The Concept of Lineaments in Geological Structural Analysis; Principles and Methods: A Review Based on Examples from Norway

: Application of lineament analysis in structural geology gained renewed interest when remote sensing data and technology became available through dedicated Earth observation satellites like Landsat in 1972. Lineament data have since been widely used in general structural investigations and resource and geohazard studies. The present contribution argues that lineament analysis remains a useful tool in structural geology research both at the regional and local scales. However, the traditional “lineament study” is only one of several methods. It is argued here that structural and lineament remote sensing studies can be separated into four distinct strategies or approaches. The general analyzing approach includes general structural analysis and identification of foliation patterns and composite structural units (mega-units). The general approach is routinely used by most geologists in preparation for field work, and it is argued that at least parts of this should be performed manually by staff who will participate in the field activity. We argue that this approach should be a cyclic process so that the lineament database is continuously revised by the integration of data acquired by field data and supplementary data sets, like geophysical geochronological data. To ensure that general geological (field) knowledge is not neglected, it is our experience that at least a part of this type of analysis should be performed manually. The statistical approach conforms with what most geologists would regard as “lineament analysis” and is based on statistical scrutiny of the available lineament data with the aim of identifying zones of an enhanced (or subdued) lineament density. It would commonly predict the general geometric characteristics and classification of individual lineaments or groups of lineaments. Due to efficiency, capacity, consistency of interpretation methods, interpretation and statistical handling, this interpretative approach may most conveniently be performed through the use of automatized methods, namely by applying algorithms for pattern recognition and machine learning. The focused and dynamic approaches focus on specified lineaments or faults and commonly include a full structural geological analysis and data acquired from field work. It is emphasized that geophysical (potential field) data should be utilized in lineament analysis wherever available in all approaches. Furthermore, great care should be taken in the construction of the database, which should be tailored for this kind of study. The database should have a 3D or even 4D capacity and be object-oriented and designed to absorb different (and even unforeseen) data types on all scales. It should also be designed to interface with shifting modeling tools and other databases. Studies of the Norwegian mainland have utilized most of these strategies in lineament studies on different scales. It is concluded that lineament studies have revealed fracture and fault systems and the geometric relations between them, which would have remained unknown without application of remote sensing data and lineament analysis.


The Concept of Lineaments
The concept of lineament analysis [1,2] has a wide application in the description and analysis of fault and fracture systems through the utilization of remote sensing data on global, regional and local scales (e.g., [3][4][5][6][7][8][9][10][11][12]).The term "lineament" was introduced by Hobbs [13] to characterize the spatial relationships of linear morphological features that included "(1) crests of ridges or boundaries of elevated areas (2) drainage lines, (3) coastlines, (4) boundary lines of formations and (5) petrographic rock types, or of lines of outcrops" [13] (p. 485).Hobbs added "ravines and valleys" to the definition in his later works [14].Utilizing morphological patterns from south Norway previously identified by Kjerulf [15], Hobbs [14] (p.227) emphasized the tectonic aspect of many lineaments and particularly stated that lineaments commonly are "lines of fracture zones and zones of fault breccia".He pointed out that "many lineaments are identical with seismotectonic lines of greatest danger from earthquake shock".The term "lineament" was less frequently applied in the geological literature for some time but was rejuvenated in connection with a new generation of satellite-generated remote sensing data becoming available in the 1970s, and it was the explicit topic in a series of dedicated conferences on "Basement Tectonics" in the period from 1980 to 2000, resulting in a series of conference proceedings (e.g., [16][17][18][19]).
In the recent literature, the term lineament is mainly used for geological features of tectonic origin.Thus, O'Leary et al. [1] used the term for "a linear or curvilinear feature, which is identified by remote sensing methods, and which is believed to represent the trace of intersection between a planar or subplanar structural inhomogeneity (such as a fault) and the surface of the Earth" (italics inserted by us).Thus, it follows from O'Leary's definition and the current use of the term that lineaments can be identified by photometric methods in topographic, bathymetrical, spectral digital and potential field data (see also [20][21][22]).A lineament data set would commonly include dyke and vein systems, since these commonly display linear structures that are detectable in optical, multispectral and potential field data sets (e.g., [23][24][25]).
Gabrielsen and Braathen [7] expanded on the definition of lineaments, using the term fracture lineament for those lineaments that can be demonstrated to coincide with a zone of a markedly enhanced fracture frequency and hence likely represent a stress-induced zone of weakness in the bedrock, such as fracture corridors [26][27][28][29] and faults.Gabrielsen and Braathen [7] emphasized that when the true nature of a lineament is established, the term should reflect the level of knowledge whenever such information is available, thus naming it, for example, a "fracture lineament", "fracture corridor", "fault" or "dyke (swarm)" rather than a "lineament".
The aim of the present contribution is to demonstrate that lineament analysis is not an obsolete analytical concept in structural geological analysis.It is emphasized that modern lineament analysis has an escalating number of data types (e.g., radiation observation bands) and computational methods at its disposal.It remains crucial that such analyses are integrated with state-of-the art field-based geological investigations (the "ground truth").

Current Remote Sensing Technology in Geological and Structural Analysis
Available remote sensing data and methods dedicated to geological study were revolutionized by satellite platforms becoming available for geological monitoring in the 1970s.Before that, lineament analysis was centered solely around areal photographic imagery recorded in the visible spectrum as well as morphological maps, perhaps supported by potential field data acquired by aircraft or ground campaigns.Although some areas were well covered by such data, standardized data were not universally available.Furthermore, processing and interpretation methods were constrained by available financial possibilities and technology.
The age of satellite technology promoted the development of remote sensing equipment which was dedicated to earth analysis, and such data soon became universally available (e.g., [2,30]).However, the data were of uneven quality in the incipient time period, and some of the available observation systems were designed for meteorological purposes rather than solid Earth observations (e.g., TIROS, NOAA, GEOs and Datta [31]).Much methodological and analytical technology dedicated to Earth observation were affiliated with remote sensing systems designed for application in the Landsat satellite series and the space shuttle, although several operators soon developed dedicated supplementary Earth observatory technology.The first Landsat satellite generation (ERTS-1, launched in 1972 and later renamed Landsat 1) utilized a multispectral scanner (MSS) which had four spectral bands between 0.5 and 1.1 µm.The MSS was supplied by and later replaced by the Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (TM+), which were in use from Landsat 4 to Landsat 7, and finally the Operational Land Imager (OLI).The Thermal Infrared Sensor (TIRS) operated in up to 11 bands in wavelengths between 0.43 and 12.51 µm on Landsat 8.Although allowing for observations and analysis of foliage or herbal significance, geological observations remained a dedicated target, with MSS and TM bands between 0.5 and 1.1 µm as the most common data type utilized [32].Current remote sensing systems commonly include both reflected and emitted radiation [33].In particular, synthetic-aperture radar (SAR) offers higher-resolution images.Such data also can be used to document surface motion and change, and several SAR-based Earth observation systems such as NASA-ISRO Synthetic Aperture Radar (NISAP) are presently in different stages of construction.Current geological studies frequently include systematic use of potential field data [20,[34][35][36][37].Furthermore, remote sensing includes dedicated methods for analyzing specific problems for recording and optimizing data, such as the processing, filtering and precision computation methods at its disposal (e.g., [38,39]).

