This work is the code used as part of my undergraduate dissertation. Scripts contain code used. However, data not added to github and therefore, will not run. Own data will need to be applied. Apologies for poor commenting throughout scripts.
These are low amplitude, high wavelength features constructed from aeolian drift snow, formed from tens thousands of years. They occupy vast fields +500,000km2 (see image below for segment example). These features are previously thought to occupy ~18% of Eastern Antarctica. It is unknown how well modern altimeters will represent the megadune morphology and if so how accurately. This work aims to assess this comparing between various forms of altimters ICESat-2 (IS2), CryoSat-2 (CS2). Therefore, assessing Radar vs laser based approaches. This was weighted and assessed against REMA a DEM for Antarctica constructed from ICESat-1 and CryoSat-2 alongside the Worldview product series and Geoeye data.
These scripts illustrate the code used to investigate Antarcticas megadunes. Megadunes are a unique wave landfrom expression formed by wind laid snow deposits.
code based from: https://portailsig.org/content/python-utilisation-des-couches-vectorielles-et-matricielles-dans-une-perspective-geologique-.html .
These REMA extracted elevations from a flightpath aligned with various altimeter satellite products to assess how well REMA accurately represents megadune features. The altimeter products used were ICESat-2 and CryoSat-2
Calculating: amplitude, wavelength.
see/ use Waveform_calc.py
This script extracts the peak, trough points along transect lines and stores these values. These values are then used to calculate amplitude and wavelengths. Plots can be created showing all transects for each site. But adds 12mins runtime to the script.
HDF5 and netcdf files containing transect paths were generated and extracted over 5-sites and contrasted with REMA.
See 'Altimeter representation of megadunes each site folder' for individual site scripts Site[x].ipynb [x] = A/B/C/D/E.
QGIS transects were drawn and the shapefiles were read and processed in python extracting each pixel along the shapefile. Therby extracting the raster slope/roughness pixel. Basic stats and ranges were quantified.
see/use 'Extract surface metric values along transects'--> slope_extraction.py & Roughness_extraction.py
These ranges were then used back in QGIS to generate an output raster calculated megadune location model (continental scale).