Selfie Drones for 3D Modelling, Geological Mapping and Data

Collection: Key Examples from Santorini Volcanic Complex, Greece

Fabio Luca Bonali

1,2 a

, Varvara Antoniou

, Othonas Vlasopoulos

, Alessandro Tibaldi

1,2 d

and Paraskevi Nomikou

Department of Earth and Environmental Sciences, University of Milano-Bicocca,

Piazza della Scienza 4 – Ed. U04, 20126, Milan, Italy

CRUST- Interuniversity Center for 3D Seismotectonics with Territorial Applications, Italy

Department of Geology and Geoenvironment, National and Kapodistrian University of Athens,

Panepistimioupoli Zografou, 15784 Athens, Greece

Keywords: Geological Mapping, Structure from Motion, Santorini Volcano, Virtual Outcrop, LBA Eruption.

Abstract: In the present work, we tested the use of selfie drones as a tool for 3D modeling, geological mapping, and

data collection. The model we used is a 0.300-kg multirotor quadcopter being equipped with a 1/2.3-inch

CMOS sensor capable of capturing 12 Megapixel pictures, attached to a 2-axis mechanical gimble and with

approximately 16 minutes of flight time. Test sites are located in Santorini and are characterised by different

settings: i) the 1570-1573 AD volcanic crater area, in Nea Kameni island, has a mostly horizontal topography;

ii) the outcrop along Vlychada beach, showing layers of the Late Bronze Age (also well-known as Minoan)

eruption, has mostly vertical topography. By applying the Structure from Motion techniques to pictures

collected using the selfie drone, we were capable of: i) reconstructing the two sites with centimetric to sub-

centimetric resolution; ii) recognizing geological features on very high-resolution Digital Surface Models and

Ortomosaics; iii) mapping vertical cliffs made up of volcanic deposits on 3D Digital Outcrops Models; iv)

collect new quantitative data for both sites.

1 INTRODUCTION

Field studies and data collection are vital for mapping

and understanding the active geological processes on

Earth, particularly for those that induced superficial

deformations like earthquakes and shallow magmatic

processes (e.g. Bonali et al., 2012; Tibaldi et al.,

2017). However, field studies and direct observations

are very often limited by specific field-related

conditions such as the inaccessibility of key outcrops

due to their location in remote or dangerous areas

(e.g. Tibaldi et al., 2008). The Structure from Motion

and Multiview stereo (SfM-MVS) photogrammetry

techniques, where photos are collected using

Unmanned Aerial Vehicles (UAVs), are nowadays

widely used in Earth and Environmental Sciences to

https://orcid.org/0000-0003-3256-0793

https://orcid.org/0000-0002-5099-0351

https://orcid.org/0000-0002-6713-9141

https://orcid.org/0000-0003-2871-8009

https://orcid.org/0000-0001-8842-9730

overcome these problems providing high-resolution

3D Digital Outcrop Models (DOMs), digital surface

models (DSMs) and Orthomosaics as results (e.g.

Bonali et al., 2019a; Fallati et al., 2019). Most people

dealing with geological and geohazard studies use

different types of UAVs: balloons, multi-rotor, fixed-

wing and hybrid. Whereas balloons do not need fuel

or battery, on the other hand they cannot be remotely

controlled. Hybrid types allow to switch between

flying like a fixed-wing aircraft and hovering like a

multi-rotor one. The fixed-wing type can cover larger

areas in a smaller time frame using high quality

cameras, but such model is more difficult to be

transported and more expensive than multi-rotor

UAVs. Based on our experience, the latter can fly at

very low heights attaining a great field resolution, and

Bonali, F., Antoniou, V., Vlasopoulos, O., Tibaldi, A. and Nomikou, P.

Selﬁe Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece.

DOI: 10.5220/0009575001190128

In Proceedings of the 6th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2020), pages 119-128

ISBN: 978-989-758-425-1

119

much more importantly, take-off and landing

operations are easier than for fixed-wing models; this

is crucial especially in difficult logistic terrains (e.g.

outcropping lavas, Bonali et al., 2019a).

In the present work we tested the use of the so

called “selfie drones” - quadcopter type - to produce

very high-detailed 3D DOMs of relevant outcrops for

geological mapping, data collection and scientific

dissemination. As case studies we selected two sites

within the Santorini volcanic complex (Fig. 1) with

different characteristics: i) the 1570-1573 AD

volcanic crater in Nea Kameni island and ii) an

outstanding vertical outcrop showing volcanic

layered deposits, along the Vlychada Beach, southern

Santorini (Fig. 1B).

2 GEOLOGICAL SETTING OF

KEY SITES

Santorini volcanic group is a ring of three islands

(Thera, Therasia, and Aspronisi) around a flooded

caldera containing the islands of Palea and Nea

Kameni, which postdate caldera collapse (3.6ka) and

are the subaerial expressions of an intracaldera,

largely submarine lava shield (Druitt, 2014). The

caldera is a 11x7km composite structure resulting

from at least four collapses over the last 200ky (Druitt

and Francaviglia, 1992), the last of which took place

during, and immediately following, the ~1630 BCE

‘Minoan’ eruption (Friedrich et al., 2006). It consists

of three flat-floored basins: a large northern 390m

deep, and two smaller ones (western: 320m and

southern: 270m respectively, Nomikou et al., 2013).

