Type of the Paper (Article) Special Issue 3D Reconstruction and Data Analysis: Cutting-Edge Techniques for [303969]

Type of the Paper (Article) Special Issue "3D Reconstruction and Data Analysis: Cutting-Edge Techniques for

Terrestrial and Marine Environments"

Digital field mapping and drone‐aided survey for

structural data collection and seismic hazard

assessment: the 2016 Central Italy Earthquakes case.

Daniele Cirillo 1-2,*

1 DiSPUTer Università G. d’Annunzio, via dei Vestini 31, 66100 Chieti, Italy

2 CRUST- Centro InteRUniversitario per l'[anonimizat].

* Correspondence: [anonimizat]; Tel.: +39-0871-355-6389

Received: date; Accepted: date; Published: date

Abstract:

In this work, a high-resolution survey of the coseismic ground ruptures due to 2016, [anonimizat], performed through a dedicated software installed on a [anonimizat] a set of images acquired by drone.

[anonimizat] (central Italy), were affected by surface faulting after the

strongest events of the sequence: the 24 August, Mw 6.0, Amatrice and 30 October, Mw 6.5, Norcia

earthquakes.

[anonimizat], vary.

[anonimizat], or dangerous areas.

[anonimizat],

dip-angles, [anonimizat].

[anonimizat], we conclude that the

structural dataset obtained through remote sensing techniques shows an excellent degree of

reliability.

Keywords: drone-acquired imagery; high-[anonimizat];

compass-clinometer integrated tablet; coseismic ruptures; aerophotogrammetry; 2016 Central Italy

seismic sequence; [anonimizat]-from-Motion.

31

32 1. Introduction

33

The new technologies that replace the 20th-[anonimizat], tablets, iPad, and Drones. [anonimizat], have the performances of a

hand-compass, plus take pictures and act as both a [anonimizat]. They can even be equipped with GIS software. The impressive

technological development characterizing the current times have brought to significant

improvement also in the field mapping techniques. The “classical” [anonimizat], and in some particular contexts by the acquisition of images by drones [1-14]. This work

provides a [anonimizat]-geological mapping which offer

significant advantages compared to the traditional fieldwork. The collected data enable us to

Appl. Sci. 2020, 10, x; doi: FOR PEER REVIEW www.mdpi.com/journal/[anonimizat],

during a [anonimizat] (cliffs or outcrops at the foot of

prominent fault scarp). Furthermore, compared to the traditional methods which employ

compass‐clinometer and field book, digital mapping, performed either with tablet and/or drone,

allow collecting an enormously greater number of data thanks to the fast and continuous acquisition

procedure and the automatic geo‐referencing. As an example, more than 4000 survey data, as type of

rupture, attitude, kinematics, slip vector (obtained from coseismic slickenlines or piercing points),

and displacement (throw, opening, and net displacement), were acquired along the Mt Vettore-Mt

Bove Fault (VBF) over a total length of about 30 km, immediately after the two major seismic events.

The survey was performed at a high resolution, with a sampling rate of at least one site per 50 m, and

locally of three sites per meter. All data reported and discussed in Brozzetti et al. [15], has been

collected through the Fieldmove app installed on iPad dispositive. During the aforementioned

survey, the traditional mapping methods were supported by digital technologies, such as digital

field mapping and drone-aided survey, which turned out to be useful to reduce sensibly the

working time and to collect such massive field datasets.

Furthermore, the use of multiple work tools, even on the same site, has allowed us to compare the

effectiveness of the various applied methods and, at the same time, making cross-checks to test their

reliability.

Here we report on the results of the high-resolution structural mapping of the coseismic ground

deformations due to surface faulting that followed the strongest events of the 2016 Central Italy

seismic sequence, which are the 24 August, Mw 6.0, Amatrice and the 30 October, Mw 6.5, Norcia

earthquakes.

In this work we explored the advantage, the conditions of applicability and the limits of the

proposed novel approach to map and parameterize the coseismic ruptures, hundreds of meters or

kilometers long, occurred along two synthetic splays of the VBF herein referred to as Colli Alti e

Bassi (CAB) and Prate Pala (PTP) faults. The CAB had been known for a long time in published

geological maps thanks to its excellent surface exposure; the second was unveiled by recent

paleoseismological research [16] which highlighted at least three reactivations during the Holocene.

Along these faults, where a detailed survey of coseismic ruptures was available [15], the quality of

the results here obtained through high-resolution aerial photogrammetric images acquired using a

small-size low-cost drone, has been tested and discussed.

75

1.1. Study area

The study areas are located near the village of Castelluccio di Norcia (42°49'42.51"N

13°12'28.00"E) in the southern part of the Sibillini Mts. chain, a deeply studied portion of the

Umbria-Marche thrust and fold belt in Central Italy [17-19]. The stratigraphy of this region includes

a Jurassic-Middle Miocene, shelf to pelagic, carbonatic succession, followed upward by Late

Miocene siliciclastic turbidites.

The contractional structures consist of a system of NW-SE to N-S-striking asymmetric box-folds

characterized by overturned eastern limbs and bounded by east-verging low-angle thrust faults,

Late Miocene-Early Pliocene in age [18,20,21].

Towards E and SE, the outer basal thrust of the chain which, from N-S direction, turns southwards to

NNE-SSW direction, causes the superposition of the Sibillini thrust sheet on the Messinian Laga

basin turbidites.

Since the Early Pleistocene, extensional fault systems displaced the contractional structures

[22-24] with associated offset, along the single faults, locally exceeding 1500 m. Despite the fault

system show a certain geometrical complexity, on the whole, the Quaternary normal faults appear

arranged in a quite a regular, NNW-SSE-striking regional alignment across the Umbria-Marche and

Laga domains.

During the Quaternary, the activity along the major extensional faults, which dip to W-SW, led to

the formation of intra-mountain basins, filled with alluvial and lacustrine deposits [25-27].

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

The easternmost extensional fault, the VBF, is a highly segmented, well-known active structure with a 6.5 ≤ Mw ≤ 6.8 maximum expected magnitude [28] (Figure 1a). It shows an along-strike length of ~30 km, a long-term maximum displacement of ~ 1500 m [21,25,29], and a maximum Late Quaternary slip rate of 1.3mm/y [15].

The Piano Grande basin developed in the VBF hanging wall during the Middle-Late Pleistocene (Figure 1b). Eluvial-colluvial deposits overlay unconformably the carbonate bedrock along the borders of the basin whereas the major depression is mainly filled by alluvial and lacustrine sediments [30-31].

Well before the 2016 Central Italy seismic sequence, the present activity of VBF had been suggested on the base of the observed displacement of Quaternary deposits and Late Pleistocene – Holocene landforms [21,23,25]; subsequently its seismogenic potential was highlighted by paleoseismological evidence [28].

The diffuse surface faulting observed along the VBF after the 2016 sequence (Figure 1b) and the good correlation between the activated fault at the surface and the hypocentral distribution of the seismicity, unquestionably confirmed the seismogenic role of this fault.

Figure 1. (a – top-right inset): location map of the study area in the central Apennines of Italy (a – main figure) Structural sketch of the Mt Vettore-Mt Bove fault (VBF), with epicenters (red stars) and focal mechanism of the three M>5.5 earthquakes of the 2016 seismic sequence, according to Chiaraluce et al [32]; Geological key color – green: Umbria-Marche Domain, brown: Laga Domain, yellow: Quaternary deposits;

(b) Geological map of the study area, after Pierantoni et al. [29], modified, and location map of the two sample

areas (1 and 2) where the Colli Alti e Bassi (CAB) and e Prate Pala (PTP) splays of the VBF crop out; top-left inset shows the stratigraphic scheme of the area extracted from Pierantoni et al. [29]; MAS: Calcare Massiccio; BUG: Bugarone Group; COI: Corniola; RSA: Rosso Ammonitico; POD: Calcari a Posidonia; CDU: Calcari Diasprigni; MAI: Maiolica; FUC: Marne Fucoidi; SBI: Scaglia Bianca; SAA: Scaglia Rossa.

All the three main segments in which the VBF is divided activated during the 2016 seismic sequence; they are referred, from south to north, to as the Vettoretto-Redentore, Bove-Porche and Cupi-Ussita segments [15].

Several synthetic and antithetic splays branch from the three segments at distance up to 4 km west of the main fault trace.

128

129

130

131

132

133

Two of the major splays of the Vettoretto-Redentore segment, located along the eastern border of the Piano Grande di Castelluccio basin, and named hereinafter the Colli Alti e Bassi Fault and Prate Pala fault (CAB and PTP respectively, Figure 1b) were affected by significant coseismic reactivation and have been studied in detail in this work.

Distribution of the Coseismic ruptures

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

The 24 August, Mw 6.0, Amatrice earthquake [32,33] caused the reactivation of the Vettoretto-Redentore segment of the VBF [VRS, Figure 1a-b, 15, 34, 35]. The 26 October (Mw 5.4 and Mw 5.9) and 30 October (Mw 6.5) events continued the reactivation of the VBF causing respectively the rupturing of its northern (Cupi-Ussita) and central (Bove-Porche) segments and reactivating again the Vettoretto-Redentore segment.

