GROUND-RUPTURING EARTHQUAKES AT THE ARCHAEOLOGICAL SITE OF QASR TILAH, DEAD SEA TRANSFORM, JORDAN

The Qasr Tilah archaeological site is located along the northern Wadi Araba fault portion of the Dead Sea transform about 8 km south of the escarpment to the Dead Sea Ghor. The site contains a fort, a water reservoir (birkeh) with aqueducts leading to it from the adjacent Wadi Tilah, and aqueducts leading from the reservoir to agricultural fields. Charcoal collected from the foundation mortar and upper wall mortar of the birkeh reveal that the water reservoir at this site was built ca. 641-687 A.D. (Niemi, 2000; 558-776 A.D. in Klinger et al., 2000), indicating its use in the Late Byzantine to Umayyad period. Repairs to the upper northwest corner wall of the birkeh that is built over the active fault suggest that seismic subsidence occurred on the fault possibly during construction of the structure. Recent detailed archaeological excavations in 2001 and 2003 and mapping of stratigraphic sections reveal evidence for at least three earthquakes that postdate construction of the reservoir and aqueduct. The most recent earthquake clearly cuts through sedimentary layers that are very near to the surface indicating a very recent age of faulting. At least two faulting events cut sedimentary layers full of mortar, charcoal, building blocks, and other tumble debris from the collapse of the birkeh. These data indicate that there are at least three earthquakes that post-date the collapse of the birkeh wall. Several earthquakes are known from historical records to have occurred in the vicinity of the southern Dead Sea, Kerak, and Wadi Araba. These include the earthquakes of A.D. 31, 363, 659/60, 1068, 1212, 1293, and 1456-59 1588 (Ambraseys et al., 1994; Amiran et al., 1994). At this time, we do not have enough age control to assign any of the known historical earthquakes to this section of the fault. However, radiocarbon samples collected from key stratigraphic horizons and ceramic analyses of potsherds collected in archaeological context should enable us to date the individual seismic events identified in the trenches.