Different Approaches and Strategies in Structural Remote Sensing
Lineament studies in support of structural geological analysis are in frequent use in academic studies, in risk analysis (e.g., earthquake hazards and in the assessment of environmental change) as well as in resource assessment (e.g., water and mineral resources).However, depending on the focus of each investigation, optical remote sensing methods may provide the most relevant data sets in general structural analysis.Passive optical remote sensing data are, however, commonly supplied by emittance remote sensing data sets, particularly in the study of lithologic alteration products and mineralization affiliated with, for example, faulting and weathering [11,[40][41][42][43].
We suggest that for practical purposes, diverse approaches can be subdivided into four categories: general, statistical, focused and dynamic.The general approach aims at collecting data that describe the general geological structural picture of the study area.Such data sets are particularly valuable in areas where structural data are sparce due to limited previous field studies.The statistical approach focuses more specifically on the identification and distribution of lineament and fault populations, whereas the focused and dynamic approaches aim to describe and analyze distinct lineaments or faults.
In the following, we describe each of the approaches and give examples of each study type using examples from Norway which we have first-hand experience with regarding the use of remote sensing data and field study.

General Study
The Norwegian mainland is predominantly composed of a Proterozoic basement that is overlain by a complex pile of nappes of Caledonian age, which include lithologies from Proterozoic to Palaeozoic origin.These rocks are crisscrossed by fracture systems of contrasting deformation styles, ages and origins (Figure 1).Lineament studies are of great value for analysis of the interplay between these structural types.Several generations of lineament studies have been performed in Norway, using the traditional statistical approach from which zones of high lineament densities were defined (Figure 2).This approach coincides with most people regarding "lineament analysis", which is described in the section below.We note, however, that traditional lineament analysis subdues important structural information such that the general structural analysis should be expanded upon more often to include additional structural information like foliation, low-angle discontinuities (e.g., thrust faults) and areas of structural contrasts affiliated with, for example, tectonostratigraphic mega-units like magmatic bodies and nappes (see also [11,[44][45][46][47]).We regard foliation mapping as giving particularly important information that provides structural information which goes beyond that obtained from traditional lineament analysis [48].Interpretations obtained through general analysis can be enhanced by the inclusion of data obtained from other sources, such as emittance remote sensing methods, analysis is performed by the active field geologist.Because the general interpretation method commonly is performed as preparation for a field campaign, it is favorable that the geologist(s) who perform the fieldwork can also retrieve a general geological overview by performing the analysis manually.The results of this analysis would, together with the state-of-the-art geological maps, provide the basis for the next generation of field investigations (see below).Examples of general structural analysis based on remote sensing data are numerous (e.g., [25,36,47,50]).One early example combining lineament mapping with foliation mapping was performed on Caledonian tectonostratigraphic units in the Norwegian Caledonides of Nordland in northern Norway [48].The nappe pile of the Scandinavian Caledonides is traditionally subdivided into the lower, middle, upper and uppermost allochthons [51,52].Corfu et al. [53] recently challenged this subdivision by introducing along-strike segmentation.In this scheme, the units referred to here belong to the Seve-Köli Nappe Complex of the central segment of the Caledonides (Figure 3).The general remote sensing study was performed manually by interpreters who had first-hand field experience with the study area, with a focus on the tectonostratigraphic architecture and fault pattern.The remote sensing data were interactively used in the correlations with the Faults in these lineament zones seem to have young (Mesozoic or younger) structural overprints.Young reactivation is less conspicuous for the bulk of the faults that define the Østfold Zone (blue).From [8].
Interpretations obtained through general analysis can be enhanced by the inclusion of data obtained from other sources, such as emittance remote sensing methods, information on lithology, mineral distribution and secondary effects like weathering, which can sometimes be extracted [49].Potential field data would be of great support in general studies in that they may constrain the distribution and geometry of lithological units.
It is therefore generally appropriate that (at least parts of) general remote sensing analysis is performed by the active field geologist.Because the general interpretation method commonly is performed as preparation for a field campaign, it is favorable that the geologist(s) who perform the fieldwork can also retrieve a general geological overview by performing the analysis manually.The results of this analysis would, together with the state-of-the-art geological maps, provide the basis for the next generation of field investigations (see below).
Examples of general structural analysis based on remote sensing data are numerous (e.g., [25,36,47,50]).One early example combining lineament mapping with foliation mapping was performed on Caledonian tectonostratigraphic units in the Norwegian Caledonides of Nordland in northern Norway [48].The nappe pile of the Scandinavian Caledonides is traditionally subdivided into the lower, middle, upper and uppermost allochthons [51,52].Corfu et al. [53] recently challenged this subdivision by introducing along-strike segmentation.In this scheme, the units referred to here belong to the Seve-Köli Nappe Complex of the central segment of the Caledonides (Figure 3).The general remote sensing study was performed manually by interpreters who had first-hand field experience with the study area, with a focus on the tectonostratigraphic architecture and fault pattern.The remote sensing data were interactively used in the correlations with the previously established regional geology, detailed interpretation of nappe units and nappe geometry, including the intrinsic nappe architecture and fault interpretation.The foliation maps and the structural contrast between the nappe units were particularly useful in the subdivision of nappe units in the structural analysis of the area.Supporting geochronological data have helped to restore the margin of the Iapetus Ocean and provided background for dating and restoration of its magmatism [54].
geometry, including the intrinsic nappe architecture and fault interpretation.The foliation maps and the structural contrast between the nappe units were particularly useful in the subdivision of nappe units in the structural analysis of the area.Supporting geochronological data have helped to restore the margin of the Iapetus Ocean and provided background for dating and restoration of its magmatism [54].
A similar approach was utilized in the study of the upper nappe units of the Caledonides in southwestern Norway (Figure 4).Here, the major nappe units were well established by previous studies (e.g., [55,56]), but through the use of detailed remote sensing data (Landsat), several subordinate nappe slivers of contrasting deformational styles and their interaction with younger fault systems were identified.The adjusted structural architecture could thereafter be utilized in geochronological studies, and a more detailed chronostructural development for this part of the Norwegian Caledonides could be established [57].A similar approach was utilized in the study of the upper nappe units of the Caledonides in southwestern Norway (Figure 4).Here, the major nappe units were well established by previous studies (e.g., [55,56]), but through the use of detailed remote sensing data (Landsat), several subordinate nappe slivers of contrasting deformational styles and their interaction with younger fault systems were identified.The adjusted structural architecture could thereafter be utilized in geochronological studies, and a more detailed chronostructural development for this part of the Norwegian Caledonides could be established [57].

Statistical Lineament Study
The statistical approach to lineament analysis usually aims at establishing the geometrical relations between and the spatial distribution of fault and fracture systems.This is probably the most common approach in the study of fracture lineaments.Traditional lineament studies are aimed at detecting structures that are believed to represent the traces of intersections between a planar or subplanar structural inhomogeneity, such as faults or dykes, and commonly include classification of the lineament type (topographic, fracture, etc.), lineament distribution (spatial distribution and density), orientation, geometrical relation and lineament architecture (e.g., segmentation).Such data sets provide the background for identification and classification of lineament zones, sets, systems and complexes [58][59][60][61][62] and are valuable in defining and assessing the tectonic interaction between ancient and current (active) fracture systems [8,63,64].Numerous studies of this type have been reported in the former and current literature [3,4,39,[58][59][60].Such studies are frequently reported in the Basement Tectonics conference series (e.g., [16][17][18][19]).
Statistical mapping of lineaments may cover large land areas with thousands of lineaments at different scales and perhaps of contrasting morphologies and architectures.Manual statistical lineament mapping can therefore be quite time-consuming and demand long-term efforts on the scale of months and years [65,66].Unfortunately, manual structural and lineament mapping often cannot be fully reproduced because it relies on the subjective skills and experience of the interpreter.This can result in errors, such as failure to identify man-made infrastructure.In addition, it may also be influenced by the use of disparate interpretation techniques, such as induced oblique illumination of varying angles.Automatized (computer-generated) interpretation methods are therefore

