Foreword by Cathay Seas, september 2007 1

Right after the tsunami of 26 december 2004, we found ourselves stuck in front of the TV screens, staring at the images of an unbelievable devastation in the Indian Ocean. Afterwards, the loss of lives has been estimated at 230.000, 170.000 of them Indonesian. In september 2007, Nature magazine published a study showing that there is a risk for another megaquake in central Sumatra : the only question is When ?. Ouch. ...But the real discovery of the study by Phil Cummins is that a new mega-quake hotspot is now identified in the Bay of Bengal, off the shorelines of Myanmar, Bangladesh and India. Should a quake there trigger a tsunami, it would directly threaten 60 millions inhabitants of the lowlying lands of these three countries and, given the long-distance nature of a tsunami wave, one million casualties is the initial estimation, covering the whole Indian Ocean, again. There are two risks in one here : - The mega-quake itself has enough power to flatten large cities, destroy roads, bridges and change the geography and the landscape : up-raising of new reefs at sea, sinking entire beaches underwater, 2 all things witnessed in the region of Nias island after the quake of march 2005. - Then the tsunami comes in and floods a population either weakened by the quake a few minutes before, or completely unaware because the wave(s) break many hundred kilometers away from their birthplace. This knowledge of the mega-quake hotspots in the Indian Ocean is of great help in terms of prevention (before) and disaster mitigation (after). The political bodies, national and international, now have enough data to start some serious work of public education and emergency planning.

But it would be a great mistake to stay focused on one single area, because the risk is worldwide : there are several mega-quake hotspots on Earth, all of which whith the potential to create a tsunami. The study proposed here is a work from the Australian Consortium for the Asian Spatial Information and Analysis Network (ACASIAN), hosted by the Griffith University in Brisbane. The author, Jason Wotherspoon, was published in 1998. This work shows that there is a mega-quake hotspot right in the middle of South China Sea, with the potential to lift the sea floor and send towards China a tsunami estimated here at 20 meters high (35 meters confirmed in the indonesian province of Aceh in 2004). Jump to page 24 to have a direct view of the Worst Case Scenario. Interestingly, the fictious earthquake described is 8.5 in magnitude, compared to the 8.7 that happened in march 2005. It is “a study of possible risks of an ocean originating inundation, such as a tsunami, upon low-lying, densely populated coastal urban and rural centres, by applying a tsunami model constructed from data on the July 1998 Papua New Guinea tsunami event and other historical examples, to the Pearl River Delta, China as an example.”

As it is said, the choosen area of this study is an example. Originating from the central South China Sea, the spectrum for a tsunami would be from Vietnam to the West, to approximately Xiamen to the East. This includes surf destinations like Hainan island and mass-tourism places like the Halong bay in the Gulf of Tonking. In terms of destruction and casualties, it would be no less devastating than what we have already experienced in the Indian Ocean.

1

The mega-quake of december 2004 was 9,3 in magnitude, one of the fifth strongest ever recorded. The quake of march 2005 was 8,7 in magnitude, six times less destructive than in 2004. People had to lay on the floor not to fall down during the tremor. On Nias and Simeulue islands, 900 people were killed. 2

Page 1 on 27

The aim of this Foreword is not to force you to cancel your next holidays... such hotspots exist in many places and not only in ‘exotic’ countries : from the Canary islands, the Cumbre Vieja mountain threatens the whole east coast of the United States. The aim is to raise your awareness, and to help spread scientific, reliable information to policymakers and local communities. With this study in mind, we hope to succeed in this task with the communities bordering the South China Sea. Don’t have fear, be prepared.

- Yannick, editor of Cathay Seas.

Overview of the concerned area : Nanshui – Hong Kong – Guangzhou.

Page 2 on 27

Source : http://www.asian.gu.edu.au © Jason Wotherspoon, 1998.

Investigation of the potential effects of a tsunami on the Pearl River Delta.

1)

Abstract :

As population growth and it's attendant problems increase exponentially, numbers in coastal areas increase in a rate above that of national averages, leading to dense and highly populated areas lying in very close proximity to ocean and estuarine bodies of water. Tsunamis occur too infrequently enough to be in the back of people's minds when they decide to relocate to coastal areas to live, yet they happen often enough, and with such devastation, that local and governmental authorities are concerned enough to begin various programmes of measurement, prediction, warning, mitigation, and emergency response. Given the shocking extent of devastation on the coastal Papua New Guinea communities by tsunami in July, 1998, and the media reaction following, it is clear that much more effort needs to be paid to these programmes by the relevant authorites. By applying data and knowledge based on experience from the PNG tsunami, and other important historical tsunami events from around the world, to the coastal populations of the Pearl River Delta, in Guangdong province, China, I hope to highlight the importance becoming more aware of this potential problem.

Page 3 on 27

2)

Introduction 2.1 The Need For A Deeper Understanding.

The July 17, 1998 tsunami of Aitape, Papua New Guinea brought to the public awareness with shocking reality not only the power of nature, but also our lack of understanding and preparedness against it. Three earthquake-triggered tsunami's, the largest only 10m in height, struck a 30 km stretch of coastline that housed 13,000 people over 7 villages, and completely wiped them off the face of the earth. While only 2,000 bodies have been found, as many as 8,000 are reported missing. Sissano Lagoon, the heart of the devastated area and where most of the casualties lie, was reported as being so contaminated by rotting corpses that all efforts to recover the bodies were officially abandoned, and a 120 sq. kilometre region was declared a "no-go zone". (Phillips, 1998) Coastal populations are growing rapidly at rates higher than the national average due to migration from inland. According to the UNEP, most people in the (Asia -Pacific) region live along the coasts, with one quarter of the world's 75 largest cities being near or on the region's coastlines (www.unep.org 1997) In a report for the American Association for the Advancement of Science, Hinrichsen observed that "of China's 1.2 billion people, over 677 million (56 percent) reside in 13 southeast and coastal provinces and two coastal municipalities -- Shanghai and Tianjin. Along much of China's 18,000 kilometers of continental coastline, population densities average over 600 per square kilometer" (www.aaas.org 1995) Earthquakes will continue to occur as a natural phenomenom, and as some suggest, they are on the increase. Tsunamis will continue to be generated by oceanic earthquakes and affect the coastal populations. Since coastal populations are growing rapidly, and the fact that there is the possibility of earthqukes of greater magnitude, I believe the threat of a huge loss of human life and damage to property is growing at a rate that far exceeds the growing capacity for disaster warning, emergency response, and clean-up procedures. This, I also believe, will lead to a repeat of the situation found at Aitape, in Papua New Guinea, but with much greater devastating effect, especially if it were to occur on a densely populated coastal region such as the Pearl River Delta, in Guangdong province, China.

2.2 Identify and discuss important aspects of the inundation scenario which need to be examined in the project The most important issue regarding the study of tsunami's is of course the destructive effects upon the human environment. Understanding tsunami's will help save lives. To assess the threat level and vulnerability of any given area, we need to take into consideration a number of factors. Proximity to earthquakes zones is the primary factor, although all countries around the Pacific Rim are likely to experience tsunamis from quakes occurring anywhere in the Pacific. Earthquake magnitude, sea-floor depth (bathymetry) coastline topography and populations are all factors that contribute to an event and determine whether or not it will have disastrous effects. Not to be ignored however, is the emergency response and clean-up procedures that follow such an event, for these factors can also contribute heavily to loss of life. If for any reason, the emergency response crews are delayed or hindered in their work, or clean-up procedures are mishandled, these factors can add to the loss of human life well after the tsunami event itself. The 1998 PNG event is a prime example of a lack of disaster warning, excessively slow emergency response times, and poorly handled clean-up precedures, which, if prperly executed, could have resulted in minimising the death toll of the event.

