They write that observations suggest a more active CO2 cycle in high northern latitude ecosystems. Evidence points to increased uptake by boreal forests and Arctic ecosystems, as well as increasing respiration, especially in autumn. However, there is currently no strong evidence of increased CH4 emissions. Long-term observations using both bottom-up (e.g., process models or flux observations) and top-down (atmospheric inversion models) approaches are essential for understanding changing carbon cycle budgets. Consideration of atmospheric transport is critical for interpretation of top-down observations of atmospheric carbon.
You can read the paper here.