Class 9
Vegetation Indices
Two-band vegetation indices
Three-band vegetation indices
Leaf area index
Structure of a Leaf
Red and blue light
largely absorbed
for use in
photosynthesis
Cuticle
Upper
Epidermis
Strong Infrared
reflectivity and
transmittance.
Palisade
Layer
Spongy
Tissue
Lower Epidermis
and Cuticle
Stomates and
Guard Cells
Campbell 16.3
50
40
Reflectance (%)
Visible
30
Near Infrared
Black Spruce Needle
Moss
20
rn
r
10
r
BLUE
0
400
GREEN
500
RED
600
700
800
Wavelength (nm)
900
1000
Vegetation Indices
• Quantitative measures for vegetation abundance and
vigour.
• Formed from combinations of two to several spectral
bands that are added, divided, or multiplied in a
manner to yield a single value that indicates the
amount or vigour of vegetation within a pixel.
Campbell 16.5
Leaf Area Index (LAI)
LAI is defined as the total one-sided (or one half of the total
all-sided) green leaf area per unit ground surface area.
It is an important biological parameter because: it defines the
area that interacts with solar radiation and provides the
remote sensing signal;
It is the surface responsible for carbon absorption and
exchange with the atmosphere.
Campbell 16.6
Spectral response to vegetation amount (grass)
Response of Red and NIR to LAI changes in crops
Martin and Heiman, 1986,
Photogrammetric Engineering and Remote Sensing
Reflectance
croplands, grasslands
Near Infrared
Red
LAI
Campbell 16.5
Response of Red and NIR to LAI Changes
Chen, 1996, Canadian Journal of Remote Sensing
Forest remote sensing
(Hyperspectral)
Measurements
0.5
Background reflectivity (moss)
Foliage reflectivity (black spruce needle)
Foliage transmittance (black spruce needle)
Canopy reflectance, nadir view
Reflectance
0.4
0.3
0.2
Simulation
0.1
0.0
400
500
600
700
800
900
1000
Wavelength (nm)
Chen and Leblanc, 2000
Forests
More trees-foliage means more shadows when
the density is low
Reflectance
Because transmittance in near-infrared is high
infrared shadows appear less shaded
than shadows in visible
Near Infrared
Red
LAI
Campbell 16.5
50
NDVI =
rn - rr
rn + rr
40
Reflectance (%)
Visible
30
Near Infrared
Black Spruce Needle
Moss
20
rn
r
10
r
BLUE
0
400
GREEN
500
RED
600
700
800
Wavelength (nm)
900
1000
Vegetation Indices
NDVI
Normalized Difference Vegetation Index (NDVI)
LAI
NIR  RED
NDVI 
NIR  RED
Saturation
problems
SR
Simple Ratio (SR)
LAI
NIR
SR 
RED
NIR = reflectance in near-infrared band
RED = reflectance in red band
Perpendicular Vegetation Index (PVI)
Based on
Eucledian
distance
S R  VR 2  S IR  VIR 2
Near Infrared reflectance
PVI 
X
S = soil, V = vegetation
Y
C
W
C: dry soil
B: wet soil
X: “pure” vegetation
Y: Partialy vegetated pixel
B
A
Red Reflectance
Campbell 16.9
Simple Ratio (SR)
SR1
>
SR2
>
SR3
>
SR4
Near Infrared reflectance
SR1
SR2
NIR
SR 
RED
SR3
SR4
Red Reflectance
Normalized Difference Vegetation Index (NDVI)
NDVI1
>
Near Infrared reflectance
NDVI1
>
NDVI3
>
NDVI4
NIR  RED
NDVI 
NIR  RED
NDVI2
NDVI3
NDVI4
Red Reflectance
NDVI2
Principles of SAVI
Huete, 1988, Remote Sensing of Environment
Near Infrared reflectance
Soil Adjusted Vegetation Index (SAVI)
L
SAVI1
SAVI1
SAVI2
SAVI2
>
SAVI3
>
SAVI4
NIR  RED
SAVI  (1  L)
NIR  RED  L
SAVI3
SAVI4
Red Reflectance
>
Two-band Vegetation Indices (1)
Name
Formula
NDVI
r
r
n
n
Rouse et al., 1974
 rr 
Jordan, 1969
n
r
rn
1
rr
rn
1
rr
MSR
Chen, 1996
rn  rr
rn  rr
RDVI
WDVI
 rr 
r
r
SR
Reference
a
r
r
n , soil
,
Roujean and Breon, 1995
r ar
n
r , soil
r
Clevers,1989
Two-band Vegetation Indices (2)
Name
Formula
 r  r 1  L
 r  r  L
SAVI
n
,
Huete, 1988
r
n
SAVI1
Reference
L  05
.
r
,
r  r 1  L
r  r  L 
n
L  1 212
.  NDVI WDVI
r
n
Qi et al., 1994
r
SAVI2
r  0.5 
n
r
 0.5  2 rn  rr 
2
n
1  0.25    ,r  0125
. 
1  r 
2 r  r   15
. r  0.5r