2.3 Methodology The nature of the problem discussed in this paper is constructed of two very different sets of information which deal with two very different types of environments, ie. that which occurs in the water, and that which occurs on land. Both of these can be observed separately, and the problem itself does not exist until the two are brought together. For reasons discussed below, the Pearl River delta and

Page 4 on 27

environs is not at as great a risk of tsunami damage as Hawaii or Japan, yet the physical conditions present at Guangzhou are representative of the typical tsunami vulnerable environment at many places around the world. These two sets of information will be discussed separately, under the headings of Water and Land. Next I will examine existing models and work done by several researchers at the forefront of the field, to use as a basis for the hypothetical model which will be later applied as a hypothetical scenario with which to illustrate the disastrous effects of inundation upon coastal populations. As a visual aid to examining the problem, a special interactive map presentation (see below) has been devised to compare various types of maps with specific sets of data. Next I will set the parameters for a hypothetical scenario, and review the inundation process step by step as it would possibly apply to the Zhujiang estuary. Finally, I will discuss interesting or valuable issues that arise from examining such an event,

3)

Water

3.1 Forms Of Inundation And The Threats They Pose Inundation of coastlines from the ocean comes in a number of forms, some more immediate and threatening than others, and I will briefly examine them here.

3.1 (a) Flood By far the most prevalent of historical accounts referring to inundation of some form is the legend of the Deluge that is mentioned in the Bible. Found, in many various yet strikingly similar forms, in such diverse countries and cultures as the Middle East, Asia, Africa, the Pacific Islands, North and South America, there are more than 500 deluge legends known around the world. While some can be shown to have stemmed from the Bible in its present form, the specialist researcher Dr. Richard Andee concluded that a major portion of the Deluge legends were entirely independant of the Mesopotamian and Hebrew accounts. (Filby, 1970) The threat posed by such an event as described in the Bible and other documents, should it possibly occur in the future, would be of such proportions that it would render the topic of this paper useless - it is beyond the scope of this project. However, there is a growing body of present day research into the matter, as can be seen in the publications of Zangger (1992), Wood & Campbell (1994), Flem-Ath & Flem-Ath (1995), Allan & Delair (1995)

3.1 (b) Rising Ocean Levels. It has now been established beyond doubt that the level of the oceans is rising. Globally, the sea-level has risen 10 - 25 cm over the last century, and is projected to rise by 15 to 95 centimetres by 2100. (Schroeder and Bassett, 1998) This is due to the release of water previously trapped in frozen form in the polar ice caps into the Earth's hydrological cycle, an effect of Global Warming due to an increase in the "greenhouse effect", where chemicals and particles in the Earth's atmosphere insulate the planet from the loss of heat received from the sun. Mean surface temperatures have increased by 0.3 to 0.6 degrees C over the last century, and are projected to increase by about 1 to 3 degrees C by 2100. (Schroeder and Bassett, 1998) There is evidence of rapid 20m changes in sea levels about 400,000 years ago, found in connection with sudden appearances of coral reefs due to a deposition of calcium carbonate in the water. (New Scientist, 31 May 1997).

3.1 (c) Land Subsidence Closely connected to Rising Ocean Levels is Coastal Land Subsidence, which combined together form what is known as passive submergence. Subsidence is caused by two factors, gradual tectonic movement, and natural wave action, the latter found to have less effect than the former. The western seaboard of the US has been particularly noteworthy of land and property losses due to subsidence, caused by active plate movements. This is a problem that has been recognised within State

Page 5 on 27

governments concerned - the Massachusetts Office of Coastal Zone Management, for example, has estimated that 65 acres of upland per year are lost to the sea. (www.whoi.edu 1994) It is generally a slow process, which varies from region to region, but is measured in amounts of less than a centimetre per year. Seismic plate movements cause land subsidence on a greater scale, but such events generate wave, or tsunami effects that are experienced synchronously but with more effect than subsidence itself. Although the subject of rising sea levels and land subsidence has important consequences for world climate, hydrological systems and loss of land in the future, it is beyond the scope of this paper. Residents and authorities will have sufficient warning when threats arise to relocate or take preventative measures such as diking or land reclamation

3.1 (d) Storm Surges Giant waves travelling up the mouths of rivers and estuaries as a result of hurricane influence, has only been recently recognised. A number of port regions around the world experience "king tides", which are simply very large changes in tide heights, and some river systems, particularly in South America, experience unusual wave effects due to a mixture of tidal changes with unique bathymetric and geographical structures. Storm surges are destructive waves caused by hurricanes, and can be as high as 5-6 metres. They are a large dome of water that sweeps into the coast as much as 5 hours before a hurricane makes its landfall, and cause most of all hurricane related deaths. The dome is a large body of agitated water that is driven at the surface by the extremely high winds of the hurricane. It also appears among rough seas of increasing intensity as a single massive wave followed by a decrease in intensity, and tends to match normal onshore speeds of breaking waves. When coinciding with high tide, their waves can be extremely destructive: over 6000 people were killed in the Galveston, Texas Hurricane of 1900, most by storm surge, and Hurricane Hugo in 1989 generated a 20-foot storm tide in South Carolina. (www.fema.gov 1997) As many as 300,000 perished when a storm surge caused by a hurricane in the Bay of Bengal swept across the Ganges Delta on 12 November, 1970 (Times Publication, 1989) Studies conducted by Purdue University (wxp.eas.purdue.edu) have found that estimates of storm surge wave heights can be based on hurricane strength.(see Fig.1) Fig. 1 : 1997 Hurricane/Tropical Data for Atlantic

Type

Category

Pressure (mb)

Winds (kts)

Surge (ft)

Depression

TD

---

980

64-82

4-5

Hurricane

2

965-980

83-95

6-8

Hurricane

3

945-965

96-112

9-12

Hurricane

4

920-945

113-134

13-18

Hurricane

5

134

>18

For storm surge waves to represent any threat, they need to be generated by very strong hurricanes. Category 4 or 5 hurricanes, which take days to grow, form waves that only begin to be comparable with wave heights of tsunami's, and thus there is plenty of early warning indicators to alert authorities of impending danger. The mechanics of storm surges are very similar to tsunami's, albeit on a smaller scale, and a model to describe the action and effects of either would be very similar.

