LEAF AREA INDEX RETRIEVAL USING IRS LISS-III SENSOR DATA AND
VALIDATION OF MODIS LAI PRODUCT OVER INDIA
M. R. Pandya
Remote Sensing Applications and Image Processing Area
Space Applications Centre, ISRO
Ahmedabad — 380 015, India
m r pandya(gyahoo.com
Commission VII, Working Group VII/2
KEYWORDS: Remote Sensing, Crop, IRS, Retrieval, Comparison, Leaf Area Index (LAI), Normalized Difference Vegetation
Index (NDVI), Aerosol Optical Thickness, Moderate Resolution Imaging Spectroradiometer (MODIS).
ABSTRACT:
This paper reports results of an experiment LRVE (Leaf Area Index Retrieval and Validation Experiment) that was conducted over
agricultural areas of Central India during winter season of 2001-02, aimed at relating field measurements of LAI to space borne IRS
LISS-III data, preparation of site-level LAI maps and validation of MODIS-based 1 km LAI global ficlds,. Measurements of field-
level LAI, aerosol optical thickness and water vapor were carried out on the day of LISS-III overpass. Empirical models based on
site-specific LAl-vegetation index relation were developed and used to generate 23-meter resolution LAI maps for two sites (Indore
and Bhopal) covering 30 km X 30 km. These LAI images were aggregated to Ikm spatial resolution and used for validation of
MODIS LAI product (MOD15A1). The results indicated significant positive correlation between LAI derived from LISS-III data and
MODIS data, with an overestimation in the MODIS product, with RMSE of 0.92 to 1.26 for Bhopal site and 0.20 to 0.33 for Indore
site. Analysis of MODIS land cover product that forms an input in MODIS LAI retrieval algorithm, indicated error in assigning land
cover class over the study sites and that could be a source of error in MODIS LAI product.
1. INTRODUCTION
The Leaf area index (LAI) is one of the most important
parameters characterizing a canopy. LAI is a dimensionless
index used to quantify the single sided vegetation leaf area per
unit of ground area in broadleaf canopies (or projected needle
leaf area in coniferous canopies). It is one of the surface
parameters that plays key role in climate, weather and
ecological studies. LAI is a biophysical variable influencing
vegetation photosynthesis, transpiration and the energy balance
of canopies (Bonan, 1993). LAI and fraction of absorbed
photosynthetically active radiation (0.4 — 0.7 mm) (fAPAR) are
important surface attribute controlling water, carbon and energy
exchanges between vegetation and the atmosphere (Running ef
al, 1996). LAI is not only an important driver of most
ecosystem productivity models operating at landscape to global
scales (Turner ef al., 1999), but also an interaction component
of general circulation models (Buermann ef al., 2001).
Amongst the many surface biogeochemical parameters, which
can be derived from satellite spectral measurements, LAI is a
vegetation structural parameter of fundamental importance for
quantitative analysis of many physical and biological processes
related to vegetation dynamics, global carbon cycle and climate.
Estimation of LAI at frequent intervals can facilitate estimates
of mass and energy exchange over a wide range of spatial scales
and with considerable temporal resolution. Ground
measurements of LAI however are cumbersome, time
consuming and impossible to obtain at global scale, while
satellite remote sensing is the most effective means of
estimating LAI global fields on a regular basis.
As part of the US Earth Observing System (EOS), the Terra
(launched in December 1999) and Aqua (launched in May
2002) satellites, carry the Moderate resolution Imaging
Spectroradiometer (MODIS) along with a host of other
144
advanced sensors. Algorithms have been developed to generate
a number of land products from MODIS, including LAUFPAR
and that have been made available through from EROS Data
Center Data Archive Center (EDC-DAAC) for
evaluation/validation and utilization. The MOD15 LAI and
FPAR are 1km products provided on a daily and 8-day basis.
The validation of LAI global fields, i.e., assessment of
uncertainty of remote sensing derived products by analytical
comparison to reference data which are presumed to represent
the target values (Justice et al., 2000), is necessary and has been
carried out for many sites over USA, Africa (Privette ef al.,
2002) and elsewhere, but no results are available over India. A
LAI Retrieval and Validation Experiment (LRVE) aiming at
development of remote sensing based site-specific vegetation
index-LAI relations and validation of MODIS LAI product was
conducted at Indore and Bhopal during the wheat-growing
season of 2001-02. The experiment had three components, (1)
the field measurements of LAI and atmospheric properties
(aerosol optical depth and water vapor); (2) generation of fine
resolution (23m) LAI map from IRS LISS-III and field data;
and (3) the generation of 1-km LAI maps and their comparison
with MODIS LAI product.
2. MATERIALS AND METHODS
2.1 Ground sites description
Two sites in the state of Madhya Pradesh of India, Indore and
Bhopal were selected for LAI measurements. These sites repre-
sent semi arid and semi humid zone respectively. Both the sites
have black cotton soil. The sites had wheat as a major crop with
small proportion of other crops like gram and pea. Indore has
rainfed agriculture and Bhopal has irrigated agricultural prac-
tice. The study area and the remote sensing data used in the
LRVE are shown in table 1.
Internat
MEC IEEE
Table
Descı
Site lc
Lon;
Date (
IDLI
acqui
IRS
Path
Da
MOD
8-day
MOD
Nu
^
* LISS-
# Overce
2.2 Sat
2.3 IRS
The data
onboard
the study
red, neat
(SWIR)
2.4 The
The MC
resolutio
period, v
with tl
photosyr
(MODI:
products
grid, wh
axis, a
approxir
À brief s
(2002).
radiative
table (L
dimensk
bands ar
prior to |
The algc
six majo
Crops, sé
up table
mode] f
During
modeled
structure
natural «
and the
solution
also app
the MO
provide
