Multi-Resolution Modeling of the Tufa Formation Dynamic using

Close-Range Photogrammetry, Handheld 3D Scanner and

Terrestrial Laser Scanner

Ivan Marić

, Lovre Panđa

and Rina Milošević

University of Zadar, Department of Geography, Trg Kneza Višeslava 9, 23000 Zadar, Croatia

Keywords: Tufa Formation Dynamic, Artec Eva, Faro Focus M70, Close-Range Photogrammetry, NP Krka.

Abstract: Advances in geospatial technologies (GST) have revolutionized the ability to quantify spatio-temporal

changes in various geomorphological forms at different scales. One of the most complex geomorphological

forms are tufa and travertine cascades whose evolution is the subject of numerous research in various scientific

fields. In this paper, we are presenting a new methodological framework for analyzing tufa formation dynamic

(TFD) at three levels of research (macro-meso-micro) using the close-range photogrammetry (CRP) method,

handheld 3D scanner, and terrestrial laser scanner (TLS). The results, 3D models and digital elevation models

(DEMs), of the first (reference) measurement at three levels of research, are presented in this paper. Reference

models were generated using Agisoft Metashape, Artec Studio Professional 15, and SCENE software.

Measurements were done in an artificial tufa tunnel, located within the Jaruga, the second oldest hydroelectric

power plant in the world constructed within National Park Krka, Croatia. This tunnel is a specific tufa-forming

environment. The subject of the next paper will be the comparison of interval tufa 3D tufa models at three

levels of research and the calculation of volumetric (mm

-1

) and linear (mm a

−1

) tufa growth rates after two

years of exposure to the Krka River. The presented methodological framework will expand the knowledge

about TFD within this specific depositional sub-environment and can be applied in the dynamic formation

analysis of other hydroprecipitates.

1 INTRODUCTION

Calcium carbonate (CaCO3) precipitation is a feature

of many freshwater systems. Various names for

hydroprecipitates formed by this process exist (Viles

and Goudie, 1990). The most often are (1) tufa and

(2) travertine (Bonacci et al., 2017) which can be

found worldwide (Viles and Pentecost, 2007).

Sometimes misinterpretation of these names occurs in

the literature, although there are clear differences

between them (Ford and Pedley, 1996; Capezzuoli,

2014; Bonacci et al., 2017). Travertine is by the most

authors associated with the precipitate sedimented in

warm and hot hydrothermal waters (Cukrov et al.,

2010), while the tufa is secondary carbonate

sedimented in freshwater at ambient temperature and

usually includes the remains of micro- and

https://orcid.org/0000-0002-9723-6778

https://orcid.org/0000-0003-4549-4481

https://orcid.org/0000-0002-5473-2579

macrophytes, invertebrates, and bacteria (Ford and

Pedley, 1996; Pedley, 2000; Capezzuoli, 2014,

Barešić et al., 2021). These hydroprecipitates

represent one of the most spectacular depositional

forms in karst landscapes (Šiljeg, et al., 2020) which

have universal aesthetic and scientific values and are

often included in the UNESCO World Heritage List.

Analysis of tufa and travertine formation

dynamics are oriented to quantification of the growth

and erosion rates which can be measured using

various direct and indirect methods (Marić et al,

2020a) and expressed using different units. Accurate

calculation of rates is important for several reasons

(Marić, et al. 2020b). Recent advances in data

acquisition sensors have revolutionized the ability to

quantify different spatial-temporal changes on a wide

range of scales. In the majority of TFD studies, very

large variability in rates was recorded as a result of

Mari

c, I., Pan

da, L. and Miloševi

c, R.

Multi-Resolution Modeling of the Tufa Formation Dynamic using Close-Range Photogrammetry, Handheld 3D Scanner and Terrestrial Laser Scanner.

DOI: 10.5220/0010887000003185

In Proceedings of the 8th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2022), pages 75-82

ISBN: 978-989-758-571-5; ISSN: 2184-500X

specific sedimentation conditions, their spatial

distribution along the water body, and selected

measurement method (Arenas et al., 2010, Arenas et

al., 2014, Auqué et al., 2014).

