Erosion is a group of processes, including chemical and mechanical weathering, by which material is worn away from the Earth's surface. Weathering of silicate rocks acts as an important sink for atmospheric CO2, and hence is thought to have played an important role in regulating the Earth’s climate over geological time. Despite the potential importance of this process, and its significance to the global carbon cycle, our ability to reconstruct past variations in silicate weathering remains limited. Another important issue is the degree to which human-induced environmental degradation has affected chemical weathering intensity on shorter time scales, from tens to thousands of years. There are clear evidence that human activities, such as land-use, agriculture, anthropogenic CO2 emission, are significantly changing the weathering rates of continental rocks, leading to increasing alkalinity export from large river systems to the ocean. A better understanding of how chemical weathering relates to climate change is necessary to further assess how sensitive our environment is to human activities. During the last decades, there has been a growing effort for unraveling the factors behind continental erosion, and for quantifying present and historical rates of weathering through experimental work, modelling, or direct investigation of river systems and soils. This was reflected in the increasing number of sessions dedicated to weathering issues at international conferences, and in the creation of large research programmes focused on Earth’s surface processes (e.g., the INSU programme ‘Reliefs de la Terre’ in France). Although this global effort led to major improvements in understanding the weathering mechanisms, and the links between erosion, tectonics, climate change, and ocean chemistry, there is still a lack of proxies for investigating silicate weathering signals from geological records. Here, we propose to explore the use of novel isotopic and molecular tracers in marine sediments as proxies for past continental weathering. Recent studies showed that the combined use of hafnium and neodymium isotopes in sediments preserved in the geological record could provide an unique tool for tracing the evolution of continental weathering through time. In addition, the emergence of molecular (biomarkers) tracers, in response to improving mass spectrometric techniques, has opened new fields of investigation in paleoclimatology in recent years, some of which (e.g., BIT-index) are particularly well-suited for the study of continental weathering through time. The complementary use of those novel isotopic and organic tracers can now provide a powerful tool to trace the evolution of continental weathering through time. In the first part of the project, we intend to further explore the use of Hf-Nd isotopes and biomarkers as paleoclimatic proxies by analysing an important set of sediments deposited worldwide on continental margins, near the mouth of rivers draining basins characterized by various geological, climatic, and land-use contexts. In the second part of the project, we intend to provide new insights into the global link between climate change and erosion, by applying our novel proxies to the study of two exceptional marine sediment records: 1) a 125 m-long drilled record from the Est-Corsican margin, providing a continuous record of river discharge for the last 500,000 years; and 2) a core recovered from the Gulf of Guinea, allowing the reconstruction of past silicate erosion in a continent-scale tropical watershed, the Congo Basin. The data acquired during the course of this project should significantly improve our understanding of the natural links between erosion of the continents and climate.