everal
'rowth
offer
frared
lies a
refore,
ase in
clearly
eira et
nporal
fficult
nce of
areas.
mands
ctions,
not be
ick of
nporal
'tation
1greau
hke et
some
> been
lation,
larms,
s have
as the
other
vegetation indices could be better tailored to burn
area discrimination. This is the case of Pereira
(1996) who compared the performance of four
indices in several fire-affected areas. He
suggested using an alternative of the GEMI (Pinty
and Verstraete, 1992) based on the reflectance of
near and middle infrared channel (instead of red
and near infrared, as the original GEMI), to better
discriminate the burned signal. Alternative
techniques for burned land mapping are based on
regression techniques, spectral mixture analysis,
principal component analysis and multitemporal
classification (Martín et al., 1994; Siljestróm and
Moreno, 1995). They generally obtain accurate
results when high-resolution images are used. As
for the use of low-resolution sensors (such as
NOAA-AVHRR) several questions for automatic
discrimination are still unsolved: (i) selection of
suitable bands, (ii) change detection techniques,
(iii) sources of noise (clouds, cloud-shadows,
agricultural areas, etc.).
Spectral. vegetation indices have also been used to
monitor fire severity levels: Milne (1986),
Chuvieco and Congalton (1988), Jakubauskas et
al. (1990), and López and Caselles (1991).
However, severe problems were found to meet
accurate results in areas of steep slopes (effects of
shades) and those in which vegetation cover is
very sparse.
Micro-wave discrimination of burned areas has
been attempted by several authors (Kasischke et
al., 1992; Kasischke et al., 1994; Malingreau et
al, 1995). Their results are somewhat
contradictory, since in several cases the
backscatter coefficient increases after fire, while
in others, it decreases. The spatial pattern of the
signal is attributed to differences in soil moisture
content, ground layer roughness, level of canopy
damage, and vegetation regrowth.
5. DISCUSSION: GENERAL VERSUS
SPECIFIC SATELLITE MISSIONS FOR
FIRE MANAGEMENT
We have offered a general review on the current
state of research in using remotely sensed data in
forest fire applications. Most current limitations
are derived for the lack of suitable spatial,
spectral, radiometric or temporal resolution of
satellite systems currently available to meet
operational requirements of fire managers.
Almost all satellite missions are designed as
general-purpose information systems. While data
generated by those sensors can be addressed to a
wide scope of applications, they also do not fulfil
the specific requirements of most. Perhaps the
best example of operational remote sensing
systems are the meteorological satellites, because
they were designed just one a single application
(although in many cases they can also be used
satisfactorily for others) and, consequently, the
requirements of that specific application are met
and they can solve real-world problems.
This is not the case of other applications, in which
general purpose systems cannot be used
operationally. For instance, in Europe an
operational fire detection system cannot miss fires
larger than 0.5 hectares. If a satellite system is
designed for such purpose, very frequent
coverage, at proper spatial and spectral resolution,
is essential to meet that requirement. If that
objective is not fulfilled, fire managers will not
rely on satellite systems, and applications of Earth
Observation data will only be developed on the
research side. Similar ideas could be presented in
burned land mapping or fire danger estimation,
although the former is better addressed by current
systems.
6. REFERENCES
Alonso, M., Camarasa A. Chuvieco E.,
Cocero, D., Kyun, I..A., Martin M.P. and Salas F.
J., 1996. Estimating temporal dynamics of Fuel
Moisture Content of Mediterranean Species from
NOAA-AVHRR data, EARSeL Advances in
Remote Sensing, 4/4: 9-24.
Belward, A.S, 1991. Remote sensing for
vegetation monitoring on regional and global
scales. In: Remote Sensing and Geographical
Information Systems for Resource Management in
Developing countries, (A.S.Belward, and
C.R.Valenzuela, editors), Kluwer Academic
Publishers, Dordrecht, pp. 169-187.
Burgan, R.E. and Hatford, R.A., 1993.
Monitoring vegetation greenness with satellite
data, Tech. Rep. INT-297, Ogden, USDA Forest
Service, Intermountain Research Station, 13p.
Carter, G.A., 1991. Primary and Secondary
effects of water content on the spectral
reflectance of leaves. American Journal of
Botany, 78, pp. 916-24.
Chuvieco, E. and R.G. Congalton, 1988.
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 643
