CHAPTER 3 THE FORMATION OF ROCKS SECTION 3.1. H O W IGNEOUS ROCKS ARE FORMED Introduction 1. Igneous rocks may be defined as those rocks which have solidified from liquid melts, or magmas as they are termed in geology. These rocks are classified in two ways, firstly by their chemical composition and secondly by their grain size. The grain size is largely dependent on the speed at which the rock has cooled; if the cooling has been very slow then the individual crystals will be large and the rock will be coarse-grained; if the cooling has been fast then the crystals will be small, even microscopic, and the rock will be fine-grained. Thus molten lava quickly extruded from a volcano will become a fine-grained igneous rock when it has cooled and solidified. 2. The chemical composition of igneous rocks varies considerably and the usual method of expressing this is by the percentage of silica ( S i 0 2 > contained. A rock with a high silica content is termed 'acid' and one with a lower silica content is known as 'basic'. These terms apply whether the grain-size is fine or coarse. It is not necessary for a rock to be sent to a laboratory for chemical analysis to determine its chemical composition; the various types may be deter mined, approximately enough, by field characteristics for almost all engineering purposes. The most common types of igneous rock are as follows (their properties and field identification are dealt with in Chapter 4): Coarse-grained Granite Diorite Gabbro 70% approx S i 0 (Acid) 60% approx Si02 (Intermediate) 50% approx S i 0 (Basic) 2 2 Fine-grained Rhyolite Andesite Basalt Igneous rocks are formed either from a complete melting of previouslyformed rock (of any type; igneous, metamorphic or sedimentary or any combina tion of these) or more often from magmas from deep in the Earth which had not previously been near the surface, and which had been molten (or solid-state above their melting point) for very long periods. If solid rock is forced down to a sufficient depth by any means then it will melt. If subsequent forces push it to the surface again then it will appear as an igneous rock. 3. Acid magma is much more viscous than basic magma, and this accounts for the fact that the two most commonly found types of igneous rock are granite (coarse-grained) and basalt (fine-grained), since basic basalts are extruded more 29 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.1 easily than acid rhyolites. Thus wide areas, such as the Deccan of India or much of Northern Ireland, may be covered by sheets of basalt, which have spread out quickly over the land or ocean floor before cooling. On the other hand acid magmas tend to cool slowly at depth. When erosion subsequently occurs the granites thus formed are exposed to view, such as the chain of granites at Dart moor, Bodmin Moor, St Austell, Land's End and the Scilly Isles, all in South west England (which may be connected at depth), or the immense Sierra Nevada granite in the Western USA. 4. When igneous rocks appear on the surface as outcrops, the magma has intruded the previously-existing rock in a number of different ways, giving rise to differently shaped igneous bodies, the chief types of which are: a. Dykes and sills (Figures 11 and 12). Both of these sheet-like structures vary from a few centimetres to more than 100 metres in thickness, a most usual thickness being about 1 to 5 metres for dykes and rather thicker for sills. They are both injections of magma from underground into the overlying rock, a dyke cutting more or less vertically through the bedding of the older rock and a sill intruding roughly parallel to the bedding. Often the heat from the liquid magma alters the surrounding rock for a short distance on either side of the contact (rock is usually a rather poor conductor of heat) and this phenomenon is known as 'contact metamorphism'. Dykes and sills are always fine-grained at their edges, but in thicker intrusions the rock may be coarse-grained towards the centre of the intrusion, where the magma has cooled slowly. b. Volcanic flows and ashes. These come from an active volcano and may occur over land or over sea floor, or even flow from land to sea. As explained above, flows of basic composition are much more common than acid flows. When later preserved in a sequence of rocks, flows may be distinguished from sills in two ways; (1) rocks above a flow show no contact meta morphism, and (2) flows do not transgress vertically from bed to bed as sills Fig 11. Cross-section through rocks intruded by a dyke. (The stipple indicates the area in which contact metamorphism is sometimes found) 30 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.1 Fig 12. Cross-section through dipping rocks intruded by a sill. (Stipple as in Figure 11) sometimes do (see Figure 71). Whilst some flows are of liquid magma, others are density flows of hot gases, which can move large quantities of ash and cinder at high speeds, and which on solidification have an aerated texture when geologically young (sometimes termed