MS Plan A Defense: Evaluation of Coastal Imaging Georectification For Measurin

Gabriel Nelson
Masters Student
Department of Ocean & Resources Engineering
University of Hawai’i at Manoa

**This defense will be held in-person (HOLM 287) and Zoom**
Meeting ID: 884 1267 3489
Passcode: GabrielMS
https://hawaii.zoom.us/j/88412673489


Coastal inundation presents a significant and escalating threat to island communities such as Hawaii, driven by rising sea levels and increasing wave activity. Under these circumstances, precise measurement of wave runup is crucial for accurate coastal hazard assessments, effective mitigation strategies, and sound engineering practices. UAV-based photogrammetry paired with the Coastal Imaging Research Network (CIRN) Toolbox offers a powerful tool for coastal image processing. However, CIRN’s standard georectification method assumes a flat reference plane, which can introduce significant elevation errors in complex topographic environments. This study evaluates the accuracy of the flat-plane approach versus a topographically-informed method that incorporates Digital Elevation Models (DEMs) for image georectification, with a focus on its applicability to Hawaiian beaches.

Deterministic analysis using surveyed ground truth data shows that the Topographically- Informed method produced three times lower elevation error than the Flat Plane method (mean absolute error of 0.063 ft vs. 0.229 ft), with improved consistency across varied sites and viewpoints. Time series comparisons of runup exceedance metrics (R1%, R2%, R10%) showed that average differences between georectification methods ranged between 2.25 to 8.45% with beach slope, beach elevation, and UAV orientation influencing results. Differences between runup statistics combined with the insights from deterministic analysis indicate that the topographic correction improves georectification accuracy for runup metrics—particularly under the steep and irregular beach conditions common in Hawai‘i.

To support survey planning, a geometric error estimation formula is developed to predict Flat Plane georectification error. With an average predictive error of less than 0.05 ft, the formula provides a validated tool for assessing and minimizing error tolerance before field deployment and deciding when topographic correction is necessary. The tools developed here enable practitioners in Hawai‘i and beyond to optimize UAV-based coastal monitoring for both efficiency and accuracy.

Event Sponsor
Ocean and Resources Engineering, Mānoa Campus

More Information
8089568110

Tweet