Image-Based Measurement of Leaf Area Index and Radiation Interception for Modelling of Oil Palm

Leaf area index (LAI) is an important parameter for precise characterization of the plant canopy structure. M I describes a fundamental property of the plant canopy that has often been used as a critical variable to simulate the growth and yield models. The present conventional method used in determining LA1 is laborious, difficult and time consuming. Thus an imagebased measurement using camera system with fish eye lens offers an alternative means for accurate indirect measurement of LA1 in oil palm. In this study leaf area index was determined by direct and indirect methods. The LAI-2000 plant canopy analyser (PCA), fish eye lens with charge couple device (CCD) camera and radiation sensor were used as indirect methods. Results show that the LA1 value was overestimated (30.8%-153%) for immature palm and underestimated (24%-52%) for mature palm as compared to direct measurement. The MI-2000 PCA reading varies according to the condition of sky, measuring technique, view cap, height of the measuring point and shade. Four models (leaflet shape factor model,frond area model, leaflet dry weight model and leaflet area model) were tested for accurate estimation of leaf area. Results show that the leaflet dry weight was strongly correlated (r = 0.98) with leaflet area. Light interception by a canopy is a fundamental requirement for crop growth and important for biomass production and plant growth modelling. In this study, two operational methods for estimating the amount of photosynthetically active radiation intercepted by a canopy of the oil palm were developed, i.e. "Triangular" method and "Circular" method. Results show that both methods were suitable for oil palm PAR measurement. A nonlinear relationship was found between radiation interception and LAI. Results show that the radiation interception decreased with increasing distance from the frond base to frond tip. Hemispherical photography was used in this study to estimate the leaf area index and gap fraction in oil palm plantation. Photographs were taken from different palm ages i.e. 2, 3, 6, 7, 9, 13 and 16-year old after field planting. The average LAI values obtained from photography method were 0.68 to 1.71 for 2 to 16-year old palms. The average LA1 value was underestimated as compared to destructive method. The LA1 values need to be multiplied by a conversion factor to get the accurate LA1 as obtained from the photography method. For palms less than 5-year olds, the photographic method gave the accurate MI value. The gap fractions obtained from photographic method ranged from 0.51 to 0.18 for palm age range from 2 to 16-year old palms. The gap fraction was linearly correlated (r = 0.99) with leaf area index.Computer simulation models have become powerful tools to enhance information derived from costly and laborious field experiments. Particularly in oil palms where field experiments are expensive, time consuming and labours intensive. A computer simulation model was developed using Visual C++ computer language for simulation of leaf area index and yield of oil palm. The simulated results were reasonably comparable to the field data for both LA1 and yield. The LA1 data was collected by field experiment for 2 to16- year old palms, whereas yield data was obtained from Malaysian Palm Oil Board (MPOB). A strong linear relationship was found between the measured LA1 and the simulated LA1 with correlation coefficient r of 0.96. A good linear relationship (r = 0.95) was found between the simulated LA1 and the simulated yield. Also a strong relationship (r = 0.98) was found between the simulated yield and observed yield. The proposed photographic method for LA1 estimation, different regression models, methodology for PAR measurement as well as computer program for LA1 and yield simulation have potential application in oil palm sector, oil palm R&D and also as teaching tools.