Statistical Lineament Study
The statistical approach to lineament analysis usually aims at establishing the geometrical relations between and the spatial distribution of fault and fracture systems.This is probably the most common approach in the study of fracture lineaments.Traditional lineament studies are aimed at detecting structures that are believed to represent the traces of intersections between a planar or subplanar structural inhomogeneity, such as faults or dykes, and commonly include classification of the lineament type (topographic, fracture, etc.), lineament distribution (spatial distribution and density), orientation, geometrical relation and lineament architecture (e.g., segmentation).Such data sets provide the background for identification and classification of lineament zones, sets, systems and complexes [58][59][60][61][62] and are valuable in defining and assessing the tectonic interaction between ancient and current (active) fracture systems [8,63,64].Numerous studies of this type have been reported in the former and current literature [3,4,39,[58][59][60].Such studies are frequently reported in the Basement Tectonics conference series (e.g., [16][17][18][19]).
Statistical mapping of lineaments may cover large land areas with thousands of lineaments at different scales and perhaps of contrasting morphologies and architectures.Manual statistical lineament mapping can therefore be quite time-consuming and demand long-term efforts on the scale of months and years [65,66].Unfortunately, manual structural and lineament mapping often cannot be fully reproduced because it relies on the subjective skills and experience of the interpreter.This can result in errors, such as failure to identify man-made infrastructure.In addition, it may also be influenced by the use of disparate interpretation techniques, such as induced oblique illumination of varying angles.Automatized (computer-generated) interpretation methods are therefore preferential for statistical lineament analysis.In most cases, this reduces the interpretation time, secures the use of stable and reproduceable interpretation criteria and facilitates statistical handling of the resulting data.Digitized lineament databases generated by algorithms that include pattern recognition and machine learning methods are optimal for statistical lineament mapping in facilitating statistic automatic imaging and contouring of structural data [67][68][69][70].Since such methods are based on digital elevation models (DEMs), the result reflects the spatial resolution of the primary data set and the quality of available complimentary and more sophisticated data sets such as ASTER-DEM (digital surface model) and LIDAR-DEM5 (true digital terrain model).In combination with various filtering techniques, such combined data sets could be utilized to obtain optimal results.The total procedure would however require comprehensive tailored testing for each case [10].
Whenever feasible, statistical topographical lineament analysis based on photometric data should be supported by potential field data (gravity and magnetics) for interpretation of the regional structural significance of specific lineaments and lineament populations [36,37,71,72].
Numerous examples of studies utilizing the approach of statistical lineament study are available in the literature.For example, the Norwegian mainland has been subject to repeated mapping of lineament complexes and systems, utilizing continuously improving technology and data through manual and computerized techniques [8,12,13,66,67,73].This effort has led to the identification of zones of high lineament intensities, probably representing regional zones of mechanical weakness and localized deformation (Figure 2).This basic database has been expanded to encompass more detailed investigations of areas of particular significance and interest (e.g., [6,60]).These studies have helped to identify the complexity of zones of particularly high lineament or fault densities in the basement, which consists mainly of Proterozoic and early Palaeozoic rocks, many of which were deformed during the Caledonian Orogeny.Some of the zones of high lineament frequencies parallel the coastlines of southern Norway (Figure 2), whereas some define the continuation of the younger structural mega-structures like the Permian Oslo Graben (Figures 1 and 2).Field studies (e.g., [7][8][9]74,75]) suggest that several of the fault systems may have roots in the deep crust and that they have been repeatedly reactivated.Several lineament populations were defined already at an early stage of manual lineament mapping and simple statistical handling of distribution patterns (e.g., by use of rose diagram-facilitated correlation to established fault systems, from which geochronological data may be available).Detailed analysis in southern Norway [6,60] confirmed that regional lineament zones like the N-S-striking Bergen Zone, which is paralleled by the offshore Øygarden Fault Complex (Figure 1 [66,67]) interfere with other regional zones of contrasting deformational styles and geometries [9,[75][76][77][78] in a way that suggests some of the deeply rooted fault systems may be hard-linked.Several fault systems have been reactivated later in the geological history of Norway than previously anticipated, to the point that some of the fault systems have been partially active in concert with Mesozoic and Cenozoic fault activity in the Norwegian continental shelf [77].Some of these zones are sites for current moderate seismic activity [9,64,73,79].For example, the Møre-Trøndelag Fault Complex (Figures 1 and 2), which coincides with the ENE-WSW coastline of central Norway, probably has a Caledonian origin but has been repeatedly reactivated [63,80,81] and has contributed to the present morphology of southern Norway by defining a staircase-like fault system of rotated fault blocks [76,82].The lineament populations define varying cross-cutting relations, and the observed intrinsic geometries of individual lineaments and relations between lineament systems are strongly scale-dependent.The data suggest that a scale-invariant lineament network is composed of lineament sets which individually exhibit a scale-dependent hierarchical distribution and represent variable structural settings and time relations (e.g., [69]) (Figure 5).Note that the scales for the regional maps (a-c) and the inset (b1) are not the same.
The regional lineament maps of southern Norway provide the background for specialized analysis of focused lineaments and faults [7,9].

Focused Lineament Study
Focused lineament studies explore and describe individual lineaments and particularly focus on defining the lineament type (fracture zone or fault), architecture (segmentation, width, and splay configuration), deformation style and orientation.Such studies are commonly based on MSS and TM imagery processed into false-color images (e.g., [9,79]).The results from focused lineament studies are therefore supplementary to data acquired in the statistical approach.For focused lineament studies, the complete database would commonly include high-resolution digital topographic data and data acquired from field work.In many cases, the interpreter would consult and update the interpretation when additional field data are acquired.The interpreter would generally identify and compare lineament segments, secondary fracture systems like fault segmentation, secondary structures like Riedel shears, splays, ramps and horse-tail structures with the intention of determining under what stress system the lineament was generated and active.This requires identification of geometrically related (e.g., conjugate) fracture systems or fracture systems related to the same fracture generation (e.g., characterized by the same type or generation of fracture fill).Furthermore, lineaments or lineament segments of a particular color signature and configuration (width, dip relations, etc.) may help in the identification of related lineament sets (e.g., [9,83,84]).This should be combined with the application of robust models for fracture generation, like Andersonian principles or established fracture geometries affiliated with strike-slip faults.Additionally, a focused approach to structural and lineament study could greatly benefit from the inclusion of potential field data (magnetics and gravity) and correlation with The regional lineament maps of southern Norway provide the background for specialized analysis of focused lineaments and faults [7,9].

