Python workflow for seamlessly merging high resolution digital elevation models

Full description

This set of python scripts and Jupyter notebooks constitutes a workflow for seamlessly merging multiple digital elevation models (DEMs) to produce a hydrologically robust high-resolution DEM for large river basins.

The DEM merging method is adapted from Gallant, J.C. (2019) Merging lidar with coarser DEMs for hydrodynamic modelling over large areas, in: El Sawah, S. (Ed.) MODSIM2019, 23rd International Congress on Modelling and Simulation. Presented at the 23rd International Congress on Modelling and Simulation (MODSIM2019), Modelling and Simulation Society of Australia and New Zealand. https://mssanz.org.au/modsim2019/K24/gallant.pdf

The workflow runs on the CSIRO EASI platform https://research.csiro.au/easi/ and expects data stored in an AWS s3 bucket. Dask is used for parallel processing.

The workflow was built to merge all the available high-resolution DEMs for the Murray Darling Basin, Australia, using 852 individual lidar and photogrammetry DEMs from the Geoscience Australia elevation data portal Elvis https://elevation.fsdf.org.au/ and the Forests and Buildings removed DEM (FABDEM; Hawker et al. 2022- https://doi.org/10.1088/1748-9326/ac4d4f), a bare-earth radar-derived, 1 arc-second resolution global elevation model. The seamless composite high-resolution Murray Darling Basin DEM datasets (5 m and 25 m resolutions) produced with this workflow can be downloaded here https://doi.org/10.25919/e1z5-mx88.

The workflow is divided into three parts: 1) Preprocessing, 2) DEM merging and 3) Postprocessing and validation. The Jupyter notebooks in the workflow are also provided in html format for initial access to the content, without needing a python kernel.