86
ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2020 Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1936 Seismic investigations and physical property studies of natural resources in Finland and Sweden Efficient exploration of groundwater and mineral resources GEORGIANA MARIES ISSN 1651-6214 ISBN 978-91-513-0953-8 urn:nbn:se:uu:diva-408712

Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 1936

Seismic investigations and physicalproperty studies of naturalresources in Finland and Sweden

Efficient exploration of groundwater and mineralresources

GEORGIANA MARIES

ISSN 1651-6214ISBN 978-91-513-0953-8urn:nbn:se:uu:diva-408712

Page 2: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Dissertation presented at Uppsala University to be publicly examined in Hambergsalen,Geocentrum, Villavägen 16, Uppsala, Friday, 5 June 2020 at 10:00 for the degree of Doctorof Philosophy. The examination will be conducted in English. Faculty examiner: AssociateProfessor Musa Manzi (University of the Witwatersrand, South Africa).

AbstractMaries, G. 2020. Seismic investigations and physical property studies of natural resourcesin Finland and Sweden. Efficient exploration of groundwater and mineral resources. DigitalComprehensive Summaries of Uppsala Dissertations from the Faculty of Science andTechnology 1936. 84 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0953-8.

Natural resources, such as mineral deposits and groundwater in particular, are crucial for oursociety, as the world prepares itself for a smooth transition towards green technologies anddecarbonization. Apart from extraction and use, innovative mineral exploration solutions areneeded to complete the full value chain and to achieve the sustainable development goals.

The application of seismic methods for both near-surface environmental and deep mineralexploration investigations is known, but high costs are associated with data acquisition andprocessing. In order to illustrate the potential of the seismic methods for efficient explorationof groundwater and mineral resources, cost-effective seismic surveys were acquired withintwo locations in Finland and Sweden, for aquifer delineation and imaging of iron-oxidemineralization in a hardrock environment, respectively. Physical properties, obtained fromgeophysical downhole logging and laboratory measurements, were analyzed for a completecharacterization of the mineralization and its host rocks. 3D ray-tracing and 2D finite-difference forward modeling were carried out for better assessing the seismic response of themineralization.

The effectiveness of these seismic surveys was revealed by the quality seismic data acquiredusing a low-cost, easily operated seismic source and different sensors, including a broadbandseismic landstreamer. In particular, the seismic source provided adequate penetration in twodifferent and challenging environments, namely soft glacial sediments at Virttaankangas,southwest Finland, and swampy glacial cover at Blötberget, south central Sweden. The large-scale units of the Virttaankangas aquifer were successfully delineated and integrated withthe hydrogeological units of the groundwater flow model. The mineralization at Blötbergetwas interpreted to further extend 300-400 m downdip, below the currently known depthfrom borehole observations. 3D processing of the 2D seismic profiles revealed a lateralextent of least 300 m, providing encouraging results for improved assessments of the mineralresources. The reflection pattern validated through forward modeling, suggested a possible newmineralized horizon below the known deposits. Physical property studies helped characterize themineralization and its host rocks in terms of seismic attenuation and rock quality. Fracture zonesdetected through sonic full-waveform logging were associated with high seismic attenuation,suggesting low mechanical competence of the mineralized rocks despite good rock qualitydesignation, providing thus important information for mine planning and exploration.

The studies presented in this thesis illustrate the potential of seismic methods and physicalproperty studies for efficient natural resources exploration in crystalline rocks and in overlyingglacial sediments.

Keywords: Seismic reflection, seismic refraction, esker, glacial sediments, iron-oxide,hardrock, efficient exploration, physical property, forward modeling, seismic quality factor,Finland, Sweden, Bergslagen

Georgiana Maries, Department of Earth Sciences, Geophysics, Villav. 16, Uppsala University,SE-75236 Uppsala, Sweden.

© Georgiana Maries 2020

ISSN 1651-6214ISBN 978-91-513-0953-8urn:nbn:se:uu:diva-408712 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-408712)

Page 3: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

To my family

Page 4: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Supervisor Professor Alireza Malehmir Department of Earth Sciences – Geophysics Program Uppsala University, Sweden

Co-supervisor Professor Christopher Juhlin Department of Earth Sciences – Geophysics Program Uppsala University, Sweden

Faculty opponent Associate Professor Musa Manzi School of Geosciences – Geophysics Program University of the Witwatersrand, South Africa

Committee members Dr. Beatriz Benjumea Department of Geophysics Cartographic Institute of Catalonia, Spain

Associate Professor Ayse Kaslilar Sisman Department of Earth Sciences – Geophysics Program Uppsala University, Sweden

Professor Ramon Carbonell Spanish National Research Council Spain

Dr. Bjarne Almqvist (reserve) Department of Earth Sciences – Geophysics Program Uppsala University, Sweden

Page 5: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Maries, G., Ahokangas, E., Mäkinen, J., Pasanen, A., Malehmir, A. (2017) Interlobate esker architecture and relatedhydrogeological features derived from a combination of high-resolution reflection seismics and refraction tomography,Virttaankangas, southwest Finland. Hydrogeology Journal, 25,829–845.

II Maries, G., Malehmir, A., Bäckström, E., Schön, M., Marsden, P. (2017) Downhole physical property logging for iron-oxideexploration, rock quality, and mining: An example from centralSweden. Ore Geology Reviews, 90, 1–13.

III Markovic, M., Maries, G., Malehmir, A., von Ketelhodt, J., Bäckström, E., Schön, M., Marsden, P. (2020) Deep reflection seismic imaging of iron-oxide deposits in the Ludvika mining area of central Sweden. Geophysical Prospecting, 68, 7–23.

IV Maries, G., Malehmir, A., Marsden, P. (2020) Cross-profile seismic data acquisition, imaging and modeling of iron-oxide deposits: A case study from Blötberget, south central Sweden. Submitted to Geophysics

Reprints were made with permission from the respective publishers. Additionally, during my PhD studies, I have contributed to the following papers that are not included in this thesis.

Ahokangas, E., Maries, G., Mäkinen, J., Pasanen, A., Malehmir, A. (2016) Seismic Imaging of Esker Sediments within the Satakunta Sandstone Depres-sion in Köyliö, SW Finland. In 22nd European Meeting of Environmental and Engineering Geophysics, Extended Abstract.

Maries, G., Malehmir, A., Backstrom, E. (2016) Downhole Physical Properties Measurements Supporting Iron-oxide Deep Exploration and Mining in Blötberget, Sweden. In 22nd European Meeting of Environmental and Engineering Geophysics, Extended Abstract.

Malehmir, A., Andersson, M., Mehta, S., Brodic, B., Munier, R., Place, J., Maries, G., Smith, C., Kamm, J., Bastani, M., & Mikko, H. (2016) Postglacial

Page 6: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

reactivation of the Bollnäs fault, central Sweden; a multidisciplinary geophysical investigation. Solid Earth, 7(2), 509–527.

Malehmir, A., Heinonen, S., Dehghannejad, M., Heino, P., Maries, G., Karell, F., Suikkanen, M., Salo, A. (2017) Landstreamer seismics and physical property measurements in the Siilinjärvi open-pit apatite (phosphate) mine, central Finland. Geophysics, 82, B29–B48.

Malehmir, A., Maries G., Backstrom, E., Schön, M., Marsden, P. (2017) Developing cost-effective seismic mineral exploration methods using a landstreamer and a drophammer. Scientific Reports, 7, 10325.

Maries, G., Malehmir, A., Backström, E., Schön, M., Marsden, P. (2017) Reflection Seismic Imaging of Iron-oxide Deposits - An Example from Bergslagen Mining District of Sweden. In 23rd European Meeting of Environmental and Engineering Geophysics, Extended Abstract.

Brodic, B., Malehmir, A., Pugin, A., Maries, G., (2018) Three-component seismic land streamer study of an esker architecture through S- and surface-wave imaging. Geophysics, 83, B339–B353.

Maries, G., Malehmir, A., Marsden, P. (2019) Cross-profile 2D recording for 3D imaging of iron-oxide mineralization in Blötberget, south-central Sweden. In 16th SAGA Biennial Conference & Exhibition 2019, Durban, South Africa, Extended Abstract.

Page 7: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Contributions

The papers included in this thesis are the result of close collaborations with several colleagues. My contributions in each paper are summarized below:

Paper I: I participated in the fieldwork in August 2014 and I processed the seismic data with the help of A. Malehmir. I wrote the initial version of the manuscript, and prepared most of the figures. My Finnish co-authors E. Ahokangas and J. Mäkinen helped with the detailed interpretation of the seismic reflection data and all improved the text through a number of iterations.

Paper II: I participated in most of the data acquisition campaigns (downhole logging) and performed all the density measurements along the drilled cores at the core facility. I plotted all the physical properties, calculated the seismic quality factor and wrote the first draft of the manuscript. Together with the co-authors, the manuscript was refined.

Paper III: I participated in data acquisition, in both field campaigns, 2015 and 2016. I obtained preliminary results from processing the 2016 seismic data. I prepared some of the figures of the manuscript and wrote the first draft of manuscript together with M. Markovic. All authors improved the manuscript through several iterations.

Paper IV: I processed the 2016 seismic data that was not included in Paper III, including the 3D processing of the merged data sets and 3D ray-tracing modeling. With the help of A. Malehmir, I performed the 2D finite-difference forward modeling. I wrote the initial version of the manuscript and together with the co-authors improved the text through several iterations.

Page 8: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015
Page 9: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Contents

1 Introduction ......................................................................................... 13 1.1 Motivation and objectives ............................................................... 13 1.2 Outline of the thesis ........................................................................ 15 1.3 Projects ............................................................................................ 16

1.3.1 Formas funded Trust GeoInfra .............................................. 16 1.3.2 ERA-MIN StartGeoDelineation ............................................ 16 1.3.3 H2020-funded Smart Exploration .......................................... 16

2 Study areas ........................................................................................... 17 2.1 Virttaankangas ................................................................................ 17

2.1.1 Geology and geomorphology ................................................. 17 2.1.2 Eskers ..................................................................................... 18 2.1.3 Eskers and groundwater resources ......................................... 19 2.1.4 Esker architecture .................................................................. 20 2.1.5 Previous and current studies .................................................. 21

2.2 Blötberget ........................................................................................ 21 2.2.1 Geology .................................................................................. 21 2.2.2 Mining history ....................................................................... 22 2.2.3 Mineral resources ................................................................... 24 2.2.4 Previous and current studies .................................................. 25

3 Methods ............................................................................................... 27 3.1 Seismic methods ............................................................................. 27

3.1.1 Seismic waves and elastic constants ...................................... 28 3.1.2 Seismic modeling ................................................................... 30

3.2 Physical property studies ................................................................ 31 3.2.1 Downhole logging.................................................................. 31 3.2.2 Seismic quality factor ............................................................ 32 3.2.3 Laboratory measurements ...................................................... 33

4 Summary of papers .............................................................................. 35 4.1 Paper I: Interlobate esker architecture and related hydrogeological features derived from a combination of high-resolution reflection seismics and refraction tomography, Virttaankangas, southwest Finland ........................................................... 35

4.1.1 Summary ................................................................................ 35 4.1.2 Conclusions ............................................................................ 44

Page 10: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

4.2 Paper II: Downhole physical property logging for iron-oxide exploration, rock quality, and mining: An example from central Sweden ..................................................................................................... 45

4.2.1 Summary ................................................................................ 45 4.2.2 Conclusions ............................................................................ 52

4.3 Paper III: Deep reflection seismic imaging of iron-oxide deposits in the Ludvika mining area of central Sweden ........................... 53

4.3.1 Summary ................................................................................ 53 4.3.2 Conclusions ............................................................................ 57

4.4 Paper IV: Cross-profile seismic data acquisition, imaging and modeling of iron-oxide deposits: A case study from Blötberget, south central Sweden ......................................................................................... 58

4.4.1 Summary ................................................................................ 58 4.4.2 Conclusions ............................................................................ 67

5 Conclusions ......................................................................................... 68

6 Outlook ................................................................................................ 70

7 Summary in Swedish ........................................................................... 71

Acknowledgements ....................................................................................... 73

References ..................................................................................................... 76

Page 11: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Abbreviations and acronyms

2D Two dimensional 3D Three dimensional a.s.l. Above sea level CDP Common depth point E Young modulus FDCT Fast discrete curvelet transform Fe Iron Ga Billion years ago GTK Geological Survey of Finland GPR Ground penetrating radar Hz Hertz k Bulk modulus kg Kilogram km Kilometer m Meter Ma Million years ago MAR Managed aquifer recharge MEMS Micro-electro mechanical system MUKH Morphologically undetected kettle hole NIO Nordic Iron Ore QP Seismic quality factor for P-waves QR Rock mass quality factor QS Seismic quality factor for S-waves RQD Rock quality designation RMS Root mean square s Second sp Seismic parameter SGU Geological Survey of Sweden TRUST Transparent Underground Structures UN United Nations VP Compressional velocity VS Shear velocity VSP Vertical seismic profiling μ Shear modulus ρ Density υ Poisson’s ratio

Page 12: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015
Page 13: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

13

1 Introduction

1.1 Motivation and objectives Natural resources, and mineral deposits in particular, are crucial for almost every sector of the global economy. Perhaps, even more important is how these resources are discovered and extracted from the earth, as there is a growing concern on the enduring environmental impacts of mining. Some estimates state that global material resource use is likely to more than double by 2050 if the current consumption trend continues (UN Environment, 2017). Between 1980 and 2005, the extraction of metal ores increased by more than 65% (World Resources Forum, 2020), and over the last decade at a faster rate than even the economic growth (UN Environment, 2020). Their importance for industrial development is undeniable, as this trend is forecasted to steadily increase, even more as worldwide transition towards renewable energy sources requires larger amounts of metals.

UN reports (International Resource Panel, 2020) suggest that more efficient extraction and use of natural resources could be “one of the most effective, and cost-effective, ways to reduce environmental impacts, including pollution, while at the same time advancing human well-being”. It is undisputable that the global economy must embrace resource efficiency to achieve its goals of a sustainable future for our society, and an efficient extraction and use of natural resources should start with effective exploration methods.

Geophysical methods offer a non-invasive way to map the subsurface for many environmental, urban, engineering or mining applications (Telford et al., 1990; Sheriff and Geldart, 1995). One of the great advantages of seismic methods stands in their versatility. With suitable acquisition set up and adequate processing techniques they can offer high-resolution imaging for both near-surface (e.g. environmental or engineering) applications (Miller and Steeples, 1994; Bradford et al., 1998; Pugin and Pullan, 2000; Juhlin et al., 2002; Pugin et al., 2014) and for deep (>500 m) investigations, such as for mineral and geothermal exploration (Milkereit et al., 1996; Eaton et al., 2003; Malehmir et al., 2012 and references therein; Buske et al., 2015 and references therein; Manzi et al., 2012, 2020; Hloušek et al., 2015). Routine application of seismic methods for these purposes can, however, be deterred by the high cost (both for data acquisition and processing) associated with seismic surveys. Seismic surveys acquired at two different study areas within this

Page 14: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

14

thesis prove that seismic data can be acquired in a cost-effective and efficient manner. The case studies sum up the challenges and solutions within different environments and applications: aquifer delineation in shallow glacial sediments (Paper I) and iron-oxide mineralization imaging in a hardrock environment (Papers III, IV). Integration with other types of data is essential, particularly for interpretation purposes, and when modeling is carried out. Paper I illustrates how seismic data are integrated with hydrogeological units of a 3D groundwater flow model, whereas Paper II brings in an additional physical property study for aiding mineral exploration and mine planning in a hardrock environment.

It is noteworthy defining hardrock environments. Geological settings where extensive igneous activity has occurred, and may have undergone metamorphism, are generally termed as hardrock (Hall, 1988). These environments can also include metamorphosed sedimentary rocks. Hardrock seismic exploration commonly refers to the application of seismic methods in such crystalline terrains with a high background medium velocity (e.g., > 5000 m/s). The potential of seismic methods for mineral exploration in hardrock environments is largely known for many years (Nelson, 1984; Durrheim and Maccelari, 1991; Reed, 1993 and references therein; Pretorius et al., 2000, 2003; Malehmir et al., 2012 and references therein; Cheraghi et al., 2020; Manzi et al., 2020), but routine application is still somewhat hindered by the large associated costs. The advance of low-cost sources and sensors in particular (e.g., Brodic et al., 2015; Malehmir et al., 2017b,c) has made it easier to employ seismic methods at the exploration stage, a trend illustrated in Papers III and IV. However, since the shallow resources became mined out or small ones near the surface are non-economic, current deep exploration strategies need to be improved to explore for deposits at greater depth. Naturally, as much information as possible should be extracted from a 2D survey before undertaking a 3D one (Milkereit et al., 1996, 2000; Matthews, 2002; Rodriguez-Tablante et al., 2007; Dehghannejad et al., 2012a; Manzi et al., 2012; Koivisto et al., 2015; Malehmir et al., 2017a). Another way to de-risk a 3D survey is by estimating the seismic response of the target deposits through forward modeling (e.g. Juhlin, 1995; Bellefleur et al., 2012; Cheraghi et al., 2013). Research related to 2D and 3D aspects of seismic surveys, along with seismic forward modeling, is reported in Paper IV.

This thesis, while illustrating the principles and applications of seismic methods and integrated physical property studies, has the following main objectives: • To delineate the large-scale units of an aquifer used for groundwater

filtering;• To test the applicability of a cost-effective seismic survey set up and low-

cost source for groundwater and mineral exploration in challenging near-surface environments (soft sediments, swampy terrain);

Page 15: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

15

• Delineate the depth extent as well as 3D geometry (i.e. lateral extent) ofiron-oxide mineralization, from 2D seismic data;

• Validate the seismic response of the mineralization through seismicmodeling and physical property studies, and extending these for rockquality and seismic attenuation characterizations that can be used at themine planning stage.