ignimbrites). After some geological time and burial, these small gas holes usually become filled with secondary recrystallisation products. Ashes are also ejected from volcanoes, sometimes to great heights, and come to rest some distance from the volcano, both on land and out to sea. Beds of ash can eventually solidify into rock, varying from many metres to less than a centimetre in thickness. Such rocks may have similar engineering properties to mudstones, siltstones or sand stones of equivalent grain size, although in some sequences of solid sedimentary rocks there are very fine-grained thin ashbands, known as 'bentonites' (for uses see Section 10.2, paragraph 6c and Section 12.7, para graph 6). c. Volcanic necks. These are near-circular structures in plan, varying from about 100 metres to over 1 kilometre in diameter, up which magma has passed, usually into a once-active volcano. They may be filled with a variety of igneous rocks, often with a conglomeratic or brecciated selection of many different rocks. Sometimes the neck contains coarse or fine-grained igneous rock crystallised from magma, which has come up at a late stage in the volcano's history and solidified before reaching the surface. d. Batholiths. These are much larger structures than necks, often elongate in plan and ranging from one or two kilometres up to many hundreds of kilometres in length. These usually represent the once-molten roots of mountain chains which, after melting, have moved upwards, first to crystal lise and solidify and finally to become exposed on the surface when the covering rock above has been eroded away. Batholiths are invariably made up of coarse-grained igneous rocks. The largest ones are usually acid in 31 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.2 composition, made up of granite formed from the melting of continental crust. Around the sides and at the roofs of batholiths (when these are exposed) a zone of contact metamorphism is usually seen, with a thickness of metres or even kilometres around a large intrusion. In addition large blocks of pre-existing rock sometimes fall into a magma whilst the latter is still molten, and may be much altered before the surrounding magma solidifies. SECTION 3.2. H O W SEDIMENTARY ROCKS ARE FORMED Introduction 1. The distinction between igneous and sedimentary rocks is that whilst the former came from beneath the Earth's surfaces and crystallised from hot magmas, sedimentary rocks were formed at the Earth's surface at normal atmospheric temperature. 2. There are two main types of sedimentary rock, which will be considered separately: a. Those whose constituent particles have been transported to the place of deposition, known as 'clastic' rocks. b. Those which have been formed from nearby, either by aggregation of organic matter or by chemical deposition. Clastic rocks 3. These rocks are classified according to their average grain size as follows: coarse-grained rocks are Conglomerate and Breccia (both consolidated) and Gravel (unconsolidated); medium-grained rocks are termed Sandstone and finergrained rocks are known as Siltstone, Shale or Mudstone. The properties and identification of these rocks are given in Chapter 4. 4. Most clastic rocks are formed under water, but some are formed on land, in particular in desert conditions and also deposited by ice sheets in colder lati tudes. Desert sandstones are usually well-sorted, that is most of the constituent particles are of approximately the same size, due to sifting by the wind. On the other hand glacial deposits have been dumped haphazardly by melting glaciers or ice sheets, and are usually un-sorted, with rock fragments of all sizes and shapes jumbled to form tillites or breccias, which may be interspersed with lenses of finer-grained rocks. 5. The clastic sedimentary rocks formed under water are usually of fairly uniform grain size since they have been sorted to a greater or lesser extent by the water currents which have transported them. The material is carried in the first instance by rivers. Nearly all the transport of sediment occurs only at periods immediately following heavy storms, since only quickly-moving water is capable of carrying anything but the finest sediment. It is only at times of really violent 32 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.2 storms, such as occur perhaps two or three times in a century, that the largest blocks are moved, e.g. Lynmouth disaster of 1953. Eventually the sediments will come to rest in a lake, delta or sea and will compact into rock when the weight of the overlying sediment has caused consolidation, thousands or even millions of years later. The degree of consolidation of a rock bears no relation to its geological age, for example Cambrian rocks laid down over 500 million years ago near Leningrad, USSR, are much softer and weaker than rocks of the same grain size laid down in the Alps less than 20 million years ago. 6. Sedimentary rocks formed under fresh water are often indistinguishable from those formed under sea water, apart