Focused Lineament Study
Focused lineament studies explore and describe individual lineaments and particularly focus on defining the lineament type (fracture zone or fault), architecture (segmentation, width, and splay configuration), deformation style and orientation.Such studies are commonly based on MSS and TM imagery processed into false-color images (e.g., [9,79]).The results from focused lineament studies are therefore supplementary to data acquired in the statistical approach.For focused lineament studies, the complete database would commonly include high-resolution digital topographic data and data acquired from field work.In many cases, the interpreter would consult and update the interpretation when additional field data are acquired.The interpreter would generally identify and compare lineament segments, secondary fracture systems like fault segmentation, secondary structures like Riedel shears, splays, ramps and horse-tail structures with the intention of determining under what stress system the lineament was generated and active.This requires identification of geometrically related (e.g., conjugate) fracture systems or fracture systems related to the same fracture generation (e.g., characterized by the same type or generation of fracture fill).Furthermore, lineaments or lineament segments of a particular color signature and configuration (width, dip relations, etc.) may help in the identification of related lineament sets (e.g., [9,83,84]).This should be combined with the application of robust models for fracture generation, like Andersonian principles or established fracture geometries affiliated with strike-slip faults.Additionally, a focused approach to structural and lineament study could greatly benefit from the inclusion of potential field data (magnetics and gravity) and correlation with seismological information.Particularly potential field data are crucial for determining the significance of structures mapped on the scale of the lithosphere (e.g., [20][21][22]74]).When feasible, traditional field data should be supplied with additional physical measurements by specialized equipment (e.g., geo-radar, hand-held XRF or porosity and magnetic susceptibility meters).
Several examples of focused lineament studies exist in studies of southern Norway.One focused lineament study was performed on one object within the framework of the regional system of the ENE-WSW-striking Agder-Telemark Lineament Zone [8,67], which is delineated to the east by the Skien-Porsgrunn Shear Zone (Figure 1).The Lista-Drangedal lineament is one of the most extensive elements within this lineament intensity zone, and a focused study utilizing Landsat imagery, aerial photography and potential field data and extensive field investigation was performed.It was demonstrated that the lineament includes at least seven distinct segments of contrasting geometries, architectures (intrinsic structure) and dip and deformation styles (Figure 6).General field work was performed along the entire structure, which was supplied by detailed field investigations in key areas [9], confirming its status as a fault of regional significance and justifying its upgrade from "the Lista-Drangedal lineament" to "the Lista-Drangedal fault".Field studies documented that the different segments contain lenses or layers of fault rocks up to 5-10 m thick that include mylonite, cataclasite, pseudotachylite and fault gouge.Thus, the study of the Lista-Drangedal Fault concluded that it is perhaps the principal element of the ENE-WSW-striking Agder-Telemark lineament intensity zone [8].Its field characteristics show that it has a history that probably extends from the late Proterozoic period (mylonitization) to the latest event dated to the Triassic period (cataclasis).It is also possibly affiliated with recent earthquake activity [9].However, it remains unclear as to whether the pattern of lineament and fault intensity zones are hard-linked and whether the system is associated or not with the Mesozoic-Cenozoic uplift of southern Norway, as can be indicated by the history of the Møre-Trøndelag Fault Zone in central Norway [76,82].
Geomatics 2024, 4, FOR PEER REVIEW seismological information.Particularly potential field data are crucial for determining t significance of structures mapped on the scale of the lithosphere (e.g., [20][21][22]74]).Wh feasible, traditional field data should be supplied with additional physical measuremen by specialized equipment (e.g., geo-radar, hand-held XRF or porosity and magne susceptibility meters).
Several examples of focused lineament studies exist in studies of southern Norwa One focused lineament study was performed on one object within the framework of t regional system of the ENE-WSW-striking Agder-Telemark Lineament Zone [8,67], whi is delineated to the east by the Skien-Porsgrunn Shear Zone (Figure 1).The Lista-Dranged lineament is one of the most extensive elements within this lineament intensity zone, and focused study utilizing Landsat imagery, aerial photography and potential field data an extensive field investigation was performed.It was demonstrated that the lineame includes at least seven distinct segments of contrasting geometries, architectures (intrins structure) and dip and deformation styles (Figure 6).General field work was perform along the entire structure, which was supplied by detailed field investigations in key are [9], confirming its status as a fault of regional significance and justifying its upgrade fro "the Lista-Drangedal lineament" to "the Lista-Drangedal fault".Field studies document that the different segments contain lenses or layers of fault rocks up to 5-10 m thick th include mylonite, cataclasite, pseudotachylite and fault gouge.Thus, the study of the List Drangedal Fault concluded that it is perhaps the principal element of the ENE-WSW striking Agder-Telemark lineament intensity zone [8].Its field characteristics show that has a history that probably extends from the late Proterozoic period (mylonitization) to t latest event dated to the Triassic period (cataclasis).It is also possibly affiliated with rece earthquake activity [9].However, it remains unclear as to whether the pattern of lineame and fault intensity zones are hard-linked and whether the system is associated or not wi the Mesozoic-Cenozoic uplift of southern Norway, as can be indicated by the history of t Møre-Trøndelag Fault Zone in central Norway [76,82].

Figure 6. (a)
The ENE-WSW-striking Lista-Drangedal Fault (yellow arrows) in southern Norway, as represented in the visual spectrum (Google Pro).Seven segments of contrasting geometries and contrasting field characteristics can be identified.This fault is a dominant element in the Agder-Telemark Lineament Zone and parallels the Skien-Porsgrunn Shear Zine (Figure 1).(b,c) Segments 1 and 2 are particularly expressed as topo-lineaments.The expression of the Lista-Drangedal Fault as (d) digital topographic, (e) aeromagnetic and (f) gravimetric data.Note the position of a recent seismic event (red spot) in inset of (d).All elements in this figure were compiled and redrafted from [9].
Another study that combined focused lineament analysis with a detailed field study and geochronology (Ar/Ar and K-Ar) in southern Norway was performed for the WNW-ESE-striking Himdalen Fault [75].In addition, the Himdalen Fault constitutes one segment of a greater structure, namely the Ørje Shear Zone [84], which is one of several major shear zones of the late Proterozoic period in southern Scandinavia (e.g., [85][86][87][88][89]).The core of the Himdalen Fault displays a series of plastic-to-brittle fault rocks (blastomylonite, cataclasite and fault gouge), which reveal a thermo-tectonic history extending from the late Proterozoic period (c.908 Ma) to the late Triassic period (c.200 Ma [75]).We cannot, however, rule out the possibility that the youngest age of the fault gouge reflects a weathering age.
Although these studies offer typical examples of focused studies of seismically extinct structures, they also contain aspects that characterize dynamic fault studies.