Page 6 on 27

3.1 (e) Catastrophic Wave events Impact Waves Impact waves caused by meteor strike, and those caused by ice-sheets breaking off from the polar ice-caps, are yet to be seen in recent history, but there is considerable we only have geological and archeological evidence for their occurence in the past. Meteor strike, similar to that which is thought to have caused the demise of the dinosaurs, has also been thought to have formed the Gulf of Mexico, while some researchers, A.T.Wilson of Victoria University, New Zealand for instance, believe that the Biblical Deluge could have been caused by a slippage of an ice sheet into the sea (Sitchin, 1976 p. 402-406) Such events, despite their potential destructive power, happen too rarely for serious study to be undertaken, although separation of glaciers from the polar ice caps is an event of growing concern due to the rising temperatures of the sea, and indeed there have been reports of a 300km ice flow finally breaking off from the North Pole ice cap in the last year. Added to that the fact that there is, at present, too little wave data available for a thorough study of this phenomenom. Seismic Waves The tsunami, after theDeluge, has the most prominent recognition of inundation events within a cultural tradition. Despite it being localised to only a physically small but densely populated island nation, Japan, their occurence and devastating effect on the culture has been so great and prolonged that they have become a part of Japanese heritage. The Japanese were the first to recognise earthquakes as being the cause of tsunamis, having historical records exist of tsunamis as early as 2000 years before present (Nakata & Kawana, 1995). The word tsunami has been created to describe the phenonmenom, they have been depicted in folk art (The cover graphic to this paper shows a 19th Century print by Hokusai), and have even been instrumental, along with earthquakes, in determining the structural design and construction of traditional Japanese buildings. Japan, being on the join of the Pacific and Asian tectonic plates has probably experienced more tsunamis than any other coastline in the world. They have become part of the Japanese psyche. Other cultures have not experienced them enough to have them become part of their historical tradition - for example, the very same community at Aitape was struck by devastating tsunami at the beginning of this century, yet noone in the community remembered, though Alaska and the Aleutians have also experienced many tsunami's as evidence shows. Statistically, earthquakes occur on a regular basis. For smaller quakes around the world it is on a daily basis, for larger quakes in certain areas of the world, they occur less occasionally, on average every 34 years. For the South China Sea region, 28 earthquakes of magnitude 7.0 or greater, which is the average minimum for tsunami potential, have occurred in the 94 years from 1900, as measured by the NOAA.(see Fig. 2) Fig. 2 : Graph by J. Wotherspoon based on NOAA data for period 1900 - 1994, magnitude greater than 7.0, in area 0.0 - 20.50 deg. N by 110 - 125 deg. E, 1998.

Page 7 on 27

Although it is difficult at present to predict with accuracy the occurence of future earthquakes, some pattern is discernible from the graph. The most obvious trend is a decrease in magnitudes over time. This is not necessarily a good indication, as after closer inspection, one notices that there appears to be an event of great magnitude followed soon after by events of lesser magnitude, eg. 1905 - 1955, followed then by a much larger quake, or if after a period of inactivity, a cluster of events, eg 1970 and 1990. After a long slow downward trend, it could be argued that a very large event of manitude 8.0 or greater is due to occur within the next 5 years, or a cluster of events after a period of 10 years. Periods of inactivity tend to build up pressure, which leads to larger and more forceful releases of energy when events do occur. The following map is a very good visual indicator of not only the high frequency of earthquake occurences, but it also shows the plate boundaries by marking the most active seismic zones.

Source: ERMOS, created online in response to query, 17th September, 1998

Page 8 on 27

The accompanying map shows the seismicity for the South East Asia region and was generated automatically in response to a web query to the IRIS online database on 17th September, 1998, showing all magnitudes for the period of 5 years.. Of interest is the magnitude 5.5 earthquake in the South China Sea directly north of East Malaysian Kalimantan, which occurred in September 1998. It's epicentre is roughly 1000kms from The Zhujiang Delta.

Waves caused by undersea avalanches and land slippages may or may not be asscociated with earthquakes. With tsunami's there is a seismic event which releases a shockwave into the water that transfers into a tsunami. An undersea avalanche or land slippage may not necessarily be associated with a seismic event, they could be a resettling of the sea bottom, known as "slope failure"(Dawson et al, 1993) as a result of centuries of slow build up of ocean deposits. The shockwave in these cases is caused by sudden changes in the sea floor, which causes rapid movement of a large body of water, and a tsunami like wave will result. Of course, such an event will leave a seismic signature on testing stations, but it will appear differently to that of earthquakes.

Page 9 on 27

3.2 Decide which type of inundation are important for the project model. By a process of elimination of the more impractical and inappropriate types of inundation, and due to the wealth of physical data available today, and due to the prevalence of the particular type of wave occurence, I will choose the tsunami event as the inundation type for this project.

3.3 Examine the mechanics of the chosen inundation type. Earthquakes of magnitude greater than 7.0 at depths of 50-100km are considered to cause tsunami risk, however "tsunamigenic" earthquakes of much less magnitudes have been known to cause destructive tsunamis. The propagation of energy in a tsunami radiates through what is described as linear wave dispersion theory. (Gonzalez & Kulikov, 1993). Tsunamis are amplified through resonation caused by decreases in bottom depth, narrowing of bays and inlets, and by coinciding with high tide periods present in the inlets. (Henry & Murty, 1993) Prior to a tsunami's landfall, sea levels decrease dramatically like a rapidly receeding tide, leaving the sub-littoral zones exposed. This is followed immediately by the first of usually 3 tsunamis. In linear wave theory, directional changes occur through simple geometric reflection, therefore transferral of wave energy is reflected by sea bottom relief and bolstering from sides of inlets etc. When the sea floor depth shallows to 0 m, a wave is produced as the end product of the transferral of shockwave energy, which is finally dissipated on land. The effect is the same as normal beach break waves, only greatly magnified. The tsunami continues inland until the forward transferral of energy is dissipated, then the sea water will succumb to the pull of gravity, and seek the lowest elevations. The resistance of buildings and structures to the force of a tsunami depends upon load ratings of the construction materials in simple equations of fluid dynamics.

Page 10 on 27

4)

Land

4.1 Topography Types Susceptible to Inundation The shape of the land under sea determines the strength and direction of the tsunami toward the coastline, while above ground it determines the run-up. Since the generation of a wave is defined by transferance of energy, or force in simple mathematical terms, the decrease in sea floor depth has the proportional effect of focusing and speeding up the movement of the wave. Thus a wave generated from a large, deep body of water will produce larger waves than those produced from a shallower body of water.

4.1 (a) Beach coastlines. As shown by Adams & Lewis, (1979) offshore coastline has a modifying effect on the wave shape as it breaks on land. The most destructive waves are where the force of the wave is focussed as it breaks on land. Beaches, with their concave sea floor relief are thus more susceptible are than promontaries to damage from a tsunami of equal force, because the convex relief of the sea floor deflects the wave off centre and out to the side. Beach breaking tsunamis have potentially more threat for two reasons. Land behind beaches tends to be lower than land behind promontaries, therefore the penetration by a wave would be greater. Secondly, populations tend to congregate around beaches for reasons of leisure, and practicality of access for fishing vessels.

4.1 (b) Estuarine shorelines The tidal period will also affect the height and run-up of a tsunami. Low tides, where the water's edge recedes off shore due to lowering of the sea level have a mitigating effect of reducing the threat of tsunami. Combined with a high tide period where the sea level is higher and closer in shore, tsunamis pose more of threat. A tsunami hitting an estuarine environment will also have devastating effects for several important reasons. Since estuarine environments are not subject to normal wave action from the sea, only indirect tidal influences, the coastline topography can be quite different, particularly with large delta regions such as Zhujiang. Most of the land mass around the shorelines of a delta region is flat, and barely above the water line, and is somewhat static. In this type of environment, populations can and do congregate very close to shorelines, due to the lesser apparent wave threat (bearing in mind that people are thinking of ocean waves, not tsunamis), and also because of the easy access to plentiful food supplies. Most of these coastal communities have a lifestyle that revolves around water, fishing, boating, washing and irrigation are all activities associated with estuarine lifestyle. This familiarity with close proximity to water would lead people to have less fear of water, and hence possibly, less caution. Coastal estuarine communities of developing countries also generally build with traditional construction techniques and materials. Simple wooden structures designed for ease of construction, from readily available materials, such as wood, twine and leaves are the norm. Communities such as the Papuans, the Javanese, still Phillipino still follow traditional building methods to this day. These types of structures have absolutely no rsistance when it comes to meeting the force of a tsunami. The Aitape tsunami of July, 1998 is a prime example of this type of situation. The coastal village communities built virtually on the beach, and around the lagoon at Sissano, and the 10m tsunami simply wiped the shorelines clear of all structures.

