12.07.2015 Views

mineralogical and fabric characterization and classification of ...

mineralogical and fabric characterization and classification of ...

mineralogical and fabric characterization and classification of ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

MINERALOGICAL ANDFABRIC CHARACTERIZATIONAND CLASSIFICATIONOF WEATHERED VOLCANICROCKS IN HONG KONGGEO REPORT No. 66T.Y. IrfanGEOTECHNICAL ENGINEERING OFFICECIVIL ENGINEERING DEPARTMENTTHE GOVERNMENT OF THE HONG KONGSPECIAL ADMINISTRATIVE REGION


MINERALOGICAL ANDFABRIC CHARACTERIZATIONAND CLASSIFICATIONOF WEATHERED VOLCANICROCKS IN HONG KONGGEO REPORT No. 66T.Y. ManThis report was originally produced in February 1997as GEO Spedal Project Report No. SPR 1/97


© The Government <strong>of</strong> the Hong Kong Special Administrative RegionFirst published, July 1998Prepared by :Geotechnical Engineering Office,Civil Engineering Department,Civil Engineering Building,101 Princess Margaret Road,Homantin, Kowloon,Hong Kong.This publication is available from :Government Publications Centre,Ground Floor, Low Block,Queensway Government Offices,66 Queensway,Hong Kong.Overseas orders should be placed with :Publications Sales Office,Information Services Department,28th Floor, Siu On Centre,188 Lockhart Road, Wan Chai,Hong Kong.Price in Hong Kong : HK$106Price overseas : US$17 (including surface postage)An additional bank charge <strong>of</strong> HK$50 or US$6.50 is required per cheque made in currenciesother than Hong Kong dollars.Cheques, bank drafts or money orders must be made payable toThe Government <strong>of</strong> the Hong Kong Special Administrative Region


- 3 -PREFACEIn keeping with our policy <strong>of</strong> releasing information,we make available some <strong>of</strong> our internal reports in a series <strong>of</strong>publications termed the GEO Report series. The reports inthis series, <strong>of</strong> which this is one, are selected from a widerange <strong>of</strong> reports produced by the staff <strong>of</strong> the Office <strong>and</strong> ourconsultants. A charge is made to cover the cost <strong>of</strong> printing.The Geotechnical Engineering Office also publishesguidance documents <strong>and</strong> presents the results <strong>of</strong> researchwork <strong>of</strong> general interest in GEO Publications. Thesepublications <strong>and</strong> the GEO Reports may be obtained from theGovernment's Information Services Department.Information on how to purchase these publications is givenon the last page <strong>of</strong> this report.R.K.S. ChanPrincipal Government Geotechnical EngineerJuly 1998


FOREWORDAs part <strong>of</strong> its research <strong>and</strong> development programme, theGeotechnical Engineering Office has been looking into ways <strong>of</strong>determining <strong>and</strong> characterising the <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong>properties <strong>of</strong> weathered rocks, in particular the soil grades, with theobjective <strong>of</strong> underst<strong>and</strong>ing their behaviour in the laboratory <strong>and</strong> inthe field.The results <strong>of</strong> the research on weathered granitic rocks havealready been published in a GEO Report by Irfan (1996a). Thisdocument summarizes the study results on the <strong>mineralogical</strong> <strong>and</strong><strong>fabric</strong> properties <strong>of</strong> the weathered volcanic rocks, in particular, thetuffs.The results <strong>of</strong> the studies on weathered granitic <strong>and</strong> volcanicrocks have already been used by the GEO in drafting local st<strong>and</strong>ardsfor soil <strong>classification</strong> <strong>and</strong> index tests based on BS1377:1990 version(see Chen, 1993). The results <strong>of</strong> the study on the rock grades arebeing incorporated into local st<strong>and</strong>ards for rock tests being preparedby the GEO.The report has been written by Dr T. Y. Irfan who alsoinitiated the study, conducted the research <strong>and</strong> carried out most <strong>of</strong>the in-house <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong> analyses, <strong>and</strong> coordinated <strong>and</strong>arranged for testing at overseas institutions <strong>and</strong> laboratories. Thechemical <strong>and</strong> X-ray diffraction analyses were carried out by theBritish Geological Survey <strong>and</strong> at three commercial laboratories inAustralia.Dr S.D.G. Campbell <strong>and</strong> Dr R. J. Sewell reviewed the draftdocument.( J. B. Massey )Government Geotechnical Engineer/Development


- 5 -ABSTRACTAs part <strong>of</strong> its research <strong>and</strong> development programme, theGeotechnical Engineering Office has been looking into ways <strong>of</strong>determining <strong>and</strong> characterising the <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong>properties <strong>of</strong> weathered rocks, in particular the soil grades, with theobjective <strong>of</strong> underst<strong>and</strong>ing their behaviour in the laboratory <strong>and</strong> inthe field.The results <strong>of</strong> the research on weathered granitic rocks havealready been described by Irfan (1996a). This document summarizesthe study results on the weathered volcanic rocks, in particular, thetuffs. It documents the typical chemical, <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong>changes that occur in these rocks due to weathering <strong>and</strong> otheralteration effects.Based on the results <strong>of</strong> the study, a model for weathering <strong>of</strong>volcanic rocks is established. The effects <strong>of</strong> hydrothermal alterationwhich are a major influence in pr<strong>of</strong>ile development <strong>and</strong> soilformation in Hong Kong are also considered. Modification, similarto those recently proposed for granites (Irfan 1996a), are proposedto the material <strong>and</strong> mass weathering schemes commonly adopted inHong Kong for the <strong>characterization</strong> <strong>of</strong> volcanic works in engineeringuses.The results <strong>of</strong> the studies on weathered granitic <strong>and</strong> volcanicrocks have already been used by the GEO in drafting local st<strong>and</strong>ardsfor soil <strong>classification</strong> <strong>and</strong> index tests.


- 6 -CONTENTSPageNo.Title Page 1PREFACE 3FOREWORD 4ABSTRACT 5CONTENTS 61. INTRODUCTION 92. GEOLOGICAL SETTING OF VOLCANIC ROCKS IN HONG KONG 92.1 General 92.2 Late-Stage Alteration <strong>and</strong> Metamorphism 102.3 Structure 103. WEATHERING PROFILE DEVELOPMENT AND ENGINEERING 10CLASSIFICATION OF WEATHERED VOLCANIC ROCKS3.1 Weathering Pr<strong>of</strong>ile in Volcanic Rocks 103.2 Weathering Grade Classification 113.2.1 Classification <strong>of</strong> the Weathered Rock Mass 113.2.2 Classification <strong>of</strong> the Weathered Rock Material 124. METHODS OF CHARACTERIZATION OF WEATHERED AND 12ALTERED ROCKS AND THE WEATHERING INDICES5. PREVIOUS WORK ON HONG KONG WEATHERED VOLCANIC 13ROCKS6. METHODS OF THE PRESENT STUDY 137. DESCRIPTION OF SAMPLING LOCATIONS AND WEATHERING 14CHARACTERISTICS7.1 Fine Ash Tuff, Ap Tsai Wan, Tseung Kwan O 147.2 Fine Ash Tuff, Tseung Kwan O 147.3 Coarse Ash Tuff, Sai Kung 15


- 7 -PageNo.8.7.4 Coarse Ash Tuff, Aberdeen7.5 Coarse Ash Tuff, Yuen LongDESCRIPTION OF MATERIAL WEATHERING GRADES IN TUFFS8.1 General8.2 Fresh Rock8.3 Slightly Decomposed8.4 Moderately Decomposed8.5 Highly Decomposed8.6 Completely Decomposed8.7 Transition Soil8.8 Residual Soil15161616161717171718189. MINERALOGY AND FABRIC CHARACTERIZATION OF 18WEATHERED TUFFS9.1 General 189.2 Fresh Rock 209.3 Slightly Decomposed 209.4 Moderately Decomposed 229.5 Highly Decomposed 239.6 Completely Decomposed 249.7 Transition <strong>and</strong> Residual Soil 269.8 Iron-oxide Minerals in Volcanic Saprolites <strong>and</strong> Residual Soils 279.9 Kaolin Veins 2810. CLASSMCATION OF WEATHERED VOLCANIC ROCKS AND 28SOILS IN HONG KONG10.1 Classification <strong>of</strong> Material Weathering Grades <strong>of</strong> Volcanic Rocks 2810.2 Classification <strong>of</strong> Volcanic Saprolitic <strong>and</strong> Residual Soils 2811. QUANTITATIVE WEATHERING INDICES 2911.1 General 2911.2 Chemical Indices 29


- 8 -PageNo.11.3 Degree <strong>of</strong> Decomposition Index 3011.4 Micropetrographic Indices 3012. CONCLUSIONS 3013. REFERENCES 32LIST OF TABLES 35LIST OF FIGURES 64LIST OF PLATES 84APPENDIX A : SAMPLE DESCRIPTIONS 97APPENDIX B : METHODS OF PETROGRAPHICAL AND CHEMICAL 106ANALYSIS AND IDENTIFICATION OF MINERALS


- 9 -1. INTRODUCTIONAs part <strong>of</strong> its research <strong>and</strong> development programme, the Geotechnical EngineeringOffice has been looking into ways <strong>of</strong> determining <strong>and</strong> characterizing the <strong>mineralogical</strong> <strong>and</strong><strong>fabric</strong> properties <strong>of</strong> weathered rocks, in particular the soil grades, with the objective <strong>of</strong>underst<strong>and</strong>ing their behaviour in the laboratory <strong>and</strong> in the field.A number <strong>of</strong> chemical, petrographical <strong>and</strong> other methods have been used to document<strong>and</strong> quantify the mineralogy <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> weathered igneous rocks, in particularthe granites, <strong>and</strong> volcanic rocks at each grade <strong>of</strong> weathering. These included chemicalanalyses, X-ray diffraction (XRD) analysis, optical microscopy, scanning <strong>and</strong> electronmicroscopy (SEM, TEM) <strong>and</strong> differential scanning calorimetry.The results <strong>of</strong> the study on the weathered granitic rocks have been described by Irfan(1996a, b). This document presents the results <strong>of</strong> the study on the weathered volcanic rocks.It documents the typical chemical, <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong> changes that occur in volcanicrocks due to weathering <strong>and</strong> other alteration effects, in particularly the tuffs which arewidespread in Hong Kong.The index <strong>and</strong> engineering properties <strong>of</strong> weathered volcanic rocks are not discussed here.A review <strong>of</strong> the mineralogy <strong>and</strong> structure <strong>of</strong> saprolitic <strong>and</strong> residual soils <strong>of</strong> volcanic <strong>and</strong>granitic rock origin is given in Irfan (1996b, c), who also discusses the effects <strong>of</strong> specificmineralogy <strong>and</strong> structure developed in these materials on the <strong>classification</strong> <strong>and</strong> indexproperties <strong>and</strong> behaviour.2* GEOLOGICAL SETTING OF VOLCANIC ROCKS IN HONG KONG2.1 GeneralGranites, granodiorites <strong>and</strong> the associated volcanic rock suite <strong>of</strong> Mesozoic age underliea greater part <strong>of</strong> the Territory <strong>of</strong> Hong Kong (Figure 1). The volcanic rocks <strong>of</strong> Jurassic toCretaceous age, several thous<strong>and</strong> metres thick, have been intruded by a number <strong>of</strong> graniticplutons <strong>of</strong> slightly younger age.The volcanic rocks, recently grouped by the Hong Kong Geological Survey under theyounger Repulse Bay <strong>and</strong> the older Tsuen Wan volcanic groups, comprise a series <strong>of</strong>lithostratigraphic formations, each with its own distinctive geological characteristics (see forexample, Langford et al. 1995). The volcanic rocks consist mainly <strong>of</strong> tuffs <strong>of</strong> pyroclasticorigin derived from ash fall <strong>and</strong> ash flow deposits with some lavas <strong>and</strong> intercalatedsedimentary horizons. The rocks are generally rhyolitic (acidic) in composition with some<strong>and</strong>esitic rocks occurring locally. Macroscopic features such as flow-b<strong>and</strong>ing, flow-folding,auto-brecciation <strong>and</strong> columnar jointing are present in the lavas as well as in the tuffs. Therecent <strong>classification</strong> <strong>of</strong> pyroclastic rocks has been in terms <strong>of</strong> grain size <strong>and</strong> physicalcomponents (Figure 2). Fine ash vitric tuffs <strong>and</strong> coarse ash crystal tuffs are the dominantrock types occurring in most formations. Other lithologies including tuff breccia, lapilli tuff,<strong>and</strong>esitic lavas <strong>and</strong> tuffites (tuffaceous sedimentary rocks), also occur as mappable units.More detailed information on the mineralogy <strong>and</strong> textural features <strong>of</strong> the common volcanicrock types is given in Section 9 <strong>and</strong> in the Geological Survey Memoirs published by GEO.


- 10 -2.2 Late-Stage Alteration <strong>and</strong> MetamorphismVolcanic rocks show local evidence <strong>of</strong> contact metamorphism close to granitic plutons.The effect <strong>of</strong> metamorphism is most intense within a few metres <strong>of</strong> the contacts, resulting indestruction <strong>and</strong> recrystallization <strong>of</strong> the original tuff texture <strong>and</strong> formation <strong>of</strong> hornblende-topyroxene-hornfels. However, in most cases, the volcanic rocks near granite contacts arehardly affected. In the Lo Wu-Tuen Mun area, the regional low to medium-grade metaorphismassociated with overthrusting <strong>of</strong> Palaeozoic basement onto Jurassic volcanic rocksalso affected the volcanic rocks resulting in mylonitization <strong>and</strong> formation <strong>of</strong> a schistosity inthe rocks (Langford et al, 1989). Volcanic rocks also show effects <strong>of</strong> metasomatic <strong>and</strong>thermal alteration (hydrothermal alteration) particularly in the vicinity <strong>of</strong> granitic plutons.2.3 StructureThe granites <strong>and</strong> volcanic rocks are affected by faulting dating back to the Mcsozoic,the faults being rejuvenated during each subsequent orogeny. The most dominant <strong>and</strong>persistent fault directions in the Territory are ENE-WSW <strong>and</strong> the most regionally extensivestructures trend NE-SW <strong>and</strong> NW-SE.Some large-scale gentle <strong>and</strong> open folding has been recognised in volcanic rocksoverlying granitic plutons (Strange et al, 1990). These are at least in part related to the rise<strong>of</strong> the underlying granitic plutons into a relatively competent volcanic cover. Two majorcollapsed caldera-type structures have been postulated in the Sai Kung area (Strange et al,1990), <strong>and</strong> a further caldera on Lantau (Langford et al, 1995), which resulted in roughlyhorizontal or gently dipping volcanic strata at the centres.Some <strong>of</strong> the joint sets in the volcanic rocks are clearly related to those formed ingranites. Additional joints sets are also developed in volcanic rocks related to folding <strong>and</strong>faulting due to the intrusion <strong>of</strong> granites, metamorphism, <strong>and</strong> the cooling <strong>and</strong> erosional history.Strong foliation or schistosity is also present in some <strong>of</strong> the volcanic rocks, in particular inthe Northwest New Territories, associated with overthrusting. The stress-release related gentlydipping joints are generally poorly developed or non-existent in volcanic rocks. The jointspacing is usually much closer than in the granitic rocks, except perhaps in some crystal tuffs,being generally between 0.1 m to 0.6 m. However, the mean joint spacing <strong>and</strong> also thespacing <strong>of</strong> individual sets may vary from one locality to another, being closer in the vicinity<strong>of</strong> faults <strong>and</strong> fault zones.3- WEATHERING PROFILEDEVELOPMENT AND ENGINEERINGCLASSIFICATIONOF WEATHERED VOLCANIC ROCKS3.1 Weathering Pr<strong>of</strong>ile in Volcanic RocksThe majority <strong>of</strong> studies on rock weathering <strong>and</strong> pr<strong>of</strong>ile descriptions have been confinedto the granitoid rocks in Hong Kong. The volcanic rocks have received less attention <strong>and</strong>only brief references have been made to pr<strong>of</strong>ile development in these rocks (Powell & Irian,1987; Irfan et al, 1987; GCO, 1988; Irfan, 1994). Recently, Irfan (1996a, b) reviewed thecommon weathering processes responsible for deep weathering pr<strong>of</strong>iles which have developedover the igneous <strong>and</strong> volcanic rocks in Hong Kong.


- 11 -In general, soil zones <strong>of</strong> the weathering pr<strong>of</strong>ile are comparatively shallow in volcanicrocks, usually less than 15 m thick, in contrast to the deep pr<strong>of</strong>iles that occur over thegranitoid rocks. However, thick soil pr<strong>of</strong>iles <strong>of</strong> up to 35 m have also developed, particularlyin the coarser grained volcanic rocks (Figure 3). Deeper soil pr<strong>of</strong>iles have been recorded, butthese are considered to be confined to faults <strong>and</strong> closely fractured zones. Corestonedevelopment, which is characteristic <strong>of</strong> the medium- <strong>and</strong> coarse-grained granites, is notcommon in weathering pr<strong>of</strong>iles developed over volcanic rocks, except in coarse ash tuffs <strong>and</strong>lapilli tuffs (Plates 1 <strong>and</strong> 2).Cataclasis <strong>and</strong> shearing <strong>of</strong> the rocks by tectonic processes <strong>and</strong> granite intrusions <strong>and</strong> theassociated hydrothermal alteration are a major influence in pr<strong>of</strong>ile development <strong>and</strong> soilformation at many locations in Hong Kong. Hydrothermal alteration generally results inweakening <strong>of</strong> rocks <strong>and</strong> deeper penetration <strong>of</strong> weathering (Irfan, 1996a; Irfan & Woods,1997). Some volcanic rocks are more resistant to weathering (e.g. lavas) than others, resultingin well-defined escarpments where resistant rocks cap <strong>and</strong> protect the s<strong>of</strong>ter more easilyweathered tuffs. For example, in the western New Territories, the slightly metamorphosedtuffs form thin weathering pr<strong>of</strong>iles with corestones, but strongly metamorphosed tuffsgenerally form resistant ridges (Langford et al, 1990).The general weathering pr<strong>of</strong>ile in lithologically uniform volcanic rocks is similar to that<strong>of</strong> granitoid rocks, <strong>and</strong> consists <strong>of</strong> four major layers grading from residual soil at the top t<strong>of</strong>resh rock at depth. In many localities in Hong Kong, the transitional rock <strong>and</strong> soil zone isusually relatively thin, except in coarse-grained volcanic rocks which have a similar pr<strong>of</strong>iledevelopment to medium to coarse-grained granites (Plate 1). In bedded sequences <strong>of</strong> varyinglithology, the rock mass may consist <strong>of</strong> relatively unweathered b<strong>and</strong>s or beds alternating withdecomposed b<strong>and</strong>s or beds. In some cases, the soil layer may be underlain by the fresh rock,with intermediate layers not present (Plate 3).The residual soil layer over the volcanic rocks is not very well developed, as is the casewith granites also. It is generally underlain by the transitional soil layer containing relict rock<strong>fabric</strong> but no structures. This zone is up to 3 m thick at many locations even on gently tomoderately sloping ground.3.2 Weathering Grade Classification3.2.1 Classification <strong>of</strong> Weathered Rock MassAs the weathering pr<strong>of</strong>iles developed over the volcanic rocks are similar to those <strong>of</strong> thegranitic rocks, the same weathering grade <strong>classification</strong> scheme is generally applicable to bothrock types. The six-fold mass grading scheme recommended by BS 5930 (BSI, 1981) hasalso found favour for classifying various components <strong>of</strong> the weathering pr<strong>of</strong>ile developed overvolcanic rocks (Powell & Irfan, 1987; GCO, 1990; Irfan, 1994) (see Table 1). The zonalscheme proposed by GCO (1988) is particularly suitable for corestone-forming, generallycoarse-grained volcanic rocks (e.g. coarse ash tuffs). However, it is less well suited forfine-grained rocks which show gradual loss <strong>of</strong> strength with weathering without forming anysignificant soil material.The schemes proposed by GCO (1988) <strong>and</strong> BSI (1981) <strong>and</strong> the most recent <strong>classification</strong>schemes proposed by the Geological Society <strong>of</strong> London (1995) fail to cater adequately in


