Elsevier

Quaternary Geochronology

Volume 61, February 2021, 101132
Quaternary Geochronology

Spatial dendrogeomorphic sampling based on the specific tree growth responses induced by the landslide mechanism

https://doi.org/10.1016/j.quageo.2020.101132Get rights and content

Abstract

The knowledge of spatial distribution of landslide events is crucial from forest management, landscape planning or hazard assessment point of view. Dendrogeomorphic approaches are actually a suitable tool for construction of past landslide behaviour due to known position of sampled trees. The position of trees on the landslide surface can influence the detection of the spatial patterns of landslide reactivation and considerably influence the amount and intensity of recorded landslide signals within tree ring series. This fact is caused by the presence of various morphological zones with different movement mechanisms that create landslide bodies.

This study used tree ring data from 1030 tree ring series coming from 515 trees (Picea abies (L.) Karst.: 482; Larix decidua L.: 33) occupying ten different complex landslides in the Outer Western Carpathians to test the spatial distribution of landslide signals during reactivation events. Next, specific effects of six different landslide morphological zones on tree growth was studied to detect the most valuable sites for future sampling on landslide surfaces. The analysis of spatial patterns can be used as a rigorous indicator for landslide events in the case of a limited number of tree ring signals. Thus, the spatial patterns of landslide reactivations were tested using the Moran index. Moreover, this analysis is suitable only for block-type landslides (not for flow-like landslides). Next, this study presents a general model of the movement mechanism effect on tree growth in individual morphological zones creating complex landslide surfaces. Finally, based on the obtained results, recommendations for future sampling strategies regarding the position of trees within specific morphological zones are presented. Generally, trees on subhorizontal landslide blocks or in the zone of plastic shallow movement should be preferred in contrast to trees on the steep parts of landslides or in the source zone, which should be avoided.

Introduction

Landslides are among the most dangerous geomorphic processes worldwide (Guzzetti et al., 1999; Geertsema et al., 2006; Pánek et al., 2008; Grahne and Jaldell, 2017). As important natural hazard, they annually cause severe damage to human infrastructure or even fatalities (Petley, 2012; Froude and Petley, 2018). The knowledge of past landslide activity is one of the most important aspects of landslide research (Pánek, 2015). Data about temporal landslide behaviour are crucial for assessing their triggers (Lopez-Saez et al., 2012a) or magnitude-frequency ratio (Corominas and Moya, 2010). The crucial aspect regarding current and future climate changes is the detailed knowledge of the relationship between landslide activations and causing meteorological triggers. This relationship can be effectively reconstructed using data about historical landslide activity (Lopez Saez et al., 2012a). Combination of climatic forecast models with knowledge of the past landslide-climate relationships is a key for prediction future landslide activity (Stoffel and Huggel, 2012). Data about the spatial distribution of landslide movements are crucial from a forest management or urban planning point of view (Martire et al., 2012). The reactivation of partial, spatially limited zones is typical for the behaviour of large complex landslides (Cockburn et al., 2016). Thus, the general view of landslide activity can be considerably influenced by the various behaviours of (in)active zones. From a risk assessment point of view, knowledge about the most active zones is crucial, particularly if accompanied by temporal characteristics of the movements (Raška et al., 2015). Data about the temporal aspects of past landslide activity are generally rare (Lopez Saez et al., 2012a), and detailed knowledge about past spatial landslide behaviour is even more rare and more difficult to obtain than knowledge about the simple chronology of past landslide events. The creation of a detailed spatial reconstruction of landslide activity is very limited if historical monitoring is missing. Historical orthophotos can serve as a possible data source, but their availability, chronological range or quality is usually limited (Walstra et al., 2007). Archives and newspaper reports are usually very incomplete and limited to only temporal information (Raška et al., 2015).

