[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ Skip to main content

Advertisement

Springer Nature Link
Log in

Evaluation of machine learning-based algorithms for landslide detection across satellite sensors for the 2019 Cyclone Idai event, Chimanimani District, Zimbabwe

  • Original Paper
  • Published:
Landslides Aims and scope Submit manuscript

Abstract

Changes in climatic patterns, manifested as intensified cyclones and torrential rainfalls in a warming world will inevitably impact the frequency of landslides. One such climatic extreme, Cyclone Idai (March 2019), caused significant havoc across southeastern Africa, including Mozambique, Malawi, and eastern Zimbabwe, by triggering thousands of landslides and widespread concurrent flooding, both of which resulted in substantial loss of life. The study was conducted in the Chimanimani District of eastern Zimbabwe to understand the impact of Cyclone Idai on landslide initiation, quantify the volume of mobilized hillslope material during the event, and compare the event-triggered material release against the annual erosional yield across the study area using machine learning and remote sensing techniques. We evaluated satellite imagery of various resolutions, namely, PlanetScope (3m/px), RapidEye (5m/px), and Sentinel-2 (10m/px) and three machine learning algorithms: artificial neural network (ANN), random forest (RF), and support vector machine (SVM) to identify landslides and compared the efficacy of the models. A total of nine predictor variables derived from the satellite imagery and a 30-m ASTER Global Digital Elevation Model were employed to identify landslides and differentiate them from concurrent hydrological flooding. The models classified the study area into three classes: (i) landslides, (ii) flooding, and (iii) unaffected area. The RF model using PlanetScope satellite data attained the highest prediction accuracy of 97.88%, whereas the accuracy of other machine learning model-satellite data combinations ranged between 94.58 and 97.27%. Subsequently, landslide size thresholds were applied on the initially mapped landslides to eliminate noise and uncertainty from the data before estimating the final Cyclone Idai event-triggered landslide volume. A probability density function, which corresponds to a logarithmic plot of non-cumulative landslide frequency against the mapped landslide area, was employed to calculate landslide size thresholds using divergence and rollover point cutoff values. Finally, a landslide area-to-volume power-law scaling relationship was exploited to derive landslide volumes in the study area that ranged between \(6.8 \times 10^{6}\) to \(14.7 \times 10^{6}\) m3 and \(9.0 \times 10^{6}\) to \(19.6 \times 10^6\) m\(^3\) for divergence and rollover point thresholds, respectively, across the different combinations of machine learning models and satellite sensors. The estimated landslide volume indicates hillslope material liberated by Cyclone Idai was 269 to 345 times greater than the estimated annual average background denudation of the study area computed from a topography-based local erosion model.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
£29.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (United Kingdom)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

References

  • Ahnert F (1970) Functional relationships between denudation, relief, and uplift in large, mid-latitude drainage basins. Am J Sci 268(3):243–263

    Google Scholar 

  • Althuwaynee OF, Pradhan B, Ahmad N (2015) Estimation of rainfall threshold and its use in landslide hazard mapping of Kuala Lumpur metropolitan and surrounding areas. Landslides 12(5):861–875

    Google Scholar 

  • Behling R, Roessner S, Kaufmann H et al (2014) Automated spatiotemporal landslide mapping over large areas using rapideye time series data. Remote Sens 6(9):8026–8055

    Google Scholar 

  • Behling R, Roessner S, Golovko D et al (2016) Derivation of long-term spatiotemporal landslide activity—a multi-sensor time series approach. Remote Sens Environ 186:88–104

    Google Scholar 

  • Borgatti L, Soldati M (2010) Landslides as a geomorphological proxy for climate change: a record from the dolomites (Northern Italy). Geomorphology 120(1–2):56–64

    Google Scholar 

  • Bucknam RC, Coe JA, Chavarría MM et al (2001) Landslides triggered by Hurricane Mitch in Guatemala–Inventory and discussion. US Geological Survey Open File Report 1