1.2 Outline of the thesis The thesis consists of two main parts: 7 chapters, summarizing the main aspects of my research, including study area information and theoretical background and a collection of papers, each focusing on specific research as summarized. Chapter 1 introduces the main goals of my research and the topics covered in each paper. Chapter 2 presents a brief geology of the study areas, including the framework for natural resources exploration, history and previous studies, for each of the sites. The methodology applied in my research is described in Chapter 3, with two main parts: seismic methods and physical property studies. Chapter 4 is a summary of each paper and finally Chapter 5 concludes all the research carried out. Chapter 6 is a summary of my work in Swedish. Chapter 7 presents some ideas for future work and advancing the research introduced during my PhD studies.

In Paper I a combination of seismic refraction and reflection methods was applied for aquifer delineation in a glacial sediment environment at Virttaankangas in southwest Finland. Seismic data acquired using the seismic landstreamer and a low-cost seismic source improved the characterization of the coarse-grained units of the aquifer, despite unconsolidated thick sediment cover on top of the bedrock. Guided by hydrogeological units within an existing 3D groundwater flow model, the interpretation provided by the seismic data helps the long-term planning and sustainable use of the managed aquifer recharge plant.

Paper II initiates the research for characterization and seismic imaging of iron-oxide mineralization at Blötberget, south central Sweden. Downhole logging measurements for physical property studies, including elastic constants and seismic quality coefficient help characterize the mineralization and its host rocks in terms of rock quality and seismic attenuation. Synthetic seismograms derived from physical properties obtained from downhole logging and laboratory measurements show that seismic surveys are expected to be successful in delineating the mineralization in this hardrock environment.

Paper III combines seismic data from two field campaigns (2015 and 2016) at Blötberget for improving the resolution and imaging the mineralization at depth. Given the challenging hardrock environment, data are

Page 16: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

16

acquired in a cost-effective manner, and the mineralization is interpreted to extend from the known depth of 800–850 m to at least 1200 m depth.

Paper IV focuses on the 2016 seismic data at Blötberget, with the aim to delineate additionally the lateral extent of the mineralization. The two seismic profiles were processed as 3D, illustrating that 3D information can be retrieved from 2D seismic profiles when cross-profile data acquisition is performed. The seismic response of the mineralization was forward modeled through 3D ray-racing and 2D finite-difference modeling, suggesting possible additional resources underneath the known ones as well as important structures that may crosscut the mineralization.

1.3 Projects The research presented in this thesis was sponsored and supported by several industry-academia consortia. The nation-wide or international collaborations were established for underground infrastructure projects in urban development, and for mineral exploration and related mine planning, aiming to support related research as well as to develop new technologies.

1.3.1 Formas funded Trust GeoInfra The seismic studies at Virttaankangas (Paper I) were initiated under Trust 2.2 Formas project (252-2012-1907). Turku Region Water Ltd, the Geological Survey of Finland (GTK) and University of Turku, Department of Geography and Geology sponsored the data acquisition and collaborated in this project.

1.3.2 ERA-MIN StartGeoDelineation A large part of my PhD studies falls under the StartGeoDelineation project funded by ERA-MIN initiative (project number 2014-06238), and in particular Vinnova, SGU, Takes, NIO and Yara. Downhole logging measurements and the physical property study at Blötberget (Paper II) as well as the seismic data acquisition within two field campaigns at Blötberget (Paper III) were supported by the ERA-MIN project.

1.3.3 H2020-funded Smart Exploration The last part of my research continued into the H2020-funded Smart Exploration project (grant agreement number 775971). The research on the 2016 seismic data (Paper IV) was carried out within this project.

Page 17: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

17

2 Study areas

The case studies that form the basis of the thesis are from two sites in the Scandinavian Peninsula, in Finland and in Sweden, respectively. The two areas are located in the Paleoproterozoic Svecofennian/Svecokarelian orogen in the Fennoscandian Shield (Figure 2.1). The targets of each of the studies represent, however, different environments.

Paper I focuses on the glaciofluvial complex Säkylänharju-Virttaankangas, that covers igneous and metamorphic rocks (1.90–1.75 Ma), in southwest Finland, close to the city of Turku. Papers II, III and IV focus on the iron-oxide mineralization and their host rocks in the Bergslagen lithotectonic unit in south central Sweden, in the Ludvika mining area.

Figure 2.1. Simplified geological map of the Fennoscandian Shield showing the two study area locations in Finland (Virttaankangas) and in Sweden (Blötberget). Modified from Stephens et al. (2009).

2.1 Virttaankangas 2.1.1 Geology and geomorphology The Säkylänharju-Virttaankangas glaciofluvial complex belongs to the 150 km long Pori-Koski esker chain, one of the most prominent glacial landforms

Page 18: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

18

in southwest Finland (Figure 2.2). The Virttaankangas esker has been referred to as an interlobate esker, formed by time-transgressive deposition within an interlobate joint between two differently behaving ice streams originating from the rapidly retreating Baltic Sea ice stream (Punkari, 1980; Mäkinen, 2003a).

Figure 2.2. Study area location and geomorphological map of southwest Finland with glaciofluvial deposits © Finnish Environment Institute (2016), coastline and water bodies © National Land Survey of Finland (2010). Modified from Paper I.

The esker covers the Svecofennian igneous and metamorphic rocks (Figure 2.1) and is 10–15 km southeast of the basement contact with the younger “Jotnian” Satakunta sandstones (1.50–1.20 Ma). The glaciofluvial complex is further divided into two morphologically different areas (Mäkinen, 2003b): the Säkylänharju ridge, and my area of interest, the Virttaankangas plain. The Säkylänharju ridge is about 1–2 km wide with a narrow 50–150 m wide summit at about 140 m a.s.l. that gradually decreases towards southeast into a 5 km long fan-like plain at about 110–115 m a.s.l. The Virttaankangas plain is 2–4 km wide and glaciofluvial sediments above the bedrock are 10–40 m thick, and reach at least 70 m in some places (Mäkinen, 2003a, b).

2.1.2 Eskers Eskers are defined as continuous or segmented winding ridges of stratified sand and gravel deposited by glacial meltwater-streams (Banerjee and

Page 19: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

19

McDonald, 1975; Gorrell and Shaw, 1991). They are aligned parallel to the ice flow direction and may reach hundreds of kilometers in length. Their width is typically from tens to hundreds of meters, and can reach tens of meters in height. Moreover, eskers can be joined by tributaries (smaller eskers), forming wide esker networks, such as in southwest Finland. Interlobate eskers are exceptionally large eskers that formed within interlobate joints between differently behaving ice streams (Punkari, 1980, 1997). They generally have a different depositional environment, compared to “regular” eskers, as they are more extensive and present complex large-scale depositional units.

2.1.3 Eskers and groundwater resources Eskers are hydrogeological features of great importance in the postglacial landscape in areas such as Finland, Sweden or North America. They are remarkable aquifers and are often used for construction purposes, as they are easily quarried and thus attractive targets for aggregate extraction.

The Virttaankangas plain hosts an operational MAR (managed aquifer recharge) plant that provides potable water for the entire city of Turku and surrounding region. Water from nearby rivers and lakes is first pumped and pre-treated and subsequently infiltrated into Virttaankangas aquifer through elaborate sprinkler systems (Artimo et al., 2003). The artificially recharged groundwater is then pumped again to a waterworks plant and finally to the city of Turku. Turku, including surrounding municipalities, has a population of 330,000 inhabitants (Statistics Finland, 2020) and relies for its drinking water on the artificially recharged groundwater.

MAR plants are an efficient way to provide drinking water for entire communities, as they are a reliable long-term water supply, less affected by seasonal variations in quantity and quality compared to river water (Artimo et al., 2003). Eskers in particular are exceptional aquifers due to their hydraulic properties. The coarse-grained units of eskers have high hydraulic conductivity and capacity to yield adequate groundwater supplies. However, the hydraulic properties of the esker units can vary rapidly, both horizontally and vertically, requiring precise geological and hydrogeological models of the aquifers. An accurate hydrogeological model, particularly in 3D, provides the fundamental base for flow and transport models within the aquifer.

For a successful groundwater flow model, the geometry, stratigraphy and sedimentology of the aquifer and its units must be integrated into the model, such that the correct hydraulic conductivities can be assigned to the hydrogeological units. At a larger scale, the depositional units of the esker (i.e., the aquifer) can be delineated also from multidisciplinary studies that combine geophysical methods and hydrogeology aspects (Rossi et al., 2014; Ulusoy et al., 2015), providing thus key hydrogeological information that is needed for modeling the groundwater flow within the aquifers.

Page 20: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

20

2.1.4 Esker architecture The internal architecture of Virttaankangas consists of several elements that are relevant for the 3D hydrogeological model for groundwater flow, with the main elements: • Esker core – course-grained sediments such as gravel and boulders,

deposited on bedrock. Continuous and thick till beds are often foundbelow the esker core, but the till beds were sometimes eroded during theesker deposition;

• Esker fans – stratified esker sediments, usually deposited on top of theesker core. They are usually ice-marginally deposited and have anoverlapping, successive pattern;

• Fan lobe channels – cross-bedded structures, usually fine-grainedsediments deposited towards the esker margins.

Special features in esker environments are morphologically undetected kettle holes (MUKHs). They are large-scale deformation structures formed when buried blocks of ice melt and subsequently get filled with collapsing esker fan sediments and shore deposits, resulting in mixed sediments with fine-grained beds.

These large-scale esker elements are the basis for the reference model of the glacifluvial complex illustrated in Figure 2.3. The model comprises the main 3D hydrogeological units of the Virttaankangas aquifer, and a base for modeling the groundwater flow (Artimo et al., 2010).

Figure 2.3. The main 3D hydrogeological units of the Virttaankangas aquifer (from Artimo et al., 2010). The esker core is contained within the glaciofluvial coarse-grained unit. MUKH structures are treated as a separate unit. Red dashed line in the bottom unit represents fractured bedrock.

Page 21: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

21

2.1.5 Previous and current studies The Säkylänharju-Virttaankangas complex has been thoroughly studied through sedimentological work, including characterization of the depositional environments and related depositional stages during the last deglaciation by several authors (Artimo et al., 2003; Mäkinen 2003a, b; Mäkinen and Räsänen 2003). The depositional model is based on an extensive ground penetrating radar (GPR) survey (21 km) verified by reference data and exposed sand and gravel pits (Mäkinen, 2003b; Mäkinen and Räsänen, 2003). Bedrock topography (Valjus, 2006) and the main hydrogeological units have been studied earlier and integrated in the 3D hydrogeological model that is the base for the groundwater modeling (Artimo et al., 2003, 2008). Geomorphological studies of the esker complex were combined with drillhole logs, seismic data, results of pumping and recharge tests and pit observations (Mäkinen, 2003b).

The current study at Virttaankangas (Paper I) focuses on seismic data acquired over the most complex part of the course-grained unit of the aquifer. Seismic refraction and reflection data were acquired along two profiles with a total length of approximately 2 km. Boreholes in the area provide information about the bedrock levels, together with lithology and description of the units. The seismic data interpretation was guided by the hydrogeological units in the existing MAR groundwater flow model.

2.2 Blötberget 2.2.1 Geology Blötberget, in south central Sweden, is located in the intensely mineralized Bergslagen lithotectonic unit, and a part of the Svecokarelian orogeny in the Fennoscandian Shield (Figure 2.4). The province is especially known for its high-quality iron-oxide deposits that formed the foundation for parts of the Swedish industry. Host rocks to the iron-oxide mineralization are metamorphosed volcano-sedimentary rocks of Palaeproterozoic age (1.91–1.87 Ga). The largest deposits in the region are represented by skarn-type iron-oxide deposits and apatite iron-oxide deposits (Magnusson, 1970; Stephens et al., 2009), the latter type accounting for more than 40% of the iron ore produced in the region (Magnusson, 1970; Stephens et al., 2009).

Blötberget deposits are high-quality apatite iron-oxide mineralization with about 50% iron content. Economic parts of the deposits are found in sheet-like horizons, within dacitic to andesitic, feldspar and porphyritic metavolcanic rocks, metamorphosed in low pressure amphibolite facies. The mineralization and host rocks are cut by several intrusions, such as coeval dacitic, andesitic and basaltic dykes and subvolcanic intrusions. These were further disturbed by synvolcanic, granitic to intermediate plutonic rocks and post-mineralization intrusions of granite–aplite–pegmatite composition. The rocks

Page 22: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

22

have been affected by several episodes of ductile deformation (Allen et al., 1996; Ripa and Kübler, 2003).

The study area is located about 500 m southeast of the village of Blötberget and 12 km southwest of the city of Ludvika. Throughout the thesis and reported research, the Blötberget deposits are referred to as Blötberget or Ludvika Mines. Figure 2.4 marks the location of the study area in Blötberget (Sweden), main geological units and geophysical surveys reported in Papers II, III and IV.

2.2.2 Mining history Bergslagen province is one of the oldest mining provinces in Sweden, with mining activities dating back to the Middle Ages. Mining and metallurgic industries have been important in the province from the industrial revolution to the end of 20th century when many mines closed due to decreased iron prices. Nowadays, the demand for minerals and metals is higher than ever, and interest has increased into reopening mines in the region.

Mining in Ludvika area started in the 1600s, and was focused on iron production through the 1900s, including the mine at Blötberget, but production started to decrease at the end of the century. Ground-based and airborne geophysical surveys (SGU) were carried out in the Blötberget area in 1950s and 1960s, respectively. From 1942 until 1977, the deposits were systematically core drilled for better definition and extension of the ore, referred in this thesis as historical drilling since only parts of the cores are currently accessible. Mining eventually stopped in 1979, a period when many mines in the Bergslagen province closed their operations due to low iron ore price globally. The deposits at Blötberget were mined down to 200–300 m and the old mining infrastructures (e.g., shaft and tunnels) are still in place.

Nordic Iron Ore (NIO) started exploration at Blötberget in 2011, and reported studies in the area include magnetic surveys and drilling programs (2012 and 2014). All the drilled cores were sampled by NIO, including core photography, point load testing, drill core structure orientation, geotechnical logging and classification. Many of the historical drilled cores were also recovered, re-logged and re-assayed.

Most of the mineral extraction from the Blötberget mine was above the 240 m level. NIO plans to restart mining operations at the 280 m level in the near future, subject to a good outcome of the feasibility study and financing of the project. Before closing in 1979, construction of a new haulage level began at 330 m and a ramp to the 160 m level; these facilities were however never put into operation. Before resumption of extraction activities, NIO will complete necessary renovations and technical upgrades to the existing underground infrastructure that remained since the mine’s closure.

Page 23: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Figu

re 2

.4. G

eolo

gica

l map

of t

he B

lötb

erge

t are

a in

clud

ing

all g

eoph

ysic

al m

easu

rem

ents

repo

rted

in P

aper

s II,

III a

nd IV

. Circ

les r

epre

sent

bo

reho

les,

with

the

ones

dow

nhol

e lo

gged

in c

olor

(Pap

er II

); se

ismic

surv

eys a

re m

arke

d in

col

or: p

rofil

es 1

5P1

(Pap

er II

I), 1

6P1

and

16P2

(P

aper

IV).

The

bedr

ock

map

was

repr

inte

d w

ith p

erm

issio

n fro

m S

GU

. Maj

or te

cton

ic u

nits

wer

e m

odifi

ed fr

om M

aleh

mir

et a

l. (2

015)

.

Page 24: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

24

2.2.3 Mineral resources The Mineral Resources in Blötberget are estimated at 45.4 million tonnes of 41.7% Fe, which are classified as Measured, and 9.6 million tonnes at 36.2% Fe classified as Indicated. An additional 11.8 million tonnes of 36.1% Fe is classified as Inferred Mineral Resources (NIO Mineral Resource Estimate Report, 2017).

The mineralization in Blötberget consists of magnetite and hematite, and additionally apatite and small amounts of quartz and silicate minerals are present. The ore contains more than 50% iron, mostly hosted in magnetite. Hematite occurs mainly in distinct horizons. Hematite ores are less massive than the magnetite ones and their host rock mineralogy is slightly different, containing more quartz and feldspar. The origin of the apatite iron-oxide deposits is considered to be syn-volcanic, although this is disputed, with a new study favoring a magmatic-hydrothermal origin (Jonsson et al., 2013). Of the Measured and Indicated Resources, 54% is magnetite and 46% is hematite.

The geological model of the deposits (Figure 2.5) is based on the borehole drilling and historical data and shows the mineralization dips towards southeast in sheet-like units (varying thicknesses) at about 55° down to 400 m after which they dip moderately at about 25° down to the known depth of 850 m given by historical boreholes. Data from the drilling campaigns show that the mineralization is intersected by the boreholes between 400–600 m depth, within layers of about 30–50 m, sometimes thinner. Figure 2.6 illustrates an example of downhole measurements of magnetic susceptibility, as reported on in Paper II. Black double arrows mark the mineralization zone in each logged borehole, including laboratory measurements on the core samples in the mineralization depth ranges of 320–570 m.

Figure 2.5. 3D visualization of the Blötberget deposits and boreholes in the area. The deposits are represented with different colors representing four distinct orebodies. The most southern deep boreholes are historical boreholes intersecting the mineralization at 850 m depth.

Page 25: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

25

Figure 2.6. Magnetic susceptibility measurements from all the seven boreholes downhole logged and reported on in Paper II. The logs are displayed in two groups based on their spatial proximity to each other. Note a marked increase in the depth of the mineralized zone (indicated with black arrows for each panel) from northwest to southeast (dip direction).

2.2.4 Previous and current studies Apart from the historical drilling data and airborne magnetic surveys reported in section 2.2.2, a series of studies were initiated at Blötberget during 2015–2016, starting with physical property studies reported in Paper II. Together with other geophysical and petrophysical studies in the area (Yehuwalashet and Malehmir, 2018; Almqvist et al., 2019), these studies form a basis for new geophysical surveys in the area, in particular seismic surveys. Seismic field campaigns at Blötberget started in 2015 and 2016, as illustrated in Figure 2.4, and are reported on in Papers III and IV.