- 12 -characterizing soil zones <strong>of</strong> the weathering pr<strong>of</strong>ile for engineering purposes, A <strong>classification</strong>scheme for soils formed from weathered rocks in tropical areas has been proposed by theGeological Society <strong>of</strong> London (1990). However, this scheme is complex <strong>and</strong> uses many termsunfamiliar to geotechnical engineers. Also, it does not account for the structuralcharacteristics <strong>of</strong> saprolitic <strong>and</strong> residual soil layers, that are incorporated in the simpler<strong>classification</strong> scheme proposed by Wesley & Irfan (1996). A brief description <strong>of</strong> this schemetogether with a critique <strong>of</strong> various mass weathering <strong>classification</strong> schemes applicable to rocksin Hong Kong is given in Irfan (1996b).3.2.2 Classification <strong>of</strong> Weathered Rock MaterialTraditionally, a six-fold decomposition grade scheme has been used in Hong Kong toclassify the weathered state <strong>of</strong> rock material (GCO, 1988). This scheme is generallyapplicable to most volcanic rocks. Various authors have suggested general characteristicsbased on observation or simple tests for assessment <strong>of</strong> decomposition grades, in particular ingranitic <strong>and</strong> volcanic rocks (Hencher & Martin, 1982; GCO, 1988; Irfan, 1988, 1996b).The most accurate method <strong>of</strong> describing the degree <strong>of</strong> decomposition, or weathering ingeneral, is to use a quantitative index. Assessment <strong>of</strong> decomposition grade should besupplemented by description <strong>of</strong> the state <strong>of</strong> disintegration <strong>of</strong> the rock material as disintegrationis also an intrinsic component <strong>of</strong> weathering. The terms used for description <strong>of</strong>decomposition grades <strong>of</strong> rock material may be used for hydrothermally altered rocks also asdecomposition is a general term used for chemical changes.Irfan (1996b) subdivided the 'completely decomposed' granite grade into; (i) completelydecomposed, (ii) (completely) disintegrated, <strong>and</strong> (iii) (completely) altered, based on thedominant type <strong>of</strong> weathering <strong>and</strong> alteration process. He also proposed the term 'transitionsoil* for material that shows partial loss <strong>of</strong> <strong>fabric</strong> before it transforms into true residual soil.These subgrades also appear to apply to volcanic rocks. Further division <strong>of</strong> each subgradecan be made, as necessary, based on relative strength (density or consistency) <strong>of</strong> the soilmaterial, which is a function <strong>of</strong> its degree <strong>of</strong> weathering <strong>and</strong> alteration.4- METHODS OF CHARACTERIZATION OF WEATHERED AND ALTERED ROCKSAND THE WEATHERING INDICESThere have been several attempts to quantify the degree <strong>of</strong> weathering <strong>and</strong> the weatheredrock <strong>fabric</strong>s using chemical, petrographical, X-ray diffraction <strong>and</strong> electron microscopemethods, or by simple field <strong>and</strong> laboratory mechanical <strong>and</strong> physical index tests. Although inmost cases, the quantitative <strong>and</strong> semi-quantitative weathering indices have been derived forquantifying changes in purely geological properties, some quantitative weathering indices havebeen used to relate <strong>and</strong> indirectly to determine the engineering properties <strong>of</strong> weathered rocks(e.g. Dearman & Irfan, 1978). Many <strong>of</strong> the weathering indices have been specificallydeveloped for characterizing coarse-grained igneous rocks, <strong>and</strong> particularly granites. Somequantitative weathering indices have also been applied to fine-grained igneous rocks includingvolcanic rocks (Weinert, 1964; Dearman et al, 1987).


- 13 -Irfan (1996a, b) reviewed the various weathering indices proposed to date <strong>and</strong> theirsuitability for granitic rocks in Hong Kong. The suitability <strong>of</strong> selective weathering indicesfor volcanic rocks is discussed in Section 10.5. PREVIOUS WORK ON HONG KONG WEATHERED VOLCANIC ROCKSLittle published data exist on the clay mineralogy <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> weatheredvolcanic rocks in Hong Kong, <strong>and</strong> these are confined to only a few samples collected froma limited number <strong>of</strong> rock types (Brock, 1943; Parham, 1969; Lumb & Lee, 1975). There isalso little <strong>mineralogical</strong> <strong>and</strong> chemical data from other countries on the weathering aspects <strong>of</strong>older volcanic rocks. In order to underst<strong>and</strong> the laboratory <strong>and</strong> field behaviour <strong>of</strong> weatheredrocks, in particular the soil grades, the basic <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> these rocks<strong>and</strong> soils need to be characterized. However, characterising these properties for engineeringpurposes is not easy because <strong>of</strong> their great variability <strong>and</strong> inhomogeneity, particularly in thecase <strong>of</strong> volcanic rocks, as a result <strong>of</strong> their geological <strong>and</strong> weathering history.The earlier studies indicated that the most abundant clay minerals in the near-surfacesoils <strong>of</strong> volcanic rock origin are the kaolin group <strong>of</strong> minerals. Both kaolinite <strong>and</strong> halloysiteare formed mainly from the decomposition <strong>of</strong> feldspars, which are abundant in these rocks.Halloysite is more common in saprolitic soils, co-existing with a minor quantity <strong>of</strong> kaolinite.This is to be expected as these soils have been formed on well-drained sideslopes <strong>of</strong> themountainous terrain in Hong Kong under a prolonged subtropical climate. Halloysite <strong>and</strong>kaolinite may give way to gibbsite under advanced weathering conditions close to the top <strong>of</strong>the weathering pr<strong>of</strong>ile, particularly in residual soils (Lumb & Lee, 1975; Irfan, 1996b). Asmall amount <strong>of</strong> the swelling mineral vermiculite may also be present (Parham, 1969).Allophane has been identified in volcanic rocks, at least in the initial stages <strong>of</strong> decomposition<strong>of</strong> feldspars, but it then rapidly changes to halloysite with further decomposition (Lumb &Lee, 1975). Recently, Irfan (1994) published the results <strong>of</strong> chemical, <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong>analyses <strong>of</strong> coarse ash tuffs from a l<strong>and</strong>slide site in Hong Kong. The results have beenincorporated into this document.6. METHODS OF THE PRESENT STUDYThe study undertaken on the volcanic rocks was less comprehensive than that in asimilar study <strong>of</strong> granitic rocks (Irfan, 1996a). The principal aim <strong>of</strong> the study is to establishthe basic mineralogy <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> selected weathered rock types, mainly the tuffs.The following methods were used to determine the mineralogy <strong>and</strong> micro<strong>fabric</strong> <strong>of</strong> the volcanicrocks:(a)(b)Detailed description <strong>of</strong> h<strong>and</strong> specimens.Optical microscopy on thin sections to determine <strong>and</strong> quantifythe mineralogy <strong>and</strong> <strong>fabric</strong> properties (e.g. grain boundaryrelationships, voids, weathering state <strong>of</strong> the mineralconstituents <strong>and</strong> their arrangement).


- 14 -(c) Chemical analyses, to determine chemical <strong>and</strong> norm<strong>mineralogical</strong> composition by X-ray fluorescence (XRF),inductively coupled plasma atomic emission spectrometrv(ICPAES) or atomic absorption spectroscopy (AAS). Loss onignition (LOI) <strong>and</strong> moisture determinations were also made.(d)X-ray diffraction (XRD) analysis, on bulk samples <strong>and</strong> clayfractions.Brief description <strong>of</strong> the methods <strong>of</strong> analysis used by the four laboratories which carriedout the analyses under items (c) <strong>and</strong> (d) is given in Appendix B. The comments by these fourlaboratories <strong>and</strong> the author on the accuracy <strong>of</strong> identification <strong>of</strong> various minerals from thetechniques used are also given in Appendix B.7. DESCRIPTION OF SAMPLING LOCATIONS AND WHM1I11IMCHARACTERISTICS7.1 Fine Ash Tuff, Ap Tsai Wan, Tsueng Kwan 0The sampling site is situated on a ridge at Ap Tsai Wan headl<strong>and</strong>, north <strong>of</strong> the town <strong>of</strong>Tseung Kwan O (Figure 4). The headl<strong>and</strong> is composed <strong>of</strong> fine ash tuff, eutaxite <strong>and</strong> tuffbreccia <strong>of</strong> the Ap Lei Chau Formation. The site has recently been formed into a series <strong>of</strong> cutslopes.The saprolitic layer is about 15 m thick on the site but the residual soil layer is missing,at least along the ridge line. There is a regular layering <strong>of</strong> the mass weathering grades at thetruncated headl<strong>and</strong> with the core <strong>of</strong> the hill composed <strong>of</strong> fresh to slightly weathered fine ashtuff <strong>and</strong> eutaxite. The regular layering is disturbed by fault zones where weatheringpenetrates to greater depth (Plate 4).H<strong>and</strong> <strong>and</strong> block samples were collected from the different material grades at the sitewith the more weathered specimens obtained from higher levels in the cut slopes. Briefdescriptions <strong>of</strong> h<strong>and</strong> samples are given in Appendix A.7.2 Fine Ash Tuff, Tseung Kwan 0The sampling site is situated at an elevation <strong>of</strong> about 130 mPD on the mid-slopes <strong>of</strong> TaiSheung Tok (elevation 400 m), west <strong>of</strong> Tseung Kwan 0 (Plate 5). Volcanic soil from thissite was used to assess the effect <strong>of</strong> matric suction on direct shear strength <strong>of</strong> Hon« Kongsoils by Gan & Fredlund (1992).The site is underlain by fine ash vitric tuff <strong>of</strong> the Ap Lei Chau formation. Theweathering pr<strong>of</strong>ile is variable <strong>and</strong> relatively thin at the site (Figure 5). Only the upper portion<strong>of</strong> the weathering pr<strong>of</strong>ile is exposed along the 5 m high road cutting where the samples weretaken. A layer <strong>of</strong> mottled brown <strong>and</strong> yellowish brown residual soil is present in some placesunderneath a thin colluvium layer. The transition from residual soil to saprolitic soil isirregular but generally occurs within a metre from the ground surface. The presence <strong>of</strong> whitekaolin veins along certain discontinuities within the rock may indicate hydrothermal alteration.


Corestones are present but poorly developed.- 15 -Three block samples were used for the direct shear strength research <strong>and</strong> <strong>mineralogical</strong><strong>and</strong> micro<strong>fabric</strong> studies were conducted on h<strong>and</strong> specimens representing different weatheringgrades. Brief descriptions <strong>of</strong> the samples are given in Appendix A.7.3 Coarse Ash Tuff, Sai KungThe sampling site is situated in a disused borrow area on top <strong>of</strong> a small isolated hill(elevation <strong>of</strong> 60 mPD), on a small peninsula in Sai Kung (Figure 1). The site is underlainby coarse ash crystal tuff, originally mapped as belonging to the Tai Mo Shan Formation byStrange et al (1990), now referred to as Sai Kung Formation.The thickness <strong>of</strong> saprolitic soil at the site is not known. However, it is considered tobe over 10 m thick, based on the coastal exposures <strong>of</strong> weathered rock. The residual soil layeris thin, less than 1 m, <strong>and</strong> irregularly developed. The rock shows typical spheroidal type <strong>of</strong>weathering commonly associated with the coarse-grained tuffs in the Territory (Plate 6).Numerous corestones, usually less than 0.5 m in diameter, are present in the saprolitic soil <strong>and</strong>occasionally in the residual soil.A limited number <strong>of</strong> specimens were collected from the saprolitic <strong>and</strong> residual soilzones. Fresh to moderately decomposed samples were obtained from the corestones in thesaprolitic soil. Brief descriptions <strong>of</strong> the samples are given in Appendix A.7.4 Coarse Ash Tuff, AberdeenThe sampling site was originally a minor ridge crest on the southern footslopes <strong>of</strong> anortheast-southwest trending major ridge in Aberdeen. Borrowing activity in the 1920sresulted in removal <strong>of</strong> the colluvium cover <strong>and</strong> top few metres <strong>of</strong> the saprolitic-residual soillayer <strong>and</strong> the formation <strong>of</strong> two dish-shaped areas.A shear plane, formed at a depth <strong>of</strong> about 4 m in the saprolitic soil as a result <strong>of</strong>continuous creep, acted as the release surface for a l<strong>and</strong>slip in 1988. A petrographical study<strong>of</strong> the weathered rocks in the slope was undertaken to determine the causes <strong>of</strong> the creepingbehaviour (Irfan, 1992, 1994).The coarse ash tuff is deeply weathered on the site, with the saprolitic soil zone 15-20 mthick (Figure 6). Numerous white clayey veins criss-cross the saprolite. The layer above thepre-existing shear plane consisted <strong>of</strong> s<strong>of</strong>t mottled volcanic soil with a deformed tuff <strong>fabric</strong>.A thin transition soil layer was observed above the saprolitic soil.Samples for the <strong>mineralogical</strong> study were collected from various soil materials on thesite as well as the less weathered rock grades recovered as cores from the boreholes (Plate 7).Brief descriptions <strong>of</strong> the samples are given in Appendix A.


- 16 -7.5 Coarse Ash Tuff, Yuen LongThe sampling site is a newly formed cut slope adjacent to a haul road in I*ai long Eastborrow area, south <strong>of</strong> Yuen Long (Figure 1). The slope, 30 to 50 ni high with a gradient <strong>of</strong>about 30°, was formed in early 1991 by cutting into the natural hillside. The slope wassubject to instability soon after cutting in June 1991. The instability is confined to a thinlayer <strong>of</strong> volcanic rocks which overlie a fine-grained granite.The volcanic rock at the site, mapped as undifferentiated tuff <strong>and</strong> tuffite on the newgeological map, is a lapilli-bearing coarse ash crystal tuff <strong>of</strong> the Shing Mun Formation. Therock is rich in quartz crystals, up to 5 mm in diameter, <strong>and</strong> mica (up to 10%) but poor infeldspar crystals. It is probably contact-metamorphosed by the granite.The tuff above the granite contact is completely to highly weathered. The granite siiowsless intense weathering, at least where it is exposed along the cutting. Occasional corestonesare present in the saprolite but these are not well developed unlike other coarse ash tuffs inthe Territory. The topmost layer <strong>of</strong> the saprolitic zone shows a highly deformed <strong>fabric</strong> <strong>and</strong>structure where it has been affected by recent slope movements.Only five samples were collected from the site, with two each from the highlydecomposed <strong>and</strong> completely decomposed tuff <strong>and</strong> one A fresh' sample from inside a corestone.Details <strong>of</strong> the samples are given in Appendix A.8. DESCRIPTION OF MATERIAL WEATHERING GRADES INjnJFFS8.1 GeneralThe material weathering grade scheme adopted in this document follows Irian (1996a).However, subdivisions based on relative strength values have not been attempted for thesevolcanic rocks due to lack <strong>of</strong> laboratory <strong>and</strong> insitu data. Plates 8 to 10 illustrate the variousmaterial grades in fine ash <strong>and</strong> coarse ash tuffs.8.2 Fresh RockFresh Fine Ash Tuffs are usually medium grey to dark grey but also light grey or black,<strong>and</strong> extremely strong. These rocks generally consist <strong>of</strong> a small amount <strong>of</strong> euhedral tosubhedral broken crystals <strong>of</strong> feldspars or quartz or both <strong>and</strong> occasional vitric fragments in avery fine vitric (glassy) matrix. If fiamme are present, then the rock is called 4 eutaxite\ Thematrix may be finely recrystallized, particularly if the rock is metamorphosed.Fresh Coarse Ash Tuffs are usually bluish grey to dark grey <strong>and</strong> very strong toextremely strong. They generally comprise abundant crystals <strong>of</strong> feldspars (up to 3-4 mm insize) with some quartz <strong>and</strong> biotite <strong>and</strong> occasional lithic fragments in a fine-grained vitric torecrystallized matrix. Welding <strong>fabric</strong> may be present in the matrix.


- 17 -8.3 Slightly DecomposedSubdivisions can be made within this grade for fine <strong>and</strong> coarse ash tuffs (<strong>and</strong> forvolcanic rocks in general) based on the degree <strong>of</strong> discoloration (Plates 8 to 10), similar to thesubgrades suggested for granitic rocks. Discoloration is usually an indication <strong>of</strong> weathering.Slightly Discoloured: The rock material is discoloured less than 10% by volume fromjoint <strong>and</strong> fracture surfaces inwards to yellowish grey to yellowish brown. The interior <strong>of</strong> thejoint-bounded block retains its original grey to dark grey colour.Moderately Discoloured: The rock material is discoloured 10% to 50% by volume toyellowish grey to light grey with dark brown in the vicinity <strong>of</strong> fractures <strong>and</strong> joints. Thefresher centre may show very light grey discoloration.Highly Discoloured: The rock material is discoloured more than 50% by volume.Colour b<strong>and</strong>ing may be present with medium grey fresher centres surrounded, in turn, by lightyellowish grey, yellowish grey <strong>and</strong> dark brown rims. The same concentric colour b<strong>and</strong>ingmay also be present in corestones which are commonly present in the saprolitic zone in coarseash tuffs, with the outermost layers being highly to completely decomposed. In thecorestones, feldspars become progressively s<strong>of</strong>ter (more decomposed) towards the exterior.There is also development <strong>of</strong> concentric micr<strong>of</strong>ractures in the discoloured rim in both coarseash <strong>and</strong> fine ash tuffs, which increases in intensity towards the exterior <strong>of</strong> a corestone or ajoint-bounded block in this subgrade.8.4 Moderately DecomposedThe rock material is completely discoloured, light yellowish to yellowish greenish grey.The joint surfaces may be stained reddish brown to dark brown. The rock is very strongwhen least weathered to moderately weak when most weathered within this grade. Somefeldspar crystals are gritty, particularly in the most weathered tuff specimens. Near jointsurfaces they may be gritty to s<strong>of</strong>t. Biotite, if present, is dull greenish grey <strong>and</strong> s<strong>of</strong>t. Therock <strong>fabric</strong> is fractured throughout by a network <strong>of</strong> tight but iron-oxide stained microcracks.8.5 Highly DecomposedThe rock material is completely discoloured, light yellowish grey or greyish yellow withyellowish brown near joints <strong>and</strong> fractures. The strength is very much reduced due toextensive microcracks. Feldspar crystals are largely gritty, some hard (in the weak rockrange) to powdery <strong>and</strong> occasionally s<strong>of</strong>t (in the extremely weak rock range).8.6 Completely DecomposedThis grade <strong>of</strong> weathered tuff could be divided into subgrades similar to those for graniticrocks, based on the type <strong>and</strong> degree <strong>of</strong> alteration <strong>and</strong> weathering <strong>and</strong> the relative density(strength), but this has not been attempted here.


- 18 -The rock material is discoloured throughout to yellowish grey, yellowish greenish greyor reddish brown, <strong>and</strong> it can also be mottled. Pinkish grey <strong>and</strong> white patches can be presentlocally if it is affected by hydrothermal alteration, usually in the vicinity <strong>of</strong> kaolin orkaolin-quartz veins. Differential weathering may be present between the lithic fragments <strong>and</strong>the matrix in lapilli-bearing tuffs. Feldspar crystals are usually powdery to s<strong>of</strong>t with degree<strong>of</strong> decomposition being more intense than in the hydrothermally altered areas. Quartz, ifpresent, is dull <strong>and</strong> biotite is transformed into colourless to silvery grey flakes. Thecompletely decomposed specimens can be easily disaggregated into soil particles by fingerpressure.8.7 Transition SoilTransition soil specimens are usually dark reddish brown, but may also he mottled withpatches <strong>of</strong> yellowish grey. The soil shows partial loss <strong>of</strong> the tuff <strong>fabric</strong>, the proportion <strong>of</strong>which increases up the weathering pr<strong>of</strong>ile. Relict discontinuities are largely absent in thetransition soil zone. Occasional partially decomposed feldspar crystals may be present in areasstill retaining the original tuff <strong>fabric</strong>.8.8 Residual SoilThe specimens are usually yellowish brown, but may also be yellowish grey whenbleached <strong>of</strong> its iron minerals. The soil comprises some s<strong>and</strong>-size quartz grains in astructureless clayey silt to silty clay matrix. Nodules or pods <strong>of</strong> iron-oxide cemented soilaggregations may be locally present. In extreme cases, cementation by iron-oxides <strong>and</strong>hydroxides may produce a weak rock (i.e. a laterite).9. MINERALOGY AND FABRIC CHARACTERIZATION OF WEATHERED TUFFS9.1 GeneralThe textural <strong>and</strong> <strong>mineralogical</strong> properties <strong>of</strong> volcanic rocks in Hong Kong are veryvariable depending upon the type <strong>of</strong> volcanic activity <strong>and</strong> subsequent processes which formedthese rocks, the mode <strong>and</strong> history <strong>of</strong> solidification <strong>and</strong> later metamorphism, hydrothermal <strong>and</strong>other alteration processes. As a result, wide variations in texture <strong>and</strong> mineralogy can evenoccur in the same rock type at each location (Table 2).Coarse ash tuffs occur in many formations. In general, they consist <strong>of</strong> crystals or crystalfragments <strong>of</strong> feldspar <strong>and</strong> quartz, up to 6 mm in size, with small biotite <strong>and</strong> occasionalhornblende <strong>and</strong> secondary rnuscovite set in a fine-grained matrix <strong>of</strong> the same minerals. Clasts<strong>of</strong> sedimentary or igneous rocks <strong>of</strong> various sizes may also be present. The matrix may bepartly or wholly recrystallized, particularly in the vicinity <strong>of</strong> granite intrusions. Some crystalalignment <strong>and</strong> weak to strong welding <strong>fabric</strong> with elongate recrystallized siliceous shards <strong>and</strong>fiamme may be present. Crystal tuffs <strong>of</strong> the Shing Mun Formation are particularly variablein grain size <strong>and</strong> texture <strong>and</strong> can show rapid variation both laterally <strong>and</strong> vertically from coarseash to fine ash to lapilli tuffs. The total crystal content can vary from 20% to over 50%.


