- Andrew Bennett,Department of Civil and Environmental Engineering, College of Engineering, UW Seattle
- Jessica Lundquist,Department of Civil and Environmental Engineering, College of Engineering, UW Seattle
- Joseph Hamman, National Center for Atmospheric Research
- Bart Nijssen, Department of Civil and Environmental Engineering, College of Engineering, UW Seattle
For the last two years we have taught an online graduate course on computational modeling for simulating snow hydrology. We used open platforms to reduce the barrier to entry for students learning advanced concepts and tools for computational hydrology. Previously, students encountered considerable hurdles that both confused them and slowed the overall pace of the course.
We designed the course using multiple open platforms that enable computational research in the sciences. We used the Jupyter platform (a set of open-source tools enabling interactive computing) as the interface to our computational tools. The computing resources were provided by Pangeo (a community platform for open and reproducible data analysis in the geosciences). Students were able to publish and share their finished analyses on HydroShare (an online collaboration environment focused on hydrology). By leveraging Jupyter, Pangeo, and HydroShare, we were able to get students up and running without wasting precious course-hours on technical support. The design of our course was inspired by the UC Berkeley “Foundations of Data Science” course and other recently popular online computing courses.
The use of these platforms streamlined the packaging of datasets, tutorials, and homework assignments in a way that made the key concepts of the course clearer and emphasized the actual usage of advanced computational techniques over logistical challenges in managing multiple computing environments. We believe that this framework for computationally-intensive courses drives their adoption for student-driven research projects. In our presentation, we will share our perspectives and provide recommendations for how instructors in other fields can leverage existing technologies and build collaborations to foster best practices in computational courses. We motivate our recommendations based on student testimonials provided as part of course evaluations.