primary production

Plant growth in most ecosystems forms the base or “primary” component of the food web. The amount and type of plant growth in an ecosystem helps to determine the amount and kind of animals (or “secondary” productivity) that can survive there.

Microbial mat biomass and Normalized Difference Vegetation Index (NDVI) values from Lake Fryxell Basin, Antarctica, January 2018


This package contains data collected from microbial mat surveys (i.e., percent cover, ash-free dry mass (AFDM), and pigment concentrations – chlorophyll-a, scytonemin, and carotenoids) associated with satellite-derived Normalized Difference Vegetation Index (NDVI) values from the Lake Fryxell Basin of Taylor Valley, located in the McMurdo Dry Valleys of Antarctica. The purpose of this study was to quantitatively compare key microbial mat characteristics to NDVI. Data were collected at seven plot locations within the Canada Glacier Antarctic Specially Protected Area (ASPA) near Canada Stream, as well as alongside Green Creek and McKnight Creek. NDVI values were derived from a WorldView-2 multispectral satellite image taken of the Lake Fryxell Basin on January 19, 2018, while biological ground surveying and sampling were conducted during the 2nd and 4th weeks of January 2018. 

LTER Core Areas: 

Dataset ID: 


Associated Personnel: 


Short name: 


Data sources: 



Locations of visually conspicuous microbial mats outside of stream channels were selected for seven 100 m2 plots beside Canada Stream in the Canada Glacier Antarctic Specially Protected Area (ASPA), Green Creek, and McKnight Creek in the Lake Fryxell Basin. Plots Canada Stream-1, -2, -3, and -4 were all located in the lower flush of the ASPA and Canada Stream-5 was in the upper flush. Hemispherical photos of ground cover along with position data were taken every 5 m at each of the seven plots. Using a random stratified sampling design, we sampled microbial mat from 9 or 10 locations within each plot for pigment chemistry and ash-free dry mass (AFDM).The hemispherical photos were analyzed for microbial mat cover using ImageJ, an image analysis software. By converting the photos to binary black and white images, the software distinguished pixels representing microbial mats (black) from those representing bare soil or rocks (white). Estimates for percent cover were quantified using a histogram. Chlorophyll-a concentration was measured with a spectrophotometer (Shimadzu UV-1601) using a protocol that corrects for scytonemin concentration. The samples were dried at 105°C for 24 hrs and extracted in 10 mL of 90% unbuffered acetone using 0.5 g of mat, based on protocols from Garcia‐Pichel and Castenholz (1991) and standard methods employed by the McMurdo Dry Valleys Long Term Ecological Research program (MCM LTER). The samples were then left for 24 hrs in a dark room at ambient temperature for complete chlorophyll extraction. After centrifuging for 10 min at 4000 RPM, they were analyzed on a spectrophotometer using 10 mL cuvettes. The absorbances contributed by chlorophyll-a, carotenoids, and scytonemin were quantified at 663, 490, and 384 nm, respectively, using the trichromatic equations outlined in Garcia‐Pichel and Castenholz (1991). One sample from Canada Stream-2 had extremely high pigment concentrations overall, with values greater than 6 standard deviations of the mean chlorophyll-a concentrations for all observations, and it was removed from the dataset. AFDM was calculated by weighing approximately 0.25 g of oven-dried sample, combusting at 550°C for 60 min using a muffle furnace, and reweighing after cooling in a desiccator. Given the extremely low clay content of soils in this region, the rehydration of clays was assumed negligible, so the methods did not include rewetting samples to account for the hydration of clays.The WorldView-2 multispectral satellite (DigitalGlobe, Inc.) collected a cloud-free image of the Lake Fryxell Basin on January 19, 2018, within 10 days of our surveying and sampling. The image was georeferenced using ground control points and processed to atmospherically corrected surface reflectance using the Environment for Visualizing Images software (ENVI, Harris Geospatial). We calculated the normalized difference vegetation index (NDVI) for the image. Using ArcMap, we converted the NDVI image raster file to points and then selected the pixels whose centroids fell within the 100 m2 plot boundaries. Each plot corresponds to approximately 14-17 pixels (5.95 m2 per pixel) with unique NDVI values.


Subscribe to RSS - primary production