REMOTE SENSING ESTIMATION OF FOREST LAI IN CLOSE CANOPY 
SITUATION 
K. S. Lee, Y. I. Park, S. H. Kim, J. H. Park, C. S. Woo, and K. C. Jang 
Inha University, Department of Geoinformatic Engineering 
253 Yonghyung-dong, Nam-gu, Incheon 401-751, S. KOREA 
ksung@inha.ac.kr 
KEY WORDS: LAI, forest, spectral reflectance, spectro-radiometer, close canopy, ETM+ 
ABSTRACT: 
This study attempts to find a new and better approach to estimate forest LAI in fully closed canopy condition. Although there 
have been many previous studies to estimate LAI using optical remote sensor data, there are not enough evidences whether the 
red and near-IR reflectance are still effective to estimate forest LAI in closed canopy situation. In this study, we have conducted a 
simple correlation analysis between LAI and spectral reflectance at two different settings: 1) spectral measurements on the 
multiple-layers of leaf samples and 2) Landsat ETM+ reflectance with field-measured LAI on the close canopy forest stands. In 
both cases, the correlation coefficients between LAI and spectral reflectance were higher in short-wave infrared (SWIR) and 
visible wavelength regions. Although the near-IR reflectance showed positive correlations with LAI the correlations strength is 
weaker than in SWIR and visible region. The higher correlations were found with the spectral reflectance data measured on the 
simulated vegetation samples than with the ETM+ reflectance on the actual forests. In addition, there was no significant 
correlation between the forest LAI and NDVI, in particular when the LAI values were over three and full canopy situation. The 
SWIR reflectance may be important factor to improve the potential of optical remote sensor data to estimate forest LAI in close 
canopy situation. 
INTRODUCTION 
Forest leaf area index (LAI) has been one of important 
structural variables to understand the process of forest 
ecosystems and can be used to measure the activities and the 
production of plant ecosystem (Pierce and Running, 1988; 
Bonan, 1993). The measurement of LAI on the ground is very 
difficult and requires a great amount of time and efforts (Gower 
et al., 1999). This is particularly true in forest where the canopy 
structure is much more complex than the grasslands and 
agriculture systems. Since plant canopy is composed of leaves, 
which is a direct source of the energy-matter interactions that 
are observed by earth-observing remote sensing systems, LAI 
has been an attractive variable of interest in vegetative remote 
sensing. 
There have been many attempts to estimate LAI using various 
types of remote sensor data since the early stage of space 
remote sensing (Badwhar et al., 1986; Peterson et al., 1987; 
Turner et al., 1999). Remote sensing estimation of LAI has 
been primarily based on the empirical relationship between the 
field-measured LAI and sensor observed spectral responses 
(Curran et al., 1992; Peddle et al., 1999). As a single value to 
represent the remotely sensed spectral responses of green 
leaves, spectral vegetation indices, such as normalized 
difference vegetation index (NDVI) or simple ratio, are 
frequently used to indirectly estimate LAL 
Normalized difference vegetation index (NDVI) has been a 
popular index with which to estimate LAI across diverse 
  
' Corresponding author 
400 
ecosystems. However, large portion of such studies to estimate 
LAI using NDVI were dealing with semi-arid vegetation and 
agricultural systems where the canopy closure is less than 
100%. Recent studies have shown that the NDVI many not be 
very sensitive to values of LAI in particular at the forest 
ecosystem having the close canopy condition that the LAI 
/alue is relatively high (Chen and Cihlar 1996, Turner et al. 
1999) 
The objectives of this study are to analyse the relationship 
between spectral reflectance and LAI in fully canopy condition 
and to find a methodology to estimate LAI in forest where the 
canopy closure is closed to 100% and LAI values are high. 
Although there were several studies dealing with the remote 
sensing estimation of LAI in forest, the study sites were 
generally not close canopy situation (Turner et al, 1999; 
Lefsky et al, 1999). The forest vegetation has very dense 
canopy closure in Korea as well as many other temperate and 
tropical forests around the world. Considering the 
environmental value of these forest ecosystems, more effective 
and accurate method to estimate forest LAI would be very 
beneficial. 
METHODS 
Spectral Measurements on Simulated LAI Samples 
Before attempting to analyze the actual satellite imagery along 
with field-measured LAI data, we decided to analyse the 
Interne 
relatio 
canop 
condu 
sample 
spectre 
Veget: 
stackir 
japoni 
leaves 
Since 
of lea 
sample 
an int 
covere 
and the 
Figure 
Reflecta 
radiome 
reflectar 
2,500nn 
with a 1 
spectro- 
trigonon 
diameter 
At each 
percent 
the way 
statistica 
correlati. 
vegetatic 
a differe 
discrepai 
provide 1 
We alsc 
differenc 
measure 
bands wi 
several 
However 
sensitive 
red and