Antarctic moss stress assessment based on chlorophyll content and leaf density retrieved from imaging spectroscopy data

• Published in: New Phytologist
• Main Authors: Malenovský, Zbyněk, Turnbull, Johanna D., Lucieer, Arko, Robinson, Sharon A.
• Other Authors: Australian Research Council and Antarctic Science
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2015
• Subjects: Antarc*; Antarctic; Schistidium antarctici
• Online Access: http://dx.doi.org/10.1111/nph.13524; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fnph.13524; https://onlinelibrary.wiley.com/doi/pdf/10.1111/nph.13524; https://onlinelibrary.wiley.com/doi/full-xml/10.1111/nph.13524; https://nph.onlinelibrary.wiley.com/doi/pdf/10.1111/nph.13524

Summary The health of several East Antarctic moss‐beds is declining as liquid water availability is reduced due to recent environmental changes. Consequently, a noninvasive and spatially explicit method is needed to assess the vigour of mosses spread throughout rocky Antarctic landscapes. Here, we explore the possibility of using near‐distance imaging spectroscopy for spatial assessment of moss‐bed health. Turf chlorophyll a and b , water content and leaf density were selected as quantitative stress indicators. Reflectance of three dominant Antarctic mosses Bryum pseudotriquetrum , Ceratodon purpureus and Schistidium antarctici was measured during a drought‐stress and recovery laboratory experiment and also with an imaging spectrometer outdoors on water‐deficient (stressed) and well‐watered (unstressed) moss test sites. The stress‐indicating moss traits were derived from visible and near infrared turf reflectance using a nonlinear support vector regression. Laboratory estimates of chlorophyll content and leaf density were achieved with the lowest systematic/unsystematic root mean square errors of 38.0/235.2 nmol g −1 DW and 0.8/1.6 leaves mm −1 , respectively. Subsequent combination of these indicators retrieved from field hyperspectral images produced small‐scale maps indicating relative moss vigour. Once applied and validated on remotely sensed airborne spectral images, this methodology could provide quantitative maps suitable for long‐term monitoring of Antarctic moss‐bed health.