Pebble Bed Modular Reactor Demonstration Power Plant (Pbmr Dpp) Environmental Impact Assessment and Environmental Management Programme

PEBBLE BED MODULAR REACTOR DEMONSTRATION POWER PLANT (PBMR DPP) ENVIRONMENTAL IMPACT ASSESSMENT AND ENVIRONMENTAL MANAGEMENT PROGRAMME

SPECIALIST STUDY FOR ENVIRONMENTAL IMPACT REPORT

SPECIALIST STUDY: GEOLOGY AND SEISMIC HAZARD

Council for Geoscience Report number: 2007-0277

CONTENTS ______Chapter Description Page

1 EXECUTIVE SUMMARY

2 INTRODUCTION

Description of Proposed Project Terms of Reference

3 BACKGROUND

Legislative Framework Assumptions and Limitations

4 DESCRIPTION OF THE DUYNEFONTEIN SITE AND SURROUNDING ENVIRONMENT

Geology Tectonics Palaeo-seismicity Seismic hazard

5 IMPACTS AND MITIGATION MEASURES 5.1 Project Impacts and Mitigation Measures

5.1.1 Project Impacts on the Environment 5.1.2 Mitigation Measures

5.2 Environmental Impacts and Mitigation Measures

5.2.1 Impacts on the Environment on the Project 5.2.2 Mitigation Measures

5 CONCLUSIONS

6 REFRENCES

GLOSSARY OF TERMS

LIST OF FIGURES

Figure 2.1.1: Position of PBMR plant and extent of regulatory radii.

Figure 4.1.1: Geological setting, structure and seismicity for the Koeberg site. Geology derived from the 1:250,000 digital database of the CGS.

LIST OF TABLES

Table 1. Geological investigations at Duynefontein(Appendix A). Table 2. Seismological investigations at Duynefontein(Appendix B)

LIST OF APPENDICES

Appendix A . Geological investigations.

Appendix B . Scope of work to determine mitigating measures to minimize geological impact on the project.

Appendix C . Seismological investigations.

Appendix D . Scope of work to determine mitigating measures to minimize seismological impact on the project.

1 EXECUTIVE SUMMARY

This report includes a specialist assessment of geological, structural, tectonic, palaeo-seismic and seismological data to be included in the Environmental Impact Assessment (EIA) Report to be compiled by ARCUS GIBB (Pty) Ltd. The report describes and assesses the scope of available data and investigations and outlines the uncertainties related to available data. The scope of investigations that must be undertaken to give a meaningful input into the EIA report for the Pebble Bed Modular Reactor Demonstration Power Plant (PBMR DPP) at the Koeberg Nuclear Power Plant Site (KNPPS) along the Western Cape coastline is also outlined.

As Duynefontein (Koeberg) houses an existing facility the site was well studied in the past. Detailed work will have to be undertaken if a new location is chosen on this site. The questions around the 1809 to 1810 seismic events and the existence of the Milnerton fault have to be further resolved.

Seismologically the site has an updated earthquake catalogue. The maximum expected earthquake was determined deterministically and the expected peak ground acceleration was determined probabilistically. The Peak Ground Acceleration (PGA) was further de-aggregated (decomposed) into its base dimensions, magnitude and distance. The current knowledge can be summarised as follows:

Koeberg Maximum Possible Earthquake in Seismotectonic Province (by 6.60 ± 0.3 procedure of Kijko (2004)) Deterministically calculated PGA 0.27 g ± 0.14

No detail work has been undertaken as yet on the site vicinity and site specific scale and the above table may change depending on the discovery of new palaeo-seismicity remnants.

Significant analyses were undertaken on the impacts of possible identified natural and man made effects on the site and can be summarised as follows:

Impact Consequence Probability Significance Confidence Vibratory low Improbable low High ground movement 1 With mitigation low Improbable low High Faulting- High Improbable High High earthquake 3 With mitigation low Improbable low High Tsunami 4 Medium Improbable Medium High With mitigation low Improbable low High Slope collapse low Improbable low High 5 With mitigation low Improbable low High

1 Mitigation measures have to be implemented on the following: • Possible short period tectonic activity in the existing geology, whether from collapse of the continental slope or other kinds of rock movement within a radius of 320 km. The PBMR DPP will have to withstand the expected maximum activity. • Possible movements along any of the known faults or a new fault within a radius of 320 km. The PBMR DPP will have to withstand the effects of the expected maximum movement. • Possible Tsunami flooding.

In conclusion, with the current knowledge of the seismic activity in the surrounding environment of the KNPS and the general geological environment, the KNPS site would be considered suitable for the location of the proposed PBMR DPP.

2 INTRODUCTION.

Background

Eskom has identified the Duynefontein site for the suitability for the establishment of the Pebble Bed Modular Nuclear Reactor Demonstration Power Plant (PBMR DPP) for an experimental power generation facility (Figure 2.1). The geological assessment forms part of the Environmental Impact Assessment (EIA) and must provide evaluations to obtain an estimate of the safe shutdown earthquake ground motion, the risk for deformation at or near the surface and to permit adequate engineering solutions to actual and potential geologic and seismic effects at the proposed site. This EIA report must also provide input for the seismic hazard analysis. The purpose of this contribution towards the Report is to; (a) state very briefly what is known about the site, and (b) what inputs must still be provided to complete the geoscientific contribution to the EIA.

Terms of Reference

Since the US Nuclear Regulatory Commission (USNCR) Standard Review Plan NUREG-800 is favoured, the responsibility lies in providing the required geoscientific information with specific reference to Chapters 2.5.1 to 2.5.5 of the NUREG-800 for Chapter 11 of a Site Safety Report (SSR). These requirements form the basis for the EIA report and entail on- and off-shore geological, geophysical, seismological and geotechnical investigations in progressively greater detail closer to the site. Radii of 320 km (regional), 40 km (semi-regional), 8 km (site vicinity) and 1 km (site specific) constrain the envelopes that describe the ever-increasing required detail of the investigations.

Figure 2.1: Geographical setting and regulatory radii for the Duynefontein Site

Legislative Framework

The project concerns a range of proposed activities that have been identified in the schedule of activities listed in terms of section 24(4)(a) and (d) of the National Environmental Management Act (1998, as amended) in Government Notice No. R. 387. Investigations required before environmental authorisation of these activities can be considered must follow the procedures outlined in regulations 27 to 36 of the Environmental Impact Assessment regulations, 2006.

The National Nuclear Regulator Act regulates the construction and running of nuclear power plants in South Africa.

Council for Geoscience (CGS) has been the past preferred supplier of geological and seismic hazard information and contract research as outlined in NSIP-SHA-014949#P1-P5. CGS as such has decided to use the US

3 Regulations for the probabilistic part of the Seismic Hazard Assessment (SHA) and that US Standards and practice be applied to the palaeoseismic- neotectonic investigations as well. This is because the US nuclear industry is considered the most advanced and the regulations more conservative as well as most readily understandable, tried and tested.

Assumptions and Limitations.

The descriptions and facts given here stem from published data and work undertaken by the CGS. In terms of the identification of faults and seismic risk the information represents the current knowledge and understanding based on a regional picture.

The site sensitivity on each of the impacts reflect only the current state of knowledge without incorporating the regulatory required nuclear licensing requirements detailed investigations, which still have to be undertaken.

3 METHODOLOGY.

The most important threat that impacts on the location and design of the PBMR DPP is the earthquake hazard and therefore great effort has to go into determining the so-called Seismic Hazard Assessment (SHA). The SHA methodology consists of estimating what the expected maximum earthquake frequency, size and ground movement spectra would be during the active and decommissioned life of the PBMR DPP. This can only be achieved through advanced statistical methodologies and although these methodologies assume that there is an incomplete earthquake catalogue the period of time in which earthquake occurrences have been recorded is, in a geological environment, extremely short. Therefore the study of palaeo-seismic (fossil- seismic) movement is very important, costly and time consuming, but absolutely necessary to boost the confidence level in the maximum estimates.

As the SHA shall make use of an earthquake catalogue for the region of study, such a catalogue, consisting of the existing national seismic data combined with the recordings of the Eskom micro-seismic network(s), must contain the locations, times of occurrence, the size (magnitude) of earthquakes together with their uncertainties. In addition, information on its completeness is required. If available, information on paleo-seismicity as well as historic seismicity must be included.