- 19 -Feldspar crystals are particularly abundant in Tai Mo Shan <strong>and</strong> Yim Tin Tsai tuffs.Quartz grains may be rounded to broken <strong>and</strong> angular, occasionally resorbed with no clearlydefined grain boundaries. Biotite occurs as individual large grains as well as smallaggregations scattered throughout the groundmass. It is usually partially chloritised. Alkalifeldspars may be cloudy <strong>and</strong> partly replaced by epidote. Plagioclases may be slightlysericitized even in the fresh rock. These <strong>mineralogical</strong> changes are probably the result <strong>of</strong> subaerialprocesses during the formation <strong>and</strong> solidification <strong>of</strong> the rock as well as the laterhydrothermal activity from intruding granites.Fine ash tuffs including eutaxites are particularly common on Hong Kong Isl<strong>and</strong> <strong>and</strong> inthe vicinity <strong>of</strong> Junk Bay <strong>and</strong> Sai Kung. They generally consist <strong>of</strong> a small amount <strong>of</strong> scatteredcrystals (up to 20% <strong>of</strong> the rock), dominantly quartz with some feldspars <strong>and</strong> occasional lithicfragments, in a very fine vitric (glassy) matrix (Strange & Shaw, 1986). The matrix may befinely recrystallized, particularly in the vicinity <strong>of</strong> igneous intrusions. In the eutaxites, awelding compaction structure arising out <strong>of</strong> alignment <strong>of</strong> flattened fiamme, stretched pumice<strong>and</strong> lithic fragments is present. Majority <strong>of</strong> crystals are usually broken <strong>and</strong> <strong>of</strong>ten resorbedwith in distinct grain boundaries.The crystal-bearing fine ash welded tuffs which occur in the vicinity <strong>of</strong> High Isl<strong>and</strong>display well-developed columnar jointing (Strange et al, 1990). The quartz <strong>and</strong> feldsparcrystals comprise up to 20% <strong>of</strong> the rock in a very fine-grained or vitric groundmass. Quartzcrystals are invariably corroded <strong>and</strong> resorbed. Feldspars occur as euhedral grains or brokencrystal fragments. The fine ash tuffs <strong>of</strong> the Silverstr<strong>and</strong> Formation have prominent eutaxitic<strong>fabric</strong>. Fiamme is usually abundant comprising up to 30% <strong>of</strong> the rock. Lenses <strong>of</strong>sedimentary rocks <strong>and</strong> chert fragments may be abundant at some localities.In the following sections, a brief summary is given <strong>of</strong> the <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong>properties <strong>of</strong> the volcanic rocks weathered to different degrees, with particular emphasis onthe fine ash <strong>and</strong> coarse ash tuffs as these were the only rocks studied in detail in this study.The <strong>mineralogical</strong> compositions <strong>of</strong> the freshest specimens from the five rock types aregiven in Table 3, with the results <strong>of</strong> quantitative petrographical determinations shown inTables 4 <strong>and</strong> 5. The results <strong>of</strong> chemical analyses, arranged in order <strong>of</strong> weathering grade, aregiven in Tables 6 to 8. The results <strong>of</strong> XRD analyses on bulk <strong>and</strong> clay fractions are presentedin Tables 9 to 14 in terms <strong>of</strong> the relative abundances <strong>of</strong> minerals identified from the traces.Figures 7 to 10 show the typical XRD traces for selected specimens. The bulk normmineralogies calculated from the results <strong>of</strong> chemical analysis using the mineral types identifiedby XRD are given in Tables 15 to 17. Schematic representations <strong>of</strong> the <strong>mineralogical</strong> changesoccurring with weathering in each <strong>of</strong> the five tuffs are given in Figures 11 to 15. Plates 11<strong>and</strong> 12 are micrographs illustrating the typical <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> fine <strong>and</strong>coarse ash tuffs weathered to various degrees.Quantitative petrographic analyses were not carried out for the finer tuffs due to thedifficulty in recognising the minerals <strong>and</strong> voids, particularly in the weathered specimens.Comments on the identification <strong>of</strong> minerals from the XRD analysis <strong>and</strong> on the accuracy <strong>of</strong>the chemical analyses results are contained in Appendix B.


- 20 -9.2 Fresh RockThe fine <strong>and</strong> coarse ash tuffs included in the study are all rhyolitic in composition. Themajor constituents are feldspars <strong>and</strong> quartz with some mica occurring as large euhedralcrystals or crystal fragments in a devitrified or recrystallized groundmass <strong>of</strong> the sameminerals. Occasionally, lithic fragments are also present. The crystal content reaches over50% in the case <strong>of</strong> coarse ash tuff from Aberdeen, but in the fine ash tuffs it is usually lessthan 25% <strong>of</strong> the rock (Table 4). Feldspars are the dominant constituents with plagioclasebeing more abundant than alkali feldspar. Overall, quartz contents vary between 25% <strong>and</strong>45%. However, the quartz crystal contents are usually low, less than a few percent, exceptin the coarse ash tuff from Aberdeen it reaches 15% <strong>of</strong> the rock. The chemical analysis <strong>and</strong>optical microscopy revealed that the specimen ATWl from Ap Tsai Wan has a significantlydifferent chemical <strong>and</strong> <strong>mineralogical</strong> composition than the other specimens collected from thesame site (Table 8).Biotite, when present, is mostly partially to completely altered to a mixture <strong>of</strong> hydrousmica, chlorite, iron oxides <strong>and</strong> sometimes to vermiculite. The mica content <strong>of</strong> the coarse ashtuffs from Yuen Long is rather high (about 10%). Small amounts <strong>of</strong> secondary minerals suchas muscovite, hydrous mica, chlorite, occasionally calcite, <strong>and</strong> in sonic cases, kaolinite areinvariably present in addition to minor amounts <strong>of</strong> iron-oxide minerals, mainly magnetite.The majority <strong>of</strong> secondary products is attributed to probable effects <strong>of</strong> epigenetic processesduring or subsequent to volcanism <strong>and</strong> solidification. Later effects <strong>of</strong> contact metamorphism<strong>and</strong> hydrothermal alteration by the intruding granitic magmas might also have contributed tothe formation <strong>of</strong> micaceous minerals in these rocks.In thin section, most plagioclases <strong>and</strong> some alkali feldspars show slight to moderatealteration to micaceous products (commonly termed sericite) in all five tuffs. Calcite alsooccurs in some crystals <strong>and</strong> also in the groundmass, possibly as a byproduct <strong>of</strong> breakdown <strong>of</strong>Na-Ca plagioclase to micaceous minerals <strong>and</strong> kaolinite. Fine-grained micaceous minerals <strong>and</strong>chlorite also occur in the groundmass. The fresh rock is unfractured except for occasionalshort <strong>and</strong> tight microcracks in quartz <strong>and</strong> feldspar crystals.The tuff from Aberdeen has the coarsest grain size, with quartz <strong>and</strong> feldspar crystals,up to 4 mm in diameter. The fine ash tuff from Ap Tsai Wan has a strong eutaxitic foliation.A faint flow <strong>fabric</strong> is also present in the tuff from Tseung Kwan 0.9.3 Slightly DecomposedNoticeable decreases occur in Na 4 * <strong>and</strong> Ca 2+ contents which indicate the commencement<strong>of</strong> decomposition <strong>of</strong> Na - Ca feldspars (plagioclases) in the slightly decomposed tuff grade.The behaviour <strong>of</strong> K + , Si 4+ <strong>and</strong> Al 3+ are not so predictable <strong>and</strong>, in some cases, these cationsshow increase or decrease against the expected trend with increased weathering. For example,fC <strong>and</strong> Si 4+ contents are higher than those <strong>of</strong> the fresh rock in some <strong>of</strong> the slightlydecomposed specimens (Table 8). These variations are attributed to variations in the chemical<strong>and</strong> <strong>mineralogical</strong> composition <strong>of</strong> the original rock within each sampling site including thosearising from hydrothermal <strong>and</strong> other alteration processes.The carbonate minerals, which may be present in small amounts in fresh tuffs, are


- 21 -largely removed by solution in this grade as indicated by significant reduction in CO 2 <strong>and</strong>Ca 2+ contents, although most <strong>of</strong> the reduction in the latter is probably due to decomposition<strong>of</strong> Ca-bearing feldspars. The optical microscopy revealed void formation within somefeldspars <strong>and</strong> in the groundmass at the locations <strong>of</strong> calcite, especially in the stained portions<strong>of</strong> the specimens.In thin section, a small amount <strong>of</strong> feldspar decomposition, mainly to micaceous minerals,is present even in the fresher portions <strong>of</strong> the specimens. The degree <strong>of</strong> decomposition <strong>of</strong>feldspars increases in the stained rims, especially in plagioclase megacrysts. There arevariations in the type <strong>and</strong> intensity <strong>of</strong> feldspar weathering amongst the tuffs. For example,all feldspar megacrysts show slight to moderate degrees <strong>of</strong> decomposition in the tuffs fromSai Kung, Tseung Kwan O <strong>and</strong> Ap Tsai Wan as reflected in their calculated mineralogies(Tables 16 <strong>and</strong> 17), but the plagioclases are only slightly decomposed <strong>and</strong> the alkali feldsparsare not generally affected by weathering in the coarse ash tuffs from Aberdeen. Themoderately <strong>and</strong> highly decomposed specimens show an increase in the calculated alkalifeldspar content (Table 15). This is considered to be mainly due to the higher alkali feldsparcontents <strong>of</strong> these specimens.In the stained rims, some areas within the partially altered feldspars appear 'cloudy' or'speckled'. These could be poorly crystalline alumina-silicate minerals transitional tokaolinite or halloysite. They might also be the locations <strong>of</strong> solution pits indicating the onset<strong>of</strong> solution weathering (Dearman & Baynes, 1978; Irfan, 1996b). Slight to noticeable solution<strong>of</strong> feldspars was observed by Kwong (1988) in slightly to moderately decomposed volcanicrocks from Hong Kong Isl<strong>and</strong> using electron microscopy. Reddish brown staining <strong>of</strong> the rock<strong>fabric</strong> by iron-oxides without any apparent micr<strong>of</strong>racturing might also be taken as evidence<strong>of</strong> solution in the feldspar megacrysts <strong>and</strong> also the groundmass.Biotite, shows partial to complete alteration to a mixture <strong>of</strong> secondary products includinghydrous mica, chlorite, iron-oxides <strong>and</strong> sometimes vermiculite. The trace amounts <strong>of</strong> chlorite,hydrobiotite <strong>and</strong> vermiculite or intermixed chlorite-vermiculite identified by XRD in sometuffs are considered to be formed from the alteration <strong>of</strong> biotite. The presence <strong>of</strong> chlorite <strong>and</strong>vermiculite also occurring in the unweathered rock suggests that these minerals are largelyformed by hydrothermal processes although vermiculite <strong>and</strong> hydrobiotite can also form fromweathering <strong>of</strong> biotite in well-drained conditions in humid climates (Gilkes & Suddhiprakaran,1981).The feldspars in the devitrified groundmass also show increased decomposition withincreased staining where the groundmass assumes a semi-opaque appearance under themicroscope. In the case <strong>of</strong> tuffs with eutaxitic texture, staining is more intense along b<strong>and</strong>sthat also show more intense alteration.Trace to small amounts <strong>of</strong> kaolin minerals were identified by XRD in the discolouredportions <strong>of</strong> specimens in some tuffs (Tables 10 <strong>and</strong> 11). Kaolin minerals are also present intbe-feesh rock or unstained portion <strong>of</strong> some <strong>of</strong> the specimens, which suggests that the majority<strong>of</strong> kaolin minerals present in the fresh <strong>and</strong> slightly decomposed tuffs are formed bypre-weathering processes such as hydrothermal alteration. The predominant kaolin mineralis kaolinite with only trace amounts <strong>of</strong> halloysite in the more weathered specimens <strong>of</strong> theslightly decomposed grade. The halloysite content increases in the moderately <strong>and</strong> highlydecomposed grades, suggesting that this mineral is formed directly from the decomposition


- 22 -<strong>of</strong> feldspars in intermediate stages <strong>of</strong> weathering. The increase in kaolin content in thediscoloured portions <strong>and</strong> in the moderately decomposed specimens is mostly derived from thealteration <strong>of</strong> plagioclases as the alkali feldspars do not show any significant petrographicchange.A trace amount <strong>of</strong> a possible smectite mineral was indicated in the coarse ash tuff fromAberdeen in addition to minor amounts <strong>of</strong> chlorite <strong>and</strong> vemiiculite or an intermediate mineralbetween the two (Table 9). The hydrothermal alteration <strong>of</strong> the rock might have resulted inthe formation <strong>of</strong> metastable smectite in the initial stages <strong>of</strong> feldspar alteration, as well asformation <strong>of</strong> chlorite <strong>and</strong> vermiculite from alteration <strong>of</strong> biotite (Man, 1994).The total amount <strong>of</strong> secondary minerals varies, up to 12% or even more in the slightlydecomposed tuffs. The proportion <strong>of</strong> microcracks <strong>and</strong> pores is very low, usually less than 1% by volume.A network <strong>of</strong> simple-branched <strong>and</strong> tight but mostly iron-oxide infilled or coatedmicrocracks is developed in the stained portions. The frequency <strong>of</strong> micr<strong>of</strong>racturing increasestowards the joint surfaces. The boundaries <strong>of</strong> quartz <strong>and</strong> feldspar megacrysts are tight in thefresher cores but they may be coated with iron-oxides in the stained rims. The staining orcoating along grain contacts may indicate the development <strong>of</strong> intergranular fracturing or theformation <strong>of</strong> dissolution pits <strong>and</strong> channels around the exterior <strong>of</strong> the larger feldspar grains,as identified by Kwong (1985).9.4 Moderately DecomposedRapid reductions occur in Na + <strong>and</strong> Ca 2+ contents with the exception <strong>of</strong> the coarse ashtuff from Yuen Long (Tables 6 to 8), indicating a significant decomposition <strong>of</strong> plagioclase.The associated increases in clay mineral contents are reflected in LOI values which are almostdouble those <strong>of</strong> the slightly decomposed tuff. These changes are confirmed by XRD results,which show that the kaolin mineral content in this grade is increased significantly from a fewpercent to up to 20% or more (Tables 15 to 17). No consistent change is apparent in K +content indicating that the alkali feldspars are not yet significantly affected by weathering inthis grade.Although there is a sharp drop in Na + content as the rock weathers from the slightlydecomposed to the moderately decomposed state, the Ca" content is unaffected in coarse ashtuff from Yuen Long (Table 7). This might be due to the presence <strong>of</strong> calcite as indicated bythe high CO 2 content or absence <strong>of</strong> Ca- bearing feldspars in this rock type.Optical microscopy revealed that the majority <strong>of</strong> large plagioclase grains are decomposedto kaolin <strong>and</strong> micaceous minerals, with alkali feldspars only slightly affected but generallyhighly micr<strong>of</strong>ractured. The feldspars in the fine-grained to cryptocrystalline groundmass alsoappear to be affected by weathering in this grade. Numerous isotropic speckles were observedin many large feldspars under the microscope, similar to those identified in granites (Man,1996b). These are probably either the locations <strong>of</strong> solution pits <strong>and</strong> trenches observed byKwong (1985) <strong>and</strong> Dearman & Baynes (1978), or some intermediate alumina-silicate growthstransitional to kaolin minerals.


- 23 -The main kaolin mineral is a poorly crystalline kaolinite with a possible trace amount<strong>of</strong> halloysite only detected in the tuff from Tseung Kwan O. Parham (1969) also reported thegrowth <strong>of</strong> flame - shaped films <strong>and</strong> halloysite tubes on feldspars in some slightly tomoderately decomposed rhyolite- porphry specimens (probably a coarse ash tuff).In addition to kaolin minerals, small amounts <strong>of</strong> sesquioxides, vermiculite <strong>and</strong> chloritewere identified in some rocks from the XRD patterns. A trace amount <strong>of</strong> a smectite mineral,or an interstratified clay mineral <strong>of</strong> possible illite-smectite composition, was also detected inthe tuff from Aberdeen. The quantitative modal <strong>and</strong> <strong>mineralogical</strong> analyses show that the clay<strong>and</strong> micaceous mineral content can be up to 25% in this grade (Tables 15 to 17).The most significant change in the rock is the formation <strong>of</strong> an extensive network <strong>of</strong> tightto slightly open microcracks in this grade. As a result, a fairly rapid reduction in intactstrength occurs across the grade. Most large feldspars <strong>and</strong> some quartz grain boundaries arestained which indicate that the grain boundaries are opening up. Apart from the pores formedat the locations <strong>of</strong> dissolved calcite grains in some specimens, solution features have not beenobserved under the microscope.9.5 Highly DecomposedFairly rapid changes occur in the major oxides contents within the highly decomposedgrade (Tables 6 to 8), indicating more intense decomposition <strong>and</strong> more rapid variation <strong>of</strong> theclay mineral content <strong>of</strong> the volcanic rocks in this grade than in the moderately decomposedgrade. Both Na + <strong>and</strong> Ca 2+ contents are reduced to very small values in all rock types exceptthe coarse tuff from Yuen Long, with complete disappearance <strong>of</strong> plagioclase in the moreintensely weathered specimens as confirmed by XRD <strong>and</strong> optical microscopy. A noticeablereduction in K + content in all rock types except the tuff from Aberdeen indicates that alkalifeldspars are now being affected by weathering within this grade. The Si 4+ content also showsa noticeable reduction except in some hydrothermally altered specimens (Tables 7 <strong>and</strong> 8).Both feldspar types show variations in their degree <strong>and</strong> type <strong>of</strong> alteration. Plagioclasecontents are less than 3% in the fine ash tuffs, but can be up to 17% in the coarse ash tufffrom Yuen Long. Although plagioclases show near-complete alteration, the type <strong>of</strong> clay<strong>fabric</strong>s formed can be different in each rock. For example, the decomposed feldsparpseudomorphs are very porous in the coarse ash tuff from Aberdeen in comparison to thedense clay <strong>fabric</strong>s produced in the coarse ash tuff from Sai Kung. The altered feldspars inthe specimens affected by hydrothermal alteration have, in general, more clayey, densermicro <strong>fabric</strong>s than those affected only by weathering.Alkali feldspars are little altered in the tuff from Aberdeen, but they are intenselymicr<strong>of</strong>ractured. In contrast, the tuffs from Sai Kung <strong>and</strong> Tseung Kwan 0 are highlydecomposed to clay minerals. The intensive solution <strong>of</strong> alkali felspars generally occurs in thecompletely decomposed grade (Plate 11). The decomposition <strong>of</strong> feldspars in the groundmassis more advanced in this grade than in moderately decomposed tuff, with specimensunaffected by hydrothermal alteration having more porous groundmass than those affected.The groundmass is particularly porous in the tuff from Aberdeen.Kaolin minerals are the dominant clay minerals, with small to accessory amounts <strong>of</strong>


