3-D visualisation of palaeoseismic trench stratigraphy and trench logging using terrestrial remote sensing and GPR – a multiparametric interpretation

Abstract

Abstract. Two normal faults on the island of Crete and mainland Greece were studied to test an innovative workflow with the goal of obtaining a more objective palaeoseismic trench log, and a 3-D view of the sedimentary architecture within the trench walls. Sedimentary feature geometries in palaeoseismic trenches are related to palaeoearthquake magnitudes which are used in seismic hazard assessments. If the geometry of these sedimentary features can be more representatively measured, seismic hazard assessments can be improved. In this study more representative measurements of sedimentary features are achieved by combining classical palaeoseismic trenching techniques with multispectral approaches. A conventional trench log was firstly compared to results of ISO (iterative self-organising) cluster analysis of a true colour photomosaic representing the spectrum of visible light. Photomosaic acquisition disadvantages (e.g. illumination) were addressed by complementing the data set with active near-infrared backscatter signal image from tLiDAR measurements. The multispectral analysis shows that distinct layers can be identified and it compares well with the conventional trench log. According to this, a distinction of adjacent stratigraphic units was enabled by their particular multispectral composition signature. Based on the trench log, a 3-D interpretation of attached 2-D ground-penetrating radar (GPR) profiles collected on the vertical trench wall was then possible. This is highly beneficial for measuring representative layer thicknesses, displacements, and geometries at depth within the trench wall. Thus, misinterpretation due to cutting effects is minimised. This manuscript combines multiparametric approaches and shows (i) how a 3-D visualisation of palaeoseismic trench stratigraphy and logging can be accomplished by combining t-LiDAR and GPR techniques, and (ii) how a multispectral digital analysis can offer additional advantages to interpret palaeoseismic and stratigraphic data. The multispectral data sets are stored allowing unbiased input for future (re)investigations.

The Seismic History of the Pisia Fault (Eastern Corinth Rift, Greece) From Fault Plane Weathering Features and Cosmogenic36Cl Dating

“…The backscatter signal intensity provides information on a mixed range of surface properties. These include moisture, roughness, the range between sensor and target, and the angle of incidence (Schneiderwind et al, ; Wiatr et al, ). The backscatter intensities were treated by hierarchical unsupervised clustering, meaning that all data points are assigned to existing cluster centers during each iteration and that the new means are then recalculated for every class (ISO, iterative self‐organizing clustering).…”

The seismic history of the Pisia fault (eastern Corinth rift, Greece) from fault plane weathering features and cosmogenic 36Cl dating

“…Active normal faults in carbonates are frequently preserved as bedrock fault scarps, particularly in the Mediterranean region. Their exhumation occurs periodically during cumulative surface rupturing earthquakes generating free-faces that are several meters high [e.g., Bosi, 1975;Stewart and Hancock, 1990;Armijo et al, 1992;Benedetti et al, 2002;Papanikolaou et al, 2005;Mason et al, 2016Mason et al, , 2017. Recently in the Italian Apennines, the M w 6.0 Amatrice earthquake (24 th August 2016) and the M w 6.5 Norcia earthquake (30 th October 2016) exhumed an additional portion of up to 15-20 cm and up to 1-2 m of the respective fault planes [Livio et al, 2017;Pucci et al, 2017;Pizzi et al, 2017].…”

“…In the past years such stripes were identified by color and roughness contrasts using different methods such as visual findings [Wallace, 1984;Caputo et al, 2004;Mildon et al, 2016], in-situ micro-roughness measurements [Stewart, 1996], photographic studies [Giaccio et al, 2003], rare-earth-element analysis [Carcaillet et al, 2008], and terrestrial laser scanning (TLS) [Wei et al, 2013;Wiatr et al, 2015;He et al, 2016]. TLS in particular has become an established tool to analyze fault scarps since it provides high spatial and temporal resolution data [e.g., Jones et al, 2009;Wilkinson et al, 2015, Mason et al, 2016Cowie et al, 2017].…”

“…A ~13 m high fault scarp is exposed for 25 m along strike, whose geometry is revealed from topographic profiles. The profiles show all the typical features of bedrock fault scarps, with the hanging wall being the shallowest slope, followed by a steep free-face, a shallower degraded upper scarp and a footwall that is slightly steeper than the hanging wall ( Figure 3) [compare to Stewart andHancock 1990b, 1991;Papanikolaou et al, 2005;Mason et al, 2016;Cowie et al, 2017].…”

“…These include moisture, roughness, the range between sensor and target, and the angle of incidence [Wiatr et al, 2015;Schneiderwind et al, 2016]. The backscatter intensities were treated by hierarchical unsupervised clustering, meaning that all data points are assigned to existing cluster centers during each iteration, and that the new means are then recalculated for every class (ISO, iterative self-organizing clustering).…”