The peak ground accelerations (PGA) caused by an earthquake is related to the size of the event, the hypocentral distance, attenuation and site response. The prediction of site-specific ground motion for particular size earthquakes as a function of distance from the event has received attention from several authors. One technique that is used is to determine the model for the prediction of the expected ground motions and their variability at a site located

4 at a certain distance from a future earthquake of given magnitude by the logic tree approach (Scherbaum et al ., 2005). The main disadvantage of this technique is that it is based on the sentiment of experts. Its major punted advantage is that it presents a consensus estimate from possibly widely differing views and neutralises dissenting views in that it is presented as the mainstream thought. It also lessens the risk on decision makers as a committee of (normally) international experts is responsible for presenting the ultimate design parameters for a NPP. The supporters of the logic tree approach, focus their current research on making the technique more transparent, rigorous and repeatable (Scherbaum et al ., 2006). Bommer et al ., (2005) the use of logic trees for ground-motion prediction equations.

Due to a lack of strong motion data in South Africa the attenuation formulas for stable continental regions (Atkinson and Boore, 1995, 1997) are still being used. As the original equation does not adequately match the shape of the attenuation curve for small to moderate earthquakes, the coefficients of the attenuation formula were re-calculated by the CGS, so that they fit the values of PGA and response spectra significantly better. This is similar to what Scherbaum et al., (2006) are trying to achieve but from a different approach.

Bommer et al ., (2003) investigated the style of faulting in ground-motion prediction equations and Youngs et al ., (2003) presented two methodologies whereby the future fault rupture hazard is identified in a probabilistic way. The CGS methodology uses all existing seismological data to determine the regional and site specific hazard through a probabilistic approach. Where available, information specific to a fault can be included, and is dealt with through a Bayesian procedure.

SHA methodology

A SHA can be either deterministic or probabilistic . The most common procedure of deterministic SHA requires the designation of active seismic sources. If faults cannot be identified, the locations of possible earthquakes must be associated with aerial sources. These are spatial areas within which earthquake characteristics are designated to be uniform. The basis for the definition of seismic sources should be a seismo-tectonic model for the region of interest.

The logic tree approach can also be used to determine a probabilistic SHA and is increasing in popularity to the extend that it is recommended as the technique in Budnitz et al ., (1997) and re-emphasized in Savy et al., (2002). Apart from the advantages and disadvantages mentioned above it distinguishes between aleatory and epistemic uncertainties (See glossary for explanation). Bommer et al ., (2004) discuss the challenges in defining upper limits on earthquake ground motions and Bommer and Abrahamson (2006) ask the question why the logic tree application to SHA lead to increased hazard estimates and present mitigating factors.

The Council for Geoscience employs a probabilistic SHA (PSHA) methodology, called the Parametric-Historic PSHA (Kijko and Graham, 1998,

5 1999; Kijko, 2004), in which the knowledge of seismic sources or detailed geological information can be incorporated if available. However, the only requirement of the CGS SHA methodology is to have a seismic catalogue of the area under investigation, and it is thus not restricted by available geological information. The development of the PSHA by the CGS was dictated by the uncertainty and incompleteness of the seismic catalogues (which is often the case). The Parametric-Historic PSHA methodology was reviewed by the National Nuclear Regulatory Commission, who expressed their satisfaction thereof. This methodology was used by the CGS to assess the seismic hazard of potential Nuclear Power Plant (NPP) sites and the existing NPP site of Koeberg.

The procedure is free from the basic disadvantages characteristic of the traditional deductive and historic procedures. The approach is parametric and its application consists essentially of two steps. The first step is applicable to the area in the vicinity of the site for which knowledge on the seismic hazard is required. In this respect the procedure is similar to Cornell's deductive approach and requires an estimation of the area-specific parameters. The parameters are the area-specific mean seismic activity rate λ, the Gutenberg- Richter parameter b and the maximum regional magnitude mmax . The second step is applicable to a specified site , and consists of a maximum likelihood assessment of the parameters of amplitude distribution of the selected ground motion parameter.

Since in each step parameters are estimated through the maximum likelihood procedure, by applying the Bayesian formalism any additional geological or geophysical information (as well as all kinds of uncertainties) can easily be incorporated. The procedure is consequently capable of giving a realistic assessment of seismic hazard in areas of both low and high seismicity, including cases where the catalogues are incomplete. In the present form, the procedure allows assessment of seismic hazard in terms of peak ground acceleration, peak ground velocity, peak ground displacement or acceleration response spectra.

The parametric-historic approach closest in conception to that of the CGS approach (Kijko and Graham, 1998, 1999) Kijko (2004) is that of Frankel (1995), where seismic hazard analysis is performed for the central and eastern United States. Frankel’s approach has been adopted by Lapajne et al . (1997) in modeling seismic hazard in Slovenia. Recently, a similar approach has been applied in seismic hazard assessments of Central America and the Caribbean as part of the Global Seismic Hazard Assessment Program, IDNDR demonstration Project of the UN/International Decade of Natural Disaster Reduction (1999).

Treatment of uncertainties

Traditionally the standard deviation of the predicted value of PGA, or acceleration response spectra (ARS), represents the observed scatter that is not explained by the physical model (Toro et al ., 1997). In addition to the

6 aleatory uncertainty there is another source of uncertainty, which comes from the possibility of erroneous determination of the earthquake magnitude and hypocentral distance. This type of uncertainty is known as epistemic uncertainties. In general, the epistemic uncertainty is due to incomplete knowledge and data about the physics of the earthquake process. In principle, epistemic uncertainty can be reduced by the collection of additional information.

The CGS approach to SHA makes it possible to assess the required value of the maximum value of PGA at the site together with its uncertainty. The maximum acceleration is calculated as a median value of the distribution of accelerations, calculated at the critical distance and generated by the largest possible earthquake magnitude.

The 84% confidence level of the results is calculated by analogy with the normal distribution, for which the upper confidence limit, with the 84% confidence level, corresponds to the mean value increased by one standard deviation.

The CGS SHA approach has a built in conservatism in that the earthquake with magnitude mmax has no return period, i.e. it should not actually occur. The calculated median PGA is based on the mmax and also on a conservative attenuation relationship. In the light of the recommendations contained in different international codes and guides regarding the use of median or median plus one standard deviation PGA, the CGS recommends that, to avoid adding further conservatism, the median of PGA (at a pre-specified probability) should be used as input into the design of existing and future NPP, when the CGS SHA methodology is employed. When median plus standard deviation is to be used, e.g. as part of a sensitivity analysis, use of all information pertaining to uncertainties must be made, not just the uncertainty in attenuation, when calculating the standard deviation.

The SHA methodology is probabilistic and allows the calculation of mmax , which is the SSE (Safe Shutdown Earthquake) according to IAEA definition, PGA, probabilistic seismic hazard curves and spectral acceleration response spectra. A rigorous treatment of all uncertainties is followed throughout the procedure.

In its approach to PSHA, the CGS follows, in most cases, the main ideas supplied in the guides issued by the U.S. Office of Nuclear Reactor Regulation. Although the regulations/codes of the mentioned Office do not apply to South Africa, adherence to the principles stated therein is also strived for.

However the CGS is open to other techniques and due to the gaining popularity of the logic tree approach for SHA will implement the logic tree approach parallel to its own Parametric-Historic PSHA.

Benchmarking of results

7 In order to check on the correctness of the results produced by the Parametric-Historic Probabilistic Seismic Hazard Assessment (PSHA) procedure, the CGS performed a benchmark study using synthetic and real data. Furthermore, in order to comply with international standards regarding SHA, the CGS also follows the main ideas on SHA as advocated by the U.S. Nuclear Regulatory Commission. These two aspects are considered important in order to provide assurance that any SHA undertaken by the CGS will be able to withstand international scrutiny.