Dendrogeomorphic methods are good alternatives for the long-term monitoring of landslide activity (Alestalo, 1971). Past landslide behaviour can be reconstructed with good precision (Lopez Saez et al., 2012a; Šilhán, 2019) based on data from the tree ring series of trees occupying landslide surfaces. Moreover, the length of reconstruction can reach several centuries up to a millennium (Šilhán et al., 2015; Zhang et al., 2019). Trees can provide high-resolution temporal and spatial information via their positions on the landslide surface (Corona et al., 2014). Several methods of using tree position data for spatial landslide reconstruction have been employed in the past. The simplest and most frequently used method is the realization of “event-response maps”. The positions of all trees containing landslide signals in a particular year are highlighted on the map (e.g., Stefanini, 2004; Wiktorowski et al., 2017). This simple concept was extended by Lopez Saez et al. (2012b), who tested the spatial position of affected trees in the case of low indicators of landslide activity using the Moran index (Moran, 1950). The mentioned approach assumes that the spatial clustering of disturbed trees is affected by landslide reactivation (i.e., secondary movement of the spatially limited area sometime after the initial landslide). As noted, this approach was used only for event years with a limited number of trees with tree ring signals. Thus, the validity of the theory of disturbed tree clustering during landslide movements should be tested for all landslide events (i.e., even for events detected in the tree rings of a high number of trees). The next possibility of past spatial landslide reconstruction is the interpolation of recurrence intervals detected between two tree ring signals in each tree (Šilhán, 2016). This approach can be used to obtain information about landslide activity even from areas not covered by trees. The most developed approach for the tree ring-based reconstruction of landslide spatial activity is the creation of probability maps, as introduced by Lopez Saez et al. (2012b), who combined the spatial positions of trees with the calculated frequency of landslide events processed with a Poisson probability model.

The spatial reconstruction of landslide movements has been performed for various types (rotational, translational, shallow, etc.) of landslides in the past. However, landslide areas often express complex geomorphology containing various types of morphological zones (Hungr et al., 2013). Each landslide type or morphological zone with a specific movement mechanism is able to induce different growth disturbances with different intensities in the tree ring series of affected trees (Bollati et al., 2012). For example, block-type movements with rotational mechanisms usually generate reaction wood in coniferous trees (Šilhán et al., 2013), whereas shallow translational movements are typical of inducing abrupt growth suppression in tree ring series (Lopez Saez et al., 2012b). Spatially limited area or specific morphological zone within the entire complex landslide area is usually activated during the landslide events (Guida et al., 2008). As different movement mechanisms in various zones will most likely impact individual growth behaviour in different ways, one can suppose that trees from different zones will variously participate in the resulting total landslide chronology. Thus, trees from some types of landslide morphological zones are able to provide more valuable data about past landslide activity than trees from other zones. The spatial reconstruction of past landslide activity should reflect the specific morphology of the landslide area. Simple spatial interpolation of point values for the entire landslide area seems to be unsuitable, particularly for complex landslides. Some studies suggest this effect and recommend the analysis of various landslide morphological zones separately (Van Den Eeckhaut et al., 2009; Šilhán et al., 2012, 2014; Lopez-Saez et al., 2017), but none have exactly verified, quantified and studied it in detail.

Regarding the abovementioned limitations and gaps in the dendrogeomorphic reconstruction of past landslide behaviour, the aims of this study are i) to test the theory of disturbed tree clustering when affected by landslide movements, ii) to distinguish types of tree growth disturbances induced by various movement mechanisms in different morphological zones of landslides and iii) to detect trees from which morphological zones are able to provide the most valuable data for the final spatiotemporal reconstruction of landslide movements. Toincrease the robustness of the obtained results, the mentioned study aims were adressed for a group of ten individual landslides in the Outer Western Carpathians.

Section snippets

Study area

All of the studied landslides are located in the Hostýnsko-vsetínská hornatina Mts. (centred at 49.4° N and 18.0° E; Fig. 1), a region with a known high frequency of slope deformations, particularly landslides (Kirchner and Krejčí, 1998). The region with a prevalence of studied landslides covers approximately 300 km2 (approximately 30 × 10 km). Its relief has typical highland to mountain character, with the highest peaks reaching 1000 m a.s.l., steep slopes reaching up to 20° and mean

Geomorphic mapping and defining the types of morphological zones

Each studied landslide was geomorphically mapped (1:500) using LiDAR (light detection and ranging)-based digital elevation model (DEM) and orthophotos as the first step. The positions of all trees expressing signs of past landslide activity (tilted or bent stems, stretched or damaged tree roots) were recorded using a GNSS (Global Navigation Satellite System) device and specified using orthophotos. Field geomorphic observation and analysis of the resulting geomorphic map were used to determine

