Full text: Mesures physiques et signatures en télédétection

Department of Thermodynamics, Faculty of Physics, University of Valencia. 
50, Dr. Moliner. 46100 Burjassot, Spain. 
In this work we present a methodology for determining and mapping land surface emissivity and temperature 
using NOAA-AVHRR data. A split-window equation has been derived for atmospheric and emissivity correction, 
which requires the knowledge of the mean emissivity and its spectral variation in the AVHRR channels 4 and 5, 
e and Ae. The spatial variation of surface characteristics (soil composition, vegetation cover, etc.) makes 
necessary the definition of the effective surface emissivity at the pixel scale. Thus, methods for retrieving both e 
and Ae at the NOAA-AVHRR spatial and spectral resolution have been developed. The objective of this work is 
twofold: 1 ) to describe the theoretical models, and 2 ) to show applications in areas with different climatic and 
surface characteristics. Field measurements of emissivity were performed using the box method for the bare soil 
and vegetated surfaces existing in the study areas, and coincident radiosonde measurements have been processed 
for radiative transfer calculations. Then, we have determined the surface emissivity using the methods developed. 
Finally, the split-window algorithm has been applied to NOAA-11 AVHRR data. 
KEY WORDS: Land Surface Temperature, Emissivity, Atmospheric Correction, Split-Window 
Land surface temperature (LST) is a key parameter for understanding the energetic and hydrological balance in the 
ground-atmosphere system. Satellite measurements provide an unique tool for mapping and monitoring LST, 
which has a great interest in hydrology, agriculture, meteorology, geology. The thermal radiation measured by a 
satellite sensor is diturbed by the effect of the intervening atmosphere and the non-blackness of the surface. The 
applicability of LST retrievals to environmental sciences depends on the degree of accuracy to which these effects 
can be removed in an operational scheme. Water vapor is the main absorber in the 10.5-12.5 |im atmospheric 
window, with minor contributions of carbon dioxide, ozone and other trace gases. These gases absorb part of the 
radiation emitted by the surface, and also emit thermal radiation at a temperature generally lower than the surface 
temperature. On the other side, the surface emissivity is lower than unity, which effect is two-fold: modification 
of the radiance emitted by the surface, and reflection of downwelling atmospheric radiance. 
The most useful method for correcting for the atmospheric attenuation in thermal infrared 
images is the split-window method, which uses two measurements within the atmospheric window. This 
technique can be applied to the AVHRR channels 4 (10.3-11.3 pm) and 5 ( 1 1.4-12.4 pm). The split-window 
method has been succesfully applied for retrieving sea surface temperature with accuracies of 0.7 K (McClain, 
1989; May 1993). The sea can be considered as an homogeneous near blacbody radiator, so that emissivity 
effects are negligible. This is not the case for the land surface, which emissivity can be less than unity and 
spatially variable. Several approaches have been developed for extend the split-window technique to LST 
retrievals, showing the large impact of both the mean surface emissivity in the 10,5-12,5 pm waveband and its 
spectral variation (Price, 1984; Becker, 1987; Becker and Li, 1990; Sobrino et al„ 1991). From these previous 
works, it becomes evident the necessity of performing measurements of the spectral signatures of land surfaces. 
Besides, different approaches for deriving surface emissivitics from NOAA-AVHRR data have been developed 
(Caselles and Sobrino, 1989; Becker and Li, 1990; Vidal, 1991; Li and Becker, 1992). 
The objective of this work is to propose a simple operational procedure for determining LST 
fom NOAA-AVHRR data. The method is based on the split-window equation and requires the knowledge of the 
mean emissivity, e, and the emissivity spectral difference in the split-window channels, Ae. Thus, we have 
developed two models for determining both e and Ae at the AVHRR spectral and spatial resolution. In the next 
section we will outline the models developed. Then we will show the results obtained in two areas with different 
climatic and surface characteristics (a n agricultural area in Spain, and the Sahelian area in Niger). As a final 
result, emissivity and temperature maps will be derived lor the areas of study, showing that the model developed 
can be easily applied in different environments.

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.