An Open-Source Based Workflow for DEM Generation from Sentinel-1 for Landslide Volume Estimation

Lecture given by Lorena Abad, PhD student in Geoinformatics Department, Paris Lodron University Salzburg

Creating a Workflow and Data Accessibility

I was very excited about attending this lecture as open-source data and software has recently become a passion of mine. I was expecting this lecture to follow the standard structure seen in this subject where the lecturer explains what the workflow is and gives a couple of examples of applications of the workflow. However, to my great delight Ms. Abad actually took us through the process of how an open-sourced workflow is created. I had never before been privy to the behind the scenes of developing open-source products. In the past when I have been creating a workflow I did so almost after the fact, where I only had a vague idea of the steps to take beforehand and no solid steps to take in order to achieve my goal. Rather, I would look back on what had worked during my analysis and written that up as my workflow. I recognize to some extent that trial and error is part of developing a workflow but Ms. Abad’s talk was the first time I was able to see how a workflow could intentionally be built from the ground up and not just retrospectively.

Accessibility was a big theme in this lecture and as such I had many thoughts on the subject. One of the key benefits and biggest necessity of open-sourced products that people most often point to is that it allows for increased access to data and technology for a greater number of people. Previously, I had looked at that access to as key to building or analyzing something new but Ms. Abad’s talk made me realize the importance of access to open-source products for quality assurance/quality control (QAQC). No matter how much you are creating, you need to have data, such as open-access spectral libraries, to check your product against or else your product will be seen as unreliable and unusable.

With the important of increased accessibility via open-sourced products established, I had some questions as to the accessibility of the workflow Ms. Abad was detailing. One of the most promising benefits of the SliDEM workflow is that everything is run through docker, which allows for greater collaboration, as well as through scripts. This removes the need to access actual servers or download all the many images related to DEM generation. This certainly makes the data more accessible especially to those with little technological resources, but because this workflow relies so heavily on programming I wonder how accessible it really is. Certainly, programming is becoming a larger part of the GIS community and in the future, it might be a necessity to know how to code in order to work in the field, but there are still many people who do not know how to code. Perhaps the issue of difficulty in implementation could be solved with detailed documentation, but it is something that is important to consider in regards to accessibility. Another concern I had in relation to the automation of this workflow, and really with any automated workflow, was if the computer is doing everything does the user really understand the process or output of the workflow? This is not so much an issue for the creator of the script or workflow as they know what they have told the computer to do, but for the users implementing the script they might not completely understand how or what they are doing which might lead to errors in analysis or incorrect conclusions. Automation is beneficial in that it reduces human error and perhaps the user does not need to know exactly how everything works to use a product successfully, but it is another consideration to keep in mind in regards to accessibility with open-sourced products.

Twitter  facebook  Email