The Kameni islands are the subaerial expression

of a 4.3 ± 0.7km

intracaldera shield, 3.5km in basal

diameter, the summit of which towers 470m above

the caldera seafloor. The magmatic vents of both, lie

within a NE-SW volcanotectonic line which controls

the magma ascent of the region.

The evolution of the Kameni islands has been

determined by 9 subaerial eruptions: 197 BCE,

AD46-47, AD726, 1570-1573, 1707-1711, 1866-70,

1925-28, 1939-41, and 1950 (Pyle and Elliot, 2006)

that discharged dacitic flows and formed domes,

channels and levees, blocky lavas, ash plumes

(Vulcanian eruptions) and ballistic ejecta.

Bathymetric imagery data have revealed unknown

submarine flows (pillow lavas) defining the actual

morphology (pillow lavas) and final volume of

products from Kameni Volcano to 4.85±0.7 km

(Nomikou et al., 2014).

The 1570-1573 AD volcanic crater is located in

the northeastern part of it. During its

surtseyan activity

which was accompanied by ash-fall and block fall-out,

a small lava dome named Mikri Kameni was extruded

(Watts et al., 2015).

Figure 1: Location of Santorini group in the Aegean Sea (A)

and selected sites belonging to Santorini Volcanic Complex

(B).

Moving to the external southern part of Thera

island, and along Vlychada beach, a very well

exposed section with pumice layers deposited during

the famous Late Bronze Age (LBA) (well-known also

as Minoan) eruption, can be seen. The LBA eruption

of Santorini has influenced the decline of the great

Minoan civilization on Crete, making it an iconic

event in both volcanology and archaeology (e.g.,

Manning et al. 2006; Druitt, 2014). It discharged

between 30 and 80km

(dense rock equivalent;

Johnston et. al., 2014) of rhyodacitic magma, mostly

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

120

as pyroclastic flows which entered the sea, and are

preserved as ignimbrite in the surrounding submarine

basins (Sigurdsson et. al., 2006). According to

numerous volcanological studies, there is a consensus

that the eruption occurred in four major phases with

an initial precursory phase (Reck, 1936; Heiken and

McCoy, 1990; Druitt, 2014). In Vlychada, volcanic

products from phases P2 and mostly P3 and P4 can be

recognized. Phase P2 products are dominated by

pyroclastic surge deposits with multiple bedsets,

dune-like bedforms with wavelengths of several

meters or more, bomb sag horizons, and TRM

temperatures of 100–250°C. Phase P3 is a coarse-

grained, massive, phreatomagmatic ignimbrite up to

55m thick (Druitt et al., 1999), still reflecting magma-

water interaction and deposited at low temperatures

(Druitt, 2014; McClelland E. & Thomas R. A.,1990).

Phase P4 is a tan- to pink- colored compound

ignimbrite (“tan ignimbrite”) (Druitt, 2014), mostly

finegrained (ash and lapilli grade), with a high

abundance of comminuted lithic debris in the ash

fraction (Bond and Sparks, 1976) (Fig. 2B).

Figure 2: (A) Panoramic view of Nea Kameni island, the

location of 1570-1573 crater is indicated. (B) UAV-

captured picture showing part of the LBA deposit

outcropping along the Vlychada beach, the recognizable

phases (P2, P3 and P4) are indicated.

3 3D MODELLING

In this section we present the used workflow, aimed

at 3D DOMs construction, which can be divided in

three parts: i) appropriate UAV selection, ii) data

collection (digital image gathering and setup of

Ground Control Points - GCPs), and iii) SfM-MVS

photogrammetry processing - data processing and

model reconstruction. The results are in the form of

Digital Surface Models (DSMs), Orthomosaics and

3D DOMs (or Virtual Outcrops).

3.1 UAV Selection and Use

For the present research, a commercial multi-rotor

vehicle has been chosen, since it can be remotely

controlled, is characterized by a stable hovering, can

be easily transported in the field and is less expensive

than hybrid and fixed-wing models. In addition, it can

fly at very low heights, thus obtaining greater field

resolution, while take-off and landing operations are

smoother compare to that of fixed-wing models and

this can be crucial especially when operating in

difficult logistic terrains, such as lava flow outcrops

or remote beach areas (Bonali et al., 2019a; Fallati et

al., 2019). Having that in mind, we selected the DJI

Spark “selfie drone” (Fig. 3), being a 0.300-kg

vehicle equipped with a 1/2.3 inch CMOS sensor

capable of capturing 12 Megapixel pictures, including

EXIF information (Exchangeable Image file Format)

GPS geographic coordinates (DATUM WGS84), and

video up to 1080p at 30 fps, while its storage capacity

is up to 64 GBs via a Micro SD card. Its flight time is

approximately 16 minutes, thus four batteries and an

external charger (since it can be also charged by USB

plug) were used for the survey. The camera is

attached to a 2-axis mechanical gimble that provides

stabilization, allowing to capture clear, stable images

and video, having a tilting range of 0-85°. Owing to

its small size and low weight, we retain that this

model is useful for field research and 3D DOM

reconstruction, particularly for outcrops located in

very remote areas where the equipment must be

carried on foot.

