Canopy hyperspectral reflectance feature of rice caused by Brown Plant-hopper (Nilaparvata lugens) infestation

Abstract

Rice Brown Plant-hopper (BPH) (Nilaparvata lugens) is one of the most serious disasters of rice production in Asia. To implement timely targeted pesticide applications, reducing input costs and benefiting the environment, an accurate early detection and quantification of damage caused by BPH infestation in rice plants is required. Currently, traditional methods, such as plant-flapping method (to investigate the population of insect-pest by macroscopic observation, the tracking down rate is between 30% and 70%), are the most common but subject to bias and can be inaccurate. Based on these imprecise and inaccurate detection and damage evaluation data may cause costly errors to variable-rate spraying in Rice Precision Agriculture. The aim of this paper is to increase the distinguish precision using a ground-based hyperspectral radiometry to detect stress in rice caused by BPH infestation. A study was conducted to investigate: 1) the feasibility of using an ASD Fieldspec 3 radiometer to identify and discern differences of rice canopies with and without BPH infestation; and 2) the sensitive bands and the useful and optimum eigenvalues for differentiating the infested rice canopies caused by BPHs from non-infested rice canopies under greenhouse conditions. Reflectance data and derived eigenvalues of the spectral curves from the radiometer were analyzed using statistical analysis procedure. Results show that it is possible to detect the stress caused by the BPH in rice using remote sensing. 1813-1836nm may be most sensitive to BPHs infestation at canopy measurement level. The sensitive bands that measured under canopy level include 550-555nm, 667-669nm, 830-924nm, 970-990nm, 1056-1100nm, 1202-1212nm, 1261-1277nm, 1788-1839nm, 1941-1945nm, and 2211-2236nm, which selected from hyperspectral spectrometry, also have the potential to detect the infestation caused by BPHs in rice. The eigenvalues, such as Crest Amplitude, Trough Amplitude, Difference between the Crest and the Trough, Ratio of the Trough to the Trough, Edge Amplitude, Peak Area of the 1st Derivation, Peak Abruptness of the 1st Derivation, which determined from the above mentioned sensitive bands, were found to be useful and optimum eigenvalues for differentiating the infested rice canopies caused by BPHs from non-infested rice canopies. However, more canopy-level-studies are needed to identify bands and indices that might have potential to differentiate the different BPH infestation levels on rice under field conditions.