Lake/reservoir controls on evasion of inland water CO2 and implications for remote sensing of network-scale CO2 emissions

Abstract:

Lakes and reservoirs are central components of the inland water system that are inextricably connected to river networks. However, existing network-scale CO₂ emission estimates from the inland water system, although robust, typically treat rivers, lakes, and reservoirs either separately or reductively. We address this gap in knowledge by building a network-scale potential evasion model for terrestrial CO₂ that accounts for lake/reservoir morphometry for over 98,000 discrete river/lake/reservoir units in the Connecticut River watershed in the northeastern United States. Because we define and model each network unit, we can explicitly parse the influence of lakes/reservoirs on potential network-scale evasion of terrestrial CO₂ and map spatial patterns and the relative importance of rivers, lakes, and reservoirs while honoring full conservation of water mass and momentum from the headwaters to the ocean. We combine this approach with classic hydrological streamflow routing to discern relative lake/reservoir influence across the entire streamflow regime. We find that lakes/reservoirs exhibit a significant influence on potential network evasion of terrestrial CO₂: they are responsible for 25-30% of potential evasion, as inversely related to streamflow, despite there being approximately 5.8 times as many rivers as lakes/reservoirs at mean annual flow. We also identify a distinct ‘lake/reservoir threshold’ for evasion wherein lakes/reservoirs act as ‘stop signs’ in the CO₂ transport network and prevent dissolved CO₂ from further advection downstream. We conclude by taking this fully coupled network and exploring the possibility of remote sensing of CO₂ emissions from it: NASA’s Surface Water and Ocean Topography (SWOT) satellite launches in 2021 and will provide measurements of river height, river width, and lake surface area at unprecedented temporal resolutions for the largest rivers/lakes/reservoirs in the system. SWOT data and river hydraulics will likely be useful in estimating near-daily CO₂ fluxes for entire river/lake/reservoir networks.