Eskom requested the CGS to perform a benchmark study (Retief and Kijko, 1999) in which the results obtained with its methodology was compared with the results obtained with other approaches of good international standing. In addition, the absolute values obtained through the CGS procedure were compared with the known values in a synthetic simulation. The benchmark study indicated that the CGS estimates, for all hazard parameters, differ minimally from the “true” synthetic values. With respect to the comparison with another approach of good international standing, it was evident that the most important difference consists in the estimation of the maximum regional magnitude mmax . In the case of the CGS technique, the estimation forms an integral part of the methodology with mmax being an output of the calculations performed, whereas in the alternative approach the value of mmax is mainly based on expert opinion and to a lesser extent on available data. It was also evident that the values of Peak Ground Acceleration (PGA) are more conservative for the CGS approach, mainly as a result of the usage of different (more appropriate) attenuation relations. The Gutenberg-Richter b- values were in close agreement.

Codes and guides followed in the CGS SHA

Two types of documents regarding criteria for the safe functioning of a NPP exist. The one type contains regulations/codes which have the force of law and are formally issued by government or by a regulatory body legally empowered to do so, while the other kind, usually known as guides/guidance documents , contains recommendations relating to the fulfilment of basic requirements of a NPP as stipulated in the regulations/codes. According to the document "Status and Content of Seismic & Geologic Siting Criteria Revision, 10 CFR Part 100 , Appendix A ", guidance documents describe procedures acceptable to the NRC (United States Nuclear Regulatory Commission) staff on how to carry out the determinations required by the regulations/codes. This division between regulations and guides is also followed by the IAEA. Their Safety Guide category of publications gives recommendations, on the basis of international experience, relating to the fulfilment of basic requirements.

The U.S. regulations pertaining to NPP are put under Title 10 – Energy , of their Code of Federal Regulations (CFR). Title 10 is subdivided into different parts, and a particular part is referred to as e.g. 10 CFR 100 (part 100 of the CFR, Title 10 ). In addition, they have a series of guides, referred to as regulatory guides. We found that the U.S. codes and guides are the most comprehensive relative to that of other countries or organisations, and the

8 guides are more specific. The NUREG 800 guide can be followed in the SHA with respect to the geologic and seismic investigations to be performed in the application for a licence to establish/operate a nuclear power plant. However, it must be noted that this guide is not primarily aimed at guidance in PSHA calculations. The NUREG/CR-6372 guide provides the most up-to-date ideas regarding PSHA, and gives by far the most comprehensive description of a PSHA. Regulatory Guide 1.165 also deals mainly with PSHA, although not to the extent of NUREG/CR-6372 . However, it compliments NUREG/CR-6372 .

The CGS is in the process of standardising on the usage of NUREG/CR-6372 and Regulatory Guide 1.208 in their PSHA. Although the current CGS Parametric-Historic procedure, used by the CGS, is not in exact agreement with the methodology proposed by these guides, the same parameters can be calculated by the CGS methodology. The CGS methodology is constantly being updated as new versions of these guides are released, in order to be able to calculate all required parameters. A short description of the mentioned family of codes and guides that originated, and is followed in the United States of America, follows.

NUREG 800 ∗ – Standard Review Plan (Revision 2 – July 1981)

This Standard Review Plan is intended to guide the U.S. Office of Nuclear Reactor Regulation staff responsible for the review of applications to construct and operate nuclear power plants. "Standard Review Plans are not substitutes for regulatory guides or the U.S. Nuclear Regulatory Commission's (NRC) regulations and compliance with them is not required".

The applicable rules and basic acceptance criteria pertinent to the areas of the Standard Review Plan are:

10 CFR Part 50, Appendix A, "General Design Criteria for Nuclear Power Plants", General Design Criterion 2 – "Design Bases for Protection Against Natural Phenomena";

10 CFR Part 100, "Reactor Site Criteria";

10 CFR Part 100, Appendix A, "Seismic and Geologic Siting Criteria for Nuclear Power Plants".

The following regulatory guides provide information, recommendations and guidance and in general describe a basis acceptable for implementing the requirements General Design Criterion 2, Part 100, and Appendix A to Part 100: Regulatory Guide 1.132, "Site Investigations for Foundations of Nuclear Power Plants";

∗ Formerly NUREG-75/087

9 Regulatory Guide 4.7, "General Site Suitability Criteria for Nuclear Power Stations".

REGULATORY GUIDE 1.165 – IDENTIFICATION AND CHARACTERIZATION OF SEISMIC SOURCES AND DETERMINATION OF SAFE SHUTDOWN EARTHQUAKE GROUND MOTION (1997)

This guide has been developed to provide general guidance on procedures acceptable to the NRC for (1) conducting geological, geophysical, seismological, and geotechnical investigations, (2) identifying and characterising sources, (3) conducting probabilistic seismic hazard analyses, and (4) determining the SSE for satisfying the requirements of 10 CFR 100.23 (i.e. 10 CFR 100 paragraph 23) . The information collections contained in this regulatory guide are covered by the requirements of 10 CFR Part 50 .

NUREG/CR-6372 Guide – Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts, Main Report (April 1997)

The project resulting in this document was directed towards providing methodological guidance on how to perform a PSHA, and was prepared by the Senior Seismic Hazard Analysis Committee (SSHAC), supported by a large number of other experts working under the Committee's guidance, under contract to Lawrence Livermore National Laboratory. It was co-sponsored by the U.S. Nuclear Regulatory Commission, the U.S. Department of Energy, and the Electric Power Research Institute.

NUREG-1.208 A Performance-Based Approach to Define the Site-Specific Earthquake Ground Motion

The purpose of this regulatory guide is to provide guidance on the development of the site-specific ground motion response spectrum. This represents the first part of the assessment of the SSE for a site as a characterization of the regional and local seismic hazard. It provides an alternative for using the requirements of 10 CFR Part 100.

4 DESCRIPTION OF THE DUYNEFONTEIN (KOEBERG) SITE AND SURROUNDING ENVIRONMENT BASELINE.

The geological and tectonic setting of the site and presence of faults or other potentially seismogenic sources in the 320 km radii from the sites are covered in De Beer (2004, 2005, 2006a and b). No detail neotectonic or palaeoseismic investigation to investigate the possibility of movement capability on each known fault inside the regional radius has been undertaken. Regional and detailed map compilations are available for Duynefontein and were completed for the purpose of the Koeberg facility.

4.1 Geology

10 The existing Nuclear Power Plant (NPP) at Koeberg is situated on Neoproterozoic rocks of the Malmesbury Group, intruded by the late Neoproterozoic Cape Granite Suite and Cretaceous dolerite dykes (Figure 4.1.1). Some 40 km to the south is the high topography of the Cape Peninsula, composed of the overlying Palaeozoic rocks of the Table Mountain Group. Most of the coastal plain around the site is covered with Cenozoic sand. Exposures of Pliocene-Pleistocene marine deposits and their overlying aeolianites are rare, being often restricted to low beach cliffs, such as at Springfontyn. Pre-Cenozoic basement rocks are cut by intense Pan-African brittle-ductile shear zones and, in places, by regional NW striking brittle shear zones. The Cape Peninsula outlier contains structural characteristics of both the western branch and the syntaxis of the Cape Fold Belt. The closest known fault of the latter type is the Mamre fault (see De Beer, 2005), whilst another such possible shear zone, tentatively called the Milnerton fault, has been proposed (see Dames and Moore, 1976) to occur between Bloubergstrand and Cape Town (Section 8.2.4.10). The existence of the Milnerton fault is not yet proven.

In the off-shore a number of formerly unknown faults have been recently identified and thought to be of cretaceous age. The offshore fault nearest to Koeberg NPS, the so called Koeberg fault falls just within the regulatory 8km site radius. Mesozoic faults in the Cape Peninsula strike predominantly WNW. Dolerite dykes have been intruded along these fractures in Early Cretaceous times; these dykes are responsible for many of the magnetic anomalies seen in geophysical surveys, and occur very close to the KNPS.

Figure 4.1.1: Geological setting, structure and seismicity for the Koeberg site. Geology derived from the 1:250,000 digital database of the CGS.

