Seismic hazard represents a major worldwide scientific issue in view of potential catastrophic consequences experienced by people and facilities. Methodological developments to improve our ability to evaluate the seismic hazard are then of particular importance. We focus here on the so-called site effects that correspond to the modification of the seismic motion by the local geological configuration and that can lead to dramatic seismic amplifications. By being related to local conditions, site effects are highly variable from one site to another. That is why site-specific studies can greatly contribute to improve the hazard prediction at a specific site in comparison to ergodic estimates based on data from global databases. However site-specific studies have historically been considered difficult to carry out in low-to-moderate areas (such as France and Germany) where moderate to large earthquakes have long return periods. The DARE project proposes to exploit data from 2 complimentary dense passive experiments (100s of captors) that will be acquired on one specific km-scale sedimentary basin in the French Rhône valley. These data will be used to investigate the contribution and interest of innovative methods combining dense array processing and the use of seismic noise to offer possibilities to perform site-specific studies using relatively short temporary experiments in low-to-moderate areas. The density of instruments proposed will help to 1) improve the spatial resolution of imaging studies, allowing for a better characterization of the basin and 2) to catch the variability and multi-dimensional features of the site effects. We propose to adopt a multi-approach estimation of site effects using different seismic observations (noise & seismicity) and approaches (numerical & empirical). This strategy will allow us to propose and confront alternative methods; evaluate their own interests, uncertainties and limitations. The application of this integrated procedure in the Rhône Valley will lead to a robust estimation of site effects in an area where many critical infrastructures are located. Beyond improving our knowledge of this specific basin, the results of the project DARE will more generally contribute to improve seismic site effect estimation in terms of 1) methodological developments, 2) understanding of physical processes leading to seismic amplifications and 3) observations on km-scale western European sedimentary basins. Site-specific results obtained in this project will be confronted to estimates based on ergodic approaches that are commonly used in seismic hazard assessment studies especially in low-to-moderate areas. This will help understand the conditions of applications and limitations of the use of such ergodic approaches in seismic site effect estimations.

Africa has been recognized by the IPCC group as one of the most vulnerable region to climate change. Past abrupt climatic and environmental events that have punctuated the recent history of the African provide insights on rapid climate dynamics in the future and are of prime interest to test the ability of models to capture the transient response of the African monsoon and vegetation to both external forcing and internal feedbacks. These changeovers have furthermore crucial impact on sustainability and cultural development of African Societies. However, up to now, humans have been viewed as passive agents in these transitions. Very recently, it was controversially suggested that humans might have play an active role in regime shifts in unbalanced ecosystems. The TAPIOCA project aims at providing detailed investigation of the two last abrupt changes in Western Central Africa: the African Humid Period (AHP) and Late Holocene Rainforest Crisis (LHRC). The most prominent environmental change over the last 20 ka was the AHP, characterized by the greening of the Sahara 15 ka ago and its later desertification after 6 ka. This climatic “optimum” was triggered by a gradual increase in Northern Hemisphere summer insolation, but many issues remain unresolved about this major perturbation (e.g. timing, stepwise or abrupt transitions, spatial expression) and the role of nonlinear land surface and vegetation feedbacks. A second major vegetation change occurred in Western Central Africa approximately 3000 years ago, forest–savannah mosaic including pioneer trees replaced the mature forests. The origin of the LHRC is still a matter of debate, the anthropogenic hypothesis (deforestation due to Bantu expansion) challenging the more accepted climatic hypothesis to explain this disruption. Most of the existing continental records of these events are based on qualitative climate indicators and the lack of quantitative reconstructions of key climatic parameters such as temperature hinders comprehensive data-model comparisons. Particularly pollen-based records are potentially confounded by human activities. To fully decipher the drivers and mechanisms controlling monsoon variability, vegetation response and human-environment interactions, there is a crucial need of new records of climate proxies independent of vegetation changes as well as new tracers of human footprints. The collaborative TAPIOCA project will provide high-resolution, quantitative time series of temperature, precipitation and vegetation type based on well-calibrated organic and isotopic biomarkers. These innovative proxies will be combined with markers of human activities (e.g., fire), and with inorganic isotopic tracers of erosion, provenance and chemical weathering to investigate key lacustrine archives yielding continuous records for the last 20 kyr. This cutting-edge geochemical toolbox will be coupled with regional archeological syntheses to paint climate changes and a prehistory of expansion and technical development of African Societies with the final objectives of testing two main hypotheses: • Continental temperature changes that have modulated climate sensitivity at a regional scale have been largely minimized leading to an underestimation of the land-ocean thermal contrast during the deglaciation up to the Holocene; • Human impact on the Central Africa ecosystem has been underestimated as inducing tipping points in environmental evolution over the last 8 ka. In the frame of international collaborations, we will investigate the Barombi and Mbalang lakes. Located in Cameroon, these lacustrine archives have already shown their potential to deliver continuous and high-resolution sedimentary sequences. While the TAPIOCA consortium already retrieved a sedimentary sequence in the Barombi Mbo lake, a coring campaign of the Mbalang in planed in this project.