r  r  0.5
r
GEMI
r
2
n
r
n
NLI
r
r
2
n
n
2
2
n
Pinty & Verstraete, 1992
r
r

r
 rr
r
Goel & Qin, 1994
Two-band Vegetation Indices:References
Chen, J. M., (1996). Evaluation of vegetation indices and a modified simple ratio for boreal applications. Can. J. Remote
Sensing. 22:229-242
Clevers, J. G. P. W. (1989). The applications of a weighted infrared-red
vegetation index for estimating leaf area index by correcting for soil
moisture. Remote Sens. Environ. 29:25-37.
Goel, N. S., and Qin, W. (1994). Influences of canopy architecture on
relationships between various vegetation indices and LAI and FPAR: a computer
Simulation, Remote Sens. Rev. 10:309-347.
Huete, A.R. (1988). A soil adjusted vegetation index (SAVI), Remote Sens.
Environ. 25:295-309.
Huete, A. R. and Liu, H. Q., (1994). An error and sensistivity anbalysis of the atmospheric- and soul-correcting variants of the
NDVI for the MODIS-EOS. IEEE Trans. Geisci. and Remote Sens. 32:897-905.
Jordan, C.F. (1969). Derivation of leaf area index from quality of light on
the forest floor. Ecology 50:663-666.
Kaufman, Y. J., and Tanre, D. (1992). Atmospherically resistant vegetation index (ARVI) for EOS-MODIS. IEEE Trans. Geosci.
Remote Sens. 30:261-270.
Pinty, B. and Verstrate, M. M. (1992). GEMI: a non-linear index to monitor global vegetation from satellites. Vegetatio 101:1520.
Qi, J., Chehbouni, A., Huete, A.R., Kerr, Y.H. and Sorooshian, S. (1994). A
modified soil adjusted vegetation index, Remote Sens. Environ. 48:119-126.
Rouse, J. W., Hass, R. H. Shell, J. A., and Deering, D. W. (1974). Monitoring vegetation systems in the Great Plains with
ERTS-1. Third Earth Resources Technology Satellite Symposium 1: 309-317.
Roujean, J.-L. and Breon, F. M. (1995). Estimating PAR absorbed by vegetation
from bidrectional reflectance measurements. Remote Sens. Environ.
51:375-384.
Some useful features of vegetation indices (1)
1. NDVI, SR, MSR are based on the ratio of
red and NIR bands. They are often preferred
because the ratio can remove much
measurement noise in individual bands
2. SAVI, SAVI1 and SAVI2 have the advantage of
considering the influence of the soil background
Effect, but it is not based on the ratio and much of
Measurement noise is retained
3. Other more complicated indices might have
Advantages in specific applications, but they have
The potential to amplify measurement noise
Chen, 1996, Canadian Journal of Remote Sensing
Effectiveness of VIs in retrieving LAI of boreal forests
Note:
The usefulness
of VIs in other
ecosystems
may differ
Satellite-based LAI algorithm development
Canada-wide LAI map validation involving all five forest research centres
and several universities
(satellite: Landsat; ground data: TRAC)
Chen et al. 2001, Remote Sensing of Environment
60
25
50
20
40
15
SR
Reflectance in the
Near-Infrared band(%)
LAI - Agriculture
30
10
20
5
10
0
0
1
2
3
4
5
6
Reflectance in the Red band (%)
7
8
9
0
0
1
2
LAI
3
4
Three-band Vegetation Indices
Name
Formula*
r
r
Reference
 rrb
n  rrb 
Kaufman and Tanre, 1992
n
ARVI
rrb  rr   ( rb  rr )
Liu and Huete, 1995
SARVI
,
r  r 1  L 
r  r  L 
n
rb
n
SARVI2
rb
L  05
.
2.5rn  rr 
(1  rn  6 rr  7.5rb )
Huete et al., 1996
Nemani et al., 1993
MNDVI
RSR
r
 rr 
r s  r s min
(1 
)
( rn  rr )
r s max  r s min
n
rn
r s  r s min
(1 
)
rr
r s max  r s min
Brown et al., 1999
Three-band Vegetation Indices (References)
Brown, L. J., J. M. Chen, S.G. Leblanc, and J. Cihlar. 2000. “Short Wave Infrared
Correction to the Simple Ratio: An Image and Model Analysis,” Remote Sens. of
Environ, . 71:16-25
Huete, A. R., C. Justice, W. van Leeuwen. 1996. “MODIS vegetation index (MOD 13)”.
EOS MODIS Algorithm-Theoretical baiss document, NASA Goddard Space Flight
Center, Greenbelt, Maryland 20771. USA. 115pp.
Kaufman, Y. J., and Tanre, D. (1992). Atmospherically resistant vegetation index
(ARVI) for EOS-MODIS. IEEE Trans. Geosci. Remote Sens. 30:261-270.
Liu, H. Q. and A. R. Huete. 1995 “A feedback based modification of the NDVI to
minimize canopy background and atmospheric noise.” IEEE Trans. Geosci. Remote
Sens. 33:481-486.
Nemani, R., L. Pierce, S. Running, and L. Band. 1993. “Forest Ecosystem Processes at
the Watershed Scale: Sensitivity to Remotely Sensed Leaf Area Index Estimates,” Intl.
J. Remote Sens., 14, Pp. 2519-2534.
Some useful features of vegetation indices (2)
1. ARVI, SARVI, and SARVI2 are able to reduce the
the influence of the atmosphere.
2. MNDVI and RSR are designed to reduce the background
effects.
The best way is to do proper atmospheric
correction and use ratio-based indices
Reduced Simple Ratio


MIR  MIRMIN  
RSR  SR  1 

 MIRMAX  MIRMIN 
The mid-infrared scales the background effect
Brown et al, 1999
RSR
SR
LAI
LAI
a = aspen
m = mixed
s = spruce
p = pine
Brown et al, 1999, Remote Sensing of Environment