4.2 Tectonics

The existing Koeberg NPS is located ~28 km north of Cape Town. Its 320 km regulatory radius includes both Bitterfontein and Oudtshoorn, and implies that its regional area of investigation contains some of the most faulted parts of the Cape Fold Belt, namely the western branch and the syntaxis, with current prominent seismicity in the Ceres–Tulbagh area. Additionally, it lies within 20 km of one of the most important NW-SE trending zones of faulting in the SW Cape, namely the Vredenburg-Stellenbosch-Colenso fault zone and its related faults, many of which are of appreciable displacement. These faults have been active from the Saldanian Orogeny (ca. 550 Ma – 500 Ma ago) to the Mesozoic breakup of Gondwana, and should probably still be regarded as a potential threat to the Koeberg site. The Colenso fault (Schoch, 1976) is the best known of them and ties up with the Kalbaskraal fault.

The Mamre fault strikes northwestwards from Mamre towards Yzerfontein. The Colenso and Mamre faults both put Cape Granite Suite against Malmesbury Group rocks, implying appreciable, but unknown vertical displacements, and suggesting that the Darling hills represent a horst block. The Darling fault does not separate Malmesbury Group from Cape Granite Suite, but its wide mylonite zone testifies to its regional importance. The Mamre fault may tie up with the Klipheuwel fault, which itself may actually continue further north-westwards, implying that it may pass within 14 km east of the Koeberg site. The nearest on land, proven faults to the SW of Koeberg are those displacing Table Mountain Group rocks in the Cape Peninsula some 30 km away from Koeberg. In the off-shore the Koeberg fault passes within 7.5 km of the NPP. The aeromagnetic study of Day (1986) revealed the presence of many NW-SE striking magnetic anomalies in the area between Koeberg and False Bay. Most of these are probably dolerite dykes of the False Bay Swarm (Reid et al., 1991) as exposed in outcrops along the peninsula coastline, but as they trend in exactly the same direction as faults in the Cape Peninsula, some of them might have intruded along pre-existing faults. A recent off-shore magnetic survey has shown displacement of dolerite dykes possibly attributable to the postulated Milnerton fault.

4.3 Palaeo-seismics

Liquefaction and intense ground deformation in the area between Melkbosstrand and Cape Town during the 1809 event are well known from historical data, but the cause of the earthquake remains uninvestigated to this day; no new information could be acquired during the regional investigations performed by De Beer (2005; 2006b). Extensive housing and industrial development in those areas necessitates that geophysical investigations and further palaeoseismic work be performed in the area without delay. In the mean time, a fault capable of creating another M 6.3 event should be inferred to pass within 10 km offshore of the Koeberg site.

Apart from a dolerite dyke displacement of unknown post cretaceous age no new data on this hazard (now termed the “Milnerton seismic source”, E.

14 Hattingh, pers. com., 2005) were acquired during this investigation (see De Beer, 2005). Reliable evidence for a large earthquake with an intensity of VIII, and M L 6.3 (Brandt et al., 2005) having occurred in 1809 within 25 km of Koeberg comes from historical records of its secondary effects (Von Buchenröder, 1830; Hartnady, 2003). The closest position to Koeberg where liquefaction features were reported is at Blauweberg’s Vlei.

Whatever the cause of the earthquake, its effects imply that peak ground accelerations (for M 6 proximal events) between 0.2 and 0.3g were attained (Talwani and Gassman, 2000) 11 km south of Koeberg.

Dames and Moore (1976) concluded that enough circumstantial evidence exists for the presence of a NW striking fault offshore of Koeberg but that it does not come closer than 8 km to the site. The author is of the opinion that such a fault could pass anywhere between 7 and 10 km offshore of Koeberg (the newly recognised Koeberg fault passes 7.5 from the KNNP) and the fact that liquefaction was reported at Blauweberg’s Vlei and at Milnerton resulted primarily from the presence of geologically susceptible materials in those areas. No new research has been performed to confirm or refute the presence of the fault or its point of closest approach to the site. The inference that the event happened closer to Milnerton than to Koeberg is based on the reported damage to the farmhouse at Jan Biesjes Kraal. The seismic hazard model should therefore take into account the possibility that a fault capable of producing an event at least equal in size to the 1809 event of inferred M 6.3 magnitude, and with a minimum recurrence interval of about 200 years, is located about 8 km SW of Koeberg.

It would also have been possible for a large event on the Mamre fault (Figure 24; De Beer, 2005), which may come to within 17 km of the site (De Beer, 2005), to have caused the liquefaction. However, the Moravian mission at Groenkloof, which became Mamre, was already in existence at the time and the area inhabited by a substantial number of Khoi people. It would therefore have been unlikely that a large earthquake on the Mamre fault would have gone unnoticed by the population of Mamre.

The Vredenburg–Stellenbosch fault zone occurs within 25 km of the site and although there is currently no evidence of it having been active in Quaternary times the presence of extensive sand cover and intense cultivation in the area can be expected to non-preservation of fault scarps. The fault system is extensive enough to be able to generate events in excess of M 7 and should therefore be taken into account in any seismic hazard assessment for Koeberg.

The only other evidence of palaeoseismic importance to the Koeberg site is minor faulting in Pleistocene aeolianites at Saldanha (De Beer, 2005) which is both too far away from Koeberg and too difficult to interpret with confidence.

There is no evidence of substantial tectonic deformation in available exposures of the post-Early Pliocene to pre-Late Pleistocene Springfontyn Formation west of Koeberg (3.6 Ma–200,000, Roberts, 2006) but exposures are discontinuous and uncertainties therefore exist as to how representative this evidence is.

The observation that a postulated NNW to NW striking offshore “Velddrif fault” (see Section 4.4.5) may currently be seismically active should be taken as a warning that such Mesozoic “geologically old” faults may not be ignored in terms of seismic risk.

4.4 Seismic Hazard

The maximum possible earthquake for this region calculated by the Parametric-Historic Procedure (Kijko and Graham, 1998, 1999) is expected to be M 6.60±0.3 and the deterministically calculated Peak Ground Acceleration is 0.27 g ±0.14

5 IMPACTS AND MITIGATION MEASURES

5.1. Possible vibratory ground motion of the site.

Description: Vibratory motion referred to as “ground motion” by seismologists may occur as a consequence of a seismic event and CGS will provide parameters to enable the proper design to counter the ground motion effects.

Proposed mitigation measures: Parameters provided that are to be used as design input for determining the Safe Shutdown Earthquake Ground Motion (SSEGM) while the site is active as well the regulatory period after its decommissioning. Facility is designed to withstand the maximum expected PGA and earthquake magnitude. Foundations of the facility to be sunk into solid bedrock.

Assessment of Impact: Impact Nature Intensity Extent Duration Probability Non- Confidence Revers Vibratory negative Very low Local Long Improbable low High ground movement With neutral Insig- local short Improbable low High mitigation nificant

Impact Nature Consequence Probability Confidence Vibratory negative low Improbable High ground movement With mitigation neutral low Improbable High

Impact Consequence Probability Significance Confidence Vibratory low Improbable low High ground movement With mitigation low Improbable low High

16 Acceptability: Vibration movement does not constitute a fatal flaw from an environmental or legal perspective if the mitigation measure is followed.

Cumulative Impacts: None are foreseen.

Indirect Impacts: None are foreseen.

5.2. Short period tectonic changes in the existing geology, whether from rock fall or other kinds of rock movement within a radius of 230 km i.e sediment displacement along the continental shelf slope.

Description: Short period changes in geology include movement along known or new faults which sometimes cause small earthquakes but larger rock fall in mountainous regions, sand boils or movement of large volumes of sediment in the ocean, resulting in seismic shockwaves and aftershocks being transmitted with velocities and amplitudes dependent on the rock media through which they travel. They are natural phenomena, impossible to predict. Movements of sediment down the continental slope can trigger a tsunami.

A PBMR can go with much more difficulty into a core meltdown in the sense that can happen to conventional reactors. However a tilting of the reactor vessel bed could cause unexpected results, starting a spontaneous reaction. Apart from the reactor, spent uranium material is stored on the Koeberg site and could be displaced. A large earthquake could cause rock movement (acceleration) with serious effects.

Assessment of Impact: Impact Nature Intensity Extent Duration Probability Non- Confidence Revers Faulting- negative High National long Improbable high High earthquake With neutral low Local short Improbable low High mitigation

The 1) nature of an acceleration faster than what the vessel is designed for would be of a negative nature, the 2) intensity high, and the rest of the parameters follow logically. The mitigating factor here is to supply the peak ground acceleration spectrum expected in the reactors life time.

