By: Alemu Gonsamo Gosa
Abstract:
Leaf area index (LAI) is an important vegetation structural variable for the quantitative analysis of biophysical processes in the terrestrial ecosystem. In particular, it can be a crucial parameter related to hydrological modeling, carbon cycle and climate study at different spatial scales. The main objective of the study was to evaluate and compare performances of LAI estimation using three selected optical field instruments namely: LAI-2000 plant canopy analyzer (PCA), TRAC and hemispherical photography. These results shall subsequently be used to calibrate and validate the estimation of LAI based on imaging spectrometer data. The study involves diverse plant functional types, namely grass, shrub and forest canopies, of a river floodplain along the river Rhine in the Netherlands.
Ground-based LAI measurements were acquired from June 19 to 30, 2005 after acquiring the Airborne Hyperspectral System (AHS) image on June 19, 2005 of the Millingerwaard, a managed natural ecosystem which consists of a wide range of plant species and plant functional types. Ground measurements were collected following the VALERI sampling scheme. The Reduced Simple Ratio (RSR) was used to derive LAI from the AHS imaging spectrometer data and was calibrated based on the ground measurements. This study compares the individual LAI estimates, and the potential advantages and disadvantages of each method are discussed in relation to its use in different plant functional types and to field data collection supporting remote sensing data
calibration and validation.
The comparison of LAI from optical field instruments indicates that TRAC and LAI-2000 PCA underestimate the LAI for grass plots when compared to hemispherical photography. This demonstrates that the LAI from TRAC and LAI-2000 PCA does not encompass the contribution of the vegetation below the sensor height for short canopies. The comparison of LAI from hemispherical photography alone and the combined method of hemispherical photography and TRAC demonstrates a good agreement (R2 = 0.74), which indicates their comparable performance. Hemispherical photography proves to be the most appropriate method to estimate LAI of short canopy vegetation, and improved classification techniques in applied software (CAN_EYE) give a good discrimination possibility for the classification of foliage elements and gaps, whereas the clumping index as derived from the TRAC instrument is more reliable in determining the effect of spatial distribution of foliage elements. The clumping index from TRAC instrument can then be used in combination with hemispherical photography or LAI-2000 for a more accurate estimation of the LAI. A key benefit, however, of all of these estimation methods is that observations can be collected in a short period of time. A poor correlation of RSR and the LAI from all three methods of ground measurements were obtained in this study for all plant functional types. A possible reason for this can be found in the low dynamics of the reflectance in the wavelength bands which are used to compute the RSR.
Keywords: Leaf Area Index, LAI-2000 PCA, TRAC, Hemispherical Photography,
Imaging Spectroscopy