- 24 -micas <strong>and</strong> trace amounts <strong>of</strong> chlorite <strong>and</strong> vermiculite or intermixed ehlorite-vermieulite in somerocks (Tables 15 to 18). Kaolin minerals in the fine ash tuffs form up to 40% o( the rock <strong>and</strong>even more in the altered specimens, but generally less than 25% in the coarse ash tuffs. Thecoarse ash tuffs have similar kaolin contents to granites (Irian, 19%a). The halloysite contentgenerally increases in this grade but not in all rock types. It forms up lo half <strong>of</strong> the kaolincontent in tuffs from Aberdeen <strong>and</strong> Ap Tsai (Tables 12 <strong>and</strong> 14). Halloysite was not detectedin tuffs from Sai Kung <strong>and</strong> Yuen Long. Due to the lack <strong>of</strong> detailed electron microscopestudies it is not known whether the halloysite in the coarse ash tuff from Aberdeen is formeddirectly from the decomposition <strong>of</strong> feldspars or through an intermediate mineral. Kaolinitemay form (i) by hydrothermal alteration <strong>of</strong> feldspars, (ii) by direct transformation o( feldspars,or (iii) through an intermediate alumina-silicate mineral. It is possible that all three sourceshave contributed to kaolinite formation in the weathered volcanic rocks. A small amount <strong>of</strong>kaolinite might also have formed from the decomposition <strong>of</strong> biotite. Trace amounts <strong>of</strong>vermiculite or an intermediate vermiculite-chlorite identified by XRI) in the tuff fromAberdeen. It is highly likely that it is formed from alteration o( biotite. Small amounts <strong>of</strong>sesquioxides occur in all rocks, generally as poorly crystalline aggregations.Thermogravimetric analysis indicates that these are mostly goethite.The rock <strong>fabric</strong> is extensively micro fractured by tight to slightly open mieroemeks,which together with voids comprise up to 20% <strong>of</strong> the rock in tuffs from Aberdeen {Table 5).Some disturbances <strong>of</strong> the fine-grainedgroundmass in the vicinity <strong>of</strong> the micro fractured quartz<strong>and</strong> feldspar grains indicate a slight expansion <strong>of</strong> the rock at this stage ol weathering.9.6 Completely DecomposedIn spite <strong>of</strong> the differences in their chemical <strong>and</strong> <strong>mineralogical</strong> compositions <strong>and</strong> textures,the completely decomposed fine <strong>and</strong> coarse ash tuffs show many similar <strong>mineralogical</strong> <strong>and</strong><strong>fabric</strong> characteristics in the completely decomposed grade. However, significant differencescan also occur, even in the same rock type, if the rock has been subjected to hydrothermal <strong>and</strong>other alteration processes prior to weathering. Variations in the present-day weatheringenvironment can lead to significant differences in void characteristics <strong>and</strong> clay mineralogy <strong>of</strong>the soil formed from the same rock at different locations.Plagioclases are completely decomposed in this grade in coarse ash tuffs from Aberdeen<strong>and</strong> Ap Tsai, whilst complete decomposition <strong>of</strong> plagioclases occur in the highly decomposedgrade in all other tuffs studied. The Ca 2+ <strong>and</strong> Na f contents are further reduced to very smallamounts compared with highly decomposed tuffs, indicating the removal <strong>of</strong> cations from thesoil, except in the tuff from Yuen Long where the Ca 2+ content is still appreciable (Table 7).This is attributed to the accumulation <strong>of</strong> Ca-bearing minerals (e.g. calcite) in the soil orabsorption <strong>of</strong> freed Ca 2+ by other mineral species rather than the presence <strong>of</strong> unaltered Cabearingplagioclases.Significant reduction in K + content in all completely decomposed tuffs compared withhighly decomposed tuffs indicates intense weathering <strong>of</strong> alkali feldspars within this grade(Tables 6 to 8). Nevertheless, there are variations in the way the alkali feldspars decompose(e.g. solution versus alteration to clay minerals), depending on the environment <strong>of</strong> weathering<strong>and</strong> the degree <strong>of</strong> hydrothermal alteration or contact metamorphism. For example, the tuffsunderlying gentle hillslope environments where active removal <strong>of</strong> cations by migrating


- 25 -groundwater occurs (e.g. Aberdeen coarse ash tuff, Tseung Kwan O fine ash tuff), showwidespread solution type weathering with advanced honeycomb <strong>fabric</strong> development in thecoarse ash tuffs (Irfan, 1994). The groundmass in these rocks also assumes a highly porousappearance.Although the Yuen Long site has also a relatively gentle hillside topography, thecompletely decomposed tuff specimens from this site do not have the highly porous <strong>fabric</strong>sdisplayed by Aberdeen tuffs. This might have been the result <strong>of</strong> contact metamorphism <strong>of</strong>the tuff by the underlying granite at this site, which had resulted in the formation <strong>of</strong> kaolinite<strong>and</strong> micaceous minerals in the fresh rock. Kaolinite, being a stable mineral would not besubject to solution type weathering, which usually results in the removal <strong>of</strong> cations,particularly from the feldspars. The tuffs from Aberdeen also show the effects <strong>of</strong>kaolinisation <strong>and</strong> sericitization due to hydrothermal alteration, but these appear to be restrictedto the vicinity <strong>of</strong> kaolin or kaolin-quartz veins (Irfan, 1994).The results <strong>of</strong> quantitative <strong>mineralogical</strong> analysis are in general agreement with theobservations made on thin sections. For example, the unaltered alkali feldspar content isreduced from about 13 to 19% <strong>and</strong> 27% in the highly decomposed tuffs from Ap Tsai <strong>and</strong>Aberdeen respectively, to less than 8% in the completely decomposed tuffs from both sites.The alkali feldspars in the completely decomposed specimens from Sai Kung, which has acoastal low hilltop environment, show chemical decomposition to clay mineral products(mainly kaolinite).Biotite usually occurs in a completely altered state comprising a mixture <strong>of</strong> fineaggregations <strong>of</strong> hydrous mica, chlorite, kaolin minerals <strong>and</strong> iron-oxides. The altered mineralstructure shows evidence <strong>of</strong> expansion <strong>and</strong>, in some cases, complete disaggregation intosmaller grains.The kaolin mineral content increases sharply to up to 50% in this grade <strong>and</strong> particularlyin the more intensely weathered specimens (Table 17). Halloysite becomes almost asabundant as kaolinite in the least weathered specimens in this grade, being particularlyabundant in the coarse ash tuff from Aberdeen (Table 12), concentrations being similar togranites (Irfan, 1996a, b). However, in the completely decomposed tuffs from Sai Kung <strong>and</strong>Yuen Long, only small amounts <strong>of</strong> halloysite were detected in the clay portions (Table 13).There may be a number <strong>of</strong> reasons for the dominance <strong>of</strong> kaolinite in these two rock types,some <strong>of</strong> which are given below :(i)(ii)(iii)The contribution from direct alteration <strong>of</strong> alkali feldspars tokaolinite. The alkali feldspars show widespread decompositionto kaolinite rather than solution type weathering in this grade.Direct formation <strong>of</strong> kaolinite from pre-weathering processessuch as contact metamorphism <strong>and</strong> hydrothermal alteration.The tuffs at both sites show evidence <strong>of</strong> contactmetamorphism, with the granite immediately underlying tuffat Yuen Long.Poor drainage in a low-lying coastal topography at Sai Kungmight also have inhibited the formation <strong>of</strong> halloysite.


- 26 -(iv)Most halloysite may be <strong>of</strong> silt size, as is the case with granites(Irfan, 1996b), <strong>and</strong> hence is not present in the clay portionsanalysed.In the more intensely weathered completely decomposed specimens <strong>and</strong> in the transitionsoil grade, kaolinite becomes dominant with only small amounts <strong>of</strong> halloysite (Tables 12 to14).The other alteration products include trace to small amounts <strong>of</strong> tine-grained micaceousminerals, sesquioxides <strong>and</strong>, in some rocks, small amounts <strong>of</strong> vermiculite or chlorite. Thechlorite content appears to decrease with increased weathering. Chlorite was not detected inthe completely decomposed <strong>and</strong> residual soil specimens from some sites. Gibbsite was alsonot identified in the XRD traces <strong>of</strong> any <strong>of</strong> the specimens in this grade, The micaceousproducts were identified by XRD as mostly illite in tuffs from Aberdeen based on the linebroadening <strong>and</strong> relative peak heights (Appendix B).The weathered rock shows extreme <strong>fabric</strong> variations in the completely decomposedgrade. In general, very fractured <strong>and</strong> porous <strong>fabric</strong> is developed in the coarse-grained tuffsfrom hillside environments. Generally, denser clayey <strong>fabric</strong>s are formed in the fine-grainedtuffs <strong>and</strong> those rocks previously affected by contact metamorphism or hydrothermal alteration.Poorly drained conditions, for example, in valley bottom type environments, may also leadto generally clayey <strong>fabric</strong>s. The cementation by iron-oxides may locally be very intense <strong>and</strong>so obliterates the soil <strong>fabric</strong>s. Organic matter may also be present in some specimens, asreflected in their higher than usual carbon contents.9.7 Transition <strong>and</strong> Residual SoilThe most significant chemical change is in K f content where it decreases appreciablywithin the residual soil stage, indicating intense decomposition <strong>of</strong> alkali feldspars. However,the K + content is still significant in the residual soil specimens <strong>of</strong> the fine ash tuff fromTseung Kwan 0 <strong>and</strong> the coarse ash tuff from Sai Kung (Tables 7 <strong>and</strong> 8). As the alkalifeldspars are completely decomposed, the major contribution to IC content is from kaolinite,which now forms over 50% <strong>of</strong> the total soil weight. The increased decomposition <strong>of</strong> alkalifeldspars to clay minerals indicates that chemical alteration <strong>of</strong> these feldspars takes place ata very advanced stage <strong>of</strong> weathering, subsequent to solution weathering in the initial <strong>and</strong>intermediate stages. The other important chemical change is in Si 44 content which shows anoticeable reduction for the first time in this grade indicating a noticeable solution <strong>of</strong> quartz.The XRD traces <strong>and</strong> semiquantitative calculations confirm some reduction in quartz contentin agreement with the microscopical observations (Tables 16 <strong>and</strong> 17). The reduction in silicaor quartz content may also be due to hydrothermal effects.Halloysite content decreases significantly in the transition soil <strong>and</strong> disappears completelyin the residual soil derived from tuffs. The halloysite in this advanced stage <strong>of</strong> weatheringwas identified as the dehydrated type. Lumb & Lee (1975) attributed the formation <strong>of</strong>dehydrated halloysite in near-surface volcanic soils to considerable seasonal desaturation,together with a much greater chance <strong>of</strong> oxidation resulting eventually in complete dehydrationto kaolinite.


- 27 -In addition to predominant kaolinite <strong>and</strong> trace amounts <strong>of</strong> halloysite, small amounts <strong>of</strong>micaceous minerals (illite <strong>and</strong> muscovite), trace chlorite <strong>and</strong>, in some soils, vermiculite <strong>and</strong>possibly interstratified clay minerals <strong>of</strong> illite-smectite composition are present. Smallquantities <strong>of</strong> gibbsite was reported by Lumb & Lee (1975) in near-surface volcanic soils <strong>and</strong>by Irfan (1996b) in residual granitic soils. Gibbsite has not been identified in any <strong>of</strong> thevolcanic soils analysed here. In some soils, small amounts <strong>of</strong> CO 2 were detected. This isattributed to the presence <strong>of</strong> organic matter in the specimens. There is a small increase in thetotal iron-oxide content, which forms up to 8% <strong>of</strong> the soil. The dominant iron mineralpresent in the volcanic soils was identified by thermogravimetric analysis as poorly crystallinegoethite with haematite in some specimens.The original rock <strong>fabric</strong> gradually disappears within the transition soil grade <strong>and</strong> theresidual soil comprises some silt-to s<strong>and</strong>-size quartz grains (usually less than 20% <strong>of</strong> the soil)in a structureless clay mineral-iron oxide mineral matrix. The coating <strong>and</strong> cementation byiron minerals are so intense, locally, that the details <strong>of</strong> soil <strong>fabric</strong> may be completelyobliterated. Occasional large pores, some <strong>of</strong> which are ant holes <strong>and</strong> decayed root locations,are present. Microcracks are much reduced in amount <strong>and</strong> most are secondary features, e.g.desiccation cracks. There may be small aggregations or 'pebbles 4 <strong>of</strong> reddish brown iron-oxideminerals throughout the soil <strong>fabric</strong> <strong>and</strong> these may be taken as an indication <strong>of</strong> commencement<strong>of</strong> laterization at some sites.9.8 Iron-oxide Minerals in Volcanic Saprolites <strong>and</strong> Residual SoilsThere are variations in the iron-oxide contents <strong>of</strong> the saprolitic <strong>and</strong> residual specimensanalysed, as reflected in their Fe 2 O 3 contents, depending on the degree <strong>of</strong> weathering <strong>and</strong> thelocation <strong>of</strong> the specimen within the weathering pr<strong>of</strong>ile. Generally, the iron-oxide content isless than 10%. However, it may be higher in areas <strong>of</strong> iron-oxide concentration, e.g. in thegroundwater fluctuation zone, as was the case with the Aberdeen l<strong>and</strong>slide site (Irfan, 1994).In thin section, iron minerals generally appear as poorly crystalline aggregationsscattered throughout the groundmass, but also as coatings to partially to completelydecomposed minerals, or as microcrack infillings. Locally, weak cementation is also present.Detailed studies on granitic rocks indicated that goethite is the main iron-oxide mineralformed via ferrihydrite or directly from the decomposition <strong>of</strong> biotite in the intermediate <strong>and</strong>advanced stages <strong>of</strong> weathering (Irfan, 1996a). This is probably the case for the coarse ashtuffs <strong>and</strong> most other volcanic rocks in Hong Kong. However, the lack <strong>of</strong> any weight lossbetween 200° <strong>and</strong> 400°C in the thermogravimetric analyses <strong>of</strong> specimens from Sai Kung <strong>and</strong>other volcanic rock sites (except Aberdeen) were interpreted as these rocks not containing anydetectable goethite or gibbsite by one <strong>of</strong> the laboratories (British Geological Survey).Haematite was considered to be the main iron mineral present in the weathered tuffs fromthese sites. The tuffs from Sai Kung contain some magnetite. Diffuse magnetite boundariesmay indicate the commencement <strong>of</strong> alteration to haematite in the moderately to highlydecomposed stage. Thermogravimetric analyses carried out for Aberdeen specimens indicatedthe presence <strong>of</strong> poorly crystalline goethite. In granites, Irfan (1996a) reported that thecrystallinity <strong>of</strong> goethite increased with weathering. This might also be the case for volcanicrocks.


- 28 -9.9 Kaolin VeinsThe XRD analysis <strong>of</strong> a number <strong>of</strong> kaolin veins taken from the l<strong>and</strong>slide sites on HongKong Isl<strong>and</strong> (Irfan & Woods, 1997) show that these veins contain dominantly kaolinite withsome illite <strong>and</strong> quartz <strong>and</strong>, in some cases, a smectite mineral (Table 18). Optical microscopyindicates that there may be an increase in the micaceous alteration products (mainly formedwithin the feldspars) adjacent to these veins <strong>and</strong> in the "pink feldspar' areas. This mayrepresent the situation described by Sales & Meyer (1940) that in the vicinity <strong>of</strong> hydrothermalveins in igneous rocks, a zonal arrangement <strong>of</strong> clay minerals may occur, with a sericitic micazone immediately adjacent to the vein followed by kaolinite <strong>and</strong> smectite - chlorite zones.Small amounts <strong>of</strong> swelling minerals (smectite) were identified in the weathered rock fromAberdeen, in addition to up to 15% smectite identified in some <strong>of</strong> the veins. It is alsopossible that smectite may have formed from weathering processes in a poorly drainedenvironment, but this is considered unlikely for the Aberdeen site.10. CLASSIFICATION OF WEATHERED VOLCANIC ROCKS AND SOILSJN^HONGKONG10.1 Classification <strong>of</strong> Material Weathering Grades <strong>of</strong> Volcanic RocksA <strong>characterization</strong> <strong>of</strong> the material grades <strong>of</strong> volcanic rocks (primarily the tuffs) in HongKong is given in Table 19 in terms <strong>of</strong> their <strong>mineralogical</strong> <strong>and</strong> micro<strong>fabric</strong> properties.10.2 Classification <strong>of</strong> Volcanic Saprolitic <strong>and</strong> Residual SoilsA <strong>classification</strong> <strong>of</strong> various soil types formed from volcanic <strong>and</strong> granitic rocks is givenin Table 20, together with a summary <strong>of</strong> their structural <strong>and</strong> <strong>mineralogical</strong> properties, inaccordance with the <strong>classification</strong> scheme proposed by Wesley & Irfan (1996). The soils <strong>of</strong>both granitic <strong>and</strong> volcanic origin fall into Group A, as structural (both micro <strong>and</strong> macrostructure)influences are dominant in controlling their behaviour. Thin layers or lenses <strong>of</strong>Group C soils rich in sesquioxides, may also form near ground surfaces, particularly in poorlydrained areas. The formation <strong>of</strong> the Group (C c ) soils in Hong Kong is considered to be <strong>of</strong>minor occurrence in comparison to those developed in tropical regions : these soils are looselyreferred to as lateritic or laterite.In volcanic saprolitic soils, the kaolinite content may be more dominant than halloysite;they may also contain appreciably more mica, some chlorite <strong>and</strong> small amounts <strong>of</strong> swellingminerals <strong>of</strong> vermiculite, smectite <strong>and</strong> layered minerals <strong>of</strong> illite-smectite composition.Kaolinite may also be more dominant in volcanic soils if the original rock had been affectedby hydrothermal <strong>and</strong> other alteration process prior to weathering.Almost all soil zones including the highly weathered volcanic rock zone in Hong Kongwould be classified as 'ferruginous (sensu stricto) soil* under the <strong>classification</strong> schemeproposed by the Geological Society (1990), except perhaps the uppermost residual soil zone,which would be described as 'ferralitic (kaolinitic)\


- 29 -11. QUANTITATIVE WEATHERING INDICES11.1 GeneralA brief review <strong>of</strong> the weathering indices proposed to assess the degree <strong>of</strong> weathering<strong>of</strong> rocks is given in Irfan (1996a, b).A large number <strong>of</strong> these weathering indices are derived from the results <strong>of</strong> chemicalanalyses. The selected chemical indices used to quantity the degree <strong>of</strong> weathering in volcanicrocks is given in Irfan (1996a). A number <strong>of</strong> petrographic indices were also determined fromthe results <strong>of</strong> the modal analysis, such as the percentages <strong>of</strong> altered <strong>and</strong> unaltered minerals,the micropetrographic index (Ip) following Irfan & Dearman (1978).As the physical <strong>and</strong> mechanical properties <strong>of</strong> the volcanic rocks studied were notavailable, the indices derived from these properties were not determined. Also no attempt wasmade to classify the weathered volcanic rocks in terms <strong>of</strong> their dry density, void ratio,Schmidt hammer values or SPT N values, similar to that done for the granites (Irfan, 1996b).11.2 Chemical IndicesThe chemical weathering indices calculated for the volcanic rocks are presented inTables 21 to 24. The mobiles index (I mob ) proposed by Irfan (1996a) serves as an index <strong>of</strong>the degree <strong>of</strong> decomposition <strong>of</strong> the feldspar-bearing rocks, particularly in reasonably welldrained(leaching) conditions, such as those found in most hillsides <strong>and</strong> hilltops in HongKong. The silica-to-alumina (SA) ratio has also been found to be a good index <strong>of</strong> chemicalweathering in most free draining, usually acid, weathering environments in humid climates,especially on acid <strong>and</strong> intermediate rocks that weather to kaolinite or illite group <strong>of</strong> minerals(Ruxton, 1968). Both indices are not considered to be suitable for weathering conditionswhich result in smectite or vermiculite minerals, especially for basic <strong>and</strong> ultrabasic rocks.These two indices <strong>and</strong> the others calculated from chemical analyses do not account for thephysical <strong>and</strong> <strong>fabric</strong> changes occurring with progressive weathering. It is the <strong>fabric</strong> propertieswhich influence the engineering properties most in weathered rocks (Massey et al, 1988; Irfan,1996b, c).The results in Tables 21 to 24 indicate that the weathering potential index (WPI),alumina-to-potassium-sodium (AKN) ratio, the loss-on-ignition (LOI) value <strong>and</strong> the mobilesindex are generally good indicators <strong>of</strong> the degree <strong>of</strong> weathering <strong>of</strong> volcanic rocks. Theseindices appear to be more suitable for characterizing the weathered materials formed over aparticular rock type at a particular weathering environment. The chemical composition <strong>of</strong>fresh volcanic rocks can vary significantly not only from formation to formation but alsowithin the same formation. In addition to chemical variation in the original rock, themineralogy <strong>and</strong> chemical composition <strong>of</strong> the weathered product are affected by theenvironment <strong>and</strong> history <strong>of</strong> weathering at each site as well as other alteration processes asdiscussed in Section 9. This is illustrated in Figure 16, which is a plot <strong>of</strong> silica-to-aluminaratio against mobiles index for the weathered coarse ash tuffs from Aberdeen <strong>and</strong> Sai Kung.Similarly, Figure 17 shows an almost linear relationship between WPI <strong>and</strong> I mob for the coarseash tuff from Aberdeen in contrast to a wide scatter <strong>of</strong> values shown in Figure 18 when thedata from all rocks are included.


