A New Metric for Evaluating Northern Hemisphere Growing Season Net Flux in Climate Models

Understanding the function of terrestrial ecosystems and their response to anthropogenic climate change requires quantification of the exchange of carbon between the atmosphere and the land surface. However, large-scale land-atmosphere fluxes, including the Northern Hemisphere growing season net flux (GSNF), are difficult to quantify because ecosystem heterogeneity requires a higher density of direct flux observations than can be made, and indirect inference from atmospheric CO₂ observations via inverse modeling is susceptible to uncertainty in atmospheric transport. Here, we develop a data-driven metric for the Northern Hemisphere GSNF from observations of atmospheric CO2 concentrations collected during the HIAPER Pole-to-Pole Observations (HIPPO) and Atmospheric Tomography (ATom) flight campaigns. These campaigns sampled the atmosphere over the remote Pacific from ~80°N to ~60°S, attaining vertical profiles from the marine boundary layer to an altitude of ~10 km. We use the data from the flight campaigns, which span all seasons over a time period of nine years, to fit a seasonal cycle of atmospheric CO₂ mass in the Northern extratropics (20° - 90°). Then we infer the growing season net flux by taking the derivative of the observed seasonal cycle after using the CarbonTracker system to account for mixing at the southern and upper boundaries of our domain. We use our observationally derived GSNF to evaluate carbon cycle simulations from the Coupled Model Intercomparison Project (CMIP) phase 6 Earth system models. While the model-to-model spread in GSNF has decreased relative to that of the (CMIP) phase 5 models, there is still disagreement on the magnitude and timing of seasonal carbon uptake. Our research shows that the flux inferred from these aircraft observations provides an additional constraint on these models that may ultimately improve our ability to accurately predict carbon-climate feedbacks.