In this research, we are presenting a new

framework for monitoring the tufa formation

dynamic (TFD) using the modern active and passive

3D coordinate measuring devices at three levels of

research within the tufa tunnel. To demonstrate the

new framework in the Jaruga hydroelectric power

plant tunnel, located within National park (NP) Krka

was chosen. The TFD will be analyzed using close-

range photogrammetry (CRP) method - micro level,

handheld 3D scanner - meso level, and terrestrial laser

scanner (TLS) - macro level of research.

The basic objective of the research was to expand

knowledge about TFD within this specific

depositional environment and then compare it with

the results of other tufa-forming systems in the world.

A detailed overview of the TFD research is available

in Marić et al. (2020a). Furthermore, the objective is

to determine are there any differences in the TFD at

different levels of research using the above-

mentioned sensors. Also, since TFD will be analyzed

on an empty and smooth limestone plate (PL) and

rough substrate of the tunnel lateral walls, which is

composed of tufa, plant fragments, microorganisms,

and moss, it will be interesting to see whether the

growth and erosion rate will be similar. Namely, in

order for tufa formation to occur, in addition to

CaCO3 oversaturation, it is desirable that the water

contains a suitable substrate for calcite deposition.

They can be provided by organisms or organic

substances such as plant fragments and moss.

Therefore, one of the basic hypotheses of this and

future research is that the expected growth rate will

be higher on rougher surfaces (meso level of

research), ie on already formed tufa (tunnel lateral

wall).

Thus, it can be tested whether the tufa growth rate

determined on the lower level of research can be

extrapolated on the highest level of research. TLS was

used on a macro level of research with the aim of

examining the possibility of measuring TFD using

this type of sensor since such research has not yet

been done according to our knowledge.

2 STUDY AREA

Hydroelectric power plant Jaruga is located on the

Krka River, in Šibenik-Knin County (Croatia). It is

the second oldest hydroelectric power plant in the

world and the first in Europe. It was built near the

waterfall Skradinski buk.

Figure 1: Location of Jaruga HE power plant in NP Krka.

The Jaruga system includes a separate supply

structure within the small closed part of the Krka in

the area of the Skradinski Buk waterfall. The supply

structure consists of a tufa tunnel (Figure 3A) with

gravity flow, a concrete channel with almost vertical

sides (Figure 3B) leading to the water tank, and two

water flow regulators leading to the turbines. The

length of the tufa tunnel is around 82.7 meters

(Holjevac and Kuzle, 2019). On these almost vertical

sides of the tunnel tufa is forming (Figure 3B) and

periodically removed, depending on the growth rate

(Figure 2).

Figure 2: Removal of the tufa formed on the lateral sides of

tunnel (URL 1).

Given the fact that during the year the flow rate

and water level in this tunnel are more or less constant

and that certain parts of the tunnel are completely in

the dark (Figure 3C), while other parts are constantly

exposed to light, it is obvious why this area is

recognized as an excellent depositional environment

for analyzing TFD using the advance active and

passive sensors at a different level of research.

GISTAM 2022 - 8th International Conference on Geographical Information Systems Theory, Applications and Management

Figure 3: (A) Lateral sides of tufa tunnel; (B) formed tufa;

3 MATERIALS AND METHODS

3.1 Data Acquisition

Data acquisition was done using (A) coordinate

measuring macro-photogrammetry device (CMD)

(Marić et al. 2020a) at micro-level of research; (B) 3D

hand-held scanner Artec Eva at meso level of

research; and (C) TLS Faro M70 at the macro level

of research. The water in the tunnel was stopped for

the conduction of the measurements which were all

made in one day.

Figure 4: Sensors used in three level of research at tufa

tunnel.

