Environmental Engineering Reference
In-Depth Information
agricultural land. After a careful analysis of the relevant error sources, he showed
that the split-window method for SST could be adopted with an accuracy of 3
C.
Since most land surface emissivities are not close to unity, Becker (1987) pointed
out that if emissivities in the two split-window channels are assumed to be 1, the
error
ΔT
in LST by the SST split-window method is significant and is of the order of
50
ð
εÞ
ε
ð
ε
1
ε
2
Þ
1
ΔT ¼
300
ε
(19.1)
ε ¼
ε
1
þ ε
2
ð
Þ
2
Becker and Li (
1990
) extended the split-window method for SST to LST and
accounted for land surface emissivity. Again, surface temperature is expressed as a
linear combination of the brightness temperature in the two split-window channels,
in a form similar to SST, but with coefficients varying with spectral emissivites.
They show that more accurate LST can be retrieved with this local split-window
method, once the surface emissivities are known with sufficient accuracy. This
approach, the so-called local split-window LST algorithm, has been widely used.
As yet, information on surface emissivities is lacking. In some algorithms (Vidal
1991
; Ulivieri et al.
1992
), the land surface emissivity correction term derived by
Becker (1987) is added to the SST split-window equation, to obtain a split-window
equation for LST.
The coefficients of the above LST split-window algorithms depend on spectral
emissivities but not on atmospheric conditions. The corrections for atmospheric
effects are limited to the use of the differential absorption of water vapor continuum
inside the atmospheric window 10.5-12.5
m. Prata (
1993
) introduced a split-
window algorithm with coefficients that not only depend on surface emissivity
but also on atmospheric transmittance. Sobrino et al. (
1994
) also developed an LST
split-window algorithm with coefficients that vary with both surface emissivity and
atmospheric transmittance. However, it is difficult to obtain information on atmo-
spheric transmittance.
Since water vapor is the major absorbing gas in the split-window channels,
Becker and Li (1995) introduced a new split-window algorithm with coefficients
depending on surface emissivity and the atmospheric water vapor content. Francois
and Ottle (
1996
) introduced another LST split-window algorithm with coefficients
being quadratic functions of the water vapor content and tabulated for different
emissivity values. Coll and Caselles (1997) developed an LST algorithm with
nonlinear brightness temperature difference term, with coefficients of emissivity
correction terms changing with atmospheric transmittance and water vapor content.
The problem with these algorithms is that the error of precipitable water itself may
become another source for LST retrieval error. Although precipitable water can be
estimated from satellites (Jedlovec 1989; Kleespies and McMillin 1990), errors are
unavoidable.
μ
Search WWH ::
Custom Search