINO 2012–2017; University of Pisa 2018–present) have revealed a long but discontinuous occupation history, from the Neolithic (Hassuna-Samarra, ca. 7300–6000 BCE) to the Ottoman period.
The most significant discovery is the Level V palace of Kuwari, governor of ancient Shusharra, destroyed by fire ca. 1780 BCE. Approximately 250 cuneiform tablets found in the palace document a dramatic year of political turmoil, when the site was caught between the expanding empire of Shamshi-Adad and rival mountain kingdoms. Recent excavations have revealed earlier Bronze Age palatial levels (VIII–IX, ca. 2100–1900 BCE) and, in 2023, the first evidence of Iron Age (Neo-Assyrian) occupation at the site.
Threats
Tell Shemshara is located close to the edge of the flood-zone, and most winters and springs turns into an island site. The lower portions of Main and North Hills have suffered severe damage from decades of intermittent flooding. The high northern part of Main Hill has lost approximately four metres of elevation since the 1950s due to erosion caused by cyclical inundation.
The east face of Main Hill in autumn 2023, showing the site’s geological setting and the dramatic down-cutting of the natural section caused by the lake.
The cut east section viewed from south.
Redeposited soil with embedded modern toy doll on the west slope of Shemshara.
Tell Basmusian
Baked brick filler-walls, presumably of the Middle Assyrian terracing operation, enveloping the original summit of the site with its early 2nd mill. BCE temples.
Tell Basmusian is one of the largest archaeological sites on the Rania Plain, comprising a high mound of approximately 2.1 ha reaching 24 m in height (511 m asl at its summit). The site can now be securely identified with ancient Pakute, a Middle Assyrian stronghold known from foundation inscriptions of Tiglath-pileser I (ca. 1114–1076 BCE).
Iraqi excavations in 1956–1958 revealed a sequence of three superimposed temples dating to the Middle Bronze Age, while lower levels documented occupation from the Halaf/Samarra period through Uruk into the third millennium BCE. The NINO project revisited the site by boat in 2015, documenting severe erosion and observing baked brick “walls” eroding from the summit.
In autumn 2025, exceptionally low water levels allowed our project to open eleven test trenches across the site. These revealed substantial architectural remains at the base of the mound, partly dated to the Iron Age. Near the summit our project could plot a system of baked brick retaining walls probably representing terracing for the fortification works described in Tiglath-pileser’s inscriptions. Test trenches on the south-eastern and western flanks exposed Late Chalcolithic/Uruk period deposits, including floors with typical fourth millennium BCE ceramics.
Threats
Tell Basmusian lies within the flood-risk zone of Lake Dokan and has suffered extensive damage from decades of cyclical inundation. The site survives as a heavily eroded island throughout most of the year, accessible only by boat, though during exceptional droughts it can be reached by car. The continuous cycle of flooding and exposure has caused severe erosion: walls, foundations, baked bricks, ovens, and other features are eroding out of the slopes, while thousands of sherds have washed out and been redeposited by the lake in thick concentrations on the lower parts of the mound. The high summit, which once featured the second millennium BCE temple platform, has retained its squarish shape but the surrounding slopes have been dramatically reduced.
Drone image of the site in October 2025. The northern ridge mainly consists of material deported from eroded slopes of the site and redeposited by the waves of Lake Dokan.
Drone oblique view of baked brick filler-walls below the SW edge of the summit.
Ninevite 5 sherds (early 3rd mill. BCE) from fill in Trench 11 on the south slope of Basmusian.
Modern impression of one cylinder seal found on surface in October 2025.
Araban
Drone image of Araban from south (October 2025).
Araban is a site now apparently only surviving as a ca. 750 m long stretch of sub-surface pits on the east side of the lake. Surface sherds on part of this area include painted Ninevite 5 examples (early 3rd mill. BCE), but the pits, many eroded almost to their ancient bottoms, contain Late Chalcolithic sherds, bones, and flints. These ancient trash pits represent what seems the largest and perhaps most important Late Uruk period site on the plain.
