stanbul 2004
as from 0.14
it the time of
) nm and the
ers across
Water
vapor
(cm)
0.71
1.15
0.51
1.23
am, p: pea
d validation
d and NIR
was higher
ed and NIR
d to generate
ly corrected
it. sites/dates
lerived from
, X-LAI
v
).58
).65
60
ind MODIS
etween LAI
ndependent)
llustrated in
nt positive
AODIS LAI
e of | and 0
v over/under
n in MODIS
^he scale of
8 and 2.49)
by LAI was
specially in
uare error of
compared to
1er range of
ODIS LAI).
roduct were
rrealistically
ar trend was
is restricted
ild arise due
, vegetation
kground or
»ver product
| algorithm,
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004
indicated error in assigning land cover class over the study sites
(broad leaf crop biome class has been considered instead of
grasses-cereal crop biome class) and that could be a source of
error in MODIS LAI product. Myneni ef al., (2002) have shown
when misclassification of land cover happens between such
classes LAI retrieval indicated overestimation of approximately
20 percentages. A preliminary analysis using all pixels
including partially cloudy pixels and LAI retrieval through
backup algorithm indicated a much higher scatter between
MODIS LAI and LISS-III estimated LAI. However, significant
correlation gives an indication of good performance of MODIS
LAI product. However, additional studies are needed before
using the product in operational use.
Table 4: Regression models to relate MODIS LAI and LAI
derived from LISS-3 data
"Equation. v2 bx Y MODIS [AT xS LISS 3 CAI
———————M——— -
Site Date a* b* R? RMSE
Bhopal 24Dec.200] 0.433 24980 061 092
(0.054) (0.073)
12 Feb. 2002 0419 1.9874. 0,615 4226
(0.089) (0.058)
Indore 02 Dec.2001 0.571 1.1643 0.52 0.20
(0.013) (0.04)
27 Dec. 2001 0.681 70.7477 051 033
(0.021) (0.026)
conne ete AP A NES ETES
* Numbers in bracket indicate RMSE
4. CONCLUSIONS
A study to compare/validate the MODIS LAI product with the
LAI images generated from IRS LISS-II1 data using regression
model between field-measured LAI and NDVI is presented. A
significant positive correlation indicates good performance of
the MODIS LAI product. However, for a few scattered pixels
of MODIS product high LAI was observed. Similar trend was
observed for other sites. The reason of difference in LAIs could
arise due to many factors such as wrong biome type, effect of
soil background, or due to aggregation. Thus different
aggregation procedures (such as fractal based aggregation
method, Milne and Cohen, 1999) have to be tested to confirm
this. Since the observations were made majorly over wheat
crop, with less LAI, observation on other crops and natural
vegetation are essential for broad scale validation of LAI
product. Detailed modeling and observation experiments using
reflectance data at two different spatial resolutions may be
necessary to identify the cause of this overestimation. It may be
pointed out that LAI is spatially very heterogeneous quantity,
and is associated with high uncertainty in field observations and
other procedures. However, additional studies, covering more
sites and vegetation types are underway before using the
product in operational use.
ACKNOWLEDGEMENTS
I am indebted to Dr. V. K. Dadhwal, Head, Crop Inventory and
Modelling Division, for his guidance in designing and carrying
out this experiment and valuable suggestions in writing the
manuscript. I am grateful to Shri J. S. Parihar, Group Director
Agricultural Resources Group for his support during the course
of work. I gratefully acknowledge the help provided by Shri R.
P. Singh and K. N. Chaudhari, Scientists-SAC, in carrying out
the field experiment and analysis. I thank Shri R. Sharma and
147
Dr. G. D. Bairagi (MPRSAC-Madhya Pradesh) for their support
during field campaign.
REFERENCES
Badhwar, G. D., R. B. MacDonald, and N. C. Mehta, 1986.
Satellite-derived LAI and vegetation maps as input to global
cycle models-a hierarchical approach. Int. JI. Remote Sensing,
7, pp. 265-281.
Bonan G., 1993. Importance of leaf area index and forest type
when estimating photosynthesis in boreal forests, Remote Sens.
Environ., 43, pp. 303-314.
Buermann, W., J. Dong, X. Zeng, , R. B. Myneni, and R. E.
Dickinson, Evaluation of the utility of satellite-based vegetation
leaf area index data for climate simulations. Journal of Climate,
2001, 14, pp. 3536-3550.
Chen, J. M., and J. Cihlar, 1996. Retrieving leaf area index of
boreal conifer forests using Landsat TM images. Remote Sens.
Environ., 55, pp. 153-162.
Gao, W. and B. M. Lesht, Model inversion of satellite-
measured reflectances to obtain surface biophysical and bi-
directional reflectance characteristics of grassland. Remote
Sens. Environ, 1997. 59, pp. 461-471.
Justice, C., A. Belward, J. Morisette, P. Lewis, J. Privette, and
F. Baret, Developments in the ‘validation’ of satellite sensor
products for the study of the land surface. International Journal
of Remote Sensing, 2000, 21, pp. 3383-3390.
Knyazikhin Y., J. V. Martonchik, R.B. Myneni, D.J. Diner, and
S.W. Running. 1998. Synergistic algorithm for estimating
vegetation canopy leaf area index and fraction of absorbed
photosynthetically active radiation from MODIS and MISR
data. J. Geophys. Res.,103, pp. 32257-32275.
Milne B. T., and W. B. Cohen, 1999. Multiscale assessment of
binary and continuous landcover variables for MODIS
validation, mapping and modelling applications. Remote Sens.
Environ, 70, pp. 82-98.
Myneni R. B., G. Asrar and S. A. W. Gerstl, (1990). Radiative
transfer in three-dimensional leaf canopies. Transport Theory
and Statistical Physics, 19, pp. 205-250.
Myneni R. B., S. Hoffman, Y. Knyazikhin, J. L. Privette, J.
Glassy, Y. Tian, Y. Wang, X. Song, Y. Zhang, G. R. Smith, A.
Lotsch, M .Friedl, J. T .Morisette, P. Votava, R. R. Nemani,
and S. W. Running, 2002. Global products of vegetation leaf
area and fraction absorbed PAR from year one of MODIS data.
Remote Sens. Environ. 83, pp. 214-231.
Myneni, R. B., R. R. Nemani and S. W. Running., Estimation
of global leaf area index and absorbed PAR using radiative
transfer model. /EEE Trans. Geoscience Remote Sensing, 1997,
35, pp. 1380-1393.
Pandya M. R., R. P. Singh, K. R. Murali, N. Babu, A. S.
Kirankumar, and V. K. Dadhwal, 2002. Band Pass Solar
Exoatmospheric Irradiance and Rayleigh Optical Thickness of
Sensors onboard Indian Remote Sensing Satellites-1B, 1C, ID
and P4. IEEE Trans. Geoscience and Remote Sensing, 40, pp.
714-718.