Massive field surveys performed by several research groups after the three earthquakes, allowed to reconstruct and compile very detailed databases of the associated primary ruptures [15,36-38].

To date, most authors [15, 39-41] agree that: i) the Mw 6.0 event caused surface faulting, along a ~ 6 km section of the VRS, with coseismic throws reaching 27 cm, without activating any other splay of VBF; ii) the 26 October Mw 5.9 event ruptured the only northern segment of the VBF, (Cupi-Ussita), with throws ranging from 8 to 10 cm; iii) the 30 October (Mw 6.5) mainshock re-activated and ruptured the entire Vettoretto-Redentore and Bove-Porche segment, with a surface rupture length 32>L>22 km and coseismic surface displacements up to 222 cm, and iv) despite most of the 30 October coseismic ruptures occurred along the main fault trace, they partially reactivated also synthetic and antithetic splays within a few kilometers off the main fault trace. On the contrary, no evidence of coseismic rupture was observed after the 24 August event along the subsidiary structures.

Selection of sample areas

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

In the Castelluccio di Norcia area (Figure 1b), two cases of significant coseismic fault-reactivation occurred, after the 30 October mainshock, along the Colli Alti e Bassi (CAB) [(42°50'10.06"N 13°13'46.35"E)] and Prate Pala (PTP) [(42°49'10.70"N 13°13'23.15"E)] normal fault

(Figure 1b). The CAB geologic (i.e. long term) expression was previously recognized and reported in

published geological maps [21, 29].

The PTP was recognized and partially mapped by Galadini and Galli [42] through an aerial photo-geologic survey, integrated by paleoseismological trenching.

Both these reactivated synthetic splays of the VRS, crop out close to the eastern border of the

Piano Grande basin. We selected the sample area 1, straddling the CAB, and the sample area 2, including the PTP ruptured during the 2016 mainshock.

In neither area evidence of ruptures were detected after 24 August and 26 October events (Figure

3a-b-e), thus the total measured displacements can be referred to the 30 October reactivation.

Digital fieldwork carried out on both the sample areas soon after the mainshock, provided coseismic displacement values of up to 101 cm along the CAB (Figure 3 c-d) and of 14 cm in the central part of the PTP (Figure 3 f-g).

169

170

171

172

173

174

175

176

177

178

179

180

181

182

Figure 2; a,b) Sample area 1, along the Colli Alti e Bassi Fault (CAB, location in Figure 1b); the two sketches (a=plan view, b= bird’s eye view) illustrates the mission flight plan for image capturing over the sample area using a manual flight, with an overlap of 80% along the path and 80% in a lateral direction, useful to have a good alignment of images and to reduce the distortions on the resulting orthomosaics;

c,d) Sample area 2, along the Prate Pala Fault; (PTP, location in Figure 1b) the two sketches (c=plan view, d=

bird’s eye view) illustrates the mission flight plan for image capturing that covers the area using north-south flight paths with 65% along the path and 60% in a lateral direction.

Legend: blue point: shot point photo; yellow point: sites of manual and digital mapping; blue line: trace of the CAB (a) and PTP (b) faults; tics indicates the hanging wall block.

Materials and Methods

Field Mapping

183

184

185

186

187

188

189

The high-resolution mapping of the surface coseismic ruptures in the field was carried out integrating the traditional survey, performed using a geologic compass-clinometer for measurement of planar and linear elements (dip azimuth/ dip-angle of fault planes, trend and plunge of slickenlines), with digital mapping technique. The digital mapping was based on a digital compass-clinometer App for portable devices, allowing the direct measurement of georeferenced planar and linear elements accompanied by written notes, sketches, and editable photographs of each survey site (Figure 4). The “Fieldmove” App for smartphones, tablets, and iPad, developed by

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

Midland Valley & Petroleum Experts [43], Glasgow-Edinburgh Scotland, was chosen and installed on an Apple iPad Air2 wifi-cellular (128 GB memory) with an internal A-GPS GLONASS.

Several preliminary trials allowed to verify that the internal sensor of IOS provides reliable and precise azimuthal and dip-angle measures. The reliability-trials consisted of repeated comparisons between measures of the same structure carried out using the digital device and the geological compass-clinometer. This testing experience highlighted that the chosen device and software assured accurate compass readings for both azimuthal and clinometric measurements that not exceed ±2 degrees. For the positioning of the elements measured, an internal GPS sensor A-GPS and GLONASS also integrated with the high-resolution images from the Google Earth™ Pro 7.3 was successfully used. The accuracy of the waypoints acquired depended on satellite coverage. When the maximum satellite coverage reached the optimum values, the error estimated for the automatic location was of 1-2 meters, whereas it reached 3-5 meters in the worst coverage conditions. The solution to minimalize the error location (well below one-meter resolution) was the manual adjusting on the iPad screen, taking into account the easily recognizable elements on the Google Earth base map imagery. A further precaution to avoid positioning inaccuracies was the use, as Fieldmove base maps, of previously acquired high-resolution images, orthorectified through ArcGis, and exported from the Move software in .mbtiles or .geotiff format file.

207

208

209

210

211

212

213

214

215

216

217

Field data were organized in a comprehensive GIS-managed dataset for subsequent elaborations. The used software aside from the Fieldmove App for Apple iPad includes Esri ArcMap 10.x and Move 2019 Petex suite [http://www.petex.com/products/move-suite/]. All the geological-structural data acquired were elaborated within the Move suite software Petroleum Experts see sect.3.1.2.

Figure 3. (a-d): sample area 1 – the same outcrop of the CAB fault scarp shooted at three temporal stages: a) before the beginning of the 2016 seismic sequence, b) after the 24 August and 26 October events but before the 30 October one, and c-d) after the 30 October mainshock (bottom left on each image, date and the hour of photo shooting) ; (e-h): sample area 2 – images at different scales, of the coseismic surface effects of the 30 October mainshock along the PTP fault trace.

218

Georeferenced field images dataset

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

The database of georeferenced images used also for reconstructing virtual outcrops, includes the photographs shooted during the digital survey, using the Fieldmove dedicated tool, and hundreds of digital photographs taken through a compact Sony camera with an integrated GPS sensor. Furthermore, it has been improved by several set of high-resolution aerial photographs acquired through a digital camera transported by a Drone (Figure 4), whose methods of acquisition and processing is reported in the following section 2.3.

Photogrammetric processing led to elaborate two Digital Surface Models (DSM) at centimetric

resolutions for both the study-cases.

For their processing, different software as Agisoft Metashape Professional [https://www.agisoft.com/], Cloud Compare [https://www.danielgm.net/cc/], Esri ArcMap v.10.x, Move 2019 Petex suite [http://www.petex.com/products/move-suite/] were used.

Figure 4. Survey tools used during different fieldwork activities: above: Traditional compass-clinometer and Digital device (Apple iPad Air2) supporting the “Fieldmove” App by Midland Valley & Petroleum Expert, in position for the measurement of a shear plane within a fault breccia; below: Drone model DJI Phantom 3 Professional and details of its components; Sony camera DSC-HX5V with integrated GPS; bottom-right: traditional papery field book.

Remotely acquired images and data

Instruments calibration and pre-drone survey

240

241

242

243

244

245

246

247

The use of cameras for the Structure From Motion (SfM) and the creation of tri-dimensional virtual outcrops is now very easy and fast because the instrumentation at our disposition is very light to transport during the field survey.

The necessity of total stations GPS for the acquisition of more precise and more accurate data positioning is generally required, but in this work, we did not use it. The method used in this research is based mainly on the placement of markers Ground Control Points (mGCPs) at a well-defined known distance during the field survey. The objectives of this work do not take into account the accuracy in terms of positioning of the virtual outcrop in the space (that is in terms of

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

absolute coordinates), but it aims at achieving high-precision in the measurement of the outcrops parameters, as f.i. height of slopes, width, and length of the virtual outcrop, amount of coseismic displacement associated with each part of the rupture. Therefore this research proposes a method of utilizing photogrammetry for creating a georeferenced and scaled virtual outcrop without the use of total stations GPS. The method is based on the use of an mGCPs of known geometries located close to outcrop during the field survey. The markers located on outcrops are used as ground control points and their distances are calculated.

The acquisitions of mGCPs are essential in the pre and post-processing phase, in order to make scale

changes, to position, orientate and rescale the three-dimensional model in the chosen Geographic Reference System. In this work 25 x 25 cm markers were used as mGCPs. They were black and yellow plasticized paper, resistant to damp and rain, fixed with silicon on small wooden boards to obtain major stability and prevent movements due to wind.

During the subsequent analysis through the Agisoft Metashape software, the placement of markers

(see inset Figure 5a) were positioned at the center of each mGCPs surveyed, to optimize precision between the images, and for subsequent rescaling of the virtual outcrop. The contrasting black and yellow rectangles allow make easily recognizable the common points on each acquired photo and also facilitate the automatic location of points belonging to different images.