Morphology of the studied landslides

All landslides expressed similar morphological structures and spatial distributions of individual morphological zones (Fig. 4). The zone of landslide blocks was always situated just below the source zone. The only exception was the Vlčice landslide, whose entire area was dominated by shallow movements. The zone of shallow movements followed the zone of landslide blocks in the lower parts of the landslide and was absent in the case of the Soláň landslide. The zone of frontal lobes was mapped in

Discussion

Tree ring-based data about past landslide behaviour were collected from ten individual landslides in the Outer Western Carpathians. This dataset was used for detailed analysis of the spatial distribution of movement signals in trees regarding their positions in six different morphological zones of landslides. The analysis of 560 signals of past landslide activity (tree growth disturbances) coming from the 1030 tree ring series of 515 disturbed trees led to the detection of 110 historical

Conclusions

This study analysed various growth responses and the spatial aspects of sampled trees within the different morphological zones of landslides. Data from 1030 increment cores from ten different complex landslides were studied in detail to increase the robustness and universality of the obtained results.

The testing of the spatial patterns of 515 studied trees on all landslides provided evidence of a limited tendency for the clustering of disturbed trees during individual reactivation events.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The Czech Science Foundation project 19-01866S and the Specific Research Project of the University of Ostrava Nr. SGS02/PřF/2019–2020 supported this study. The language was reviewed by American Journal Experts.

References (56)

  • J.D. Phillips et al.

    Domination of hillslope denudation by tree uprooting in an old-growth forest

    Geomorphology

    (2017)
  • H. Přecechtělová et al.

    Tree ring based estimation of the landslide movement velocity: evaluation of new experimental approach

    Geomorphology

    (2019)
  • J.F. Shroder

    Dendrogeomorphological analysis of mass movement on table Cliffs plateau, Utah

    Quat. Res.

    (1978)
  • M.C. Stefanini

    Spatio-temporal analysis of a complex landslide in the Northern Apennines (Italy) by means of dendrochronology

    Geomorphology

    (2004)
  • M. Stoffel et al.

    Mass movements and tree rings: a guide to dendrogeomorphic field sampling and dating

    Geomorphology

    (2013)
  • P. Šamonil et al.

    The role of tree uprooting in Cambisol development

    Geoderma

    (2010)
  • K. Šilhán

    How different are the results acquired from mathematical and subjective methods in dendrogeomorphology? Insights from landslide movements

    Geomorphology

    (2016)
  • K. Šilhán

    Tree ring eccentricity in the dendrogeomorphic analysis of landslides - a comparative study

    Catena

    (2019)
  • K. Šilhán et al.

    Tree-ring analysis in the reconstruction of slope instabilities associated with earthquakes and precipitation (the Crimean Mountains, Ukraine)

    Geomorphology

    (2012)
  • K. Šilhán et al.

    The dating of bedrock landslide reactivations using dendrogeomorphic techniques: the Mazák landslide, Outer Western Carpathians (Czech Republic)

    Catena

    (2013)
  • K. Šilhán et al.

    Spatio-temporal patterns of recurrent slope instabilities affecting undercut slopes in flysch: a dendrogeomorphic approach using broad-leaved trees

    Geomorphology

    (2014)
  • K. Šilhán et al.

    Regional, tree-ring based chronology of landslides in the Outer Western Carpathians

    Geomorphology

    (2018)
  • Th.W.J. Van Asch et al.

    Problems in predicting the mobility of slow-moving landslides

    Eng. Geol.

    (2007)
  • S. Wang et al.

    Creep properties of clastic soil in a reactivated slow-moving landslide in the Three Gorges Reservoir Region, China

    Eng. Geol.

    (2020)
  • Y. Zhang et al.

    Centennial-scale process activity in a complex landslide body in the Qilian Mountains, northeast Tibetan Plateau, China

    Catena

    (2019)
  • J. Alestalo

    Dendrochronological interpretation of geomorphic processes

    Fennia

    (1971)
  • I. Baroň et al.

    Structure and dynamics of deep-seated slope failures in the Magura flysch nappe, outer western Carpathians (Czech Republic)

    Nat. Hazards Earth Syst. Sci.

    (2004)
  • J.M.H. Cockburn et al.

    Tree-ring evidence linking late twentieth century changes in precipitation to slope instability, central New York state, USA

    Phys. Geogr.

    (2016)
  • Cited by (0)

    View full text