Dynamic Lineament Analysis
Dynamic analysis demands that time series describing the incremental structural development of the structure in question are available.Dynamic fault and lineament analysis can be subdivided into two groups, namely those of geohistorical significance and those of contemporaneous significance.
For geohistorical studies of faults, these require palinspastic knowledge of the regional configuration and development, fault architecture, deformation products (fault rocks) and geochronological information.Such data are mainly collected from detailed field studies that would follow general statistical or focused lineament studies and would be separated from such studies by a period of dedicated field work that includes careful sampling.Samples taken during the field study would be subject to painstaking laboratory work, not least of which would be the identification of different faultrock types and which would preferentially be followed by mineralogical and geochronological studies when feasible.Such work is commonly complex and time-consuming (e.g., [75]).In areas affected by relatively recent fault activity, topography and morphological studies may be beneficial in establishing the full fault story (e.g., [76,82,90]).
Examples of dynamic analysis of paleostress for distinct faults has become more frequent in Norway, parallel with the development of methods like apatite fission track (AFT) analysis and techniques for the geochronological study of faults [91,92] The AFT method is affiliated with relatively high statistical uncertainties (e.g., [93]) and sometimes a modest resolution, but it is still of great help when tectonic units like hanging walls and footwalls of established faults are well exposed and where vertical sections can be sampled systematically (e.g., [76,78,82,94]), as well as in cases where the geometry and offset of dateable strata across the fault can be determined.
Examples of contemporaneous dynamic lineament and fault studies are scarce in Norway due to modest neotectonic (post-glacial) activity.Norway is characterized by modest seismic activity and micro-earthquakes (M = 1-4) that occur regularly (e.g., [64,[95][96][97]), sometimes interrupted by more significant events.For example, the eastern margin of the Oslo Graben, which does not coincide with any of the master faults in the Permian Oslo Graben, experienced an earthquake with a magnitude M = 5.4 in 1904 [97].Additionally, the magnitude of the seismic events associated with the post-glacial reverse displacement of the Stuoragurra Fault Complex in Finnmark in northern Norway has been calculated to be M = c.7 [98].Most current seismic activity is believed to be related to remnantlocalized long-term domal activity and marginal uplift in several parts of Norway [99][100][101][102] as well as Pliocene-Pleistocene sedimentary loading and unloading, ongoing post-glacial rebound (e.g., [103][104][105][106]) and, in some cases, loading associated with natural and manmade water reservoirs.Current seismic activity in Norway is mainly constrained to the continental shelf.Here, the pre-present stress configuration is monitored and well understood (e.g., [96,103,107]), providing a firm background for the analysis of fault reactivation (e.g., [63]).In areas of good exposure, post-glacial tectonic activity opens the opportunity to combine remote sensing methods with detailed field study, obtaining data which can be used for dynamic analysis of faults.One example is the Stuoragurra Fault Complex of Finnmark in northern Norway, where sections across the fault trace have been excavated to facilitate the dating of strata and study of the fault geometry and hanging wall and footwall deformations, including strain markers [108,109].The Stuoragurra Fault is a key element in the Mierujav'ri-Svaerholt Fault Zone (Figure 7).Field work supported by drilling, trenching, the use of geo-radar and seismic refraction has shown that it defines an approximately 90 km-long, SE-dipping reverse fault with a reverse displacement to the order of 10 m.The fault core is characterized by an approximately one-meter-wide layer of clay-altered fault gouge.It is marked by a more than 100 m-wide low-resistivity zone, probably reflecting fracturing in the footwall and hanging wall of the fault.Radiocarbon dating has shown that the largest earthquake along the Stuoragurra Fault Complex occurred less than 500 years ago [98,110].The recurrence interval of large earthquakes in intraplate regions like Norway can be several hundred years or even longer.Rock avalanches and landslides, potentially triggered by earthquakes, can generate tsunamis in fjords and lakes.Such information is therefore of high societal importance for hazard assessments and mitigating the effects of future large earthquakes on built-up areas and infrastructure.
remnant-localized long-term domal activity and marginal uplift in several parts of Norway [99][100][101][102] as well as Pliocene-Pleistocene sedimentary loading and unloading, ongoing post-glacial rebound (e.g., [103][104][105][106]) and, in some cases, loading associated with natural and man-made water reservoirs.Current seismic activity in Norway is mainly constrained to the continental shelf.Here, the pre-present stress configuration is monitored and well understood (e.g., [96,103,107]), providing a firm background for the analysis of fault reactivation (e.g., [63]).In areas of good exposure, post-glacial tectonic activity opens the opportunity to combine remote sensing methods with detailed field study, obtaining data which can be used for dynamic analysis of faults.One example is the Stuoragurra Fault Complex of Finnmark in northern Norway, where sections across the fault trace have been excavated to facilitate the dating of strata and study of the fault geometry and hanging wall and footwall deformations, including strain markers [108,109].The Stuoragurra Fault is a key element in the Mierujav ri-Svaerholt Fault Zone (Figure 7).Field work supported by drilling, trenching, the use of geo-radar and seismic refraction has shown that it defines an approximately 90 km-long, SE-dipping reverse fault with a reverse displacement to the order of 10 m.The fault core is characterized by an approximately one-meter-wide layer of clay-altered fault gouge.It is marked by a more than 100 m-wide low-resistivity zone, probably reflecting fracturing in the footwall and hanging wall of the fault.Radiocarbon dating has shown that the largest earthquake along the Stuoragurra Fault Complex occurred less than 500 years ago [98,110].The recurrence interval of large earthquakes in intraplate regions like Norway can be several hundred years or even longer.Rock avalanches and landslides, potentially triggered by earthquakes, can generate tsunamis in fjords and lakes.Such information is therefore of high societal importance for hazard assessments and mitigating the effects of future large earthquakes on built-up areas and infrastructure.For currently active faults, seismological data should usually be available and supplied by direct field observations while the deformation parameters are still fresh and available [111].The current development of SAR technology opens new possibilities for this kind of lineament and fault analysis but is not feasible for studying the present seismicity in Norway.Where such data exist, regular repetition of satellite recordings opens the possibility of dynamic analysis of contemporaneous stress configuration and tectono-magmatic processes [39].Combined with dynamic seismological data and fringe patterns in differential interferogram data (SAR, InSAR or Sentinel-1 (e.g., [112][113][114])), such information can be used to construct dynamic time series for tectonic events.Combined with field-controlled lineament analysis, this promotes a combined palaeo-and current tectonic history and also helps with defining the present stress configuration, and it can therefore be used to support seismic hazard assessment.
Thus, studies on the Stuoragurra Fault Complex has produced significant data for understanding of the post-glacial rebound, the total stress system and the recent and present seismicity of Norway.