3.1.1 CMD: Micro Level

On the micro-level of research, the TFD was

monitored by CMD on an area of 16 cm

. The CMD

consists of six main parts and was designed by Marić

et al. (2020a). It minimizes the frequent problems that

occur in the CRP process and when using a modified

micro-erosion meter (Drysdale, Gillieson, 1997). The

area of 16 cm² represents the upper part of the

limestone plate (PL) mounted on the lateral sides of

the tufa tunnel (Figure 5A). A total of three PLs were

mounted, one in the always illuminated area, the

second in constant dark (Figure 5B), and the third at

location with interchangeable conditions. The PL

design is described in Marić et al. (2020a). PLs were

measured with CMD before installation.

Figure 5: (A) Limestone PL measured with CMD; (B)

instalation of PL on the lateral sides of tunnel in no-light

condition.

3.1.2 Artec Eva: Meso Level

Artec Eva was used at the meso level of research. It is

a lightweight and compact structured light 3D hand-

held scanner that uses triangulation and structured

light for data collection. Its 3D resolution is up to 0.5

mm at a distance of 40 cm to 1 m. The accuracy of the

3D is up to 0.1 mm (Marić et al., 2020b).

Figure 6: (A) Square surfaces scanned using Artec Eva; (B)

colored caps used as GCPs.

Multi-Resolution Modeling of the Tufa Formation Dynamic using Close-Range Photogrammetry, Handheld 3D Scanner and Terrestrial

Laser Scanner

The TFD was monitored by Artec Eva on three

surfaces of around 1 m

. Three square surfaces were

marked and scanned on the lateral sides of the tunnel

(Figure 6A). The frame square surfaces were marked

with four screws that had colored caps (Figure 6B).

The screws were used as ground control points

(GCPs). Also, they were useful for the scanning

procedure. GCPs were necessary because the 3D

scanner axes (X, Y, Z) are relative to the scene being

scanned. Therefore, the GCPs eases the alignment of

the interval tufa 3D model enabling the interval

comparison of the formed tufa and quantification of

growth and erosion rates between interval models.

A distance adjustment indicator was used while

scanning marked test surfaces. Localization of Artec

Eva was achieved by moving it away or closer to

surfaces in order to get the best scanning quality

(Figure 7). The speed of scanning was around 13.5

fps. Scanning of the one test surface lasted around 3

minutes.

Figure 7: Scanning of test surface using Artec Eva.

3.1.3 Faro Focus M70: Macro Level

At the macro level of the research, terrestrial scanning

of the whole tufa tunnel on October 10, 2020, was

performed using a FARO Focus M70 laser scanner

(Figure 8). FARO Focus M70 is a phase laser 3D

scanner with a measuring range from 0.6 m to 70 m,

with a measuring accuracy of ± 3 mm. It collects

about 488,000 points per second and has a level of

protection against water and dust.

There were a total of nine scanning locations. The

total number of scans was determined with respect to

available survey time which was only one day. Within

that time, data acquisition had to be made at all three

levels of research. Ideally, the number of TLS

scanning locations can be higher. The optimal

positions for survey reference targets were identified

in the field following examples of good practice

(Domazetović et al., 2020). Survey reference targets

(spheres) were used for faster and easier registration

of the multiple scans. Spheres were set at a distance

of 15–20 meters to enable target registration. Since,

this type of research requires an interval comparison

of 3D tufa models on all three levels of research, four

targets (spheres) have been fixed on exact XYZ

location. The X, Y, and Z coordinates of spheres will

not differ between interval scanning. The XYZ

coordinates of the spheres were not measured using

GNSS receivers because in this type of research the

location of the tunnel or survey area within the global

coordinate system is not important. However, the

relative positional accuracy is very important, ie

overlapping of interval models. Therefore, all spheres

were fixed and their location will not change between

interval scanning.

The following scanning settings were set: 1/4

resolution (with a point distance of approximately 6

mm at 10 m distance) and 2× quality. Also, the HDR

texture was set. The scanning per one location took

6–7 min.

Figure 8: Terestrial laser scanning of tufa tunnel.

