Hyperspectral Spectrometry as a Means to Differentiate Uninfested and Infested Winter Wheat by Greenbug (Hemiptera: Aphididae)

Abstract

Although spectral remote sensing techniques have been used to study many ecological variables and biotic and abiotic stresses to agricultural crops over decades, the potential use of these techniques for greenbug, Schizaphis graminum (Rondani) (Hemiptera: Aphididae) infestations and damage to wheat, Triticum aestivum L., under field conditions is unknown. Hence, this research was conducted to investigate: 1) the applicability and feasibility of using a portable narrow-banded (hyperspectral) remote sensing instrument to identify and discern differences in spectral reflection patterns (spectral signatures) of winter wheat canopies with and without greenbug damage; and 2) the relationship between miscellaneous spectral vegetation indices and greenbug density in wheat canopies growing in two fields and under greenhouse conditions. Both greenbug and reflectance data were collected from 0.25-, 0.37-, and 1-m2 plots in one of the fields, greenhouse, and the other field, respectively. Regardless of the growth conditions, greenbug-damaged wheat canopies had higher reflectance in the visible range and less in the near infrared regions of the spectrum when compared with undamaged canopies. In addition to percentage of reflectance comparison, a large number of spectral vegetation indices drawn from the literature were calculated and correlated with greenbug density. Linear regression analyses revealed high relationships (R2 ranged from 0.62 to 0.85) between greenbug density and spectral vegetation indices. These results indicate that hyperspectral remotely sensed data with an appropriate pixel size have the potential to portray greenbug density and discriminate its damage to wheat with repeated accuracy and precision.

Airborne Remote Sensing to Detect Greenbug1Stress to Wheat

“…Yang et al (2005) demonstrated on wheat grown in 0.75 x 0.75 m flats in a greenhouse that a hand-held radiometer with bands centered at 694 and 800 nm was sensitive to damage caused by greenbugs. Mirik et al (2006) used a hand-held hyper-spectral spectrometer at wavelengths from 400-900 nm and found differences in reflectance between greenbug-damaged and nondamaged wheat fields. The greatest differences in reflectance between infested and noninfested wheat occurred between 600-700 and 750-900 nm.…”

“…In a laboratory study, Riedell and Blackmer (1999) found that reflectance in the 625-635 and 680-696 nm ranges from detached wheat leaves infested with greenbugs was significantly correlated with greenbug feeding. Multispectral remote sensing studies in the greenhouse and field have been conducted at the whole plant and multiple plant levels to detect stress caused by greenbug infestation of wheat (Yang et al 2005, Mirik et al 2006). These greenhouse (Yang et al 2005) and field (Mirik et al 2006) studies used handheld radiometers and confirmed that stress caused by greenbugs on wheat plants can be detected as altered reflectance patterns of visible and near-infrared light.…”

Sensorik für einen präzisierten Pflanzenschutz

“…Der Befall von Tomaten durch Blattminierer lässt sich im Nahinfrarotbereich, insbesondere bei Wellenlängen von 1.450 und 1.900 nm, quantitativ erfassen (Xu et al 2007). Mit hyperspektralen Messungen konnte bei Weizen durch Befall mit der Blattlaus Schizaphis graminum Veränderungen in der Reflektion im sichtbaren und im NIR-Bereich nachgewiesen werden, die zur genauen Erfassung des Schadens genutzt werden sollen (Mirik et al 2006).…”

Rice plant-hopper infestation detection and classification algorithms based on fractal dimension values and fuzzy C-means

“…The remote sensing (RS) method based on spectral scanning is an alternative method for mapping the spatial distribution and severity of anomalies [6][7][8][9]. Crop plants under stress, i.e.…”

“…Many studies have shown that crop plants under stress have a significant fluctuation in reflectance of the infrared band and the chlorophyll active band, and a consequent shift of the red edge and vegetation indices [10][11][12]. As a harmless, rapid, and costeffective technology, remote sensing has been extensively used by many researchers to study insect-pest-caused stress, such as aphids on wheat [7,[12][13][14][15][16][17][18], rice [11,19,20], cotton [21][22][23], tobacco [24], corn [25], and sorghum [26]. However, an actual rice-paddy system is an extremely complex ecosystem, involving several cross-hazards: the pests themselves, diseases, and nutritional stress simultaneously.…”