Stordalen Mire landcover classifications map (2014) and associated code

Stordalen Mire landcover classifications (including 3 permafrost thaw stages: Palsa, Bog, and Fen), based on WorldView-2 satellite imagery (WV2) acquired on August 8, 2014. 

This image is a GeoTIFF with embedded georeferencing information, and was generated using the following steps:

1. Task the satellite to get the imagery (WV2)
2. Orthorectify imagery - to drap imagery over the topography - this was done by the company we bought the imagery from
3. Run some texture analysis on the imagery - entropy, std. Dev.
4. Take vegetation plots and locations to extract pixel values
5. Take pixels and associated vegetation plots and develop an artificial neural network (ANN) in JMP 9.0 statistical software
6. Export the ANN as SAS Code
7. Recode or translate from SAS to Python 2.8
8. Run code on the whole image to classify the imagery.  Include the probability of each vegetation class in our output.  

This Python code is included as code_vegetation_landcover_map.py, and includes the ANN recoded. This code exports the completed map image to the filename "mire_classification.tif;" this image was then re-named to vegetation cover.tif.

FUNDING:

• National Aeronautics and Space Administration, Interdisciplinary Science program: From Archaea to the Atmosphere (award # NNX17AK10G)
• National Science Foundation, Biology Integration Institutes Program: EMERGE Biology Integration Institute (award # 2022070)
• United States Department of Energy Office of Biological and Environmental Research, Genomic Science Program: The IsoGenie Project (grant #s DE-SC0004632, DE-SC0010580, and DE-SC0016440)
• National Science Foundation, MacroSystems program (grant # EF-1241037)