3.2 Data Processing

3.2.1 Micro Level Data Processing

At the micro-level of research image workflow

process (capturing + processing) was done following

the guidelines proposed by James et al. (2019). Image

capture was done using CMD which contained

system sensors NIKON D5300 and macro-lens Venus

LAOWA 60-mm f/2.8. Each PL was represented with

more than 180 overlapping (>80%) images. Initial

(reference) 3D tufa models and digital elevation

models (DEMs) were derived in Agisoft Metashape

1.5.1. The overall methodological process is shown in

Figure 9. Four GCS and four CP (checkpoint) were

used to georeference and assess the X, Y, and Z

accuracy of the models.

GISTAM 2022 - 8th International Conference on Geographical Information Systems Theory, Applications and Management

Figure 9: Image processing in Agisoft Metashape.

3.2.2 Meso Level Data Processing

At the meso level of research, scanned surfaces were

processed using Artec Studio 15 Professional,

software for professional 3D data processing. The

overall methodological process is given in Figure 10.

Figure 10: Data processing in Artec Studio 15 Professional.

After the second and subsequent scans, which will be

presented in future research, the data processing will

include the sixth step (alignment). It refers to the

placement of the interval 3D tufa models in the same

coordinate system. Align of interval models will be

done using the Rigid Align option. The center of the

colored caps mounted on top of the screws (Figure

11) will serve as fixed GCPs for each reconstructed

3D model. Then using the Measure option which

contains the Distance map and Section, volume, tufa

growth, and erosion rates will be measured.

Figure 11: Fixed GCP used in meso level of research.

3.2.3 Macro Level Data Processing

At the macro level of research collected 9 scans were

processed using FARO SCENE 2019 software. It was

used for scan registration and creation of dense point

cloud and polygonal model of whole scanned site

(tufa channel). Scan registration was done using

Target Based mode in the Automatic Registration

option. This setting causes SCENE to look only for

artificial targets (spheres) in scans. Figure 12 shows

recognized spheres at the eighth scan location.

Figure 12: Recognized spheres at eighth scan station.

4 RESULTS

This paper was conceived as an introduction to the

possibility of using different sensors in analyzing the

TFD and to expand knowledge about TFD through

various aspects mentioned in the introduction.

Therefore, the results section contains only the initial

models that will be used in further research. Namely,

numerical results comparisons between derived

models were not done because it requires another data

acquisition process at all three levels of research that

has not yet been carried out.

4.1 3D Tufa: Micro Level

3D tufa models at the micro-level of research were

created in Agisoft Metashape. Polygonal mesh

models of PLs were generated based on the

reconstructed dense point cloud. The Surface type

parameter was set to Arbitrary (3D). The Face count

parameter to "0" to skip the decimation stage so the

model has all the reconstructed faces. Based on this

reference and new interval 3D models which will be

recorded in the following periods (2-year interval),

volumetric (mm

-1

) TGRs will be calculated. Also,

DEMs of PLs were derived. Based on them and new

interval tufa DEMs linear TGRs (mm a

-1

) will be

calculated (Figure 13). Volumetric (mm

-1

) TGRs

will be calculated in Artec Studio 15 Professional

after two years of exposure PLs to water. Linear

TGRs (mm a

-1

) will be calculated in ArcMap

software.

Multi-Resolution Modeling of the Tufa Formation Dynamic using Close-Range Photogrammetry, Handheld 3D Scanner and Terrestrial

Laser Scanner

Figure 13: Derived reference 3D models and tufa DEMs on

micro level of research.

A field survey was done one year (October 2021)

after the PLs were mounted and a few interesting

things were spotted. Tufa-forming organisms of

tubular shape were observed on the PL which was

placed in complete darkness (Figure 14B). No

organisms were observed on the other PLs. We

visually estimated that the highest TGR was on the

PL that was most exposed to the light (Figure 14A).

Figure 14: (A) PL with highest increment after one-year;

(B) observed tufa-forming organisms.

4.2 3D Tufa: Meso Level

3D tufa models at meso level of research were created

in Artec Studio 15 Professional. Figure 15. shows

reconstructed models of two scanned surfaces (A and

B).

