The site of Tell Shemshara itself has suffered heavy erosion since the formation of Lake Dokan. The 2012/3 survey of the site shows an estimated 164.000 cubic meters loss from the area mapped in 1957 alone! The entire site covers three natural hills: Main Hill, North Hill, and Camp Hill, which form a triangular area originally framed by the Lower Zab to the east and the Boskin Wadi to the west. Most of the site shows traces of the earliest Neolithic and Hassuna occupation exposed in small areas by the Danish 1957 excavation, and currently investigated by a team from the University of Reading (directed by R. and W. Matthews). After a long interval Shemshara again saw major occupation from the late 3rd into the 1st quarter of the 2nd Milllennium BC. The old excavations famously exposed portions of an 18th cent. palace on Main Hill, Level V, where two separate archives of cuneiform tablets, in total ca. 250, were found. The new excavations have exposed/re-exposed a major part of this structure, and probed the earlier levels VI-IX, all apparently featuring major administrative structures. Finds (as of early 2014) include two cuneiform tablets retrieved in Levels VIIIa and VIIIb. The early Level IX seems to define this era of occupation, but has as yet only been exposed in small areas. After another long interval Shemshara was again occupied in the Medieval period (Levels I-III), and numerous clay tobacco pipes indicate extensive Ottoman period squatter(?) occupation.