Other
Scientific paper
Dec 2002
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002agufm.p72a0481m&link_type=abstract
American Geophysical Union, Fall Meeting 2002, abstract #P72A-0481
Other
1640 Remote Sensing, 1694 Instruments And Techniques
Scientific paper
Determining surface-temperature (T) and emissivity spectra (ɛ) is significant to many aspects of Earth and planetary sciences, and has motivated the design and utilization of terrestrial remote-sensing instruments. However, as the amount of radiation emitted from a surface is dependant on both its T and ɛ, their estimation from remotely sensed data remains an underdetermined problem, requiring additional assumptions in order to constrain the inherent extra degree of freedom. A simple algorithm, based on the normalized emissivity method, enables the retrieval of terrestrial surface temperatures (260-330K) and emissivities to within +/-1K, and +/-0.02 respectively, with data from the Multispectral Thermal Imager (MTI) thermal bands. Using the normalized emissivity method, the extra degree of freedom in determining T and ɛ from remotely sensed data is reduced by assigning a maximum ɛ value to each band, and assuming that the true emissivity in one of the bands is within an acceptable range from the assumed ɛ. Using Planck's function and the assumed emissivities, apparent temperatures are calculated for each band. The maximum apparent temperature is then taken as the estimated surface-temperature, which is subsequently used to estimate emissivities in the other bands. Designed as a research and demonstration satellite, MTI offers five thermal bands: two in the mid-infrared wavelengths (MIR: 3-5 μm), and three in the thermal infrared wavelengths (TIR: 8-12 μm). Each of these bands has a unique dL/dT (where L is emitted radiance, and T is temperature), which is defined by Planck's function. The MTI T/ɛ separation algorithm utilizes the fact that dL/dT in the MIR bands is significantly higher than in the TIR bands. Consequently, T estimates using the MIR bands are less sensitive to errors in emissivity assumptions, and can be used in some cases to improve T recovery. However, this also implies that MIR emissivities, in atmosphere-free numerical simulations, are recovered to within +/-0.04, as opposed to +/-0.02 ɛ recovery in the TIR bands in similar simulations. Night-time validation of the MTI T/ɛ separation algorithm at Mauna Loa Caldera in Hawaii suggests that the algorithm performs within similar error thresholds in real conditions. Day-time implementation of the algorithm requires significant correction for reflected solar irradiance in the 3.87-μm MIR band, but only minor correction in the other 4.97-μm MIR band. Emissivity recovery in the MIR range may enable improved surface-composition and particle-size estimations.
Balick L. K.
Gillespie Alan R.
Mushkin Amit
No associations
LandOfFree
Surface-Temperature and Emissivity Recovery using the Spaceborne Multispectral Thermal Imager (MTI) Sensor does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Surface-Temperature and Emissivity Recovery using the Spaceborne Multispectral Thermal Imager (MTI) Sensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Surface-Temperature and Emissivity Recovery using the Spaceborne Multispectral Thermal Imager (MTI) Sensor will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1892368