- 30 -The alumina-to-calcium-sodium (ACN) ratio appears to be useful only for indexing therock grades <strong>of</strong> weathering <strong>and</strong> not sensitive to chemical changes occurring in the soil grades.The lixiviation index (jS) is not a good indicator <strong>of</strong> weathering even for the soil grades, whichit was originally proposed for by Rocha Filho et al (1985).11.3 Degree <strong>of</strong> Decomposition IndexThe degree <strong>of</strong> decomposition index (Lumb, 1965) nor the modified degree <strong>of</strong>decomposition index (Irfan, 1988) were determined for volcanic rocks as these indices wereconsidered inappropriate for fine-grained volcanic rocks, especially those with a high content<strong>of</strong> cryptocrystalline or microcrystalline feldspar <strong>and</strong> quartz. Also, as discussed by Irfan(1996b), the degree <strong>of</strong> decomposition index is not a very reliable indicator <strong>of</strong> feldspardecomposition <strong>and</strong> it does not take into account the structural <strong>fabric</strong> elements such as voids<strong>and</strong> microcracks.11-4 Micropetrographic IndicesPetrographical methods using an optical microscope <strong>and</strong> good quality thin sections arevery useful in studying weathering effects. Not only do they allow cheap <strong>and</strong> rapid mineralidentification but also the micro<strong>fabric</strong> properties <strong>of</strong> the materials can be studied <strong>and</strong>characterized for the whole range <strong>of</strong> the weathering spectrum.The point counting technique described by Irfan <strong>and</strong> Dearman (1978) was employed todetermine the modal compositions <strong>of</strong> the mineral constituents, voids <strong>and</strong> microcracks for theweathered coarse-grained crystal tuffs only. Because <strong>of</strong> the extremely fine-grained nature <strong>of</strong>the matrix minerals, no individual mineral determinations were made <strong>and</strong> these were lumpedtogether as the 'fresh matrix' or 'altered matrix' minerals. The results <strong>of</strong> quantitativedeterminations are presented in Table 5. It was not possible to differentiate between variousfine-grained alteration products <strong>and</strong> small pores in the more weathered specimens under thenormal resolution <strong>of</strong> an optical microscope. The results are, however, still useful as anappropriate means <strong>of</strong> quantifying the observed <strong>fabric</strong> changes occurring in each material.The technique was found inappropriate for fine ash tuffs <strong>and</strong> the coarse ash tuffs witha large cryptocrystalline mineral contents, especially when they were weathered beyond theslightly decomposed grade. Also, general iron-staining or cementation <strong>of</strong> the <strong>fabric</strong> inmoderately <strong>and</strong> more intensely decomposed volcanic rocks made the identification <strong>of</strong> mineralconstituents very difficult.12. CONCLUSIONSAlthough chemical weathering has been the dominant process in pr<strong>of</strong>ile development <strong>and</strong>soil formation in volcanic rocks in Hong Kong, cataclasis <strong>and</strong> shearing by tectonic processes<strong>and</strong> granite intrusions <strong>and</strong> the associated hydrothermal alteration have also been important.The general weathering pr<strong>of</strong>ile in lithologically uniform volcanic rocks is similar to that <strong>of</strong>granitoid rocks, although the soil zones are generally thinner <strong>and</strong> the transition from 'all soillayer' to 'all rock layer* is generally more gradual <strong>and</strong> less distinct.


- 31 -The recent <strong>classification</strong> scheme proposed by Wesley & Irfan (1996), which groups thesoils on the basis <strong>of</strong> their <strong>mineralogical</strong> <strong>and</strong> structural characteristics, appears to be suitablefor classifying the saprolitic <strong>and</strong> residual soils <strong>of</strong> volcanic rock origin.The six-fold material weathering grade scheme, based on that adopted by GeotechnicalEngineering Office (1988) but with further subdivisions <strong>and</strong> refinements that are proposedhere can be used to describe <strong>and</strong> classify the weathered volcanic rock materials encounteredwithin each broad soil group or rock mass zone.A large number <strong>of</strong> mechanical, physical, chemical <strong>and</strong> petrographical indices areproposed to quantify the degree <strong>of</strong> weathering <strong>and</strong> weathered rock <strong>fabric</strong>s <strong>and</strong> some <strong>of</strong> thesehave been used to relate <strong>and</strong> indirectly determine engineering properties <strong>of</strong> weathered rocks.Only the chemical <strong>and</strong> petrographical indices have been determined for weathered volcanicrocks due to lack <strong>of</strong> data on physical <strong>and</strong> mechanical properties <strong>of</strong> the rocks studied here.Most proposed chemical indices, in particular the mobiles index, the weathering potentialindex <strong>and</strong> the alumina-to potassium-ratio, have been found to be good indicators <strong>of</strong> the degree<strong>of</strong> chemical weathering in a particular volcanic rock type, but direct comparison <strong>of</strong> indexvalues at a particular weathering grade is not generally possible even for the same rockoccurring in different formations. This is due to variations in the chemical composition <strong>and</strong>mineralogy <strong>of</strong> the weathering products arising from variations in the fresh rock <strong>and</strong> theenvironment <strong>and</strong> history <strong>of</strong> weathering at each site. The micropetrographic indices are onlyappropriate for the coarse-grained crystal tuffs due to the difficulty in recognising various finemineral <strong>and</strong> <strong>fabric</strong> components under the microscope in fine-grained rocks, particularly thosewith vitrified <strong>and</strong> cryptocrystalline textures.Although only the mineralogy <strong>and</strong> <strong>fabric</strong> properties <strong>of</strong> the fine <strong>and</strong> coarse ash tuffs werestudied in detail, the results <strong>of</strong> the study <strong>and</strong> the general mineralogy <strong>and</strong> <strong>fabric</strong> characteristicsobserved in these rocks are considered to apply to other volcanic rocks in the Territory, exceptperhaps the more basic types.In spite <strong>of</strong> great variation in their mineralogy <strong>and</strong> chemical composition <strong>and</strong> textures,the volcanic rocks show many similar mineralogy <strong>and</strong> <strong>fabric</strong> characteristics in the way theyweather. However, significant differences can also occur, even in the same rock, if the rockhas been affected by metamorphism <strong>and</strong> hydrothermal or other alteration processes prior toweathering. The unweathered rock may contain small amounts <strong>of</strong> secondary alterationproducts, mostly micaceous minerals, but also in some cases, kaolinite, calcite, chlorite <strong>and</strong>vermiculite. These secondary minerals are attributed to effects <strong>of</strong> epigenetic processes duringvolcanism <strong>and</strong> solidification, as well as the effects <strong>of</strong> contact metamorphism <strong>and</strong> hydrothermalalteration by the intruding magmas. Variations in the present day weathering <strong>and</strong>topographical environment can also lead to significant differences in void characteristics <strong>and</strong>clay mineralogy <strong>of</strong> the soils formed from the same rock at different locations.In general, similar to granites, halloysite tends to form first, mainly from thedecomposition <strong>of</strong> feldspars, unless the rock has been affected by other alteration processes,in which case, kaolinite is dominant, in the early stages <strong>of</strong> weathering. Halloysite becomesthe dominant clay mineral with some muscovite/illite in the intermediate stages <strong>of</strong> weathering,which is then replaced by kaolinite in the near-surface saprolitic <strong>and</strong> residual soils. Smallamounts <strong>of</strong> gibbsite may also be present in near-surface residual soils <strong>of</strong> volcanic rock origin.


- 32 -Similar to granites, alkali feldspars in the tuffs are generally affected b\ weathering,initially by solution, much later than plagioclases, usually in the completely decomposedgrade. Alteration may commence earlier in the highly decomposed grade m some tuffs. Thetuffs underlying gentle hilisiope <strong>and</strong> hilltop environments, where ieachinti i»t cations bygroundwater occurs actively, show widespread solution type weathering ot' both the alkalifeldspars <strong>and</strong> the fine-grained matrix with advanced honeycomb <strong>fabric</strong> de\ el* »pmcnt Solutionweathering seems to be more common in the coarse ash crystal tu(t* than the fine ash tuffs.The kaolin veins <strong>of</strong> both weathering <strong>and</strong> hydrothermal or mixed origin arc common inthe saprolitic soil zone. Those <strong>of</strong> hydrothermal origin, as well as the weathered rock itself,may contain some smectite minerals.The presence <strong>of</strong> halloysite <strong>and</strong> sesquioxide minerals <strong>and</strong> the specific microstructuredeveloped in the soil grades, require that special precautions are taken in the preparation <strong>of</strong>volcanic soil specimens for conventional laboratory testing <strong>and</strong> in the interpretation <strong>of</strong> the testresults as discussed by Irfan (1996b, c).13. REFERENCESBaynes, F.J. & Dearman, W.R. (1978). The micro<strong>fabric</strong> <strong>of</strong> a chemically weathered granite.Bulletin <strong>of</strong> the International Association n <strong>of</strong> Engineering Geology, No. IS, pp 91-100.Brock, R.W. (1943). Weathering <strong>of</strong> igneous rocks near Hong Kong. Bulletin_<strong>of</strong>_theGeological Society <strong>of</strong> America, vol. 54, pp 717-738.BSI (1981). Code <strong>of</strong> Practice for Site Investigations, BS.,,S_93Q ; 1 981,Institution, London, 149 p.British St<strong>and</strong>ardsDearman, W.R. & Irfan, T.Y. (1978). Assessment <strong>of</strong> the degree <strong>of</strong> weathering in graniteusing petrographic <strong>and</strong> physical index tests. Intenia^[letcrkiTation<strong>and</strong> Protection <strong>of</strong> Stone Monuments, Unesco, Paris, Paper 2.3.Dearman, W.R., Turk, N., Irfan, T.Y. & Rowshanei, H. (l l )87). Detection <strong>of</strong>' rock materialvariation by sonic velocity zoning. Bulletin <strong>of</strong> the International Association jrfEngineering Geology, No. 35, pp 3-8.Gan, J.K. & Fredlund, D.G. (1992). Direct Shear Testing <strong>of</strong> a Hong Kong SoiJJLJnderVarious Applied Matric Suctions. Geotechnical Engineering Office, Hong Kong, 217 p.(GEO Report No. 11)GCO (1988). Guide to Rock <strong>and</strong> Soil Descriptions (Geoguide 3).Office, 189 p.Geotechnical ControlGCO (1990). Foundation Properties <strong>of</strong> Marble <strong>and</strong> Other Rocks in the Yuen Lonn-Tuen MunArea. Geotechnical Control Office, Hong Kong, 117 p, (GCO Publication No. 2/90).Geological Society <strong>of</strong> London (1990). Tropical residual soils. Geological Society WorkingParty Report. Quarterly Journal <strong>of</strong> Engineering Geology, vol. 23, pp 1-102.


- 33 -Geological Society <strong>of</strong> London (1995). The description <strong>and</strong> <strong>classification</strong> <strong>of</strong> weathered rocksfor engineering purposes. Geological Society Working Party Report. Quarterly Journal<strong>of</strong> Engineering Geology, vol. 28, pp 207-242.Gilkes, RJ. & Suddhiprakaran, A. (1981). Mineralogical <strong>and</strong> chemical aspects <strong>of</strong>lateritisation in Southwestern Australia. Laterisation Processes. Proceedings <strong>of</strong> theInternational Seminar on Laterisation Processes. Balkema, pp 34-44.Hencher, S.R. & Martin, R.P. (1982). The description <strong>and</strong> <strong>classification</strong> <strong>of</strong> weathered rocksin Hong Kong for engineering purposes. Proceedings <strong>of</strong> the Seventh South-east AsianGeotechnical Conference, Hong Kong, vol. 1, pp 125-142.Irfan, T.Y. (1988). Fabric variability <strong>and</strong> index testing <strong>of</strong> a granitic saprolite. Proceedings<strong>of</strong> the Second International Conference on Geomechanics in Tropical Soils, Singapore,vol. 1, pp 25-35.Irfan, T.Y. (1992). Mineralogical Assessment <strong>of</strong> Creep-Type Instability at Two L<strong>and</strong>slipSites. Geotechnical Engineering Office, Hong Kong, 143 p. (GEO Report No. 13.)Irfan, T.Y. (1994). Mechanism <strong>of</strong> creep in a volcanic saprolite. Quarterly Journal <strong>of</strong>Engineering Geology, vol. 27, pp 211-230.Irfan . T.Y. (1996a). Mineralogy <strong>and</strong> Fabric Characterization <strong>and</strong> Classification <strong>of</strong> WeatheredGranitic Rocks in Hong Kong. Geotechnical Engineering Office, Hong Kong, 158 p.(GEO Report No. 41)Irfan . T.Y. (1996b). Mineralogy, <strong>fabric</strong> properties <strong>and</strong> <strong>classification</strong> <strong>of</strong> weathered graniticin Hong Kong. Quarterly Journal <strong>of</strong> Engineering Geology, vol. 20, pp 5-35.Irfan . T.Y. (1996c). Suitability <strong>of</strong> st<strong>and</strong>ard <strong>classification</strong> <strong>and</strong> index tests for saprolitic <strong>and</strong>residual soils in Hong Kong. Proceedings <strong>of</strong> the Twelfth Southeast Asian GeotechnicalConference <strong>and</strong> the Fourth Conference on Tropical Soils, Kuala Lumpur, vol. 1, pp 559-568.Irfan, T.Y. & Dearman, W.R. (1978). The engineering petrography <strong>of</strong> a weathered granitein Cornwall, Engl<strong>and</strong>. Quarterly Journal <strong>of</strong> Engineering Geology, vol. 11, pp 233-244.Irfan, T.Y., Koirala, N. & Tang, K.Y. (1987). A complex slope failure in a highlyweathered rock mass. Proceedings <strong>of</strong> the Sixth International Congress on RockMechanics, Montreal, pp 397-402.Irfan, T.Y. & Woods, N. (1997). Discontinuity controlled l<strong>and</strong>slides in weathered rocks inHong Kong. Transactions <strong>of</strong> the Hong Kong Institution <strong>of</strong> Engineers (in press).Kwong, J.K.P. (1985). Effects <strong>of</strong> Geological <strong>and</strong> Environmental Factors on the EngineeringProperties <strong>of</strong> Weathered Igneous Rocks in Hong Kong. Ph.D. Thesis, University <strong>of</strong>Leeds, UK, 291 p, unpublished.


. 34 -Langford, R.L., Lai, K.W., Arthurton, R.S. & Shim, R. ilWK iieoli^ouh^^New Territories. Geotechnical Control Office, Hong Kong, 140 p. (Hong KongGeological Survey Memoir No. 3).Langford, R.L., James, J.W.C., Shaw, R., Campbell, S.D.(L Kirk, P.A.


- 35 -LIST OF TABLESTableNo.PageNo.1 Comparison <strong>of</strong> Various Engineering <strong>and</strong> Geological 37Schemes Proposed for Classifying Weathering Pr<strong>of</strong>ile inIgneous <strong>and</strong> Volcanic Rocks (after Irfan, 1996b)2 Chemical Composition <strong>of</strong> Some Fine Ash <strong>and</strong> Coarse Ash 38Tuffs in Hong Kong3 Mineralogical Composition <strong>of</strong> 'Fresh' Volcanic Rocks 39(Based on Chemical Analysis <strong>and</strong> XRD)4 Composition <strong>of</strong> 'Fresh' Volcanic Rocks Determined from 40Modal Analysis5 Results <strong>of</strong> Modal Mineralogy <strong>and</strong> Quantitative Micro<strong>fabric</strong> 41Determination, Weathered Coarse Ash Tuff, Aberdeen6 Results <strong>of</strong> Chemical Analysis, Coarse Ash Tuff, Aberdeen 427 Results <strong>of</strong> Chemical Analysis, Coarse Ash Tuffs, Sai Kung 43<strong>and</strong> Yuen Long8 Results <strong>of</strong> Chemical Analysis, Fine Ash Tuff <strong>and</strong> Eutaxite, 44Tseung Kwan O9 Results <strong>of</strong> Semi-quantitative Bulk XRD Analysis, Coarse 45Ash Tuff, Aberdeen10 Results <strong>of</strong> Semi-quantitative Bulk XRD Analysis, Coarse 46Ash Tuffs, Sai Kung <strong>and</strong> Yuen Long11 Results <strong>of</strong> Semi-quantitative Bulk XRD Analysis, Fine Ash 47Tuff <strong>and</strong> Eutaxite, Tseung Kwan O12 Results <strong>of</strong> Semi-quantitative Clay Mineralogy Determined 48from XRD, Coarse Ash Tuff, Aberdeen13 Results <strong>of</strong> Semi-quantitative Clay Mineralogy Determined 49from XRD, Coarse Ash Tuffs, Sai Kung <strong>and</strong> Yuen Long14 Results <strong>of</strong> Semi-quantitative Clay Mineralogy Determined 50from XRD, Fine Ash Tuff <strong>and</strong> Eutaxite, Tseung Kwan O


36TableNo.PageNo.15 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy Coarse Ash 51Tuff, Aberdeen (Based on Chemical Analysis)16 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy, Coarse Ash 52Tuffs, Sai Kung <strong>and</strong> Yuen Long (Based on ChemicalAnalysis)17 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy, Fine Ash 53Tuff <strong>and</strong> Eutaxite, Tseung Kwan O (Based on ChemicalAnalysis)18 Mineralogy <strong>of</strong> Kaolin Veins in Hong Kong (after Irfan & 54Woods, 1997)19 Classification <strong>of</strong> Material Weathering Grades <strong>of</strong> Volcanic 55Rocks in Terms <strong>of</strong> Petrological Change20 Classification <strong>of</strong> Granitic <strong>and</strong> Volcanic Soils in Hong Kong 58(Wan, 1996c)21 Chemical Weathering Indices, Weathered Coarse Ash Tuff, 59Aberbeen22 Chemical Weathering Indices, Weathered Coarse Ash Tuff, 60Sai Kung23 Chemical Weathering Indices, Weathered Fine Ash Tuff, 61Tseung Kwan O24 Chemical Weathering Indices, Weathered Fine Ash Tuff; Ap 62Tsai, Tseung Kwan O25 Micropetrographic Indices for Weathered Coarse Ash Tuff, 63Aberdeen


- 36 -TableNo.PageNo.15 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy, Coarse Ash 51Tuff, Aberdeen (Based on Chemical Analysis)16 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy, Coarse Ash 52Tuffs, Sai Kung <strong>and</strong> Yuen Long (Based on ChemicalAnalysis)17 Results <strong>of</strong> Quantitative Bulk Norm Mineralogy, Fine Ash 53Tuff <strong>and</strong> Eutaxite, Tseung Kwan O (Based on ChemicalAnalysis)18 Mineralogy <strong>of</strong> Kaolin Veins in Hong Kong (after Irfan & 54Woods, 1997)19 Classification <strong>of</strong> Material Weathering Grades <strong>of</strong> Volcanic 55Rocks in Terms <strong>of</strong> Petrological Change20 Classification <strong>of</strong> Granitic <strong>and</strong> Volcanic Soils in Hong Kong 58(Irfan, 1996c)21 Chemical Weathering Indices, Weathered Coarse Ash Tuff, 59Aberbeen22 Chemical Weathering Indices, Weathered Coarse Ash Tuff, 60Sai Kung23 Chemical Weathering Indices, Weathered Fine Ash Tuff, 61Tseung Kwan O24 Chemical Weathering Indices, Weathered Fine Ash Tuff, Ap 62Tsai, Tseung Kwan O25 Micropetrographic Indices for Weathered Coarse Ash Tuff, 63Aberdeen