Impact Nature Consequence Probability Confidence Faulting- negative High Improbable High earthquake With mitigation neutral low Improbable High

Impact Consequence Probability Significance Confidence Faulting- High Improbable High High earthquake With mitigation low Improbable low High

Acceptability: The possibility of the PBMR DPP experiencing an earthquake or ground movement during its active and decommissioned life does not constitute a fatal flaw from an environmental or legal perspective if the mitigation measures are followed and the facility is designed to accommodate the maximum expected earthquake or ground motion.

Cumulative Impacts: None are foreseen.

Indirect Impacts: None are foreseen.

5.3. Movements along any of the known faults or a new fault within a radius of 320 km.

Description: Ridge push from the mid Atlantic Ridge causes the African Plate to be pushed north and north-eastwards causing stress build-up on the African Plate. Therefore stress release causes movement along known or new faults at surface or rock stress release at depth resulting in earthquakes with noticeable to severe ground movement especially in unconsolidated media, resulting in seismic shockwaves and aftershocks being transmitted with velocities and amplitudes dependent on the rock media through which they travel. They are natural phenomena, impossible to predict. Earthquakes can trigger a tsunami.

Proposed mitigation measures: To provide the expected maximum capable frequency dependent Peak Ground Acceleration (PGA) seismic design parameters based on rock movement events in the immediate past 300 000 years and present. These parameters are to be used as design input for determining the Safe Shutdown Earthquake Ground Motion (SSEGM) while the site is active as well the regulatory period after its decommissioning. CGS to provide the best estimate for the SSEGM the earthquake catalogue i.t.o. palaeo-seismic events should be completed.

Impact Nature Consequence Probability Confidence Faulting- High Improbable High earthquake With mitigation low Improbable High

Impact Consequence Probability Significance Confidence Faulting- High Improbable High High earthquake With mitigation low Improbable low High

Cumulative Impacts: None are foreseen.

Indirect Impacts: None are foreseen.

5.4. Tsunami flooding.

Description: Faulting or sediment movement causes water displacement resulting in a water wave with substantial larger amplitude than encountered due to for instance spring tide. The sea overflowing the PBMR DPP facility will probably not cause the reactor vessel to be damaged, but could dislodge spent fuel if not housed in reinforced concrete facilities and wash it back to sea with devastating consequences, spent uranium fuel being “hot” for at least 10 000 years. The last tsunami on the west coast was in 1969 and was recorded in “Die Burger” of Saturday 19 March 2005.

Proposed mitigation measures: Study the effects of possible earthquakes in the Indian and south Atlantic oceans on the South African coastline. Study especially Quaternary deposits for evidence of palaeo-tsunami events. Design to exceed most likely worst case scenario. The reactor should be situated above the expected tsunami water level during the reactors life time. Spent fuel to be stored such that it can not easily be swept away by receding waters. Nothing is known about tsunami impacts on our coast and a complete study of this will have to be undertaken to determine the reasonable expected maximum tsunami effect occurring with spring tide.

Assessment of Impact: Impact Nature Intensity Extent Duration Probability Non- Confidence Revers Tsunami negative Medium Regional long Improbable high High With neutral low Local short Improbable medium High mitigation

19 Any radioactive spillage from spent high level waste will have a 1) negative impact on the environment, will be of 2) high intensity , will have a 3) regional to even national extent (uranium is chemically very mobile and fish and other species could be contaminated over a wide area), the 4) duration of the impact would be very long, 5) probability of occurring improbable and the 6) non-reversibility high.

Impact Nature Consequence Probability Confidence Tsunami negative Medium Improbable High With mitigation neutral low Improbable High

Impact Consequence Probability Significance Confidence Tsunami Medium Improbable Medium High With mitigation low Improbable low High

Acceptability: Tsunami are natural events, unpredictable and do not constitute a fatal flaw from an environmental or legal perspective if the mitigation measure is followed.

Cumulative Impacts: None are foreseen.

Indirect Impacts: None are foreseen.

5.5. Collapsing rock slopes within a radius of 8 km.

Description: No collapsing rock formations or high mountains are present within 8 km of the site.

Proposed mitigation measures: Foundations of the structures to be sunk into solid bedrock where possible.

Assessment of Impact: Impact Nature Intensity Extent Duration Probability Non- Confidence Revers Rock negative Insig- local long Improbable low High movement nificant With neutral Insig- local short Improbable low High mitigation nificant

The likelihood of this happening is extremely low.

Impact Nature Consequence Probability Confidence Rock negative low Improbable High movement With mitigation neutral low Improbable High

Impact Consequence Probability Significance Confidence Rock low Improbable low High movement With mitigation low Improbable low High

Acceptability: Collapsing rock formations do not constitute a fatal flaw from an environmental or legal perspective if the mitigation measure is followed.

Cumulative Impacts: None are foreseen.

Indirect Impacts: None are foreseen.

The presented results reflect current knowledge without actually having investigated the proposed site for past palaeo-seismic events within the regulatory 8 km and 1 km radii.

To provide a more complete historic seismic catalogue a number of tasks still have to be completed and some of them have already been alluded to. These tasks have been separated as geological and seismological tasks and are given in Tables 1 and 2. Although Seismology is a branch of Geophysics the term “geophysical studies” in the tables refer to all other methods and techniques in Geophysics, with the exception of Seismology.

Regional and semi-regional geological information has been collected and studied but since the detailed work done for the KNPS no new geological work within the 8 and 1 km radius of the proposed sites has been undertaken. Where historic data at a better resolution than the 1:50 000 scale maps are available this data has to be updated.

To summarise the status quo in terms of geology, Table 1 in Appendix A provides an overview of what has been done at the Koeberg site. Appendix B lists the work that still needs to be done at the site.

The Probabilistic Seismic Hazard Assessment (PSHA) depends in part on the historical record of ground movement before the use of seismometers and depending on what further evidence is located in the site vicinity and site specific locations. Therefore, the PSHA and its derivatives will have to be updated. In a similar vein, Table 2 in Appendix C summarise the status quo for Seismology and Appendix D lists the work that still needs to be done.

6 CONCLUSIONS

This solicited EIA report presents specialist assessments of geological, structural, tectonic, palaeo-seismic and seismological data to be included in the EIA Report, to be compiled by ARCUS GIBB (Pty) Ltd. The report describes and assesses the scope of published data and investigations and outlines the uncertainties related to available data. The scope of investigations that must still be undertaken requires detailed investigations at the site with respect to tectonics, palaeo-seismicity, continued monitoring of current seismicity, finding palaeo-tsunami evidence.

Duynefontein (Koeberg site) houses an existing facility, and the site was well studied in the past. Detailed work will have to be undertaken if a new location is chosen on this site. The questions around the 1809 to 1810 seismic events and the existence of the Milnerton fault have to be resolved.

Seismologically the site has an updated earthquake catalogue. The maximum expected earthquake was determined deterministically for the site and expected peak ground acceleration was determined probabilistically. The PGA was further de-aggregated (decomposed) into its base dimensions, magnitude and distance. The current knowledge can be summarised as follows:

Koeberg Maximum Possible Earthquake in Seismotectonic 6.60 ± 0.3 Province (by procedure of Kijko (2004)) Deterministically calculated PGA 0.27 g ± 0.14 (Bejaichund et al., 2005; 2006a,b)

No detail work has been undertaken as yet on a site vicinity and site specific scale and the above table can change depending on the discovery of new palaeo-seismicity remnants.

A direct impact analysis was undertaken on the site and can be summarised as follows:

Impact Nature Intensity Extent Duration Probability Non- Confidence Revers Vibratory negativ low Local Long Improbable low medium ground e movement With mitigation neutral Insig- local short Improbable low medium nificant Rock movement negativ High Regional long Improbable high High e With mitigation neutral low Regional short Improbable low High Faulting- negativ High National long Improbable high High earthquake e With mitigation neutral low Local short Improbable low High Tsunami negativ Low Regional long Improbable high High e With mitigation neutral low Local short Improbable low High

Mitigation measures have to be implemented on the following: • Short period tectonic changes in the existing geology, whether from rock fall or other kinds of rock movement within a radius of 230 km. Completed project will have to withstand these changes. • Movements along any of the known faults or a new fault within a radius of 320 km. Completed project will have to withstand these changes.

