Concept &
Institutions &
Virtual tour
Events &
english | deutsch

Project C1: Paleoclimate reconstruction for the East African Rift from geochemical studies of mammalian teeth

Figure 1: Variations in 87Sr/86Sr ratios (A) and trace element concentrations (B) along a profile from the outside enamel rim towards the dentin in a 2.3 Ma old molar tooth of a hippopotamid from Lake Albert, Uganda. Element concentration minimum ranges defining plateaus are <10 ppb for REE, Y and U, 200-800 ppm for Sr, 100-500 ppm for Ba, 0.01-0.2 ppb for Pb, 0.05-3.0 ppm for Cu, and 15-50 ppm for Zn (Legend in figure: Nu Plasma = laser-ablation multi-collector inductively coupled plasma mass spectrometer, TIMS = thermal ionisation mass spectrometer).


Figure 2: Evolution of average δ13C (A), δ18O (B) and 87Sr/86Sr ratios (C) measured on Hippopotamid tooth enamel from northern Albertine Rift specimens, for the period from ca. 7 Ma to recent.

New proxy data (δ18O, δ13C, 87Sr/86Sr, trace element concentrations) measured on tooth enamel of Hippopotamids (Mammalia) are used to evaluate long-term climate and watershed dynamics controlling the evolution of paleolake systems in the western branch of the East African Rift System (Lake Albert, Uganda) since ca. 7 Ma (Late Neogene).

Profiles of 87Sr/86Sr ratios (Figure 1A) and trace element concentrations (Figure 1B) in enamel measured from the outside rim towards the dentin show characteristic patterns. Concentrations develop asymmetrically and continuously decrease by up to 5 orders of magnitude within a distance of about 1 mm from the rim until a minimum is reached which continuous by forming a plateau. Ca. 100 µm before the enamel-dentin junction, concentrations steeply rise. The plateau represents the least altered part of the enamel and appears to have preserved the primary geochemical signal and hence the information on environment and diet. However, the width of the altered rim defined by decreasing concentrations is specific for varying elements which also form different trajectories from the cingulum to the apex of a single tooth. The primary Sr isotope composition in the outer enamel rim has also been extensively altered during diagenesis. This is shown for example by 87Sr/86Sr ratios decreasing from 0.7181 at the outer rim to a plateau-type range of ca. 0.7155 (Hippo tooth 5306 in Figure 1A) considered to represent non-altered enamel.

Our systematic approach indicates that spacially high-resoluted trace element concentration and isotope ratio distributions in each tooth are necessary in order to distinguish diagenetic from primary signals. Although much analytical effort is required, this approach is rewarded by revealing the ambient environmental conditions during both, tooth growth and subsequent diagenesis. Also, our results compromise interpretations regarding migration or provenance on the basis of bulk sample analyses.

Carbon isotope evolution in Hippopotamid enamel (Figure 2A) through geological times document diets entirely comprising C3 biomass during 7 to 4 Ma and a maximum of C4 biomass in diets from 2.3 to 1.3 Ma. Mixed diet composed of C3 and C4 vegetation is documented for the recent geological past and compatible with moist climates during the early Holocene. Oxygen isotope data reflect a trend of hydrological isolation and evaporation of lakes within the Albertine Rift from ca. 7 Ma to a maximum at ca. 2.3 to 1.5 Ma (regression in Figure 2B corresponds to samples within this time span and do not include the two younger samples). Long-term variations in isotope compositions of Hippopotamid enamel indicate that maxima in δ18O and δ13C are correlated with high-radiogenic 87Sr/86Sr up to 0.717 (Figure 2C) at 2.5 to 1.5 Ma. This consistency is interpreted in terms of rift shoulder uplift and subsidence of the rift valley floor inducing maximum dryness and maximum hydrological isolation from rift-external river catchment areas. Structural rearrangements starting at ca. 2.5 Ma within the northern segment of the Albertine Rift are well constrained by reversals in river flow, cannibalisation of catchments and biogeographic turnover.

Our data represent integrate climate signals which are influenced by regional East African as well as local rift valley geodynamic processes. It appears that during tectonic reorganisation of rift systems, paleoclimate archives rather include information on rift dynamics than of global climate change.