3.2 Flight Mission and Data Collection

The first step has been devoted to defining the area to

be surveyed and to planning the details of the flight

missions, such as path orientation. In doing this, care

must be taken of wind direction, which may affect

UAV flight performance. As the surveyed geological

objects are situated in very remote areas, we made use

of the smaller DJI Spark, managed through the DJI

Selﬁe Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece

121

GO App (https://www.dji.com/it/goapp). Generally,

mission planning involves fundamental parameters

like path orientation, overlaps of images, flight

height, flight speed, also depending on camera

characteristics (e.g. Bonali et al., 2019). Such

parameters influence the quality of the generated

products (3D point cloud, DSM, orthomosaic, 3D

Model). As suggested in recent works (Gerloni et al.,

2018; Antoniou et al. 2019; Bonali et al. 2019a;

Krokos et al. 2019), UAV-captured photos should

have an overlap of 90% along single paths and 80%

in a lateral direction, so as to obtain a better alignment

of the images and reduce distortions on the resulting

orthomosaics. The UAV we tested does not use any

autopilot system, so that it has been manually

controlled by the pilot for the entire duration of the

mission. We are aware that not using mission

planning software can affect the final quality of the

model, but it was one of the challenge of the present

work).

Figure 3: The selfie drone used in the present work, also

equipped with propeller guards, person for scale.

In order to reach the goals of the present work,

during image collection pictures were taken from a

height lower than 30 m, the drone flew at a speed of

2 m/s with an overlap consistently in a range of 90-

85% along paths and 80-75% in a lateral direction;

images were captured every 2 seconds (equal time

interval mode), and in optimal light conditions,

suitable for the camera ISO range (100-1600). This

was done to minimize the motion blur, to avoid the

rolling shutter effect, and to achieve well-balanced

camera settings (exposure time, ISO, aperture), thus

ensuring sharp and correctly exposed images (e.g.

Vollgger et al. 2016). Moreover, to reduce shadows

around elevated features, drone was operated when

the sun was straight overhead (at zenith). In order to

allow the co-registration of datasets and the

calibration of models resulting from SfM-MVS

photogrammetry processing (e.g. James and Robson,

2012; Turner et al., 2012; Westoby et al., 2012; James

et al., 2017), World Geodetic System (WGS84)

coordinates of, at least, four artificial Ground Control

Points (GCPs) were fixed near every corner of each

surveyed area (an additional one was selected in the

central part) reducing the ‘doming’ effect resulting

from SfM processing.

3.3 Photogrammetry Processing

After photo collection, the next step is dedicated to

data processing aimed at 3D DOMs, DSMs and

Orthomosaics generation. The collected images have

been processed through Agisoft Metashape

(http://www.agisoft.com/), a commercial Structure

from Motion software (SfM). This application has

been increasingly used for both UAV and field based

SfM reconstructions, owing to its user-friendly

interface, intuitive workflow and high quality of point

clouds (Burns et al., 2017).

Figure 4: (a) UAV-based SfM-MVS Workflow used, (b)

UAV-collected pictures already aligned with the processed

sparse cloud. (c) Dense cloud generated by the SfM

software showing the 1570-1573 Volcanic Crater.

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

122

SfM-MVS techniques allowed us to identify

matching features in different photos and combine

them to create a sparse and a dense cloud, an

orthomosaic, a DSM, and eventually a 3D DOMs as

final products (Stal et al., 2012; Westoby et al., 2012).

The steps leading to model construction are shown in

Figure 4A; further details are provided hereunder.

The first step was to obtain an initial low-quality

photo alignment, only considering measured camera

locations. Thereafter, we excluded the photos with

quality value ˂0.8 (or out of focus) from any further

processing, using the tool provided by the software

that is designed to detect poorly focused images

(Agisoft L.L.C, 2020). Following this initial quality

check, Ground Control Points (GCPs) were added to

all photos, where available, so as to: i) scale and

georeference the point cloud (and thus the resulting

model); ii) optimize extrinsic parameters, such as

estimated camera locations and orientations; iii)

improve the accuracy of the final model.

Photos were

then realigned in high quality setting, camera

locations and orientations were better established, and

the sparse point cloud was computed by the software

(e.g. Fig. 4B). The next phase consisted in

reconstructing the dense point cloud (e.g. Fig. 4C)

from the sparse point cloud, using a mild depth

filtering and medium quality settings. The 3D DOMs,

DSMs and orthomosaics were finally created through

the Agisoft Metashape software. The resulting 3D

DOMs are characterized by a high-resolution texture

with a pixel size < 1.0 cm/pixel.

3.4 Georeferencing of GCPS

In order to allow the co-registration of datasets and

calibration of models as well as for reducing the

‘doming’ effect resulting from SfM processing (e.g.

