Estimativa da radiação de ondas longas incidente na Amazônia: uma análise espectral.

Abstract

The downward longwave radiation (L↓) component really is difficult of the net radiation the measured. Therefore, different parameterizations have been proposed to estimate it. Thus, in this study we evaluated the performance of various parameterizations for the L↓ flows estimated and their interaction with other variables measured as well as the air temperature and water vapour pressure interactions with L↓. In this study we used by wavelet transform (WT) analysis as mathematical tool for energy of local variations analysis in the studied time series here by their frequency time-space decomposition. To investigate the correlation between two series (L↓ with air temperature and between L↓ measured with estimated), we used crosswavelet transform and coherency wavelets and phase angle. The datasets used in this study are of the January 01 to December 31 of 2009, measured from micrometeorological tower (named K34) controlled by Large Scale Biosphere Atmosphere experiment in Amazonia (LBA) from Manaus city, Amazonas. The WT showed that in 36 hours period, for the entire period analyzed, a persistent, smooth and strong energy spectrum between L↓ and the air temperature, which demonstrated a synergy between these variables. It was also found that precipitation and water vapour pressure exert enough on L↓ flows influence, especially in the wet season. The crosswavelet transform and coherency wavelets and phase angle analysis at 32 to 64 hours period. Already in adjusted models were in the 16-64 hours period. With the adjustable coefficients, according to the environmental study area conditions, the models presents an improvement in the L↓ flows estimates. The cloudy skies models proposed, they exhibited a regular performance.