The dimensions and the double color of mGCPs made them visible from heights up to over 100

meters.

To improve the reliability of the measurement performed on a virtual outcrop, different calibers oriented in the third dimension (zy/zx plane=height, yx plane=width and length), perpendicular to each other were added.

The caliber instruments as hand-ruler 25 meters long, aluminum stadia graduated rod 5m long, and

markers positioned at predetermined distances were used.

This method provided good precision in evaluating the virtual outcrop parameters, but it does not take into account the absolute accuracy in terms of geographical positioning, therefore the whole model could be shifted also by some decimeters. In this work, all the markers were positioned at a distance of 10 meters from each other and arranged in two preferential orthogonal directions North-South and West-East.

In the sample area 1, they were positioned at the intersection points of a 10 m sided square grid

oriented in N-S and E-W directions (Figures. 1b, 2a-b). Overall, in this area, 15 mGCPs, 7 of which in the fault footwall and 8 in the hanging wall, and different calibers were used for scaling the obtained model.

The area 2, characterized by a narrow and elongated shape (1 x 0.13 km), to reduce the 'doming'

effect (eg 10, 12, 13), a total of 20 mGCPs were arranged in the fault footwall, in the hanging wall, and also at corners and central part of the investigated area (Figures 1b, 2c-d).

Drone Survey

The aircraft used to perform the survey was a quadcopter Drone that is generally thought for hobbyist use and accordingly, has a user-friendly control system. It also offers several other functions, such as autopilot and auto-return home in case of loss of signal. This Drone has a flight range of approximately 25 min with a single battery charge and flies up to 3.1 miles (5 km) from the controller. These characteristics are sufficient to cover a relatively small to medium area. The Drone is equipped with an autopilot package that, in the presence of internet connection, follows a pre-programmed flight-path using the appropriate app, enabling the user to choose the number of waypoints, the photo shooting, and the desired flight altitude.

A Drone camera with sensor 1/2.3” CMOS, 12.4 M Effective pixels, lens FOV 94° 20 mm (35 mm format equivalent), Image Size of 4000×3000 was used for the survey, also for more detailed resolution, a camera Sony DSC-HX5V with 10 megapixels and internal GPS sensor was used.

The flight was followed through a remote controller wirelessly connected with an Apple iPad Air 2. The mission program for both the study areas included a detailed flight plan, including the flight direction, number of flight lines, flying altitude, and orientation of the camera respect to the flight

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

path. The program was thought taking into account that the awaited resolution of the post-processing images strongly depends on the overlap, side lap, and ground sampling distance. The overlap (in flying direction) and side lap (between adjacent flight lines) were set considering the width parallel to the flying direction.

In the study area 1, a manually piloted flight-mode was carried out due to the lack of coverage of UMTS signal causing, in turn, the complete absence of internet connection. The manual flight paths were organized into three different acquisition method: a) three aerial strips at three different altitudes: 30, 60 and 120 meters above ground level, shooting the photos perpendicular to the flight direction; b) two aerial strips very close (3 and 5 meters) to the fault plane, with the oblique-oriented camera; c) two aerial strips farther from the previous one, with photos shooted perpendicular to the outcrop (Figure 2a-b). A total of 268 photos were acquired, with a minimum front and side overlapping of 80%. During half of the flight, the camera orientation was maintained perpendicular to the main path whereas, for the second half-flight, a path-parallel arrangement was set. The need to adopt different shooting angles was due to the high slope-angle characterizing the outcrops of this sample area. The flight time required to cover the entire area was 20 minutes.

In the study area 2, the constantly good internet connection, allowed to plan an automatic-mode

guided flight through a dedicated application.

In this case, the flight path was planned to maintain a constant elevation of 70 m above ground level (Figure 2c-d). In the second study area, four aerial sweeps were carried out, leading to the acquisition of 100 photos, with overlapping (both in front and side) ranging from 60% to 65%. During the flight, being this area completely flat, the camera orientation was maintained perpendicular to the main path, and the images were acquired in automatic. The flight time required to cover the entire area was 15 minutes.

324

325

326

327

328

329

330

331

332

333

334

335

Photogrammetric processing

Photogrammetric processing was performed using a Laptop HP Omen 17-an002nl with 3.8 a GHz (in Turbo Boost, 6 MB of cache, 4 core) Intel Core i7-7700HQ, 32 GB RAM (2×16 GB DDR4-2400), NVIDIA GeForce GTX 1070A (8 GB GDDR5) and SSD 512 GB PCIe NVMe M.2, running Agisoft Metashape Professional Software.

The processing with Agisoft Metashape software was necessary to obtain a spatial resolution of ~1

cm/pixels for virtual outcrops reconstruction. Digital photogrammetry, obtained from the sequential drone-acquired overlapping images and through the use of Agisoft Metashape software led to create the following elaborations: a) sparse point clouds, b) dense point clouds, c) meshes and textures, d) 3D models, e) tiled models, f) DEM and orthomosaic model.

Figure 5, shows all the processing steps carried out with Agisoft Metashape software (details in the

caption of Figure 5).

336

337

338

339

340

341

342

343

344

345

346

347

348

Figure 5. workflow of the used digital photogrammetry procedure (a to g are in-sequence steps): (a) overlapping digital aerial photographs and ground control points are imported into the software package; features within the images are automatically detected, matched and used to calculate locations and orientations of camera shooting points; (b) points that have been used for the image alignment form an initial sparse-point cloud (109.802 points); the inclusion of ground control points make the sparse-point cloud automatically georeferenced; (c) point densification (multi-view stereo algorithm): points that were manually classified as representing the ground surface are exported as a dense-point cloud (over 13 Millions points); (d) 3D model construction obtained building meshes between dense-point cloud, with 1,930,896 faces; (e) a tiled model is obtained by interpolations of all image acquired during drone survey with faces of 3D model, the final resolution is 3,75 mm/pixel; (f) Construction of DSM containing all height values through rasterizing the dense-point cloud or from meshes. The obtained Digital Surface Model has a resolution of 1,00 cm/pixel; while (g) the georeferenced orthomosaic with a high resolution of 3,75 mm/pixel was reconstructed interpolating all the

349

350

photos shooted from the drone. Additional datasets were computed for fracture mapping in a Cloud Compare, ArcGIS, and Move Petex suite software.

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

Extraction of fault slip data from Cloud Compare

The Cloud Compare [http://www.danielgm.net/cc/] open-source software allowed to perform structural geological analysis of the surveyed coseismic ruptures, directly on a desktop PC. The generated 3D point cloud, including more than 13 million points (Figure 5c and 6), georeferenced in the previous steps, were used to reconstruct the virtual outcrops and for the subsequent kinematic analysis on the fault planes.

After the processing of the CAB virtual outcrop, the surface of the ~ 15 m long fault mirror was manually extracted from the density point cloud.

Using the Cloud Compare software, and particularly the plugin qFacets [44]

(https://www.cloudcompare.org/doc/wiki/index.php?title=Facets_(plugin)), it was possible to make observable the long-term striae through an “interactive stereogram” (see inset in Figure 6c-d). In fact, by selecting a narrow range of values of dip direction and dip angle, the plugin makes visible only few facets of fault plane (i.e. as the ridges in grooves lineations). This application of interactive filtering, is greatly useful to understand the trend of the lineations on the outcrop in a tridimensional space. After choosing and comparing them with the 30 October coseismic slip vector, measured directly in the field (correlating hanging wall and footwall piercing points), the associated net displacement was assessed. Through a "virtual compass" on Cloud Compare the outcrop orientation was determined and the use of the "lineation tool" allowed the measurement of the height and dip directions of the coseismic slickenlines was performed. Finally, through Cloud Compare software, all the extracted data obtained from the measurements on the virtual outcrop, were exported in .csv format file, as dip, dip-azimuth, trend and plunge, and plotted onto stereonets for structural analysis.

The described steps are shown in Figure 6, from dense-point cloud import (a) to the extraction of the fault plane (b), to the measurement of coseismic elements (total or net displacement, c-g ) until export the database (h) in excel, .csv or .txt format.

376

377

378

379

380

381

382

Figure 6. The main steps carried out in Cloud Compare (CC): (a) dense-points cloud imported into a CC as ascii file; (b) selection of the interest features; (c) contour and selection of all the dense-point cloud around the fault plane and extraction, through the plugin qFacets, of all plunge and dip direction in a stereogram; (d) from the stereogram the dip directions of the striae, evidenced by grooves, are excluded; (e) and (f) picking of net displacement (mm), striae attitude (trend and plunge), dip direction and dip-ange of the fault plane; (g) view of the net displacement and the striae detected on the fault plane; (e) export all data in excel, .csv or .txt format.