Orthomosaics and DSMs), World Geodetic System

(WGS84) coordinates of, at least, four artificial

Ground Control Points (GCPs) were established near

each corner and another one in the central part of each

surveyed area (e.g. James and Robson, 2012; Turner

et al., 2012; Westoby et al., 2012; James et al., 2017).

For surveying GCPs, we placed well visible artificial

markers, as well as natural targets (e.g. lava flow

borders). GCPs were surveyed with the Emlid Reach

RS©, low-cost single frequency receivers (Rover and

Base) in RTK configuration (with centimetre-level

accuracy). In regard to the surveys in Santorini, we

used the Long-Range Radio (LoRa 868/915 MHz)

connection mode where the base was set on a fix

position and sent a real-time correction to the Rover.

The LoRa mode can be advantageous in the absence

of international GNSS service or CORS network. All

the z (altitude a.s.l.) values of the GCPs were

corrected using the regional geoid model to obtain the

orthometric height for the models.

4 RESULTS

In this section we provide all details regarding the 3D

DOMs, DSMs and orthomosaic for the two studied

areas, including new 2D and 3D maps, data and

interpretations.

4.1 Volcanic Crater in Nea Kameni

A total number of 1522 of pictures have been

collected using nadir camera orientation and 1231 of

them have been correctly aligned and used for the

dense cloud generation. The remnant 291 have been

excluded because they resulted out of focus, too dark

or too white, or below the quality threshold value of

0.8 applied for this model.

The resulting model for the 1570-1573 volcanic

crater has an overall extent of 207x 291 m, a

resolution of 3.79 cm/pix for the DSM and of 9.47

mm/pix for both the orthomosaic and the texture of

the 3D DOM. The DSM values are in the range of

11.02- 49.15 m above the sea level (a.s.l.) (Fig. 4). By

analysing the Orthomosaic, DSM and the 3D DOM

we also recognised the following features: i) the

crater is slightly elongated along N5°E direction

(102.52 x 95.33 m); ii) by 3D analysis it is possible to

trace the line connecting the two crater rim depressed

points (e.g. Tibaldi, 1995; Bonali et al., 2011) in the

same direction; iii) in the northernmost part of the

crater an open fracture with a dilation of 4.7-5.3 m is

present and iv) the crater has a depth of 31.43 m.

4.2 Volcanic Deposit along the

Vlychada Beach

A total number of 568 pictures have been collected

and considering the geometry of the outcrops, that is

almost vertical, the majority of them have been

collected with an oblique camera orientation.

439 of them have been correctly aligned and used

for the dense cloud generation, whereas 129 have

been excluded, using the same method as in the

previous mentioned model. In addition, a mask has

been applied to all photos, in order to exclude the sky

and the sea water from the processing, to avoid

artefacts and noises. The model for the volcanic

deposit surveyed along the Vlychada beach has an

overall extent of 157 x 128 m, a resolution of 1.67

Selﬁe Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece

123

cm/pix for the DSM and of 8.37 cm/pix for both the

orthomosaic and the texture of the 3D DOM.

The DSM values are in the range -1.61- 38.3m

a.s.l.. Regarding the volcanic phases that can be

recognised, it resulted very useful to collect the

thickness of them directly on the 3D DOM, because

of the vertical geometry of the outcrop. These

measurements have been collected in Agisoft

Metashape, using the ruler tool: the phase 2 has a

thickness of 2m, phase 3 ranges between 7 and 10 m

and phase 4 has an outcropping thickness of 25m. In

addition, it is possible to appreciate the presence of

several blocks as well as to quantify their dimension

as shown in Figure 5D.

5 DISCUSSION

In the present section we discuss the use of selfie

drones for 3D modelling and mapping, as well as we

present new outcomes for the studied areas.

5.1 Selfie Drones for Surveying, 3D

Modelling and Mapping

At a general level, UAVs are excellent instruments to

collect highly detailed pictures and videos from the

above of the key sites (e.g. Fig. 2A), which is

impossible using only classical field activity. These

images can be used for research activity and better

interpretation as well as for outreach activity (e.g.

Bonali et al., 2019a; Pasquaré Mariotto et al., 2020).

Furthermore, a field view from the drone during field

surveys can also help in planning further steps of

exploration. In view of the above, a selfie drone, due

to its small dimension and weight, is useful in

supporting field exploration and photo/video

collection, even though it is recommended to work in

the height and distance range suggested by local laws

and manufacture’s technical manual.

Regarding the 3D modelling, and consequent

mapping and data collection activity, both models

have great resolution, in terms of DSM and

Orthomosaic. In particular, the latter and the texture

of the 3D DOM, reach a resolution greater than 1 cm.

This excellent resolution is helpful for classical

mapping and data collection on DSM/Orthomosaic,

as well as is crucial for mapping 3D vertical outcrops,

including the quantitative characterisation of small

objects like the ones included in the volcanic layers

outcropping along the Vlychada beach. Based on the

experience gained from the present work we were

capable of flying up to 50 m from the ground to

collect some pictures for the latter model.