Seismic data acquisition consisted of 2D seismic reflection surveys. In 2015, a 3.5 km long seismic profile was acquired with a combination of

Page 26: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

26

landstreamer (MEMS sensors) and wireless recorders connected to 10 Hz geophones. In 2016, two seismic profiles were acquired: a 2.4 km long profile consisting of cabled 10 Hz geophones and a short (700 m) profile consisting of 10 Hz geophones connected to wireless recorders. Results from the 2015 seismic survey and from one profile of the 2016 seismic survey are reported on in Paper III. Processing of the 2016 seismic data focused on 3D swath processing and seismic modeling, as reported on in Paper IV.

Page 27: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

27

3 Methods

3.1 Seismic methods Seismic surveys are one of the most widely used geophysical methods for investigating the subsurface. In the simplest way, seismic methods, like all geophysical methods, rely on detecting differences of the subsurface materials (Telford et al., 1990; Sheriff and Geldart, 1995), or the locations where one region differs from the other in their properties (i.e., elastic properties of rocks, for seismic methods). The most widespread use of seismic methods is traditionally related to petroleum exploration, but they are applied for other purposes as well, as the research presented in this thesis illustrates. The greatest advantages of seismic over other geophysical methods are higher resolution and greater penetration depth, making them attractive for both shallow, e.g., groundwater and/or engineering applications (Steeples and Miller, 1988; Miller and Steeples, 1994; Bradford et al., 1998; Baker, 1999; Pugin and Pullan, 2000; Francese et al., 2002; Juhlin et al., 2002; Pugin et al., 2014; Brodic et al., 2017, 2018), as well as for deeper studies, e.g., mineral exploration (Milkereit et al., 1996; Eaton et al., 2003a; Schmitt et al., 2003; Malehmir et al., 2012, 2014; Heinonen et al., 2013; Manzi et al., 2015, 2020) or crustal studies (Allmendinger et al., 1987; Chadwick and Pharaoh, 1998; Singh et al., 2006; Chowdhury, 2011; Dehghannejad et al., 2012b; Hloušek et al., 2015; Buntin et al., 2019). Papers I, III and IV illustrate applications of seismic methods for groundwater and mineral exploration purposes.

Seismic methods can have active or passive sources. The fundamental principle of the active seismic methods (the focus of this thesis) is the controlled generation of elastic waves from a seismic source and measuring the time required for the waves to travel through the subsurface. Sensors at the ground surface (or in boreholes) record the propagation of the waves through different materials or structures in the subsurface. Seismic methods derive structural information principally from two types of recorded waves: refracted or reflected. Seismic refraction deals with the refracted waves that travel nearly horizontally along the boundary between layers of different elastic properties, while seismic reflection deals with waves that travel initially downward and reflect back to the surface (essentially vertical). The traveltimes always depend on the physical properties of rocks and geological settings that control the seismic velocity.

Page 28: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

28

One of the restricting factors in the use of seismic methods is the perceived high acquisition and processing costs involved. However, recent advances in sensor development, less expensive manufacturing and the use of dense arrays such as landstreamers for high-resolution applications (e.g., Brodic et al., 2015, 2018; Malehmir et al., 2017b, 2017c), have brought down considerably the associated costs. Moreover, a low-cost and effective seismic source contributes to further making seismic methods accessible and efficient in different types of environments, as reported in the case studies in Papers I and III.

Depending on the acquisition setup, investigated target and application, seismic methods can be applied in 2D or 3D. Higher costs associated with acquiring 3D data place the focus primarily on obtaining as much information as possible from 2D seismic surveys, both for subsurface characterization and effective 3D survey planning and design (Zaleski et al., 1997; Urosevic and Evans, 1998; Adam et al., 2000; Milkereit et al., 2000; Eaton et al., 2003b; Nedimović and West, 2003a, b; Urosevic et al., 2007, 2012; Malehmir and Bellefleur, 2009; Koivisto et al., 2012; Malehmir et al., 2017a; Manzi et al., 2012, 2020; Heinonen et al., 2013; Manzi and Durrheim, 2015). The seismic data acquired within the case studies in this thesis are essentially 2D. While the seismic surveys proved to be cost-efficient and effective in the target delineation (Paper III), it was also shown that information about the 3D geometry of the target could be retrieved from 2D seismic profiles when cross-profile recording and adequate 3D swath processing was applied (Paper IV).

3.1.1 Seismic waves and elastic constants There are two main types of waves generated by seismic sources (natural or passive and man-made or active), divided by the way the seismic energy travels away from the source: • Surface waves - the energy travels near or along the ground surface;• Body waves - the energy travels through the interior (bulk) of the earth.

Body waves are of prime interest for the seismic refraction and reflection methods and can be divided into two types, depending on the direction of ground movement relative to the propagation direction: • P-waves (primary, longitudinal or compressional waves) - the particle

motion is parallel to the direction of propagation;• S-waves (secondary, shear or transverse waves) - the particle motion is

perpendicular to the direction of propagation, with:• SV-waves or radial - when the particle motion is in the vertical

plane,• SH-waves or transversal - when the particle motion is in the

horizontal plane.

Page 29: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

29

The seismic velocity is the most fundamental parameter in seismic methods and depends on the density and elastic properties of the material the seismic wave propagates through. Considering an elastically isotropic medium, the P-wave velocity (VP) and the S-wave velocity (VS) are defined as:

= (3.1)

= (3.2)

where k and µ are the bulk and shear moduli, respectively, and ρ is the density. Depending on the complexity of the rocks and of the medium, such as parallel bedding or intense fracturing, the elastic constants and thus the velocities may be directionally dependent, and anisotropy should be considered. Nevertheless, a great part of seismic data analysis works on the assumption that the subsurface is seismically isotropic (Yilmaz, 2001).

Density and seismic velocity are the most important physical properties influencing the overall seismic response of a target (Salisbury et al., 2000, 2003; Schmitt et al., 2003). Understanding the parameters controlling the elastic properties is translated to knowledge of elastic constants of the materials the waves propagate through. Apart from mineralogical composition, factors such as porosity and pore content, fracturing and their frequency directly influence the elastic constants, hence are directly related to characterizing the rock behavior.

Table 3.1 summarizes the main elastic parameters of rocks, as derived from seismic velocities (Birch, 1939; Mavko et al., 2009), and for isotropic materials. While most of the elastic constants can be derived from density and velocity, the seismic parameter (sp), introduced in Paper II, is only derived from compressional and shear wave velocities. The seismic parameter is linked to the bulk modulus by density, and is used to characterize high-density outliers that do not follow a linear increase with increasing velocity (Milkereit and Wu, 2015; Kassam et al., 2016).

Table 3.1. Elastic constants influencing the seismic response of rocks (Paper II). Elastic constant Equation

Seismic parameter sp=VP2 – 4/3∙VS2 Bulk modulus k=ρ∙sp Shear modulus µ=ρ∙VS2 Young’s modulus E=ρ∙VS{(3∙VP2 – 2∙VS2)/ (VP2 – 1/3∙VS2)} Poisson’s ratio ν=(VP2 – 2∙VS2)/2(VP2 – VS2)

Page 30: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

30

3.1.2 Seismic modeling Seismic forward modeling is a valuable tool that helps to understand the seismic response of the investigated target through comparison of synthetic and real seismic data. The modeled seismic response of a known target not only allows validating the interpretation of real data, but also verifying estimates of physical properties and geometry of the target, as well as the geological setting (Carcione et al., 2002; Bellefleur et al., 2012; Dehghannejad et al., 2012a; Malinowski et al., 2012). There are many approaches to seismic modeling, such as direct methods (finite-difference) and ray-tracing methods. In this thesis, constant velocity 3D ray-tracing reflector modeling and 2D finite-difference forward modeling are used (Paper IV).

3.1.2.1 Reflector modeling Reflection modeling helps understanding the 3D geometry of reflectors by matching the modeled traveltimes to those observed in the seismic data. Assuming a 3D planar structure with certain geometry (dip, strike), in a constant velocity medium, simple 3D ray-tracing (straight ray) traveltime modeling can be used to fit the reflections from the modeled reflectors to those observed on shot gathers and stacked sections (Ayarza et al., 2000; Juhlin and Stephens, 2006). Traveltime calculation of the modeled reflectors relies on the real data acquisition setup geometry, with a priori or iteratively set density and seismic velocity (for reflector and the medium hosting it). The response is then modeled using receiver coordinates for reflector modeling on the shot gathers or CDP coordinates for the stacked section. When there is a match between the calculated and the observed traveltimes on the shot gathers (or stacked section), a potential geometry that can explain the observed reflections is found.

3.1.2.2 Finite-difference forward modeling Finite-difference forward modeling is commonly used for studying the seismic response of mineral deposits, with the aim of aiding seismic survey design and subsequent seismic interpretation (Bellefleur et al., 2012; Cheraghi et al., 2013). It is essential in validating the seismic data against geological models and understanding the geometry of orebodies that can have their seismic response easily scattered in a hardrock environment (L’Heureux et al., 2009; Cheraghi et al., 2013). In crystalline rocks, the composition and shape of orebodies is best studied through 3D finite-difference modeling (Bohlen et al., 2003), but 2D algorithms, computationally less expensive, also provide reliable results that improve the knowledge of the expected seismic response.

Page 31: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

31

3.2 Physical property studies The study of physical properties of rocks comprises a wide range of physical properties that can be obtained from surface seismics, downhole logging or laboratory measurements. These measurements represent different scales of investigation. Seismic velocities obtained from sonic logging are considered representative for the in-situ conditions of the rock formation, as the spatial variation in velocity along the boreholes is measured, and traditionally, can be used to calibrate surface seismic data (McKenzie et al., 1982; Serra, 2008). It is important to note, however, that the frequency range in a seismic survey goes to 200-300 Hz, while the sonic logging tool uses frequencies in the range of 20 to 40 kHz, hence, shorter wavelengths, resulting in higher recorded velocities (Goetz et al., 1979; Sams et al., 1997). A similar observation applies to laboratory ultrasonic measurements of seismic velocity, employing even higher frequencies, in the MHz range.

3.2.1 Downhole logging Downhole logging consists of measuring various physical, chemical or structural properties of a geological formation through a tool sampling the borehole along its length (Hearst and Nelson, 1985; Serra, 2008). Data obtained through downhole logging have the advantage of representing the in-situ physical properties of the rocks and their fluids where the recorded logs can be directly compared to lithology information if the drilled cores are recovered and available. The scale of investigation is intermediate between the laboratory measured properties and field-scale geophysical surveys, providing thus an important link in integrating different datasets.

There is a variety of logging tools available (Serra, 2008), and the ones employed in Paper II are fluid conductivity and temperature, formation resistivity, magnetic susceptibility and full-waveform sonic logging. The logging tools in Paper II include natural gamma radiation measurements as well.

One of the logging tools that is particularly useful for lithological characterization, for mineral or engineering applications, is the triple full-waveform sonic log. This tool (e.g., the one used in Paper II) uses a piezoelectric transmitter to emit a high frequency mechanical pulse (approximately 20 kHz) that is recorded at three receivers, near, mid and far, spaced at 20 cm from each other. The traveltime (interval transit time) that the seismic waves take to reach these three receivers is recorded and the seismic velocity for the P- and for the S-waves can be calculated.

An example of the waveform recorded by the full-waveform triple sonic probe is illustrated in Figure 3.1. The principles of the sonic logging are based on the theory of the wave propagation in an elastic medium, just like for the surface seismic methods, discussed in the previous section. The recorded

Page 32: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

32

waveforms are compressional (P) and shear (S) waves, and high amplitude late arrivals tube (Stoneley) waves specific for acoustic logging. Note that the Stoneley waves appear clipped in Figure 3.1, due to equipment settings.

Figure 3.1. Examples of waveforms recorded by the full-waveform triple sonic probe for (a) mid-receiver and (b) far receiver. The traveltimes for the first arrival of P- and S-waves are indicated by red and blue stars, respectively. The adjacent small windowsnext to each waveform represent an example of the amplitudes preserved for eacharrival time, for seismic QP calculation (amplitude-decay curve method) discussed inthe next section.

3.2.2 Seismic quality factor Seismic quality, Q (Knopoff, 1964), is a dimensionless quality factor that describes the ability of a material (intrinsic) to attenuate the passing seismic energy. The physical mechanisms that govern the reduction in the amplitude of the seismic energy are essentially three-fold: (1) geometrical spreading (the amplitude loss as waves move away from the source), (2) intrinsic attenuation (energy loss due to inelastic processes or internal friction during wave propagation) and (3) scattering due to heterogeneity.

Seismic quality quantifies the intrinsic attenuation of a material due to inelastic processes and it is inversely related to the attenuation strength of the materials; regions with low Q are more attenuative than regions with high Q values. It is a significant additional seismic parameter used in seismic interpretations and can be a reliable indicator for different lithologies (Hamilton, 1972; Hauge, 1981) and subsurface conditions such as inhomogeneities, fracture zones or water saturation. Conventional Q analysis, obtained from surface seismic data or borehole data (Tonn, 1991; Dasgupta

Page 33: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

33

and Clark, 1998), is used for reservoir characterization in sedimentary environments (Hauge, 1981; Dasios et al., 2001) but also in tunneling and occasionally in mineral exploration applications (Sjogren et al., 1979; Barton, 2002; King, 2009; Brodic et al., 2017), as a petrophysical parameter itself.

The seismic quality factor can describe the attenuation of both P- and S-waves. QP denotes the seismic quality factor when P-waves are concerned, while for S-waves QS is used. As shown in Paper II, attenuation can be relatively easy to determine from full-waveform sonic data after the identification of P- and S-wave energy traveling along the borehole walls. There are several computational methods for determining seismic Q, from wavelet modeling to amplitude-decay methods (Tonn, 1991). The results presented in Paper II are derived from the amplitude-decay method for the P-wave arrivals from the full-waveform data. The amplitude-decay method uses the ratio of the maximum amplitudes of the P-wave (can also be S-wave) at two different distances, x1 and x2, that represent two receivers of the sonic probe. In our case, these are mid and the far receivers (20 cm spacing), whereas the amplitude window for each receiver is chosen such that it contains only the first P-wave arrivals, by tapering the data before and after. An example is illustrated in Figure 3.1. The seismic quality factor is given by:

= (3.3)

where x1 and x2 are two different distances where the amplitudes A1 and A2 of the seismic wave are measured. The velocity of the wave is denoted by c and ω=2πf represents the angular frequency or essentially the dominant frequency. The QP values obtained in this way (and as used in Paper II) are considered frequency independent.

3.2.3 Laboratory measurements A series of laboratory measurements were included for complete characterization of the host rocks and mineralization, as reported in Paper II.

Rock quality designation (RQD) is a standard technique in the construction industry (both tunneling and mining) for the qualitative and quantitative assessment of rock quality and degree of jointing, fracturing, and shearing in a rock mass (Deere, 1962; Haldar, 2013). Although it is criticized for its lack of accuracy, it is still widely used due to its simple estimation during core drilling. It classifies the overall quality of rocks by measuring in percentages the degree of jointing or fracturing in a rock mass. A rock quality designation of 75% or more shows good quality hard rock and less than 50% shows low quality, possibly fractured or weathered rocks. RQD measurements were available for the majority of the logged boreholes. Table 3.2 summarizes the descriptive character of the rock quality based on the measured RQD.

Page 34: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

34

Table 3.2. Rock quality as assessed by RQD measurements.

Rock Quality RQD (%)

Excellent 90-100Good 75-90Fair 50-75Poor 25-50Very poor 0-25

Density laboratory measurements were performed on the cores at every 2–3 m, using a suspension method based on Archimedes' principle (Hughes, 2005). Measurements were performed for six out of seven logged boreholes.

Laboratory measurements of magnetic susceptibility, magnetite and iron content (as well as other minerals), and density as well, were performed by NIO on all the core samples for the mineralization depth range.

The rock engineering classification of rock mass quality (Barton et al., 1974), referred to as QR, is described using six parameters, from RQD to a number of joint sets detailing the degree of weathering, fractures fillings and weakness zones. This measurement is not included in the reported studies but it is relevant, as described in the Outlook Chapter. Highly jointed rock masses with low QR values are expected to show increased attenuation of the seismic energy, compared to an intact hard rock mass that will attenuate less the seismic waves (Barton, 2006).

Page 35: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

35

4 Summary of papers

4.1 Paper I: Interlobate esker architecture and related hydrogeological features derived from a combination of high-resolution reflection seismics and refraction tomography, Virttaankangas, southwest Finland

A seismic refraction and reflection study was carried out at Virttaankangas, close to the city of Turku, southwest Finland, for esker characterization and testing the potential of a 200 m long broadband seismic landstreamer developed by Uppsala University (Brodic et al., 2015) for efficient aquifer delineation over thick glacial sediments.

4.1.1 Summary The seismic data acquired in August 2014 at Virttaankangas consisted of two seismic profiles, each about 1 km long, in the proximity of the MAR (managed aquifer recharged) infiltration pools (Figure 4.1). The main objectives of the study were to locate the highly hydraulically conductive esker core as the target for new pumping station locations and delineate large-scale architectural esker elements (i.e. esker core, fan lobe channels) within the coarse-grained hydrogeological unit of the aquifer. An additional goal was to define large-scale deformation structures (morphologically undetectable kettle holes, MUKHs) and characterize the lateral relationships of the esker elements and the underlying bedrock surface.

The two profiles were acquired along sandy roads and forest trails with the same geometry spread for both seismic refraction and reflection data. The acquisition setup consisted of moving the landstreamer, a 200 m long array of 80 MEMS sensors (Brodic et al., 2015), until the entire profile length was covered. The array consisted of four segments, each containing 20 units; one segment with units spaced at 4 m and three with sensors 2 m apart, respectively. Additionally, 51 wireless sensors connected to 10 Hz geophones, spaced at about every 20 m, were deployed along each profile for increasing the low data fold at the end points of each landstreamer segment. Table 4.1 summarizes the key acquisition parameters. The survey was carried out within

Page 36: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

36

5 days, with a low-cost seismic source (500 kg Bobcat-mounted drophammer) and a small number of personnel.

Figure 4.1. Quaternary map of the study area with the seismic profiles, main esker core, Löytäne tributary esker, Myllylähde spring and fractured bedrock zone. The extent of the MAR plant area with infrastructure (observation wells and infiltration pools) and the coarse-grained unit provided by Turku Region Water Ltd. The map of superficial deposits of Finland © Geological Survey of Finland and the base map and LiDAR © National Land Survey of Finland (2010).

