Geoglam Fractional Cover - MODIS, CSIRO Land and Water Algorithm

Vegetation fractional cover represents the exposed proportion of Photosynthetic Vegetation (PV), Non-Photosynthetic Vegetation (NPV) and Bare Soil (BS) within each pixel. In forested canopies the photosynthetic or non-photosynthetic portions of trees may obscure those of the grass layer and/or bare soil. The MODIS Fractional Cover product is derived from the MODIS Nadir BRDF-Adjusted Reflectance (NBAR) product (MCD43A4, collection 6). A suite of derivative are also produced, namely total vegetation cover (PV+NPV), monthly fractional cover and total vegetation cover, monthly anomaly of total cover against the time series, and three-monthly total cover difference. MODIS fractional cover has been validated for Australia.

Enivronment setup

• Run git clone https://github.com/nci/geoglam.git into your local directory
• Run setup_env.sh (i.e. bash setup_env.sh) to set up miniconda environment. You only need to run this step once
• Open submit_pbs_jobs.sh and fill up NCI_PROJECT and BASE_DIR. NCI_PROJECT is your NCI project number, fr1, for example. BASE_DIR is the full path of the to-be-generated geoglam products. For example, /short/<project no.>/<user id>/geoglam_data

Generating Geoglam netCDF products

• Once the miniconda environment is set up, run submit_pbs_jobs.sh <tile csv file> to submit PBS jobs for each tile in the tile csv file. There are a few example tile csv files under the tile_files directory.
• Once the PBS jobs finished, run utils/parse_logs.py (i.e. python utils/parse_logs.py <log directory>) to extract any failed tiles. If there are failed tiles, a csv file that contains the failed tiles will be written under the current working directory.
• You might want to archive the log files for future references. To do so, run utils/archive_log_files.sh

Updating netCDF metadata

If the metadata of the netCDF product files need to be updated, please do the following steps:

• Enter the metadata directory (i.e. cd metadata)
• Run bash batch_update_metadata.sh <metadata json file>. The <metadata json file> has the same format as nc_metadata.json located under the geoglam root directory.

Optionally, one might want to open batch_update_metadata.sh and edit the UPDATE_DIR and the PYTHON variables to point to alternative locations.