383

2.3.5 High-resolution topographic profiling

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

To optimize the quality of the result of our topographic analysis, we used two different software to extrapolate profiles from the Digital Surface Model which led to obtaining 1 cm resolution over area 1 and a resolution of 3 cm in area 2 (Figure 7a-d). In particular, serial profiling across the coseismic ruptures of both CAB and PTP fault was performed through the Esri ArcGis v.10.x “3DAnalyst” and through the “Create Sections” tool of Move 2019. From both the sets of profiles we estimated the fault-scarp height (Figure 7a-d)

To compare our digital field data with those obtained by aero-photogrammetric 3Dmodel, several fault attitudes, along the CAB were manually collected and the height of the scarps measured (Figure 7b).

The “surface control” is essential to estimate the errors between analogical (real outcrop) and digital (virtual outcrop) measurement data.

In the field, a meter stick was used to estimate all along the fault trace, the height of the scarp, associated throw, and net displacement, both on the long term and coseismic free face. The slope dip angle was measured using a clinometer (Figure 7b).

Then, we compared the morphological height, the strike, the dip and dip azimuth of the fault plane,

measured in the field, with those obtained from the virtual outcrop 3D model.

The observed deviation between the height values of the fault scarp directly measured in the field and those extracted from virtual outcrop never exceed 2 centimeters (compare the deviations between the curves and the colored dots of Figure 8b), confirming the reliability of the reconstructed digital topography.

404

405

406

407

408

Figure 7. Georeferenced orthomosaic of CAB. (a) Digital Surface Model (DSM) of CAB and trace of topographic profile (from 1 to 8); (b) manual measuring of the scarp height associated to the CAB; (c) topographic profiles obtained using 3D Analyst extension of ESRI ArcGis; (d) topographic profiles obtained through move software.

409

Results

410

Study area 1 – Colli Alti e Bassi Fault (CAB)

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

Topographic analysis of CAB

The graphs in Figure 8 show a comparison between the measures of the Total Topographic Offset (TTO = blue line), the Height of the preserved Fault Scarp (HFS=red line) and the 30 October 2016 Coseismic Ruptures (net displacement) from Remote Analysis (CRRA= violet line) measured on the CAB in the sample area 1.

The estimated values of these features (on the Y axis, in meters) are plotted against the along-strike distance of each measurement point (abscissa) from the northern tip point of the fault-scarp outcrop (assumed as X= 0).

Dots in the Figure 8 graph mark the values, manually measured on the control points, respectively

for the height of the fault scarp, measured with hand ruler (green dots) and the 30 October 2016 coseismic ruptures (net displacement) from the manual survey (yellow dots). The good agreement between the field data and the assessments obtained from the topographic analysis on the virtual outcrop, confirms the reliability of the adopted method.

The TTO and HFS curves show very similar trends but differ in several points because the former

includes the part of the scarp removed by erosion.

Interestingly, the CRRA curve, even if characterized by lower values, (higher observed net slip = 1.25

m) shows a good correlation with TTO and HFS curves, as regards the position of the maxima and minima. This remarkable consistence suggests that the along-strike height profile of the current fault scarp derives from the add up of a certain number of coseismic displacements, each characterized by slip-profiles similar to the one caused by the mainshock of the 2016 sequence.

433

434

435

436

437

Figure 8. (a) Outcrop image of the CAB fault; (b) Comparison of different types of topographic profiles; blue line: height of long term slope; red line: effective present fault scarp; violet: coseismic net displacement. Both along-strike distance (abscissa) and height values (ordinate) are expressed in meters. Colored dots refer to field measurements collected at control points.

438

439

440

441

442

443

444

445

446

447

448

449

450

451

Structural and kinematics analysis of the CAB

During the fieldwork, several attitudes of the CAB were measured as a dip and dip azimuth, using either a traditional compass-clinometer and a Fieldmove-equipped Apple iPad. After data collection, the whole dataset was managed in a georeferenced digital database.

Subsequently, we extracted attitude and kinematic data from the virtual outcrop reconstructed through remotely acquired images (see sec. 2.3.4.)

A comparison between the three differently attitude dataset is shown in the plots of Figure 9.

Figure 9. Comparison between the fault attitude and striations trends measured through high-resolution sampling on the CAB using traditional compass-clinometer (a,d), Fieldmove App on a digital device (b,e), and virtual outcrop reconstructed from drone-acquired imagery (c,f). The upper plots (a-c) shows the fault dip-azimuth; the lower plots (d-f) shows the trend of associated striae; for each rose diagram the mean value resulting from the statistical analysis of the dataset is reported in the bottom-right rectangle Rose diagrams were extracted through Move 2019 software and redrawn through Corel Draw X4 software.

452

453

454

455

456

457

458

459

460

The hand-compass collected dataset shows that on six measurements on the fault plane, the trend of dip-azimuth is variable between N220 and N260, with a mean resultant plane orientation of dip-azimuth at N237 and dip-angle of 68 degrees (Figure 9a). In the same points, the measurements with the Fieldmove app show a dip-azimuth trend on a fault plane, comprised between N220 and N260 with a mean resultant plane orientation of dip-azimuth at N235 and dip-angle of 70 degrees (Figure 9b). While the results obtained from virtual outcrop, clearly with a huge number of data (77 points measurements), shows a main dip-azimuth trend between N220 and N240 and only a few data comprised between N240 and N260; also in this last case mean dip-azimuth of the fault plane orientation shows a trend of N233 and dip-angle of 70 degrees (Figure 9c).

461

462

463

464

465

466

In order to compare the measures of striation trend collected through three different used tools (Figure 9d-f), a number of six data were also acquired using hand-compass and Fieldmove integrated compass. In both cases, the dispersion of data varies in the range between N180 and N240 except for the huge data extracted from the virtual outcrop that shows an attitude between N180 and N220 and the mean direction dipping toward N207, well comparable with data acquired with the previously described tools.

467

468

469

470

471

472

473

As regards the determination of fault attitude data, the good degree of mutual correlability between the differently acquired data is evident and encouraging. In fact they mostly agree in providing an average fault attitude of~ a 235/70 (dip-notation value, compare rose diagrams Figure 9a-c and 9d-f). From such a point of view, the remote sensing survey is clearly advantageous, considering the large number of data collected over a very short time in comparison with both the traditional and digital field mapping techniques.

474

Study area 2 – Prate Pala Fault (PTP)

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

Topographic analysis of PTP

Despite the small offset associated with the 30 October coseismic rupture, the reactivation of the PTP was highlighted soon after the mainshock thanks to the displacement of the paved road (S.P.477) from Forca di Presta to Castelluccio di Norcia (Figure 3g-h). A maximum offset of 14 cm was measured in the field in the central part of the coseismic rupture (Figure 3f) whose total length was estimated in ~1.3 km.

The performed drone-aided mapping, which allowed the detection of tectonic scarps displacing the flat topography of the Castelluccio di Norcia plain, led to reconstruct a 3 cm resolution Digital Surface Model (DSM).

DSM models elaborated using this technique do not permit to remove the vegetation noise.

Fortunately, in the studied cases, this problem was negligible because the periods in which the aerial images were acquired (beginning of November 2016 and afterward in 2018), were preceded by several months of intense grazing with poor grass regrowth, due to very low rainfall.

Further processing of the DSM allowed to explore and analyze through the functions of the ArcGIS tools, the slope analysis, the shaded relief and the aspect of the fault scarp.

A set of 27 topographic profiles, perpendicular to the fault scarp, were drawn to investigate the morpho-structural expression of the PTP (Figure 10) and to reconstruct the along-fault coseismic slip profiles, from north to south.

Interesting is the systematical counter-slope (i.e. dipping towards east, thus opposite to the fault dip)

topography observed west of the PTP trace, suggesting a back tilting of the hanging wall block which is coherent with the normal kinematic of the fault.

This observation, confirmed by the persistence of a long-term topographic scarp resistant to the

intense breeding and plowing practiced in the area, provides further evidence of the present activity of the PTP.

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

Figure 10. Topographic profiles generated using “3D analyst” in the ArcGis platform (Orthomosaic Photo and DSM have a resolution of 3 cm per pixel.

A comparison between the along-strike DSM-derived topographic offset and the coseismic slip profile is shown in Figure 11 .

In the first case (Figure 11a), from the northern edge of PTP, the profile increases with a medium-high slope reaching the maximum height where the topographic scarp is ~2.3 meters. Moving south, the scarp shows a gradual decrease to zero at the southernmost surface tip of the PTP. The observed asymmetric trend of the topographic offset suggests, according to Manighetti [45] and Cappa et al. [46], that, during previous events, surface ground ruptures propagated mainly from north to south. On the contrary, in Figure 11b, the trend of the coseismic slips shows an opposite triangular shape compared to the previous one, culminating at a maximum value of 14 cm (Figure 3f).

The aforesaid works [45] and [46], investigating the distribution of the slip along with the fault segments during the coseismic rupture, interpreted the triangular shape, characterizing both the profiles as due to the dynamics of rupture propagation.