Table 4.1. Key seismic data acquisition parameters (August 2014).

Spread parameters Profile 1 Profile 2

Recording system Sercel Lite 428 Receivers DSU3 (MEMs), 10 Hz vertical-type geophones No. of receiver locations 371 451 Landstreamer 320 (80 moving) 400 (80 moving) Wireless recorders 51 (fixed) 51 (fixed) No. of shot points 192 (3 records/point) 248 (3 records/point) Receiver spacing Landstreamer 2/4 m Wireless recorders 20 m Shot spacing 4 m Maximum offset ~ 1000 m Source type Bobcat-mounted drophammer Sampling interval 1 ms

Page 37: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

37

Refraction tomography was carried out using a diving-wave (or turning-ray) first break traveltime tomographic inversion algorithm (Tryggvason et al., 2002) and cells of 2 and 1 m in horizontal and vertical directions, respectively. In the lateral direction cells were kept large to force the model to look 2D although the modeling uses a 3D algorithm for traveltime calculations. The results are shown together with available borehole data providing bedrock confirmation at several places (Figure 4.2). One notices a large velocity gradient, from about 500 m/s to about 1500 m/s, typical for sandy or silty sediments (Malehmir et al., 2013b; Salas-Romero et al., 2015), when they become water saturated (note the water table). This change in velocity generally matches closely the water table, except at a few places where coarse-grained or highly disturbed sediments are expected to be found (e.g. 200–300 m along profile 1 or the northeastern end of profile 2).

Figure 4.2. Seismic refraction tomography sections along (a) profile 1 and (b) profile 2. Boreholes near the profiles are indicated using red bars. A wide bedrock fracturezone is estimated from 300 to 600 m distance along profile 1. The blue dashed linerepresents the water table at the time of the seismic acquisition (August 2014),matching well the velocity transition from 400–500 m/s (unsaturated sediments) to1,500–1,800 m/s (saturated sediments).

Bedrock surface is easily identifiable by the high (>5000 m/s) velocity regions. The central part of profile 1 (Figure 4.2a) where still relatively high seismic velocity (> 3000 m/s) is observed, represents a major bedrock fracture zone as suggested by previous studies (Valjus, 2006) in the area, and corresponds to a bedrock depression as well. A bedrock depression is observed also at 300–400 m distance along profile 2. The boreholes reaching bedrock show an excellent match with the bedrock delineated from the seismic velocities, apart from borehole VI596 that reaches the bedrock fracture zone (Figure 4.2a).

Despite the dry and unconsolidated glacial sediments that can highly attenuate the seismic response, the recorded seismic reflection data are of excellent quality. First breaks, already employed in the seismic refraction tomography, are visible along the entire profile length, with a few exceptions, and reflections are visible already in the raw shot gathers (Figure 4.3).

Page 38: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

38

Figure 4.3. Example raw shot gathers and apparent velocities of direct and refracted arrivals from (a) profile 1 and (b) profile 2. The red arrows indicate distinct reflections identifiable already in the raw data. The wireless sensors are noticeable at far offsets and tend to be in general noisier and have lower frequency content; however, the first breaks are identifiable almost on all traces. Note that for display purposes the offset distance between receivers is not scaled.

The most shallow reflections are interpreted to originate from the water table, even though that reflections marking the water table boundary are not common in shallow reflection surveys (Steeples and Miller, 1988). The depositional environment at Virttaankangas is characterized by large and rapidly changing velocity gradients, further complicated by the transition from dry to water saturated sediments. Thus, even though the reflection data was rich in reflections, a careful velocity analysis was needed. An approach based on combining different stacked sections (Miller and Xia, 1998; Juhlin et al., 2010; Malehmir et al., 2013b) was taken: two different stacks, one imaging the shallowest reflections and one focusing on bedrock reflections, were combined and thus, 10–20 m shallow structures could be identified along profile 1. Water table reflections are identified in the stacked sections as well, but are masked in some places by the complex depositional environment. In a similar manner, deeper high-amplitude reflections, interpreted to originate from the bedrock boundary, are discontinuous at places as the seismic signal can be easily scattered in such a gravel and sandy environment. However, the refraction tomography sections together with existing borehole data and a priori knowledge of the depositional environment helps limit uncertainties.

In order to compare and validate the seismic imaging, the seismic tomography results are overlaid on the seismic reflection sections (Figure 4.4 and Figure 4.5). In accordance with the high seismic velocity in the refraction tomography models, the bedrock surface is identified from high-amplitude reflections in the stacked section. The bedrock reflections have an undulating character and are mostly continuous for profile 1, but only partly along profile 2. Delineating the bedrock contact is important for defining the boundary forthe groundwater flow models. Along profile 2, the bedrock boundary is highlydiscontinuous, especially at the southwestern part. This zone is interpreted torepresent esker fan sediments, characterized by coarse-grained units.According to borehole VI282, just above the bedrock there is a zone of sand

Page 39: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

39

and bouldery gravel, thus it is not surprising that the seismic signal would be scattered and strong reflections are not observed from the underlying bedrock.

The interpretation of the shallow seismic reflections above the bedrock is based on the hydrogeological and depositional model previously studied (Artimo et al., 2003, 2010; Mäkinen, 2003a). This reference model is based on three large-scale esker elements: esker core, MUKH structures and fan lobe channels, the major units that affect the artificial recharge and groundwater flow within the MAR plant. Thus, the interpretation of the seismic profiles is divided into several zones for each profile.

Profile 1 (Figure 4.4) has four depositional zones based on the reflection pattern and the hydrogeological model, with the main elements and reflection characteristics summarized below: Zone 1:

• a small MUKH structure and deformed sediments (chaotic anddeformed reflections),

• esker core and fan sediments (arched geometry, convex reflectionsin contact with the bedrock);

Zone 2: • steep bedrock (high-amplitude, but vertically dislocated

reflections),• subaqueous esker fan sediments (shallow reflections);

Zone 3: • a wide bedrock fracture zone overlaid by a till layer (strong

horizontal reflection above the bedrock),• esker core overlaying the bedrock (its shape is not distinctly arched

due to the diagonal position to the profile),• shallow esker fan sediments and esker fan lobe channels (shallow

reflections);Zone 4:

• distal esker fan sediments (shallow reflections overlaying thebedrock).

Boreholes present along the profiles provide a control for the interpretation, especially when identifying the lithology and the zonation of the course-grained units, most relevant for the hydrogeological model.

Page 40: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Figu

re 4

.4. (

a) S

eism

ic r

efle

ctio

n se

ctio

n al

ong

prof

ile 1

with

the

pro

ject

ed t

omog

raph

y se

ctio

n on

top

, inc

ludi

ng e

xisti

ng b

oreh

ole

data

(in

dica

ted

with

red

bars

); (b

) the

seism

ic re

flect

ion

sect

ion

with

inte

rpre

ted

depo

sitio

nal z

ones

, bor

ehol

e str

atig

raph

y an

d in

dica

tion

of b

edro

ck

from

the

bore

hole

s; (c

) the

inte

rpre

tatio

n of

dep

ositi

onal

uni

ts an

d in

dica

tion

of b

edro

ck fr

om b

oreh

oles

alo

ng p

rofil

e 1.

Page 41: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

41

Profile 2 (Figure 4.5), perpendicular to the esker system (Figure 4.1) lies across the main aquifer and its depositional zonation is summarized below: Zone 1:

• Cross-bedded fan lobe channels from the main esker (shallowreflections),

• Proximal tributary fan sediments (profile runs parallel to thetributary esker),

• Bedrock reflection is lost under the shallow sediments;Zone 2:

• Large MUKH structure with collapsed/deformed esker fansediments (concave feature overlaying the bedrock);

Zone 3 (main aquifer of the glaciofluvial complex): • Main esker core and tributary esker core (arched geometry),• Proximal fan structure between the cores (through-shaped);

Zone 4: • End of another large MUKH structure (vaguely dislocated

reflections),Zone 5:

• Widening of the coarse-grained unit (mainly based on boreholeVI251 and the hydrogeological unit from Artimo et al., 2003).

Additionally, thick and continuous till beds bellow the esker core are identified, relevant for the groundwater flow model as they are poorly conductive sediments. The till beds were however partly eroded due to the esker deposition lateral to the main core.

Page 42: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Figu

re 4

.5. (

a) S

eism

ic r

efle

ctio

n se

ctio

n al

ong

prof

ile 2

with

the

pro

ject

ed t

omog

raph

y se

ctio

n on

top

, inc

ludi

ng e

xisti

ng b

oreh

ole

data

(in

dica

ted

with

red

bars

); (b

) the

seism

ic re

flect

ion

sect

ion

with

inte

rpre

ted

depo

sitio

nal z

ones

, bor

ehol

e str

atig

raph

y an

d in

dica

tion

of b

edro

ck

from

the

bore

hole

s; (c

) the

inte

rpre

tatio

n of

dep

ositi

onal

uni

ts an

d in

dica

tion

of b

edro

ck fr

om b

oreh

oles

alo

ng p

rofil

e 2.

Page 43: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

43

The seismic profiles at Virttaankangas confirm the location and dimension of the main esker core as well as large MUKH structures and their contact with the bedrock. The interpreted esker elements are illustrated in Figure 4.6 with an emphasis on the newly identified elements in color. The position of the esker core is confirmed along profile 1 (Figure 4.6a) together with a proximal fan deposition, indicative of a high conductivity zone within the fractured bedrock zone. Along profile 2 (Figure 4.6b) possible routes of the tributary esker are identified, relevant for improving the high hydraulic conductivity close to the main core.

The high-resolution seismic survey was successful in delineating main structures associated with stratigraphically and hydrogeological relevant units in a complex aquifer. Sedimentological and lithological data interpreted only from open pits and sparse boreholes might be inadequate, requiring methods that can offer continuous subsurface information. Esker cores and esker fans can reach up to 500 m width and several km in length, thus acquiring data over such wide study areas should involve fast and effective field campaigns. The seismic survey at Virttaankangas was acquired during just a few days with an easily operated and cost-effective energy source. Moreover, at locations where streams or springs require an interruption of the seismic profiles, wireless sensors assure continuity in the recorded data. Soft, thick glacial sediments are challenging environments for seismic surveys, especially when the water table is deep (i.e. about 40 m) and sediments are unconsolidated (Steeples and Miller, 1988; Bachrach and Nur, 1998). However, the seismic reflection processing accounted for the rapid velocity changes in the near-surface, offering valuable information down to bedrock. The high-resolution seismic survey also resolved the stratigraphy of 70–100 m thick glacial sediments, generally a challenging environment for near-surface seismic surveys.

Page 44: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

44

Figure 4.6. The interpreted main esker elements (esker core, fans, MUKH structures; shown in color) and bedrock characteristics based on profile 1 compared to the earlier research (Mäkinen 2003b; shown in grayscale) for (a) profile 1 and (b) profile 2. Bedrock elevation contours modified from Valjus (2006).

4.1.2 Conclusions The main findings of this study are valuable for the future sustainable use of the Virttaankangas aquifer. Seismic reflection sections combined with refraction tomography offer continuous depth information and a good overview of the depositional environments. Together with the existing depositional model, the current study efficiently delineated the main architecture of the interlobate esker complex, highly relevant for the hydrogeological model as well as for groundwater flow modeling down to the bedrock level. In summary:

• The cost- and time-effective seismic survey enabled thecharacterization of the hydrogeologically relevant large-scale eskerarchitectural elements that impact the groundwater infiltration andflow;

• Deep fractured bedrock with complex overlying stratigraphy wasdelineated and the hydraulic connection between the hydraulicallyconductive unit and bedrock was identified;

• Scattered geological data was integrated, as the seismic sectionsimproved the determination of hydraulic conductivity distribution inthe groundwater flow model down to the bedrock level;

• New data enable accurate planning for future pumping wells andinfiltration areas in the most complex and widest part of the coarse-grained unit (within the esker core sediments).

Page 45: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

45

4.2 Paper II: Downhole physical property logging for iron-oxide exploration, rock quality, and mining: An example from central Sweden

One of the first in a series of geophysical studies in Blötberget area, Paper II reports an analysis of the physical properties of the host rocks and mineralization with a focus on rock quality and seismic quality factor for exploration and mine planning applications.

4.2.1 Summary During several field campaigns, between 2015 and 2016, downhole physical properties were measured in seven boreholes at Blötberget, with the following main objectives:

• Characterize the mineralization and its host rocks;• Link downhole physical properties to rock quality estimation;• Find controls for future geophysical deep exploration studies.

The physical properties measured were magnetic susceptibility and natural gamma radiation for each borehole and additionally, depending on the site and borehole condition: formation resistivity, fluid temperature, fluid conductivity and triple sonic full-waveform logging. Densely sampled (every 2 or 3 m) density measurements were carried out on the core samples from most of the logged boreholes. Laboratory measurements of density, magnetic susceptibil-ity, magnetite and iron content, performed by Nordic Iron Ore (NIO), are available in the mineralization depth range. Geomechanical measurements were performed as well by NIO, and RQD (rock quality designation) assess-ments are available for some of the boreholes as well. An example of the full range of physical properties measured is illustrated in Figure 4.7.

The mineralization zone is distinctly characterized by high magnetic susceptibly, magnetite content, and density. P-wave velocity does not show a significant increase in the mineralized zones (ranging from 5600 m/s to 6100 m/s) despite the fact that the density increases significantly. A similar observation applies to the S-wave velocities.

Page 46: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Figu

re 4

.7. C

ompl

ete d

ownh

ole l

oggi

ng d

ata f

rom

bor

ehol

e BB1

4005

show

ing

diffe

rent

phy

sical

pro

perti

es m

easu

red

alon

g th

e bor

ehol

e dep

th,

colo

r co

ded

base

d on

the

litho

logy

logg

ed f

rom

the

core

sam

ples

. Ref

lect

ion

time

serie

s w

ere

calc

ulat

ed f

rom

the

acou

stic

impe

danc

e, a

nd

synt

hetic

trac

es w

ere

gene

rate

d by

con

volv

ing

the

refle

ctio

n se

ries w

ith a

100

Hz

Rick

er w

avel

et.

Page 47: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

47

Another observation from the mineralization zone, in particular from the deeper segment, 420–450 m depth, is a decrease in the resistivity from 12000 Ohm-m to about 100–1000 Ohm-m. The host rocks have high resistivities typical for igneous and metamorphic rocks (Telford et al., 1990) and a decrease in the resistivity is usually indicative of metallic minerals. However, the values measured here are at least one magnitude higher than values encountered for massive sulphide deposits, as an example (Nabighian, 1988). Resistivity-based geophysical methods alone will unlikely be able to delineate magnetite mineralization at depth at Blötberget. Moreover, the shallower mineralization zones in Figure 4.7 do not show such a distinctly sharp decrease in the resistivity, suggesting that fracturing and possibly more fluid inflow is to be expected deeper within the mineralization zone. A rapid increase in fluid conductivity is observed at 330 m depth, just before the mineralization zone. This zone is associated with poor RQD values and a washed-up amplitude zone in the far receiver waveform record, suggesting that this zone could represent a fracture zone. The decreased resistivity and increased fluid conductivity in the mineralization zone suggest, however, fractured mineralization, highlighting the importance of linking physical properties with rock quality assessment.

A first step for assessing how elastic properties from downhole logging data may provide information for mineral exploration and planning was to calculate the acoustic impedance using the laboratory density measurements and the corresponding sonic log derived P-wave velocity. Reflectivity series were calculated and convolved with a 100 Hz Ricker wavelet (commonly used for synthetic seismograms), indicating that a strong seismic response might be expected from the mineralized zone (see last plot in Figure 4.7). The strong seismic response clearly separates the mineralization from the host rocks, but it appears that this is primarily due to the increased density of magnetite and hematite. Figure 4.8 shows density against both P- and S-wave velocity from all the logged boreholes, with constant impedance lines. The mineralized rocks are found along lines of increased impedance, confirming that a seismic survey would be successful in delineating the mineralization, with a correct acquisition set-up. However, the mineralization does not follow Birch’s law of linear increase in P-wave velocity with density (Birch, 1939; Nafe and Drake, 1961).

Page 48: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

48

Figure 4.8. (a) Density versus P-wave velocity and (b) density versus S-wave velocity plots. A wide range is observed for the velocities for essentially all the studied rocks types, while magnetite and hematite show distinctly high densities followed by metavolcano-sedimentary rocks.

The previously discussed fracturing within the mineralization zone might explain why the P-wave velocity does not show particularly high values for the ore-bearing rocks. Moreover, high-density rocks, such as magnetite and hematite, have been reported to show a different behavior of elastic parameters (Milkereit and Wu, 2015). The seismic parameter (Kassam et al., 2016) aims to reduce the high-density outliers and explain better the links between elastic moduli and density. Figure 4.9 shows 3D scatter plots of the seismic parameter, bulk modulus and density, color-coded by lithology or by different elastic moduli. The high-density outliers appear to correspond to high seismic parameters values and increased shear modulus, Young’s modulus and Poisson’s ratio. Higher values of these elastic constants suggest less susceptibility to deformation and failure under stress, crucial information for describing material strength for many applications, from engineering to mining.

Page 49: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

49

Figure 4.9. Three parameter (density-bulk modulus-seismic parameter) cross plots color coded by (a) lithology, (b) Poisson’s ratio, (c) shear modulus, and (d) Young’s modulus. Black arrows indicate clusters of outliers that are discussed in the text.

For a full characterization of the host rocks and mineralization at Blötberget, full-waveform data was used for estimating seismic quality QP and finding a correlation between QP and RQD. Seismic quality QP (Knopoff, 1964) was calculated for the boreholes where full-waveform sonic logging was performed and if available, it was compared to RQD assessments as well. An example is shown in Figure 4.10. Several zones where there is a strong decrease in seismic amplitude are identified and marked on the full-waveform plot. These zones correlate with a decrease in the seismic quality factor QP and with poor-quality rocks (< 50% RQD). Especially below 350 m depth, the seismic quality QP has a sharp decrease, suggesting that the mineralization is less competent than the host rocks, despite the increased elastic moduli discussed and previously illustrated in Figure 4.9.