4.1 (c) Rivers Secondary to tsunami effects on coastlines, is the effect of the wave as it travels up rivers. Henry & Murty (1995) described the process of "resonance amplification", where the wave grows as it travels higher upstrem. Another effect is wave shear, where the river banks and their features are "scraped"

Page 11 on 27

by trapped "edge waves" as it travels upstream. This effect is almost identical in mechanics to tide surges and storm surges, and has been mentioned by Carrier in a paper on transverse waves travelling parallel to coastlines (Carrier, 1995)

4.1 (d) Sloped Shorelines Another feature of topography that adds to tsunami threat is the presence of hills or slopes behind built structures lying on coastlines. Slopes cause a very destructive tsunami phenomenon known as run-up, where the wave will wash in over land, and the inertia will carry the water inland and push it up the slopes of a hill or mountain range, where it will then wash down, back into the sea. Historically, some absolutely incredible elevations have been reached by run-up. The July 9 tsunami in 1958 at Lituya Bay, Alaska recorded scouring by waves at the height of 1720 ft above normal sea level, (Halacy, 1974). This case was unique as Lituya Bay is shaped like a cul-de-sac with a ring of high and steep mountains around the foreshores. Nevertheless, the force necessary to create such a wave was tremendous, the triggering earthquake recorded a magnitude of 7.9, and bathymetric conditions present only compounded the effect. Other events have recorded run-up heights as much as 100m and 50m, however the frequency of such events, fortunately is quite low. Another effect of run-up is run-down, where the water naturally runs back to the sea. This has a doubling action against built structures, attacking them from both sides, but it also has the effect of washing people and structures out to sea. In the 1998 PNG event, only 2000 bodies from 8,000 missing were recovered. While many were left to rot in Sissano Lagoon due to the overwhelming situation, many were washed out to sea. According to the Sydney Morning Herald, 5 weeks after the event, reports came from Indonesia of bodies being washed ashore (www.smh.com.au)

4.2 Reasons For Selecting The Zhujiang Delta For The Project This project is built on the premise that there will be a very large tsunami event in the South China Sea in the not too distant future. The area is not free from earthquakes, events have been recorded as recently as September 1998 (see IRIS) While the Zhujiang delta is not under as serious threat from tsunami's as is Japan or Hawaii, there are many good reasons for my selection of the area for such a project. Japan and Hawaii have an efficient and effective tsunami warning and mitigation system in place, albeit in the early stages. The system has worked, as already the Tsunami Warning Centre has had to issue evacuation warnings to residents, which has had the effect of preventing loss of life. Japan, with it's long history of tsunami's, has naturally become wary of the dangers of living close to the coast, and have established similar warning and mitigation systems to that of Hawaii. The Zhujiang delta on the other hand, does not. The communities are only just coming out of development stage to modernisation, and undoubtedly no thought has yet been paid to tsunami warning or mitigation. At least, no published work is available to date. The coastline of Guangzhou is still subject to some measure of threat, much more so than many populated coastal regions around the world. It lies less than 1,000 kms from the "Ring Of Fire", the edge of the Pacific Rim to the east, and has bathymetric conditions in the South China Sea that are typical of the majority of events. The low-lying plains and shorelines of the estuary are becoming heavily populated; and the shape of the delta itself has the conditions to create a funneling effect in focusing a wave impinging upon the shorelines. Potentially, there is great cause for concern for the region. The delta region is very representative of the type of coastal development of populations around many parts of the world, especially developing countries, or those countries becoming modernised. China's huge population is growing at an enormous rate, and many coastal regions along the Chinese mainland are following the trend found at Guangzhou. We have the advantage of having much more data available on the Guangzhou region than we do other coastal regions of China, which enables us to make a deeper examination of the problem than we could if we were looking elsewhere.

Page 12 on 27

5)

Existing models

5.1 A Review of Models To construct a model to predict the effects of an inundation, data on a variety of topographies and environments must be combined, extending from the earthquake epicenter through to the houses themselves. The closer to the land one focuses their attention, the more localised the predictive models need to become. In order to arrive at predictions of any accuracy, one must operate models within models, and take into account the following series of events and environments, Earthquake, Tsunami, Bathymetry, Coastline, Topography, Landuse (population, buildings and roads). The Center for Tsunami Inundation Mapping Efforts (TIME) is a good example of state of the art mapping and topographical modelling efforts. It was created to assist the Pacific States in the development, maintenance, and upgrading of maps which identify areas of potential tsunami flooding. Their system is to first create earthquake event models and then map the wave propagation of these events in order to create a set of Tsunami Hazard Maps pertaining to specific areas. These are then made commercially available.

5.1 (a) Earthquakes Earthquake models are generally simple and large scale, they exhibit similar figures worldwide, as it is a standard geological occurrence, though of course there are variations inherent, such as magnitude, depth, and fault type. For the purposes of this project, modelling an earthquake is not necessary. The approach will simply be to arbitrarily, though not unrealistically choose an epicentre and a variety of magnitudes from which a tsunami and destructiveness model will be created.

5.1 (b) Tsunamis Tsunami models gain an extra complexity from sea floor topography, and position within the lunar tide period. There are two major factors that contribute to the amplitude of a tsunami; magnitude of triggering earthquake, and sea-floor topography. Depth is also a large contributing factor, the highest incidence of tsunamis occur from earthquakes at depths of 50 - 100km, though larger quakes at greater depth will have similar effect. Other contributing factors such as Coriolis force, linear and nonlinear inertia force, bottom friction, frequency dispersion should be considered but play more or less importance depending upon wave travel distance. The 1960 Chilean tsunami that travelled 17,000kms and caused fatalities in Japan would have Coriolis force playing a larger part in calculations than would a locally triggered tsunami, while bottom friction and non-linear inertia force are insignificant. (Liu P.L.F et al., 1993) Professor Nobuo Shuto of the Disaster Control Research Center, Tohoku University, Japan (www.geophys.washington.edu 1998) has created a computer generated animation of the 1960 Chilean tsunami. Logarithmic models were created using known data recorded at the event, as well as applying standard laws of wave motion theory. The US Geological Survey has also produced a number of computer generated animations of the July 17, 1998 tsunami, using separate data collected from seismic stations of the NEIC, Harvard University, and the Earthquake Research Institute http://walrus.wr.usgs.gov/docs/tsunami/low_res_agif.html

There are two approaches to creating a predictive model; statistical, or mathematical. Statistical models are constructed solely from measured event data for a particular region, within a set of parameters and are somewhat limiting. In other words, they are estimations of wave heights based upon experience in a given area. Mathematical models are formulae constructed to describe in mathematical terms the physics of any given event, based upon measured data from given areas, but attempting to be applied to all possible areas. These are somewhat limited by the complexity of the variables, and their inaccuracy. The ideal method is to combine these two, first using existing data to

Page 13 on 27

create mathematical models, which are then tuned or modified by real-time events to create realistic and accurate models. (Abe, 1995). Abe found when matching these equations against real-time events, there seems to be an uncertainty factor of about 1.5, particularly with a certain type of tsunamigenic earthquake known as a "tsunami earthquake", one which produces unusually large tsunamis for their relatively low magnitude. These comprise of about 10% of tsunamigenic earthquakes. He felt that these type of earthquakes should fit outside this mathematical model in order for it to be more accurate, and states that research needs to be done on tsunami earthquakes.

