Use of thermal imaging and Fourier transform infrared spectroscopy for the pre-symptomatic detection of cucumber downy mildew

Abstract

Most cultivars of cucumber are susceptible to Pseudoperonospora cubensis, the causal agent of cucurbit downy mildew. This study aimed to evaluate the potential of two methods for the pre-symptomatic detection of downy mildew in cucumber. First, we observed the infection process of downy mildew using digital infrared thermography, as measured by the maximum temperature difference (MTD) and spot temperature of symptomatic areas. Under controlled conditions, visible symptoms were observed 5.36 ± 0.10 days after infection with P. cubensis, while thermal differences were detected 4.42 ± 0.16 days after infection. The MTD values of infected leaves were higher compared to those of non-infected leaves, but the MTD decreased sharply before the symptoms of downy mildew became visible. Significant differences in the spot temperatures of the symptomatic areas were apparent between the infected and healthy leaves from 4 days after infection. Second, we observed a changes in the Fourier transform infrared spectroscopy (FTIR) spectra of infected leaves and three characteristic wavenumbers—2977 cm−1, 1544 cm−1, and 1050 cm−1—were selected for the pre-symptomatic detection of cucumber downy mildew. According to the peak areas at these wavenumbers, pre-symptomatic and symptomatic samples could be classified correctly. These results clearly demonstrate that both thermal infrared imaging and FTIR spectra allow for the discrimination of healthy and infected leaves before there are visible symptoms. The pre-symptomatic data may be useful for scheduling timely fungicide applications for early detection and subsequent control of this destructive disease of cucumber.