Landslide detection with ALOS-2/PALSAR-2 data using convolutional neural networks: a case study of 2018 Hokkaido Eastern Iburi earthquake

Konishi, Tomohisa; Suga, Yuki

doi:10.1117/12.2531695

Cited by 7 publications

(9 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Snow is a potential source of error when using SAR in landslide detection. There is a strong difference in SAR backscattering properties between wet snow, dry snow and no snow (Koskinen et al, 1997). As such, decreased coherence can be caused by snowmelt, drift or fall between image acquisitions.…”

Section: Snowmentioning

confidence: 99%

“…However, with the exception of Mondini et al (2019), who used a global selection of landslides, these studies are generally tested on a single landslide event and use a single radar sensor. For example, Aimaiti et al (2019), Konishi and Suga (2019), Jung and Yun (2019), and Yamaguchi et al (2019) tested their methods using ALOS-2 imagery of the 2018 Hokkaido earthquake. If such methods are to be applied in future events, wider testing is needed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A systematic exploration of satellite radar coherence methods for rapid landslide detection

Burrows

Walters

Milledge

et al. 2020

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Emergency responders require information on the distribution of triggered landslides within 2 weeks of an earthquake or storm. Useable satellite radar imagery is acquired within days of any such event worldwide. Recently, several landslide detection methods that use these data have been developed, but testing of these methods has been limited in each case to a single event and satellite sensor. Here we systematically test five methods using ALOS-2 and Sentinel-1 data across four triggering earthquakes. The best-performing method was dependent on the satellite sensor. For three of our four case study events, an initial ALOS-2 image was acquired within 2 weeks, and with these data, co-event coherence loss (CECL) is the best-performing method. Using a single post-event Sentinel-1 image, the best-performing method was the boxcar–sibling (Bx–S) method. We also present three new methods which incorporate a second post-event image. While the waiting time for this second post-event image is disadvantageous for emergency response, these methods perform more consistently and on average 10 % better across event and sensor type than the boxcar–sibling and CECL methods. Thus, our results demonstrate that useful landslide density information can be generated on the timescale of emergency response and allow us to make recommendations on the best method based on the availability and latency of post-event radar data.

show abstract

Section: Snowmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A systematic exploration of satellite radar coherence methods for rapid landslide detection

Burrows

Walters

Milledge

et al. 2020

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…The second is to estimate landslide locations based on their signal in satellite synthetic aperture radar (SAR) data, which can be acquired through cloud cover and so is often able to provide more complete spatial coverage than optical satellite imagery in the critical 2-week response window (e.g. Aimaiti et al, 2019;Burrows et al, 2019Burrows et al, , 2020Jung and Yun, 2019;Konishi and Suga, 2019;Mondini et al, 2019Mondini et al, , 2021.…”

Section: Introductionmentioning

confidence: 99%

“…ground shaking estimates, and a model is produced that predicts landslide likelihood based on these inputs. A range of methods have been used to generate landslide susceptibility models, including fuzzy logic (Kirschbaum and Stanley, 2018;Kritikos et al, 2015;Robinson et al, 2017), logistic regression (Cui et al, 2020;Nowicki Jessee et al, 2018;Tanyas et al, 2019) and random forests (Catani et al, 2013;Chen et al, 2017;Fan et al, 2020). When generating a susceptibility map for emergency response, the training dataset can be either a collection of landslide inventories triggered by multiple earthquakes worldwide (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…When generating a susceptibility map for emergency response, the training dataset can be either a collection of landslide inventories triggered by multiple earthquakes worldwide (e.g. Kritikos et al, 2015;Nowicki Jessee et al, 2018;Tanyas et al, 2019) or a small sample of the affected area mapped immediately following the earthquake (e.g. Robinson et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Integrating empirical models and satellite radar can improve landslide detection for emergency response

Burrows

Milledge

Walters

et al. 2021

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Information on the spatial distribution of triggered landslides following an earthquake is invaluable to emergency responders. Manual mapping using optical satellite imagery, which is currently the most common method of generating this landslide information, is extremely time consuming and can be disrupted by cloud cover. Empirical models of landslide probability and landslide detection with satellite radar data are two alternative methods of generating information on triggered landslides that overcome these limitations. Here we assess the potential of a combined approach, in which we generate an empirical model of the landslides using data available immediately following the earthquake using the random forest technique and then progressively add landslide indicators derived from Sentinel-1 and ALOS-2 satellite radar data to this model in the order they were acquired following the earthquake. We use three large case study earthquakes and test two model types: first, a model that is trained on a small part of the study area and used to predict the remainder of the landslides and, second, a preliminary global model that is trained on the landslide data from two earthquakes and used to predict the third. We assess model performance using receiver operating characteristic analysis and r2, and we find that the addition of the radar data can considerably improve model performance and robustness within 2 weeks of the earthquake. In particular, we observed a large improvement in model performance when the first ALOS-2 image was added and recommend that these data or similar data from other L-band radar satellites be routinely incorporated in future empirical models.

show abstract

Sentinel-1 P-SBAS data for the update of the state of activity of national landslide inventory maps

et al. 2023

View full text Add to dashboard Cite

The redaction of landslide inventory is a fundamental task for risk management and territorial planning activities. The availability of synthetic aperture radar imagery, especially after the launch of Sentinel-1 mission, enables to systematically update landslide inventories covering wide areas in a reduced time frame and at different scales of analysis. In this work, SAR data processed from the fully automatic P-SBAS pipeline have been adopted to update the Italian national landslide database. Specifically, a matrix has been introduced by comparing past landslide state of activity obtained with Envisat data (2003–2010) and that computed with Sentinel-1 (2014–2018). The state of activity was defined by obtaining the projected velocity along the slope dip direction. The analysis involved about 56,000 landslides which showed at least one Sentinel-1 measurement point, of which 74% were classified as dormant, having annual average velocity < 7 mm/year (considering a value of two times the standard deviation) and 26% as active (mean velocity > 7 mm/year). Furthermore, a landslide reliability matrix was introduced on the landslide inventory updated with S1 data, using the measurement point (MP) density within each landslide and the standard deviation of the mean Vslope value of each landslide. In this case, the analysis revealed that more than 80% of landslides has values of reliability from average to very high. Finally, the 2D horizontal and vertical components were computed to characterize magnitude and direction of every type of landslides included in this work, showing that spreadings, deep-seated gravitation slope deformations, and slow flows showed a main horizontal movement, while complex and translational/rotational slides had more heterogeneity in terms of deformation direction. Hence, the work demonstrated that the application of fast and automatically nationwide Sentinel-1 MTInSAR (multi-temporal interferometry SAR) may provide a fundamental aid for landslide inventory update.

show abstract

Landslide detection with ALOS-2/PALSAR-2 data using convolutional neural networks: a case study of 2018 Hokkaido Eastern Iburi earthquake

Cited by 7 publications

References 9 publications

A systematic exploration of satellite radar coherence methods for rapid landslide detection

A systematic exploration of satellite radar coherence methods for rapid landslide detection

Integrating empirical models and satellite radar can improve landslide detection for emergency response

Sentinel-1 P-SBAS data for the update of the state of activity of national landslide inventory maps

Contact Info

Product

Resources

About