 |
| Titel |
Reconstructing landslide dynamics and characteristics using remote sensing data (photogrammetry, LiDAR and seismic data): comparison between different techniques and complementary data analysis |
| VerfasserIn |
Marta Torné, Marta Guinau, Mar Tapia, Cristina Pérez, Manuel Jesús Royan, Anna Echeverria, Pere Roig, Emma Suriñach |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250104034
|
| Publikation (Nr.) |
 EGU/EGU2015-3454.pdf |
|
|
|
|
|
| Zusammenfassung |
| The purpose of this study is to characterize the rock planar landslide that occurred in the
village of La Riba (Catalonia) on May 5th 2013, using different techniques such as
photogrammetry, terrestrial LiDAR data, and seismic data. Advantages and disadvantages of
these techniques were evaluated.
Back-analysis and characterization of landslides allow us to better understand their
behaviour. This information could be used to protect areas affected by similar hazards.
Remote techniques are an excellent tool to obtain data and to reduce the exposure of
technicians in unstable (or inaccessible) areas.
After the May 5th natural landslide, a controlled blasting was carried out to stabilize the
slope. Using this programmed blasting as a benchmark, two photogrammetric models and
two terrestrial LiDAR data models corresponding to the pre and post blast were made to
compute the rock volume involved in the blast. The blasting process was recorded with two
HD video cameras and by two temporary seismic stations deployed close to the
site. Both the seismic and video records enabled us to reconstruct the details of the
blasted landslide. The volumes obtained from seismic data were compared with the
total volumes computed by LiDAR and photogrammetry. Moreover, information
about the natural landslide was obtained from the records of a permanent seismic
station 10 km from the site. Data such as the estimated fallen volume, the landslide
mechanism and time of occurrence are information that would otherwise not be
obtained.
Six discontinuity families were detected and characterized in the rock slope using the
photogrammetric and LiDAR models with a software developed by the Institut de Recerca de
Geomodels of the Universitat de Barcelona. Similar results were obtained from the two
models, but the higher point density of the LiDAR data enabled us to detect more
discontinuity surfaces and in greater detail.
The volume involved in the blast was calculated using two methods: 1) the construction of
simplified surfaces enclosing the rock mass using the software Gocad (Paradigm -
www.pdgm.com) and 2) the comparison of pre and post blast models to detect the differences
with the software PolyWorks (InnovMetric - www.innovmetric.com). The advantage of the
simplified surfaces method is the ability to compute the volume of current stable blocks that
may be part of future landslides. In contrast, the comparison method is faster and more
accurate than the first method, but the need to remove the vegetation could increase the error
of the computed volume in some areas.
Although the two techniques provided similar results, the differences between the data
acquisition and processing are significant. Terrestrial LiDAR data are more accurate and the
process to construct the model is shorter.. By contrast, photogrammetric technique is cheaper
and the data acquisition in the field is faster. Another aspect that affects the two techniques is
the need to have outcrops without or with sparse vegetation to obtain optimum results. |
|
|
|
|
|
|