Figure 5: (A) Orthomosaic and DSM (B) of the 1570-1573

volcanic crater in Nea Kameni; a.s.l.: above the sea level.

(C-D) Detailed view of the northernmost fracture. (E) 3D

Virtual Outcrop of the volcanic crater available online

(https://skfb.ly/6PUNt).

Respect to other models, there are some negative

aspects that we highlighted. Selfie drones must be

manually controlled during the survey and they can

be more affected by the wind; so, an expert pilot is

recommended to collect pictures in the best way

possible. This can result in a larger number of

collected pictures compared to those really needed for

SfM-MVS processing, since some of them must be

removed due to out of focus condition or incorrect

white balance. To fly above 50 m and to cover larger

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

124

areas, other types of UAVs can be used: balloons,

larger multi-rotor, fixed-wing and hybrid. Balloons

do not need fuel or a battery, but they cannot be

remotely controlled. The fixed-wing type can cover

larger areas in a smaller time frame using high quality

cameras, but such model is more difficult to be

transported in the field. Hybrid types allow to switch

between flying like a fixed-wing aircraft and hovering

like a multi-rotor one. The latter two are more

expensive than commercial multi-rotor type. Larger

multi-rotor can fly for a longer time respect to a selfie

drone, at very low heights attaining a great field

resolution, and much more importantly, take-off and

landing operations are easier than for fixed-wing

models; this is crucial especially in difficult logistic

terrains.

Figure 6: (A) Orthomosaic and DSM (B) of the outcrop

related to LBA deposits along the Vlychada beach; a.s.l.:

above the sea level. (C) 3D Virtual Outcrop available online

(https://skfb.ly/6PZPO), where it is possible to recognise

the different phases with very high detail (D).

5.2 New Outcomes for the Studied

Areas

Regarding the 1570-1573 volcanic crater in Nea

Kameni, we defined its dimension and depth with

very high accuracy, also discovering a N5°E

elongation trend that matches with the line connecting

the two most depressed points along the crater rim.

As suggested by Tibaldi (1995) and Bonali et al.

(2011), such line can represent, with low discrepancy,

the direction of magma feeding fractures, suggesting

that the 1570-1573 AD eruptions were driven by a

fissure. Dykes outcropping along the Northern

Caldera Wall have a NE-SW dominant strike, even

though the N-S direction is also represented in the

dataset (Browning et al., 2015), suggesting that

magma can reach the surface also along this direction

in central Santorini. Finally, the fracture located just

north of the crater has a dilation of about 5 m that is

consistent with dyke-induced fractures in Northern

Iceland (e.g. Bonali et al., 2019a,b).

Regarding the volcanic deposits measured along

the Vlychada beach, we defined the boundaries of

different phases of the LBA deposits on the 3D DOM,

as well as we measured their thickness with very high

accuracy. Also, it was possible to observe the cliff

closely enough to distinguish individual and small

components such as volcanic blocks within the bulk

of each phase and to quantify their dimensions.

5.3 3D DOMs and Virtual Outcrops for

Teaching and Dissemination

The two presented sites can be used also for teaching

and outreach activity, as well as they can be both

suggested as geosites. In fact, they have a

considerable scientific value and a potentially high

educational value, enhanced by their accessibility and

safety (e.g. Pasquaré Mariotto et al., 2020).

Recent improvements in Geographic Information

Systems (GIS) technologies can provide new

opportunities for immersive and wide engaging

public audiences. Story Maps being interactive

webGIS applications can provide support for

scientific storytelling in a compelling and

straightforward way (Antoniou, et al., 2019) using

multi-media assets (e.g. photos, videos, 3D DOMs)

and narrative texts with the aim of visualizing spatial

data effectively.

As previously tested (Antoniou et al., 2018; 2019,

2019a, 2019b) these applications can represent an

interactive way for presenting the geological and

geomorphological characteristics of places which can

be defined as geotopes or protected areas worldwide,

providing a quick access of all useful data to a wide

audience and thus developing the interest and

possibly motivating people to learn more and visit the

area.

With the purpose of enhancing the popularization

and fruition of these two sites, we published them as

“Virtual Outcrops” (Xu et al. 1999; Trinks et al. 2005;

Tavani et al. 2014) on the web. The 1570-1573 AD

volcanic crater is available as Virtual Outcrop at

https://skfb.ly/6PUNt and the volcanic deposits

section in Vlychada beach, at https://skfb.ly/6PZPO.

Both of them can be visualized using a laptop or

Selﬁe Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece

125

mobile phone (in 2D) or using a Mobile VR headset

to appreciate the third dimension (e.g. Fig. 6A).

Finally, as suggested by Gerloni et al. (2018) and

Krokos et al. (2019), SfM-MVS-derived 3D DOMs

can be imported in a game engine to build fully

navigable immersive Virtual Reality systems. Such

approach has been firstly used for teaching activity

(e.g. Fig 6B) and dissemination activity to popularise

geosciences to non-academic audiences (e.g.

citizens).