- 63 -Table 25 - Micropetrographic Indices for Weathered Coarse Ash Tuff, Aberdeenr_ . . SampleDescription ,. rSlightlyDecomposedModeratelyDecomposedHighlyDecomposedCompletelyDecomposedTransitionSoilNotes : *IR1IR2IR4IR5IR6Percentage <strong>of</strong> poresareas <strong>and</strong> pores ismicroscope.AlteredMinerals%11.824.943.058.164.5SoundMinerals%88.173.734.929.619.8Microcracks<strong>and</strong> Voids%0.11.322.012.2*15.6*MicropetrographicIndexIp7.872.810.540.420.25could be higher. Differentiation between minute alterednot possible under the normal resolution power <strong>of</strong>


- 64 -FigureNo.PageNo.1 Geological Map <strong>of</strong> Hong Kong <strong>and</strong> Sampling Locations 662 Classification <strong>of</strong> Pyroclastic Volcanic Rocks 673 Weathering Pr<strong>of</strong>ile in Coarse Ash Tuff, The Peak 684 Sketch <strong>of</strong> Weathering Pr<strong>of</strong>ile in Fine Ash Tuff, Tseung 69Kwan O5 Sketch <strong>of</strong> Weathering Pr<strong>of</strong>ile in Fine Ash Tuff, Ap Tsai, 70Tsueng Kwan 06 Weathering Pr<strong>of</strong>ile in Coarse Ash Tuff, Aberdeen (Prior to 71the L<strong>and</strong>slide)7 Whole Rock XRD Scans <strong>of</strong> Slightly, Moderately <strong>and</strong> Highly 72Decomposed Coarse Ash Tuff8 Whole Rock XRD Scans <strong>of</strong> Completely Decomposed <strong>and</strong> 73Residual Soil <strong>of</strong> Coarse Ash Tuff9 Clay Portion XRD Scans <strong>of</strong> Weathered Coarse Ash Tuff, 74Yuen Long10 Clay Portion XRD Scans <strong>of</strong> Weathered Fine Ash Tuff, 75Ap Tsai, Tseung Kwan 011 Variation in Mineralogical Composition with Weathering, 76Coarse Ash Tuff, Aberdeen12 Variation in Mineralogical Composition with Weathering, 77Coarse Ash Tuff, Sai Kung13 Variation in Mineralogical Composition with Weathering 78Coarse Ash Tuff, Yuen Long14 Variation in Mineralogical Composition with Weathering, 79Fine Ash Tuff, Tseung Kwan O15 Variation in Mineralogical Composition with Weathering 80Fine Ash Tuff, Ap Tsai, Tseung Kwan 0


- 65 -FigureNo.PageNo.16 Variation in Chemical Indices for Weathered Coarse Ash 81Tuffs from Different Localities17 Weathering Potential Index Versus Mobiles Index, Coarse 82Ash Tuff, Aberdeen18 Weathering Potential Index Versus Mobiles Index, All 83Rocks


- 67 -


- 68 -


- 72 -


- 73 -


- 74 -


- 75 -


- 76 -


- 78 -


- 84 -LIST OF PLATESPlateNo.No.1 Weathering Pr<strong>of</strong>ile in Coarse Ash Tuff, Mount Davis, Hong 85Kong Isl<strong>and</strong>2 Complex Weathering Pr<strong>of</strong>ile in Fine Ash Tuff, Chai Wan, 86Hong Kong Isl<strong>and</strong>3 Soil Layer Overlying Rock Layer in Fine Ash Tuff, Chai 87Wan, Hong Kong Isl<strong>and</strong>4 Weathering Pr<strong>of</strong>ile in Fine Ash Tuff, Ap Tsai, Tseung 88Kwan O5 Sampling Site at Tseung Kwan O 896 Sampling Site at Sai Kung 907 Weathered Coarse Ash Tuff Samples, Abedeen 918 Material Weathering Grades in Fine Ash Tuff, Tsueng 92Kwan O9 Material Weathering Grades in Fine Ash Tuff (Eutaxite),Ap 93Tsai, Tseung Kwan 010 Material Weathering Grades in Coarse Ash Tuff, Sai Kung 9411 Micrographs Illustrating Mineralogical <strong>and</strong> Fabric Properties 95<strong>of</strong> Coarse Ash Tuff Weathered to Different Degrees12 Micrographs Dlustrating Mineralogical <strong>and</strong> Fabric Properties 96<strong>of</strong> Fine Ash Tuffs Weathered to Different Degrees


- 91 -(Negative No. SP8907803 & 04).Plate 7 - Weathered Coarse Ash Tuff Samples, Aberdeen


- 97 -APPENDIX ASAMPLE DESCRIPTIONS


- 98 -APPENDIX A : SAMPLE DESCRIPTIONSLocality : New Slope Cutting at Ap Tsai Wan, Tseung Kwan OSample No.ATW1(JB8)ATW2(JB5)ATW3(JB5)ATW3FATW3WATW4(JB3)ATW5(JB3)LocationSample taken from freshrock zone near base <strong>of</strong>50-60 m high cut.Sample taken from lowerpart <strong>of</strong> moderatelyweathered zone in cutslope, half way up the hill.Adjacent to sample ATW2.Sample taken from closelyjointed, highly weatheredrock zone, in cut slope neartop <strong>of</strong> hill (about 5-7 mbelow original groundsurface before cutting).Adjacent to ATW4.Description <strong>of</strong> H<strong>and</strong> SpecimenFresh. Very strong, dark grey, veryfine grained matrix with up to 20%feldspar <strong>and</strong> occasional quartzmegacrysts (max. 2 mm), occasionallapilli, fine ash TUFF.Slightly decomposed (slightlydiscoloured). Very strong, yellowishbrown discoloured rim, up to 10 mmwide with very strong, medium greyfresh core, fine ash TUFF.Slightly decomposed (moderatelydiscoloured). Strong, yellowish brownto dark brown discoloured rim, up to20 mm wide with very strong,medium grey, fresher core, fine ashTUFF, occasional feldspars are grittyin discoloured rim; a series <strong>of</strong> tightmicrocracks parallel to discolouredzone.Fresh Core.Discoloured rim (ModeratelyDecomposed).Highly Decomposed. Weak to veryweak, completely discolouredyellowish grey with brown <strong>and</strong> patchyblack staining along fractures, fine ashTUFF; large feldspars are grittyoccasionally powdery; white kaolin(?)infill (vein?), macro-fractures at 5 to20 mm spacing.Highly Decomposed (SlightlyAltered). Very weak, completelydiscoloured, light yellowish grey withpatchy brown <strong>and</strong> black on fracturesurfaces, fine ash TUFF; feldspars aregritty to powdery; kaolinized adjacentto one face; macro-fractures at 3 to40 mm spacing.


- 99 -Sample No.ATW6(JB1)LocationSample taken fromcompletely weathered rockzone in cut slope (approx.2 m below original groundsurface before cutting).Description <strong>of</strong> H<strong>and</strong> SpecimenCompletely Decomposed. Extremelyweak, yellowish brown with pinkishorange <strong>and</strong> olive green patches, fineash TUFF (eutaxitic?); feldsparsclayey to powdery, loss <strong>of</strong> tuff <strong>fabric</strong>at places, particularly around rootlets(Clayey s<strong>and</strong>y silt).


100 -Locality : Access Road Cutting, above Tseung Kwan 0 New TownSample No,,TKO1TKO2FTKO2WTKO3TKO4TKO5LocationSample taken from freshcore <strong>of</strong> a small corestone inmoderately weathered tuffzone cutting; along accessroad.Sample taken from partiallydiscoloured portion <strong>of</strong>above corestone.Same as TKO2F(discoloured rim <strong>of</strong>corestone).Adjacent to TKO1, smallcorestone.Sample taken from highlyweathered rock zone alongaccess road cutting,adjacent to TKO4.Sample taken fromcompletely weathered zone.DgscriptionFresh core <strong>of</strong> slightly discolouredblock. Very strong to extremelystrong, dark grey, crystal-bearing fineash TUFF.Fresher core <strong>of</strong> moderately discolouredblock. Very strong, light to mediumgrey (faintly discoloured) core withyellowish grey to yellowish brown rim(TKO2W\ 20 mm wide, crystalbearing fine ash TUFF; black oxidecoating on joint surface.Completely Discoloured (moderatelydecomposed) rim <strong>of</strong> TKO2. Lightgrey with yellowish grey to yellowishbrown (on joint surface), moderatelystrong to strong, fine ash TUFF.Moderately Decomposed. Moderatelyweak, light yellowish grey withreddish brown <strong>and</strong> black on jointsurface, completely discoloured fineash TUFF; some feldspars are gritty,irregular fractures at 10 to 80 mmspacing near joint surface.Highly Decomposed. Weak, yellowishgrey with yellowish brown on jointsurfaces, patches <strong>of</strong> black, fine ashTUFF; micro-fractured throughout at 5to 40 mm spacing. Light pinkfeldspars on one side <strong>of</strong> block(altered?); most large feldspars arepowdery.Highly Decomposed. Very weak,yellowish grey with reddish brown <strong>and</strong>patches <strong>of</strong> black, fine ash TUFF;micro-fractured at 10 -20 mm spacing,irregular pattern; veinlets <strong>of</strong> kaolin(altered?)(Can be crumbled to s<strong>and</strong>ysilt with strong finger pressure).


- 101 -Sample No.TKO6TKO7(S2)*TKO8LocationSample taken fromcompletely weathered zone,a few metres West <strong>of</strong>TKO7.Sample taken from trial pitin transition soil zone.Sample taken from residualsoil zone, about 1.5 mbelow ground surface.Description <strong>of</strong> H<strong>and</strong> SpecimenCompletely Decomposed (very dense).Extremely weak, mottled yellowishgrey <strong>and</strong> reddish brown, fine ashTUFF; all large feldspars are powderyto s<strong>of</strong>t; easily crumbled to individualgrains (S<strong>and</strong>y silt with some clay).Transition soil with partial tuff <strong>fabric</strong>(about 30%). Loose to dense, reddishbrown <strong>and</strong> yellowish brown, fine ashTUFF (Clayey silt).Residual soil. Reddish brown withpatches <strong>of</strong> yellowish brown, smallpatches <strong>of</strong> tuff <strong>fabric</strong> (Clayey silt).


- 102 -Locality : L<strong>and</strong>slip Site behind Isl<strong>and</strong> Road Government School AberdeenNote : For details, see Irfan (1994)Sample No.IR1IR7IR2IR8IR4IR9LocationCore sample from boreholeBH1, at 9.6 m depth inhighly weathered rock zone.Core sample from boreholeBH1, at 6.5 m depth, inhighly weathered rock zone.Core sample from boreholeBH1, at 8.6 m depth inhighly weathered rock zone.Core sample from boreholeBH1, at 5.3 m depth inhighly weathered rock zone.Core sample from boreholeBH1, at 8.8 m depth.Core sample from boreholeBH1, at 7.3 m depth inhighly weathered rock zone.Description <strong>of</strong> H<strong>and</strong> SpecimenSlightly Decomposed (SlightlyDiscoloured). Strong, medium greywith light yellowish grey in thevicinity <strong>of</strong> joint surfaces, coarse ashTUFF; Megacrysts (up to 4 mm) <strong>of</strong>quartz <strong>and</strong> feldspars, in a finelycrystalline matrix; biotite rich, 5 to10%; feldspars gritty in stained rim.Slightly Decomposed (HighlyDiscoloured). Moderately strong tostrong, light grey core with yellowishgrey rim (up to 4 mm wide), coarseash TUFF; most large feldspars grittyin the discoloured rim; very closelyspaced fractures parallel to jointsurface.Moderately Decomposed. Moderatelyweak to strong, light yellowish greywith yellowish brown secondarystaining along joint surface; coarse ashTUFF; feldspars with varying degrees<strong>of</strong> decomposition but mostly hard togritty, except near joint surfaces theyare powdery.Moderately Decomposed. Moderatelyweak, light yellowish greenish grey,coarse ash TUFF; feldspars powderyto gritty; micro-fractured by a network<strong>of</strong> irregular, branched microcracks.Highly Decomposed. Weak, lightyellowish grey, coarse ash TUFF;feldspars powdery to gritty, some s<strong>of</strong>t.Highly Decomposed. Very weak,light yellowish grey, coarse ash TUFF;feldspars powdery, some s<strong>of</strong>t.


- 103 -Sample No.IRSIR3IR6LocationCore sample from blocksample IRS in a trial pit(TP3) at 0.3 m belowl<strong>and</strong>slip surface incompletely weathered zone.Sample contains up to30 mm thick kaolin veins<strong>and</strong> relict joints.Block sample taken acrossshear plane comprising boththe undisturbed saproliticzone containing kaolinveins <strong>and</strong> creeping zone.Block sample taken from atrial pit, at 1.3 m depth, inresidual (transition) soilzone.Description <strong>of</strong> H<strong>and</strong> SpecimenCompletely Decomposed <strong>and</strong>hydrothermaily altered. Extremelyweak, light yellowish grey to pinkishgrey (altered portion), coarse ashTUFF with kaolin veins; feldsparspowdery to s<strong>of</strong>t (in the vicinity <strong>of</strong>kaolin veins (clayey s<strong>and</strong>y silt).Specimens IR51 <strong>and</strong> IR55 are kaolinveins.Completely Decomposed. Extremelyweak, yellowish grey with pinkishgrey <strong>and</strong> white patches, coarse ashTUFF with relict joints <strong>and</strong> kaolinveins; feldspars powdery to s<strong>of</strong>t <strong>and</strong>biotite greenish grey to silvery grey(Clayey s<strong>and</strong>y silt).Transition soil. Mottled reddishbrown <strong>and</strong> yellowish grey with thindiscontinuous kaolin veinlets but norelict discontinuities; partial loss <strong>of</strong>tuff <strong>fabric</strong> throughout block sample(Clayey s<strong>and</strong>y silt).


104Locality : Sai Kung, Old Borrow AreaSample No.SK7(CT5)SK6SK5(CT4)SK4(CT4)SK2 & SK3(CT3)SKI(CT1)LocationSample taken from acorestone, about 3 m beloworiginal ground surface, incompletely weathered rockzone.Sample taken from acorestone in completelyweathered rock zone.Sample taken from core <strong>of</strong>a corestone, about 3 mbelow original groundsurface, in completelyweathered rock zone, 1 mto the west <strong>of</strong> SK7.Sample taken from outerrim <strong>of</strong> the same corestoneas SK5.Sample taken in completelyweathered rock zone, about2 m below original groundsurface.Sample taken from residualsoil zone, about 0.8 mbelow ground surface.n <strong>of</strong> H<strong>and</strong> SpecimenSlightly Decomposed. Very strong,medium to dark grey fresher core withlight yellowish grey rim, up to 10 mmwide, coarse ash TUFF withoccasional lithic fragments <strong>and</strong>abundant quart/, <strong>and</strong> feldspar crystals,up to 4 mm in diameter; feldsparsgritty.Slightly Decomposed, Strong, darkgrey core surrounded by light greyinner rim <strong>and</strong> light yellowish greyouter core; flow-b<strong>and</strong>ed porphyriticLAVA with cluster <strong>of</strong> quartz <strong>and</strong>feldspars (The sample is possibly froma lava b<strong>and</strong> within coarse ash TUFF).Moderately Decomposed, Weak, lightyellowish grey with brown <strong>and</strong> blackpatches, completely discoloured,coarse ash TUFF; most feldsparsgritty, some hard; occasional quartz upto 6 mm.Highly Decomposed. Very weak,light yellowish grey, coarse ash TUFFwith very closely spaced fractures; thinclay infill along some fractures;feldspars powdery to gritty, some s<strong>of</strong>t.Completely Decomposed. Extremelyweak, yellowish brown with yellowishgrey, coarse ash TUFF; with patchy<strong>fabric</strong> loss; most feldspars powdery tos<strong>of</strong>t (Clayey s<strong>and</strong>y silt).Residual soil. Reddish brown, s<strong>and</strong>yclayey silt with small patches <strong>of</strong> tuffmatrix; numerous pores, up to 2 mmin diameter <strong>and</strong> desiccation cracks.


- 105 -Locality : Tai Tong East Borrow Area, Yuen LongSample No.YL1YL6YL3YL5YL4LocationSample taken from freshcore <strong>of</strong> a large corestone, incompletely weathered rockzone, within the l<strong>and</strong>sliparea.Sample taken from a highlydecomposed corestone,adjacent to trial pit TP1 inthe l<strong>and</strong>slip area.Sample taken at 1 m depthin trial pit TP1 in thel<strong>and</strong>slip area, in completelyweathered rock zone.Sample taken at 1.5 mdepth in trial pit TP4 in thel<strong>and</strong>slip area, in completelyweathered rock zone.Sample taken fromdeformed saprolite layer at0.5 m depth in trial pitTP1, containing polishedshears surfaces.Description <strong>of</strong> H<strong>and</strong> SpecimenFresh. Very strong, medium grey,lapilli bearing coarse ash TUFF withoccasional sedimentary rockfragments, epidote-chlorite veined(altered), abundant feldspars(dominantly plagioclase) with quartzcrystals.Highly Decomposed. Extremely weakto very weak, dark greenish yellowishgrey, coarse ash TUFF containing aseries <strong>of</strong> parallel micro-fracturesspaced at 5 mm, feldspars powdery togritty; quartz dull, brownish stained;biotite greenish grey, powdery.Completely decomposed. Extremelyweak, light greenish yellowish grey,coarse ash TUFF; feldspars gritty topowdery, occasionally s<strong>of</strong>t; quartzvery dull (S<strong>and</strong>y silt with little clay).Completely Decomposed. Very dense,greenish grey with patches <strong>of</strong>yellowish grey, coarse ash TUFF;feldspars powdery, occasionally s<strong>of</strong>t;quartz very dull; biotite powdery,yellowish greenish grey; groundmasspowdery to s<strong>of</strong>t (Clayey s<strong>and</strong>y silt).Deformed completely decomposedtuff. Light greenish grey with patches<strong>of</strong> yellowish brown; no relict jointsbut with deformed disturbed tuff<strong>fabric</strong>.


106 -APPENDIX BMETHODS OF PETROGRAPHICAL AND CHEMICAL ANALYSISAND IDENTIFICATION OF MINERALS


- 107 -B.IGeneralThe chemical <strong>and</strong> XRD analyses <strong>of</strong> the coarse ash tuff specimens from the l<strong>and</strong>slip siteat Aberdeen were carried out on duplicate specimens at Amdei <strong>and</strong> Analabs laboratories inAustralia in 1989. The <strong>mineralogical</strong> interpretations from XRD traces were carried out byDr R. Brown <strong>of</strong> Amdei <strong>and</strong> Dr R. J. Gilkes <strong>and</strong> Mr G.D. Moore <strong>of</strong> Analabs.The chemical analyses by XRF were carried out by Sietronics Pty in Australia on thetuff specimens from Tseung Kwan 0 (TK series), Sai Kung (SK series) <strong>and</strong> Ap Tsai Headlead(ATW series), which also carried out the XRD analyses on the same specimens. The XRDanalyses on these rocks using duplicate specimens were repeated by Messrs S. V. Prior <strong>and</strong>S. J. Kemp <strong>of</strong> the British Geological Survey (BGS) in the U.K. due to uncertainty with some<strong>of</strong> the interpretations made by Sietronics Pty. The BGS determined the quantitative normativebulk mineralogies <strong>of</strong> these rocks using the chemical analyses results <strong>of</strong> Sietronics Pty.The BGS also carried out the chemical <strong>and</strong> XRD analyses <strong>of</strong> the specimens from YuenLong (YL series). The <strong>mineralogical</strong> <strong>and</strong> <strong>fabric</strong> description <strong>of</strong> all rocks were carried out bythe author (T. Y. Irfan), using thin sections <strong>and</strong> h<strong>and</strong> specimens.B.2 Methods <strong>of</strong> Analyses Used by AmdeiB.2.1Clay Mineralogy bv XRDPortion <strong>of</strong> each sample was powdered finely <strong>and</strong> used to prepare an X-ray diffractometertrace which was interpreted by st<strong>and</strong>ard procedures.Further, weighed, lightly pre-ground subsamples were taken <strong>and</strong> dispersed in water withthe aid <strong>of</strong> deflocculants <strong>and</strong> an electric blender, <strong>and</strong> allowed to sediment to produce-2 /xm e.s.d size fractions by the pipette method. The resulting dispersions were examinedby plummet balance to determine their solids contents, <strong>and</strong> were then used to produce orientedclay preparations on ceramic plates. Two plates were prepared per sample, both beingsaturated with Mg ++ ions, <strong>and</strong> one in addition being treated with glycerol. When air-dry, thesewere examined in the X-ray diffractometer. Additional diagnostic examinations carried outconsisted <strong>of</strong> examination <strong>of</strong> the glycerol-free plate after heating for one hour at 110°C, aftertreatment with formamide or glycol, <strong>and</strong> after heating for one hour at 550°C (varioustreatments according to individual requirements). The radiation used was CoKa.B.2.2Chemical AnalysesWhole rock chemical analyses were carried out by Inductively Coupled Plasma (ICP)Atomic Absorption Spectroscopy.The loss on ignition (LOI) was determined by heating representative finely groundspecimens to 13G0°C <strong>and</strong> relating back weight losses to original sample weights.


