Land-Ocean Interactions in Arctic Coastal Waters: Ocean Colour Remote Sensing and Current Carbon Fluxes to the Arctic Ocean

Arctic rivers carry about 40 Tg of organic carbon per year into the Arctic Ocean, enough to change the colour of the surface water over entire shelf seas. Ongoing permafrost thaw mobilizes ancient organic matter in the Arctic Ocean’s watershed and, in particular, organic carbon that was previously preserved in the perennially frozen soils. Whereas the particulate fraction of organic matter is prone to settling and subsequent burial, the dissolved fraction of organic matter (DOM) can be transported over large distances and is quickly integrated and cycled within the aquatic environment. Therefore, monitoring of DOM and its carbon (DOC) in terms of fluxes, quality, transport routes and ultimate fate in the Arctic Ocean, is one of the goals of current polar research. In situ observations in the Arctic are challenging and costly and hold tremendous scientific value. Ocean Colour Remote Sensing (OCRS) is a powerful tool that can complement in situ observations by providing frequent and synoptic estimates of surface water DOM and DOC concentration via the coloured fraction of DOM (CDOM). However, use of OCRS in Arctic organic-rich waters is hampered by uncertainties and needs further evaluation and development. The goal of this thesis is to advance our knowledge of the quantity, origin, seasonal variability and fate of DOM and carbon transported from land to sea in the Arctic. Biogeochemical and bio-optical parameters of water across the fluvial and marine zones in two Arctic regions were collected. These in situ datasets include: 1) Lena River DOM measured at least bi-weekly for one full year, 2) Lena River and Laptev Sea Shelf DOM and optical parameters measured intermittently over 11 years and 3) a suite of water column optical, radiometric, and biogeochemical measurements from spring to fall in the Mackenzie River Delta and on the Beaufort Sea Shelf. These data are a unique and novel resource for testing OCRS atmospheric correction and CDOM retrieval algorithms and for improving satellite-derived DOC estimates across the fluvial-marine transition zone. Frequent monitoring of the Lena River revealed that three source water types determine the strong seasonality of fluvial DOM: 1) melt water, 2) rain water and 3) subsurface water. The improved estimation of annual Lena River DOC flux was 6.79 Tg C, most of which (84%) was transported into the Lena River by melt and rain water. Optical properties of the DOM indicated that, in spring, the Lena River dominantly transports young carbon originating from degrading vegetation from land surfaces. With rising air temperatures in summer and fall, optical properties indicated an increasing fraction of older DOM originating from deeper soil horizons and thawing permafrost deposits. Salinity and DOM were strongly correlated (r²>0.8) in both shelf regions, indicating a dominant terrigenous source of DOM and a conservative mixing of DOM-rich river water with DOM-poor water from the Arctic Ocean. Both in situ and space-borne observations of surface waters revealed a strong seasonal variability of river plume propagation and DOC distribution on both shelves. The evaluation of several OCRS algorithms with in situ data showed that the OLCI (Ocean and Land Colour Instrument) neural network swarm (ONNS) algorithm performed best for the retrieval of CDOM in the Lena – Laptev Sea region (r²=0.72, mean percentage error=58.4%), whereas the semi-analytical algorithm “gsmA” performed best in the Mackenzie – Beaufort Sea region (r²=0.52, mean percentage error=24.1%). Furthermore, the Polymer atmospheric correction algorithm resulted in better match-up correlations than either the WFR or the C2RCC atmospheric corrections. For both regions, new DOC – CDOM models, based on the in situ observations, expand the applicability of OCRS to monitor DOC in surface waters to the entire fluvial-marine transition zone and improve the accuracy of DOC retrieval. Overall, the studies of this thesis demonstrated the capability of OCRS to monitor the propagation and distribution of DOM on Arctic shelves on large spatial and temporal scales. In the future, high frequency sampling in combination with OCRS of major Arctic rivers have the potential to improve quantification of DOC export into the Arctic Ocean and reduce current uncertainties due to the lack of data. Long-term OCRS time series merged from multiple satellites can help in identifying trends of land-sea carbon fluxes and their impact on the global carbon cycle and climate in a rapidly changing Arctic.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.