  • Button A (1977) Stratigraphic history of the Middle Proterozoic Umkondo Basin in the Chipinga area, Southeastern Rhodesia. University of the Witwatersrand, Economic Geology Research Unit, Johannesburg

  • Campforts B, Shobe CM, Steer P et al (2020) Hylands 1.0: a hybrid landscape evolution model to simulate the impact of landslides and landslide-derived sediment on landscape evolution. Geosci Model Dev 13(9):3863–3886

  • Cardinali M, Galli M, Guzzetti F et al (2006) Rainfall induced landslides in December 2004 in South-Western Umbria, Central Italy: types, extent, damage and risk assessment. Nat Hazard 6(2):237–260

    Google Scholar 

  • Chanza N, Siyongwana PQ, Williams-Bruinders L et al (2020) Closing the gaps in disaster management and response: drawing on local experiences with Cyclone Idai in Chimanimani, Zimbabwe. International Journal of Disaster Risk Science 11(5):655–666

    Google Scholar 

  • Chingombe W, Musarandega H (2021) Understanding the logic of climate change adaptation: unpacking barriers to climate change adaptation by smallholder farmers in Chimanimani District, Zimbabwe. Sustainability 13(7):3773

    Google Scholar 

  • Ciabatta L, Camici S, Brocca L et al (2016) Assessing the impact of climate-change scenarios on landslide occurrence in Umbria Region, Italy. J Hydrol 541:285–295

    Google Scholar 

  • Ciervo F, Rianna G, Mercogliano P et al (2017) Effects of climate change on shallow landslides in a small coastal catchment in southern Italy. Landslides 14(3):1043–1055

    Google Scholar 

  • Crozier MJ (2010) Deciphering the effect of climate change on landslide activity: a review. Geomorphology 124(3–4):260–267

    Google Scholar 

  • Das R, Wegmann K (2021) Landslide mapping using satellite imagery and machine learning algorithms. https://doi.org/10.5281/zenodo.5544698

    Article  Google Scholar 

  • Drǎguţ L, Tiede D, Levick SR (2010) ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data. Int J Geogr Inf Sci 24(6):859–871

    Google Scholar 

  • Foody GM (2010) Assessing the accuracy of land cover change with imperfect ground reference data. Remote Sens Environ 114(10):2271–2285

    Google Scholar 

  • Froude MJ, Petley DN (2018) Global fatal landslide occurrence from 2004 to 2016. Nat Hazard 18(8):2161–2181

    Google Scholar 

  • Gariano SL, Guzzetti F (2016) Landslides in a changing climate. Earth Sci Rev 162:227–252

    Google Scholar 

  • Geertsema M, Highland L, Vaugeouis L (2009) Environmental impact of landslides. In: Sassa K, Canuti P (eds) Landslides-disaster risk reduction. Springer, Berlin, Heidelberg, pp 589–607

    Google Scholar 

  • Ghosh S, van Westen CJ, Carranza EJM et al (2012) Generating event-based landslide maps in a data-scarce Himalayan environment for estimating temporal and magnitude probabilities. Eng Geol 128:49–62

    Google Scholar 

  • Gold R (2004) A comparative study of aerial photographs and lidar imagery for landslide detection in the Puget Lowland, Washington. Tech. rep., Department of Natural Resources, Division of Geology and Earth Resources, Olympia, WA, United States

  • Gorelick N, Hancher M, Dixon M et al (2017) Google earth engine: planetary-scale geospatial analysis for everyone. Remote Sens Environ 202:18–27

    Google Scholar 

  • Guzzetti F, Mondini AC, Cardinali M et al (2012) Landslide inventory maps: new tools for an old problem. Earth Sci Rev 112(1–2):42–66

    Google Scholar 

  • Haque U, Da Silva PF, Devoli G et al (2019) The human cost of global warming: deadly landslides and their triggers (1995–2014). Sci Total Environ 682:673–684