from the fossil animals and plants which they may contain. Some types of animal and plant thrive in fresh water or on land, others in sea water; only a small proportion of living things can tolerate both fresh and salt water. Fossils are also useful in determining the age of the rock, since different animals and plants existed at different evolutionary stages in the Earth's history. However the age of the rock is not usually of direct importance to the engineer, since its physical properties do not correlate with its age. 7. Although much rock-forming sediment is deposited directly by rivers, ocean currents and waves are also important in shifting loose sediment on the sea floor, and even eroding submarine and coastal rock outcrops. Banks of sediment near the edges of ocean basins or continental shelves are often unstable, and the addition of more sediment, storm conditions, or perhaps an earthquake shock, may cause such banks to slide into the depressions. Such subaqueous slides often form turbidity currents of sedimentary particles in water suspension which can reach speeds up to 50 km/h and thus spread the sediment evenly over a large area, perhaps some tens of kilometres wide. The rocks eventually formed from these deposits, 'turbidites', usually consist of a mixture of grain sizes, each turbidite bed having coarser particles at its base and finer particles at its top, termed a 'graded bed'. The tops and bottoms of beds in a turbidite sequence are usually more parallel than other types of sedimentary rock, with each bed extending as far as the original turbidite flow. Rocks formed in situ 8. These fall under four headings, each with a different mode of origin: a. Limestones, dolostones and cherts. The first two rocks are mainly com posed of the carbonate minerals calcite (CaCCfe) and dolomite CaMg(COa)2 respectively. Most are organic in origin, being largely made up of the remains of fossil animals and plants; some are inorganic, made up of chemicals precipitated from sea water in shallow high temperature conditions, some are a mixture of the two. However they are often recrystallised during subsequent geological time, so that both organic and inorganic limestones may be considered to be alike by the engineer. Many limestones contain small spherical objects, varying much in size but often about 1 mm in diameter, called ooliths. These are carbonate aggregations formed by 33 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 33 crystallisation around a minute central particle, and kept rounded by con tinual wave or current action during growth. Some limestones, such as the Jurassic limestones forming much of the Cotswold Hills in England, are largely made up of ooliths. Chalk is another type of limestone, formed by immense numbers of algal skeletons so small as to be invisible to the human eye (about 10 microns diameter). Chert, including its variety flint, consists of amorphous silica (SiC>2) often occurring as bands or nodules within limestone sequences. The silica is deposited originally under sea water, and concentrated into the bands or nodules during the rock-forming period subsequent to deposition. b. Evaporites. When sea water evaporates, usually by the sun's heat acting in an enclosed, or semi-enclosed basin, then the various salts remain behind, which can eventually accumulate as rock. This process occurs on the shores of the Persian Gulf today, when sea water occasionally covers wide coastal areas and subsequently dries out. Thick beds of salt (NaCl) and other evaporite minerals exist under many parts of the Earth. They are not seen at the surface except in arid areas, since the salts will have been dissolved and carried away by circulating groundwater. c. Coal. Some large rivers, such as the Nile, form deltas at their mouths where sediment spreads out over a considerable area. At some periods in the Earth's history very large deltas have been covered by vegetation, which on death has become buried. Thick deposits of dead vegetation become coal seams when the subsiding deltaic deposits subsequently become preserved as rock. These seams are usually interbedded with the sand, silt and mud brought down by the river. Coal-forming forests flourished at particular times, for example nearly all the coal in Britain and the eastern USA is of Carboniferous age (about 300 million years ago) and is formed from large extinct trees and fern-like plants. However coal of other geological periods from the Devonian onwards is to be found in many parts of the world, for example in the Tertiary of Spain. d. Oil and natural gas. Since they occur in rock, oil and gas are mentioned here for completeness as geological phenomena. Both are formed from the decay of microscopic marine animals and plants without hard parts, leaving organic hydrocarbons which are deposited in most types of sedimentary rocks. To become of economic importance they have to be concentrated by subsequent geological processes into traps capable of being tapped by drill ing. Gas often originates by emigration from coal deposits. SECTION 3.3. H O W ROCKS MAY BE CHANGED UNDERGROUND Faulting, folding, jointing and earthquakes 1. When rocks lie at an angle to the horizontal they are known as 'dipping rocks' (see Figure 13). The dip angle is expressed in degrees from the horizontal, followed by the compass orientation of the maximum dip, for example 'the limestone dips 45 degrees at 120 degrees' or sometimes simply 'Dip 45/120'. 