ALBERT

Description: Sedimentary ancient DNA (sedaDNA) has proven to be a useful tool for palaeoenvironmental studies, but only a handful of studies exist so far for tropical regions. In this study we used sedaDNA to study the temporal succession of Brachionus spp. rotifer mitochondrial DNA haplotypes using two sediment cores from two climatically different alkaline-saline crater lakes from the Kenyan Rift Valley. Data were retrieved from a sediment core (dating back to AD 1800) from Lake Kageinya, located in the remote, hot and hyper-arid Suguta Valley. sedaDNA was used to study the temporal succession of mitochondrial DNA haplotypes of Brachionus spp. rotifers. The results were compared to previously published data from Lake Sonachi, a well-studied lake in the humid and colder mountainous region of Kenya near the town of Naivasha, now supported by a 210Pb age chronology. Both records extend back before the onset of substantial anthropogenic impact in these regions. The results revealed that climate—rather than anthropogenic impact—was most strongly correlated with haplotype changes in both lakes. During prolonged dry periods (such as from AD 1910 to the late AD 1960s), certain haplotypes persisted. Sudden changes and the emergence of alter native haplotypes were observed when climate became more humid or during episodes of highly variable climate (before AD 1910 and from AD 1960s onwards). Progressive changes in prevailing haplotypes during periods with variable climate could reflect local adaptation and/or be the result of immigration of new haplotypes after the eradication of previous populations during extreme environmental conditions (high temperatures, UV irradiation, pH and salinity). The results indicate that, despite adverse chemical conditions, sedaDNA in tropical lake sediments is preserved for at least a few hundred years. Therefore, its analysis provides a useful complementary palaeoenvironmental proxy for palaeolimnological reconstructions and novel insights on changes in rotifer populations through time.