Occupation from the same period (LC 3–4, late 4th mill. BCE) is found on many smaller sites and clearly cultural influence from the Mesopotamian area was strong in the Rania Plain. In 2015 our project conducted a small test excavation in a large trash deposit on the northern part of Araban. In 2025 we were able to reassess the actual extent of the site and collect additional surface sherds.
Threats
Most of the Araban site is seasonally inundated. Since fairly briefly occupied in antiquity Araban was probably already much eroded by the 1950s, but is now facing slow and inevitable erasure. In many of the pits pottery is visible on the surface and tempts casual visitors to illicit digging.
Typical pit surface on the Araban site.
Example of Uruk sherds excavated in 2015.
Ninevite 5 sherds from surface of Araban, collected in 2023.
Complete Late Uruk jar excavated in 2015.
Baiz Agha
Drone image of the central part of Baiz Agha, with test excavations 2025 in progress.
Baiz Agha was not identified as an archaeological site by the 1955 Iraqi survey, probably because it was already then a very eroded and flat plateau, and not an elevated mounded site. We first heard of Baiz Agha from an employee in the local Raparin Antiquities’ Directorate, and visited it by boat in autumn 2015, when it appeared as a roughly triangular island. On the higher central area were numerous stone foundations of ancient buildings and surface sherds clearly datable to the Middle Assyrian period (ca. 1400–1100 BCE).
Baiz Agha was visited also in December 2023, again by boat, and more materials collected. In autumn 2025 finally the site was also accessible by car, and in late September our team spent a week at the site to do small test excavations. These operations showed that the very denuded stone foundations covered an older level, again much eroded and apparently levelled in preparation for a new occupation. Our ceramicist, C. Coppini, confirmed a date for the occupations on Baiz Agha to the later part of the Middle Assyrian period (ca. 1250–1100 BCE).
Viewed in conjunction with the Middle Assyrian occupation at Basmusian we now have the contours of an Assyrian attempt to control the strategically important Rania Plain. An interesting feature of the ceramic material from Baiz Agha is a local component of very rough reddish pottery, which we have identified on several other Rania Plain sites without specific Assyrian ceramics. This discovery now provides an approximate date for such ceramics and the sites where it occurs.
Threats
We have registered little change of preservation between visits and estimate that the shallow stone foundations on the surface are well-anchored and basically likely to remain in place for many years. The most imminent threat to the archaeological levels is looting, and in 2025 we noticed several robber holes made by recent visitors.
Baked brick pavement of Level I in Trench 1 (September 2025).
Example of Middle Assyrian sherds excavated in 2025.
NN_3
Drone image of NN_3 with remains of stone foundations (October 2025).
This site, located on the south edge of the lake, was first identified in October 2025, and we have not yet been able to establish a name for it. It appears but seldom above the waters of the lake, but then as a roughly triangular and fairly low site. On the east slope are still preserved some stone foundations, and strewn around the upper part of the site are surface sherds in medium-low concentration.
An interesting feature is the presence of the same roughly made and reddish pottery which our project has retrieved on Baiz Agha, and the site can be added to the still very modest late 2nd mill. BCE occupation identified on the Rania Plain.
Threats
As far as we know the site was never registered previously and this renders it difficult to estimate the degree of erosion, but judging by the shallow stone foundations now surviving it must have been severe.
Late 2nd mill. BCE sherds from surface (October 2025).
Gird Mamand
View north from Mamand towards Basmusian (2014).
Mamand is a large, flat and very eroded site chiefly occupied in the Sassanian period. Our project conducted extensive mapping and survey on Mamand in October 2014.
Threats
Mamand is located on the west bank of Lake Dokan and within the flood-zone. It is thus prone to further erosion from the passing water of the lake and is also damaged by seasonal cultivation.
Survey at Mamand in 2014.
Stamped Sassanian sherd from Mamand.
Stamped Sassanian sherd from Mamand.
Kendi Rash
Drone image of Kendi Rash in upper left (September 2025).
A low-contour, oval site on the west edge of Lake Dokan, first identified and sampled in autumn 2025. The collected ceramics seem predominantly from the Sassanian period.
Threats
Mostly inundated and in danger of progressive erosion by the wave action of the lake.
Sample of surface sherds collected in 2025.