Geoscience Field Issues Using High-Resolution Topography to Understand Earth Surface Processes
Public four-year institution, primarily undergraduate
Online (Zoom), Canvas
College Upper (15-16)
This course was originally intended to be conducted in-person with both field data collection and analysis to meet the "field" component of the University of Northern Colorado's Earth Science degrees (Environmental and Geology). With the 2020 world pandemic and cancellation of most in-person activities for 2020 the course was adapted for 100% online implementation with an optional one-day field campaign. As such, UNAVCO (in collaboration with the GETSI Project) and University of Northern Colorado embarked on a data collection campaign for a summer field course entitled "Geoscience Field Issues Using High-Resolution Topography to Understand Earth Surface Processes". The team collected GNSS data, drone imagery for use in structure from motion, and terrestrial laser scanning from a site near Greeley, Colorado on the Poudre River. These data were used in mock field campaigns and real analyses implemented by students virtually through Zoom and Canvas. Students from throughout North America attended the course, most meeting Field Camp requirements required for graduation. A few graduate students attended the course for training in their research areas in high-resolution 3D dataset acquisition and analysis.
Some of the individual activities for the course have been separately submitted to the NAGT Online Field Experiences Activity Collection and are linked below.
- Day 1 - Getting started with Structure from Motion (SfM) photogrammetry
- Day 2 - Introduction to GPS/GNSS
- Unit 1: GPS/GNSS Fundamentals - from the GETSI module High Precision Positioning with Static and Kinematic GPS/GNSS.
- Post-processing GPS/GNSS Base Station Position
- Day 3 - SfM of Poudre River at Sheep Draw Reach
- Day 4 - Working with Point Clouds in CloudCompare and Classifying with CANUPO
- Day 5 - Report 1 - SfM Feasibility Report assignment (Microsoft Word 2007 (.docx) 46kB Oct29 20)
- Day 6 - Optional Field Day (collected additional dataset with local students who could participate) - no associated assignment
- Day 7 - Introduction to Terrestrial Laser Scanning (TLS)
- Day 8 - Point Cloud/Raster Differencing and Change Detection
- Day 9 - OpenTopography Data Sources and Topographic Differencing
- Day 10 - Report 2 - Methods Comparison Report.docx (Microsoft Word 2007 (.docx) 46kB Oct29 20)
- Day 11 - Presentations - students gave short presentations to the group based on their final report
The course was originally envisioned as a hybrid-type course (all in person) with data collection and then data analysis (50/50) but ended up being almost 20% lecture, 20% demonstration/tutorial, 10% group work and 50% individual work. Students learned more data source and analysis content than originally planned, but less on actual field campaign implementation. We made field videos and mock field scenarios as a surrogate, but students did not actually use any of the field equipment or troubleshoot field campaign issues unless they completed the optional field demo day.
Students are able to:
- Make necessary calculations to determine the optimal survey parameters and survey design based on site and available time.
- Integrate GNSS targets with ground-based LiDAR and SfM workflows to conduct a geodetic survey.
- Process raw point cloud data and transform a point cloud into a digital elevation model (DEM).
- Conduct an appropriate geomorphic analysis, such as geomorphic change detection.
- Justify which survey tools and techniques are most appropriate for a scientific question.
Before the workshop-style portion of the course, students were assigned papers on a topographic data collection method and posted a reading summary and then a response to the post of others (implemented as a discussion post) for each reading. During the course, mock field campaigns were implemented in Zoom breakout groups. Teams discussed and planned components of field work and then shared with the group. Students assessed how their plan compared with the actual implementation and discussed with their teams and the entire class. We also posted questions as they arose on a Slack site.
Discussions, daily activities (1-2 per day), and reports (Report 1 and Report 2) were used as assessment, along with a daily reflection.
- Adaptations from the original plan were made to allow this course to be successful in an online environment
- This course relied heavily on data post-processing and analysis that could be done 100% on a computer, virtually. It was successful with the caveat that students needed a good internet connection, the ~2-week timeframe free from work and other commitments, and a reliable computer. Students needed to be able to install a number of software onto their computers (MetaShape, CloudCompare, ArcGIS). Trouble shooting and accommodations for hiccups had to be made often.
- The most successful elements of this course are:
- Structure from Motion (introductory SfM activity and Sheep Draw activity) was really successful. Students showed successful 3D models in activities and final reports.
- Exploration of other datasets and use of OpenTopography data and tools also successful (people turned in nice results here in activities and final reports)
- Recommendations for faculty who teach a course like this:
- If entire content is used, spread out over longer period.
- If preparation allows, build more asynchronous elements into the course and allow more time to complete assignments.
- Pick fewer things to cover if limited to one or two weeks time.
- Terrestrial laser scanning model was not as robust as Structure from Motion model - in this case it was used to compare and contrast two different methods at the same site. This was successful, but an alternative would be do use a temporal dataset with only one method.
- Many people struggled with ArcGIS installation (Mac users) and for some reason CloudCompare worked wonderfully for some and not for others. I do not yet have a fix for this. The associated activities are really valuable, so recommendations/other options here would be helpful.
Geoscience Field Issues Syllabus (Microsoft Word 2007 (.docx) 66kB Oct26 20)
- Tarolli, P. (2014). High-resolution topography for understanding Earth surface processes: Opportunities and challenges. Geomorphology, 216, 295-312.
- Westoby, Matthew J., et al. "'Structure-from-Motion'photogrammetry: A low-cost, effective tool for geoscience applications." Geomorphology 179 (2012): 300-314.
- James, Mike R., et al. "Guidelines on the use of structure‐from‐motion photogrammetry in geomorphic research." Earth Surface Processes and Landforms 44.10 (2019): 2081-2084.
- Sanz‐Ablanedo, Enoc, et al. "Reducing systematic dome errors in digital elevation models through better UAV flight design." Earth Surface Processes and Landforms (2020).
- Brodu, Nicolas, and Dimitri Lague. "3D terrestrial lidar data classification of complex natural scenes using a multi-scale dimensionality criterion: Applications in geomorphology." ISPRS Journal of Photogrammetry and Remote Sensing 68 (2012): 121-134.
- Heritage, George, and David Hetherington. "Towards a protocol for laser scanning in fluvial geomorphology." Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group 32.1 (2007): 66-74.