, Source, transport et enfouissement du carbone organique lors de l'érosion continentale : l'exemple du système Himalayen.
The burial of organic carbon (Corg) in oceanic sediments is a major sink of atmospheric CO2 and has a significant impact on the global C cycle. Himalayan erosion is dominated by physical transport and each year 1 to 2 billion tons of sediments eroded from the Himalaya are delivered to the Bengal Fan through the Ganga-Brahmaputra (G-B) fluvial system. The aim of this Ph.D. is to study the export and burial of terrestrial organic carbon generated by Himalayan erosion.
In the G-B system, suspended sediments are highly heterogenous in response to grain size and mineral sorting in the river channel. Detailed depth sampling of large rivers in Bangladesh coupled with ADCP measurement of current velocity allows average compositions of exported suspended sediment to be calculated. These compositions are significantly different from that of surface suspended sediments. Geochemical mass balance of insoluble elements indicates that the flux of bed sediment + floodplain deposition represents ~ 50 % of the suspended sediment flux. In the Ganga basin, the alkaline weathering budget is imbalanced due to an unusual proportion of weathered material in the Ganga sediments indicating that soil erosion is enhanced in the floodplain. This likely results from intense deforestation and intensive landuse in the basin.
In the river channel, the total organic carbon content (TOC) is also variable and decreases towards depth; surface suspended sediments have the highest values whereas bed sediments have very low TOC. In the whole G-B system, TOC is positively correlated to Al/Si ratio that characterises the mineral and grain size sorting. TOC linearly increases with the relative proportion of philosilicates and fine grain minerals, suggesting that Corg and mineral particles have similar response to sorting processes in the river. Sediments from Ganga, Brahmaputra and Lower Meghna define similar trends between TOC and Al/Si, and thus have similar Corg loading. Nevertheless, sediments from Ganga and Brahmaputra have very different specific area due to large amounts of smectite in the Ganga sediments. This indicates that specific area does not exert a primary control on Corg loading.
River sediments have highly variable Δ14Corg from -180 ‰ to - 815 ‰. Depth profile sediments define a mixing trend between bed sediments dominated by inherited fossil Corg and surface suspended sediments dominated by recent Corg. The proportion of fossil Corg in the suspended and bed sediments was roughly estimated to be respectively ~ 20 % and > 50 % of the TOC.
At the outflow of the Himalayan range, δ13Corg values reveal a mixing between C3 plant inputs and inherited fossil Corg. In the delta of Bangladesh, Corg transported by the Brahmaputra results from similar mixing while Corg transported by the Ganga results from the mixing between fossil Corg and a C4 plant rich end-member mainly originating from the floodplain. During the Gangetic floodplain transit, more than 50 % of recent Corg derived from the Himalaya is oxidised and is replaced by Corg derived from the floodplain. This renewal process is likely related to channel avulsion, generating re-erosion of sediments previously deposited in the plain.
In Bengal Fan sediments, TOC is variable but positively correlated to Al/Si. The trend between TOC and Al/Si defined by Bengal Fan sediments is statistically identical to that defined by river sediments. The Corg loadings of river and recent Bengal Fan sediments are therefore comparable. Biomarker abundance and δ13C show that Corg is dominated by terrestrial inputs. Consequently, the terrestrial Corg burial efficiency must be around 100 %. This strongly contrasts with other large deltaic system on earth, where ~ 70 % of terrestrial Corg is oxidised prior to burial. This extreme burial efficiency is sustained by high erosion rate in Himalaya that generates high sedimentation rate and low oxygen availability in the Bay of Bengal.
Vegetation changes in the Himalayan basin are well recorded in the sediments deposited in the Fan. δ13C values of both bulk Corg and higher plant biomarkers reveal a decrease of the C4 plants proportion in the floodplain since the LGM that is consistent with an increase of humidity due to a strengthening of the Asian monsoon. However, the terrestrial Corg burial efficiency appears to be rather constant throughout the last glacial-interglacial cycle.
In the Himalayan basin, we estimate the burial fluxes or recent and fossil Corg to be respectively 3.1±0.3 × 1011 mol/yr and 0.9±0.4 × 1011 mol/yr. Corg burial therefore account for ~ 80 % of atmospheric CO2 consumption generated by Himalayan erosion. Efficient burial of Corg is likely a characteristic of high physical erosion typical of active orogenic systems. Enhanced physical erosion and consequent Corg burial must buffer atmospheric CO2 thereby exerting a negative feedback on the long-term climate.
FILE » PhD Thesis Galy