- 108 -B.2.3Mineral Calculations from Chemical AnalysesA suitable computer program was used for the calculation <strong>of</strong> mineral proportions usingthe chemical analysis results. Appropriate theoretical, or quoted practical compositions forthe component minerals were used in the calculations, <strong>and</strong> in some instances these were variedslightly when the calculated fit for minor minerals (e.g. biotite, vermictilite) was poor.Minerals which were not observed by XRD were not introduced into the calculations, althoughin some cases this would have resulted in a considerably improved agreement. In any case,the effect on the calculated percentages <strong>of</strong> the major minerals is usually negligible. Thefigures given in the tables are quoted to the nearest percent, but the actual accuracy, althoughunknown, is not expected that level. It is possible that appreciable amorphous material(amorphous clay or volcanic glass) is present. If that is the case, the calculated mineralpercentages could be considerably in error.Iron was calculated to goethite on the basis <strong>of</strong> the observation <strong>of</strong> goethite in somesamples, especially in the clay fractions.B.2.3Comments on Mineral Determinations by XRDAll three kaolin minerals (kaolinite, hydrated halloysite <strong>and</strong> halloysite or dehydratedhalloysite) are present in many samples. There is no known method to estimate theproportion <strong>of</strong> hydrated <strong>and</strong> dehydrated halloysite by XRD. The presence <strong>of</strong> hydratedhalloysite was detected from peak shift with glycerol. The approximate proportion <strong>of</strong>halloysite (including both types) in the total kaolin was estimated by formamide treatment <strong>of</strong>the clay fractions. Whether the relative proportions are maintained in the bulk sample is <strong>of</strong>course uncertain as this would depend on the grain size <strong>of</strong> kaolinite <strong>and</strong> halloysite particlesin the bulk material.Smectite was only identified in kaolin veins. In the instances where interstratified claywas identified, the clay layers making up the material were not able to be determined withcertainty. An illite-vermiculite interstratification has been suggested but the evidence is notgood.Muscovite <strong>and</strong> illite merge into one another, the dividing line being indefinite <strong>and</strong> thedifferences being essentially those <strong>of</strong> crystallinity, potassium content, hydration <strong>and</strong> probablyparticle size. Because <strong>of</strong> their close relationship, the XRD results are usually reported asmica/illite or muscovite/illite. Based on diffraction peak breadths, which were generally weak,the micaceous minerals in the tuffs from Aberdeen are probably illite.B - 3Methods <strong>of</strong> Analyses Used by AnalabsB.3.1 Bulk Mineralogy bv XRDFinely pulverised portions <strong>of</strong> each rock sample were placed in the diffractometer asloose packed r<strong>and</strong>omly orientated samples using the back loading technique to reduceorientation problems. A diffractrogram was run each sample covering the angular range <strong>of</strong>4 to 65 degrees (d spacings <strong>of</strong> 22A to 1.43A) <strong>and</strong> using Cu radiation.


- 109 -Each scan was then examined using transparent overlays to identify the presence orabsence <strong>of</strong> mineral species. Quantification is based on typical analyses for each type <strong>of</strong>mineral. Many minerals such as the clays <strong>and</strong> feldspars do not have set stoichiometry <strong>and</strong>therefore the data must be regarded as semi-quantitative.B.3.2.Clav Mineralogy by XRD on Clay FractionSoil specimens were dispersed in a dilute alkali solution using on ultrasonic probe. Theclay fraction (less than 2 microns) was collected by sedimentation <strong>and</strong> the clay coated ontothe porous ceramic plates under suction. The clay was saturated with :a. Mg solutionb. Mg-glycerolc. K-550°C prior to XRD analysis using a Philips goniometerwith diffracted beam monochromator.B.3.3Whole Rock Analyses by XRFX-ray Fluorescent Spectroscopy (XRF) was used for the determination <strong>of</strong> all oxidesexcept for Na 2 O which was determined by Atomic Absorption Spectroscopy (AAS).For the XRF analysis, the samples were prepared as glass fusion discs which reduces oreliminates the inherent effects; the glass eliminates particle size effects <strong>and</strong> partiallynormalises absorption effects through dilution. An accurately weighed portion <strong>of</strong> thepulverised sample was added to a preweighed amount <strong>of</strong> lithium borate flux which containedthe heavy X-Ray absorber lanthanum oxide. These were carefully mixed <strong>and</strong> then fused inplatinum gold alloy crucibles to produce a homogeneous melt. The molten glass was thenpoured onto a graphite platten <strong>and</strong> pressed out to form a solid glass disc which after coolingwas presented to the XRF for analysis.The calibration for the XRF is based on a synthetic master st<strong>and</strong>ard but for "fine tuning"<strong>of</strong> the data a series <strong>of</strong> certified reference materials were also prepared <strong>and</strong> run with thesamples. Appropriate corrections were applied to correct for backgrounds, peak overlaps <strong>and</strong>absorption <strong>and</strong> enhancement effects.Because sodium is not a particularly sensitive element by XRF this cation wasdetermined by AAS using a mixed acid digest. An accurately weighed portion <strong>of</strong> the samplewas digested in heated teflon beakers using perchloric, nitric, hydrochloric <strong>and</strong> hydr<strong>of</strong>luoricacids. Following dissolution the solution was made up to volume in volumetric flasks usingdeionised water. Analysis by AAS with calibration was based on "Volucon" st<strong>and</strong>ardsolutions <strong>and</strong> certified reference materials used as checks.Moisture <strong>and</strong> loss on ignition (LOI) were determined by gravimetric methods wherebyweight losses at specific temperatures, 105°C for moisture <strong>and</strong> 1050°C for LOI, were recorded<strong>and</strong> related back to the original sample weights. Two <strong>of</strong> the samples TLC15 <strong>and</strong> IR39 were


110extremely wet <strong>and</strong> before any analyses or preparation were carried out, w~ere dried. Theresults reported for these two samples are on the "dried sample' basis. All other samples arereported on the 'as-received' basis.B.4 Methods <strong>of</strong> Analyses Used_byjkitij^B.4.1Sample PreparationA representative 20 g portion <strong>of</strong> each sample was dried at 55°C crushed <strong>and</strong> hammermilledto pass a 200 /xm screen. This material w r as used for thermogravimetric (TG) analyses.To provide a finer <strong>and</strong> uniform particle size for whole-rock X-ray diffraction (XRD) analysis,approximately 3 g <strong>of</strong> the less than 200 fim material was subsampled, wet micronised underacetone for 5 minutes <strong>and</strong> dried at 55°C.Oriented XRD mounts were prepared from weathered samples to provide a more detailedanalysis <strong>of</strong> the clay minerals present. 20 g <strong>of</strong> each sample was dispersed in distilled waterusing an ultrasonic probe <strong>and</strong> laboratory stirrer. The resulting suspension was then passedthrough a 63 /xm sieve <strong>and</strong> the


- Ill -Approximately 5 g <strong>of</strong> tema-ground sample was dried for 24 hours at 105°C. Loss onignition was calculated from any further weight loss from 1 g <strong>of</strong> sample heated at 1050°C for1 hour. Glass beads were then prepared by fusing 0.9 g <strong>of</strong> each sample with 9 g (correctedfor loss on fusion) <strong>of</strong> Spectr<strong>of</strong>lux-100 (Li 2 B 4 O 7 ) at approximately 1200°C (in a PhilipsPerl'X-2 microprocessor-controlled automatic fused bead maker) <strong>and</strong> pouring the melt intoa platinum casting dish. Lithium iodide (Lil) was added to all samples before fusion to actas a releasing agent.Samples were then analysed using a Philips PW1480/10 sequential, fully microprocessorcontrolled wavelength-dispersive XRF spectrometer equipped with a sc<strong>and</strong>ium anode tube.Analyte angles were calibrated from international <strong>and</strong> in-house st<strong>and</strong>ard reference materials,prepared as fused glass beads. Drift correction was catered for by an external ratio monitor<strong>and</strong> background correction was applied where necessary. Ten major elements were calibratedas oxides using the de Jongh calibration algorithm <strong>and</strong> the alpha coefficients generatedempirically with one regression based line overlap (Ca on Ng). Oxide concentrations weremerged with loss on ignition figures <strong>and</strong> the results totalled. No lower limits <strong>of</strong> detection arequoted for major elements but reporting limits are 0.01% for all oxides with the exception <strong>of</strong>MnO which is reported to 0.001%.B.4.5Mineral Quantification from Chemical AnalysesBoth K-feldspar <strong>and</strong> plagioclase contents were calculated from the XRF K 2 O <strong>and</strong> Na 2 Ofigures assuming a stoichiometry <strong>of</strong> KAlSi 3 O 8 respectively. The plagioclase concentrationcalculation necessarily assumes the absence <strong>of</strong> anorthite (Ca-rich) or mixed cation (Ca/Na)species. The K-feldspar concentration calculation also assumes the absence <strong>of</strong> otherpotassium-bearing phases such as muscovite or illite. Therefore where both K-feldspar <strong>and</strong>significant 'mica* were detected by XRD, the K 2 O figure was apportioned on the basis <strong>of</strong>K-feldspar XRD peak height data. A calibration curve was constructed from the 3.30 AK-feldspar peak data from those samples where only trace 'mica* content could then bedetermined. The K 2 O contribution from the K-feldspar was then determined, <strong>and</strong> the 'mica*concentration calculated from the residual K 2 O figure.However, it was not possible to apply this XRD peak-height method <strong>of</strong> quantificationto the 'mica* phase(s) as broadening <strong>of</strong> the 10 A peak indicates a change in crystallinity<strong>and</strong>/or phase change during weathering.Quartz concentrations were estimated from the residual wt% SiO 2 figure obtained aftersubtraction <strong>of</strong> the SiO 2 due to 'kaolin*, plagioclase <strong>and</strong> K-feldspar, <strong>and</strong> 'mica* assuming thatthese were the only silica-bearing phases present. The hematite concentration was taken asthe wt% Fe 2 G 3 .The total wt% figures for quantitative phases are generally 100+/-10%. Where totalsexceed this limit the sources <strong>of</strong> error are probably due to the increased concentration <strong>of</strong> nonquantifiedminerals, such as 'mica', chlorite, vermiculite, hydrobiotite, ?carbonates <strong>and</strong>?titanium oxides as well as calculation errors due to the stoichiometries assumed.


- 112 -B.4.6Comments on Mineral Identification from XRDDiagnostic testing indicates that the


- 113 -consistent,(b)X-Rav DiffractionUnoriented powder mounts <strong>of</strong> the total sample were preparedfor bulk mineralogy. Oriented sedimented mounts were madefor examination <strong>of</strong> clay minerals (


GEOTECHNICAL ENGINEERING OFFICE PUBLICATIONSGeotechnical Manual for Slopes, 2nd Edition (1984), 295 p.(English Version), (Reprinted, 1997).308Guide to Retaining Wall Design, 2nd Edition (1993), 258 p. Geoguide 1(Reprinted, 1998).Guide to Site Investigation (1987), 359 p. (Reprinted, 1996). Geoguide 2Guide to Rock <strong>and</strong> Soil Descriptions (1988), 186 p. (Reprinted, Geoguide 31997).Guide to Cavern Engineering (1992), 159 p. (Reprinted, 1994). Geoguide 4Guide to Slope Maintenance (1995), 91 p. (English Version). Geoguide 5HK$90(US$20)HK$90(US$20)HK$60(US$13)HK$83(US$17.5)HK$58(US$12.6)HK$36(US$13.5)HK$30(US$6.5)HK$40(US$7.5)Layman's Guide to Slope Maintenance (1995), 56 p.(Bilingual), (Reprinted, 1997).(1997.WFreeModel Specification for Prestressed Ground Anchors, 2nd Geospec 1 HK$62Edition (1989), 164 p. (Reprinted, 1997). (US$ 11)Model Specification for Reinforced Fill Structures (1989), Geospec 2 HK$58135 p. (Reprinted, 1997). (US$ 10.5)Mid-levels Study : Report on Geology, Hydrology <strong>and</strong> Soil - HK$534Properties (1982), 265 p. plus 54 drgs. (Reprinted, 1997).(US$86)Prediction <strong>of</strong> Soil Suction for Slopes in Hong Kong, by M.G. GCO Publication HK$ 132Anderson (1984), 242 p. (Reprinted, 1996). No. 1/84 (US$24)(Superseded by GCO Publication No. 1/85)GCO PublicationNo. 2/84


(Superseded by Geospec 1)GCO PublicationNo. 3/84Review <strong>of</strong> Superficial Deposits <strong>and</strong> Weathering in Hong Kong, GCO Publication HK$40by J.D. Bennett (1984), 58 p. (Reprinted, 1993). No. 4/84 (US$8)Review <strong>of</strong> Hong Kong Stratigraphy, by J.D. Bennett (1984), GCO Publication HK$2586 p. No. 5/84 (US$5.5)Review <strong>of</strong> Tectonic History, Structure <strong>and</strong> Metamorphism <strong>of</strong> GCO Publication HK$20Hong Kong, by J.D. Bennett (1984), 63 p. No. 6/84 (US$5)(Superseded by GCO Publication No. 1/88)GCO PublicationNo. 1/85Groundwater Lowering by Horizontal Drains, by D.J. Craig & GCO Publication HK$74I. Gray (1985), 123 p. (Reprinted, 1990). No. 2/85 (US$12)(Superseded by GEO Report No. 9)GCO PublicationNo. 1/88Review <strong>of</strong> Design Methods for Excavations (1990), 187 p. GCO Publication HK$40(Reprinted, 1996). No. 1/90 (US$10)Foundation Properties <strong>of</strong> Marble <strong>and</strong> Other Rocks in the Yuen GCO Publication HK$58Long - Tuen Mun Area (1990), 117 p. No. 2/90 (US$10)Review <strong>of</strong> Earthquake Data for the Hong Kong Region (1991), GCO Publication HK$42115 p. No. 1/91 (US$11.5)Review <strong>of</strong> Granular <strong>and</strong> Geotextile Filters (1993), 141 p. GEO Publication HK$32No. 1/93 (US$19)Pile Design <strong>and</strong> Construction (1996), 348 p. (Reprinted, 1997). GEO Publication HK$62No. 1/96 (US$13.5)Report on the Kwun Lung Lau L<strong>and</strong>slide <strong>of</strong> 23 July 1994, 2 - FreeVolumes, 400 p. (English Version), (Reprinted, 1996).Report on the Fei Tsui Road L<strong>and</strong>slide <strong>of</strong> 13 August 1995, - Free2 Volumes, 81 p. (Bilingual). %>MH a $£mmmm\


Report on the Shum Wan Road L<strong>and</strong>slide <strong>of</strong> 13 August 1995,2 Volumes, 63 p. (Bilingual).FreeWhat to Do When You Receive a Dangerous Hillside Order(1996), 16 p. (Bilingual).Free(Hong Kong) Rainfall <strong>and</strong> L<strong>and</strong>slides in 1984, by J. Premchitt(1991), 91 p. plus 1 drg. (Reprinted, 1995).(Hong Kong) Rainfall <strong>and</strong> L<strong>and</strong>slides in 1985, by J. Premchitt(1991), 108 p. plus 1 drg. (Reprinted, 1995).(Hong Kong) Rainfall <strong>and</strong> L<strong>and</strong>slides in 1986, by J. Premchitt(1991), 113 p. plus 1 drg. (Reprinted, 1995).Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1987, by J. Premchitt(1991), 101 p. plus 1 drg. (Reprinted, 1995).Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1988, by J. Premchitt(1991), 64 p. plus 1 drg. (Reprinted, 1995).Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1989, by K.L. Siu(1991), 114 p. plus 1 drg. (Reprinted, 1995).Aggregate Properties <strong>of</strong> Some Hong Kong Rocks, by T.Y.Man, A. Cipullo, A.D. Burnett & J.M. Nash (1992), 212 p.(Reprinted, 1995).Foundation Design <strong>of</strong> Caissons on Granitic <strong>and</strong> VolcanicRocks, by T.Y. Man & G.E. Powell (1991), 85 p. (Reprinted,1995).GEO ReportNo. 1GEO ReportNo. 2GEO ReportNo. 3GEO ReportNo. 4GEO ReportNo. 5GEO ReportNo. 6GEO ReportNo. 7GEO ReportNo. 8HK$118(US$17.5)HKS126(US$20)HKS126(US$20)HKS122(US$19.5)HKSI06(US$16)HK$126(US$20)HK$120(US$19.5)HK$62(US$10.5)Bibliography on the Geology <strong>and</strong> Geotechnical Engineering <strong>of</strong>Hong Kong to December 1991, by E.W. Br<strong>and</strong> (1992), 186 p.(Superseded by GEO Report No.39)GEO ReportNo. 9Bibliography on Settlements Caused by Tunnelling, by E.W.Br<strong>and</strong> (1992), 50 p. (Reprinted, 1995). (Superseded by GEOReport No.51)GEO ReportNo. 10HK$48(US$8.5)Direct Shear Testing <strong>of</strong> a Hong Kong Soil under VariousApplied Matric Suctions, by J.K. Gan & D.G. Fredlund (1992),241 p. (Reprinted, 1995).GEO ReportNo. 11HK$136(US$21.5)


Rainstorm Run<strong>of</strong>f on Slopes, by J. Premchitt, T.S.K. Lam, J.M. GEO Report HK$121Shen <strong>and</strong> H.F. Lam (1992), 211 p. (Reprinted, 1995). No. 12 (US$19.5)Mineralogical Assessment <strong>of</strong> Creep-type Instability at Two GEO Report HK$87L<strong>and</strong>slip Sites, by T.Y. Irfan (1992), 136 p. (Reprinted, 1995). No. 13 (US$15)Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1990, by K.Y. Tang GEO Report HK$112(1992), 78 p. plus 1 drg. (Reprinted, 1995). No. 14 (US$17)Assessment <strong>of</strong> Stability <strong>of</strong> Slopes Subjected to Blasting GEO Report HK$75Vibration, by H.N. Wong & P.L.R. Pang (1992), 112 p. No. 15 (US$12)(Reprinted, 1995).Earthquake Resistance <strong>of</strong> Buildings <strong>and</strong> Marine Reclamation GEO Report HK$48Fills in Hong Kong, by W.K. Pun (1992), 48 p. (Reprinted, No. 16 (US$8.5)1995).Review <strong>of</strong> Dredging Practice in the Netherl<strong>and</strong>s, by S.T. GEO Report HK$76Gilbert & P.W.T. To (1992), 112 p. (Reprinted, 1995). No. 17 (US$12)Backfilled Mud Anchor Trials Feasibility Study, by C.K. Wong GEO Report HK$50& C.B.B. Thorley (1992), 55 p. (Reprinted, 1995). No. 18 (US$9)A Review <strong>of</strong> the Phenomenon <strong>of</strong> Stress Rupture in HDPE GEO Report HK$56Geogrids, by G.D. Small & J.H. Greenwood (1993), 68 p. No. 19 (US$9.5)(Reprinted, 1995).Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1991, by N.C. Evans GEO Report HK$111(1992), 76 p. plus 1 drg. (Reprinted, 1995). No. 20 (US$ 16.5)Horizontal Subgrade Reaction for Cantilevered Retaining Wall GEO Report HK$50Analysis, by W.K. Pun & P.L.R. Pang (1993), 41 p. (Reprinted, No. 21 (US$9.5)1998).Report on the Rainstorm <strong>of</strong> 8 May 1992, by N.C. Evans (1993), GEO Report HK$126109 p. plus 2 drgs. (Reprinted, 1995). No. 22 (US$20)Effect <strong>of</strong> the Coarse Fractions on the Shear Strength <strong>of</strong> GEO Report HK$126Colluvium, by T.Y. Irfan & K.Y. Tang (1993), 223 p. No. 23 (US$20)(Reprinted, 1995).The Use <strong>of</strong> PFA in Reclamation, by J. Premchitt & N.C. Evans GEO Report HK$52(1993), 59 p. (Reprinted, 1995). No. 24 (US$9)Report on the Rainstorm <strong>of</strong> May 1982, by M.C. Tang (1993), GEO Report HK$ 135129 p. plus 1 drg. (Reprinted, 1995). No. 25 (US$21)


