SNAP data QA/QC refers to multiple threads of collections of R code in the form of scripts and functions across various projects utilizing SNAP data sets. However, there is particular focus on SNAP’s gridded climate- and fire-related data products, including:

The two most prominent threads of QA/QC relate to the climate data and ALFRESCO outputs noted above. The QA/QC process involves extraction of data from geotiff layers of the above data products. Data are extracted based on:

The data extractions are intensive processes. Depending on user configurations, these processes can take a while to execute. Most of the scripts, particularly those upstream in the R code process chain which entail the heaviest I/O involving thousands of high-resolution geotiff files, are called via SLURM scripts and handled by the SLURM job manager on the Atlas cluster. This is considered best practice. The code is executed in non-interactive mode R and does not require an existing connection to the server by the user except to initially launch the SLURM job. Some of the downstream processes that perform relatively light organizational work on already parsed outputs do not use SLURM.

The process chain involves the following:

Currently, here you can find documentation, including R code, which covers most of the primary (extraction) and secondary (organization) stages pertaining to the two main threads of SNAP data QA/QC. Before looking at the code, it may be helpful to capture the overall picture of the process chain by looking at the code flow diagram.