Spider mites suck: detection of two spotted spider mite infestation using an ASD FieldSpec® instrument
Malvern Panalytical recently hosted an ASD webinar presentation titled, “Precision Agriculture & Ground Truthing using Spectral Signatures.” This webinar was delivered by Keynote Speaker Christopher Crockett of Highland Precision Ag, and featured a section focusing on Christopher’s research involving the detection and prediction of two-spotted spider mite (TSSM) infestation levels on strawberry leaves using the ASD FieldSpec® 4 High-Resolution Spectroradiometer.
TSSM’s are mite pests that can cause significant damage to plants – these are some of the most pervasive and ubiquitously distributed arthropod pests on the entire globe. They are found on thousands of different field and greenhouse food crop plant hosts, including over 1,300 economically important global hosts. The feeding of this pest negatively affects leaf development and subsequent fruit reproduction by inhibiting photosynthesis.
Christopher’s research explores the early identification of potentially useful spectral wavelength ranges that are significantly correlated with TSSM damage.
As TSSM infestations spread rapidly, early detection is critical to limiting the reduction of crop yield.
A publication by Herrmann et al. (2015) builds on their past work in the early detection of two-spotted spider mites (TSSM) on agricultural crops.
Past work (e.g. Herrmann 
et al. 2011 & 2012) established the link between leaf spectral characteristics and TSSM infestation. The spectral changes observed (see figure below) are similar to other vegetation stressors that damage chloroplasts: the shorter wavelength chlorophyll absorption features decrease in strength more rapidly than the dominant 680 nm feature. This results in a steady increase in the slope of the reflectance spectrum in the 580 to 680 nm range with increasing cellular damage.
| Caption: Black = not damage; Red = low damage; Green = Moderate damage; Purple = High damage; Note the increasing slope in the 580 to 680 nm range that, in other studies, has been found to be associated with chloroplast damage. From Herrmann et al. (2011) |
In his work, Herrmann et al. (2015) developed a full spectrum partial least squares discriminant function that provided 100% detection of even the earliest stages of infestation.
Herrmann, I., Berenstein, M., Paz-Kagan, T., Sade, A., Karnieli, A. (2015) Early detection of two-spotted spider mite damage to pepper leaves by spectral means in Stafford, J.V., ed., Precision agriculture ’15, Wageningen Academic Publishers, 661-666
Links to past work, referenced above, by Herrmann et al.:
Herrmann, I., Berenstein, M., Sade, A., Karnieli, A., Bonfil, D.J., Weintraub, P.G. (2011) Evaluation of spider mite damage to greenhouse pepper leaves by spectral assessment Proc. 7th EARSeL Workshop on Imaging Spectroscopy, Edinburgh, Scotland, UK.
Herrmann, I., Berenstein, M., T., Sade, A., Karnieli, A., Bonfil, D.J., Weintraub, P.G. (2012) Spectral monitoring of two-spotted spider mite damage to pepper leaves. Remote Sensing Letters, Vol. 3, No. 4, 277–283