Abstract

Before implementing any further precision agriculture applications, the crucial task is to determine the rubber (Hevea brasiliensis) clones. This paper addresses the potential of using canopy spectral indices (CSIs) to delineate mature rubber clones. Each rubber clone responds differently to its local environment. Unfortunately, checking a sizeable mature rubber plantation and determining each rubber tree’s clone is tedious and time-consuming. It is, therefore, of interest and objective to classify and predict rubber clones via CSIs. This study highlighted statistically significant differences in the physical and spectral functional traits’ variability among nine mature rubber clones in a micro-agroclimatic environment with uniform management practices. We analysed the trunk girth, leaf physical properties and multispectral remote sensing data collected by a low-cost unmanned aerial vehicle (UAV) to explore the mature rubber canopy’s spectral indices, which are a crucial unique trait of rubber clones. We found that rubber leaf area increases with leaf mass. There is a positive relationship between trunk girth and leaf mass per area (LMA). Furthermore, leaf density and NDVI have a strong relationship that potentially links both traits. The additional assessment indicates that none of the 89 CSIs independently could discriminate between all nine rubber clones. However, several CSIs could distinguish several individual rubber clones or two or more groups of rubber clones. The classification results selected B1, NDVI, NDRE, RB3B2B4, GARI and BWDRVI as the most important CSIs. The generated models estimated that individual rubber clone prediction by CSIs provided low to moderate overall accuracy between 41.8 ± 5.3% and 82.3 ± 3.9%. Despite the variability, our findings highlighted a promising approach using CSI traits as a practical way to achieve accurate rubber clone predictions because of their uniqueness for specific clones or groups of clones.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call