In this paper, it is shown the importance of thermal measurements to characterize different surfaces carried out in boreal environment in the SIFLEX (Solar Induced and Fluorescence Experiment) campaign. The data was acquired in Sodankyla (Finland), over the boreal forest, from 23rd April to 10th June 2002. Bio-geophysical parameters such as land surface temperature and emissivity were retrieved in relationship with other parameters from fluorescence measurements made with a Passive Multiwavelength Fluorescence Detector (PMFD).
The thermal measurements of different targets (soil, vegetation, sky) under different observation angles have been carried out using a four-band field radiometer (CIMEL CE312) and two single band radiometers (EVEREST 3000.4ZLC and RAYTEK ST6). Angular measurements and transects have been also carried out concurrently to the satellites flights over the region.
12 MODIS (Moderate Resolution Imaging Spectroradiometer) was launched on board the NASA's Terra Earth Observing System (EOS AM-1) Satellite on December 18, 1999. We propose in this work operative split-window algorithms for retrieving sea surface temperature (SST) and land surface temperature (LST) using MODIS data. In order to attain our goal, the MODTRAN 3.5 radiative transfer code was used to predict radiances for MODIS channels 31 and 32. To analyze atmospheric effects, a set of radiosoundings was used to cover the variability of surface temperature and water vapor concentration on a worldwide scale. These simulated data were split into two sets which have very similar distributions in space and time. The first one was used to fit the coefficients of the algorithms for retrieving both SST and LST, while the second one was used to test the fitted coefficients. Later, a sensibility study, including the effects of noise, emissivity and water vapor content uncertainties, has been done using the error theory.