34 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.3 'Strike' is by definition the direction at right angles to the maximum dip when seen in plan view, so that on a geological map the line of outcrop at a level surface of a thin rock bed follows the strike. Fig 13. Block diagram to demonstrate dip and strike 2. When rocks fracture and move this is termed 'faulting' (see Figure 14). The plane of fracture is termed a 'fault'. The amount by which the rocks appear to have been displaced is termed 'throw', and is expressed in linear measurements, usually metres. There are many different types of fault (see also Chapter 7), but the most common are: a. Normal (or gravity) fault (as in Figure 14), where relative movement between two blocks of rock is vertical along the fault plane. b. Tear fault, where relative movement between two blocks of rock is horizontal along the fault plane. c. Thrust fault, where one block of rock has been driven over another. Fig 14. Cross-section through faulted rocks 35 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 33 3. In the first two types, the fault will appear as an approximately straight line on a small-scale geological map; in thrust faults the line on the map will be most irregular, since the fault plane will be approximately horizontal, and its outcrop pattern will be very dependent on the local topography. Some tear faults may be extremely large, for example the San Andreas fault in California USA is over 1000 km long, and has a lateral throw of more than 200 km separating two con tinental plates (see Chapter 2). Movement along this fault caused an earthquake, followed by a disastrous fire in San Francisco in 1906, when the relative lateral movement was 6 metres at one time along a large part of the fault. However many faults may be seen in quarry faces which are completely stable today, and whose throw may be as little as 10 mm. 4. Often when rock strata have been subjected to pressure, usually horizontal; instead of breaking they will have buckled. This is termed 'folding' (Figure 15). The scale of folding varies widely, on the one hand it is possible to find a piece of gneiss perhaps 100 mm across with twenty or thirty small folds across its face; on the other hand some folds have amplitudes of several hundred kilometres. A syncline and anticline are illustrated in Figure 15. Sometimes folding is associated with faulting, at other times there is no such relationship (see also Chapter 2). When pressure has occurred from more than one direction, then domes and basins are formed, the first with the rocks dipping outwards on all sides, the second with the rocks dipping inwards towards the centre. When the core of a dome is formed of a material such as salt which is plastic at the depth and pressure concerned, this may be squeezed upwards into the overlying rock to form a diapiric structure. Anticline Syncline Fig 15. Cross-section through folded rocks 5. When folding takes place, very often cracks develop in the strata, even though no movement may take place along them. Such cracks are termed 'joints', and joint patterns often make up a meshwork of cracks, usually at right angles to each other. The cracks are often enlarged, particularly in limestones, by the 36 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.3 subsequent passage of groundwater. Joints are usually at right angles to the bedding planes, and can also form in thick beds without external pressure, simply due to the internal stresses which exist in the very first stages of rock formation, at the same time as the pore water is driven out. Jointing can also be formed when igneous rocks cool, a common form being the hexagonal-sided columnar jointing to be seen in many basalts. This is the origin of the impressive structures at the Giant's Causeway, Ireland, and Fingal's Cave, Scotland. 6. After rocks have been folded, perhaps faulted, uplifted and partially eroded, then very often they are subsequently submerged and again have fresh sediment deposited upon them. When this second sedimentary sequence has been changed into rock, both sequences together may be brought up to the surface again and exposed as a rock outcrop. The junction between the first sequence and the second sequence is known as an 'unconformity' (Figure 16). An unconformity is recognisable in even a small outcrop by the difference in the angle of the dips above and below the plane of unconformity. The rocks beneath an unconformity do not necessarily have to be sedimentary rocks; where sedimentary rocks rest upon igneous or metamorphic rocks, that is also known as an unconformity. The