Report on the Rainstorm <strong>of</strong> August 1982, by R.R. Hudson(1993), 93 p. plus 1 drg. (Reprinted, 1995).L<strong>and</strong>slips Caused by the June 1983 Rainstorm, by E.B. Choot(1993), 124 p. (Reprinted, 1995).Factors Affecting Sinkhole Formation, by Y.C. Chan (1994),37 p. (Reprinted, 1995).Classification <strong>and</strong> Zoning <strong>of</strong> Marble Sites, by Y.C. Chan(1994), 37 p. (Reprinted, 1995).Hong Kong Seawall Design Study, by P.M. Aas & A. Engen(1993), 94 p. (Reprinted, 1995).Study <strong>of</strong> Old Masonry Retaining Walls in Hong Kong, by Y.C.Chan (1996), 225 p.Karst Morphology for Foundation Design, by Y.C. Chan &W.K. Pun (1994), 90 p. plus 1 drg. (Reprinted, 1995).An Evaluation <strong>of</strong> the Suitability <strong>of</strong> Decomposed Granite asFoundation Backfill for Gravity Seawalls in Hong Kong, byE.B. Choot (1993), 34 p. (Reprinted, 1995).A Partial Factor Method for Reinforced Fill Slope Design, byH.N. Wong (1993), 55 p. (Reprinted, 1995).Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1992, by P.K.H. Chen(1993), 201 p. plus 2 drgs. (Reprinted, 1995).Methods <strong>of</strong> Test for Soils in Hong Kong for Civil EngineeringPurposes (Phase I Tests), by P.Y.M. Chen, 1996 Edition, 90 p.Creep, Stress Rupture <strong>and</strong> Hydrolysis <strong>of</strong> Polyester ReinforcedGeogrids, by J.H. Greenwood (1995), 67 p.Skin Friction on Piles at the New Public Works CentralLaboratory, by J. Premchitt, I. Gray & K.K.S. Ho (1994), 158 p.(Reprinted, 1995).GEO ReportNo. 26GEO ReportNo. 27GEO ReportNo. 28GEO ReportNo. 29GEO ReportNo. 30GEO ReportNo. 31GEO ReportNo. 32GEO ReportNo. 33GEO ReportNo. 34GEO ReportNo. 35GEO ReportNo. 36GEO ReportNo. 37GEO ReportNo. 38HKS118(US$17.5)HKS83(US$13)HKS40(US$7.5)HKS40(US$7.5)HKS68(US$11)HKS130(US$21)HK$118(US$17.5)HK$38(US$7)HK$50(US$9)HK$I67(US$25.5)1IK$22(US$5.5)HK$38(US$7.5)HK$97(US$16.5)Bibliography on the Geology <strong>and</strong> Geotechnical Engineering <strong>of</strong>Hong Kong to May 1994, by E.W. Br<strong>and</strong> (1994), 202 p.(Reprinted, 1995). (Superseded by GEO Report No.50)Hydraulic Fill Performance in Hong Kong, by C.K. Shen &K.M. Lee (1995), 199 p.GEO ReportNo. 39GEO ReportNo. 40HK$I18(US$19)HK$90(US$16)


Mineralogy <strong>and</strong> Fabric Characterization <strong>and</strong> Classification <strong>of</strong>Weathered Granitic Rocks in Hong Kong, by T.Y. frfan (1996),158 p.Performance <strong>of</strong> Horizontal Drains in Hong Kong, by R.P.Martin, K.L. Siu & J. Premchitt (1995), 109 p.Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1993, by W.L. Chan(1995), 214 p. plus 1 drg.General Report on L<strong>and</strong>slips on 5 November 1993 atMan-made Features in Lantau, by H.N. Wong & K.K.S. Ho(1995), 78 p. plus 1 drg.Gravity Retaining Walls Subject to Seismic Loading, by Y.S.Au-Yeung & K.K.S. Ho (1995), 63 p.Direct Shear <strong>and</strong> Triaxial Testing <strong>of</strong> a Hong Kong Soil underSaturated <strong>and</strong> Unsaturated Conditions, by J.K.M. Gan & D.G.Fredlund(1996),217p.Stability <strong>of</strong> Submarine Slopes, by N.C. Evans (1995), 51 p.Strength Development <strong>of</strong> High PFA Content Concrete, byW.C.^Leung & W.L. Tse (1995), 84 p.AAR Potential <strong>of</strong> Volcanic Rocks from Anderson RoadQuarries, by W.C. Leung, W.L. Tse, C.S. Mok & S.T. Gilbert(1995), 78 p.Bibliography on the Geology <strong>and</strong> Geotechnical Engineering <strong>of</strong>Hong Kong to March 1996, by E.W. Br<strong>and</strong> (1996), 111 p.Bibliography on Settlements Caused by Tunnelling to March1996, by E.W. Br<strong>and</strong> (1996), 70 p.Investigation <strong>of</strong> Some Major Slope Failures between 1992 <strong>and</strong>1995, by Y.C. Chan, W.K. Pun, H.N. Wong, A.C.O. Li & K.C.Yeo(l996),97p.Environmental Aspects <strong>of</strong> Using Fresh PFA as Fill inReclamation, by K.S. Ho & P.Y.M. Chen (1996), 46 p.Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1994, by W.L. Chan(1996), 161 p. plus 1 drg.GEO ReportNo. 41GEO ReportNo. 42GEO ReportNo. 43GEO ReportNo. 44GEO ReportNo. 45GEO ReportNo. 46GEO ReportNo. 47GEO ReportNo. 48GEO ReportNo. 49GEO ReportNo. 50GEO ReportNo. 51GEO ReportNo. 52GEO ReportNo. 53GEO ReportNo. 54HK$70(US$13.5)HK$65(US$17.2)HK$110(US$18.5)HK$64(US$17)HK$40(US$8)HK$65(US$12.5)HK$46(US$8.5)HK$60(US$10.5)HK$58(US$10)HK$45(US$9)HK$31(US$6.5)HK$44(US$8.5)HK$30(US$5.5)HK$70(US$13.5)


Conventional <strong>and</strong> CRS Rowe Cell Consolidation Test on SomeHong Kong Clays, by J. Premchitt, K.S. Ho & N.C. Evans(1996), 93 p.Application <strong>of</strong> Prescriptive Measures to Soil Cut Slopes, byH.N. Wong & L.S. Pang (1996), 52 p.Study <strong>of</strong> Rainfall Induced L<strong>and</strong>slides on Natural Slopes in theVicinity <strong>of</strong> Tung Chung New Town, Lantau Isl<strong>and</strong>, by C.A.M.Franks (1998), 102 p. plus 3 drgs. under preparation.Tsing Shan Debris Flow <strong>and</strong> Debris Flood, by J.P. King(1998), 215 p. plus 9 drgs. under preparation.Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1995, by C.K.L. Wong(1997), 125 p. plus 1 drg.Assessment <strong>of</strong> Geological Features Related to RecentL<strong>and</strong>slides in Volcanic Rocks <strong>of</strong> Hong Kong Phase 2 A - ChaiWan Study Area, by S.D.G. Campbell & N.P. Koor (1998),78 p. plus 6 drgs. under preparation.Factual Report on the November 1993 Natural TerrainL<strong>and</strong>slides in Three Study Areas on Lantau Isl<strong>and</strong>, by H.N.Wong, Y.M. Chen & K.C. Lam (1997), 42 p.Areal Extent <strong>of</strong> Intense Rainfall in Hong Kong 1979 to 1995,byA.W.Malone(1997),85p.A Review <strong>of</strong> Some Drained Reclamation Works in HongKong, by J.S.M. Kwong (1997), 53 p.A Study <strong>of</strong> Hydraulic Fill Performance in Hong Kong -Phase 2, by C.K. Shen, K.M. Lee & X.S. Li (1997)^265 p."Seismic Hazard Analysis <strong>of</strong> the Hong Kong Region, by C.F.Lee, Y.Z. Ding, R.H. Huang, Y.B. Yu, G.A. Guo, P.L. Chen &X.H. Huang (1998), 145 p. (Bilingual).GEO ReportNo. 55GEO ReportNo. 56GEO ReportNo. 57GEO ReportNo. 58GEO ReportNo. 59GEO ReportNo. 60GEO ReportNo. 61GEO ReportNo. 62GEO ReportNo. 63GEO ReportNo. 64GEO ReportNo. 65HK$35(US$7)HKS12(US$3.5)HK$264(US$39.5)HK$__(US$_JHKS70(US$14.5)HK.S296(US$43.5)HK$92(US$13.5)HK$43(US$8.5)HK$36(US$6.5)HK$150(US$25)HK$«0(US$15.5)Mineralogical <strong>and</strong> Fabric Characterization <strong>and</strong> Classification <strong>of</strong> GEO Report H K$ 106Weathered Volcanic Rocks in Hong Kong, by T. Y. Man (1998), No. 66 (LJS$ 17)113 p.


Assessment <strong>of</strong> Geological Features Related to RecentL<strong>and</strong>slides in Volcanic Rocks <strong>of</strong> Hong Kong Phase 2B~Aberdeen Study Area, by C.A.M. Franks, S.D.G. Campbell &W.W.L. Shum (1998), under preparation.The New Priority Classification Systems for Slopes <strong>and</strong>Retaining Walls, by C.K.L. Wong (1998), 117 p.Diagnostic Report on the November 1993 Natural TerrainL<strong>and</strong>slides on Lantau Isl<strong>and</strong>, by H.N. Wong, K.C. Lam &K.K.S. Ho (1998), 98 p. plus 1 drg. under preparation.Hong Kong Rainfall <strong>and</strong> L<strong>and</strong>slides in 1996, by C.K.L. Wong(1998), 84 p. plus 1 drg. under preparation.Site Characterisation Study - Phases 1 <strong>and</strong> 2, by N.P. Koor(1998), under preparation.Long-term Consolidation Tests on Clays from the Chek LapKok Formation, by D.O.K. Lo & J. Premchitt (1998), underpreparation.The Natural Terrain L<strong>and</strong>slide Study Phases I <strong>and</strong> n, by N.C.Evans, S.W. Huang & J.P. King (1998), under preparation.Natural Terrain L<strong>and</strong>slide Study the Natural Terrain L<strong>and</strong>slideInventory, by J.P. King (1998), under preparation.Geotechnical Area Studies Programme - Hong Kong <strong>and</strong>Kowloon (1987), 170 p. plus 4 maps.Geotechnical Area Studies Programme - Central NewTerritories (1987), 165 p. plus 4 maps.Geotechnical Area Studies Programme - West New Territories(1987), 155 p. plus 4 maps.Geotechnical Area Studies Programme - North West NewTerritories (1987), 120 p. plus 3 maps.Geotechnical Area Studies Programme - North New Territories(1988), 134 p. plus 3 maps.Geotechnical Area Studies Programme - North Lantau (1988),124 p. plus 3 maps.GEO ReportNo. 67GEO ReportNo. 68GEO ReportNo. 69GEO ReportNo. 70GEO ReportNo. 71GEO ReportNo. 72GEO ReportNo. 73GEO ReportNo. 74GASP IGASPHGASPIEGASP IVGASPVGASP VIHK$66(US$11.5)HK$240(US$40)HK$150(US$25)HK$150(US$25)HK$150(US$25)HK$150(US$25)HK$150(US$25)


Geotechnical Area Studies Programme - Clear Water Bay(1988), 144 p. plus 4 maps.GASP VIIHKS150(US$25)Geotechnical Area Studies Programme - North East NewTerritories (1988), 144 p. plus 4 maps.GASP VIIIHKS150(US$25)Geotechnical Area Studies Programme - East New Territories(1988), 141 p. plus 4 maps.GASP IXHK5150(US$25)Geotechnical Area Studies Programme - Isl<strong>and</strong>s (1988), 142 p.plus 4 maps.GASP XHK$15()(US$25)Geotechnical Area Studies Programme - South Lantau (1988),148 p. plus 4 maps.GASP XII IKS 150(US$25)Geotechnical Area Studies Programme - Territory <strong>of</strong> HongKong (1989), 346 p. plus 14 maps.GASP XIIHKS150(US$25)Geology <strong>of</strong> Sha Tin, by R. Addison (1986), 85 p.Geology <strong>of</strong> Hong Kong Isl<strong>and</strong> <strong>and</strong> Kowloon, by P.J. Strange &R.Shaw (1986), 134 p.GeologicalMemoir No. 1GeologicalMemoir No. 2HKS50(US$9)HKS78(US$12.5)Geology <strong>of</strong> the Western New Territories, by R.L. Langford,K.W. Lai, R.S. Arthurton & R. Shaw (1989), 140 p.GeologicalMemoir No. 3HK$


Geology <strong>of</strong> North Lantau Isl<strong>and</strong> <strong>and</strong> Ma Wan by R. J. Sewell Sheet Report Free&J.W.C. James (1995), 46p. No. 4Geology <strong>of</strong> Ma On Shan by RJ. Sewell (1996), 45p. Sheet Report FreeNo. 5Geological L<strong>and</strong>scapes <strong>of</strong> Hong Kong (1998), 61p. (Bilingual). - HK$13(San Tin : Solid <strong>and</strong> Superficial Geology (1:20 000 map) MapHGM20, HK$80(1989), 1 map. Sheet 2Sheung Shui : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1992), 1 map. Sheet 3Kat O Chau : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1993), 1 map. Sheet 4Tsing Shan (Castle Peak) : Solid <strong>and</strong> Superficial Geology Map HGM 20, HK$80(1:20 000 map) (1988), 1 map. Sheet 5Yuen Long : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1988), 1 map. Sheet 6Sha Tin : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1986), 1 map. Sheet 7Sai Kung Peninsula : Solid <strong>and</strong> Superficial Geology (1:20 000 Map HGM 20, HK$80map) (1989), 1 map. Sheet 8Tung Chung : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1994), 1 map. Sheet 9Silver Mine Bay : Solid <strong>and</strong> Superficial Geology (1:20 000 Map HGM 20, HK$80map) (1992), 1 map. Sheet 10Hong Kong <strong>and</strong> Kowloon : Solid <strong>and</strong> Superficial Geology Map HGM 20, HK$80(1:20 000 map) (1986), 1 map. Sheet 11Clear Water Bay : Solid <strong>and</strong> Superficial Geology (1:20 000 Map HGM 20, HK$80map) (1989), 1 map. Sheet 12Shek Pik : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HKS80(1995), 1 map. Sheet 13Cheung Chau : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HKS80(1995)"l map. Sheet 14


Hong Kong South <strong>and</strong> Lamma Isl<strong>and</strong> : Solid <strong>and</strong> Superficial Map HGM 20, HK$80Geology (1:20 000 map) (1987), 1 map. Sheet 15Waglan Isl<strong>and</strong> : Solid <strong>and</strong> Superficial Geology (1:20 000 map) Map HGM 20, HK$80(1989), 1 map. Sheet 16San Tin: Solid Geology (1 : 20 000 map) (1994), 1 map. Map HGM20S HK$80Lo Wu : Superficial Geology (1:5 000 map) (1990). 1 map. Map HGP 5A. HKSI00Sheet 2-NE-DLo Wu : Solid Geology (1:5 000 map) (1990), 1 map. Map HGP 5B. HK$10()Sheet 2-NE-DDeep Bay: Superficial Geology (1:5 000 map) (1989), 1 map. Map HGP 5A. HK$! 00Sheet 2-SW-CDeep Bay : Solid Geology (1:5 000 map) (1989), 1 map. Map HGP 5B, HK$100Sheet 2-SW-CShan Pui: Superficial Geology (1:5 000 map) (1989), 1 map. Map HGP 5A, HK$ 100Sheet 2-SW-DShan Pui: Solid Geology (1:5 000 map) (1989), 1 map. Map HGP 5B, HK$100Sheet 2-SW-DMai Po: Superficial Geology (1:5 000 map) (1990), 1 map. Map HGP 5A, HK$100Sheet 2-SE-AMai Po : Solid Geology (1:5 000 map) (1990), 1 map. Map HGP 5B. HK$10()Sheet 2-SE-ALok Ma Chau : Superficial Geology (1:5 000 map) (1990), Map HGP 5 A, HK$1001 map. Sheet 2-SE-BLok Ma Chau: Solid Geology (1:5 000 map) (1990), 1 map. Map HGP 5B, HK$1()()Sheet 2-SE-BMan Kam To : Superficial Geology (1:5 000 map) (1990), Map HGP 5A, HK$l()01 map. Sheet 3-NW-CMan Kam To: Solid Geology (1:5 000 map) (1990), 1 map. Map HGP 5B, HK$I00Sheet 3-NW-CTin Shui Wai : Superficial Geology (1:5 000 map) (1989), Map HGP 5A, HK$1001 map. Sheet 6-NW-A


Tin Shui Wai: Solid Geology (1:5 000 map) (1989), 1 map. Map HGP 5B, HK$100Sheet 6-NW-AYuen Long : Superficial Geology (1:5 000 map) (1989), 1 map. Map HGP 5A, HK$ 100Sheet 6-NW-BYuen Long : Solid Geology (1:5 000 map) (1989), 1 map. Map HGP 5B, HK$ 100Sheet 6-NW-BHung Shui Kiu : Superficial Geology (1:5 000 map) (1989), Map HGP 5A, HK$1001 map. Sheet 6-NW-CHung Shui Kiu : Solid Geology (1:5 000 map) (1989), 1 map. Map HGP 5B, HK$100Sheet 6-NW-CMuk Kiu Tau : Superficial Geology (1:5 000 map) (1990), Map HGP 5A, HK$1001 map. Sheet 6-NW-DMuk Kiu Tau : Solid Geology (1:5 000 map) (1990), 1 map. Map HGP 5B, HK$100Sheet 6-NW-DTsuen Wan (Part): Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(1995), 1 map. Sheet 6-SE-DMa On Shan : Solid Geology (1:5 000 map) (1996), 1 map. Map HGP 5B, HK$100Sheet 7-NE-D,C (part)Click Lap Kok : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(1993), 1 map. Sheet 9-NE-C/D'lung Chung Wan : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(1995), 1 map Sheet 9-SE-APok To Yan : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(1W7). 1 map.Sheet 9-SE-BLantau Peak : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(19%), 1 map. Sheet 9-SE-CSunset Peak : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100( 1996), 1 map. Sheet 9-SE-DYam O Wan : Solid & Superficial Geology (1:5 000 map) Map HGP 5, HK$100(1995), 1 map. Sheet 10-NW-B


Siu Ho : Solid & Superficial Geology (1:5 000 map) (1994), Map HGP 5, HK$1001 map. Sheet 10-NW-CChok Ko Wan (Penny's Bay) : Solid & Superficial Geology Map HGP 5, HK$100(1:5 000 map) (1994), 1 map. Sheet 10-NW-DMa Wan : Solid <strong>and</strong> Superficial Geology (1:5 000 map) (1994), Map HGP 5, HK$ 1001 map. Sheet 10-NE-ATsing Yi : Solid & Superfical Geology (1:5 000 map) (1995). Map HGP 5. HK$ 1001 map. Sheet 10-NE-B/DPa Tau Kwu : Solid <strong>and</strong> Superficial Geology (1:5 000 map) Map HGP 5. HK$l00(1994), 1 map. Sheet 10-NE-CTai Ho : Solid <strong>and</strong> Superficial Geology (1:5 000 map) (1995), Map HGP 5. H K$ 1001 map. Sheet 10-SW-AORDERING INFORMATION IS GIVEN ON THE NEXT PAGE

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!