Abstract :
There is still no consensus on the impact of cloud on terrestrial carbon sequestration. Nevertheless, the fraction of sky irradiance which is diffuse (fDIF) is close to half as a global annual average, owing mainly to the presence of clouds. Furthermore, as a consequence of human-induced perturbations, the occurrence and opacity of cloud is changing both regionally (due to deforestation and drainage) and globally (shortwave “solar” dimming). In this study, we quantify the impact of cloud on carbon assimilation at an unprecedented number of FLUXNET sites (38) and for six plant functional types (PFTs). We compare results from previously established empirical and statistical methods with novel land-surface and three-dimensional (3D) radiative-transfer (RT) simulations that take explicit account of diffuse sunlight. We record a much lower enhancement in canopy light-use efficiency (LUE) under diffuse sunlight relative to direct sunlight (factor 1.12–1.80) compared to previous authors (factors 2–3). Increased radiation-sharing, due to varied leaf orientation within the canopy, is the primary cause of LUE-enhancement rather than beam penetration within an open crown structure. Under cloud, we consistently record a decrease in primary productivity (≥10–40%) and an unequivocal decline in daily carbon sequestration (60–80%), owing to the dramatic reduction in total (direct plus diffuse) irradiance that occurs when clouds obscure the solar disk (≥60% attenuation). A cooling-induced reduction in ecosystem respiration offsets the decline in primary productivity by about one third.