difference in age between rocks below and above an unconformity varies considerably, anything between perhaps 50 000 years and thousands of millions of years. Fig 16. Cross-section through an unconformity 7. Earthquakes are sudden tremors in the Earth, and usually result from the movement of faults. The chief occurrences of earthquakes may be mapped out as a series of belts (see Figure 17). These earthquake belts mark the margins of continental plates (see Chapter 2). Thus in stable areas, not near the edges of plates, such as Britain, the danger from earthquakes is small; but in areas of high seismic activity, such as Japan or Greece, then extra care is needed in the design and construction of any permanent, or even semi-permanent structure. 37 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.3 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 33 8. Earthquakes are measured either by local intensity or by magnitude. The scale used for intensity is the modified Mercali scale, which is a largely subjective scale based on the actual effect of the earthquake at the place of measurement. The scale runs from I (detected only by seismographs), through VI (slight damage), to XII (catastrophic). Thus the intensity of a single earthquake will differ in value from place to place, depending on its varied local effects. The magnitude scale, on the other hand, reflects the total energy released by an earthquake, and will only have a single value for each shock. The magnitude is calculated by complex equations relating the ground motions recorded by a seis mograph to the distance of the instrument from the epicentre. Several magnitude scales have been proposed, but that most commonly used is by Gutenberg and Richter, which is logarithmic. Small detectable disturbances, with energy releases of about 6-3 x lO^ergs, have a magnitude of 0, whilst the largest, with an energy release of about 2 x lO^ergs, have a magnitude of 8-5. There are more than a million earthquakes each year, ranging from an annual average of one earthquake with magnitude 8 or more, to a majority with magnitude less than 3. Metamorphism and hydrothermal activity 9. When rocks of all types are subjected to very high temperature and pressure, they will melt and remobilise to form igneous magmas. However since the melting points of each of the many constituents within a rock varies widely (and also the melting point increases with increase of pressure), some constituents will recrystallise and reform before others. Rocks which have been noticeably altered, but which still reflect some traces of their original bedding and structure are termed 'metamorphic' rocks. A special case of metamorphism, contact meta morphism, has been mentioned above, but this is only a local phenomenon, occurring close to intruded igneous rocks. Most metamorphic rocks are found in much larger areas, which have been pushed down to substantial depths in the Earth's crust, perhaps 20 km deep or more, and then raised again and finally eroded in some subsequent period of Earth history. The whole of the Highlands of Scotland is one such area, where regional metamorphism has occurred, inter spersed by a few granites where the rock has been completely melted and remobilised upwards into the metamorphic horizons. Much larger areas of metamorphic rocks occur for example in Canada and Central Africa. 10. The four principal types of metamorphic rock are as follows: a. Slate. Shales may have the orientation of their constituent particles altered by a relatively small increase in pressure and temperature to form slate. When this has occurred the rocks will become hard and brittle, and split along directions often at angles unrelated to their original bedding planes. This property is termed rock cleavage. Slate from North Wales was used extensively as roofing material during the nineteenth century, and is still common. b. Schist. Finer-grained rocks which have become completely recrystallised are termed 'schist'. The minerals recrystallise under high temperatures and pressures parallel to one another; especially conspicuous in most schists 39 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Section 3.4 are shiny flakes of mica minerals, but many other minerals also occur. The schistosity of a rock is similar to the cleavage of slates, but on a coarser and less uniform scale. Sometimes the schistosity parallels the bedding planes, at other times it lies oblique to the bedding planes, depending on the direc tion of the pressure at the time of recrystallisation. c. Gneiss. When rock is coarsely recrystallised it is termed 'gneiss' (pro nounced 'nice'). The crystal size may be as coarse as igneous rocks such as granite, but gneiss is distinguished from granite by banding, which may reflect the relic structure of bedding planes and by the fact that the crystals are preferentially orientated in one direction, as opposed to the crystals in granite, which grow slowly from small nuclei in the liquid magmas and have random orientation. d. Marble. When limestones are metamorphosed, they recrystallise to form marble. The streaks often seen in true marble are caused either by the local separation of original impurities in the limestone, excluded on recrystallisa tion of the calcite, or else by the cracking or brecciation of the limestone under stress. This geological usage of 'marble' is more restricted than the commercial use, which incorporates any kind of limestone used for orna mental work under the term, whether metamorphosed or not. It is even sometimes erroneously used to describe any sort of polished stone, perhaps granite or even slate. 