Agree, useful: Photogrammetric engineering and remote sensing:: torrent download
Photogrammetric engineering and remote sensing:: torrent download | 286 |
Photogrammetric engineering and remote sensing:: torrent download | 709 |
Photogrammetric engineering and remote sensing:: torrent download | 530 |
Photogrammetric engineering and remote sensing:: torrent download | 138 |
Modelling the dynamics of a large rock landslide in the Dolomites (eastern Italian Alps) using multi-temporal DEMs.
Results
DoD analysis
Starting from the integrated LiDAR and pre-event cartography pre-event and LiDAR post-event DEMs of the Le Laste landslide, an elevation change distribution (Fig. 5) and a DoD map (Fig. 6) were computed and analysed in order to highlight erosion and deposition areas and to estimate their respective volume. A second DoD analysis was based on the 5 m pre-event DEM derived from technical cartography and the LiDAR-derived post-event DEM (Figs. 7 and and8).8). The main results of the two DoD analyses are listed in Tables 1 and and22.
Areal (A) and volumetric (B) distributions of deposition (blue) and erosion (red). Values that were discarded during the DoD analysis are shown in grey.
Erosion (red) and deposition (blue) patterns were computed from integrated pre-event LiDAR and technical cartography and post-event LiDAR data.
Areal (A) and volumetric (V) distributions of deposition (blue) and erosion (red). Values that were discarded during the DoD analysis are shown in grey.
In blue deposition and in red erosion.
Table 1
Attribute | Raw | Thresholded DoD estimate | ||
---|---|---|---|---|
Areal | ||||
Total area of erosion (m2) | 179,294 | 150,156 | ||
Total area of deposition (m2) | 248,911 | 199,748 | ||
Volumetric | ±Error volume | % Error | ||
Total volume of erosion (m3) | 683,789 | 677,456 | ±67,570 | 10% |
Total volume of deposition (m3) | 813,820 | 801,952 | ±89,887 | 11% |
Total volume of difference (m3) | 1,497,608 | 1,479,406 | ±157,457 | 11% |
Total net volume difference (m3) | 130,031 | 124,496 | ±112,452 | 90% |
Percentages (by volume) | ||||
Percent erosion | 46% | |||
Percent deposition | 54% | |||
Percent imbalance (departure from equilibrium) | 4% |
Table 2
Attribute | Raw | Thresholded DoD estimate | ||
---|---|---|---|---|
Areal | ||||
Total area of erosion (m2) | 178,325 | 154,675 | ||
Total area of deposition (m2) | 202,800 | 179,650 | ||
Volumetric | ±Error volume | % Error | ||
Total volume of erosion (m3) | 1,268,862 | 1 256,094 | ±170,143 | 14% |
Total volume of deposition (m3) | 1,077,935 | 1,064,942 | ±197,615 | 19% |
Total volume of difference (m3) | 2 346 797 | 2 321 017 | ±367,758 | 16% |
Total net volume difference (m3) | 190 927 | 191 170 | ±260,768 | −136% |
Percentages (by volume) | ||||
Percent erosion | 54% | |||
Percent deposition | 46% | |||
Percent imbalance (departure from equilibrium) | −4% |
In the first analysis, erosion and deposition volumes related to the landslide deviate by 4% from one another, which accounts for a net volume difference of 125,000 m3. However, it has to be noted that a high error of ±112,452 is associated with the total net volume. The original rockslide that detached from the summit of Mt. Antelao had an initial failure volume of 365,000 ± 13,463 m3. During its runout, the slide progressively transformed into a debris avalanche and entrained more material, increasing the total erosion volume to about 680,000 m3.
The travel path of the Le Laste landslide stretches over more than 2,000 m horizontally and about 1,600 m vertically. From the DoD map and aerial photographs (see Fig. 2), five discrete zones can be recognised. They are divided into: (1) The initial failure zone at the summit of Mt. Antelao, (2) a subvertical wall that connects the summit and the Antrimoia Valley, (3) the “Upper Deposit”, filling the Antrimoia Valley, (4) a rocky step, which separates the upper from the lower deposit, and (5) the “Lower Deposit”, which was accumulated along the Ru de Salvela.
The initial rockslide detached as one coherent block from Mt. Antelao. After sliding off the initial failure zone, between 2,800 and 3,100 m a.s.l., the block fell about 600 m down a 53° inclined slope onto the Antrimoia Valley. This fall caused the rockslide to break up completely and changed its runout behaviour into that of a debris avalanche. Before entering the Antrimoia Valley, an additional 255,000 m3 of material were entrained, while about 85,000 m3 were deposited along the path. The resultant debris avalanche that entered the Antrimoia Valley had a volume of 535,000 m3.
According to the DoD analysis, approximately 630,000 m3 of sediment accumulated within the Antrimoia Valley, forming the “Upper Deposit”. The debris avalanche filled a small depression right at the foot of the subvertical wall, before spreading out along the entire valley. This caused local erosion (35,000 m3) along the 32° inclined track, especially along the margins of the small depression in the upper part of the valley. After moving over the high gradient step which separates the valley from the Ru de Salvela, the material accumulated along the 18° inclined Ru de Salvela valley and formed the “Lower Deposit”. The “Lower Deposit” constitutes a mass of 80,000 m3, adding to the total deposit volume of over 800,000 m3. However, this data needs to be evaluated taking into consideration the debris flow that occurred on August 4, 2015. The debris flow initiated from the Antrimoia Valley and mobilised a volume of about 50,000 m3. Along its routing, it eroded parts of the deposit in the Ru Salvela creek and caused some fatalities downstream (Gregoretti et al. in press).
The comparison between the LiDAR-based DoD and “cartographic” DoD highlights a general overestimation of both erosion and deposition volumes in case of the “cartographic” DoD. A total erosion volume of 1,255,000 m3 and a total deposition volume of 1,065,000 m3 were computed with the “cartographic” DoD analysis. Although the imbalance between erosion and deposition is the same in both DoDs (i.e., ±4%), they have opposing signs. In the first analysis, the volume of the deposited material is higher than the volume of the eroded material. It is the other way around in the second analysis.
This could be linked to both, the errors related directly to the quality of the technical cartography and LiDAR data, and also to the fact that the scree slopes and the deposit areas in the Antrimoia Valley are subject to mass movements that may have likely occurred between 2007 and 2011.
DAN3D modelling
To describe the runout dynamics of the Le Laste landslide a frictional rheology was selected for the numerical model during the calibration process:
-
-