In particular Manighetti et al. [45] show, from the analysis of several normalized displacement

profiles, that the geometry of the coseismic slip profile gives information on the polarity of the rupture propagation and the location of the epicenter from the fault tips and the point of maximum displacement.

Brozzetti et al [15] comparing the 2016 Central Italy coseismic ruptures with the surface faulting of the global earthquakes, agreed with the suggestion of Manighetti et al. [45] showing that the

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

coseismic slip profiles obtained from the field survey have an asymmetric shape (from south to north) for the whole VBF, and also for the single VRS, VBS, and CUS segments of the master fault. Similarly, the here reported data, suggest a polarity of fracture propagation from south to north as shown by the asymmetry of the graphs Figure 11b.

Possibly the difference in the observed polarity of propagation may depend on the hypocenter location and the magnitude of the earthquakes that gave surface faulting.

Moreover, according to Tao et al. [47] and Iezzi et al. [41,48], the aspect of the surface faulting is influenced by fault geometry and in particular by fault segmentation and/or change of direction (bending). It is finally to take into account that this work had the opportunity to analyze only a short section of the PTP fault, whose effective length is reasonably much higher than the portion ruptured at the surface during the 30 October mainshock.

Figure 11. Comparison between the topographic fault scarp (a, in meters) and the coseismic slip profile (b, centimeters) surveyed following the 30 October 2016 earthquake (Mw 6.5) along the PTP.

In Figure 12b,c the measures of coseismic ruptures from hand compass and tablet device are showed whereas figure 12d displays the direction of fault scarp attitude obtained from DSM analysis. Comparing the three rose diagrams we observe that the coseismic ruptures mostly strike in the same direction of the scarp generated from the ancient earthquakes even if, in a few cases, they show oblique directions. The fact that the coseismic ruptures are localized on the long term fault scarp, suggests a preferential growth of the central part of the fault during each seismic event.

The three rose diagrams of Figure 12b-d, reporting on the PTP attitudes measured with the different methods, show a substantial agreement (particularly the b-c ones) but also a certain dispersion. We interpret this latter as due to the remarkable differences in the number of acquired data. In fact, the major deviations occur when the hand compass and Fieldmove collected data are compared with data extracted from virtual outcrops. To notice that the 30 October coseismic ruptures were not measured from DSM because the aerophotogrammetric image resolution was insufficient to detect them. A lower elevation flight would have been necessary, especially in the areas of maximum slip, but unfortunately when this high-precision mission was planned, the area had been erased by agricultural activities, hampering the acquisition of new images useful for making these measurements.

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

Figure 12. Sample area 2: perspective 3D view extracted from Move Software of the PTP fault scarp in the Piano Grande near Castelluccio di Norcia; arrows show the alignment of the coseismic surface ruptures; in orange the fault scarp of PTP is highlighted, black rectangles indicate the location of Figure 3f,g of coseismic ruptures measured during the field survey after 30 October 2016 Mw 6.5 earthquake.

3.3. Accuracy and Precision

The integration of images acquired from a GPS Drone has allowed placing the virtual outcrop in its geographical position, with some limitations in terms of absolute accuracy. Anyhow, for most of structural geology analysis, that need no very-high accuracy in terms of geographical positioning, the described method can be considered satisfactory.

It is emphasized that this approach has to be considered only an integration to the “traditional”

structural geological analysis. From the previous sections it is evident that, in the two sample areas,

the obtained result, from aerophotogrammetrical processing and direct field measurements, are comparable and that their precision is good. The maximum error measured along the fault scarps was close to 1 centimeter in the first area (CAB), and around 1-3 centimeters in the PTP area. Higher accuracy and precision could be obtained using a differential GPS station integrated with the mGCPs and geotagged Drone images. This would improve significantly the virtual outcrop models and their accuracy in terms of geographical positioning (commonly close to few millimeters), but using the GPS total station, cause a considerable lengthening of work time, as well as very high survey costs, a condition which cannot frequently be accepted during routine fieldworks.

There is also to consider that many sites, which may be potentially studied using the proposed

approach, are unreachable with bulky and heavy instrumentation.

In figure 13 the comparison between methods using or not of GPS, calibers, and their combination are shown. In our case studies the low performance of the GPS installed on the drone allowed the correct positioning in a geographical system, but without a centimetric or millimetric accuracy. In any case, the precision of the height, width, and length of virtual outcrop was satisfactory thanks to

579

580

581

582

583

584

585

586

the correct use of caliber instruments. In conclusion, the results here obtained made it possible to create a valid and reliable model.

Figure 13. The synoptic diagram shows a combination between Accuracy and Precision, white and red circles represent the target, black points show an example of the measurements; to the right of each target, a schematization of an example between real outcrop (black fault trace) and virtual outcrop model (red fault trace) derived from aero-photogrammetric analysis is shown.

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

Discussion

The application of our working method to two study areas affected by coseismic surface faulting, during 2016, Central Italy seismic sequence, demonstrates that the virtual outcrops models represent an important tool for structural-geological and geomorphological research aimed at identifying and characterizing active tectonic structures.

Furthermore, comparing the information obtained from the morphological analysis of the fault

scarp of CAB, with the configuration of the coseismic surface faulting of the 2016 Central Italy seismic sequence, it is possible to make inferences on the rupturing history of the studied faults.

The good fit between the profile of the height of fault scarp and the 2016 coseismic slip profile, (Figure 8b) suggests that the long-term fault scarp may have formed as a result of several earthquakes each causing a similar amount of surface displacement, It is also possible to infer for these events comparable values of magnitude.

In particular, taking into account: i) the height of the entire fault scarp ii) the amount of the 30

October 2016 coseismic slip, iii) the geometry of the slope profiles also including the possible eroded free face, and iv) the assumpiont that earthquakes of equal/similar magnitude generated equal/similar dislocations (compared to 2016 one), we hypothesize that at least 5 events are needed to reach the observed values of fault scarp height.

Previous paleoseismological works [16,49,50] highlighted several seismic events, some of which caused the PTP reactivation, occurred in the Pano Grande-Mt Vettore area. In particular four events have been recognized in the last 8000 years, in addition to the earthquake of 2016. These four earthquakes, would have been occurred (from the oldest): i) between IV and VI millennium BC, ii) at the end of IV millennium BC, iii) between II and III millennium BC, iv) between the end of III and IV century AC (maybe in 270 AC). Furthermore, accepting the inferences, which implies that the 4.5 m

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

outcropping fault scarp of the CAB was exhumed after the Last Glacial Maximum L.G.M. (~18.000 ky), a preliminary estimate of the the slip rates, of 0.25 mm/y is obtained.

Of course, the analyzed fault scarps might have recorded a higher number of coseismic slip events which could have been followed by erosional episodes in a longer geological time, thus the above estimate must be considered purely indicative.

Several authors studied the causes/effects of 2016 earthquakes from many points of view, as

structural-geology, geophysics, seismology, paleoseismology, geomorphology, interferometry, geodesy, hydrogeology, engineering [15,31,32,34,36,39,40,51-90], but only a few authors have provided and elaborated data acquired through drones [91-94]. Also no author produced image analyses of coseismic ruptures employing the techniques we propose here. Our results, obtained from the elaboration of high-resolution topography, fit well with the data published in Villani et al. [36], Brozzetti et al. [15], and Perouse et al. [94] acquired through traditional or digital mapping techniques. At the same time the present work has improved the details of the surveyed geological features some of which were not appreciable directly in the field. Among these latter, the degradation and erosion of the fault plane as well as the tilting of the topographic surface, the dip-angle and the off-fault width of the tilted area.

The here obtained results showed that the use of the drone-derived photogrammetry presents acceptable errors for all those structural-geological applications which need not very high precision, (near centimetric resolution) and that might be an integration to the traditional field survey.

Figure 14. Comparison of DEM and detailed topographic profile obtained from: a) TINITALY/01 http://tinitaly.pi.ingv.it/ [95] with a resolution of 10 meters per pixel; b) Cartografia Tecnica Regionale vectorial 1:5000 scale from Regione Umbria [http://www.umbriageo.regione.umbria.it/pagina/distribuzione-carta-tecnica-regionale-vettoriale- 1-000], with a resolution of 5 meters per pixel; c) 1 centimeter per pixel obtained from this work.

In Figure 14 three types of DEMs with different resolutions (10 m, 5 m, and 1 cm) are compared.

637

638

639

640

641

642

643

644

645

The significant image resolution improvement of the topographic model obtained in this work (Figure 14c) is evident as well as the significant improvement of the geological interpretation, for example in the correct positioning of tectonic structures.

High-resolution topography allows us to delineate with great precision the fault trace and in some

conditions, such as in the CAB case, to identify the long term and recent kinematic indicators.

In the cases where active faults displace post-LGM slopes, the applied procedure, improve the estimates of the scarp height, thus helping to make inferences on the associated slip rates. Moreover, improving the precision of the fault mapping (e.g. in Figure 14c) can, in turn, facilitate the identification of the sites aimed at paleoseismological trenching.

