Early indication of plant stresses by changes in vegetation indices and fluorescence

Photosynthesis dynamically adapts to changing environmental conditions. Extreme environmental conditions such as high temperature and water limitation directly reflect in the photosynthetic performance as vegetation stress. To monitor photosynthesis dynamics at large vegetation scales, indicators of stresses in vegetation extracted by non-invasive techniques can be used.Spectral reflectance indices such as NDVI and PRI have been tested to provide rapid and non-invasive estimation of photosynthesis at regional scales. Unfortunately, these indices lack sensitivity to short-term vegetation physiological changes. Alternatively, a passive retrieval of sun-induced chlorophyll fluorescence (SIF) using high performance imaging spectroscopy has been suggested. SIF emission is closely related to the status of photosynthesis and therefore has the potential to track adaptation of this process to changes in environmental conditions and plant status. Although SIF emission represents a weak signal in comparison to the reflected radiation, recent research provides evidence that SIF can be measured by exploiting solar and atmospheric absorption lines using high performance spectrometers.In this contribution, we present the design and first results of an experiment that aimed to evaluate the capability of the new remote observable SIF and common reflectance based indices to detect plant stress. In the experiment, carried out in June and July 2014 in Latisana (Italy), we treated homogeneous lawn plots to provoke different level of stress in plants. In particular, we applied the herbicide (Dicuran) in different concentrations to four plots, three other plots were treated with the anti-transpirant Vaporgard; while three remaining plots were not treated and used as control. Diurnal variations in SIF and surface reflectance were evaluated for period of two weeks, using the high performance imaging spectrometer HyPlant. HyPlant is a dedicated fluorescence spectrometer and allows measuring radiance in the wavelength range between 400 nm and 2500 nm with a high spectral resolution of 0.26 nm between 670 nm and 780 nm. Data was recorded for spatial resolution of 1 meter per pixel.Airborne data was calibrated and validated using high resolution top of canopy measurements of reflectance and SIF. Vegetation indices and SIF were calculated for treated and non-treated areas and compared with corresponding ground measurements. Additional measurements of CO2 and H2O fluxes using chambers and an eddy covariance tower to monitor photosynthetic activity were performed.First results show contrasting responses of emitted SIF and vegetation indices for plant stress: SIF strongly increased after the treatment with Dicuran followed by a slow decrease down to the initial values. The Meris Terrestrial Chlorophyll Index (MTCI) showed a clear decrease after the herbicide treatment. We conclude that SIF is a promising new observable sensitive to plant stress which provides complementary information to commonly used remote observables origin form reflectance spectroscopy. We suggest a combined use of SIF and reflectance spectroscopy to establish reliable approaches for rapid, early, and non-invasive plant stress detection