Optical leaf area assessment supports chlorophyll estimation from UAV images

• In canopy with LAI < 2.7, LCC may be sufficient to characterise chlorophyll content. • In LAI > 2.7 canopy, it is worthwhile to calculate CCC based on LAI and LCC. • For LAI < 2.7 canopy, NDRE seems to be the best predictor of chlorophyll content. • For LAI > 2.7 canopy, CI RE , SR RE and NDRE were the best predictors of CCC. Measurement of crop chlorophyll content provides information on expected yield at an early stage of vegetation development. Spectral vegetation indices (VIs) are closely related with crop chlorophyll content and nowadays they became common tools for estimating parameters of vegetation monitoring in field scale. Thus, the objectives of this study were to validate the correlation of VIs (calculated from drone-based hyperspectral images) with leaf chlorophyll content (LCC) and canopy chlorophyll content (CCC) of crops grown at three different nitrogen levels at two experimental sites. LCC and leaf area index (LAI) were measured with handheld devices. The effect of vegetation size, expressed as two LAI ranges of canopy, on the magnitude of the resulting correlations was investigated as well. Our results showed that for less developed vegetation (LAI < 2.7), all studied VIs are suitable for assessing chlorophyll content. However, at higher LAI values, some VIs had no significant correlation with either LCC or CCC. Based on linear regression, NDRE for less developed vegetation (LAI < 2.7), as well as NDRE, CI RE or SR RE for closed vegetation (LAI > 2.7), are recommended for monitoring chlorophyll content when the LAI of the vegetation is known and therefore the CCC can be derived. We conclude that drone imagery may greatly assists farmers in observing biophysical characteristics, but is limited for observing chlorophyll status within crops of closed vegetation size.