22 • Tsunami flooding.

In conclusion, with the current knowledge of the seismic activity in the surrounding environment of the KNPS and the general geological, the KNPS site would be considered suitable for the location of the proposed PBMR DPP.

7 REFERENCES

Bejaichund, M., Kijko, A. and Hattingh, E., 2005 . Re-Assessment of the Seismic Hazard Parameters for the site of Koeberg NPP and the NSIP sites of Bantamsklip and Thyspunt. Volume 1: Updating of the SHA with recent seismic event catalogues. Volume II: Integration of Palaeoseismic and Neotectonic Information into the SHA. Council for Geoscience, Report No 2005-0274.

Bejaichund, M., Kijko, A. and Hattingh, E., 2006a. Sensitivity analysis. Follow-Up to Recommendations of Integration Workshop Three for the Assessment of PGA for the Koeberg NPP Sites and the NSIP Sites of Brazil– Schulpfontein, Thyspunt and Bantamsklip , Council for Geoscience rep. no. 2006-0229

Bejaichund, M., Kijko, A. and Hattingh, E., 2006b. Final report on the design spectra. Brief Summary and Final Seismic Hazard Assessment for the Koeberg NPP Sites and the NSIP Sites of Brazil-Schulpfontien, Thyspunt and Bantamsklip. Council for Geoscience Rep. No. 2006-0???

Brandt, D.; Andreoli, M.A.G. and McCarthy, T.S. (2005). The late Mesozoic palaeosoils and Cenozoic fluvial deposits at Vaalputs, Namaqualand, South Africa: Possible depositional mechanisms and their bearing on the evolution of the continental margin. South Africa Journal of Geology 108: 271-284.

Brandt, M.B.C., Bejaichund, M., Kgaswane, E.M., Hattingh, E. and Roblin, D. L. (2005). Seismic history of Southern Africa. Council for Geoscience, Seismological Series, 37, 32 p.

Dames and Moore. (1976). Geologic Report. Koeberg Power Station, Cape Province, South Africa. Report prepared for The Electricity Supply Commission. Job No.: 9629-014-45.

Day, R.W. (1986). Magnetometric mapping of the False Bay dolerites. Joint Geological Survey/University of Cape Town Marine Geoscience Unit Technical report 16: 217-227.

23 De Beer C.H. (2004). Investigation into evidence for neotectonic deformation within onland Neogene to Quaternary deposits between Alexander Bay and Port Elizabeth- Desk study report. CGS rep no. 2004-0226. ESKOM NSIP- SHA-014382#P1-187.

De Beer C.H. (2005) Investigation into evidence for neotectonic deformation within onland Neogene to Quaternary deposits between Alexander Bay and Port Elizabeth – South Coast Report. CGS 2005-0180. NSIP-SHA- 016311#P1-197.

De Beer, C.H. (2006a). Investigation into evidence for neotectonic deformation within onland Neogene to Quaternary deposits between Alexander Bay and Port Elizabeth – West Coast Report. CGS 2006-0287. NSIP-SHA-017388#P1-196.

De Beer C.H. (2006b). Investigation into evidence for neotectonic deformation within onland Neogene to Quaternary deposits between Alexander Bay and Port Elizabeth – Executive summary report. CGS 2006-0065. NSIP-SHA- 017800#P1-209.

Hartnady (2003). Earthquake risk for Cape Town. Umvoto News Collumn, http://www.umvoto.com/news_full.asp?ID=42 .

Kijko, A. and Graham G. (1998) . “Parametric-Historic” procedure for probabilistic seismic hazard analysis. Part I: Assessment of maximum regional magnitude mmax Pure Appl. Geophys 152,413-442

Kijko, A. and Graham G. (1999) . “Parametric-Historic” procedure for probabilistic seismic hazard analysis. Part II: Assessment of Seismic hazard at a specific site. Pure Appl. Geophys 154,1-22

Kijko, A (2004). Estimation of the maximum earthquake magnitude M max. Pageoph, 161, 1655-1681

Reid, D.L.; Erlank, A.J. and Rex, D.C. (1991). Age and correlation of the False Bay dolerite dyke swarm, south-western Cape, Cape Province. South African Journal of Geology 94: 155-158.

Roberts, D.L. (2006). Dating and preliminary correlation of raised marine and estuarine terraces on the western and southern coasts of South Africa. Council for Geoscience Report No. 2005-0183, NSIP-SHA-016344#P1-141 (Rev.1).

Schoch, A.E. (1976). Die geografiese posisie van die Colenso-verskuiwing in die Vredenburg-Saldanha gebied. SA Geograaf 4: 303-309.

Scherbaum, F., J. J. Bommer, H. Bungum, F. Cotton, and N. A. Abrahamson (2005) . Composite ground-motion models and logic trees: methodology, sensitivities, and uncertainties. Bulletin of the Seismological Society of America 95 (5), 1575-1593.

Scherbaum, F., J.J. Bommer, F. Cotton, H. Bungum & F. Sabetta (2006a) . Ground-motion prediction in PSHA: A post-PEGASOS perspective. Proceedings of the First European Conference on Earthquake Engineering , Geneva, Paper no. 1312.

Scherbaum, F., J. Schmedes & F. Cotton (2004a) . On the conversion of source-to-site distance measures for extended earthquake source models. Bulletin of the Seismological Society of America 94 (3), 1053-1069.

Scherbaum, F., F. Cotton & P. Smit (2004b) . On the use of response spectral reference data for the selection of ground-motion models for seismic hazard analysis: the case of rock motions. Bulletin of the Seismological Society of America 94 (6), 2164-2185.

Scherbaum, F., F. Cotton & H. Staedtke (2006b) . The estimation of minimum-misfit stochastic models from empirical ground-motion prediction equations. Bulletin of the Seismological Society of America 96 (2), 427-445.

Talwani, P. and Gassman, S. (2000) Magnitudes of Prehistoric Earthquakes from a Study of Paleoliquefaction Features. Report submitted to USGS, November 1, 2000, 7 p. Available at URL http://www.erp- web.er.usgs.gov/reports/annsum/vol42/ni/g0032.htm

Von Buchenröder, W.L. (1830) . An account of earthquakes which occurred at the Cape of Good Hope during the month of December 1809, etc. South African Quarterly Journal, Cape Town.

Youngs, R.R., W.J. Arabasz, R.E. Anderson, A.R. Ramelli, J.P. Ake, D.B. Slemmons, J.P. McCalpin, D.I. Doser, C.J. Fridich, F.H. Swan III, A.M. Rogers, J.C. Yount, L.W. Anderson, K.D. Smith, R.L. Bruhn, P.L.K. Knuepfer, R.B. Smith, C.M. dePolo, D.W. O’Leary, K.J. Coppersmith, S.K. Pezzopane, D.P. Schwartz, J.W. Whitney, S.S. Olig & G.R. Toro (2003). A methodology for probabilistic fault displacement hazard analysis (PFDHA). Earthquake Spectra 19 (1), 191-219.

GLOSSARY OF TERMS

Term Definition Aeolian Transported and deposited by wind. A rock formed by the solidification of aeolian sediment is known as an aeolianite Aeolianite Cemented dune Aleatory uncertainty Part of the standard deviation where the uncertainty is inherent to the unpredictable nature of future events, independent of eg earthquake magnitude and distance. Given a specific model, one cannot reduce the uncertainty by collection of additional information.

