Detecting symptoms of Phytophthora cinnamomi infection in Australian native vegetation using reflectance spectrometry: complex effects of water stress and species susceptibility

Diseases in natural and agricultural systems have been linked to species of the Oomycete genus Phytophthora, around the world. Direct detection of the pathogen requires sampling of soil or plant material, which can be expensive, difficult to obtain and error-prone. As an alternative, reflectance spectroscopy provides a potential indirect method for detecting symptoms of infection by P. cinnamomi. Here we evaluate the use of reflectance spectroscopy to detect physiological changes associated with infection in host plants using spectral indices designed to quantify changes in plant pigments (pigment indices), leaf water content (water indices) and fluorescence (fluorescence indices). Two grasses and two tree species with different susceptibilities to P. cinnamomi were inoculated and/or exposed to water stress in a glasshouse experiment. Inoculated plants were detected using pigment and fluorescence indices, which also had the capacity to separate inoculated plants from water stressed uninoculated plants. While inoculation may have caused an opposing spectral response to water stress in some indices, plants that were both water stressed and inoculated then demonstrated an intermediate response. Water stress was detected using the water indices in all four species, and spectroscopic changes associated with inoculation were often greater in the susceptible species. Our results indicate that reflectance spectroscopy at the leaf scale detects the effects of P. cinnamomi infection in native vegetation. Extending these results has the potential to improve early detection of disease in natural vegetation and avoiding manual sampling, thus improving management of the disease.