Page 50: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

50

Figure 4.10. Full-waveform sonic log (first panel to the left) and derived P- and S-wave velocities (second panel) for borehole BB14005 together with seismic QP (third panel) and RQD data. Zones of washed-up amplitude in the far receiver full-waveform log correspond to zones of poor or very poor-quality rocks (RQD < 50%) and are indicated with black arrows. Zones where no travel time could be identified reliably for S-waves are indicated with “NaN” in VP/VS plot (last panel).

Three parameter cross plots offer a complete view of seismic QP against RQD, incorporating seismic velocities (Figure 4.11). Unfortunately, only two boreholes had complete measurements of RQD available, and sonic logging

Page 51: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

51

measurements from where QP could be derived. When VP/VS is plotted against RQD and QP (Figure 4.11a) it appears that most rocks, including mineralization rocks, display fair or good RQD, but since highly fractured zones could not have VS clearly identified (Figure 4.10), information is missing for a complete characterization. Figure 4.11b shows only VP incorporated when QP is plotted against RQD. Most of the rocks show good RQD, yet magnetite still shows low seismic quality factor (QP below 20). The attenuation properties of the host rocks and of the mineralization pose thus an interesting question regarding the competence of the ore-bearing rocks in terms of mine planning and method of ore extraction (block caving, room and pillar or cut and fill, etc.).

Figure 4.11. Three parameter cross plots of QP against RQD against (a) VP/VS and (b) VP, for data obtained from boreholes BB14005 and BB14008, showing low QP values for the mineralized rocks despite good RQD assessment. Red dashed lines denote the threshold for low QP observed in Figure 4.10.

Elastic parameters are essential as they describe the behavior of the rocks in terms of deformation and strength, especially at planning stage for a mine. The downhole logging offers continuous measurements along the borehole depth, a clear advantage compared to scattered laboratory measurements, and it is an efficient way to asses in-situ fluid saturation, stress state or fracture zones. At zones of intense fracturing, like the ones identified in Figure 4.10, static tests could not be carried out on the drilled cores.

The downhole measurements should still be linked with static tests performed in the laboratory (King, 1983; Eissa and Kazi, 1988) as the duration of downhole dynamic testing is shorter compared to that of static tests performed in the laboratory. Therefore, rock mechanical properties measured in boreholes suggest higher rock quality estimations compared to laboratory testing. Nevertheless, since crystalline rocks have generally low porosities (Malehmir et al., 2013a; Yavuz et al., 2015), the estimated elastic properties will not be significantly influenced by pressure and frequency. The dynamic elastic properties can be considered close to the static ones and representative for the in-situ rock masses. The rock strength assessment would thus help

Page 52: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

52

estimate if and what kind of reinforcement would be required when mining will commence in the area.

At an exploration stage, the analyzed elastic properties also reveal a strong contrast between the mineralization and their host rocks, indicating that a strong seismic response is expected from a seismic survey in the area. Synthetic seismograms calculated from the density logs and the sonic logging derived P-wave velocities from four boreholes suggest that the mineralization will be successfully delineated if seismic data is acquired with sufficient signal-to-noise ratio, encouraging further field campaigns for delineating the depth extent of the mineralization.

4.2.2 Conclusions Several boreholes have been downhole logged for physical property studies supporting iron-oxide exploration and mining in the Blötberget mining area in central Sweden. The potential for future deep seismic exploration surveys has been illustrated, as the analysis of elastic properties showed a strong contrast between the mineralization and its host rocks.

Additionally, the full-waveform triple sonic log provided key information, as dynamic elastic constants and the seismic quality factor were calculated. Zones of washed-up amplitude in the full-waveform log correlate with zones of poor-quality rocks (low RQD values). However, sonic logging derived seismic attenuation shows that even fair or good quality rocks according to the RQD assessments, could in fact be less competent than expected. A decrease in the seismic quality factor was observed in the mineralized zones, suggesting that despite increased elastic moduli and density, the ore-bearing rocks might be less competent than the surrounding host rocks. Thus, mining operations should be carried out considering the eventual need for rock support and reinforcement.

The key findings of this study are summarized further: • Downhole sonic logging and geomechanical measurements are linked by

correlating attenuation of seismic waves with RQD assessments;• The downhole physical property data suggest mechanically low

competent rocks in the mineralization zone;• An immediate implication for future mineral exploration is that seismic

surveys are expected to be successful in delineating the mineralization.

Page 53: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

53

4.3 Paper III: Deep reflection seismic imaging of iron-oxide deposits in the Ludvika mining area of central Sweden

After an encouraging physical property study and an estimated strong seismic response expected from the mineralization, reported in Paper II, several geophysical surveys followed at Blötberget for delineation of the mineralization. In Paper III reflection seismic data acquired within two field campaigns (2015 and 2016) were merged and processed together for an improved resolution for deep imaging of the mineralization at Blötberget, while employing cost-effective 2D seismic surveys.

4.3.1 Summary The mineralization at Blötberget is known to extend down to about 800–850 m based on primarily borehole observations at the site. An initial seismic survey (Malehmir et al., 2017b) identified strong reflections originating from the mineralization zone and confirmed the depth extent suggesting a likely downdip continuation of the mineralization. The seismic survey was initially designed as a pilot test for testing the capability the landstreamer (Brodic et al., 2015) to image mineral deposits at depth in a hardrock environment, and led to a second seismic survey in 2016, for imaging the mineralization and geological structures at greater depth in the area. Table 4.2 summarizes the main acquisition parameters within the two seismic campaigns.

Table 4.2. Summary of main seismic acquisition parameters (2015 and 2016).

Field campaigns October 2015 Profile 15P1

September 2016 Profile 16P1

Acquisition type Move along using landstreamer and fixed wireless recorders

Fixed-line geometry and cross-line shooting

Receivers DSU3 (3C MEMS) 10 Hz geophones

10 Hz geophones

No. of receiver locations 1049 451 No. of shot points 533 387 Receiver interval 2-4 m (10 m for wireless receiver

location)5 m (10 m for wireless receiver location)

Shot interval 4 m 5 m Maximum source-receiver offset

~2500 m ~2200 m

Source 500 kg Bobcat-mounted drophammer and explosives (charges ranging from 50-100 g)

500 kg Bobcat-mounted drophammer

Sampling rate 1 ms 1 ms

Page 54: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

54

The two seismic profiles were partly overlapping, but profile 15P1, with a total length of 3500 m extended more to the south compared to profile 16P1 (2400 m length). Profile 15P1 consisted of 100 MEMS sensors spaced at 2-4 m that were moved along the profile and 75 wireless sensors connected to 10 Hz geophones spaced at 10 m that were also moved, but only once, from the southern end of the profile to the northern end. Profile 16P1 had a fixed geometry, consisting of cabled 10 Hz geophones spaced at 5 m and twenty-four wireless sensors connected to 10 Hz geophones spaced at 10 m.

The seismic data were initially processed following a conventional prestack workflow (Malehmir et al., 2013a) and post-stack migration. In order to increase the data fold, and hence, the resolution, the two datasets from 2015 and 2016 were merged and processed together. The sensors used are different in the two field campaigns (Table 4.2), and even though the two types of sensors (i.e. MEMS and geophones – acceleration vs. velocity domains, respectively) are integrated by filtering the data – balancing spectra, gapped deconvolution (Malehmir et al., 2017c), the merged data set consisted only of geophone data.

Figure 4.12 shows the stacked sections of each dataset and of the merged datasets, together with the CDP fold for each stacked section. High-amplitude reflections are visible in the 2015 data alone (Malehmir et al., 2017b) and are associated with the mineralization at Blötberget (Figure 4.12a). While the 2015 data provide a good insight of a possible extension below the known 850 m depth, the 2016 seismic data provide considerably more detail (Figure 4.12b). The merged datasets (Figure 4.12c) show more continuity and a second set of reflections associated with a new mineralization horizon.

When further processing (FDCT and residual static corrections) is applied, the reflectivity pattern is more continuous with a higher amplitude content (Figure 4.12d). Fast discrete curvelet transform (FDCT) was applied on the merged datasets through a curvelet de-noising algorithm for further attenuation of surface waves that mask the reflections associated with mineralization. Curvelet de-noising has not been tested specifically for removing surface waves, but it has been applied in structurally complex environments (Neelamani et al., 2008; Górszczyk et al., 2014; Ketelhodt et al., 2019).

To support the claim of improved signal-to-noise ratio in Figure 4.12d, the RMS amplitude was averaged for a time window above the main reflective package (i.e., 0–220 ms) and for another one within the reflective zone (i.e., 220–500 ms) and within the same CDP range (500–680). Their ratio was calculated before and after curvelet de-noising was applied, resulting in an improvement from 2.2 to 2.8, approximately 30% signal enhancement.

Page 55: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

55

Figure 4.12. Unmigrated stacked sections of (a) the 2015 dataset, (b) the 2016 dataset, (c) merged 2015 and 2016 datasets with conventional processing and (d) mergeddatasets after FDCT and residual static corrections.

Figure 4.13 shows the entire merged section with a length of 3600 m, which gives an overview of the large-scale structures along the seismic profiles. Reflections M1 and M2 represent the interpreted mineralization with a possible underlying mineralized horizon. Oppositely dipping to northwest, several sets of reflections can be identified (R1, R2 and R3). The apparent gap in the data at 2000–2200 m along the profile is due to a gap in receivers at a high-traffic road crossing resulting in low data fold. This region coincides with the flattening out of the mineralization such that it is unclear whether R1 and R2 intersect the mineralization or the mineralization stops. Due to the lack of boreholes and outcrops in the southern end of the profile, the reason for the flattening out of the mineralization remains unclear. Possible scenarios could

Page 56: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

56

include a northwest dipping fault contact or a granitic intrusion into the volcanic rocks.

Figure 4.13. Unmigrated stacked sections along the entire portion of the merged 15P1 and 16P1 datasets after FDCT and residual static corrections.

After migration (Figure 4.14a), the reflections associated with the mineralization extend well below the known depth of 850 m, and appear to flatten out at 1200 m depth, which is where the seismic fold is at its lowest as previously described. Possible additional mineral resources below M1 and M2 are likely justifying additional drilling and exploration investment in the area.

The 3D visualization illustrates an excellent match with the known deposits (Figure 4.14b). The previously logged boreholes (Paper II) show increased density values where the main reflection package is intersected (see an example in the inset in Figure 4.14). A possible continuation of the mineralization from the known 850 m depth to 1200 m depth implies an increase up to 30% in the mineral resources at Blötberget, assuming the same layer thickness and lateral extent.

Such estimation from seismic data is highly encouraging for the use of seismic methods for mineral exploration and targeting, especially when a low-cost seismic source was employed. The terrain conditions at Blötberget (soft and swampy ground) are expected to be highly attenuative, but the Bobcat-mounted drophammer, 500 kg, proved to be an effective source for both seismic campaigns. Already tested as a practical and efficient source for the 2015 field campaign, the source’s penetration depth was tested again with a conventional receiver setup (i.e. cabled geophones). Moreover, the Bobcat-

Page 57: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

57

mounted drophammer employs considerably less logistics and costs than a commercial Vibro-truck, even though the 2016 survey has a receiver set up comparable to a commercial survey. The highest costs associated with a commercial source, the source mobilization and demobilization at site, were avoided in both seismic campaigns. At this initial exploration stage, both surveys serve for future feasibility studies for optimizing drilling programmes and resource assessment for the future mine.

Figure 4.14. 3D visualization of the migrated seismic section of the merged datasets with (a) recent boreholes downhole logged with density measurements (see the closer view on the right side) and (b) the known ore bodies (blue and red surfaces) suggesting their possible continuation down to about 1200 m depth.

4.3.2 Conclusions Two reflection seismic datasets from two different field campaigns (2015 and 2016) using different setups have been merged and processed together for an improved delineation of iron-oxide mineralization in Blötberget, Ludvika mining area of south central Sweden.

Southeast dipping strong reflections in the data are associated with the mineralization and continue beyond the known 850 m depth, down to 1200 m, suggesting several additional resources downdip that might be worth investigating in future seismic or geophysical studies. The oppositely dipping reflections mark possibly large-scale structures in the area that if further investigated might explain the apparent flattening out of the mineralization.

Page 58: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

58

This study illustrates the use of seismic methods for mineral exploration and in particular their cost-effectiveness when a low-cost seismic source is used.

4.4 Paper IV: Cross-profile seismic data acquisition, imaging and modeling of iron-oxide deposits: A case study from Blötberget, south central Sweden

The iron-oxide mineralization at Blötberget was successfully inferred down to 1200 m depth from the 2D seismic profiles of the two field campaigns (2015 and 2016), as described in Paper III. One of the two profiles acquired in 2016 was previously processed and discussed together with the 2015 seismic data, in Paper III, while Paper IV focuses on both seismic profiles from the 2016 campaign. Cross-profile recording was performed from each profile thus four datasets were recorded and all data were processed as swath 3D.

4.4.1 Summary Two 2D profiles were acquired at Blötberget in 2016 with the main objective of delineating the depth and lateral extent of the iron-oxide deposits. Profile 1, approximately 2400 m long, was perpendicular to the strike of the known mineralization, along the downdip direction. Processing of profile 1 was discussed in detail in Paper III and the mineralization was imaged in the downdip direction to at least 1200 m depth.

Profile 2, consisting of 10 Hz geophones connected to wireless recorders spaced at 10 m, was about 700 m long and was crossing profile 1 from west to east. Due to terrain accessibility, shooting was performed only on the eastern part of profile 2. The main objective of this additional profile was to estimate the lateral extent of the mineralization and investigate whether a cross-dip (out-of-the-plane feature) can be seen in the data. However, due to the short length and unfavorable ground conditions (close to a water stream, soft and swampy ground conditions), profile 2 is poor in quality and reflectivity, both in shot gathers and the stacked section.

During the data acquisition both profiles were active, thus, while shooting along one profile, recording was performed along both profiles. Four datasets were acquired in total and in a novel approach to extract 3D information about the depth and lateral extent of the mineralization, the 2D profiles were binned together in a 3D grid and processed as 3D. Figure 4.15 illustrates the two seismic profiles and the 3D grid with the CDP data fold.

Page 59: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

59

Figure 4.15. (a) The two 2D seismic profiles with the CDP fold and 3D grid showing inline and crossline directions. For display purposes, the CDP fold is limited to 150, as most of the data are found within this limit; (b) Offset azimuth distribution of the seismic data showing most of the data follow the NW–SE dip direction.

The initial 2D processing of the seismic data revealed several reflections associated with the mineralization along profile 1, identified in Figure 4.16 as M1–M4. Another set of reflections, R1 and R2 (identified in the stacked section) are associated with northwest dipping large-scale structures, and are discussed later in a 3D visualization.

In order to investigate the seismic response of the mineralization and validate the origins of the reflections observed in the data, several planar reflectors were modeled using a 3D ray-tracing approach (Ayarza et al., 2000), assuming a 3D geometry similar to that of the known mineralization. The traveltime modeling assumes a constant velocity medium and 3D planar surfaces with contrasting properties. The calculation of the traveltimes of the modeled reflectors is based on the real data acquisition setup geometry: receiver coordinates for reflector modeling on the shot gathers and CDP coordinates for the stacked section. The background velocity was set to 5600 m/s and the density 2700 kg/m3, whereas for the reflectors, the velocity was set to 6000 m/s and 4000 kg/m3 for the density, based on mean values from existing downhole logging measurements at the site.

The modeled traveltimes were overlaid on the shot gathers and stacked section along profile 1 (Figure 4.16). M1 and M2 match well the strong reflections observed in the shot gathers (Figure 4.16c). The later reflections matching M3 and M4 are weaker. However, the stacked section shows a more repetitive reflection pattern and all modeled traveltimes are matched by the traveltimes of the observed reflections (Figure 4.16d). There is a change in the dip of the observed reflections, as the mineralization flattens out with depth and thus the match is not accurate after the first 0.3 s. Nevertheless, the modeling work is important as it indicates that new mineralized horizons

Page 60: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

60

might exist below the known ones. The 3D geometry of the modeled reflectors is summarized in Table 4.3.

Figure 4.16. Processed data along profile 1: (a) Shot gather and (b) stacked section (modified from Paper III) with modeled reflections (in color) overlaid on top (c and d). Arrows indicate reflections originating from the mineralization, labeled M1–M4. Oppositely dipping structures, discussed later, are identified from reflections R1 and R2.

Page 61: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

61

Table 4.3. Reflector geometry as determined from the seismic modeling. The distance refers to the distance from a reference point to the point where the reflector would intersect the surface. The reference point is the northwestern end of profile 1.

Reflector Strike (°) Dip (°) Distance (m)

M1 55 35 200 M2 55 35 400 M3 55 30 800 M4 55 27 1200 R1 235 38 2100 R2 235 30 3200

For consistency, the same reflector modeling was performed for receivers along profile 2, despite the lack of reflections along this dataset. Given the geometry of the reflectors and receivers, nearly horizontal reflections are expected. Information on the extent of the mineralization could be inferred from such reflections, if they were observed in the data.

However, a different approach was followed for obtaining information from profile 2 instead, merging the 2D seismic profiles and processing them as 3D. Data were binned together in a 3D grid (Figure 4.15), consisting of 60 inlines and 436 crosslines, with the inline direction along profile 1. Crossline spacing (distance between adjacent crosslines) was set to 5 m and inline spacing to 10 m. After individual processing, 3D processing was performed on the merged pre-stack processed 2D profiles.

The cross profile (i.e. profile 2) on its own is poor in reflectivity, but integration into a semi 3D cube provides additional information with regards to the lateral extent, showing at least 300 m width of the deposits. Results show that several mineralized horizons might underlie the known ones. Strong reflections are observed throughout the inline sections (Figure 4.17), from inline 1005 to inline 1037. At 10 m inline spacing, we can estimate a 300 m lateral extent of the mineralization. Consistent with the previous notation, the known mineralization is identified as M1 and a possible new one as M3. The 300 m lateral extent is essentially limited by the length of profile 2, illustrating that in spite of this limitation, 3D information can be extracted from 2D seismic profiles.