Technical Presumptions and Requirements in Lineament Analysis
A statistical lineament analysis alone is of limited value beyond the identification of zones of variable deformation intensities, and the lineament data need to be supplied with all available geological information about the study area.The geological interpretation should be continuously updated and adjusted so that several interpretation cycles and the inclusion of current data can be achieved (Figure 7).Increasing the quality of the geological model with each interpretation step should be the aim.This requires that the quality of the database is correctly understood and that relevant data sets are included in the model.Thus, remote sensing analysis requires additional existing data sets from, for example, magmatic history (plutons and dyke systems), structural analysis (fault history) and applied geology (mineralization and water resources) to localize the strain elements [115,116].A selection of items that should be observed, which may influence the lineament interpretation and the final geological model and which accordingly should be considered by the interpreter are presented in the following.
Magmatic systems (dykes and dyke swarms): Beyond helping to define plate-tectonic and thermo-tectonic environments, individual dykes or dyke swarms and associated joint systems in particular can be essential in the definition of palaeostress systems at the time of emplacement (e.g., [117][118][119]).Several examples of such palaeostress estimates are known from studies of dyke systems in Norway (e.g., [24,[120][121][122]).
Geochronology applied to fault analysis: As already emphasized under the section covering focused and dynamic lineament studies, detailed geochronological data are of the uttermost importance in the study of different fault systems.Recent developments in geochronology methods for the study of faults utilize 40 Ar/ 39 Ar-and K/Ar systems on various generations of white mica in different types of fault rock (see [92] for a summary of the methods) and opens the opportunity for dynamic fault analysis on a scale which was hitherto impossible.This underlines the importance of acquiring high-resolution field data in the study of lineament and fault systems.
Holocene reactivation of ancient faults (e.g., the Stuoragurra Fault Complex) can be studied by trenching across fault scarps.Radiocarbon dating of organic matter located in buried and severely deformed sediment horizons provides reliable ages for the maximum age of faulting [98,110].
The significance of erosion and deep weathering: From the Triassic period to the Jurassic period, eastern Norway was on the same latitude as India and exposed to climate conditions warmer than those found today.Subtropical deep weathering due to acidic and "aggressive" water drained from large rain forests affected the entire surface.In zones of weakness formed during previous faulting activity, volcanism or hydrothermal alteration, the water was therefore able to penetrate deep into the bedrock [106,123].When the sea rose by 300-400 m and flooded the mainland in the Cretaceous period, the surface was buried by clay and chalk [124], similar to that seen in the North Sea.The uplift and erosion of eastern Norway began in the late Tertiary period.The products of the subtropical weathering were probably preserved until the Pleistocene glaciations and subsequent sea-level fall, causing erosion of the Cretaceous sediments and renewed weathering of the basement [106].In some depressions, products of the deep weathering are preserved (e.g., Djupdalen near Larvik and Lieråsen in the Lier-Asker area (Figure 8)).
and "aggressive" water drained from large rain forests affected the entire surface.In zones of weakness formed during previous faulting activity, volcanism or hydrothermal alteration, the water was therefore able to penetrate deep into the bedrock [106,123].When the sea rose by 300-400 m and flooded the mainland in the Cretaceous period, the surface was buried by clay and chalk [124], similar to that seen in the North Sea.The uplift and erosion of eastern Norway began in the late Tertiary period.The products of the subtropical weathering were probably preserved until the Pleistocene glaciations and subsequent sea-level fall, causing erosion of the Cretaceous sediments and renewed weathering of the basement [106].In some depressions, products of the deep weathering are preserved (e.g., Djupdalen near Larvik and Lieråsen in the Lier-Asker area (Figure 8)).Age dating of weathering can be carried out by applying the same isotope techniques (Ar/Ar and K-Ar) as those used for dating fault gouge.Palynological studies of spores and pollen in the saprolites can separate weathering from hydrothermal alteration [123].
The physical properties of clay, like its electrical resistance, seismic velocity, density and magnetic susceptibility, are different from those of crystalline rocks.Aerial magnetic measurements offer, for example, a fast and cost-effective way to map zones with a high clay content.Earth s magnetic field induces secondary magnetization in the bedrock, which in turn will contribute to the geomagnetic field that is measured.Deep weathering will cause a negative deviation in the measured magnetic field, and a filtering method has been developed to make the signal from deep weathering clearer (Figure 8 [43]).
Coincident negative anomalies in the magnetic fields and depressions in the topography can be used as indicators of clay alteration in the aeromagnetic and geomorphological relations (AMAGER) method.The method seems to work for most magnetic and metamorphic rock types in central eastern Norway [43,127].Its application for low magnetic sedimentary rock types seems to be more limited.The AMAGER method Age dating of weathering can be carried out by applying the same isotope techniques (Ar/Ar and K-Ar) as those used for dating fault gouge.Palynological studies of spores and pollen in the saprolites can separate weathering from hydrothermal alteration [123].
The physical properties of clay, like its electrical resistance, seismic velocity, density and magnetic susceptibility, are different from those of crystalline rocks.Aerial magnetic measurements offer, for example, a fast and cost-effective way to map zones with a high clay content.Earth's magnetic field induces secondary magnetization in the bedrock, which in turn will contribute to the geomagnetic field that is measured.Deep weathering will cause a negative deviation in the measured magnetic field, and a filtering method has been developed to make the signal from deep weathering clearer (Figure 8 [43]).
Coincident negative anomalies in the magnetic fields and depressions in the topography can be used as indicators of clay alteration in the aeromagnetic and geomorphological relations (AMAGER) method.The method seems to work for most magnetic and metamorphic rock types in central eastern Norway [43,127].Its application for low magnetic sedimentary rock types seems to be more limited.The AMAGER method can also be used to demonstrate clay alteration caused by circulating hydrothermal solutions in the rock, but these clay zones will probably be more localized than the deep weathering zones.The method is robust for the mapping of eroded deep weathering in central eastern Norway.
It is our experience that deep weathering is an important factor in enhancing the signatures of lineament and fracture patterns in remote sensing data, including (digital) topographical, photometric, multispectral and potential field (gravimetric and magnetic) data.
The lineament and structural geology database: Systematic acquisition of diverse data sets derived from remote sensing (including optical and multispectral, emittance, gravimetric, magnetic, geochronological, mineralogical and traditional structural geological field data) requires a combined map and digital or analogue database system.The database needs to be georeferenced, and its architecture should be robust, flexible and designed to absorb and merge old and new information with increasing detail and complexity over time to meet the demand for revised, new and dynamic data sets [128,129].The basal layer of the database should therefore contain regional tectonic and lithological information, and it is important that it can be expanded to include updated information when such is available (Figure 9).General data acquisition should combine traditional fieldwork and data obtained by general remote sensing studies as described above.Important elements in this part of data acquisition particularly include the definition of lithological and structural mega-units, including nappes, duplexes and fold systems of regional importance and should therefore include careful foliation mapping.The use of statistical remote sensing (mainly lineaments) as defined above should be used to strengthen a basic database.Although the bulk of such data could be acquired by automatized (lineament) analysis, the data would need a manual quality control before they are accepted as a data set to support the basal tectono-lithological database.The focus of this part of the database would be the definition of regional zones of weakness.Hence, it is necessary that the design of the database is open to a general or dedicated correlation with (regional) potential field data and seismological information.
The special part of the structural geology database should be open to the identification of individual structural elements (e.g., lineaments) and should be supplied with information on each element that requires full field control.Such information should include the orientation (strike and dip), dimensions, segmentation, deformation products (fault rocks), mineralogical and geochemical characteristics, weathering products and age.A complete set of information of this kind may require comprehensive fieldwork and be available for only some structures.It should still be an ambition to fulfil these data requirements for all structures of some importance.A natural starting point would be to extract existing information included in available geological maps and publications.
Of equal importance is linking dedicated parts of the database to other geological, mineralogical and geochemical data (e.g., [35,40,41]) as well as to potential field and other geophysical data (Figure 10).

Discussion
Remote sensing-based structural and lineament mapping have become routine when preparing for field-based structural geology studies, and systematic structural geological analysis is commonly performed in the framework of a continuous work cycle, which reflects the data available and the stage of knowledge at any point in the work cycle (Figure 11).In our opinion, lineament analysis still has significant untapped potential for identifying everything from structural features on the mega-scale (e.g., lithological units, plutons, wide shear zones, regional scale fracture systems and nappes) down to the individual structure (faults and folds).It is unlikely that the regional frameworks would have been identified without the use of remote sensing tools and lineament mapping techniques for general and statistical investigation to begin with.It is still realized that the full potential of remote sensing data and lineament mapping is not yet fully exploited, and great potential rests in further development of technology including sensor systems and new data types.Additional challenges include the homogenization and definition of quality standards, flexibility of digital databases and systems for integrating standardized high-quality and high-resolution field data.On the Norwegian mainland, a framework of lineament zones consisting of linked N-S-and ENE-WSW-striking lineament swarms have been identified through the use of general and statistical approaches as described above.Mapping them resulted in the databases used in focused and dynamic studies.In some cases, the focused lineament studies are significant to applied geology topics, like resource mapping and geohazard assessment.
individual structure (faults and folds).It is unlikely that the regional frameworks would have been identified without the use of remote sensing tools and lineament mapping techniques for general and statistical investigation to begin with.It is still realized that the full potential of remote sensing data and lineament mapping is not yet fully exploited, and great potential rests in further development of technology including sensor systems and new data types.Additional challenges include the homogenization and definition of quality standards, flexibility of digital databases and systems for integrating standardized high-quality and high-resolution field data.On the Norwegian mainland, a framework of lineament zones consisting of linked N-S-and ENE-WSW-striking lineament swarms have been identified through the use of general and statistical approaches as described above.Mapping them resulted in the databases used in focused and dynamic studies.In some cases, the focused lineament studies are significant to applied geology topics, like resource mapping and geohazard assessment.Among the lineament systems and individual lineaments included in Norwegian lineament databases, several display complex architectures and morphologies, suggesting that they represent region-scale fault zones.In many cases, this picture has been supported by field studies, which confirms that the lineaments represent normal or strike-slip faults, perhaps in different stages of development and with varying degrees of reactivation.It is acknowledged that many of these structures coincide with previously known faults in Norway, some of which have already been described in the ancient (e.g., [13,130,131]) and contemporaneous literature [8,79,80,82,108].Several of the studies based on modern lineament methods have, however, added much to the description and interpretation of these structures, particularly in cases where the structures have been extensively studied for decades.
Field studies have revealed that several of the master faults share certain characteristics: Many have cores consisting of plastic fault rocks (different types of mylonite) which likely reflect a (late) Proterozoic origin and medium-to-high-grade metamorphism.In several cases, the mylonitic cores are overprinted by brittle fault rock, which includes cataclasite and pseudotachylite, some of which suggest post-Caledonian reactivation [75,92,132,133] and that this deformation was post-dating crustal uplift.In some cases, the distal parts of the fault cores (see, for example, [7,134] for a description of the fault core nomenclature) may also contain zones of cohesionless material commonly representing fault gouge and, in some cases, deeply weathered zones which are remnants from Mesozoic saprolites [43,123].Altogether, this documents that many large-fault systems in the Proteozoic basement of Norway are structures of great potential for reactivation stretching from the (late) Proterozoic period to the present, demonstrating that these may represent stable zones of weakness on the crustal and perhaps lithospheric scales.It is also recognized that the expression of lineaments in an optical spectrum and potential field data may be heavily influenced by secondary processes like mineralization and weathering [40,41,43].
Still, much work remains to fully understand the deep structure and history of the major fault systems of Norway and the connections between them and remote sensing data.Lineament analysis at various scales will continue to be an important research tool.A key element in this is the development of a consistent structural database.To utilize its full potential, however, it is necessary to combine automatized (computer-generated) manual interpretations and field data.The major goals of such interpretations include the identification of isolated and related lineament and fracture populations, age determinations and separation between thick-skinned and thin-skinned zones of weakness.Once established, profound zones of weakness have a well-documented capacity for reactivation, even in cases where the secondary stress situation is different from the primary one.Accordingly, such zones tend to remain active over long geological time spans [9,26,75,98,110,135,136], accumulating varying types of fault rocks and fracture systems such that originally deepseated, high pressure-temperature, quasi-plastic deformation products become overprinted by brittle structures generated at shallower levels of burial [136][137][138].High-resolution chronological work is crucial in the documentation of such developments (e.g., [78,92]) and therefore needs to be included in the database on such structural elements.
The structural geological database is a key element in determining the significance and development of the tectonostratigraphic units in question.The database therefore needs to be robust, flexible and scale-independent so that it can absorb data sets from mega-to microscales and preferentially in four dimensions (including the time dimension) (Figures 7 and 10).One key element is careful design of the model mesh, which should be prepared so that it can absorb all datasets inside the study area (independent of scale) [129].Ideally, the mesh should be adaptable to 3D mechanical algorithms (e.g., finite element), such as those utilized for the mid-Norwegian shelf [63], and chemical and mineralogical datasets and modeling tools as well.This means that the database should be designed to serve current and future modeling tools (object-oriented algorithms) [139].Simultaneously, it is important that the database is continuously updated and complete so that it is updated at any time and reflects all the current knowledge of the area in question.In Norway, a structural and lineament database has been under development for five decades, and it a priority to update this with state-of-the art structural data from current field campaigns.
In our opinion, it is important that database and interpretation tools have options for manual interaction which enable the responsible interpreter (geologist or geophysicist) to use data from ongoing work, whether they be from the field or from the laboratory, so that the geological experience of the interpreter is also allowed to play a role in the interpretation.This requires several steps in the interpretation, some of which are performed manually whereas others can be handled by algorithms involving pattern recognition and machine learning.This can probably be achieved in a three-step procedure that involves general structural interpretation (manual), statistical lineament interpretation (computer-generated) and special interpretation, which includes field control and traditional structural geological methods (manual).
We argue that the concept of lineament analysis is not obsolete; rather, the concept should be expanded upon to include state-of-the art acquisition technologies for optimizing the analysis of data.Distinct structural units might, in many cases, receive more focused attention in the analysis of lineament data.

