Balancing Atmospheric Top-Down Carbon Estimates with High-Resolution Bottom-Up Scaling of Fluxes from Heterogeneous Tundra, Wetlands, and Waterbodies

As the Arctic is warming at an accelerated rate, recent research has shown that the Arctic is switching from a net sink to a net source of carbon to the atmosphere in some locations. The Arctic carbon balance is a critical component of the global carbon cycle, yet it remains highly uncertain with mismatches between carbon budgets derived from atmospheric top-down studies and bottom-up scaling studies. Tundra ecosystems are heterogeneous at multiple scales, with numerous small waterbodies, and variations in plant functional types, soil moisture, and microtopography, for example, influencing carbon dioxide (CO2) and methane (CH4) emissions to the atmosphere. We can reduce uncertainty in the Arctic carbon budget by using high-resolution scaling where waterbodies, wetlands, and terrestrial landscapes are not double-counted and fluxes can be attributed meaningfully to atmospheric top-down comparisons. In this study we compare top-down and bottom-up carbon estimates for the Yukon-Kuskokwim (YK) Delta, Alaska. We create bottom-up carbon fluxes using a high resolution (10 m) landcover map to identify heterogeneous landcovers, small waterbodies, and wetlands. We use machine learning in an integrated terrestrial-aquatic observation-based method to scale small waterbody carbon emissions. We use un-mixed eddy covariance fluxes to scale landcover-specific fluxes from wetlands, tundra vegetation, and areas of degrading permafrost. We convolve landcover fluxes with WRF-STILT footprints from NASA CARVE (2012-2015) and NASA ABoVE Arctic-CAP (2017) airborne campaigns to calculate fractional contributions to atmospheric enhancements of CO2 and CH4. These are then compared to observed enhancements of CO2 and CH4. We discuss how small-scale landcover heterogeneity and small waterbodies affect carbon budgets for the YK Delta. We investigate how spatial resolution affects landcover C attributions by rescaling using 30 m and 1 km resolution landcover maps with the same flux models. These results demonstrate the importance of landscape heterogeneity and combining atmospheric and ecosystem perspectives when scaling C emissions in the arctic.