Contrasting carbon cycle along tropical forest aridity gradients in W Africa and Amazonia

Abstract

Tropical forests cover large areas of equatorial Africa and play a significant role in the global carbon cycle. However, there has been a lack of in-situ measurements to understand the forests’ gross and net primary productivity (GPP and NPP) and their allocation. Here we present the first detailed field assessment of the carbon budget of multiple forest sites in Africa, by monitoring 14 one-hectare plots along an aridity gradient in Ghana. When compared with an equivalent aridity gradient in Amazonia using the same measurement protocol, the studied West African forests generally had higher GPP and NPP and lower carbon use efficiency (CUE). The West African aridity gradient consistently shows the highest NPP, CUE, GPP, and autotrophic respiration at a medium-aridity site, Bobiri. Notably, NPP and GPP of the site are the highest yet reported anywhere in the tropics using similar methods. Widely used data products (MODIS and FLUXCOM) substantially underestimate productivity when compared to in situ measurements, in Amazonia and especially in Africa. Our analysis suggests that the high productivity of the African forests is linked to their large GPP allocation to canopy and semi-deciduous characteristics , which may be a result of a seasonal climate coupled with high soil fertility.

Why models underestimate West African tropical forest productivity

“…The LUE (or photosynthetic capacity) discrepancy is strong and consistent with previous literature, since a previous study found higher Asat and Amax values in West African species than in Amazonian ones (50). Higher Asat and Amax are found in BOB and KOG (drier sites), but not ANK (the wet evergreen forest on poor soils more similar to much of lowland Amazonia) (21,25), making LUE the dominating source of GPP bias in BOB and KOG but less dominant in ANK. The high photosynthetic capacity in semi-arid forest or savanna is consistent with other field studies in West Africa (49).…”

“…Biometric GPP was measured in the field using the Global Ecosystem Monitoring protocol, which sums measurements of the components of NPP and autotrophic respiration to calculate GPP (26) and is described in detail in previous study (21). Biometric GPP was originally quantified at the plot scale.…”

“…The uncertainty in Figure 2 represents measurements error as calculated by error propagation from each GPP component, instead of spatial variation of GPP. See Supplementary method and (21) for more data and error propagation associated with the study sites.…”

“…For both TRENDY and FLUXCOM, we extracted the grid cells where the three study sites are located, as an average from 2009 to 2019. S3) (21). The grey areas denote measurement uncertainty, not standard error.…”

“…PFT noted as 'not classified probably bare soil' implies that frac of the pixel does not sum up to 1, the remaining was assumed bare soil as told by LPJ-GUESS user manual. Only a screenshot of part of the The following study sites description information is also available at (21).…”