It will be very interesting to measure the erosion

of tufa in interval scanning periods. Namely, erosion

occurs mainly due to intensive flow due to increased

precipitation, dry periods during summer that leads to

drying and weakening the tufa structure, human

factor, measurement which coincides with the phase

of periphyton loss due to emigration or more

intensive grazing, etc. The PLs and test surfaces at the

micro and meso level of research are located at a

depth of a few meters while the water level and water

flow rate are constant, which is why no significant

erosion rates are expected. Also, it will be interesting

to see is there any difference in the erosion of tufa

located in constant darkness (inside the tunnel) for

two years and tufa exposed to light.

Figure 15: 3D model of two scanned surfaces in tufa tunnel.

One year after the scanning it was observed that the

tufa grew faster on the lateral walls which are more

exposed to the light and rich with moss (Figure 16A).

Also, differences in tufa hardness are obvious. The

tufa that forms in conditions of constant darkness is

much harder (Figure 16B), while the others are softer

and more porous.

Figure 16: (A) Test surface with highest increment; (B)

hard sedimented tufa inside the tunnel.

4.3 3D Tufa: Macro Level

At the macro-level of research, scans were registered

using manual registration mode. The mean horizontal

target error was 0.4 mm, while the mean vertical

target error was 2.0 mm. Registered scans were used

for the creation of dense point cloud (Figure 17). It

covered the entire area of the tufa tunnels' lateral

sides. After the two years of exposure to water TFD

on these lateral sides will be calculated using

CloudCompare software.

GISTAM 2022 - 8th International Conference on Geographical Information Systems Theory, Applications and Management

Figure 17: Point cloud of tufa tunnel section with four scan

location.

5 CONCLUSIONS

In this paper, we presented a new methodological

framework for analyzing (TFD) at three levels of

research using state-of-art active and passive sensors.

The 3D models of the first (reference)

measurement are presented fot three levels of

research. Although the next measurements will be

made in a year, interesting occurrences have already

been noticed. Tufa-forming organisms of tubular

shape were observed on the PL which was placed in

complete darkness. Furthermore, tufa grew faster on

the lateral walls which are more exposed to the light

and rich with moss than in the locations within

complete darkness and on smooth surfaces of PLs.

The subject of the next paper will be the

comparison of derived 3D models at all levels of

research and the calculation of the volumetric (mm

-1

) and linear (mm a

−1

) tufa growth rates. Also, in

future research, we are considering implementing a

hyperspectral camera at the micro and meso level of

research in order to analyze spectral reflection and

detect the plant fragments and macroinvertebrates

accumulated on the tufa surface.

ACKNOWLEDGEMENTS

This research was performed within the project UIP-

2017-05-2694 financially supported by the Croatian

Science Foundation. Authors would like to thank

administration of the NP Krka.

REFERENCES

Arenas, C., Vázquez‐Urbez, M., Auqué, L., Sancho, C.,

Osácar, C.,Pardo, G. (2014): Intrinsic and extrinsic

controls of spatial and temporal variations in modern

fluvial tufa sedimentation: A thirteen - year record from

a semi-arid environment, Sedimentology, 61(1), 90-132.

Arenas, C., Osácar, C., Sancho, C., Vázquez-Urbez, M.,

Auqué, L., Pardo, G. (2010). Seasonal record from

recent fluvial tufa deposits (Monasterio de Piedra, NE

Spain): sedimentological and stable isotope data,

Geological Society, London, Special Publications,

336(1), 119-142.

Auqué, L., Arenas, C., Osácar, C., Pardo, G., Sancho, C.,

Vázquez-Urbez, M. (2014). Current tufa sedimentation

in a changing-slope valley: the River Añamaza (Iberian

Range, NE Spain), Sedimentary Geology, 303, 26-48.

Barešić, J., Faivre, S., Sironić, A., Borković, D., Lovrenčić

Mikelić, I., Drysdale, R.N., Krajcar Bronić, I. (2021):

The potential of tufa as a tool for paleoenvironmental

research-a study of tufa from the Zrmanja river canyon,

Croatia, Geoscience, 11, 376.