Figure 7: (A) An example of a research group observing the

same virtual outcrop using a mobile VR headset. (B) An

example of an outreach event carried out using the

Immersive Virtual Reality (Bonali et al., 2019c).

5.4 Future Developments

Regarding the future roadmap for our approach, we

expect a contribution from Neanias project

(https://www.neanias.eu/) regarding online services

devoted to SfM-MVS photogrammetry processing, in

order to enlarge the community working with this

technique, possibly involving also non-scientists and

non-academics. Up to now, we cannot find a robust

and efficient open source software for SfM-MVS

processing, that can be applied to both terrestrial and

marine environments; for the latter case, pictures are

usually collected using (Remotely Operated Vehicle)

ROVs. From another side, the Selfie drone-based

approach can motivate scientists to collect data in

remote and dangerous areas, thus multiplying the

amount of available 3D DOMs for the scientific

community as well as the storage dimension needed

for the DSMs, Orthomosaics and the dense clouds. In

view of the above, an online service for data storage

and 3D scientific visualisation is also recommended.

For example, the model related to the entire project of

the 1570-1573 AD crater is as large as 54 GBs, and it

is difficult to be shared with the scientific community,

even though the high resolution of the 3D DOM,

DSM and Orthomosaic can be crucial for better

understanding volcano dynamics, relationships with

release of volcanic gases, tectonic settings, magmatic

intrusions.

6 FINAL REMARKS

In the present work we tested the use of selfie drones

as a tool for 3D modelling, geological mapping and

data collection. Test sites are located in Santorini and

are characterised by different settings: i) the Crater

area in Nea Kameni has a mostly horizontal

topography; ii) the Vlychada beach has a dominant

vertical topography. By applying the Structure from

Motion techniques to pictures collected using the

selfie drone, we were capable of: i) reconstructing the

two sites with centimetric to sub-centimetric

resolution; ii) recognizing geological features on very

high resolution DSMs and Ortomosaics; iii) mapping

vertical volcanic deposits on 3D DOMs; and iv)

collecting new quantitative data for both sites.

ACKNOWLEDGEMENTS

This work is supported by: i) MIUR Argo3D project

ACPR15T4_00098 and ii) 3DTeLC Erasmus+

Project 2017-1-UK01-KA203-036719. This article is

also an outcome of Project MIUR – Dipartimenti di

Eccellenza 2018–2022 and of GeoVires, the Virtual

Reality Lab for Earth Sciences host at Department of

Earth and Environmental Sciences, University of

Milan Bicocca, U4, Piazza della Scienza 4, 20126

Milan, Italy. Agisoft Metashape is acknowledged for

photogrammetric data processing. Finally,

NEANIAS project is acknowledged for financial

support for the submission of this paper.

REFERENCES

Agisoft, L. L. C. "Agisoft metashape user manual,

Professional edition, Version 1.6." Agisoft LLC, St.

Petersburg, Russia, from https://www.agisoft.com/

pdf/metashape-pro_1_6_en.pdf, accessed March 12

(2020):2020.

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

126

Antoniou, V., Nomikou, P., Bardouli, P., Lampridou, D.,

Ioannou, Th., Kalisperakis, I., Stentoumis, Ch.,

Whitworth, M., Krokos, M. & Ragia, L. (2018). An

Interactive Story Map for the Methana Volcanic

Peninsula. In Proceedings of the 4th International

Conference on Geographical Information Systems

Theory, Applications and Management (GISTAM

2018), pages 68-78, ISBN: 978-989-758-294-3.

Antoniou, V.; Ragia, L.; Nomikou, P.; Bardouli, P.;

Lampridou, D.; Ioannou, T.; Kalisperakis, I.;

Stentoumis, C., 2018. Creating a Story Map Using

Geographic Information Systems to Explore

Geomorphology and History of Methana Peninsula.

ISPRS Int. J. Geo-Inf. 2018, 7, 484.

Antoniou, V., Nomikou, P., Bardouli, P., Sorotou, P.,

Bonali, F.L., Ragia, L. and Metaxas, A., 2019. The

Story Map for Metaxa Mine (Santorini, Greece): A

Unique Site Where History and Volcanology Meet

Each Other. In Proceedings of the 5th International

Conference on Geographical Information Systems

Theory, Applications and Management - Volume 1:

GISTAM, 212-219.

Antoniou, V., Nomikou, P., Zafeirakopoulou, E., Bardouli,

P., Ioannou, T., 2019a. Geo-biodiversity and cultural

environment of Nisyros volcano. 15th International

Congress of the Geological Society of Greece, Sp. Pub.

7 Ext. Abs. GSG2019-195, Athens, 22-24 May, 2019,

p. 716-717.

Antoniou, V., Nomikou, P., Papaspyropoulos, K.,

Vlasopoulos, O., Zafeirakopoulou, E., Bardouli, P.,

Chrysopoulou, E., 2019b. Geo-biodiversity and

Cultural Environment of the regions surrounding the

Corinth Gulf (Greece). Regional Conference on

Geomorphology, Focal Theme: Geomorphology of

Climatically and Tectonically Sensitive Areas,

UNESCO Global Geoparks: geoheritage assessment

and management - geo-tourism development, Abstract

book, p.256, Athens 2019.