Modelling for bathymetry has usually been included within tsunami modelling, in order to increase the accuracy of localised predictive models. The Pacific Ocean has a mostly flat sea bottom, averaging 6000 7000m, with continental shelves to 200m encircling all continental land masses at Hong Kong roughly 200kms distance from the coaust, with some exception, particularly countries lying adjacent to marine trenches. Tsunamis at sea are practically unnoticeable at sea level, they are a ripple barely a centimetre or two in height. Compression of the amplitude wave due to decreasing depth will proportionately increase wave height and speed, so by the time the wave makes it's land fall, it is travelling up to 650km/h and is as high as 10-30m. The shape of the sea floor and its cross section to the coastline will greatly determine the nature of the wave. Islands and sea mounts will cause drag on the shape of the wave with some dissipation of energy, but as can be seen in the 1960 Chilean event, (Liu et al, 1993) this did not appear to apply here. The continental rise is the first major obstruction that a wave will experience, and actually acts as a buffer to dissipate a large amount of inertia, and no doubt protects coastlines from much worse waves. However the simple transferral of energy becomes reflective, ie. the initial long waves of the tsunami bounce off the continental shelf, and reflect back to sea, into the radiating wave, hence getting bounced back in shore. This is the cause of multiple waves in tsunami events. The recent PNG tsunami was particularly large due to several factors. The sea floor off the coast of Aitape drops away very quickly, to a depth of 3000m, and has an even slope from the epicentre that was 12 kms off shore. Intimately related to bathymetry, modeling of coastlines becomes more localised than the previous factors in tsunami's. Adams & Lewis (1979) made a fair attempt to categorise coastlines into nine distinguishable types, including offshore type and coastline proper. The point at which the transferance of energy travelling through the water reaches the coastline and causes the water to physically move into a breaking wave is of crucial concern, because the shape of the offshore coast immediately before tideline will determine the focus of the wave as it breaks on land. For example, if the offshore coastline is convex, the force of the wave will be dissipated out to the sides as the wave breaks, and if the offshore coastline is concave, it will focus the breaking wave into the centre of the point of landfall. Adams & Lewis chose 3 simple shapes, linear, concave, and convex for both the offshore coastline and the shoreline, creating 9 possible combinations. An interesting issue was raised by Adams & Lewis in that a model for a complicated surface can be synthesised, with some decomposition, from a model for a linear surface. This means that the basic wave properties of a

Page 14 on 27

simple, clean coastline will remain the same for a coastline with small islands and rough sea-floor, with some extra distortion and minor reflecting waves. This is important in studying the Zhujiang region, as the mouth of the estuary is dotted with various islands that will possibly modify the tsunami properties as it makes it's landfall further up stream. Topography The action of the tsunami on land, and the destruction it causes, becomes specific to the immediate topography of the area experiencing the tsunami. Detailed topographical data for any part of China is difficult to obtain, for military reasons etc, but there are in other parts of the world, particularly those prone to tsunami, systems in place to model coastlines and topographies of various communities, on a council level. To achieve this, land surveying and mapping with the specific purpose of estimating wave actions upon land in mind must be carried out. There is quite a large body of data in the form of historical accounts and eyewitness reports that can be drawn upon in the construction of wave destruction models. Earthquake scientists have paid a lot of attention to eyewitness reports, and have generally found that such reports have later been confirmed by breakthroughs in understanding from a scientific basis. In other words, people have tended to give pretty accurate reports of events, however much out of the ordinary they may have initially appeared to be. For the Guangzhou region we have no topographic and land data more detailed than contained in the maps of this paper, so construction of a model will be largely speculative, manual and one based upon observational abilities. The basis of the physical model is very general, which is less than ideal, being unable to use hard data relating in detail to the area. Using information and reports from the July 17, 1998 PNG tsunami, and other well documented events, it is hoped to present here a scenario with enough details for it to have some value. Land Use/Buildings In terms of constructing a model to predict destructiveness upon land and built environment, detailed and specific localised information is required on the nature of built structures and roads as these house and service the populations. Construction materials and design of buildings is important to how a structure will resist or give way to the destructive force of a tsunami. Of secondary importance is the road system and communications, as these will determine the emergency response times and effectivenss in clean-up procedures. (see below for mapping Land use)

Page 15 on 27

Hong Kong

Page 16 on 27

6)

Map analysis

The area chosen for this study is a rectangular region bounded by the longitudes 112.5 deg to 114.3 deg. East, and 21.75 deg. to 23.5 deg North, an area of roughly 400 by 550 kms. All maps, (except where noted) were created by MapInfo Professional, v.4, using copyright data from ACASIAN. The images were then processed in PaintShop Pro v.5. The overlaying maps were automatically registered to latitude and longitude in MapInfo, while the Landsat and Population Density maps were registered manually, as they came from different souces. The Interactive Map Presentation was specifically designed to use the "frames" facility of Java enabled broswers, to quickly and accurately compare maps showing different data.

Landsat Image The LandsatTM Image was adapted from a map produced in 1992 by Geocarto International Centre using Landsat data processed by RSGS China. (www.rsgs.ac.cn ) The colours are digitally enhanced to emulate the light frequencies as seen by the human eye, ie. it is an attempt to create a true colour image. It was produced on a different projection to the other maps, hence the 13 degree rotation from vertical, and the slight error in registration to the other maps. It offers a superb "photo-realistic" aerial view of the region, in colours that are quite accurate to the ground features, and provides physical details not present in the maps produced by vector graphics software. Most importantly, it offers a very accurate and detailed topographic picture of the region not found in the other maps. By taking note of the shadows and light source, it is possible to recognise individual mountain ranges, land contouring and relief features which will assist in determining the areas susceptible to tsunamis.

Features The Zhujiang Delta, or the mouth of the Pearl River, is the ocean outfall for practically all the major river systems of Guangdong Province. The estuary is 35 - 40kms wide and narrows to Zhujiang River 60kms inland, and has a depth of no more than 20m. A further 50 kms up river lies the province level administrative centre, the city of Guangzhou, with more than 3 million inhabitants. On the western heads is Macau, on the eastern is Hong Kong, and along its shorelines lie at least 25 towns with populations greater than 10,000 registered inhabitants. Between 10 and 20 kms offshore lie the small island systems of Wanshan and Dangan, and 20 kms inland lie the islands of Nei Ling Ding, and Qiao which occupies about 12 sq. kms. The region lies centrally at the top of the South China Sea, boxed in by the Phillipines to the east, Hainan and Vietnam to the west, and Kalimantan, more than 2000kms to the the south. Two motorways, one running up the coast 12kms inland on the west side, and one 6kms on the east side, connect Macau and Hong Kong to Guangzhou in the north. At Humen the motorway crosses the Zhujiang R. on a bridge that appears to be several km in length. There is no data available on the bridge height or construction, but it appears the engineering is taking advantage of 2 hills of greater than 30m either side of the river mouth. Between the west shoreline and the Macau-Guanzhou motorway lies a road classed as highway in the Atlas (CCPH, 1989), though whether this is accurate remains to be seen.

Land Use This region is one of the major rice producing regions, as well as one of the most productive economic zones of China. Very little land that is not hilly or mountainous remains untilled. The predominant land use of the delta region is rice farming, in the form of plain and terraced paddies and diked ponds, which lies under 20m. These areas occupy the low-lying flat and nutrient rich land formed by estuarine deposits of the rivers, and are very labour intensive; most of the work would be done by hand, with the odd mechanical hoe finding use. On the higher areas, above 30m, the land is mostly left to forest and woodland. To the north near and around Guangdong is vegetable plots and orchard farming. The

Page 17 on 27

region is very densely populated, with practically all the land under 15m being inhabited by at least 300 - 500 persons per sq.km., but mostly 500 - 1000 per sq.km. Along the shores of the delta, for roughly 50 kms on the east bank from Nantou to Songgang, and 40kms on the west bank from Tangjia, is muddy flats, which would be exposed during low tides, and covered during high tides. The whole region is dotted with small townships only 2-10 kms apart. Adjacent to the muddy flats near the mouth of the estuary are several areas of saltwater breeding plots.

