Résumé
This thesis brings a new understanding of the geological controls on the accumulation and preservation of sedimentary phosphates through the sedimentological, sequential, and geochemical study of the Upper Cretaceous-Paleogene series of the High Atlas. The petro-sedimentary analysis of the High Atlas phosphate series revealed the presence of five different types of genetically-related phosphate. The primary "pristine" phosphate corresponds to phosphatic marls formed by the authigenesis of francolite within the outer platform zone, below the storm wave base (primary phosphogenesis zone). This lithofacies can undergo a sedimentary differentiation by the interaction of autocyclic hydrodynamic processes and an allocyclic forcing. Consequently, granular phosphate is formed by winnowing of the primary lithofacies by storm and bottom currents. Granular phosphate, composed of densely-packed peloids, has relatively high concentrations of P2O5. Turbiditic phosphate results from the transport and deposition of other types of phosphate towards the basin by gravity flows leading to normally-graded phosphate sediments. However, phosphate lags are formed by the wave-reworking and transport of the primary facies and associated granular phosphates at the inner platform. Ultimately, karst-filling phosphate is formed from previously formed phosphate that is transported by the first transgressive currents and trapped in the karst pockets as microconglomeratic phosphate. The genetic evolution of the different phosphate facies reflects a natural enrichment process of the primary phosphate. This natural enrichment process consists of the removal of the fine detrital fraction of the sediment by in-situ winnowing of the primary phosphate. This enrichment is mirrored by increasing P2O5 concentrations and a significant decrease in the concentrations of detrital phase-associated chemical elements. This process of natural enrichment of the phosphate also allows the elimination or reduction of the uranium and cadmium contents present in the sediment. The phosphates of the High Atlas present a geochemical signature typical of present-day sea water with a negative cerium anomaly that indicates the oxidizing character of the sea water where phosphogenesis took place. Sequence analysis of the phosphate series of the High Atlas revealed the links between phosphate accumulation and variations in relative sea level. The spatial distribution of the different phosphate types is controlled by tectono-eustatic allocyclic factors. The position of the phosphogenic window is controlled by variations in relative sea level. The lateral shifting of the phosphogenic window has been continuous over time. This shifting phosphogenic zone remains active as long as favorable conditions for phosphogenesis, including upwellings and phosphorus availability, are maintained.