Hyperspectral vegetation indices offer insights for determining economically optimal time of harvest in Mentha arvensis

Continual crop monitoring and harvesting at optimum time to ensure maximum economic returns is a huge challenge for farmers having small landholdings in Indo-Gangetic plains where Mentha arvensis is a popular cash crop. The present study evaluates different hyperspectral indices for phenotyping the canopy of M. arvensis winter crop (cv. CIM-Kranti) for detection of optimum harvest time using ground-based plant traits. Field experiments were conducted during mid-crop growth stage to late crop growth period for two consecutive years using hyperspectral camera (400-940 nm) to acquire images representing the spectral reflectance of crop canopies. Hyperspectral vegetation indices (HVIs) derived from these images were further evaluated with an aim to identify those hyperspectral indices showing a strong correlation to measured crop parameters. Out of the 18 HVIs, 4 HVI's i.e. photochemical reflectance index (PRI), carotenoids reflectance index (CRIr), hyperspectral normalized difference vegetation index (HNDVI) and plant senescence reflectance index (PSRI) were recognized as potential predictors for sucker traits (weight and length) and thereby indicators of optimum harvesting time for M. arvensis winter crop (cv. CIM-Kranti) cultivated primarily to produce suckers for crop propagation. Our results show promise for the development of non-destructive methods for the detection of optimum harvest time using crop reflectance. This study holds importance for the menthol mint farmers, who cultivate the crop for suckers during winter months for additional income generation.