Close Window

Temporal scale of the nocturnal turbulent CO2 flux at a forested LBA site

José G Campos, UFSM, (Presenting)
Otávio C Acevedo, UFSM,
Antônio O Manzi, INPA,
Julio Tota, INPA,
Maria B L Oliveira, INPA,
Hardiney S Martins, UFSM,

In this study, we determine the nocturnal carbon dioxide fluxes from a forested site in the Amazon region, using the multiresolution decomposition technique. The commonly used eddy covariance method is successful when the atmospheric flow is characterized by fully developed turbulence. However, when stable conditions exist and, specially in strongly stable cases, the fluxes become largely dependent on many methodology parameters, such as the time period used for mean computations. Usually, this problem leads to flux underestimation. The usage of the multiresolution decomposition allows the identification of the time scales on which the turbulent transfer occurs, showing the existence of organized fluxes, highly dependent on turbulent intensity. Therefore, its application to the determination of nocturnal fluxes allows the determination of the exact time scale of the turbulent processes, distinguishing them from the more erratic, larger-scale, mesoscale exchange. The technique has only been used successfully for the nocturnal carbon dioxide flux estimate for data originated at the extremely stable pasture/agricultural LBA site, in Santarém. It remains important to quantify the improvement it will cause when used at more turbulent, forested sites. This is performed in the present study, using data from the Manaus ZF-2 tower. Results show that the turbulence has a highly intermittent character at the site. The temporal scale of the nocturnal CO2 exchange ranges from tens to hundreds of seconds, a value much smaller than those usually used in eddy covariance studies. There is also evidence of consistently positive mesoscale fluxes, in much larger scales, and this fact may explain why the increase of the averaging window usually leads to larger fluxes.

Science Theme:  CD (Carbon Dynamics)

Presentation Type:  Poster (view presentation (796 KB))

Abstract ID: 66

Close Window