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

Conclusions

During routine geological fieldwork, faults are generally surveyed on topographic and geological maps using a traditional compass-clinometer or, when a high-resolution survey is needed, using a digital device (smartphone o tablet) with a dedicated application.

In this paper, we have shown that the use of Fieldmove app for digital device integrated with the use

of Drone-acquired imagery is a very performing combination which speedily provides large sets of good-quality data improving sensibly the topographic and geological models that are subsequently reconstructed.

The presented high-resolution aerial survey represents the first attempt to digitally reconstruct

recent coseismic fault ruptures in variable geomorphological and outcrop conditions and to compare them with large published datasets collected through traditional and digital mapping.

In the study cases, which refer to two active splays of the Vettore-Bove Fault affected by surface faulting during the 2016 Central Italy seismic sequence, we reconstructed the fault traces, the height of the fault scarps, the amount of associated coseismic slip and its along-strike variations, with a precision which is much higher compared to previous literature reconstructions [15,29,36,37,96].

For all the aforesaid parameters the estimated errors range within a few centimeters, a value that is

well below the standards required by routine structural geological survey and is also acceptable for precision estimates of coseismic ground deformations.

A complete workflow in which the entire procedure is described in detail is reported. It is divided

into successive steps specifying all the necessary software and tools to be applied to obtain the topographic model of a study area.

Moreover an assessment of the precision and applicability of point clouds derived by Drone imagery is provided, and is also compared with the performances of the digital and analogic compass-clinometer.

Some further possible applications of the reconstructed models on the investigated active fault, f.i.

slip rate estimate or reconstruction of the rupture history is described, suggesting some possible correlations and integrations with the available paleoseismological data.

The results of this work highlight that photogrammetry can cover and improve the map resolution,

where necessary, with low-costs compared to more expensive techniques like the LiDAR data acquisition which are very costly and logistically complex to apply in many locations.

In other words, these methodologies undoubtedly can make a significant step forward to geological mapping and also to structural and geomorphological research.

They represent a field of work susceptible to have remarkable development and perspectives in the near future.

Supplementary Materials:

The following are available online at www.mdpi.com/xxx/s1

Author Contributions: Daniele Cirillo (D.C.) conceived the work. D.C. performed the fieldwork. D.C. prepared the GIS geological database and the figures. D.C. carries out the acquisition of images from Drone. D.C. used the new acquisition methodologies for the field survey, from traditional compass-clinometer to tablet to photogrammetry analysis. D.C. wrote the manuscript.

687

688

689

690

691

692

693

694

695

696

697

698

Funding: Acknowledgments:

This work, presents the main results of a research carried out by the authors in the frame of the CRUST

(Inter-Universitary Center for Tridimensional Seismotectonic Analysis https://www.crust.unich.it/) activities, following the 2016 Central Italy Seismic Sequence; this work realized thank it was supported by DiSPUTer Departmental 60% and Brozzetti research 30% funds for the renewal of research grant of Daniele Cirillo. I am grateful to Petex who provided us the Move, 2019.1 suite software Licence. I am grateful to my supervisor F. Brozzetti for the useful discussions, the suggestions, and the critical review of the manuscript.

Conflicts of Interest: The authors declare no conflict of interest.

699

References

727

728

729

730

731

732

733

734

735

736

737

pp. 163-178, 10.1016/j.jsg.2014.10.007

James, M.R.; Robson, S. Mitigating systematic error in topographic models derived from UAV and ground-based image networks. Earth Surface Processes and Landforms 2014, 39, 1413-1420.

Johnson, K.; Nissen, E.; Saripalli, S.; Arrowsmith, J.R.; McGarey, P.; Scharer, K.; Williams, P.; Blisniuk, K.

Rapid mapping of ultrafine fault zone topography with structure from motion. Geosphere, 2014.

http://dx.doi.org/10.1130/GES01017.1.

James, M.R.; Robson, S. Straightforward reconstruction of 3d surfaces and topography with a camera: Accuracy and geoscience application. Journal of Geophysical Research: Earth Surface 2012, 117, 17pp.

Westoby, M.J.; Brasington, J.; Glasser, N.F.; Hambrey, M.J.; Reynolds, J.M. ‘Structure-from-motion’

photogrammetry: A low-cost, effective tool for geoscience applications. Geomorphology 2012, 179, 300-314.

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

McCaffrey, K.J.W.; Jones, R.R.; Holdsworth, R.E.; Wilson, R.W.; Clegg, P.; Imber, J.; Holliman, N. Trinks, I.; Unlocking the spatial dimension: Digital technologies and the future of geoscience fieldwork. Journal of the Geological Society, London, 2005, 162, p. 927–938, doi:10 .1144/0016-764905-017

Brozzetti, F.; Boncio, P.; Cirillo, D.; Ferrarini, F.; de Nardis, R.; Testa, A. High‐resolution field mapping and

analysis of the August–October 2016 coseismic surface faulting (central Italy earthquakes): Slip distribution, parameterization, and comparison with global earthquakes. Tectonics, 2019, 38, 417–439. https://doi.org/10.1029/2018TC005305

Galadini, F.; Galli, P. Paleoseismology of silent faults in the Central Apennines (Italy): the Mt. Vettore and Laga Mts. Faults. Annals of Geophysics, 2003, 46, n.5, 815-836, 2017TC004583

Koopman, A. Detachment tectonics in the central Apennines, Italy. Geologica Eltraiectina, 1983, 30,1–155.

Lavecchia, G. Il sovrascorrimento dei Monti Sibillini: Analisi cinematica e strutturale. Bollettino della Societa Geologica Italiana, 1985, 104, 161–194

Bally, A. W.; Burbi, L.; Cooper, C.; Ghelardoni, R. Balanced cross-sections and seismic reflection profiles

across the central Apennines. Memorie della Societa Geologica Italiana, 1986, 35, 257–310.

Barchi, M. Integration of a seismic profile with surface and subsurface geology in a cross-section through the Umbria-Marche Apennines. Bollettino della Societa Geologica Italiana, 1991, 110, 469–479

Calamita, F.; Pizzi, A.; Roscioni, M. I "fasci" di faglie recenti ed attive di m vettore – m bove e di m castello

– m cardosa (appennino umbro-marchigiano). Studi Geologici Camerti 1992, volume speciale (1992/1), 81-95.

Coltorti, M.; Consoli, M.; Dramis, F.; Gentili, B.; Pambianchi G. Evoluzione geomorfologica delle piane alluvionali delle Marche centro-meridionali Geogr. Fis. Dinam. Quat., 14 (1) 1991, pp. 87-100

Blumetti, A. M.; Dramis, F.; Michetti, A. M. Fault-generated mountain fronts in the central Apennines

(Central Italy): Geomorphological features and seismotectonic implications. Earth Surface Processes and Landforms, 1993, 18(3), 203–223. https://doi.org/10.1002/esp.3290180304

Pizzi, A.; Calamita, F.; Coltorti, M.; Pieruccini, P. Quaternary normal faults, intramontane basins and

seismicity in the Umbria-Marche-Abruzzi Apennine Ridge (Italy): Contribution of neotectonic analysis to seismic hazard assessment. Bollettino Società Geologica Italiana Special Publication, 2002, 1(January), 923–929.

Brozzetti, F.; Lavecchia, G. Seismicity and related extensional stress field: the case of the Norcia seismic

zone. Annales Tectonicae, 1994, 8, 38–57.

Calamita, F.; Pizzi, A. Recent and active extensional tectonics in the southern Umbro-Marchean Apennines (Central Italy). Memorie della Societa Geologica Italiana, 1994, 48, 541–548

Cavinato, G. P.; De Celles, P. G. Extensional basins in the tectonically bimodal central Apennines

fold-thrust belt, Italy: Response to corner flow above a subducting slab in retrograde motion. Geology,

1999, 27(10), 955–958. https://doi.org/10.1130/0091-7613(1999)027%3C0955:EBITTB%3E2.3.CO;2

Galadini, F.; Galli, P. Active tectonics in the central Apennines (italy); input data for seismic hazard assessment. Natural Hazards 2000, 22, 225-270.

Pierantoni, P.; Deiana, G.; Galdenzi, S. Stratigraphic and structural features of the Sibillini Mountains

(Umbria-Marche Apennines, Italy), Italy. Journal of Geosciences, 2013 132(3), 497–520. doi:10.3301/IJG.2013.08

Biella, G.; Lavecchia, G.; Lozei, A.; Pialli, G.; Scarascia, S. In Primi risultati di un'indagine geofisica e interpretazione geologica del piano di s. Scolastica e del piano grande (norcia, pg), I Convegno Annuale G.N.G.T.S., Roma, 1981; Roma, I-TECT-UM.

Villani, F.; Sapia, V.; Baccheschi, P.; Civico, R.; Di Giulio, G.; Vassallo, M.; Marchetti, M.; Pantosti, D.

Geometry and structure of a fault‐bounded extensional basin by integrating geophysical surveys and seismic anisotropy across the 30 October 2016 Mw 6.5 earthquake fault (central Italy): The pian grande di Castelluccio basin. Tectonics 2019.