Conclusions
The present contribution argues for the use of remote sensing data in a wider context in regional studies on structural geology, of which dedicated lineament study is one element.It is suggested that such studies should include up to four stages or approaches, including general structural mapping and identification of structural mega-units and foliation patterns, which should be performed at least in part manually and before and in preparation of fieldwork.The second approach should be statistical scrutiny of the available remote sensing data to identify zones of high lineament density and the general geometric characteristics and classification of lineaments.For reasons regarding capacity, efficiency, consistency, interpretation and statistical handling, this interpretation stage should be performed automatically by applying algorithms for pattern recognition and machine learning.The final (focused) approach should address specified elements and should include a full structural geological analysis, including field verification and preferentially geochronological and dynamic aspects when possible.
Geophysical (potential field) data yield important additional information on structural and lineament interpretations and should be utilized in all three stages with the specific aim to identify thin-skinned and thick-skinned structures and for the assessment of structures as seen on the regional and lithospheric scales.
It is emphasized that great care is taken in the design of the database for structural geological and lineament data with respect to flexibility, robustness and continuation so that the database can be continuously expanded regarding both data types and resolution.The key factors in the construction of the database are object orientation, a flexible, GIS-oriented mesh and having the capacity to accumulate the escalating volumes of data.Although it is scarcely realistic that all options for the database description above can be fulfilled from the start, options should be available for all elements to be included during the life of the database.The database provides the core of all analyses and modeling in a scientific environment.However, options for data analysis and modeling tools depend on the computational capacity, and technologies and will come and go [127,139,140].Some organizations in a modern society have been assigned the responsibility of managing ever-growing amounts of scientific data and knowledge on the behalf of decision makers and society in general.To ensure the accessibility, availability, quality and relevance of the database, it must be utilized within an organization that acquires data, maintains the database and uses it for scientific purposes.The Geological Survey of Norway is an example of such an organization.

5 Figure 1 .
Figure 1.Main geological provinces of Scandinavia.Yellow frame shows focused central study area in south Norway, and positions of all examples described in the text are marked with red frames and yellow numbers: 1 = Seve-Köli Nappe Complex in Børgefjell, Nordland, 2 = Hardanger-Ryfylke Nappe Complex in Seljestad, 3 = Bømlo, Bergen Fault Zone, 4 = Lista-Drangedal Fault, 5 = Stuoragurra Fault, 6 = Lieråsen.Fault complexes and fault zones of particular interest to the present study are marked in yellow letters.Map modified from [8].

Figure 1 .
Figure 1.Main geological provinces of Scandinavia.Yellow frame shows focused central study area in south Norway, and positions of all examples described in the text are marked with red frames and yellow numbers: 1 = Seve-Köli Nappe Complex in Børgefjell, Nordland, 2 = Hardanger-Ryfylke Nappe Complex in Seljestad, 3 = Bømlo, Bergen Fault Zone, 4 = Lista-Drangedal Fault, 5 = Stuoragurra Fault, 6 = Lieråsen.Fault complexes and fault zones of particular interest to the present study are marked in yellow letters.Map modified from [8].

Figure 2 .
Figure 2. Lineament intensity zones in southern Norway.The coastlines of southern Norway are strongly influenced by the NE-SW-striking Møre-Trøndelag Fault Complex (MTFC), shown in dark yellow, and the Agder-Telemark Zone and N-S-striking Bergen Zone (yellow).Stars indicate position of key locality in the lineament intensity zones (see text for description).Faults in these lineament zones seem to have young (Mesozoic or younger) structural overprints.Young reactivation is less conspicuous for the bulk of the faults that define the Østfold Zone (blue).From [8].

Figure 2 .
Figure 2. Lineament intensity zones in southern Norway.The coastlines of southern Norway are strongly influenced by the NE-SW-striking Møre-Trøndelag Fault Complex (MTFC), shown in dark yellow, and the Agder-Telemark Zone and N-S-striking Bergen Zone (yellow).Stars indicate position of key locality in the lineament intensity zones (see text for description).Faults in these lineament zones seem to have young (Mesozoic or younger) structural overprints.Young reactivation is less conspicuous for the bulk of the faults that define the Østfold Zone (blue).From [8].

Figure 3 .
Figure 3.Comparison between (a) foliation map of [48] and (b) the tectonic map of Nordland in north Norway.For reference, dark green (also marked with the letter "e") and light green indicate the positions of the Seve-and Köli nappe systems of the upper allochthon.Note the structural contrast between the foliation map (a) and lineament map (c).The lineament map shows examples from two selected parts (subareas) of the area shown in the foliation map (red box).(d) Rose diagrams show the dominant lineament trends in the subareas.Study area shown in Figure 1 (red box 1).Note that the framed area here covers only a part of the study area shown in Figure 1.

Figure 3 .
Figure 3.Comparison between (a) foliation map of [48] and (b) the tectonic map of Nordland in north Norway.For reference, dark green (also marked with the letter "e") and light green indicate the positions of the Seve-and Köli nappe systems of the upper allochthon.Note the structural contrast between the foliation map (a) and lineament map (c).The lineament map shows examples from two selected parts (subareas) of the area shown in the foliation map (red box).(d) Rose diagrams show the dominant lineament trends in the subareas.Study area shown in Figure 1 (red box 1).Note that the framed area here covers only a part of the study area shown in Figure 1.