    Google Scholar 

  • Huffman G, Stocker E, Bolvin D et al (2019) GPM IMERG final precipitation l3 half hourly 0.1 degree x 0.1 degree v06. Greenbelt, MD, Goddard Earth Sciences Data and Information Services Center (GES DISC)

  • Huggel C, Clague JJ, Korup O (2012) Is climate change responsible for changing landslide activity in high mountains? Earth Surf Proc Land 37(1):77–91

    Google Scholar 

  • Hungr O (2018) A review of landslide hazard and risk assessment methodology. Landslides and engineered slopes Experience, theory and practice pp 3–27

  • Jaboyedoff M, Oppikofer T, Abellán A et al (2012) Use of lidar in landslide investigations: a review. Nat Hazards 61(1):5–28

    Google Scholar 

  • Kasai M, Ikeda M, Asahina T et al (2009) LiDAR-derived DEM evaluation of deep-seated landslides in a steep and rocky region of Japan. Geomorphology 113(1–2):57–69

    Google Scholar 

  • Köppen W (1884) Die wärmezonen der erde, nach der dauer der heissen, gemässigten und kalten zeit und nach der wirkung der wärme auf die organische welt betrachtet. Meteorol Z 1(21):5–226

    Google Scholar 

  • Korup O (2009) Linking landslides, hillslope erosion, and landscape evolution. Earth Surf Proc Land 34(9):1315–1317

    Google Scholar 

  • Kunedzimwe F, Gandiwa E, Kupika OL et al (2021) Impact of Cyclone Idai on Biodiversity and Natural Resources in Chimanimani District, Zimbabwe. In: Nhamo G, Chikodzi D (eds) Cyclones in Southern Africa. Springer, Cham, pp 229–244

    Google Scholar 

  • Kusangaya S, Warburton ML, Van Garderen EA et al (2014) Impacts of climate change on water resources in Southern Africa: a review. Physics and Chemistry of the Earth, Parts A/B/C 67:47–54

    Google Scholar 

  • LaHusen SR, Duvall AR, Booth AM et al (2020) Rainfall triggers more deep-seated landslides than Cascadia earthquakes in the Oregon Coast Range, USA. Science advances 6(38):eaba6790

  • Larsen IJ, Montgomery DR, Korup O (2010) Landslide erosion controlled by hillslope material. Nat Geosci 3(4):247–251

    Google Scholar 

  • Larsen MC (2012) Landslides and sediment budgets in four watersheds in Eastern Puerto Rico. In: Murphy SF, Stallard RF (eds) Water quality and landscape processes of four watersheds in Eastern Puerto Rico, vol 1789. p 153–178

  • Lee S, Lee MJ (2006) Detecting landslide location using KOMPSAT 1 and its application to landslide-susceptibility mapping at the Gangneung Area, Korea. Adv Space Res 38(10):2261–2271

    Google Scholar 

  • Li Z, Shi W, Lu P et al (2016a) Landslide mapping from aerial photographs using change detection-based Markov random field. Remote Sens Environ 187:76–90

  • Li Z, Shi W, Myint SW et al (2016b) Semi-automated landslide inventory mapping from bitemporal aerial photographs using change detection and level set method. Remote Sens Environ 175:215–230

  • Lu P, Qin Y, Li Z et al (2019) Landslide mapping from multi-sensor data through improved change detection-based Markov random field. Remote Sens Environ 231(111):235

    Google Scholar 

  • Malamud BD, Turcotte DL, Guzzetti F et al (2004) Landslide inventories and their statistical properties. Earth Surf Proc Land 29(6):687–711

    Google Scholar 

  • Malherbe J, Engelbrecht FA, Landman WA (2013) Projected changes in tropical cyclone climatology and landfall in the southwest Indian ocean region under enhanced anthropogenic forcing. Clim Dyn 40(11):2867–2886