Page 62: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

62

Figure 4.17. A series of inlines from the 3D seismic cube showing the change in the reflection pattern from inline 1005 to inline 1037: (a, b, c) unmigrated sections and (d, e, f) migrated sections. Based on the reflection package identified in Figure 4.16, the inline reflections are marked as M1, corresponding to the mineralization, and M3, a possible new mineralization layer.

An overview of the 3D seismic cube (Figure 4.18) shows that the crosslines are visibly smeared due to the migration process, but along the inline, the observed reflections (M1 and M2) match the mineralization with more accuracy after migration. A different dip is observed for the second set of reflectors, M3 and M4, associated with new mineralization layers, both in the inline and crossline direction.

Page 63: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

63

Figure 4.18. A visualization of the 3D cube along inline 1010 and crossline 1272: (a) unmigrated; (b) migrated. M1 and M2 indicate reflections originating from the mineralization, clearly matching the ore block model in the migrated seismic 3D cube view. M3 and M4 appear as distinct reflections with different dips.

In order to further validate the interpretation of the seismic data, two scenarios regarding the geometry of the mineralization were modeled using an elastic 2D finite-difference algorithm (Juhlin, 1995). Based on the known information on the geometry of the deposits at Blötberget, and the reflector modeling presented previously, the two models (Figure 4.19) were designed to investigate the seismic response of a hypothetical new mineralization layer underlying the existing ones and for comparison with real data.

Page 64: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

64

Figure 4.19. Constructed P-wave velocity models used in the finite-difference modeling, with the location of profile 1 overlaid. The models are based on a cross-section through the orebody model along the 3D cube inline direction, for two scenarios: (a) two mineralized layers and (b) three mineralized layers.

The physical properties of the background medium and of the mineralization were the same as for the 3D ray-tracing modeling, based on the mean values from downhole logging. The model was constructed from a 2D cross section through the mineralization and along profile 1 (as well as inline direction of the 3D cube), but extended to 2800 m horizontally. Depth of the model was set to 2400 m. The model had a free surface condition at zero depth and absorbing boundaries, with a 2 m cell size and 280 receivers (vertical geophones) with shots placed at every 10 m at 280 positions.

Figure 4.20 shows the unmigrated stacked sections using synthetic data generated from the models presented in Figure 4.19, and the stacked section along profile 1 for comparison. Despite some backscattered noise from the model boundaries, the target response is easily identified for each model as strong reflections arriving at similar times as in the real data. It is evident that the extra mineralization layer in the model in Figure 4.19b generates a reflection pattern more similar to the one recorded in the real data, especially between 0.2–0.4 s.

Both the traveltime and finite-difference modeling show that the physical properties of both the host rocks and mineralization are well understood, and synthetic data show an excellent match with the real data. Moreover, the forward modeling strongly indicates that the repetition of the reflections in the unmigrated stacked section is not an artifact or due to processing pitfalls, but a reasonable scenario, should another mineralized layer exist below the known one.

The modeled reflectors (Figure 4.16) are constructed as 3D surfaces and are visualized together with the migrated section along profile 1 (Paper III), with the aim of integrating the seismic data with the surface geology (Figure 4.21). Figure 4.21a focuses on the reflectors associated with the mineralization, M1–M4. Their match with the reflections observed in the

Page 65: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

65

seismic data is excellent down to about 400 m, after which depth, the dip of the reflections in the seismic section is becoming gentler. We interpret M1 as the orebody (depicted with grey), since reflector M1 is projected to the surface at the exact location of the orebody. Reflector M2 appears to be at the same location as M1 along the seismic data (Figure 4.21a), but on the surface geological map (Figure 4.21c) and based on the reflector geometry described in Table 4.3, M1 and M2 are projected at the surface 200 m apart. Note that the 3D traveltime modeling assumes planar reflectors, while the orebody has a more complex 3D geometry. Reflector M4 matches the surface contact between the granite and volcanic rocks, marking a possible granite intrusion contact. Despite no clear surface marking, M3 is inferred to originate from another mineralized horizon, a scenario suggested by the finite-difference modeled synthetic data (Figure 4.20).

Another view of the migrated section (Figure 4.21b) focuses on the oppositely dipping reflectors R1 and R2. Reflector R1 projects at the surface at 2100 m from the reference point towards southwest, at the southern granite contact with the volcanic rocks. Reflector R2 projects further south from the end of the profile, and even though it is uncertain at this point, it is possible that it represents a large-scale structural feature such as a fault system as speculated in Paper III and by others (e.g., Bräunig et al., 2020). The geological implications of these features oppositely dipping to the mineralization are currently unclear, however, they can be significant and play a role for the geological setting of the deposits or their post-emplacement geometry.

Page 66: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

66

Figure 4.20. (a) Unmigrated stacked section of the synthetic data generated from the model in Figure 4.19a, with reflections M1 and M2. (b) Stacked section of the synthetic data generated from the model in Figure 4.19b, with a new reflection identified as M3. Only data along profile 1 are displayed. (c) Unmigrated stacked section along profile 1 modified from Paper III, with the interpreted mineralization reflections.

Page 67: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

67

Figure 4.21. 3D visualization of the orebody, 2016 migrated seismic data (modified from Markovic et al., 2020) and the reflectors modeled and displayed in Figure 4.16, constructed as surfaces in GOCAD. (a) Overview of reflectors M1-M4 and their match with the seismic data. (b) Overview of reflectors R1 and R2 and their match with the seismic data. (c) Geological map showing the seismic profiles, profile 1 and 2, and the surface projection of all reflectors with respect to the reference point described in Table 4.3.

4.4.2 Conclusions Two 2D seismic profiles, including cross-profile data acquisition, have been acquired for seismic imaging of iron-oxide mineralization in Blötberget, south central Sweden.

All the data followed conventional 2D processing and subsequently were binned together in a 3D grid and processed accordingly. The mineralization identified as a set of southeast dipping reflections is successfully imaged in the semi-3D cube, downdip to 1200 m and 300 m in lateral extent, confirming that this approach for estimating 3D geometry is preferable to processing individual 2D profiles. Reflector modeling of several 3D planar surfaces provides an insight to possible new mineralization, as well as potential large-scale geological features. Finite-difference elastic modeling helps explain the seismic response of the mineralization and the nature of the reflectivity pattern. Results are encouraging for the use of both 2D and 3D seismic surveys for mineral exploration, and particularly for de-risking a future 3D survey in the area.

Page 68: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

68

5 Conclusions

The potential of seismic methods for both near-surface environments and deep mineral exploration applications has been illustrated through the research presented in this thesis and a number of case studies. Seismic data were acquired for efficient exploration of groundwater and mineral resources at two sites in Finland and Sweden. The effectiveness of these seismic surveys was revealed by good quality seismic data acquired using a low-cost, easily operated seismic source and different sensors, including a broadband seismic landstreamer. Physical property studies were integrated with seismic methods for validating the seismic response of iron-oxide mineralization and to characterize the mineralization and its host rocks. The seismic attenuation was calculated from full-waveform sonic data and its correlation with rock quality was analyzed, with the goal to aid mine planning and exploration.

The main objectives of the thesis were attained through various research summarized in the four papers. In Paper I the main objective of the seismic data was to delineate the large-scale units of an aquifer used for groundwater filtering in southwest Finland. Combining seismic first-break traveltime tomography with seismic reflection imaging, the hydrogeological relevant units of a complex esker system were characterized for their impact on groundwater infiltration and flow. The seismic data interpretation was guided by hydrogeological units within an existing 3D groundwater flow model, helping the long-term planning and sustainable use of the managed aquifer recharge plant. An additional goal of Paper I was to test the applicability of the cost-effective seismic survey set up (e.g. seismic landstreamer) and low-cost seismic source for a challenging near-surface environment, consisting of soft and thick glacial sediments. The seismic data acquired proved to be of excellent quality and was acquired in a time-effective manner with an easily operated seismic source.

This goal was extended to Papers III and IV, where the same seismic source and similar acquisition set up was used in a hardrock environment, for iron-oxide delineation at Blötberget, Sweden. Despite the swampy terrain covering the bedrock, the seismic data acquired had adequate penetration depth and proved successful in delineating the mineralization at depth, the next objective of this thesis. The seismic data at Blötberget were acquired within two field campaigns (2015 and 2016). Merging the seismic data from the two surveys helped image the mineralization at depth, as it was interpreted

Page 69: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

69

to possibly extend below the known depth of 800–850 m, to 1200 m depth, as detailed in Paper III.

Before acquiring the seismic surveys at Blötberget, an important objective was to validate the seismic response of the mineralization through physical property studies. Paper II presents results from downhole logging measurements at Blötberget and how a strong seismic response was estimated from synthetic seismograms based on seismic velocity and density measurements at the site. Additionally, the elastic property study was extended for rock quality and seismic attenuation characterization that can be used at the mine planning stage. Several fracture zones were identified from the full-waveform sonic logging, suggesting low mechanical competence for the mineralization. A decrease in the seismic quality factor, derived from the full-waveform sonic logging data, indicated that the mineralized zones, despite showing good rock quality assessments, might require additional support and reinforcement, compared to the host rocks, when mining operations begin.

Paper IV reports 3D processing of the two 2D seismic profiles acquired in 2016, including cross-profile data, aimed at delineating the 3D geometry (i.e. lateral extent) of the mineralization. The seismic data suggested a lateral extent of at least 300 m, based on the 3D seismic cube. In order to understand whether the repeated reflection pattern identified in the data could represent new mineralized resources, the seismic response of the mineralization was forward modeled using 3D ray-tracing and 2D finite-difference algorithms. Synthetic data from a scenario consisting of a new mineralized layer below the existing ones showed a similar reflection pattern when compared to the real seismic data, and thus additional mineral resources were inferred below the known mineralization. Integration with surface geology suggested as well important structures that may crosscut the mineralization.

The studies presented in this thesis illustrate the potential of seismic methods for efficient natural resources exploration in crystalline rocks and in overlying glacial sediments. Achieving resource efficiency, often considered only in terms of extraction and use, should start with efficient exploration that utilizes non-invasive, cost-effective geophysical methods. A complete characterization of the subsurface is necessary for successful exploration and exploitation of natural resources. Furthermore, seismic data can successfully be integrated with available borehole data, physical property studies or validated by forward modeling the seismic response of the targets.

Page 70: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

70

6 Outlook

The mining at Blötberget is planned to resume in the near future, pending new assessments of the potential mineral resources and upgrading of the mining infrastructure at the site. Since it is an ongoing project, ideas for future work are mostly related to this study area and could be continued in several directions.

Essential information could be further obtained from connecting seismic Q (Paper II) and rock mass quality QR (Barton et al., 1974), that has been shown to correlate with seismic P-wave velocity (Barton, 2006). A direct comparison between QR and QP or QP/QS might enable an improved understanding of the causes of seismic attenuation in the mineralized zone, as well as a better rock characterization, including additional estimates for rock competence as required for mining operations and for improving the design for underground support in the mine.

The mineralization in Ludvika has a significant contrast in the physical properties compared to their host rocks, particularly in terms of density. It is known that low impedances due to similar velocities in crystalline environments can make the detection of deposits difficult (e.g., Salisbury et al., 2003). Since seismic velocities in Ludvika do not show a significant increase for the mineralization zone, further statistical analysis will help estimate the scale lengths of the heterogeneity of the background medium and can be used in further modeling of the seismic response of the mineralization. Even though heterogeneous medium synthetic modeling is in general applied for estimating the seismic response from small ore bodies (L’Heureux et al., 2009; Cheraghi et al., 2013), it is worth investigating the effects of seismic scattering at Blötberget.

Since full-waveform sonic logging was performed at Blötberget in several boreholes, there is a good estimate of the seismic velocities of the subsurface. These estimates, however, represent only the immediate vicinity of these particular boreholes. Accurate velocity information that can correlate the boreholes with surface seismics could be further obtained from vertical seismic profiling (VSP) in the area. VSP data can detect velocity anomalies not recorded by sonic logging, and can help interpretation of the surface seismic data, especially from a 3D seismic survey.

Page 71: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

71

7 Summary in Swedish

Naturresurser i allmänhet, och mineralfyndigheter i synnerhet, är mycket vik-tiga för alla sektorer inom den globala ekonomin. Utvecklingen av industrin blir mer och mer beroende av olika metaller, i synnerhet genom den världs-omspännande övergången till fossilfria energikällor. En hållbar framtid för vårt samhälle kräver ett effektivt resursutnyttjande, inte bara vid gruvprodukt-ion, utan redan på prospekteringsstadiet.

Möjligheterna med att använda seismiska metoder både vid ytnära projekt och för prospektering på större djup presenteras i denna avhandling med ett antal exempel. Seismisk datainsamling användes för att effektivt prospektera för grundvatten och mineralfyndigheter på två platser i Finland och Sverige. Avhandlingen visar att dessa mätkampanjer effektivtkunde samla in högkva-litativa seismiska data med en relativt billig och lättmanövrerad seismisk källa samt med olika sensorer, bland annat en bredbandig seismisk landstrea-mer. Fysikaliska egenskaper studerades och integrerades med seismiska me-toder för validering av den seismiska responsen från järnoxidminerali-seringar och för att karaktärisera mineraliseringens och sidobergets egen-skaper såsom seismisk dämpning och bergkvalitet, vilket är ett stöd för gruv-planering och prospektering.

Syftena med avhandlingen uppnåddes genom forskning som har samman-fattats i de fyra artiklarna. I Artikel I var det huvudsakliga målet att avbilda de storskaliga enheterna av en akvifär som används för att filtrera grundvatten i sydvästra Finland. Genom att kombinera first-break traveltime tomografi och seismisk reflektionsavbildning så kunde de hydrogeologiskt relevanta enhet-erna av ett komplicerat eskersystem karaktäriseras och dess påverkan på grundvatteninfiltration och vattenflöden bestämmas. Tolkningen av seismisk data styrdes av hydrogeologiska enheter i en existerande tredimensionell grundvattenflödesmodell, vilket hjälper den långsiktiga planeringen och ger ett uthålligt vattenuttag från akvifären. Ett annat mål med Artikel I var att testa användningen av ett kostnadseffektivt seismiskt system (t.ex. en seismisk landstreamer) och ett lågkostnadsalternativ för seismisk källa för en utma-nande ytnära undersökning i en miljö bestående av mjuka och tjocka glaciala avlagringar.

Det senare målet utvecklades även i Artiklarna III och IV. Samma seism-iska källa och ett liknande datainsamlingssystem användes i kristallint berg för att avbilda järnoxidmineraliseringar i Blötberget, Sverige. Trots att en blöt-

Page 72: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

72

myr täckte berggrunden så gav den seismiska datainsamlingen adekvat djup-penetration och visade sig framgångsrik i att avbilda mineraliseringen på dju-pet, vilket var det andra målet med dessa två artiklar. Seismiska data från Blöt-berget samlades in under två fältkampanjer (2015 och 2016). Genom att sätta ihop de två dataseten kunde mineraliseringen avbildas på djupet, och tolk-ningen var att mineraliseringen kanske fortsätter nedanför det redan kända djupet 800 – 850 m, ända ner till 1200 m under markytan, vilket är beskrivet i Artikel III.

Ett viktigt delmål innan det seismiska fältarbetet i Blötberget var att utvär-dera den seismiska responsen från mineraliseringen genom studier av de fysi-kaliska egenskaperna hos bergarterna. I Artikel II presenteras resultaten från borrhålsloggning i Blötberget, resultaten visar i syntetiska seismogram att en stark seismisk respons kunde förväntas, vilket baseras på mätningar av seism-isk hastighet och densitetsmätningar på plats. Vidare utvecklades studierna av elastiska egenskaper till att också omfatta bergkvalitet och seismisk dämp-ningskaraktärisering vilket kan användas vid gruvplanering. Flera sprickzoner identifierades från sonisk fullvågform-borrhålsloggning, vilket indikerar låg mekanisk kompentens i mineraliseringen. En uträknad lägre seismisk kvali-tetsfaktor från fullvågforms-borrhålsloggningen för mineraliseringen jämfört med för sidoberget, trots till synes bra bergkvalitet, visar att man kan förvänta sig att ett större behov av bergmekanisk förstärkning kan behövas vid drivning i mineraliseringen jämfört med drivning i sidoberget.

Artikel IV behandlar 3D-processering av de två 2D-seismiska profilerna som mättes 2016. Detta inkluderar mätningar mellan profilerna i syfte att av-bilda 3D-geometrier (lateral utsträckning) av mineraliseringen. Resultaten in-dikerar en lateral utbredning på åtminstone 300 m baserat på 3D-seismiska data. För att förstå om repeterade reflektionsmönster som identifierades i dessa data kunde representera tidigare okända mineraliseringar så utfördes forwardmodellering med 3D-ray-tracing och 2D-finit-differans algoritmer. Syntetisk data från ett scenario med en ny mineralisering under den redan kända visade reflektionsmönster som liknade de som kom från den verkliga seimiska avbildningen, vilket implikerar att det verkligen finns ytterligare en mineralisering som tidigare varit okänd. Integration med ytgeologi indikerar också viktiga strukturer som kan korsa mineraliseringen.

Studierna som presenteras i denna avhandling illustrerar potentialen med seismiska metoder för att effektivt prospektera för olika typer av naturresurser i både berggrunden och de överliggande glaciala avlagringarna. Kampen för att nyttja naturresurser effektivt brukar fokuseras på uttag och användande av dem, men borde börja redan i prospekteringsskedet genom användning av icke-förstörande och kostnadseffektiva geofysiska metoder. En fullständig ka-raktärisering av underjorden är nödvändig för framgångsrik prospektering och nyttjande av naturresurser. Seismisk data kan framgångsrikt integreras med tillgänglig borrhålsdata och fysikaliska egenskaper eller valideras med hjälp av forwardmodellering av responsen från det undersökta objektet.

Page 73: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

73

Acknowledgements

My PhD journey has come to an end and it is filled with so many great memories and yet, it all passed so fast. I still remember the first day I visited Geocentrum and how welcoming everyone was. I am truly happy to have spent these last years in Uppsala and especially to have been so lucky to meet so many great people along the way. This exciting and challenging journey would not have been so wonderful without all of you.

