Comparative analysis of EO-1 ALI and Hyperion data for estimate leaf area index of rubber plantation

Abstract

In this paper, the ability to estimate the rubber plantation’s leaf area index (LAI) of hyperspectral remote sensing with Hyperion satellite and multispectral remote sensing with Advance Land Image (ALI) satellite were compared. LAI was estimated by NDVI. Many mathematical models such as Linear, polynomial, Logarithm, Exponential and Power functions were used to determine the correlation between NDVI and LAI from field survey. Sixty eight LAI data from field survey was divided half for calibration and half for evaluation data sets. From many mathematical models, the power function give maximum R2 both Hyperion and ALI, Hyperion give R2 = 0.738 (RMSE = 0.089 m2m-2) and R2 = 0.624 (RMSE = 0.119 m2m-2) for calibration and evaluation datasets respectively and R2 = 0.582 (RMSE = 0.165 m2m-2) and R2 = 0.482 (RMSE = 0.280 m2m-2) for ALI. The result shows hyperspectral remote sensing is suitable to estimate LAI of rubber plantation than multispectral remote sensing