Chiaraluce, L., Di Stefano, R.; Tinti, E.; Scognamiglio, L.; Michele, M.; Casarotti, E.; Cattaneo, M.; De Gori,

P.; Chiarabba, C.; Monachesi, G.; Lombardi, A.; Valoroso, L.; Latorre, D.; Marzorati, S. The 2016 Central Italy Seismic Sequence: A First Look at the Mainshocks, Aftershocks, and Source Models. Seismol. Res. Lett. 2017, 88 (3), 757–771, doi:10.1785/0220160221

Michele, M.; Di Stefano, R.; Chiaraluce, L.; Cattaneo, M.; De Gori, P.; Monachesi, G.; Latorre, D.; Marzorati,

S.; Valoroso, L.; Ladina, C. et al. The Amatrice 2016 seismic sequence: A preliminary look at the mainshock and aftershocks distribution, Ann. Geophys., 2016, 59, doi: 10.4401/ag-7227.

846

847

848

849

850

851

852

853

854

855

856

857

858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

Iezzi, F.; Mildon, Z.; Walker, J.F.; Roberts, G.; Goodall, H.; Wilkinson, M.; Robertson, J. Coseismic throw variation across along-strike bends on active normal faults: Implications for displacement versus length scaling of earthquake ruptures. Journal of Geophysical Research: Solid Earth 2018, 123, 9817-9841.

Galadini, F.; Galli, P. Paleoseismology of silent faults in the central Apennines (Italy): the Mt. Vettore and

Laga Mts. Faults, Annals of Geophysics, 2003, 46, 815-836.

Fieldmove user guide Midland Valley-Petroleum Experts. Available online (February 11, 2020)

https://www.petex.com/media/2578/fieldmove_user_guide.pdf

Dewez, T. J. B.; Girardeau-Montaut, D.; Allanic, C.; Rohmer, J. Facets: a cloudcompare plugin to extract geological planes from unstructured 3d point clouds, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., 2016, xli-b5, 799-804, doi:10.5194/isprs-archives-XLI-B5-799-2016.

Manighetti, I. Evidence for self-similar, triangular slip distributions on earthquakes: Implications for

earthquake and fault mechanics. Journal of Geophysical Research 2005, 110.

Cappa, F.; Perrin, C.; Manighetti, I.; Delor, E. Off-fault long-term damage: A condition to account for generic, triangular earthquake slip profiles, Geochem. Geophys. Geosyst., 2014 15, 1476–1493, doi:10.1002/2013GC005182.

Tao, Z.; Alves, T.M. Impacts of data sampling on the interpretation of normal fault propagation and segment linkage. Tectonophysics 2019, 762, 79-96.

Iezzi, F.; Roberts, G.; Walker, J.F.; Papanikolaou, I. Occurrence of partial and total coseismic ruptures of

segmented normal fault systems: Insights from the central Apennines, Italy. Journal of Structural Geology

2019, 126, 83-99.

Galli, P.; Castenetto, S.; Peronace, E. The macroseismic intensity distribution of the 30 October 2016 earthquake in central Italy (Mw 6.6): Seismotectonic implications. Tectonics, 2017, 36, 2179–2191. https://doi.org/10.1002/

Galli, P.; Galderisi, A.; Peronace, E.; Giaccio, B.; Hajdas, I.; Messina, P.; Polpetta, F. Quante volte figliola?

Confessioni Sibilline di una giovane faglia. In riassunti estesi della comunicazione, 36° convegno nazionale Gruppo Nazionale di Geofisica della Terra Solida, Trieste, 14-16 November 2017 sessioni 1.1, 41-45, volume ISBN 978-88-940442-5-6, raccolta ISBN 978-88-940442-4-9

http://www3.ogs.trieste.it/gngts/files/2017/S11/Riassunti/Galli2.pdf

Bignami, C.; Valerio, E.; Carminati, E.; Doglioni, C.; Tizzani, P.; Lanari, R. Volume unbalance on the 2016 Amatrice – Norcia (central italy) seismic sequence and insights on normal fault earthquake mechanism. Scientific reports 2019, 9, 4250.

Brozzetti, F.; Mondini, A.C.; Pauselli, C.; Mancinelli, P.; Cirillo, D.; Guzzetti, F.; Lavecchia, G. Mainshock

Anticipated by Intra-Sequence Ground Deformations: Insights from Multiscale Field and SAR Interferometric Measurements. Geosciences 2020, 10(5), 186 https://doi.org/10.3390/geosciences10050186.

Cheloni, D.; De Novellis, V.; Albano, M.; Antonioli, A.; Anzidei, M.; Atzori, S.; Avallone, A.; Bignami, C.;

Bonano, M.; Calcaterra, S., et al. Geodetic model of the 2016 central Italy earthquake sequence inferred from insar and gps data. Geophysical Research Letters 2017, 44, 6778-6787.

De Guidi, G.; Vecchio, A.; Brighenti, F.; Caputo, R.; Carnemolla, F.; Di Pietro, A.; Lupo, M.; Maggini, M.; Marchese, S.; Messina, D., et al. Brief communication co-seismic displacement on October 26 and 30, 2016 (Mw 5.9 and 6.5) – earthquakes in central Italy from the analysis of a local GNSS network. Natural Hazards and Earth System Sciences 2017, 17, 1885-1892.

De Luca, G.; Di Carlo, G.; Tallini, M. A record of changes in the gran sasso groundwater before, during and

after the 2016 Amatrice earthquake, central Italy. Scientific reports 2018, 8, 15982.

Di Naccio D.; Kastelic V.; Carafa M. M. C.; Esposito C.; Milillo P.; Di Lorenzo C. Gravity versus tectonics: The case of 2016 Amatrice and Norcia (central Italy) earthquakes surface coseismic fractures. Journal of Geophysical Research: Earth Surface, 2019, 124, https://doi.org/10.1029/ 2018JF004762.

Durante, M.G.; Di Sarno, L.; Zimmaro, P.; Stewart, J.P. Damage to roadway infrastructure from 2016

central Italy earthquake sequence. Earthquake Spectra 2018, 34, 1721-1737.

Ercoli, M.; Forte, E.; Porreca, M.; Carbonell, R.; Pauselli, C.; Minelli, G.; Barchi, M. R. Using Seismic Attributes in seismotectonic research: an application to the Norcia Mw = 6.5 earthquake (30th October 2016) in Central Italy. Solid Earth 2020, 11, 329-348 https://doi.org/10.5194/se-11-329-2020.

Ferrario, M.F.; Livio, F. Characterizing the distributed faulting during the 30 October 2016, central Italy

earthquake: A reference for fault displacement hazard assessment. Tectonics 2018, 37, 1256-1273.

903

904

905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

https://doi.org/10.1016/j.jsg.2020.104014

Galli, P.; Galderisi, A.; Peronace, E.; Giaccio, B.; Hajdas, I.; Messina, P.; Pileggi, D.; Polpetta, F. The awakening of the dormant mount Vettore fault (2016 central Italy earthquake, Mw 6.6): Paleoseismic clues on its millennial silences. Tectonics 2019, 38, 687-705.

Gentili, S.; Di Giovambattista, R.; Peresan, A. Seismic quiescence preceding the 2016 central italy

earthquakes. Physics of the Earth and Planetary Interiors 2017, 272, 27-33.

Huang, M.-H.; Fielding, E.J.; Liang, C.; Milillo, P.; Bekaert, D.; Dreger, D.; Salzer, J. Coseismic deformation and triggered landslides of the 2016 Mw 6.2 Amatrice earthquake in italy. Geophysical Research Letters 2017, 44, 1266-1274.

Liu, C.; Zheng, Y.; Xie, Z.; Xiong, X. Rupture features of the 2016 mw 6.2 Norcia earthquake and its

possible relationship with strong seismic hazards. Geophysical Research Letters 2017, 44, 1320-1328.

Masi, A.; Chiauzzi, L.; Santarsiero, G.; Manfredi, V.; Biondi, S.; Spacone, E.; Del Gaudio, C.; Ricci, P.; Manfredi, G.; Verderame, G.M. Seismic response of rc buildings during the Mw 6.0 august 24, 2016 central Italy earthquake: The Amatrice case study. Bulletin of Earthquake Engineering 2017.

Mazzoni, S.; Castori, G.; Galasso, C.; Calvi, P.; Dreyer, R.; Fischer, E.; Fulco, A.; Sorrentino, L.; Wilson, J.;

Penna, A., et al. 2016–2017 central Italy earthquake sequence: Seismic retrofit policy and effectiveness.

Earthquake Spectra 2018, 34, 1671-1691.

Mildon, Z.K.; Roberts, G.P.; Faure Walker, J.P.; Iezzi, F. Coulomb stress transfer and fault interaction over millennia on non-planar active normal faults: The Mw 6.5–5.0 seismic sequence of 2016–2017, central italy. Geophysical Journal International 2017, 210, 1206-1218.

