A Bayesian approach to archaeological survey in north-west Jordan

Jordan. This survey's main goal was to discover evidence of Late Neolithic habitation and landscape use, which typical archaeological surveys in the region very rarely find. To improve the efficiency and frequency with which sites are located, we experimented with Bayesian optimal allocation methods, originally designed for naval searches during the 1940s. The survey's design began with a predictive model in a GIS environment in which we assigned prior probabilities to landscape elements thought most likely to have survived the severe alteration of the landscape that has occurred since the Neolithic in this highly eroded region. Optimal-allocation algorithms, performed iteratively on the basis of each day's survey results, guided the allocation of survey effort to spaces with high probability densities, given past survey coverage. After two survey seasons, we conducted small-scale excavations in 2014 on "candidate sites" located during survey. This paper will discuss the survey methods and briefly touch on the results of these excavations, which verified that two "candidates" were Yarmoukian and Wadi Rabah Late Neolithic sites.

ADAJ, 2013

Archaeopress Publishing Ltd eBooks, 2023

Archaeological sites exist within landscapesthe surrounding physical, cultural and biological environments that provide them with context and driving factors for development. Landscape archaeologists ask questions like "What is the importance of water in determining site locations?", "What determines the location of roads?", or "How did people in ancient settlements use natural resources?" Since 2009, our team at the German Protestant Institute of Archaeology has been examining an area in northwestern Jordan with these questions in mind (Figure 1, and Vieweger and Häser 2017). The aim was to get a thorough understanding of the archaeological site. Geography Arab area in northern Jordan is an area rich in historical and prehistorical settlement. 328 sites are known from this area, of which 197 were relocated and documented in this study. The rest were known from literature from previous surveys, but were either destroyed in the last decades or we were not able to relocate them. Thus, the current study includes 197 sites ranging from the lithic epochs (until 3600 BCE) to the Ottoman era (ending 1918, at least 6500 years) were found occupation history (Vieweger and Häser 2017). Arab is a typical Levantine landscape on the edge of the Jordan Valley. It is rich in plant species, and there are many springs-an important consideration for settlements. Away from the valleys, the landscape is a mixture of steppe, shrub land and oak woodland, and water sources are scarce. (Shmida et al. 2020).

Annual of the Department of Antiquities of Jordan, 2023

Our project, “Climate, environment, and early farming societies: Late Neolithic settlement patterns on the Karak Plateau, Jordan”, aims to study the resilience to climate change of early farming communities in Jordan. It will achieve this by studying Late Neolithic settlement patterns and chronology through ground survey in combination with experimentation in the use of remote sensing and predictive modelling to aid in the discovery of these sites. Late Neolithic sites are under‑represented in the record not just because of a research bias (for example not collecting or studying chipped stone during surveys), but also because of other factors, including: a lack of training of survey crews in prehistory; the poor preservation of Neolithic pottery; the scarcity of diagnostic tools; the sites often being small; and site location being especially prone to have been covered by colluvium or later occupation. While deflated sites on hilltops and in steppe and desert areas may be found during intensive survey, in wetter areas many parts of the Neolithic landscape will have been destroyed The Karak ‘Neolithic’ Survey: PILOT SEASON, OCTOBER 2021 ADAJ 61 – 166 – by erosion (wadi downcutting) or covered by colluvium (Banning 2015). An approach that uses an iterative Bayesian allocation approach to target areas with a higher probability of containing preserved prehistoric remains has been successful in northern Jordan (Hitchings et al. 2016). We aim to use the same approach, although the pilot season presented here was used mainly to gather more data to assign such probabilities within the study region.

Geoarchaeological survey conducted in Wadi Ziqlab focused on reconstructing landscape change in the river valley and relating these changes to shifts in prehistoric settlement. Micromorphology was used to identify a Late Pleistocene paleosol that once comprised a relatively continuous soil cover throughout the main valley and allowed stratigraphic correlations between examined sections. Warmer and wetter conditions during the Late Pleistocene supported substantial occupation of the wadi during this time at the Epipaleolithic site of 'Uyun al-Hammam, and other contemporary sites in the wadi. The complete lack of late EP and early Neolithic sites throughout the wadi result from large-scale erosion during the Younger Dryas, which removed any traces of these sites. Later Neolithic and subsequent occupation of the valley occurred within a notably different climatic regime. The pattern of prehistoric occupation and differential preservation of the archaeological record emphasizes the importance of local geoarchaeological investigations as the depositional sequence found in any one wadi is not necessarily traceable to other wadis.

Journal of Landscape Ecology, 2017

As an integral part of the Gadara-Region-Project, a survey of the Wādī al-'Arab region was conducted during the years 2009-2012, by the Biblical-Archaeological Institute Wuppertal and the German Protestant Institute for Archaeology in order to achieve a better understanding of the hinterland of the main study site Tall Zirāʿa and to provide answers concerning settlement pattern, trade relationships and the importance of sites throughout time. On the basis of this survey we used ecological approaches to see what correlation might exist between archaeological sites and habitat. Since more than half the sites in this survey had Roman occupation, we asked what difference, if any, was there in the distribution of Roman sites compared to previous occupations. A comparison was made of "new" Roman sites (those not previously occupied in the Hellenistic period) with those that had both Roman and Hellenistic occupation. Open water, riverine habitats, and large archaeological sites all seemed connected. In addition, analysis indicated a correlation of older (more successful or established?) sites with open water and new Roman sites were less related to water. We knew that Roman engineering both of cistern systems and aqueducts opened new areas (such as plateaus) for settlement and exploitation. Hence the weaker correlation of new Roman sites with natural water was reasonable.

Levant, 1998