Paper Review: Zhang et al., (2018). domimant_timescales_and_regions_of_global_NEE

· ☕ 2 min read · ✍️ Hoontaek Lee

24 April, 2022 (Zhang, gcb, domimant_timescales_and_regions_of_global_NEE)

Zhang, X., Wang, Y.-P., Peng, S., Rayner, P. J., Ciais, P., Silver, J. D., Piao, S., Zhu, Z., Lu, X., & Zheng, X. (2018). Dominant regions and drivers of the variability of the global land carbon sink across timescales. Global Change Biology, 24(9), 3954–3968. https://doi.org/10.1111/gcb.14275

Zhang et al. used the Fourier transformation to attribute variance of carbon cycle components to multiple time scales (from interannual to multi-decadal scales). Their study is based on mathematics, making the results very sound. A three-pools box model was formulated using the matrix representation by Prof. Luo to simplify the carbon cycle in ecosystem models of the Trendy project.

Message: Semi-arid regions and NPP dominate the global NEE variability at the interannual scale (2-10 yrs), while Tropical regions and Rh dominate at the multi-decadal scale (over 30 yrs).

At the interannual scale, semi-arid regions popped up, like many other studies support its importance. Tropical regions also contributed significantly or even as much as semi-arid regions did for an inversion product.

At the multi-decadal scale, tropical regions emerged as the strongest contributor. The large contribution of tropical regions at the multi-decadal scale was partly attributed to the effect of multi-decadal climate mode (i.e., the Pacific Decadal Oscillation, PDO).

In terms of the dominant NEE component, NPP drove the interannual NEE variation, while Rh drove the multi-decadal NEE.

Globally, the CO2 growth rate was dominated by terrestrial ecosystems, according to a result of comparing variability of CO2 growth rate, the global carbon project (an independent estimate of global land carbon sink), and a land carbon CO2 inversion product.

The increase of atmospheric CO2 concentration affected the variability of NPP and Rh at long-term time scales (>10 yr).

etc

I liked that their theory was based on mathematics.

Their method can attribute the variability of target variables to multiple time scales. Can this method be used to attribute to each timestep or pixel?

Share on

Hoontaek Lee
WRITTEN BY
Hoontaek Lee
Tree-Forest-Climate Researcher

What's on this Page