    Google Scholar 

  • Martha TR, Kerle N, Jetten V et al (2010) Characterising spectral, spatial and morphometric properties of landslides for semi-automatic detection using object-oriented methods. Geomorphology 116(1–2):24–36

    Google Scholar 

  • Martha TR, Kerle N, Van Westen CJ et al (2012) Object-oriented analysis of multi-temporal panchromatic images for creation of historical landslide inventories. ISPRS J Photogramm Remote Sens 67:105–119

    Google Scholar 

  • Martha TR, van Westen CJ, Kerle N et al (2013) Landslide hazard and risk assessment using semi-automatically created landslide inventories. Geomorphology 184:139–150

    Google Scholar 

  • Mondini A, Guzzetti F, Reichenbach P et al (2011a) Semi-automatic recognition and mapping of rainfall induced shallow landslides using optical satellite images. Remote Sens Environ 115(7):1743–1757

  • Mondini AC, Chang KT, Yin HY (2011b) Combining multiple change detection indices for mapping landslides triggered by typhoons. Geomorphology 134(3-4):440–451

  • Montgomery DR, Brandon MT (2002) Topographic controls on erosion rates in tectonically active mountain ranges. Earth Planet Sci Lett 201(3–4):481–489

    Google Scholar 

  • Mukwakwami J (2012) Structural geology of the Umkondo Group in Eastern Zimbabwe and Geochronology of Associated Mafic Rocks and Possible Correlatives in Zimbabwe

  • Munsaka E, Mudavanhu C, Sakala L et al (2021) When disaster risk management systems fail: the case of cyclone Idai in Chimanimani District, Zimbabwe. Int J Disaster Risk Sci 12(5):689–699

    Google Scholar 

  • Muthige MS, Malherbe J, Englebrecht F et al (2018) Projected changes in tropical cyclones over the south west Indian ocean under different extents of global warming. Environ Res Lett 13(6):065,019

  • Niang I, Ruppel O, Abdrabo M et al (2014) Africa, climate change 2014: impacts, adaptation and vulnerability–contributions of the working Group II to the fifth assessment report of the intergovernmental panel on climate change. 1199–1265

  • Nichol J, Wong MS (2005) Satellite remote sensing for detailed landslide inventories using change detection and image fusion. Int J Remote Sens 26(9):1913–1926

    Google Scholar 

  • Olden JD, Joy MK, Death RG (2004) An accurate comparison of methods for quantifying variable importance in artificial neural networks using simulated data. Ecol Model 178(3–4):389–397

    Google Scholar 

  • Olofsson P, Foody GM, Herold M et al (2014) Good practices for estimating area and assessing accuracy of land change. Remote Sens Environ 148:42–57

    Google Scholar 

  • Parker R, Hancox G, Petley D et al (2015) Spatial distributions of earthquake-induced landslides and hillslope preconditioning in the Northwest South Island, New Zealand. Earth Surface Dynamics 3(4):501–525

    Google Scholar 

  • Pearce AJ, Watson AJ (1986) Effects of earthquake-induced landslides on sediment budget and transport over a 50-yr period. Geology 14(1):52–55

    Google Scholar 

  • Pillay MT, Fitchett JM (2019) Tropical cyclone landfalls south of the Tropic of Capricorn, southwest Indian Ocean. Climate Res 79(1):23–37

    Google Scholar 

  • Planet Team (2017) Planet application program interface: in space for life on earth. Planet Team, San Francisco, CA

  • Sangelantoni L, Gioia E, Marincioni F (2018) Impact of climate change on landslides frequency: the Esino river basin case study (Central Italy). Nat Hazards 93(2):849–884

    Google Scholar 

  • Scheip CM, Wegmann KW (2021) HazMapper: a global open-source natural hazard mapping application in Google Earth Engine. Nat Hazard 21(5):1495–1511