Capezzuoli, E., Gandin, A., Pedley, M. (2014): Decoding

tufa and travertine (fresh water carbonates) in the

sedimentary record: the state of the art, Sedimentology,

61(1), 1-21.

Cukrov, N., Surić, M., Fuček, L., Ćosović, V., Korbar, T.,

Juračić, M. (2010). Geology of Krka River estuary, u:

Vodič ekskurzija–Excursion Guide-book (4. hrvatski

geološki kongres), Hrvatski geološki institut.

Bonacci, O., Roje-Bonacci, R., Andrić, I. (2017). Prilog

izučavanju hidrologije Skradinskog buka na rijeci Krki,

Hrvatske vode, 25(99), 27-36.

Domazetović, F., Šiljeg, A., Marić, I. (2020) Guidelines for

optimization of terrestrial laser scanning surveys over

gully erosion affected areas. In: Massimiliano Alvioli,

Ivan Marchesini, Laura Melelli & Peter Guth, eds.,

Proceedings of the Geomorphometry 2020 Conference,

220-223.

Drysdale, R., Gillieson, D. (1997). Micro‐erosion meter

measurements of travertine deposition rates: a case

study from Louie Creek, Northwest Queensland,

Australia. Earth Surface Processes and Landforms: The

Journal of the British Geomorphological Group,

22(11), 1037-1051.

Ford, T. D., Pedley, H. M. (1996). A review of tufa and

travertine deposits of the world, Earth-Science

Reviews, 41(3-4), 117-175.

Holjevac, N., Kuzle, I. (2019). Prvi cjeloviti višefazni

elektroenergetski sustav na svijetu–Krka Šibenik.

Annual of the Croatian Academy of Engineering,

2019(1), 162-174.

James, M. R., Chandler, J. H., Eltner, A., Fraser, C., Miller,

P. E., Mills, J. P., Noble, T., Robson, S., Lane, S. N.

(2019). Guidelines on the use of structure from motion

Multi-Resolution Modeling of the Tufa Formation Dynamic using Close-Range Photogrammetry, Handheld 3D Scanner and Terrestrial

Laser Scanner

photogrammetry in geomorphic research. Earth

Surface Processes and Landforms, 44(10), 2081-2084.

Marić, I., Šiljeg, A., Cukrov, N., Roland, V., Domazetović,

F. (2020a). How fast does tufa grow? Very high -

resolution measurement of the tufa growth rate on

artificial substrates by the development of a contactless

image-based modelling device. Earth Surface

Processes and Landforms, 45(10), 2331-2349.

Marić, I., Šiljeg, A., Domazetović, F., Cukrov, N. (2020b).

A framework for using handheld 3D surface scanners

in quantifying the volumetric tufa growth. In:

Massimiliano Alvioli, Ivan Marchesini, Laura Melelli

& Peter Guth, eds., Proceedings of the

Geomorphometry 2020 Conference, 18-21.

Pedley, M. (2000). Ambient temperature freshwater

microbial tufas. In Microbial sediments (pp. 179-186).

Springer, Berlin, Heidelberg.

Šiljeg, A., Marić, I., Cukrov, N., Domazetović, F., Roland,

V. (2020). A Multiscale Framework for Sustainable

Management of Tufa-Forming Watercourses: A Case

Study of National Park “Krka”, Croatia. Water, 12(11),

3096.

URL 1 http://st-gramit.com/projekti_jaruga.html, 6 June,

2021.

Viles, H. A., Goudie, A. S. (1990). Tufas, travertines and

allied carbonate deposits, Progress in Physical

Geography, 14(1), 19-41.

Viles, H., Pentecost, A. (2007). Tufa and travertine, in:

Geochemical Sediments and Landscapes, eds. D. J. Nash,

S. J. McLaren, Blackwell Publishing Ltd., 173-199.

GISTAM 2022 - 8th International Conference on Geographical Information Systems Theory, Applications and Management