I would like to first thank my supervisor Alireza Malehmir for giving me this opportunity and always pushing me to do more. I am thankful for your support and guidance from seismic data processing and manuscript writing, to encouraging me to participate in project meetings and conferences. Thank you for the productive discussions, your enthusiasm and energy, advice and friendship along the years. I am truly thankful for all your support.

Thank you, Christopher Juhlin, my second supervisor. I learned from you simply by attending the Friday Applied Geophysics meetings and seeing how you get across the essential every time. Thank you for your comments and suggestions, always to the point and very helpful.

There are so many people in the Geophysics corridor I would like to thank for all the discussions, advice, fieldworks and conferences, coffee breaks, TGIFs, parties, trips and so much more, it would be hard to know how to sum up everything. I am happy I got to know, work and have fun during these years with Bojan, Omid, Daniel, Suman, Shunguo, Magnus, Emil, Azita, Joachim, Peter, Maria, Thomas, Ayse, Bjarne, Monika, Sahar, Silvia, Fred, Sebastian, Tegan, Rémi, Zeinab, Angeliki, Claudia, Samar and the “new ones”: Mohsen, Laura, Ruth, Paula, Thorben, Alba, Michael, Magdalena, Tatiana, George, Jan, Joshi, Alex. I appreciate my office mates, Nehemiah, Ruixue, Rodolphe, for funny and friendly conversations. Thank you Bojan, for always being there to lend a hand, be it Claritas, Matlab, fixing a bike or just being a great friend. Thank you, Alba for the nice chats and coming all the way to my office to remind me about the coffee breaks. Thank you everyone for the e-fikas in the last months!

Geocentrum is a place full of friends, not only from the Geophysics corridor: Iwa, Ruta, Pauline, Canadian Chris, Jarek, Weronika. Thank you for the awesome Sunnersta parties, board games, trips, and all the other great memories. Even those not working in Geocentrum are still geo-related, and I am very happy to have shared lots of fun times with Marta, Dragos, Nada,

Page 74: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

74

Christos, Simon, Rudi. I am sorry if I am forgetting someone. A huge thank you to all of you.

I enjoyed participating in all the fun, interesting and sometimes exhausting fieldworks. I will never forget how exciting (“Vad spännande!”) it is, at least in Varberg, southwest Sweden, to measure the depth to bedrock. Thank you, Silvia, Bojan, Suman and Emil for helping in the Turku fieldwork for my esker seismic data. It was a great fieldwork with long but sunny days, and nice swims in the Finnish lakes. Thank you Bojan for encouraging me to drive the Bobcat and not miss out on the fun! Thank you, Joachim for helping with the downhole logging measurements. Thank you, Sara Eklöf, for accompanying me so many times at Blötberget, and especially for all the time spent doing density measurements.

I would like to acknowledge the consortia that made the research presented in this thesis possible: Trust 2.2 Formas project with Turku Region Water Ltd, GTK and University of Turku, ERA-MIN StartGeoDelineation project sponsored by Vinnova, SGU, Tekes, Nordic Iron Ore (NIO), Yara and H2020 Smart Exploration project. I thank NIO, specially Emma, Monika and Paul, for their collaborations in all studies. Thank you, Emma for checking the Blötberget section of the thesis. And thank you everyone involved in these three projects, for interesting work, discussions and project meetings during the years. Thank you Mehrdad for the nice discussions at the various meetings and workshops.

Thank you to my Finnish co-authors, Elina Ahokangas and Joni Mäkinen, for sharing their extensive knowledge and enthusiasm for eskers. Special thanks go to the people who took time to read this thesis and provided very helpful feedback and comments: Bojan, Alba, Silvia, Fred, Alireza and Chris. Thank you, Karin for improving the geology parts. A huge thanks goes to Magnus and Emil for helping with the Swedish summary.

Thank you, Anna, Fred and Erik for help with printing posters, TGIFs or other little things, Leif for cheerful news and stories and Fatima for all the help navigating through the administrative issues. Thank you, Karin for your precious help when bringing Geologiska Sektionen and Otterborg stipend back to life. And thank you, all friends involved in the fun GS activities!

Thank you Iwa and Bojan for being so welcoming in the first year of my PhD, when I still didn’t live in Uppsala. With you, I already had a home in Uppsala, long before moving here. Thank you, for waiting for us at our new house together with Ruta and Shunguo, as we had just a “little bit” left to pack and we were a “little” late Thank you, and Anna and Chris for great times in Sunnersta, always! Thank you, Magnus for fun chats, debating Swedish grammar among others, and offering me a nice place to stay in Flogsta.

I am so happy to always find good advice and chats from Silvia, Fred and Tristan, Daniel and his family, Francisco and his family, Ruth, Lars and the twins, Monika, Christos and little Astrid, or simply enjoy and have fun with the kids at the playground or elsewhere.

Page 75: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

75

I am grateful to have good friends both in Romania and Portugal, and thank you for always making us feel welcome and happy, even over short visits. A big thank you goes to João’s family, especially to Teresa, for a warm welcome whenever we traveled to Portugal. I am most grateful for my family in Romania, for your encouragement and love. Thank you for taking care of us every time we visit, especially in the last two years. Special thanks go to my brother and his family, for making everything fun and kid friendly during our visits!

Sweet little Sara, you fill my heart with joy, with your laughter and happiness. I hope that one day you will read this thesis. Last but not least, thank you my dear João for your continuous love and support, especially in these last months.

Georgiana Mărieș, Uppsala, April 2020

Page 76: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

76

References

Adam, E., G. Perron, B. Milkereit, J. Wu, A. J. Calvert, M. Salisbury, P. Verpaelst, and D.-J. Dion, 2000, A review of high-resolution seismic profiling across the Sudbury, Selbaie, Noranda, and Matagami mining camps: Canadian Journal of Earth Sciences, 37, 503–516.

Allen, R. L., I. Lundstrom, M. Ripa, and H. Christofferson, 1996, Facies analysis of a 1.9 Ga, continental margin, back-arc, felsic caldera province with diverse Zn-Pb-Ag-(Cu-Au) sulfide and Fe oxide deposits, Bergslagen region, Sweden: Economic Geology, 91, 979–1008.

Allmendinger, R. W., K. D. Nelson, C. J. Potter, M. Barazangi, L. D. Brown, and J. E. Oliver, 1987, Deep seismic reflection characteristics of the continental crust:Geology, 15, 304–310.

Almqvist, B. S. G., A. Björk, H. B. Mattsson, D. Hedlund, K. Gunnarsson, A. Malehmir, K. Högdahl, E. Bäckström, and P. Marsden, 2019, Magnetic characterisation of magnetite and hematite from the Blötberget apatite-iron-oxide deposits (Bergslagen), south-central Sweden: Canadian Journal of Earth Sciences.

Artimo, A., J. Mäkinen, R. C. Berg, C. C. Abert, and V.-P. Salonen, 2003, Three-dimensional geologic modeling and visualization of the Virttaankangas aquifer, southwestern Finland: Hydrogeology Journal, 11, 378–386.

Artimo, A., S. Saraperä, and I. Ylander, 2008, Methods for Integrating an Extensive Geodatabase with 3D Modeling and Data Management Tools for the Virttaankangas Artificial Recharge Project, Southwestern Finland: Water Resources Management, 22, 1723–1739.

Artimo, A., S. Saraperä, O. Puurunen, and J. Mäkinen, 2010, The Turku Region Artificial Infiltration Project, Finland – Tools for Enhanced Aquifer Characterization: 93–100.

Ayarza, P., C. Juhlin, D. Brown, M. Beckholmen, G. Kimbell, R. Pechnig, L. Pevzner, R. Pevzner, C. Ayala, M. Bliznetsov, A. Glushkov, and A. Rybalka, 2000,Integrated geological and geophysical studies in the SG4 borehole area, TagilVolcanic Arc, Middle Urals: Location of seismic reflectors and source of thereflectivity: Journal of Geophysical Research: Solid Earth, 105, 21333–21352.

Bachrach, R., and A. Nur, 1998, High‐resolution shallow‐seismic experiments in sand, Part I: Water table, fluid flow, and saturation: Geophysics, 63, 1225–1233.

Baker, G. S., 1999, Processing Near-Surface Seismic-Reflection Data: A Primer: Society of Exploration Geophysicists Tulsa.

Banerjee, I., and B. McDonald, 1975, Nature of Esker Sedimentation, Glaciofluvial and glaciolacustrine sedimentation.: Jopling AV & McDonald BC (eds).

Barton, N., R. Lien, and J. Lunde, 1974, Engineering classification of rock masses for the design of tunnel support: Rock Mechanics Felsmechanik Mecanique Des Roches, 6, 189–236.

Page 77: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

77

Barton, N., 2002, Some new Q-value correlations to assist in site characterisation and tunnel design: International Journal of Rock Mechanics and Mining Sciences, 39, 185–216.

Barton, N., 2006, Rock Quality, Seismic Velocity, Attenuation and Anisotropy. CRC Press. Available at https://www.crcpress.com/Rock-Quality-Seismic-Velocity-Attenuation-and-Anisotropy/Barton/p/book/9780415394413.

Bellefleur, G., A. Malehmir, and C. Müller, 2012, Elastic finite-difference modeling of volcanic-hosted massive sulfide deposits: A case study from Half Mile Lake, New Brunswick, Canada: Geophysics, 77, WC25–WC36.

Birch, F., 1939, The variation of seismic velocities within a simplified earth model, in accordance with the theory of finite strain: Bulletin of the Seismological Society of America, 29, 463–479.

Bohlen, T., C. Müller, and B. Milkereit, 2003, Elastic Seismic-Wave Scattering from Massive Sulfide Orebodies: On the Role of Composition and Shape, in Hardrock Seismic Exploration, Geophysical Developments Series, Society of Exploration Geophysicists, 70–89.

Bradford, J. H., D. S. Sawyer, C. A. Zelt, and J. S. Oldow, 1998, Imaging a shallow aquifer in temperate glacial sediments using seismic reflection profiling with DMO processing: Geophysics, 63, 1248–1256.

Bräunig, L., S. Buske, A. Malehmir, E. Bäckström, M. Schön, and P. Marsden, 2020, Seismic depth imaging of iron-oxide deposits and their host rocks in the Ludvika mining area of central Sweden: Geophysical Prospecting, 68, 24–43.

Brodic, B., A. Malehmir, C. Juhlin, L. Dynesius, M. Bastani, and H. Palm, 2015, Multicomponent broadband digital-based seismic landstreamer for near-surface applications: Journal of Applied Geophysics, 123, 227–241.

Brodic, B., A. Malehmir, and C. Juhlin, 2017, Delineating fracture zones using surface-tunnel-surface seismic data, P - S, and S - P mode conversions: Seismic Response of Fractures: Journal of Geophysical Research: Solid Earth, 122, 5493–5516.

Brodic, B., A. Malehmir, A. Pugin, and G. Maries, 2018, Three-component seismic land streamer study of an esker architecture through S- and surface-wave imaging: Geophysics, 83, B339–B353.

Buntin, S., A. Malehmir, H. Koyi, K. Högdahl, M. Malinowski, S. Å. Larsson, H. Thybo, C. Juhlin, A. Korja, and A. Górszczyk, 2019, Emplacement and 3D geometry of crustal-scale saucer-shaped intrusions in the Fennoscandian Shield: Scientific Reports, 9, 1–11.

Carcione, J. M., G. C. Herman, and A. P. E. ten Kroode, 2002, Seismic modeling: Geophysics, 67, 1304–1325.

Chadwick, R. A., and T. C. Pharaoh, 1998, The seismic reflection Moho beneath the United Kingdom and adjacent areas: Tectonophysics, 299, 255–279.

Cheraghi, S., A. Malehmir, G. Bellefleur, E. Bongajum, and M. Bastani, 2013, Scaling behavior and the effects of heterogeneity on shallow seismic imaging of mineral deposits: A case study from Brunswick No. 6 mining area, Canada: Journal of Applied Geophysics, 90, 1–18.

Cheraghi, S., M. Naghizadeh, D. Snyder, R. Haugaard, and T. Gemmell, 2020, High-resolution seismic imaging of crooked two-dimensional profiles in greenstone belts of the Canadian shield: results from the Swayze area, Ontario, Canada: Geophysical Prospecting, 68, 62–81.

Chowdhury, K. R., 2011, Deep Seismic Reflection and Refraction Profiling, in H. K. Gupta, ed., Encyclopedia of Solid Earth Geophysics, Springer Netherlands, 103–118.

Page 78: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

78

Dasgupta, R., and R. A. Clark, 1998, Estimation of Q from surface seismic reflection data: Geophysics, 63, 2120–2128.

Dasios, A., T. R. Astin, and C. McCann, 2001, Compressional-wave Q estimation from full-waveform sonic data: Geophysical Prospecting, 49, 353–373.

Deere, D. U., 1962, Technical Description of Rock Cores for Engineering Purposes: University of Illinois.

Dehghannejad, M., A. Malehmir, C. Juhlin, and P. Skyttä, 2012a, 3D constraints and finite-difference modeling of massive sulfide deposits: The Kristineberg seismic lines revisited, northern Sweden: Geophysics, 77, WC69–WC79.

Dehghannejad, M., T. E. Bauer, A. Malehmir, C. Juhlin, and P. Weihed, 2012b, Crustal geometry of the central Skellefte district, northern Sweden – constraints from reflection seismic investigations: Tectonophysics, 524–525, 87–99.

Durrheim, R. J., and M. J. Maccelari, 1991, Seismic exploration for precious metals in the hard rock environment: SEG Technical Program Expanded Abstracts 1991, 159–162.

Eaton, D. W., B. Milkereit, and M. H. Salisbury (eds.), 2003a, Hardrock Seismic Exploration: Society of Exploration Geophysicists.

Eaton, D. W., B. Milkereit, and M. Salisbury, 2003b, Seismic methods for deep mineral exploration: Mature technologies adapted to new targets: The Leading Edge, 22, 580–585.

Eissa, E. A., and A. Kazi, 1988, Relation between static and dynamic Young’s moduli of rocks: International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 25, 479–482.

Francese, R. G., Z. Hajnal, and A. Prugger, 2002, High‐resoluti on images of shallow aquifers—A challenge in near‐surface seismology: Geophysics, 67, 177–187.

Goetz, J. F., L. Dupal, and J. Bowler, 1979, An investigation into discrepancies between sonic log and seismic check spot velocities: The APPEA Journal, 19, 131–141.

Gorrell, G., and J. Shaw, 1991, Deposition in an esker, bead and fan complex, Lanark, Ontario, Canada: Sedimentary Geology, 72, 285–314.

Górszczyk, A., A. Adamczyk, and M. Malinowski, 2014, Application of curvelet denoising to 2D and 3D seismic data — Practical considerations: Journal of Applied Geophysics, 105, 78–94.

Haldar, S. K., 2013, Chapter 8 - Mineral Resource and Ore Reserve Estimation, in Mineral Exploration, Elsevier, 137–155.

Hall, R. B., 1988, Hardrock versus softrock geology, in General Geology, Springer US, 331–331.

Hamilton, E. L., 1972, Compressional‐wave attenuation in marine sediments: Geophysics, 37, 620–646.

Hauge, P. S., 1981, Measurements of attenuation from vertical seismic profiles: Geophysics, 46, 1548–1558.

Hearst, J. R., and P. H. Nelson, 1985, Well Logging for Physical Properties: McGraw-Hill.

Heinonen, S., P. J. Heikkinen, J. Kousa, I. T. Kukkonen, and D. B. Snyder, 2013, Enhancing hardrock seismic images: Reprocessing of high-resolution seismic reflection data from Vihanti, Finland: Journal of Applied Geophysics, 93, 1–11.

Hloušek, F., O. Hellwig, and S. Buske, 2015, Improved structural characterization of the Earth’s crust at the German Continental Deep Drilling Site using advanced seismic imaging techniques: Journal of Geophysical Research: Solid Earth, 120, 6943–6959.

Page 79: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

79

Hughes, S. W., 2005, Archimedes revisited: a faster, better, cheaper method of accurately measuring the volume of small objects: Physics Education, 40, 468–474.

International Resource Panel, 2020, Mineral Resource Governance in the 21st Century. A Report by the International Resource Panel. United Nations Environment Programme, Nairobi, Kenya.

Jonsson, E., Troll, V.R., Högdahl, K., Harris, C., Weis, F., Nilsson, K.P., Skelton, A., 2013, Magmatic origin of giant ‘Kiruna-type’ apatite-iron-oxide ores in Central Sweden: Scientific Reports 3.

Juhlin, C., 1995, Finite-difference elastic wave propagation in 2D heterogeneous transversely isotropic media1: Geophysical Prospecting, 43, 843–858.

Juhlin, C., and M. B. Stephens, 2006, Gently dipping fracture zones in Paleoproterozoic metagranite, Sweden: Evidence from reflection seismic and cored borehole data and implications for the disposal of nuclear waste: Journal of Geophysical Research: Solid Earth, 111.

Juhlin, C., H. Palm, C.-F. Müllern, and B. Wållberg, 2002, Imaging of groundwater resources in glacial deposits using high-resolution reflection seismics, Sweden: Journal of Applied Geophysics, 51, 107–120.

Juhlin, C., M. Dehghannejad, B. Lund, A. Malehmir, and G. Pratt, 2010, Reflection seismic imaging of the end-glacial Pärvie Fault system, northern Sweden: Journal of Applied Geophysics, 70, 307–316.

Kassam, A., B. Milkereit, and D. Shi, 2016, Analysis and Visualization of Dynamic Elastic Properties in 3D Space for High Density Minerals: Extended Abstract. Near Surface Geoscience.

Ketelhodt, J. K. von, M. S. D. Manzi, and R. J. Durrheim, 2019, Post-stack denoising of legacy reflection seismic data: implications for coalbed methane exploration, Kalahari Karoo Basin, Botswana: Exploration Geophysics, 50, 667–682.