Basement Eroded ‘foundation’ of pre-existing rocks on which younger strata have been deposited; a well consolidated geological formation which can be considered as homogenous with respect to seismic wave transmission and response Brittle-ductile Transitional conditions between brittle and ductile or plastic flow Cenozoic Last 65 million years; an era of geologic time from the beginning of the Tertiary period (65 million years ago) to the present. Its name is from Greek and means "new life." Cretaceous The final period of the Mesozoic era, spanning the time between 145 and 65 million years ago. Deaggregation To separate the exceedance contributions to the hazard from earthquake sources into its base dimensions, magnitude and distance (Frankel et al., 1999) Deformation Deformation is a change in the original shape of a material. Deterministic hazard assessment An assessment that specifies single-valued parameters such as maximum earthquake magnitude or peak ground acceleration, without consideration of likelihood Dolerite A fine-grained mafic intrusive rock, usually

26 occurring as dykes or sills, mineralogically equivalent to a basalt. Dyke A discordant intrusive body that is substantially longer than it is wide. Dikes are often steeply inclined or nearly vertical. A dyke is a tabular (sheet-like) igneous intrusion that cuts the surrounding strata at an angle. Earthquake The release of stored elastic energy caused by sudden fracture and movement of rocks inside the Earth, which causes ground accelerations that can damage property and threaten life. Earthquake hazard Earthquake hazard is anything associated with an earthquake that may affect the normal activities of people. This includes surface faulting , ground shaking , landslides , liquefaction , tectonic deformation, tsunamis , and seiches . Epicentre The location on the surface of the Earth directly above the focus, or place where an earthquake originates within the earth’s crust. Epistemic uncertainty Fault A fault is a fracture or fracture zone, along which movement has taken place. Sudden movement along a fault produces earthquakes. Slow movement produces aseismic creep. A fault is a tectonic structure along which differential slippage of the adjacent earth materials has occurred parallel to the fracture plane. It is distinct from other types of ground disruptions such as landslides, fissures and craters. A fault may have gouge or breccia between its two walls and includes any associated monoclinal flexure or other similar geologic structural feature. Fault plane The fault plane is the planar (flat) surface along which slip occurs Fault scarp The fault scarp is the feature on the surface of the earth that looks like a step caused by slip on the fault; it is a topographically visible feature. Fault trace The fault trace is the intersection of a fault with the ground surface; also, the line commonly plotted on geologic maps to represent a fault. Gneiss A rock formed by regional metamorphism in which bands or lenticles of granular minerals alternate with bands or lenticles characterised by minerals having flaky or elongate prismatic shapes Ground motion Ground motion is the pervasive movement of the earth's surface from earthquakes or explosions. Ground motion is produced by waves that are generated by sudden slip on a fault or sudden pressure at the explosive source and travel through the earth and along its surface. Intensity (of an earthquake) A measure of the severity of shaking at a

27 particular site. It is usually estimated from descriptions of damage to buildings and terrain. The intensity is often greatest near the earthquake epicenter. Today, the Modified Mercalli Scale is commonly used to rank the intensity from I to XII according to the kind and amount of damage produced.

Intensity is an indicator of the physical effects of an earthquake on humans, or structures built by humans, and on the Earth’s surface. The indicator comprises a set of numerical indices that is based on subjective judgments, not instrumental records. Liquefaction A process, in which, during ground shaking, some sandy, water-saturated soils can behave like liquids rather than solids. Liquefaction is caused by a sudden loss of shear strength and rigidity of saturated, cohesionless soils, due to vibratory ground motion. Ma Abbreviation for million years Magnitude A quantity characteristic of the total energy released by an earthquake, as contrasted with intensity, which describes its effects at a particular place. A number of earthquake magnitude scales exist, including local (or Richter) magnitude (ML), body wave magnitude (Mb), surface wave magnitude (Ms), moment magnitude (Mw), and coda magnitude (Mc). As a general rule, an increase of one magnitude unit corresponds to ten times greater ground motion, an increase of two magnitude units corresponds to 100 times greater ground motion, and so on in a logarithmic series. Maximum credible earthquake The maximum earthquake that is capable of occurring in a given area or along a given fault during the current tectonic regime. “Credibility” is in the eyes of the user of the term. Mesozoic Period from 65 -150million years ago

Neotectonics The study of post-Miocene structures and structural geology of the Earth’s crust Normal fault A fault in which the hanging wall (the block above the fault plane) appears to have moved downward relative to the footwall. The fault angle is usually 45-90° Orogeny ‘Mountain building’. A period of tectonic activity giving rise to large-scale folding and faulting typical of mountain belts. Palaeoseismology The study of prehistoric earthquakes, especially their location, timing and size. Palaeoseismic evidence Refers to earthquakes recorded geologically, most of them unknown from human descriptions or seismograms. Geologic records of past earthquakes can include faulted layers of sediment and rock, injections of liquefied sand, landslides,

28 abruptly raised or lowered shorelines, and tsunami deposits. Permo -Triassic Period around 250 million years ago

PGA Peak Ground Acceleration

29 relatively uniform Stable Continental Region (SCR) A SCR is composed of continental crust, including continental shelves, slopes and attenuated continental crust, and excludes active plate boundaries and zones of currently active tectonics directly influenced by plate margin processes. It exhibits no significant deformation associated with the major Mesozoic-to-Cenozoic (last 240 million years) orogenic belts. It excludes major zones of Neogene (last 25 million years) rifting, volcanism, or suturing. Strike The direction traced by a planar feature, such as a bed or dyke, as it intersects a horizontal surface, measured relative to geographic north. Surface faulting Differential ground displacement at or near the surface caused directly by fault movement and is distinct from non-tectonic types of ground disruptions, such as landslides, fissures and craters; The permanent offsetting or tearing of the ground surface by differential movement across a fault during an earthquake. Tertiary Period from 65 -1.6 million years ago; The first period of the Cenozoic era (after the Mesozoic era and before the Quaternary period), spanning the time between 65 and 1.8 million years ago. Unconformably Said of a younger rock formation overlying an older rock formation with an erosional surface separating the two; this implies a significant time break between the two formations

30 GLOSSARY OF TERMS

Term Definition Aeolian Transported and deposited by wind. A rock formed by the solidification of aeolian sediment is known as an aeolianite Aeolianite Cemented dune Basement Eroded ‘foundation’ of pre-existing rocks on which younger strata have been deposited; a well consolidated geological formation which can be considered as homogenous with respect to seismic wave transmission and response Brittle-ductile Transitional conditions between brittle and ductile or plastic flow Cenozoic Last 65 million years; an era of geologic time from the beginning of the Tertiary period (65 million years ago) to the present. Its name is from Greek and means "new life." Cretaceous The final period of the Mesozoic era, spanning the time between 145 and 65 million years ago. Deaggregation To separate the exceedance contributions to the hazard from earthquake sources into its base dimensions, magnitude and distance (Frankel et al., 1999) Deformation Deformation is a change in the original shape of a material. Deterministic hazard assessment An assessment that specifies single-valued parameters such as maximum earthquake magnitude or peak ground acceleration, without consideration of likelihood Dolerite A fine-grained mafic intrusive rock, usually occurring as dykes or sills, mineralogically equivalent to a basalt. Dyke A discordant intrusive body that is substantially longer than it is wide. Dikes are often steeply inclined or nearly vertical. A dyke is a tabular (sheet-like) igneous intrusion that cuts the surrounding strata at an angle. Earthquake The release of stored elastic energy caused by sudden fracture and movement of rocks inside the Earth, which causes ground accelerations that can damage property and threaten life. Earthquake hazard Earthquake hazard is anything associated with an earthquake that may affect the normal activities of people. This includes surface faulting , ground shaking , landslides , liquefaction , tectonic deformation, tsunamis , and seiches . Epicentre The location on the surface of the Earth directly above the focus, or place where an earthquake originates within the earth’s crust. Fault A fault is a fracture or fracture zone, along which movement has taken place. Sudden movement along a fault produces earthquakes. Slow movement produces aseismic creep. A fault is a tectonic structure along which differential slippage of the adjacent earth materials has occurred parallel to the fracture plane. It is distinct from other types of ground disruptions such as landslides, fissures and craters. A fault may have gouge or breccia between its two walls and includes any associated monoclinal flexure or other similar geologic structural feature. Fault plane The fault plane is the planar (flat) surface along which slip occurs Fault scarp The fault scarp is the feature on the surface of the earth that looks like a step caused by slip on the fault; it is a topographically visible feature. Fault trace The fault trace is the intersection of a fault with the ground surface; also, the line commonly plotted on geologic maps to represent a fault. Gneiss A rock formed by regional metamorphism in which bands or lenticles of granular minerals alternate with bands or lenticles characterised by minerals having flaky or elongate prismatic shapes Ground motion Ground motion is the pervasive movement of the earth's surface from earthquakes or explosions. Ground motion is produced by waves that are generated by sudden slip on a fault or sudden pressure at the explosive source and travel through the earth and along its surface. Intensity (of an earthquake) A measure of the severity of shaking at a particular site. It is usually estimated from descriptions of damage to buildings and terrain. The intensity is often greatest near the earthquake epicenter. Today, the Modified Mercalli Scale is commonly used to rank the intensity from I to XII according to the kind and amount of damage produced.