Papadopoulos, G.A.; Ganas, A.; Agalos, A.; Papageorgiou, A.; Triantafyllou, I.; Kontoes, C.; Papoutsis, I.;

Diakogianni, G. Earthquake triggering inferred from rupture histories, DINSAR ground deformation and stress-transfer modelling: The case of central Italy during august 2016–January 2017. Pure and Applied Geophysics 2017, 174, 3689-3711.

Picozzi, M.; Bindi, D.; Brondi, P.; Di Giacomo, D.; Parolai, S.; Zollo, A. Rapid determination of p

wave-based energy magnitude: Insights on source parameter scaling of the 2016 central Italy earthquake sequence. Geophysical Research Letters 2017, 44, 4036-4045

Polcari, M.; Montuori, A.; Bignami, C.; Moro, M.; Stramondo, S.; Tolomei, C. Using multi-band insar data

for detecting local deformation phenomena induced by the 2016–2017 central Italy seismic sequence.

Remote Sensing of Environment 2017, 201, 234-242.

Porreca, M.; Minelli, G.; Ercoli, M.; Brobia, A.; Mancinelli, P.; Cruciani, F.; Giorgetti, C.; Carboni, F.; Mirabella, F.; Cavinato, G., et al. Seismic reflection profiles and subsurface geology of the area interested by the 2016-2017 earthquake sequence (central Italy). Tectonics 2018, 37, 1116-1137.

Ren, Y.; Wang, H.; Wen, R. Imprint of rupture directivity from ground motions of the 24 august 2016 Mw

6.2 central Italy earthquake. Tectonics 2017, 36, 3178-3191.

Scognamiglio, L.; Tinti, E.; Casarotti, E.; Pucci, S.; Villani, F.; Cocco, M.; Magnoni, F.; Michelini, A.; Dreger,

D. Complex fault geometry and rupture dynamics of the Mw 6.5, 30 October 2016, central Italy earthquake. Journal of Geophysical Research: Solid Earth 2018, 123, 2943-2964.

Sextos, A.; De Risi, R.; Pagliaroli, A.; Foti, S.; Passeri, F.; Ausilio, E.; Cairo, R.; Capatti, M.C.; Chiabrando,

F.; Chiaradonna, A., et al. Local site effects and incremental damage of buildings during the 2016 central Italy earthquake sequence. Earthquake Spectra 2018, 34, 1639-1669.

Smeraglia, L.; Billi, A.; Carminati, E.; Cavallo, A.; Doglioni, C. Field- to nano-scale evidence for weakening

mechanisms along the fault of the 2016 Amatrice and Norcia earthquakes, Italy. Tectonophysics 2017,

712-713, 156-169.

Stewart, J.P.; Zimmaro, P.; Lanzo, G.; Mazzoni, S.; Ausilio, E.; Aversa, S.; Bozzoni, F.; Cairo, R.; Capatti, M.C.; Castiglia, M., et al. Reconnaissance of 2016 central Italy earthquake sequence. Earthquake Spectra 2018, 34, 1547-1555.

Suteanu, C.; Liucci, L.; Melelli, L. The central Italy seismic sequence (2016): Spatial patterns and dynamic

fingerprints. Pure and Applied Geophysics 2017, 175, 1-24.

952

953

954

955

956

957

958

959

960

961

962

963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

991

992

993

994

995

996

997

998

999

1000

1001

1002

Testa, A., Boncio, P., Di Donato, M., Mataloni, G., Brozzetti, F., Cirillo D. Mapping the geology of the 2016 Central Italy earthquake fault (Mt. Vettore – Mt. Bove fault, Sibillini Mts.): geological details on the Cupi – Ussita and Mt. Bove – Mt. Porche segments and overall pattern of coseismic surface faulting. Geological Field Trips and Maps, 2019, https://doi.org/10.3301/GFT.2019.03

Tinti, E.; Scognamiglio, L.; Michelini, A.; Cocco, M. Slip heterogeneity and directivity of the ml 6.0, 2016,

Amatrice earthquake estimated with rapid finite-fault inversion. Geophysical Research Letters 2016, 43, 10,745-710,752.

Tung, S.; Masterlark, T. Delayed poroelastic triggering of the 2016 october visso earthquake by the august Amatrice earthquake, Italy. Geophysical Research Letters 2018, 45, 2221-2229.

Tung, S.; Masterlark, T. Resolving source geometry of the 24 august 2016 Amatrice, central Italy,

earthquake from insar data and 3d finite‐element modeling. Bulletin of the Seismological Society of America

2018, 108, 553-572.

Valentini, A.; Pace, B.; Boncio, P.; Visini, F.; Pagliaroli, A.; Pergalani, F. Definition of seismic input from

fault‐based PSHA: Remarks after the 2016 central Italy earthquake sequence. Tectonics 2019, 38, 595-620.

Valerio, E.; Tizzani, P.; Carminati, E.; Doglioni, C.; Pepe, S.; Petricca, P.; De Luca, C.; Bignami, C.; Solaro, G.; Castaldo, R., et al. Ground deformation and source geometry of the 30 October 2016 Mw 6.5 Norcia earthquake (central Italy) investigated through seismological data, dinsar measurements, and numerical modelling. Remote Sensing 2018, 10, 1901.

Walters, R.J.; Gregory, L.C.; Wedmore, L.N.J.; Craig, T.J.; McCaffrey, K.; Wilkinson, M.; Chen, J.; Li, Z.;

Elliott, J.R.; Goodall, H., et al. Dual control of fault intersections on stop-start rupture in the 2016 central Italy seismic sequence. Earth and Planetary Science Letters 2018, 500, 1-14.

Wang, L.; Gao, H.; Feng, G.; Xu, W. Source parameters and triggering links of the earthquake sequence in

central Italy from 2009 to 2016 analyzed with gps and insar data. Tectonophysics 2018, 744, 285-295.

Whitney, R. Approximate recovery of residual displacement from the strong motion recordings of the 24 august 2016 Amatrice, Italy earthquake. Bulletin of Earthquake Engineering 2017, 16, 1847-1868.

Wilkinson, M.W.; McCaffrey, K.J.W.; Jones, R.R.; Roberts, G.P.; Holdsworth, R.E.; Gregory, L.C.; Walters,

R.J.; Wedmore, L.; Goodall, H.; Iezzi, F. Near-field fault slip of the 2016 Vettore Mw 6.6 earthquake (central Italy) measured using low-cost GNSS. Scientific reports 2017, 7, 4612.

Xu, G.; Xu, C.; Wen, Y.; Jiang, G. Source parameters of the 2016–2017 central italy earthquake sequence

from the sentinel-1, alos-2 and gps data. Remote Sensing 2017, 9, 1182.

Zimmaro, P.; Scasserra, G.; Stewart, J.P.; Kishida, T.; Tropeano, G.; Castiglia, M.; Pelekis, P. Strong ground motion characteristics from 2016 central Italy earthquake sequence. Earthquake Spectra 2018, 34, 1611-1637.

Franke, K.W.; Lingwall, B.N.; Zimmaro, P.; Kayen, R.E.; Tommasi, P.; Chiabrando, F.; Santo, A. Phased

reconnaissance approach to documenting landslides following the 2016 central Italy earthquakes.

Earthquake Spectra 2018, 34, 1693-1719.

Gori, S.; Falcucci, E.; Galadini, F.; Zimmaro, P.; Pizzi, A.; Kayen, R. E.; Lingwall, B. N.; Moro, M.; Saroli, M.; Fubelli, G.; Di Domenica A. and Stewart J. P.; Surface Faulting Caused by the 2016 Central Italy Seismic Sequence: Field Mapping and LiDAR/UAV Imaging. Earthquake Spectra, 2018, Volume 34, No. 4, pages 1585–1610.

Pavlides, S.; Chatzipetros, A.; Papathanasiou, G.; Georgiadis, G.; Sboras, S.; Valkaniotis, S. Ground

deformation and fault modeling of the 2016 sequence (24 Aug. – 30 Oct.) in central Apennines (central Italy). Bulletin of the Geological Society of Greece 2017, 51, 76.

Perouse, E.; Benedetti, L.; Fleury, J.; Rizza, M.; Puliti, I.; Billant, J.; et al. Coseismic slip vectors of 24 August

and 30 October 2016 earthquakes in Central Italy: Oblique slip and regional kinematic implications. Tectonics, 2018, 37. https://doi.org/10.1029/2018TC005083

Tarquini, S.; Vinci, S.; Favalli, M.; Doumaz, F.; Fornaciai, A.; Nannipieri, L. Release of a 10-m-resolution

DEM for the Italian territory: Comparison with global-coverage DEMs and anaglyph-mode exploration via the web, Computers & Geosciences , 2012, 38, 168-170. doi: doi:10.1016/j.cageo.2011.04.018

La carta geologica d'Italia – scala 1:100.000 – Foglio Geologico n. 132. Norcia

© 2020 by the authors. Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).