Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2010, Vol. 7, No. 2, pp. 67-74
Monitoring of land surface temperatures based on data from new
generation geostationary satellites
V.I. Solovjiev
, S.A. Uspensky
, A.B. Uspensky
SRC Planeta
The new methodology is presented developed for operational retrieving of land surface skin temperature (LST or Ts)
and emissivity (LSE or e) from SEVIRI/Meteosat-9 data. The clear sky IR brightness temperatures measured at
three different times (image cycles) in the SEVIRI split-window (SW) channels N9 (IR 10.8) and N10 (IR 12.0) are
converted to the LSTs and LSEs estimates using well-known local SW techniques in combination with the twotemperature
method. The LSEs are specified as channel 9 and 10 emissivities e1 and e2. The developed approach
employs the hypothesis that e1 and e2.remain constant during the time interval between the first and the last image
cycles used; moreover, accurate e1 and e2.first guesses are not required.
The performance of proposed retrieval methodology has been evaluated in the experiments with synthetic and actual
SEVIRI/ Meteosat-9 data covering most of Europe for the sample of days in time period May-November 2009. For
each date three SEVIRI imaging cycles have been used - at 11.00, 12.00, and 13.00 UTC. The cross-validation of
LST retrievals has been performed through comparison to independent (synchronous and collocated) SEVIRI-based
LST estimates produced at LSA SAF. The results show the RMS deviations in the range 1.4-2.4K. This can be
treated as an indirect proof of methods efficiency.
Keywords: land surface skin temperature, emissivity, SEVIRI instrument, split-window techniques, twotemperature method, cross-validation
Full text