×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Remote sensing for resources development and environmental management
Author
Damen, M. C. J.

TABLE 1
Landsat MSS and TM Spectral Band Vegetation Response Characteristics
MSS or TM
Band
Bandwidth
Interval
(nm)
Plant-Energy Relationships
Band
4
500-600
Spectral region of moderate leaf pigment absorption and
moderately low plant reflectance.
Band
5
600-700
Strong leaf chlorophyll absorption region and low plant
reflectance.
Band
6
700-800
Crossover region from strong chlorophyll absorption to
high leaf cell reflection.
Band
7
800-1100
Region of high leaf cell reflection and plant reflec
tance, and shows a strong sensitivity to foliage density
TM 1
450-520
Region of combined leaf chlorophyll and carotene absorp
tion and low plant reflectance.
TM 2
520-600
Spans the reflectance peak of green vegetation at 550 nm
and shows a slight sensitivity to leaf pigment concentra
tion .
TM 3
630-690
Region of maximum leaf' chlorophyll absorption and mini
mum leaf reflectance.
TM 4
760-900
Covers the high vegetation reflectance region of the
near-infrared and shows a strong correlation to changes
in vegetation density and vigor.
TM 5
TM 7
1550-1750
2080-2350
Contain the two reflectance peaks within the water ab
sorption region of the shortwave infrared. Both bands
are highly sensitive to leaf water content and related
stresses.
Many of the shortcomings of the MSS
spectral bands are ameliorated in the selec
tion of the TM bandwidths, which are narrower
and more precisely delineate the characteris
tic relationships between vegetation and so
lar energy. The increase in spatial resolu
tion from 76 m to 30 m and radiometric reso
lution from 64 to 256 quantisation levels
over the MSS sensor system further enhances
the capacity of TM data to discriminate
stress conditions in vegetation.
TABLE 1 lists the important Landsat MSS
and TM band responses as they pertain to ve
getation .
5. DESCRIPTION OF TEST SITE
A Norway spruce stand growing in copper-
lead-zinc enriched soils provided the ground
data used for establishing the relationship
between soil and needle metal content and
Landsat MSS and TM acquired canopy reflec
tance data. The spruce stand lies between
500 and 600 metres within an upland region
consisting of hills and low lying mountains
rising to 1200 m in elevation. The prevail
ing climatic conditions are those of the
humid continental zone.
Low-grade metamorphic rocks composed of
greenstone, marble, black schist, and car
bonate phyllite comprise the underlying bed
rock of the test site. Mineralisation con
sists of small but rich concentrations of
massive galena, sphalerite, and chalcopyrite.
The overlying soils are residual in origin
and exhibit a well-developed profile. Soil
depths range from 35 cm to over 100 cm.
Norway spruce (Picea abies, P. excelsa)
is the dominant tree type growing at the
test site, and occurs as a dense, mature
stand interspersed with small amounts of fir,
pine, and larch. Deciduous trees, such as
beech, ash, maple, and oak, are present as
a minor stand constituent. Forest understory
consists mainly of forbs, grasses, and small
bushes. No manifestations of stress are
readily apparent in the forest stand.
6. LANDSAT MSS AND TM SCENE DATA
Two Landsat MSS and two Landsat TM
scenes comprise the spectral data used in
the study. The MSS scenes are from mid-Sep
tember 1976 and 1981, and the TM scenes are
from early June and August 1984. All scenes
are essentially cloud free and have been
corrected for atmospheric haze using the
darkest object subtraction method of Crane
(1971).
7. GROUND DATA COLLECTION
Soil and vegetation samples form the
main ground data set used in the study. A
survey grid comprising 50 m line and station
spacings provided the required ground con
trol for the sampling. Soil samples were
collected at the base of the B-horizon and
needle samples gathered from the lower
branches of the spruce trees. Tree sampling
took place in September and consisted of ap
proximately 500 g samples of needles and
twigs containing up to five or six year old
needles.
A
collect*
analyse^
content
photome-
from 10-
60-6300
20-940 ]
vary fr<
30-340 ]
2-5 ppm
8. ANAL 1
Th<
the rel;
and spe<
dif ferei
applyin«
various
referenc
of the :
spatial
data sei
merging
the groi
correspc
TM pixel
had to 1
age soil
values c
present:
Soil anc
of the 1
merging
Landsat
the tes1
dot gric
76 metre
instante
MSS and
The
pixels c
1981 MSE
and in i
9. STAT1
Lir
lysis oi
establie
needle,
the cori