Rockslope and landslide monitoring using high temporal resolution terrestrial structure from motion photogrammetry: a case study of a landslide in Majes zone, Peru using multi-epoch photogrammetric techniques

Monitoring the movement of landslides and rock slopes is critical when there is potential for injury of persons or damage to property or infrastructure. There has been extensive research into the processes that drive slope movements, but most have been based on monitoring data that is either limited in temporal resolution or limited in spatial resolution. Point clouds created from lidar scanning or photogrammetric techniques have seen increased use in recent years due to their potential to capture surficial slope movements at both high spatial and temporal resolutions. Much of the existing body of literature, specifically research related to photogrammetric monitoring, has evaluated various monitoring system setups and processing methods, but has not assessed system performance at high monitoring frequencies and over long monitoring periods in detail. This thesis uses automatically captured photos taken at a sub-daily frequency by five fixed-base cameras in conjunction with Structure-from-Motion (SfM) photogrammetric processing techniques to evaluate changes of a large landslide and rock slope in Majes, Arequipa, Peru. The challenges and limitations of the monitoring method specific to the Majes study site and the implications with respect to landslide displacement trends and rock fall magnitude-frequency relationships are explored. This thesis presents modifications to recently developed photogrammetric processing methods such that, for monitored areas of widespread instability, 3D point clouds can be produced that are of high enough quality and consistency that reliable change detection over time can be performed. This thesis also shows that even after refining the photogrammetric and subsequent change detection methods, annual variations in solar incidence angle and variations in lighting conditions due to variable atmospheric conditions significantly limit the overall consistency and precision of high frequency photogrammetric monitoring in natural environments. In periods of ideal lighting and atmospheric conditions, landslide displacement rates roughly between one and three meters per year, measured in increments as small as 0.1 m, were reliably identified. Regarding rock slope monitoring, this thesis shows that there are potential issues with the common notion that higher frequency monitoring is always superior to lower frequency monitoring. For lower spatial resolutions or when only large changes are of concern, using a high frequency monitoring method may cause small volume changes that eventually aggrade into larger areas of change to be missed, whereas most of the total volume change would be captured with lower frequency monitoring intervals (all else held equal). High frequency monitoring is still the most accurate approach for evaluating slope evolution from a rockfall frequency standpoint, but longer-term comparison intervals may be required in some cases to more accurately reflect the total volume change of a given rock slope over a long period of time.

Name thesis:degree_name

Master of Science (M.S.)

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Geology and Geological Engineering

Grantor dc:publisher

Colorado School of Mines. Arthur Lakes Library

Year dc:date.issued

2023

Author dc:creator

• Butcher, Bradford C.

Advisors dc:contributor.advisor

• Walton, Gabriel
• Kromer, Ryan

Committee members dc:contributor.committeemember

• Roth, Danica
• Zhou, Wendy

• landslide
• monitoring
• multi-epoch
• photogrammetry
• rockfall
• structure-from-motion

Statement dc:rights

• Copyright of the original work is retained by the author.

Language dc:language.iso

eng, English

Identifier

T 9546

OAI identifier oai:identifier

oai:repository.mines.edu:11124/178490