The Antarctic has experienced major changes in temperature, wind speed and stratospheric ozone levels during the last 50 years. However, detecting the biological effects of Antarctic climate change is hindered by the paucity of long-term data sets, particularly for organisms that have been exposed to these changes throughout their lives. In a manner similar to tree rings, old growth shoots of Antarctic mosses also preserve a climate record of their surrounding environment. Stable carbon isotope analyses of Antarctic moss shoots show that their growth rates correlate with key climatic variables and that balance between warming, high-wind speeds and elevated UV irradiation may determine the fate of these oases of Antarctic biodiversity (Clarke et al., 2011).
The push-broom mini-Hyperspec spectrometer unit in Antarctica (left) and during a ground-based imaging spectroscopy investigation of a coastal moss bed at Robinson Reach 2012 (right).
Remote sensing of Antarctic moss-beds
Large-scale carbon isotope probing of spatially extensive moss beds is unfeasible due to the short Antarctic growing season lasting only few weeks. Therefore, we urgently need a new efficient method able to monitor spatially climate change induced stress of the Antarctic moss flora in short time frame. Cloudy weather and spatial fragmentation of the moss beds makes satellite images unsuitable for this task. Unmanned aerial vehicles (UAV), flying at low altitudes and collecting image data even under full overcast, can, however, overcome the insufficiency of satellite remote sensing. TerraLuma is currently developing a UAV system consisting of a remote-controlled microcopter carrying on-board different remote sensing imaging sensors (visible colour camera, VNIR spectroradiometer, and thermal infrared camera). This system is tailored to perform fast and cost-effective mapping of Antarctic flora at ultra-high spatial resolution (1-10 cm depending on flight altitude).
Ground measured reflectance signatures (in digital numbers) of vigorous (left), slightly stressed (middle) and highly environmentally stressed (right) Antarctic moss. Note the significant reflectance change between 700 and 800 nm depending on moss stress level.
Ground spectroscopy
First ground-based spectral measurements acquired during the short 2012 Antarctic summer on the Windmill Islands reveal a high potential of imaging spectroscopy for physiological stress monitoring of the Antarctic moss beds. Images recorded with a mini-Hyperspec imaging spectrometer (Headwall Inc., US) in the surroundings of the Australian station Casey contain from 162 up to 324 narrow spectral bands of wavelengths between 399 and 998 nm. A first analysis of uncalibrated spectral image data indicates a significant difference in near-infrared reflectance of differently vigorous moss patches. Highly vigorous moss exhibits high reflectance intensity between 700 and 800 nm, while the same wavelength’s reflectance of highly stressed moss is more than twofold lower. The narrow-band Normalized Difference Vegetation Index (NDVI; Tucker, 1979) computed from red and near infrared wavelengths seems to be, therefore, a simple, but reliable spectral indicator of the Antarctic moss stress state. A more thorough analysis of the calibrated UAV acquired spectral images is, nevertheless, needed to spatially assess the overall stress load of the coastal Antarctic moss beds. Our investigation continued during the Antarctic expedition 2013.
Spectral scan of moss bed at the Antarctic protected area ASPA 135 (Casey station, Antarctica, 29-01-2012) in false colours (upper part), and a first qualitative assessment of moss vigour (lower part) derived from the spectral image using the Normalized Difference Vegetation Index (Tucker, 1979).
References
Clarke L.J., Robinson S.A., Hua Q., Yare D.J., and Fink D., 2011. Radiocarbon bomb spike reveals biological effects of Antarctic climate change, Global Change Biology 18(1), 301–310.
Tucker, C. J. (1979). Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing of Environment, 8, 27−150.
