Spatial analysis and neotaphonomy data from two landscape bone assemblages in the Ngorongoro

Two 100 m x 200 m survey blocks, located approximately one kilometer from each other and referred to as Olduvai Transect 1 (OT1) and Olduvai Transect 2 (OT2), were visited in 2015 (OT1) and 2018 (OT2) at the height of the dry season in mid-June or July. OT1 occurs in a treeless section of the plain , while OT2 contains a stand of trees surrounding a seasonal waterhole (Figure 2). A series of abandoned Maasai homesteads (bomas) were located to the south and to the west of OT2 at the time of data collection. The survey blocks’ borders lie along an arbitrary grid system established with a laser total station. Systematic pedestrian survey was carried out to document the spatial distribution of bone specimens within these borders. The survey team was arranged in a line and spaced one arm’s length apart (~65 cm). The survey line progressed back and forth in transects at a slow walk. Individuals at either end of the survey line used compasses and regularly checked and adjusted spacing to ensure that parallel, non-overlapping transects were maintained. Vegetation is sparse in both survey areas, so ground visibility was generally excellent. All specimens—whether fully exposed or partially buried—were marked with colored pin flags (in some cases, tight clusters of bone specimens were marked with a single pin flag). Partially buried bones were generally quite conspicuous in the soft, fine sand substrate. No attempt was made to identify or recover fully buried material through excavation. While most specimens were encountered during systematic transect survey, a handful were discovered ad hoc, either as the survey team roamed the block between transect runs or as previously flagged bones were being recorded. Each specimen was plotted to the nearest millimeter at its center point with the total station and then collected. Articulated remains were plotted separately. Given the close spacing and slow pace of the survey teams and the excellent visibility, we are confident that high recovery rates were attained for all fully exposed and partially buried remains. Our analysis focuses on all recovered specimens > 2 cm in maximum dimension. To our knowledge, neither survey block experienced any systematic collection in the past, and the presence of bones in weathering stage 5 suggest that both faunal assemblages represent at least 12 years of deposition (Behrensmeyer, 1978). At OT2, all trees equal to or greater than three meters in height were also plotted with the total station by measuring three to five points around the circumference. The resulting points were imported into ArcGIS Pro (version 3.2.0) and connected to form individual polygons. The “Feature to Point” geoprocessing tool calculated the centroid of each polygon. These tree centroids were used for all distance-to-tree calculations (see below). Topographic points were recorded in UTM coordinates with a Trimble GeoXH 6000 and then converted into local total station coordinates. Elevations were recorded as meters above mean sea level.