King, M. S., 1983, Static and dynamic elastic properties of rocks from the Canadian shield: International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 20, 237–241.

King, M. S., 2009, Recent developments in seismic rock physics: International Journal of Rock Mechanics and Mining Sciences, 46, 1341–1348.

Knopoff, L., 1964, Q: Reviews of Geophysics, 2, 625. Koivisto, E., A. Malehmir, P. Heikkinen, S. Heinonen, and I. Kukkonen, 2012, 2D

reflection seismic investigations at the Kevitsa Ni-Cu-PGE deposit, northern Finland: GEOPHYSICS, 77, WC149–WC162.

Koivisto, E., A. Malehmir, N. Hellqvist, T. Voipio, and C. Wijns, 2015, Building a 3D model of lithological contacts and near-mine structures in the Kevitsa mining and exploration site, Northern Finland: constraints from 2D and 3D reflection seismic data: Geophysical Prospecting, 63, 754–773.

L’Heureux, E., B. Milkereit, and K. Vasudevan, 2009, Heterogeneity and seismic scattering in exploration environments: Tectonophysics, 472, 264–272.

Mäkinen, J., 2003a, Time‐transgressive deposits of repeated depositional sequences within interlobate glaciofluvial (esker) sediments in Köyliö, SW Finland: Sedimentology, 50, 327–360.

Mäkinen, J., 2003b, The Development of Depositional Environments Within the Interlobate Säkylänharju-Virttaankangas Glaciofluvial Complex in SW Finland: Suomalainen Tiedeakatemia.

Mäkinen, J., and M. Räsänen, 2003, Early Holocene regressive spit-platform and nearshore sedimentation on a glaciofluvial complex during the Yoldia Sea and the Ancylus Lake phases of the Baltic Basin, SW Finland: Sedimentary Geology, 158, 25–56.

Page 80: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

80

Magnusson, N.H., 1970. The origin of the iron ores in central Sweden and the history of their alterations: P.1, Sveriges geologiska undersökning, Serie C, Avhandlingar och uppsatser, Stockholm.

Malehmir, A., and G. Bellefleur, 2009, 3D seismic reflection imaging of volcanic-hosted massive sulfide deposits: Insights from reprocessing Halfmile Lake data, New Brunswick, Canada: Geophysics, 74, B209–B219.

Malehmir, A., R. Durrheim, G. Bellefleur, M. Urosevic, C. Juhlin, D. J. White, B. Milkereit, and G. Campbell, 2012, Seismic methods in mineral exploration and mine planning: A general overview of past and present case histories and a look into the future: Geophysics, 77, WC173–WC190.

Malehmir, A., M. Andersson, M. Lebedev, M. Urosevic, and V. Mikhaltsevitch, 2013a, Experimental estimation of velocities and anisotropy of a series of Swedish crystalline rocks and ores: Velocities and anisotropy of a series of crystalline rocks and ores: Geophysical Prospecting, 61, 153–167.

Malehmir, A., M. Bastani, C. M. Krawczyk, M. Gurk, N. Ismail, U. Polom, and L. Perss, 2013b, Geophysical assessment and geotechnical investigation of quick-clay landslides – a Swedish case study: Near Surface Geophysics, 11, 341–352.

Malehmir, A., E. Koivisto, M. Manzi, S. Cheraghi, R. J. Durrheim, G. Bellefleur, C. Wijns, K. A. A. Hein, and N. King, 2014, A review of reflection seismic investigations in three major metallogenic regions: The Kevitsa Ni–Cu–PGE district (Finland), Witwatersrand goldfields (South Africa), and the Bathurst Mining Camp (Canada): Ore Geology Reviews, 56, 423–441.

Malehmir, A., S. Wang, J. Lamminen, B. Brodic, M. Bastani, K. Vaittinen, C. Juhlin, and J. Place, 2015, Delineating structures controlling sandstone-hosted base-metal deposits using high-resolution multicomponent seismic and radio-magnetotelluric methods: a case study from Northern Sweden: Geophysical Prospecting, 63, 774–797.

Malehmir, A., G. Bellefleur, E. Koivista, and C. Juhlin, 2017a, Pros and cons of 2D vs 3D seismic mineral exploration surveys: First Break, 35, 49–55.

Malehmir, A., G. Maries, E. Bäckström, M. Schön, and P. Marsden, 2017b, Developing cost-effective seismic mineral exploration methods using a landstreamer and a drophammer: Scientific Reports, 7, 10325.

Malehmir, A., S. Heinonen, M. Dehghannejad, P. Heino, G. Maries, F. Karell, M. Suikkanen, and A. Salo, 2017c, Landstreamer seismics and physical property measurements in the Siilinjärvi open-pit apatite (phosphate) mine, central Finland: Geophysics, 82, B29–B48.

Malinowski, M., E. Schetselaar, and D. J. White, 2012, 3D seismic imaging of volcanogenic massive sulfide deposits in the Flin Flon mining camp, Canada: Part 2 — Forward modeling: Geophysics, 77, WC81–WC93.

Manzi, M. S. D., M. A. S. Gibson, K. A. A. Hein, N. King, and R. J. Durrheim, 2012, Application of 3D seismic techniques to evaluate ore resources in the West Wits Line goldfield and portions of the West Rand goldfield, South Africa: Geophysics, 77, WC163–WC171.

Manzi, M., and R. J. Durrheim, 2015, Reflection Seismic Applications in Coalbed Methane Exploration - Kalahari Karoo Basin, Botswana:

Manzi, M., G. Cooper, A. Malehmir, R. Durrheim, and Z. Nkosi, 2015, Integrated interpretation of 3D seismic data to enhance the detection of the gold-bearing reef: Mponeng Gold mine, Witwatersrand Basin (South Africa): Geophysical Prospecting, 63, 881–902.

Page 81: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

81

Manzi, M. S. D., G. R. J. Cooper, A. Malehmir, and R. J. Durrheim, 2020, Improved structural interpretation of legacy 3D seismic data from Karee platinum mine (South Africa) through the application of novel seismic attributes: Geophysical Prospecting, 68, 145–163.

Matthews, L., 2002, Base Metal Exploration: Looking Deeper and Adding Value with Seismic Data: CSEG Recorder Magazine, 37–43.

Mavko, G., T. Mukerji, and J. Dvorkin, 2009, The Rock Physics Handbook: Tools for Seismic Analysis of Porous Media: Cambridge University Press.

McKenzie, C. K., G. P. Stacey, and M. T. Gladwin, 1982, Ultrasonic characteristics of a rock mass: International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 19, 25–30.

Milkereit, B., and M. Wu, 2015, The Velocity – Density Relationship Revisited: Extended Abstract. At 3rd International Workshop on Rock Physics.

Milkereit, B., D. Eaton, J. Wu, G. Perron, M. H. Salisbury, E. K. Berrer, and G. Morrison, 1996, Seismic imaging of massive sulfide deposits; Part II, Reflection seismic profiling: Economic Geology, 91, 829–834.

Milkereit, B., E. K. Berrer, A. R. King, A. H. Watts, B. Roberts, A. Erick, D. W. Eaton, J. Wu, and M. H. Salisbury, 2000, Development of 3-D seismic exploration technology for deep nickel-copper deposits—A case history from the Sudbury basin, Canada 3-D Seismic Exploration Technology: Geophysics, 65, 1890–1899.

Miller, R. D., and D. W. Steeples, 1994, Applications of shallow high-resolution seismic reflection to various environmental problems: Journal of Applied Geophysics, 31, 65–72.

Miller, R. D., and J. Xia, 1998, Large near‐surface velocity gr adients on shallow seismic reflection data: Geophysics, 63, 1348–1356.

Nabighian, M. N., 1988, Electromagnetic Methods in Applied Geophysics: Volume 1, Theory: Society of Exploration Geophysicists.

Nafe, J. E., and C. L. Drake, 1961, Physical Properties of Marine Sediments: DTIC Document.

Nedimović, M. R., and G. F. West, 2003a, Crooked‐line 2D seismi c reflection imaging in crystalline terrains: Part 1, data processing: Geophysics, 68, 274–285.

Nedimović, M. R., and G. F. West, 2003b, Crooked‐line 2D seismi c reflection imaging in crystalline terrains: Part 2, migration: Geophysics, 68, 286–296.

Neelamani, R., A. I. Baumstein, D. G. Gillard, M. T. Hadidi, and W. L. Soroka, 2008, Coherent and random noise attenuation using the curvelet transform: The Leading Edge, 27, 240–248.

Nelson, R. G., 1984, Seismic reflection and mineral prospecting: Exploration Geophysics, 15, 229–250.

NIO Mineral Resource Estimate Report, 2017, https://nordicironore.se/wp-content/uploads/2017/07/mineral-resource-estimate-update-2017_final.pdf

Pretorius, C. C., W. F. Trewick, A. Fourie, and C. Irons, 2000, Application of 3-D seismics to mine planning at Vaal Reefs gold mine, number 10 shaft, Republic of South Africa: GEOPHYSICS, 65, 1862–1870.

Pretorius, C. C., M. R. Muller, M. Larroque, and C. Wilkins, 2003, 16. A Review of 16 Years of Hardrock Seismics on the Kaapvaal Craton, in Hardrock Seismic Exploration. Geophysical Developments Series Society of Exploration Geophysicists, 247–268.

Pugin, A., and S. E. Pullan, 2000, First-arrival alignment static corrections applied to shallow seismic reflection data 1: Journal of Environmental & Engineering Geophysics, 5, 7–15.

Page 82: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

82

Pugin, A. J., J. Mäkinen, E. Ahokangas, A. Artimo, H. Vanhala, A. Pasanen, and K. Moisio, 2014, High-resolution seismic reflection survey with landstreamer on the characterization of the Virttaankangas aquifer, SW Finland, in J. Virtasalo and M. Tuusjärvi, eds., Abstract book. 1st Finnish National Colloqvium of Geosciences, Espoo, 19–20 March 2014. Geological Survey of Finland, Guide 58., , 63–64.

Punkari, M., 1980, The ice lobes of the Scandinavian ice sheet during the deglaciation in Finland.: Boreas, 9, 307–310.

Punkari, M., 1997, Glacial and glaciofluvial deposits in the interlobate areas of the Scandinavian ice sheet: Quaternary Science Reviews, 16, 741–753.

Reed, K., 1993, Seismic reflection surveying for mining exploration applications, A review of practice past and current with an outlook for the future: Mineral Industry Technology Council of Canada, (unpublished report): .

Ripa, M., and L. Kübler, 2003, Apatite-Bearing Iron Ores in the Bergslagen Region of South-Central Sweden (M. Ripa, ed.,): Sveriges Geologiska Undersökning.

Rodriguez-Tablante, J., A. Tryggvason, A. Malehmir, C. Juhlin, and H. Palm, 2007, Cross-profile acquisition and cross-dip analysis for extracting 3D information from 2D surveys, a case study from the western Skellefte District, northern Sweden: Journal of Applied Geophysics, 63, 1–12.

Rossi, P. M., P. Ala-aho, J. Doherty, and B. Kløve, 2014, Impact of peatland drainage and restoration on esker groundwater resources: modeling future scenarios for management: Hydrogeology Journal, 22, 1131–1145.

Salas-Romero, S., A. Malehmir, I. Snowball, B. C. Lougheed, and M. Hellqvist, 2015, Identifying landslide preconditions in Swedish quick clays—insights from integration of surface geophysical, core sample- and downhole property measurements: Landslides.

Salisbury, M. H., C. W. Harvey, and L. Matthews, 2003, 1. The Acoustic Properties of Ores and Host Rocks in Hardrock Terranes, in D. W. Eaton, B. Milkereit, and M. H. Salisbury, eds., Hardrock Seismic Exploration, Society of Exploration Geophysicists, 9–19.

Salisbury, M. H., B. Milkereit, G. Ascough, R. Adair, L. Matthews, D. R. Schmitt, J. Mwenifumbo, D. W. Eaton, and J. Wu, 2000, Physical properties and seismic imaging of massive sulfides: Geophysics, 65, 1882–1889.

Sams, M. S., J. P. Neep, M. H. Worthington, and M. S. King, 1997, The measurement of velocity dispersion and frequency‐dependent intrinsic attenu ation in sedimentary rocks: Geophysics, 62, 1456–1464.

Schmitt, D. R., C. J. Mwenifumbo, K. A. Pflug, and I. L. Meglis, 2003, 2. Geophysical Logging for Elastic Properties in Hard Rock: A Tutorial, in D. W. Eaton, B. Milkereit, and M. H. Salisbury, eds., Hardrock Seismic Exploration, Society of Exploration Geophysicists, 20–41.

Serra, O., 2008, Well Logging Handbook: Editions OPHRYS. Sheriff, R. E., and L. P. Geldart, 1995, Exploration Seismology: Cambridge

University Press. Singh, S. C., A. J. Harding, G. M. Kent, M. C. Sinha, V. Combier, S. Bazin, C. H.

Tong, J. W. Pye, P. J. Barton, R. W. Hobbs, R. S. White, and J. A. Orcutt, 2006, Seismic reflection images of the Moho underlying melt sills at the East Pacific Rise: Nature, 442, 287–290.

Sjogren, B., A. Ofsthus, and J. Sandberg, 1979, Seismic classification of rock mass qualities: Geophysical Prospecting, 27, 409–442.

Statistics Finland, 2020, Retrieved from https://www.stat.fi/tup/suoluk/suoluk_vaesto_en.html

Page 83: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

83

Steeples, D. W., and R. D. Miller, 1988, Seismic reflection methods applied to engineering, environmental, and groundwater problems: Proc.(1st) Symp. on the Application of Geophysics to Engineering and Environmental Problems, Sot. Eng. and Mineral Exploration Geophysicists, Golden, CO, 409–461.

Stephens, M. B., M. Ripa, I. Lundström, L. Persson, T. Bergman, M. Ahl, C. H. Wahlgren, P. Persson, and L. Wickström, 2009, Synthesis of the Bedrock Geology in the Bergslagen Region, Fennoscandian Shield, South-Central Sweden: Sveriges geologiska undersökning (SGU).

Smart Exploration Available at https://smartexploration.eu Telford, W. M., W. M. Telford, L. P. Geldart, and R. E. Sheriff, 1990, Applied

Geophysics: Cambridge University Press. Tonn, R., 1991, The determination of the seismic quality factor Q from VSP data: a

comparison of different computational methods: Geophysical Prospecting, 39, 1–27. Tryggvason, A., S. T. Rögnvaldsson, and Ó. G. Flóvenz, 2002, Three-dimensional

imaging of the P- and S-wave velocity structure and earthquake locations beneath Southwest Iceland: Geophysical Journal International, 151, 848–866.

TRUST Available at http://www.trust-geoinfra.se/ Ulusoy, İ., T. Dahlin, and B. Bergman, 2015, Time-lapse electrical resistivity

tomography of a water infiltration test on Johannishus Esker, Sweden: Hydrogeology Journal, 23, 551–566.

UN Environment, 2017, Assessing global resource use: A systems approach to resource efficiency and pollution reduction. A Report of the International Resource Panel. United Nations Environment Programme. Nairobi, Kenya. UNEP - UN Environment Programme. Available at http://www.unenvironment.org/news-and-stories/press-release/resource-use-expected-double-2050-better-natural-resource-use. Accessed April 10, 2020.

UN Environment, 2020, How minerals and metals companies can help achieve 2030 Agenda for Sustainable Development. UN Environ. Available at http://www.unenvironment.org/news-and-stories/story/how-minerals-and-metals-companies-can-help-achieve-2030-agenda-sustainable. Accessed April 7, 2020.

Urosevic, M., and B. J. Evans, 1998, Seismic methods for the detection of kimberlite pipes: Exploration Geophysics, 29, 632–635.

Urosevic, M., G. Bhat, and M. H. Grochau, 2012, Targeting nickel sulfide deposits from 3D seismicreflection data at Kambalda, AustraliaTargeting Ni deposits from 3D seismic: Geophysics, 77, WC123–WC132.

Urosevic, M., E. Stolz, A. Kepic, and C. Juhlin, 2007, Seismic exploration of ore deposits in Western Australia: Exploration in the New Millennium - Volume 1, 525–534.

Valjus, T., 2006, Turun Seudun Vesi.Pohjavesialueen kallionpinnan tason määritys painovoimamittausten avulla. 41 pp. Geologian tutkimuskeskus, Tutkimusraportti 13.01.2006.show2.pdf: Geological Survey of Finland.

World Resources Forum, 2020, Facts and Figures: https://www.wrforum.org/publications-2/publications/ Yavuz, S., J. Kinkela, A. Dzunic, M. Penney, R. Neto, V. Araújo, S. Ziramov, R.

Pevzner, and M. Urosevic, 2015, Physical property analysis and preserved relative amplitude processed seismic imaging of volcanogenic massive sulfides-a case study from Neves-Corvo, Portugal: Analysis and seismic imaging of VMS: Geophysical Prospecting, 63, 798–812.

Yehuwalashet, E., and A. Malehmir, 2018, Gravity and magnetic survey, modeling and interpretation in the Blötberget iron-oxide mining area of central Sweden, in SEG Technical Program Expanded Abstracts 2018, SEG Technical Program Expanded AbstractsSociety of Exploration Geophysicists, 1479–1483.

Page 84: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

84

Yilmaz, Ö., 2001, Seismic Data Analysis: Society of Exploration Geophysicists. Zaleski, E., D. W. Eaton, B. Milkereit, B. Roberts, M. Salisbury, and L. Petrie, 1997,

Seismic reflections from subvertical diabase dikes in an Archean terrane: Geology, 25, 707–710.

Page 85: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015
Page 86: Seismic investigations and physical property …uu.diva-portal.org/smash/get/diva2:1423815/FULLTEXT02.pdfPaper III: I participated in data acquisition, in both field campaigns, 2015

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 1936

Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science andTechnology, Uppsala University, is usually a summary of anumber of papers. A few copies of the complete dissertationare kept at major Swedish research libraries, while thesummary alone is distributed internationally throughthe series Digital Comprehensive Summaries of UppsalaDissertations from the Faculty of Science and Technology.(Prior to January, 2005, the series was published under thetitle “Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-408712

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020