Bonali, F. L., Corazzato, C., Tibaldi, A., 2011. Identifying

rift zones on volcanoes: an example from La Réunion

island, Indian Ocean. Bulletin of volcanology, 73(3),

347-366.

Bonali, F. L., Corazzato, C., Tibaldi, A., 2012. Elastic stress

interaction between faulting and volcanism in the

Olacapato–San Antonio de Los Cobres area (Puna

plateau, Argentina). Global and planetary change, 90,

104-120.

Bonali, F. L., Tibaldi, A., Marchese, F., Fallati, L., Russo,

E., Corselli, C., Savini, A., 2019a. UAV-based

surveying in volcano-tectonics: An example from the

Iceland rift. Journal of Structural Geology, 121, 46-64.

Bonali, F. L., Tibaldi, A., Mariotto, F. P., Saviano, D.,

Meloni, A., Sajovitz, P., 2019. Geometry, oblique

kinematics and extensional strain variation along a

diverging plate boundary: The example of the northern

Theistareykir Fissure Swarm, NE Iceland.

Tectonophysics, 756, 57-72.

Bonali, F. L., Russo, E., Savini, A., Marchese, F., Fallati,

L., Paraskevi, N., ... Garzonio, R., 2019c. Learning

outcomes from the EGU 2018 Public Engagement

grant" Shaping geological 3D virtual field-surveys for

overcoming motor disabilities". In Geophysical

Research Abstracts, Vol. 21.

Bond, A. & Sparks, R.S.J., 1976. The Minoan eruption of

Santorini, Greece, Journal of the Geological Society,

London, 132, 1–16.

Browning, J., Drymoni, K., Gudmundsson, A., 2015.

Forecasting magma-chamber rupture at Santorini

volcano, Greece. Scientific reports, 5, 15785.

Burns, J.H.R., Delparte, D., 2017. Comparison of

commercial structure-from-motion photogrammetry

software used for underwater three-dimensional

modeling of coral reef environments. In International

Archives of the Photogrammetry, Remote Sensing and

Spatial Information Sciences - ISPRS Archives (Vol.

42, pp. 127–131).

Druitt, T. H. & Francaviglia, V., 1992. Caldera formation

on Santorini and the physiography of the islands in the

late Bronze Age, Bull. Volcanol., 54, 484–493.

Druitt, T. H., 2014. New insights into the initiation and

venting of the Bronze-Age eruption of Santorini

(Greece), from component analysis, Bull. Volcanol., 76,

794.

Fallati, L., Polidori, A., Salvatore, C., Saponari, L., Savini,

A., Galli, P., 2019. Anthropogenic Marine Debris

assessment with Unmanned Aerial Vehicle imagery

and deep learning: A case study along the beaches of

the Republic of Maldives. Science of The Total

Environment, 693, 133581.

Friedrich, W., Kromer, B., Friedrich, M., Heinemeier, J.,

Pfeiffer, T., Talamo, S., 2006. Santorini eruption

radiocarbon dated to 1627–1600 Bc., Science, 312, 548.

Gerloni, I.G., Carchiolo, V., Vitello, F.R., Sciacca, E.,

Becciani, U., Costa, A., Riggi S., Bonali F.L., Russo E.,

Fallati L., Marchese F., Tibaldi A., 2018. Immersive

Virtual Reality for Earth Sciences. In 2018 Federated

Conference on Computer Science and Information

Systems (FedCSIS) (pp. 527-534). IEEE.

James, M.R., Robson, S., 2012. Straightforward

reconstruction of 3D surfaces and topography with a

camera: accuracy and geoscience application. Journal

of Geophysical Research, 117, F03017.

James, M.R., Robson, S., d'Oleire-Oltmanns, S.,

Niethammer, U., 2017. Optimising UAV topographic

surveys processed with structure-from-motion: ground

control quality, quantity and bundle adjustment.

Geomorphology, 280, 51-66.

Johnston, E. N., Sparks, R.S.J., Phillips, J.C. & Carey, S.,

2014. Revised estimates for the volume of the Late

Bronze Age Minoan eruption, Santorini. Greece, J.

Geol. Soc. Lond., 171, 583–590.

Krokos M., Bonali F., Vitello F., Antoniou V., Becciani U.,

Russo E., Marchese F., Fallati L., Nomikou P., Kearl

M., Sciacca, E., Whitworth M., 2019. Workflows for

virtual reality visualisation and navigation scenarios in

earth sciences. In 5th International Conference on

Geographical Information Systems Theory,

Applications and Management: GISTAM 2019.

SciTePress.

Selﬁe Drones for 3D Modelling, Geological Mapping and Data Collection: Key Examples from Santorini Volcanic Complex, Greece

127

Manning, S.W., Ramsey, C.B., Kutschera, W., Higham, T.,

Kromer, B., Steier, P., Wild, E.M.,2006. Chronology

for the Aegean Late Bronze Age 1700–1400 B.C.,

Science, 312, 565–569.