Elevation The elevation data used for the creation of the topographic map was provided free by internet download from the EROS Data Centre website, and processed in ArcView and MapInfo GIS systems at ACASIAN. Since the information is available to the public over the web, it is of a lower resolution than commercial or military use data sets, being provided at 30 arc second intervals, as opposed to 5 second intervals that EROS can resolve to. Each square pixel seen in the map represents a 1 sq. km area of the Earth's surface, and the elevation given is measured at the centre of the square. The artificial colouring of the map has been grouped into 3 different hues, reds, yellows, and greens, to correlate to various important elevations related to tsunami heights. According to run-up data in the NOAA databases, an average maximum run-up height for tsunamis is 30m, so elevations at 30m and above have been coloured various shades of green, to indicate relative safety. Elevations of less than 10m have been coloured shades of red, a colour easily recognisable as danger. Simple single storey residential and urban structures are generally less than 5-6m in height, and the frequency of tsunamis of 10m height is much higher than waves of greater than 10m.

Page 18 on 27

The mouth of the estuary is protected by mountainous headlands on the east and west banks, with Wu Gui Mtn. north of Macau at 530m, and Lantou Mtn on Lantou (H.K.) at 935m. 3km inland from Houjie, on the east bank an unnamed mountain reaches 376m, while 15km inland from Humen on the NE banks lies another mountain at 530m. Guangzhou, a river city with elevations no more than 15m, lies almost 50kms upriver at the foot of a small range of mountains, varying from 372m to 433m. Two very large basins, which lie on either side of the Zhujiang R., are less than 10m in elevation. These are the rice paddies, which are low-lying, flat and featureless tracts of land, are generally inundated with water most of the year round. They are intersected with a very complex and constantly changing network of water channels. The area to the east of Guangzhou in particular, occupying more than 250sq.km is very low land, less than 3m. Along the east bank of the estuary, from Nantou through Humen to Xintang, ranging from 6 - 12 kms inland, the shoreline elevations are less than 7m. Across the river, lie 2 arms extending out from the shoreline, at Nansha, Wangqinsha and Xinken, also on low land. The Wangshan Islands at offshore from the mouth of the estuary, are generally 30m to 75m in elevation, while the Dangan Islands are very low lying, at roughly 3m, save for a few mountainous peaks to about 75-100m. Shenzen, buffering the New Territories and mainland China, despite lying in very low land, is protected from the ocean by Lantau Island, while the western banks of the estuary from Xinsha north to Nanlang are protected from the sea by the Promontory at Tangjia. The island of Qiao isn't so fortunate...although it has 2 hills over 30m, the southern shore and central inland is less than 15m, which is where most of the population is concentrated. The aspect and topography of Qiao will make it incredibly vulnerable, with quite high run-up heights, due to the relative steepness of foreshores. 50kms west of Macau, the Tanjiang R. runs through very low land, near the towns of Xin, and Jiangmen where the land is 1-2m. This low land lies at the head of the Ya Delta, and would replicate the effects found in the Zhujiang Delta.

County Populations County and Town population figures are taken from the 1990 census of urban residents, ie. those inhabitants who are registered as living in the censused area. These figures do not take into account population growth of the past 8 years (no data is available), nor does it accommodate the "floating" population, those inhabitants who are not registered as urban residents. These are itinerant workers, people on the move, visiting relatives, and persons inhabiting the region without registration...which is quite common. Although there is no data for these floating populations, estimates are that the percentages are quite high. By calculating total populations from the towns map and comparing it with the county populations figures, it will be noticed that the towns map can only account for little over half of the stated county populations. This is because only towns of greater than 10,000 inhabitants are shown. What is not shown, are the hundreds of small villages and hamlets, as well as residences on private plots of land that are interspersed among these towns. No data is available for these villages either, but estimations will have to be made based upon photographic evidence and first hand experience of rural landuse and residential distributions in these areas. Six counties immediate to the waters of the Zhujiang estuary will be vulnerable; Zhuhai Shi, Zhengshan Xian, Panyu Xian, Dongguan Shi, Bao'an Xian, and to a lesser extent Shenzen Shi. The southern shores of Hong Kong will also be vulnerable, though theyare somewhat protected by Dangan Islands. Dongguan Shi. This county has about 20 towns of at least 10,000 person, and a large city of at least 290,000 persons. This would account for more than 520,000 persons, but the county register has a population of 965,470. How do we account for the remaining 445,000? Even if those 20 towns held 20,000, we still cannot account for 215,000 persons. There must be large numbers people living

Page 19 on 27

outside the towns. If we refer to the Landsat map, we can see that there are built up areas not indicated on the Landuse map, Towns map, or the Features map. Take, for example, the triangular promontary of land with Humen near the mouth of the Zhujiang R. The Landsat Image reveals a rather large township on the southern shore of the promontary. It's not Shajiao, as it lies on the north western side of the promontary, south west of Humen. This built up area is on the water line and occupies perhaps 4 sq.kms. By comparing its size with a town of equal area, say Donglicun on the southbank of the Hong Men R., we can see that it would have a population of around 10,000 as well. So the Landsat Image is indispensable in helping us form a realistic picture of the state of the Zhujiang delta.

Town Populations The towns map and features map both lack detail for showing the number of towns and buildings, but interspersed among these large townships of 10,000, lie smaller towns, and between them, the countryside is dotted with small farm residences and houses on private plots of land at 50m distance apart. Along the shoreline of the estuary from Nantou and Macau upriver 20kms to Xintang, more than 25 towns of at least 10,000 inhabitants lie less than 4 kms inland. Most of these towns are situated on land less than 10m, except for a stretch of about 35kms from Tiangjia to Minhe, west of Xinken, which is 10-15m in elevation, yet fairly protected from the ocean, anyway.

Population Densities

Hong Kong

This map was adapted from the Population Atlas of China, and registered manually to the other maps created in MapInfo, and is accurate to about 1km. The coloured squares are bounded by a longitude interval of 1' 52.5", and a latitude interval of 1'15", giving boxes of roughly 7.34 sq. kms maximum at the top of the map. The vertical line down the centre of the map is a discolouration caused during the scanning process by merging two large map scans together. All attempt has

Page 20 on 27

been made to adjust the brightness and hues to a standardised index in the legend. The population density figures are based upon data available in 1987 so it would be safe to use the maximum figures in the ranges contained in the legend, possibly even to inflate them somewhat. Coastal regions are more desirable to live, as the weather tends to be more temperate, and it is close to a plentiful food supply. The delta land, being rich, flat and close to a source of water makes it ideal for heavy cropping. Moreover, according to a report from the American Association for the Advancement of Science, nearly 100 million Chinese are thought to have moved from the poorer provinces in the central and western regions to coastal areas in search of better economic opportunities for themselves and their families. At any given time, somewhere between 20 and 40 million Chinese are on the move, a population the size of Spain. The bulk of this large, "floating population" is concentrated in coastal provinces, precisely those areas with the highest economic growth rates. (www.aaas.org/ 1995)

