Figure 1 - uploaded by Shyama Vermeersch
Content may be subject to copyright.
Map of the southern and central Levant showing the sites used for our integrated dataset.

Map of the southern and central Levant showing the sites used for our integrated dataset.

Source publication
Thesis
Full-text available
The Bronze (3,600 BCE – 1,200 BCE) and Iron (1,200 – 586 BCE) Ages in the southern Levant witnessed major social, political, and cultural changes. These include the first development of complex urban-based settlements, and the genesis of new cultural identities. Such changes are deeply entangled with, and often driven by, developments in agricultur...

Contexts in source publication

Context 1
... southern Levant is situated in south-western Asia and corresponds to modern-day Jordan, Palestine, Israel, and southern Lebanon ( fig. 1). Although this region only encompasses a small territory, it is home to highly diverse topographical conditions and climatological differences, the latter will be discussed more in-depth in section '1.2. Climatic Background'. There are four different longitudinal topographic belts in the southern Levant: the coastal plain, the western ...
Context 2
... P, which is known as the palace area, contains a stratigraphic sequence from the MBA to the LBA and consists of an MBA palace, and domestic structures dating to the LBA ( Streit et al., 2018). An overview of the relative dating and corresponding strata, a map of Lachish showing areas P and S, and a map showing the location of Lachish, can be found in table 1, figure 1, and figure 8, respectively, of Vermeersch et al. (2021b). ...
Context 3
... CCA of all taxa against MAP shows there are differences in taxa composition in different zones of MAP (pseudo-F = 2.0, P = 0.002). The CA (see fig. 11 of Vermeersch et al. [2021a]) of all taxa against MAP shows three clusters: 200 -400 mm, 400 -600 mm, and 600 ...
Context 4
... abundance. Besides this, we believe NISP is the most appropriate counterpart to seed count in that both are primary and observational data, but a more in-depth discussion on this can be found in part 4.3. In the end, only 16 sites fulfil our criteria, represented by 24 samples spanning a time period from Early Bronze Age I to Iron Age II ( fig. 1) (3,600-586 BCE). These samples consist of 21 faunal taxa (36,626 specimens) and 111 botanical taxa (199,669 specimens). First, we analyse the botanical and faunal datasets separately for trends and check to what extent they are similar, then we analyse the integrated dataset and compare the trends found here with the trends found in ...
Context 5
... axis. They mostly contain red deer, boar, felids and fallow deer. Among the crops freethreshing wheat, flax, pomegranate, grape and bitter vetch are abundant, all being well-known for their preponderance in sites dating to the Iron Age II. We see a similar chronological pattern in the CA of the integrated dataset containing only domesticates ( fig. 10). The cumulative explained variation for the four axes is 58.71%. The EBA and MBA samples plot separately from the LBA and IA samples. The EBA and MBA samples have large numbers of cattle, ovicaprid, but also donkey, dog as well as by the glume wheats. The LBA and IA samples contain higher proportions of horse, free-threshing wheat, ...
Context 6
... can see weak patterns related to ranges in modern mean annual precipitation when looking at the CA plot for the integrated dataset containing all taxa ( fig. 11). Sites with precipitation between 200-400mm contain higher proportions of fox, horse, pig, cattle, ovicaprid, gazelle, donkey, roe deer, and dromedary. They have low amounts of seed records which are often dominated by barley, to a lesser extent olive, and in EBA sites by glume ...
Context 7
... in the pollen diagrams from this region, which also outline an increased olive cultivation with the beginning of Iron Age I, reaching almost the same extent as during the moister EBA ( . The diachronic similarities in the abundance of this important crop species may also influence the association of some EBA and IA sites in the CA plot ( fig. ...
Context 8
... as early as the Pottery Neolithic (Ussishkin 2004 We discuss the faunal remains from the Austrian-Israeli Expedition to Tel Lachish (2017)(2018)(2019), led by Streit and Hö f lm ay er , in the f r amewo r k of the ' T r acin g Transformations' project. The excavation focusses on two areas of the tell area S (deep section) and area P (palace area) (Fig. 1). Area S contains a long stratigraphic sequence from the LBA to the IA representing a settlement area (Table 1): strata S-1 and S-2 contain several pits and densely laminated layers rich in seeds, and stratum S-3 contains a monumental building consisting of several walls. Area S contains settlement assemblages, and Ussishkin (2004) ...
Context 9
... two axes. There are no statistical tests which can provide the significance of the results for CA and DCA, so it relies on a critical, visual assessment of the data. It is useful to visually classify the samples per parameter to aid in interpretation. For example, in our dataset, we give samples with a different relative dating a different symbol (Fig. 1). The closer a taxon is located to the origin of the biplot (the point where the two axes cross), the more common it is throughout all samples and as such is not ideal to differentiate between samples. Samples with similar taxa composition and proportions will plot in close vicinity to each other, essentially forming a group or cluster. ...
Context 10
... the distribution of the parameters used in our dataset ( Table 2). The raw data we used for the correspondence analyses and the legend of the sites can be found in SI 1. We use the Canoco 5 software to perform a detrended correspondence analysis on our integrative dataset and classify our data using relative chronology. The biplot presented here (Fig. 1) has been cleaned and is not the original output from the Canaco 5 software (we provide a step-by-step manual on how to use Canoco 5 most efficiently in the additional information section of this paper). First, we try to observe clustering of samples, or relationships in the biplot representing the first two axes. We see that the ...