Close Window

Data rejection and gap filling in micrometeorology measures of an Amazonian transitional forest

Camila Isabel de Menezes Fraga, UFMT, camilafraga@gmail.com (Presenting)
Luciana Sanches, UFMT, lsanches@cpd.ufmt.br
Maricéia Tatiana Vilani, UFMT, mariceia@cpd.ufmt.br
Nara Luísa Reis de Andrade, UFMT, naraluisar@gmail.com

Sensor and/or infrastructure (i.e., power) failures caused unavoidable gaps in the data collection, while short-term events such as driving rainfall and/or poor turbulent mixing lead to the rejection of data. The goal of this work was (a) to quantity of data rejection, (b) to quantity gap filling, and (c) to analyze the monthly average (±SD) between gap filled and gap not filled of micrometeorology measures of an Amazonian transitional forest. We use micrometeorology measures (air temperature, humidity, radiation and wind speed) of the tower-based installed in an Amazonian transitional forest (11º24.75’S; 55º19.50’W) on the period 2001 to 2003, concluding a total of 20 months (n=7847). Gaps in data were filled using different interpolation treatments, according to gaps minor than 3 hours; or gaps between 3 hours and 24 hours; or gaps greater than 24 hours. The monthly average of micrometeorology measures were assumed as a diurnal average calculated by averaging each 30 min measurement for a particular time (e.g., 10:00-10:30 hours). These results of analysis of gap filling for treatment interpolation in micrometeorological data suggest that including in months where data recovery was on the order of 66% the monthly average (±SD) was satisfactory when compared with monthly average (±SD) of gap not filled data. Studies of data rejection and gap filling contribute with the use of models of CO2, water vapor and energy fluxes developed to study the response of tropical ecosystems related to the environmental conditions.

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

Presentation Type:  Poster

Abstract ID: 137

Close Window