Ocean Bathymetry The map of the South China Sea bathymetry has been adapted from a special map produced by the NOAA. The pixel resolution is not known, and the depth legend has been prepared from several sources (The Times Atlas of the Sea, 1989, Atlas of China, 1989) South China Sea is roughly trapezoid, being 2000kms in length, and a little over 1000kms wide. The eastern boundary of the Sea lies with the Phillipines, extending south to Kalimantan, placing it on the edge of the "Ring Of Fire", the earthquake zone which marks the edge of the Pacific Plate. There is no virtually no continental shelf on the west coast of the Phillipines, whereas adjacent to the Guangdong coastline, the shelf extends about 230kms. The shelf at 200m depth drops away relatively evenly parallel to the coastline to a fairly flat and even floor at 5000-6000m. 500 kms directly south to south-west of the Zhujiang lie two ocean rises, the Xisha and Qunsha Qundao, or Macclesfield Bank. This small rise comes to at least 200m and possibly has some reefs. This site was near the epicentre of a magnitude 5.5 earthquake within only 15 days of the 17 September 1998 IRIS event query made by the author (see IRIS Map) 1200kms south-south-west lies another rise north of the nearby Kalimantan coast, the Nansha Qundao, which is more open and scattered than the previous two. It also rises to 200m and displays some coral reefs. If one draws a line perpendicular to the coast and extending south east out to sea, at the continental shelf one can see a steep corridor roughly 200kms wide with protruding buttresses of shallower slope on each side of the corridor, at depths of around 500-1000m. Adams & Lewis have labelled this sea floor shaping as "flaring cosine relief" (Adams & Lewis, 1979) This would have a focusing effect on a tsunami wave. At 100-200m however, the silt deposits of the river outfall have created a minor protruding slope, which would also have a modifying effect on the tsunami wave. This, Adams & Lewis label as a "tilted cosine relief" (See below)

Coast Bathymetry The coastline of Guangdong is relatively straight for about 1200kms, and has a relatively even and perpendicular sloping sea floor above the edge of the continental shelf, that drops away 20m roughly every 35kms giving a 1:1750 slope. The long slope would have the effect of slowing down the tsunami, while a steeper slope would have the effect of speeding it up. From the continental shelf, the sea floor drops away quite steeply from 200-6000m in roughly 210km, giving a 1:36 slope. This demarcation of flaring relief to tilted relief will surely have a modifying effect upon an incoming tsunami.

Wave Propagation The wave propagation shown in the accompanying map is a rough estimation of the projection and radiation of a tsunami caused by an earthquake in the South China Sea. As mentioned previously, wave energy radiates following by simple laws of fluid dynamics. The long waves radiate out from the epicentre, and as they approach the coastline, compress and flatten out to follow the relief of the coast. Small islands in the mouth of the delta would not receive as high a wave as the coast proper, as the transferral of energy is still being contained under the sea level. There would be minor, localised waves on shore as the shockwave of the tsunami passed by. As Adams & Lewis (1979) have shown, these small islands have little modifying effect on the tsunami, they would rather contribute to some measure of noise in the shape of the long wave.

Page 21 on 27

As the long waves bounce off the continental shelf back to sea, the tide appears to recede. Once the long wave reaches the coast and mouth of the delta, the shallows compress the water into a rising wall and the shockwave starts to cause water to move forward. The banks of the delta at the mouth of the estuary will then compress the wave inwards on a horizontal plane by reflection, much the same way normal beach waves break around rocky headlands, and curve inwards to the beach. This sideways compression, combined with compression from the decreasing estuary bottom, can only cause an increase in forward momentum and wave height, and thus the tsunami effect is under way. Bearing in mind that the 1998 PNG tsunami struck a 30km stretch of coastline, the 40km wide mouth of the delta would absorb the entire force of such a wave. This huge body of water will then be compressed into a space only 18 kms wide, and less than 10m in depth. The inertia would carry the wave up several of the river mouths at the reach of the estuary, and would probably extend many kilometres up the rivers, much like a hurricane driven storm surge. Henry & Murty (1995) describe this effect, known as "resonance amplification" observed in the March 1964 Alaska Earthquake, where an earthquake generated wave travelled up Barkley Sound on Vancouver Island, 65kms inland to Port Alberni. This compression by the inlet caused a 3-fold amplification of the measured tsunami height at the head of the Sound.

Page 22 on 27

Run-up

Hong Kong

In the accompanying map, run-up is shown as bright red, and the areas likely to experience subsequent flooding are shown as pink. Blue represents the hydrology of the Delta region, which is derived from ACASIAN data and processed in MapInfo. This map is produced as an estimation of tsunami hazard, based on Elevation, Topography, and Wave Propagation data shown in their respective maps. The red colouring also will indicate run-down, where the run-up gets pulled by gravity downhill and back out to sea. It also indicates the areas likely to experience a wash over by the wave, where the water finds low ground inland and causes flooding. The south coastlines of Hong Kong and Macau show little red, because their shores are steeply sloped with mountains close to the water line. Run-up effects will be immediate and dramatic. Places that will receive full force of the tsunami are those areas that lie parallel to the face of the wave. In this case, the wave will have a generally eastwest orientation, so coastlines and shorelines facing south will be the worst hit. Shorelines that run perpendicular to the face of the wave will experience a less destructive effect of the tsunami, which will have a "shearing" action as it moves along the banks of the rivers upstream.

Page 23 on 27

7)

The hypothetical scenario

7.1 Set The Parameters For A Worse Case Scenario. An earthquake of magnitude 8.5 occurring 100kms below the Earth's surface is recorded just south east of the Macclesfield Bank, in the central South China Sea. It caused a massive land slippage 250kms offshore, where a large body of silt deposition from the Pearl River Delta dislodged and resettled, causing a 100m drop in the sea floor over an area of almost 20,000 sq.kms. It's 1.00 o'clock on a Monday morning. Calls are placed to Hong Kong, Macau and Guangzhou to warn local authorites of the recorded event, but they are unsure how to respond. Twenty minutes later, a 20m wall of water stretching 60kms along the coast of Guangdong is approaching the Pearl River Delta. The tide is rising, and the outflow of water from the Pearl River delta has slowed due to the incoming tide. The wave enters West Lamma Channel between Hong Kong and Lantou, breaking on the New Territories just west of Kowloon. It travels up the inlets of the Pearl, Xi, and Tan Rivers, increasing in height due to compression of the river bottoms and sides, and breaking on all shorelines facing south. When it reaches the mouth of the Zhujiang River, the wave is 30m tall. The main wave travels up the Pearl River, past Huangpu and surges as far as the city of Guangzhou. Secondary waves travel up the Xi and Tan Rivers, with similar effects. Most of the encroaching water finds low ground and floods the region, while the rest recedes as run-down, back to sea. 20 minutes later, a second, larger wave generates off the coast, and makes its way inland in the same way, penetrating further than the first. A third and final wave, of lesser size than the first two arrives another 20 minutes later

7.2 Effects on Land And Populations 7.2 (a) Dynamics of the wave action against land. The East Side. At the entrance to the Zhujiang estuary, the 20m wave is just forming. It breaks on the southern shores of the islands of Lantou, Hong Kong, Lamma, and Po Toi on the west side. Lantou and Hong Kong Islands shield the city of Kowloon and most of the New Territories, but a smaller wave of several metres passes through the Lamma channel and breaks on land at Stonecutters Island, and Kao of the New Territories. The city of Shenzen is not affected, but the towns of Baishizhou and Rongshujiao experience smaller waves of 5m. There is extensive flooding along the low-lying land of the Shenzen River as the 5m surge travels upstream towards Shenzen. The towns of Xin'an and Ziyou along the shore north to Shiwei, all on land under 6m in elevation, experience 10-15m waves that break 100m inland after travelling 1km across the freshwater breeding plots, and encroach as wash up to 1km inland. The road from Nantou to Humen is damaged in many places. One of the worst hit shorelines is the promontary at the mouth the Pearl River itself, where there is a large town on the southern shore, and several towns, on the other side including Humen. Here the wave had grown to 30m, and since the land is all under 6m at this point, the wave breaks and washes over the area to 1.5km inland. The wave travels 20kms up the Pearl River as a 30m surge, having a shearing effect on the banks of both sides of the river. The land on the east bank is more than 1-3m in elevation, and the soft irrigated soils that support the rice paddies get washed away into the surge. The wave spreads through the intricate network of river systems towards Dongguan, inundating the whole low-lying paddy region with a muddy mixture of estuarine and salt water. This surge and shear effect was also replicated up the Xi River as far as the city of Jiangmen, and the Tan River as far as the city of Xin. As the surge wave reaches Huangpu and Xintang, the Pearl River turns to the west. Here the forward momentum would carry most of the water onto land where it would break on another large town, also on the banks of the river. The Guangzhou railway experienced flood damage between Xintang and Huangpu, effectively cutting off rail access. Since the wave, as it travelled up the river lost a lot of it's size and force due to dissipating into the side rivers and inundating the low-lying paddies, the resonance amplification effect

