Using VNIR and SWIR field imaging spectroscopy for drought stress monitoring of beech seedlings

Abstract

Drought stress is expected to become a recurrent problem for central European forests due to regional climate change. In order to study the effects on one of the most common tree species in Germany, the European beech (Fagus sylvatica), young potted beech trees were exposed to drought stress in a controlled experiment and their reaction was observed using visible/near-infrared (VNIR) and shortwave infrared (SWIR) field imaging spectroscopy cameras mounted on a platform. Equivalent water thickness (EWT) and leaf chlorophyll content (LCC) were measured and partial least squares (PLS) regression models were trained using these reference measurements and reflectance spectra of the trees. The models were applied to create maps of these properties with a spatial resolution in the millimetre range. These maps can be used to study the spatial distribution of EWT and LCC for single leaves or even for intra-leaf variability. Both properties can be estimated using only the VNIR sensor, but EWT estimation improves considerably by also incorporating SWIR data. LCC estimations with SWIR data alone do not work satisfactorily.

Acknowledgements

The authors thank Max Gerhards, Mario Gilcher, Alexander Mandelkow, Mónica Pérez Sáiz, Pyare Püschel, Gilles Rock, Simon Schreiner, Thomas Sachtleber, and Marion Stellmes for their help during the experiment.

ORCID

Henning Buddenbaum http://orcid.org/0000-0002-0956-5628

Oksana Stern http://orcid.org/0000-0002-2658-2438

Barbara Paschmionka http://orcid.org/0000-0002-0758-5932

Erik Hass http://orcid.org/0000-0002-6777-0005

Thomas Gattung http://orcid.org/0000-0002-1229-235X

Johannes Stoffels http://orcid.org/0000-0003-3509-7773

Additional information

Funding

This work was supported within the framework of the EnMAP project by the German Aerospace Center (DLR) and the Federal Ministry for Economic Affairs and Energy under grant [50 EE 1258].