International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
into apparent radiance by the following formula (Markham &
Barker, 1986):
RAD (4) = (Lmax À — Lmin A) 255" - DN + Lmin X
where RAD (1) 7 spectral radiance (W * m? e ster! * um
h Lmin A = spectral radiance range minimum (high gain)
Lmax À = spectral radiance range maximum (high
gain) DN = digital number of the considered pixel
Hereupon, the apparent radiance values were converted into
reflectance by the following formula (Moreira, 2001):
p=m-RADA-d?- Esol A - cos 0s”
where p = top of (planetary) reflectance, RADA = spectral
radiance at the sensor’s aperture (W : m? ster! : um!)
d = earth-sun distance, in astronomical units:
1,0109 for the 10" of may (NASA, 2003)
EsolA = mean solar irradiance (W - m” - um”')
(NASA, 2003), 0s 7 solar zenith angle (50,6^)
2.5.4 Comparative Statistical Analysis A linear regression
analysis was performed to determine the correlation degree
between SOC/LAI and SOC/spectral reflectance of the
pastures. Correlation degrees with 95% or higher were
considered significant and were plotted in this article.
3. RESULTS AND DISCUSSION
3.1 LAI of pastures
Figure 1. Leaf Area Index of the four pastures
The four pastures show differences in LAI, ranging from
0,150 to 1,103m*m?, which refers to 635% higher LAI for
the pasture with the highest LAI compared to the lowest. The
overall low LAI levels of the four pastures are associated to
their sandy soil texture. The LAI differences between the four
pastures are mainly the result of different management
practices as the bio-physical factors, which influence the
LAI, are very similar among all pastures (same soil type,
forage, climate, topography). The management practices
differ in manuring and overgrazing. The two pastures with
lower LAI have no manuring and the pasture with the lowest
LAI also suffers overgrazing. The high standard deviation of
pasture Bondade is related to a not entirely plant cover,
showing partly bare soil spots.
60
55
50
SOC(Mg:.h) 4 s ^
40
35
30
3.2 SOC of pastures
C: prepare tete EA
20
Descalvado — Bondade — Barreiro - Monjelada
Figure 2. Soil carbon stock (SOC) of soil depth 0 -50cm
Observation: The individual soil layers (0-5, 5-10, 10-20, 20-
30, 30-40 and 40-50 were treated as one soil layer 0-50cm).
The four pastures show differences in SOC, ranging from
32,0 to 54,41Mg ha”', which refers to 70,1% more SOC in
the pasture with highest SOC compared to the lowest. The
overall low SOC stocks of the four pastures depend
principally on their sandy soil texture. The differences in
their SOC reflect different management practices, as seen for
LAI differences. Both, SOC and LAI, have almost the same
determining parameters (climate, soil type, native vegetation,
forage specie, topography) and differ only in relation to
native or former land use, which is only sensitive to SOC.
The following paragraph explains this relation.
3.3 Correlation between SOC and LAI
SOC and the LAI showed the expected positive correlation
between each other. Both tend to have an almost linear
behavior. As the [AF varies seasonally, it is of future interest
to obtain the IAF from other seasons and evaluate its
correlation over the year in terms of SOC (which varies
seasonally very little), to define the most suited “season” for
SOC/LAI correlation determination.
E
0 03 06 09 12
LAI (m? : m?)
Figure 3. Correlation between SOC (Soil Organic Carbon)
and LAI (Leaf Area Index)
Observation: The high standard deviation of Bondade refers
to partly uncovered plant cover at this site
Intern
3.4 Re
—
LAI =
m Co
where
The €
correla
3.5 Pa:
3.51€
CASS
Band
Blue
Green
Red
NIR
SWIR |
SWIR II
Te
Observ.
3.82 ]
reflecta
linear 1
bands.
meme
Reflec
Blue
Green
Red
NIR
SWIR
_SWIR |
ce
where
Two co
0.95, inc
at least ‘
and SW,
