Tesis
Efeitos da correção atmosférica em imagens Landsat 8 e diferentes modelos de radiação solar global na estimativa do saldo de radiação superficial
Fecha
2016-09-19Registro en:
PAVÃO, Vagner Marques. Efeitos da correção atmosférica em imagens Landsat 8 e diferentes modelos de radiação solar global na estimativa do saldo de radiação superficial. 2016. 46 f. Dissertação (Mestrado em Física Ambiental) - Universidade Federal de Mato Grosso, Instituto de Física, Cuiabá, 2016.
Autor
Biudes, Marcelo Sacardi
http://lattes.cnpq.br/7273935697798004
Biudes, Marcelo Sacardi
893.219.011-91
http://lattes.cnpq.br/7273935697798004
Machado, Nadja Gomes
713.049.401-87
http://lattes.cnpq.br/9058690473723423
893.219.011-91
Pereira, Osvaldo Alves
688.606.091-00
http://lattes.cnpq.br/4202607809882017
Institución
Resumen
The Pantanal and Mato Grosso Cerrado has experienced strong changes in
land use and occupation over their territory. These changes, when performed on a
large scale can change regional climate variables such as albedo, temperature and
radiation balance. The radiation balance (Rn) is the difference between the
radiation balance of the short-wave flows and long the Earth's surface and can be
obtained in field measurements or by estimates from satellite images, such as the
Landsat 8. However, one of the main drawbacks in this type of study is the effect
of atmospheric scattering in radiance data recorded by the sensor. In this sense,
the objective of this study was to evaluate the performance of the estimate of Rn
with Landsat 8 images (with and without atmospheric correction) from solar
radiation data (Rg) measured and estimated by different models in the Cerrado-
Pantanal interface. The data used as input, as well as validation of the models
were obtained by micrometeorological towers in Baía das Pedras (BPE), Farm
Miranda (IMF) and experimental farm (FEX). All estimates Rg models suggested
by Allen, were used (2007) and Zillman et al., (1972). In the Rn calculation were
Landsat 8 reflectance letters downloaded at the top of the atmosphere (lane 10),
subsequently corrected atmospherically and surface reflectance maps (band 2 to
7), which were estimated following parameters: albedo, vegetation indexes
(NDVI, SAVI and LAI), surface temperature, longwave radiation emitted by the
surface and end the radiation balance. In terms of local micrometeorology, all
variables have seasonal variation, with the highest rainfall in the IMF and FEX
and vapor pressure deficit equal at both sites. Already the air temperature and the
variability of the albedo tended to be lower in BPE presented the highest Rg and
Rn values. In general, the best estimates were obtained with instant Rn Rg
measured data, followed by Allen model. Already the estimates of daily Rn were
satisfactory and consistent with those seen in other studies in all experimental
sites. So in terms of modeling of the instantaneous net radiation (Rn), data-based
entries in Rg measured surface, where available, are given for best estimates of
Rn. However, there is no Rg measured using equations such as Allen and Zillman
used in this work, have satisfactory Rn estimates. However, these equations
suggest the need for calibration region.