Fig 1.
The study area and location of sampling sites.
Left: The study area in north-east Tanzania, also showing Ngorongoro and the Mbulu Plateau. Right, sites from which molluscs were collected in 2016. The circle indicates the base camp on the Garusi River. Contains information from OpenStreetMap and OpenStreetMap Foundation, which is made available under the Open Database License.
Fig 2.
Stratigraphic scheme and geochronology of the main lithological units at Laetoli (adapted from [20]).
Table 1.
Pliocene fossil molluscs recorded from the Laetolil and Ndolanya Beds [1,14,18,19].
Table 2.
Mollusc sampling sites at Laetoli in 2016, and on the Mbulu Plateau in 1998.
Table 3.
Number of species and individuals found at the main Laetoli sites in 2016, with Chao-1 richness estimates.
Table 4.
Total abundance for each habitat type, with overall number of species and Chao-1 richness estimates.
Fig 3.
Axes 1 and 2 of the Detrended correspondence analysis (DCA) of sampling sites and mollusc species.
Singletons and species occurring on only one site were excluded. Polygons enclose all sites of the same habitat type, as classified in the field, green (forest (3 sites); purple–woodland/bushland (9 sites); yellow–open (3 sites). Eigenvalues–Axis 1: 0.748, Axis 2: 0.169.
Table 5.
ANOSIM R (top right) and P (lower left) values, based on Bray-Curtis coefficient, for pairwise comparisons between the three habitat categories.
Fig 4.
Division of the 15 sites into five groups (coloured) created by TWINSPAN.
Table 6.
Ordered two-way classification of TWINSPAN results also identifying site groupings.
Table 7.
Mean values of species richness, abundance and environmental and habitat variables in the five TWINSPAN groups. Note that some environmental data were unavailable for some sites.
Table 8.
List of all species showing strong associations with each of the five TWINSPAN groups, and those with no clear association.
Data are mean number of specimens per site.
Table 9.
Pearson correlation coefficients between local environmental variables and Chao-1 species richness and abundance (no. individuals collected).
Fig 5.
Canonical correspondence analysis (CCA) triplot of mollusc and environmental data from 15 sites at Laetoli and 5 forest sites on the Mbulu Plateau.
Table 10.
Eigenvalues and % variation accounted for by axes 1–5 of the CCA.
Table 11.
Correlation of the CCA axes with environmental variables.
Fig 6.
Relationship between CCA Axis 2 and temperature annual range (BIO7).
Table 12.
Attribution of extant species to fossil species, and their distribution in the Laetoli fossil beds (fossil data from Table 1).
Fig 7.
Correspondence analysis (Axis 2 versus Axis 1) of mollusc faunas from Laetoli recent sites and fossil beds, based on 11 species and their fossil equivalents.
Fig 8.
Relationship between CA Axis 1 and aridity index (AI_ET0).Based on 11 species and their fossil equivalents, to which Axis 1 weightings of faunas from the LLB, ULB and UNB stratigraphic units and subunits have been added.
Horizontal lines show boundaries of climate categories based on Aridity Index values [42]. The regression line is highly significant r2 = 0.732 P<0.001.
Table 13.
Estimates of Mean Annual Precipitation (MAP, mm/yr) and Aridity Index (AI_ET0 x 10000) based on fossil mollusc faunas recorded from six stratigraphic units/subunits.
Fig 9.
Overlaps with the mollusc faunas of Ngorongoro and the Mbulu Plateau.
Table 14.
Pliocene fossil taxa with their extant analogues, their potential value as habitat indicators, and their stratigraphic occurrences.
Occurrences are from [1] with the record of Pupoides coenopictus from [14]. The Upper Laetolil Beds are here treated as a single unit.