Solar Induced Chlorophyll Fluorescence and Vegetation Indices for Heat
Stress Assessment in Three Crops at Different Geophysics-Derived Soil
Units
Abstract
Remotely-sensed Solar Induced chlorophyll Fluorescence (SIF) is a novel
promising tool to retrieve information on plants’ physiological status
due to its direct link with the photosynthetic process. At the same
time, narrow band Vegetation Indices (VIs) such as the MERIS Terrestrial
chlorophyll index (MTCI), and the Photochemical Reflectance Index (PRI),
as well as broad band VIs like the Normalized Difference Vegetation
Index (NDVI), have been widely used for crop stress assessment. A match
between these remote sensing products and the spatial distribution of
soil units is expected; nevertheless, an in-depth analysis of such
relationship has been rarely performed so that additional studies are
required. In this contribution, we aimed at the comparison in the use of
normalized SIF (SIF = SIF/PAR; computed with the Spectral Fitting
Method, SFM) and VIs (MTCI, PRI and NDVI) for heat stress assessment in
corn, sugar beet and potato at the beginning and towards the end of a
heatwave occurring in Selhausen, Germany, 2018. Data were acquired with
the HyPlant airborne sensor, which is a high performance imaging
spectrometer with around 0.30 nm of spectral resolution in the Oxygen
absorption bands. We compared different plots located in the upper
(poorer soil characteristics for agriculture such as water holding
capacity and content of coarse sediments) or lower landscape terraces;
we also evaluated the different remote sensing products in comparison
with site specific geophysics-based soil maps. At the beginning of the
heat wave we found that, compared with VIs, SIF data showed a clearer
differentiation of the stress conditions at a terrace level in potato
and sugar beet. However, towards the end of the wave a significant
decrease of MTCI and NDVI contrasted with higher SIF in sugar beet and
corn. Nonetheless, those crops (sugar beet and corn) did not show
significant SIF differences between terraces. A significant spatial
match was found between SIF and geophysics-derived soil spatial patterns
(p = 0.004-0.030) in fields where NDVI was more homogeneous (p =
0.028-0.499, respectively). This suggests the higher sensitivity of SIF
to monitor heat stress compared with common VIs.