Page 24 on 27

is not as pronounced as the surge that travelled up the Xi River to Jiangmen. Guangzhou experienced a minor surge where water breached some of the roads and pavements. The West Side. On the west side the 20m wave breaks on the southern shorelines of Hengqin, Coloane and the Gaolan islands. As these islands have sloped shorelines, as do Lantou and Hong Kong, they experience run-up effects to the height of more than 50m. These islands shield the city of Macau proper, but not Gongbei, Zhouzai and Zhuhai east of Macau. Tiangjia, in the center of a small bay experiences a 25m wave that completely washes over Tonggu Point. The estuarine island of Qiao experiences a 25m wave also, with a run-up of up to 50m. Townships along the western shorelines from Xiazha to Linx 20kms north, experience only a minor change in wave heights due to their aspect, but Xinken, Xin'an, Tongxing and Wangchinsha bear the brunt of a 30m wave, depite an elevation of 15m. The wave runs up the slope on the southern tip, and washes over into the towns situated among the paddies. The landsat image shows an island in the stream south of Nansha that does not appear on the MapInfo maps. This is probably reclaimed land converted to paddies, at barely 1m elevation. The wave washes over this 5km island and breaks on Nansha and Nanheng, which experience run-up to 50m. The tributary arm that runs into Heng Men (entrance) experiences minor changes in water height as it is protected at the mouth by an island, and runs perpendicular to the direction of the wave. The river entrances at Minhe and Nansha, however, do catch the force of wave, and it surges up stream past Xinan and Nansha, inundating the large basin at Shunde and south of Nanhai.

7.2 (b) Effects on Population. Review. Effects of a major tsunami on a population should be measured in terms of (1) immediate effects, (2) short term effects, and (3) long term effects. Immediate effects are deaths and injuries in the first few hours, as a direct result of the wave, such as drowning and physical trauma caused by impact with water borne objects. A series of three waves would have a combined effect much worse than 3 separate waves at different times, because survivors of the first wave, weakened by their struggle to survive, would probably not last through a second and third wave. Short term effects would be experienced in 1-3 days subsequent, where stranded survivors, too weak to move and in need of urgent medical attention, are unable to withstand the elements before help arrived. Long term effects will be found 5 days - 2 weeks afterwards, as homeless and hungry survivors struggle to stay alive with the limited assistance of the emergency authorites. In the PNG tsunami, death tolls were rising up even 2-5 weeks after the event, where survivors were dying from hunger, exposure, infection and disease (Phillips, 1998, and BBC News, 1998). Also in the long term, is the effect on the food supply of the region, as this area is one of the country's main producers of rice. Inundation by salt water will make the soil untillable for a long time, unless efforts are made to divert the flow of rivers to continually flood the land in order to leach the soil of the salt. Estimations. A few locations pinpointed around the delta will be examined to estimate effects on populations. At the mouth of the Zhujiang River, on the east bank, where a triangular point of no more than 6m elevation sticks out into the estuary, there is a large (unnamed) town of at least 10,000 south east of Shajiao. The 1990 population densitiy figures for this area show up to 1000 persons per sq.km over an 7.3 sq km section. The wave is estmated to wash up to 1.5 kms inland, which would destroy the town and residences along the coast. Here the immediate toll could be as much 10,000. Across the river mouth to Nanheng and Nansha is another area of concentrated population, roughly 15 sq. kms with 500 1000 persons per sq. km. Upstream to a point midway between Huangpu and Xintang, is a large town on the banks of the river just as it starts to narrow, and turn west. A lot of water would wash over this area, as the wave compressed as it surged upstream. This township, shown on the Landsat, but not the MapInfo maps, appear to occupy an area equal to the city of Macau proper. There are perhaps 20,000 persons within a 3km arc from the tip of the point. This town is backed by mountains at least to 100m, so run-up and run-down wil be experienced.

Page 25 on 27

7.2 (c) Effects on the Natural and Built Landscape Housing Only structures that lay in the path of the wave itself, and the backwash would be destroyed outright. Photographic documentation of building types in intensive subsistence regions such as the Zhujiang Delta show few wooden structures, but mostly brick and cement rendered houses with tiled roofing. Brick houses have some measure of protection against the action of water, but should the structure give way, injury to occupants is great due to the nature of the building materials. Wooden structures, such as those mostly employed by the PNG residents, are flimsy and easily washed away, but due to their flimsy nature, are not as great a threat to the occupants. Structures that may have only been weakened by the first wave yet still remained standing, may then be demolished by the second and third waves. Structures ruined due to water damage or flooding would be rendered unusable until repairs and clean-up efforts could be made. Space for makeshift medical facilities and emergency housing serviced by road access needs to be created Transport Roads do not stand up well to water damage; the scouring action of flowing water erodes the surface rapidly, and a smooth surface is needed for them to be practical. As most of the population, being surrounded by so much water, would travel by small boat, transport will be major concern for emergency services in reaching survivors with aid. Boats will be good forshort distance travel, but road travle will be necessary to service all areas of the Zhujiang Delta effectively.There are a number of bridges evident on some maps, and the Landsat image, but data on these is not available. Transport will be crucial for emergency services to reach devastated areas, and service them Agriculture. Destruction of the rice paddies would be a major concern. Inundation by a mixture of salty and estuarine water, as well as the scouring action of the waves which would upset all channeling and landscaping work, would have a major impact on the food producing potential of the region. As the region relies heavily on intensive farming production, upset to the normal day to day routines that maintain the upkeep of the food growing areas will cause undamaged areas to be unattended, thus raising the possibility of losing perfectly good crops.

Page 26 on 27

8)

Lessons

It must be remembered that the above hypothetical is a worst case scenario. The maximum end of estimations has been taken to illustrate the argument of this project, which is that with such large concentrations of populations so close to areas of possible threat from tsunamis, and without adequate warning and mitigation systems, and with difficult terrain for emergency services to access, there lies the very real possibility of a disaster of enormous proportions. Hopefully this project will highlight the problem and generate interest in beginning the process of preparedness. Some of the historical examples given, however, show that the enormous wave heights and the destructive effects of a tsunami are very real, and their importanceshould not be overlooked. Risk assessment relies on information and data from a wide variety of sources, spanning a number of fields of scholarship, and needs to be combined in an effective way with the specific purpose of tsunami preparedness in mind. The July 1998 Papua New Guinea tsunami has been a valuable model for all aspects of a tsunami disaster, and application of events and lessons learned from it could be applied effectively to the Zhujiang Delta. Sadly, developing countries, and individual communities within newly developed countries, that have not yet reached the standard of living comparable to that of developed countries often get overlooked, not only in plans for preparedness, but also in allocation of aid and funding for restoration.

© Jason Wotherspoon, 1998.

Page 27 on 27