GOSHAWK v1.0.5

Optical shape of grains now follow recent paper titled, "Unraveling the optical shape of snow", Alvaro Robledano 2023.

https://doi.org/10.1038/s41467-023-39671-3

GOSHAWK v1.0.4

• following recommendations from Ned Bair et al. (2024), we have derived suitable background reflectance to better inform diffuse irradiance in libRadtran.
• further cleaning of the bash/SLURM scripts

GOSHAWK v1.0.3

• version used in our Cryosphere manuscript
• For our paper, radiation from slopes is diffuse only (the correct way). See Picard, 2020 for a much better paper/explanation https://doi.org/10.5194/tc-14-1497-2020, 2020.

GOSHAWK v1.0.2

• Greatly speeding up k-means clustering by using Dask
• Impurities type parameter: allows for Dust (China PM-2.5 tabulated in Caponi 2017) or Soot (Bond 2006)
• Faster/better .yml files for rebuilding on Borah and similar Linux systems

GOSHAWK v1.0.1

• Now using 30 m canopy height dataset from "Global Forest Cover Change (GFCC) Tree Cover Multi-Year Global 30m"
• Now using updated landcover map from V200 ESA WorldCover
• Now pulling average weekly ozone from S5P product. This is input into libRadtran instead of fixed 300 DU.
• Improved Dozier,2022 sky view function (includes the slope from the pixel itself). Also parallelized.
• Small usability improvements
• Missing escape function in Alex K. equation
• zonal stats/PySpark part in clustering includes arctan2 to account for aspect ranging from 0-360

Initial Release

v1.0.0-beta

upward transmittance