Nomikou P., Papanikolaou D., Alexandri M., Sakellariou

D., Rousakis G., 2013. Submarine volcanoes along the

Aegean volcanic arc. Tectonophysics, 597-598, pp.123.

Nomikou, P., Parks M.Μ., Papanikolaou D., Pyle D.Μ.,

Mather T.Α., Carey S., Watts A.Β., Paulatto M.,

Kalnins M., Livanos I., Bejelou K., Simou E., Perros I.,

2014. The emergence and growth of a submarine

volcano: The Kameni islands, Santorini (Greece).

GeoResJ, 1-2, pp.8.

Pasquaré Mariotto, F., Bonali, F.L., Venturini, C., 2020.

Iceland, an open-air museum for geoheritage and Earth

Science communication purposes. Special Issue

"Geoheritage and Geotourism Resources", accepted for

publication.

Pyle, D.M., Elliott, J.R., 2006. Quantitative morphology,

recent evolution, and future activity of the Kameni

Islands volcano, Santorini, Greece. Geosphere, 2(5),

253-268.

Reck, H., 1936. Santorini: Der Werdegang eines

Inselvulkans und sein Ausbruch, 1925–1928, Reinser,

Berlin.

Sigurdsson, H., Carey, S., Alexandri, M., Vougioukalakis,

G., Croff, K., Roman, C., Sakellariou, D., Anagnostou,

C., Rousakis, G., Ioakim, C., Gogou, A., Ballas, D.,

Misaridis, T., Nomikou, P., (2006). Marine

Investigations of Greece’s Santorini Volcanic Field.

Eos 87 (34), pp.337.

Stal, C., Bourgeois, J., De Maeyer, P., De Mulder, G., De

Wulf, A., Goossens, R., Hendrickx, M., Nuttens, T.,

Stichelbaut, B., 2012. Test case on the quality analysis

of structure from motion in airborne applications. In

32nd EARSeL Symposium: Advances in geosciences.

European Association of Remote Sensing Laboratories

(EARSeL).

Tavani, S., Granado, P., Corradetti, A., Girundo, M.,

Iannace, A., Arbués, P., Muñozb J.A., Mazzoli, S.,

2014. Building a virtual outcrop, extracting geological

information from it, and sharing the results in Google

Earth via OpenPlot and Photoscan: An example from

the Khaviz Anticline (Iran). Computers & Geosciences,

63, 44-53.

Tibaldi, A., 1995. Morphology of pyroclastic cones and

tectonics. Journal of Geophysical Research: Solid

Earth, 100 (B12), 24521-24535.

Tibaldi, A., Pasquarè, F. A., Papanikolaou, D., Nomikou,

P., 2008. Discovery of a huge sector collapse at the

Nisyros volcano, Greece, by on-land and offshore

geological-structural data. Journal of volcanology and

geothermal research, 177 (2), 485-499.

Tibaldi, A., Alania, V., Bonali, F. L., Enukidze, O.,

Tsereteli, N., Kvavadze, N., Varazanashvili, O., 2017.

Active inversion tectonics, simple shear folding and

back-thrusting at Rioni Basin, Georgia. Journal of

Structural Geology, 96, 35-53.

Trinks, I., Clegg, P., McCaffrey, K., Jones, R., Hobbs, R.,

Holdsworth, B., Holliman N., Imber, J., Waggott, S.,

Wilson, R., 2005. Mapping and analysing virtual

outcrops. Visual Geosciences, 10(1), 13-19.

Turner, D., Lucieer, A., Watson, C., 2012. An automated

technique for generating georectified mosaics from

ultra-high resolution unmanned aerial vehicle (UAV)

imagery, based on structure from motion (SfM) Point

clouds. Remote sensing, 4, 1392-1410.

Vollgger, S.A. and Cruden, A.R., 2016. Mapping folds and

fractures in basement and cover rocks using UAV

photogrammetry, Cape Liptrap and Cape Paterson,

Victoria, Australia. Journal of Structural Geology, 85,

168-187.

Watts, A.B., Nomikou P., Moore J.D.P., Parks M.M.,

Alexandri M., 2015. Historical bathymetric charts and

the evolution of Santorini submarine volcano, Greece.

Geochemistry, Geophysics, Geosystems, 16 (3), pp.847.

Westoby, M.J., Brasington, J., Glasser, N.F., Hambrey,

M.J., Reynolds, J.M., 2012. “Structure-from-Motion”

photogrammetry: a low-cost, effective tool for

geoscience applications. Geomorphology, 179, 300-

314.

Westoby, M.J., Brasington, J., Glasser, N.F., Hambrey,

M.J., Reynolds, J.M., 2012. “Structure-from-Motion”

photogrammetry: a low-cost, effective tool for

geoscience applications. Geomorphology, 179, 300-

314.

Xu, X., Aiken, C.L., Nielsen, K.C., 1999. Real time and the

virtual outcrop improve geological field mapping. Eos,

Transactions American Geophysical Union, 80(29),

317-324.

GISTAM 2020 - 6th International Conference on Geographical Information Systems Theory, Applications and Management

128