Close Window

Detecting phenology and relating to turbulent fluxes over an intensive agriculture field in the Amazon

Ricardo K. Sakai, University at Albany, SUNY, sakai@asrc.cestm.albany.edu
David R. Fitzjarrald, University at Albany, SUNY, fitz@asrc.cestm.albany.edu
Otavio C. Acevedo, Universidade Federal de Santa Maria, acevedo@pesquisador.cnpq.br (Presenting)
Osvaldo M. Moraes, Universidade Federal de Santa Maria, osvaldo.moraes@pesquisador.cnpq.br
Matt Czikwosky, University at Albany, SUNY, matt@asrc.cestm.albany.edu
Troy Beldini, Escritorio do LBA em Santarem, beldini@lbasantarem.com.br

We present analysis of 5 years of micrometeorological data made in agricultural field in the eastern Amazon. We develop relationships between a remotely-sensed observations (enhanced vegetation index, EVI) and in situ measurements such as the radiometric and turbulent fluxes. There are clear differences that appear in the turbulent fluxes (CO2, H2O, and sensible heat), radiative parameters (albedo and PAR-albedo) due to the landscape changing from pasture to crop field. Since September 2000, agricultural practices in the Santarém, Pará, have changed rapidly from cattle grazing, to upland (non-irrigated) rice cultivation, and then to soybean cultivation. The pattern we witnessed in the sampled field is characteristic of the entire region along the BR-163 highway that runs south from the city of Santarém. Seasonal changes in greenness and reflectivity measured in situ follow the patterns of daytime evaporation and carbon uptake, which depend on crop type. For instance, the lowest values of the Bowen ratio were observed during the wet season during rice plantation. After plowing and tilling, the bare field emits a small efflux of CO2 comparable to nocturnal respiration rate during pasture conditions. Changes in the albedo depend on crop type, and they not only indicate changes the net radiation regime, but they also mirror changes in energy partition and CO2 fluxes. Data obtained from MODIS sensor on the Terra satellite indicate that the enhanced vegetation index (EVI) can detect seasonal changes, but it cannot clearly distinguish the crop type nor follow the rapid rate of crop phenology as well as do in situ measurements. Therefore, to estimate NEE based on remote sensing platforms, better techniques for the detection of the field state are needed. However, the in situ radiative parameter PAR-albedo, the ratio between the upward and downward PAR radiation, does not yield a unique pattern for each crop (rice or soil). The relationship between net ecosystem exchange (NEE) and PAR-albedo resulted in different relationships each year, indicating that other parameters than radiative fluxes must be included in parameterizations.

Science Theme:  LC (Land Use and Land Cover Change)

Presentation Type:  Poster

Abstract ID: 81

Close Window