Figure 4 .
Figure 4. Identification of lithological units and nappe elements (master thrust zones and intranappe duplexes) of the Kvitenut and Revseggi nappe system of the Hardangarevidda-Ryfylke Nappe Complex and its interference with younger (post-Caledonian), linear brittle faults [50] based on Landsat imagery.Note the contrasting foliation patterns affiliated with the nappe units.Both of these data sets were produced by remote sensing techniques alone and later controlled by field mapping.The greater position of the study area is shown in Figure 1 (red box 2).Note that Figure 4 is tilted and is detailed inside this box.

Figure 4 .
Figure 4. Identification of lithological units and nappe elements (master thrust zones and intranappe duplexes) of the Kvitenut and Revseggi nappe system of the Hardangarevidda-Ryfylke Nappe Complex and its interference with younger (post-Caledonian), linear brittle faults [50] based on Landsat imagery.Note the contrasting foliation patterns affiliated with the nappe units.Both of these data sets were produced by remote sensing techniques alone and later controlled by field mapping.The greater position of the study area is shown in Figure 1 (red box 2).Note that Figure 4 is tilted and is detailed inside this box.

Figure 5 .
Figure 5. Data from statistical lineament analysis for the west coast of southern Norway.(a) Visual spectral data.Area subject to detailed analysis (Bømlo) indicated by blue arrow.(b1,b2) Results from detailed lineament study in the Bømlo area interpreted on a scale of 1:25,000 [60].The color code in (b2) indicates lineament sectors separated by the azimuth.(c) Manual lineament interpretation of Bergen Zone (from [67]) with (d) rose diagram showing the lineament orientation in the central part of the Bergen lineament zone [79].Note the contrast in lineament orientation between the interpretation performed at 1:25,000 (b1,b2) and 1:250,000 scales (d).(e) Contoured lineament density diagram for the Bergen Zone.Yellow-red colors indicate areas of high lineament density.Note that the scales for the regional maps (a-c) and the inset (b1) are not the same.

Figure 5 .
Figure 5. Data from statistical lineament analysis for the west coast of southern Norway.(a) Visual spectral data.Area subject to detailed analysis (Bømlo) indicated by blue arrow.(b1,b2) Results from detailed lineament study in the Bømlo area interpreted on a scale of 1:25,000 [60].The color code in (b2) indicates lineament sectors separated by the azimuth.(c) Manual lineament interpretation of Bergen Zone (from [67]) with (d) rose diagram showing the lineament orientation in the central part of the Bergen lineament zone [79].Note the contrast in lineament orientation between the interpretation performed at 1:25,000 (b1,b2) and 1:250,000 scales (d).(e) Contoured lineament density diagram for the Bergen Zone.Yellow-red colors indicate areas of high lineament density.Note that the scales for the regional maps (a-c) and the inset (b1) are not the same.

Figure 7 .
Figure 7. (a) Simplified geological map of Finnmarksvidda with mapped postglacial faults (modified from [98,110]).The 90 km-long Stuoragurra Fault Complex (SFC) consists of two separate fault systems: the Máze Fault System to the south and the Iešjávri Fault System to the north.The SFC occurs within the 4-5 km-wide Mierojávri-Svaerholt shear zone (MSSZ) located along the northwestern boundary of the Jergul Gneiss Complex.The MSSZ is also characterized by magnetic anomalies produced by highly magnetic mafic intrusions (diabase, albite diabase and gabbro).Evidence of a total of 60 landslides was found within 20 km of the fault scarps.A total of approximately 80 earthquakes were registered along the SFC between 1991 and 2019.Most occurred to the southeast of the fault scarps and less than 10 km from the extrapolated Mierojávri-Svaerholt shear zone at depth.The maximum moment magnitude was 4.0.(b) Aerial photograph (SE view) of the northernmost section of the Máze Fault System draped over digital topography (from

Figure 7 .
Figure 7. (a) Simplified geological map of Finnmarksvidda with mapped postglacial faults (modified from[98,110]).The 90 km-long Stuoragurra Fault Complex (SFC) consists of two separate fault systems: the Máze Fault System to the south and the Iešjávri Fault System to the north.The SFC occurs within the 4-5 km-wide Mierojávri-Svaerholt shear zone (MSSZ) located along the north-western boundary of the Jergul Gneiss Complex.The MSSZ is also characterized by magnetic anomalies produced by highly magnetic mafic intrusions (diabase, albite diabase and gabbro).Evidence of a total of 60 landslides was found within 20 km of the fault scarps.A total of approximately 80 earthquakes were registered along the SFC between 1991 and 2019.Most occurred to the southeast of the fault scarps and less than 10 km from the extrapolated Mierojávri-Svaerholt shear zone at depth.The maximum moment magnitude was 4.0.(b) Aerial photograph (SE view) of the northernmost section of the Máze Fault System draped over digital topography (from www.norgeibilder.no).This fault segment at Stuoragurra is located approximately 10 km north-northeast of the Masi or Máze settlement and cuts across to the glacifluvial deposits, and it is highlighted by a red circle.

Figure 8 .
Figure 8.(a) Interpretation map of Lier-Asker area with the 10.7 km-long Lieråsen railway tunnel.Previously mapped weakness zones [125] are marked with purple polygons and lines, while predicted zones of deep weathering [43] based on processed magnetic and topographic data are shown in cyan (probable) and yellow (possible).Violet and purple dots represent locations of wells from NGU s national well database drilled before and after 1980, respectively.(b) Geological profile along the Lieråsen railway tunnel, adapted from [126].Predicted zones of deep weathering from coinciding magnetic and topographic lows are shown in cyan (probable) and yellow (possible).(c) Conceptual model for the present occurrence of deep weathering in Norway [123].(d-f) Examples of deep weathering types in Norway [123].

8 .
(a) Interpretation map of Lier-Asker area with the 10.7 km-long Lieråsen railway tunnel.Previously mapped weakness zones[125] are marked with purple polygons and lines, while predicted zones of deep weathering[43] based on processed magnetic and topographic data are shown in cyan (probable) and yellow (possible).Violet and purple dots represent locations of wells from NGU's national well database drilled before and after 1980, respectively.(b) Geological profile along the Lieråsen railway tunnel, adapted from[126].Predicted zones of deep weathering from coinciding magnetic and topographic lows are shown in cyan (probable) and yellow (possible).(c) Conceptual model for the present occurrence of deep weathering in Norway[123].(d-f) Examples of deep weathering types in Norway[123].

Figure 9 .
Figure 9.A GIS-based mesh should provide the framework for the structural database.The database should be designed to accumulate several datasets and have the capacity to include data types not foreseen today.The mesh should have the capacity to store data at different scales in targeted areas.

Figure 9 .
Figure 9.A GIS-based mesh should provide the framework for the structural database.The database should be designed to accumulate several datasets and have the capacity to include data types not foreseen today.The mesh should have the capacity to store data at different scales in targeted areas.

Figure 9 .
Figure 9.A GIS-based mesh should provide the framework for the structural database.The database should be designed to accumulate several datasets and have the capacity to include data types not foreseen today.The mesh should have the capacity to store data at different scales in targeted areas.

Figure 10 .
Figure 10.Schematic relation between central 3D database and other available databases and affiliated interpretation tools.Modified from [129].

Figure 11 .
Figure 11.Procedure for structural analysis of remote sensing data which includes three stages.Stage 1 (general stage) features an exhaustive structural analysis that targets the identification of Figure 11.Procedure for structural analysis of remote sensing data which includes three stages.Stage 1 (general stage) features an exhaustive structural analysis that targets the identification of tectonic mega-structures.It is suggested that this stage should be performed manually.Stage 2 (statistical stage; computerized interpretation) would concentrate on the identification of lineament zones and populations.Stage 3 (special analysis) should be performed manually and concentrate on specified lineaments believed to represent fracture corridors and faults.