1 1 . Rocks may be slightly affected by hydrothermal activity which is the action of water circulating deep enough to become first hot and then often superheated under pressure, penetrating the pore spaces of rocks and acting as a lubricant in joints. Occasionally water reaches the surface in the form of hot springs, which are usually of local occurrence and not necessarily associated with active vol canic areas. Geysers, which are hot fountains of waters ejecting at sporadic intervals, are usually confined to volcanically active areas. SECTION 3 . 4 . H O W R O C K S M A Y B E C H A N G E D THE SURFACE NEAR 1 . Rock decay with little or no transport of the products is termed 'weathering'; when the rock is simultaneously removed this is termed 'erosion'. The effects of weathering on a rock may be considered on the large scale and on the local scale. The large scale is discussed in Chapter 5 in the description of landforms, which also covers erosion. The local scale is also discussed in Chapter 5 under the forma tion of soils. 2. All rocks near the surface are affected to a greater or lesser degree by weathering. The principal types of weathering are: a. Surface weathering. The zone of surface weathering varies greatly, from less than 1 mm in some rocks to perhaps as much as 2 0 0 metres in another. Climatic conditions affect the depth of weathering; in tropical climates the 40 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved. Chapter 3 zone of weathering tends to be deeper than in temperate climates, but there is also a difference in weathering between extremes of climate. Weathering is achieved firstly by physical means, in which rocks may be shattered by temperature changes, by gravity effects, and by the pressure of growing organisms in cracks; and secondly by chemical means, in which the individual minerals that make up a rock are dissolved or decomposed by the water, oxygen and carbon dioxide of the atmosphere, and also by the chemicals produced by live or decaying organisms. The usual effect is to make a weathered zone structurally much weaker from the engineering point of view, and thus in all site work it is most important firstly to differentiate between weathered and unweathered rock, and secondly to try to find out the depth of weathering, which may vary between different parts of the site. b. Weathering below the surface. Joint systems occur in most types of rock, both igneous and sedimentary, and joint planes make natural zones of attack along which and down which water and chemicals in solution can penetrate. Limestones are particularly prone to subsurface weathering of this sort, since calcium carbonate is water soluble, and large caves and channels can form underground. Well-known examples include the Cheddar Caves, England, and the Carlsbad Caverns, USA. Such caves in limestones are an engineering hazard, since the building of any heavy surface structure can cause the sudden collapse of cave roofs. 3. Diagenesis is the name given to the processes which alter the character of a sedimentary rock after it has been deposited, either by the reactions between the various constituent minerals and particles with each other, or by the reactions of the various constituents with pore or circulating fluids. Such reactions are termed diagenetic at the lower temperature range, and metamorphic at the higher temperature range, but there is no rigid division between the two. Diagenesis occurs at two principal times, firstly when the original deposit is still in contact with sea or lake water soon after the time of formation, and secondly, after this period, when such direct contact with the original water has been removed. Diagenesis usually goes on until the constituents and pore fluids are all in chemical equilibrium. 4. Water also percolates through all types of rock, firstly through pores in the rock itself and secondly along fault zones and other surfaces, such as bedding planes and unconformities, which can lead to planes of weakness in apparently massive rock. Professional geological advice should always be sought in regard to large civil engineering works. REFERENCE LIST—CHAPTER 3 BARAZANGI M and DORM AN J, 1969 — World Seismicity Map of ESSA Coast and Geodetic Survey epicenter data for 1961-1967. Bulletin of the Seismological Society of America, Volume 59, No. 1. 41 Downloaded by [ Griffith University] on [25/10/17]. Copyright © ICE Publishing, all rights reserved.