    Google Scholar 

  • Schlögel R, Doubre C, Malet JP et al (2015) Landslide deformation monitoring with ALOS/PALSAR imagery: a D-InSAR geomorphological interpretation method. Geomorphology 231:314–330

    Google Scholar 

  • Schulz WH (2004) Landslides mapped using LIDAR imagery, Seattle, Washington. US Geological Survey Open-File Report 1396(11)

  • Sidle R, Ochiai H (2006) Processes, prediction, and land use. Water resources monograph American Geophysical Union, Washington 18:312

    Google Scholar 

  • Simonett DS (1967) Landslide distribution and earthquakes in the Bavani and Torricelli Mountains, New Guinea. Landform Studies from Australia and New Guinea 64–84. https://cir.nii.ac.jp/crid/1571698600949436032. Accessed 17 Aug 2021

  • Soldati M, Corsini A, Pasuto A (2004) Landslides and climate change in the Italian Dolomites since the Late glacial. Catena 55(2):141–161

    Google Scholar 

  • Stocklmayer V (1978) The geology of the country around Inyanga

  • Stumpf A, Kerle N (2011) Object-oriented mapping of landslides using Random Forests. Remote Sens Environ 115(10):2564–2577

    Google Scholar 

  • Tanyaş H, Allstadt KE, van Westen CJ (2018) An updated method for estimating landslide-event magnitude. Earth Surf Proc Land 43(9):1836–1847

    Google Scholar 

  • Tanyaş H, van Westen CJ, Allstadt KE et al (2019) Factors controlling landslide frequency-area distributions. Earth Surf Proc Land 44(4):900–917

    Google Scholar 

  • Tsai F, Hwang JH, Chen LC et al (2010) Post-disaster assessment of landslides in southern Taiwan after 2009 typhoon Morakot using remote sensing and spatial analysis. Nat Hazard 10(10):2179–2190

    Google Scholar 

  • UNICEF (2019) UNICEF Zimbabwe cyclone Idai situation report. https://reliefweb.int/report/zimbabwe/unicef-zimbabwe-cyclone-idai-situation-report-7-31-may-2019. Accessed 2 Feb 2022

  • UNICEF (2020) One year after cyclone Idai, 2.5 million people in Mozambique remain in need of humanitarian assistance. https://www.unicef.org/press-releases/one-year-after-cyclone-idai-25-million-people-mozambique-remain-need-humanitarian. Accessed 2 Feb 2022

  • Van Den Eeckhaut M, Poesen J, Govers G et al (2007) Characteristics of the size distribution of recent and historical landslides in a populated hilly region. Earth Planet Sci Lett 256(3–4):588–603

    Google Scholar 

  • Warren M (2019) Why Cyclone Idai is one of the Southern Hemisphere’s most devastating storms. https://doi.org/10.1038/d41586-019-00981-6. Accessed 26 Mar 2019

  • Watson R (1969) The geology of the Cashel, Melsetter and Chipinga areas, Government Printer, South Africa

  • Weirich F, Blesius L (2007) Comparison of satellite and air photo based landslide susceptibility maps. Geomorphology 87(4):352–364

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Center for Geospatial Analytics, North Carolina State University, 2800 Faucette Dr, Raleigh, 27695, NC, USA

    Raja Das & Karl W. Wegmann

  2. Department of Marine, Earth, and Atmospheric Sciences, North Carolina State University, 2800 Faucette Dr, Raleigh, 27695, NC, USA

    Karl W. Wegmann

Authors
  1. Raja Das
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Karl W. Wegmann
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Raja Das.

Ethics declarations

Conflict of interest

The authors declare no competing interests.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 42.5 MB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Das, R., Wegmann, K. Evaluation of machine learning-based algorithms for landslide detection across satellite sensors for the 2019 Cyclone Idai event, Chimanimani District, Zimbabwe. Landslides 19, 2965–2981 (2022). https://doi.org/10.1007/s10346-022-01912-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10346-022-01912-9

Keywords

Search

Navigation