PGA Peak Ground Acceleration

PNI&I Palaeoseismic-neotectonic investigations and integration with probabilistic hazard analysis Probabilistic seismic assessment An assessment which stipulates quantitative probabilities of the occurrences of specified hazards, usually within a specified time period Probability The ratio of the chances favouring a certain event to all the chances for and against it Recurrence interval The recurrence interval, or return period, is the average time span between large earthquakes at a particular site. Rift A long, narrow crack in the entire thickness of the Earth's crust, which is bounded by normal faults on either side and forms as the crust is pulled apart Scarp A long, more or less continuous cliff-face or steep slope/ridge. A fault scarp is formed by sudden earth movements (usually vertical) along fault lines. Scarps may also be formed by horizontal movement where a hill or ridge have been broken open, exposing a steep interior face along the line of rupture. Secondary effects Nontectonic surface or near-surface processes that are directly related to earthquake shaking or tsunamis Seismic hazard The physical effects such as ground shaking, faulting, landsliding, and liquefaction that underlie the earthquake’s potential danger Seismic source A general term referring to both seismogenic sources and capable tectonic sources Seismogenic source A seismogenic source is a portion of the earth that we assume has uniform earthquake potential (same expected maximum earthquake and recurrence frequency), distinct from the seismicity of the surrounding regions. A seismogenic source will generate vibratory ground motion but is assumed not to cause surface displacement. Seismogenic sources cover a wide range of possibilities from a well-defined tectonic structure to simply a large region of diffuse seismicity (seismotectonic province) thought to be characterized by its involvement in the current tectonic regime (the Quaternary, or approximately the last 2 million years). Seismogenic structures Structures that display earthquake activity or that manifest historical surface rupture or effects of palaeoseismicity. Seismogenic structures are those considered likely to generate macro-earthquakes in a period of concern. Seismotectonic province A region within which the active geologic and seismic processes are considered to be relatively uniform Stable Continental Region (SCR) A SCR is composed of continental crust, including continental shelves, slopes and attenuated continental crust, and excludes active plate boundaries and zones of currently active tectonics directly influenced by plate margin processes. It exhibits no significant deformation associated with the major Mesozoic-to-Cenozoic (last 240 million years) orogenic belts. It excludes major zones of Neogene (last 25 million years) rifting, volcanism, or suturing. Strike The direction traced by a planar feature, such as a bed or dyke, as it intersects a horizontal surface, measured relative to geographic north. Surface faulting Differential ground displacement at or near the surface caused directly by fault movement and is distinct from non-tectonic types of ground disruptions, such as landslides, fissures and craters; The permanent offsetting or tearing of the ground surface by differential movement across a fault during an earthquake. Tertiary Period from 65 -1.6 million years ago; The first period of the Cenozoic era (after the Mesozoic era and before the Quaternary period), spanning the time between 65 and 1.8 million years ago. Unconformably Said of a younger rock formation overlying an older rock formation with an erosional surface separating the two; this implies a significant time break between the two formations

APPENDIX A

Geological investigations

Table 1: Geological investigations at Duynefontein (Koeberg) Task Completed Quality Comments Regional and semi-regional geology. Yes A Needs reviewing and updating Site vicinity and site specific geology. Yes A Needs review, add palaeo tsunami study Data collection - existing geology coverages (digital), topographic and Yes for A Hydroclimate, land- topocadastral information (digital), air photos (colour digital, if available), available data use and vegetation satellite imagery, hydroclimatic coverages, land-use and vegetation-type not part of CGS coverages.

Geographic Information System (GIS) compilation of coverages and base plans Yes for A Needs reviewing containing above information - required for site reconnaissance. available data and updating

Remote Sensing interpretation to identify land facets, site aspects, quarries and Yes A Needs reviewing cuttings, and other relevant surface features to familiarise oneself with the and updating expected ground conditions.

Site reconnaissance: field inspection and documentation of relevant surface Yes A Needs reviewing features, exposures (road cuttings, outcrops areas, accessibility, potential and updating problem areas etc) as identified in RS & GIS-based desk -top surveys.

GIS-based mapping of soil and rock-type distributions around the (selected) Yes A Needs reviewing sites. and updating

Field structural mapping of outcrop-scale bed-rock fracturing. Yes Needs reviewing and updating GIS-compilation and interpretation of geological and structural data. Yes Needs reviewing and updating GIS-compilation and interpretation of geophysical data. Yes Needs reviewing and updating Identification of selected sites for pit sampling and trench-profiling No

Logging of pits and trenches No

GIS compilation and map integration of pit and trench data. No

APPENDIX B Scope of work to determine mitigating measures to minimize the geological impact on the project.

Scope of work that needs to be done at Koeberg to determine mitigating measures to minimize the geological impact on the project.

a) Review and update all the available geological data and maps in possession of Eskom and the existing 1:50 000 geological maps of the CGS for this area. Perform 1: 2000 scale geological and structural mapping within the 8km radius and 1:1000 mapping within the 1km radius around the site.. Focus on the Quaternary geology and tectonics. b) Resolve the existing problems surrounding the so called “ Milnerton fault” and reduce the uncertainty as to the source of the 1809-1810 seismicity . c) Identification of selected sites for pit sampling or vibro-coring and trench- profiling to try determine the recurrence intervals of large seismic events in this area. d) Logging of named pits and trenches. e) GIS compilation and map integration of pit and trench data. f) Specialist structural geological interpretation of the marine geophysical data.

APPENDIX C

Seismological investigations

Table 2. Seismological investigations for Koeberg

Task Completed Quality Comments Updating and standardization of regional earthquake catalogues Yes according to modem international conventions of format and earthquake- size (magnitude / seismic moment) homogeneity. Re-assessment of earthquake frequency-size distributions and statistics. Yes

Re-determination of "maximum regional magnitude/moment" in tapered No Not CGS Pareto formulation.

Selection of appropriate "scenario earthquake" parameters (epicentre Yes latitude and longitude hypocentre depth, moment magnitude, seismic- attenuation formulae) from historical and instrumental catalogue records (e.g. the 1809 Milnerton earthquake scenario for the Koeberg location).

Quantitative loss-estimation modelling for selected scenario earthquakes Not CGS function (using RADIUS, QuakeLoss, ShakeMap, and/or other suitable software)

Palaeoseismic investigations (selected trench-profiling) for identification of No possible great prehistoric earthquakes and determination of site susceptibility to liquefaction during severe ground shaking.

Review of far-field, trans-Indian Ocean impact predictions from numerical Partially Formalise, write-up modeling of the next (imminent?) Great Sumatran earthquake tsunami (1797/1833 Mentawai source zone), with special reference to the Thyspunt, Bantamsklip and Koeberg sites.

Review of far-field-impact predictions from numerical modeling of a trans- Partially Formalise, write-up Atlantic ocean (South Sandwich Trench earthquake/tsunami source, with special reference to the Koeberg and Northern Cape sites.

Review of coastal-zone geomorphology, sedimentology, onland and No offshore (continental slope) Quaternary stratigraphy, for palaeo-tsunami indicators from either Atlantic or Indian trans-oceanic sources and for local tsunami sources due to (earthquake-triggered?) submarine slump/slide activity on the continental margin of the Cape Basin (as actually recorded along the Table Bav shoreline immediately following the M6+ event historical event on 4 December 1809).

APPENDIX D

Scope of work that needs to be done at sites to determine mitigating measures to minimize the seismological impact on the project.

Scope of work that needs to be done at Duynefontein to determine mitigating measures to minimize the seismological impact on the project.

a) Update the seismological catalogue for site. b) Add additional palaeo-seismic events to catalogue. c) Determine new PSHA. d) Determine PSHA by Logic Tree technique. e) Describe the uncertainties surrounding (a) to (d) above.