Integrating Disaster-management Information: Mauritius

62 downloads 0 Views 394KB Size Report
industrial and nuclear in disasters. • Nuclear disasters;. • Wars;. • Pollution;. • Deforestation; ..... rocky cliffs, where oil cannot penetrate into the rock and ... to many kinds of plants and animals. More recently ... Coral reefs. 9. Sheltered tidal flats.
7

Integrating Disaster-management Information: Mauritius

GENERAL INFORMATION ◆

Implementing Institution: Faculty of Science, University of Mauritius



Head : Prof. Indur Fagoonee (Vice Chancellor)



Details of Institution: Address: University of Mauritius, Réduit, Mauritius Tel.: (+230) 454 1041, (+230) 464 9958 Fax.: (+230) 465 6928 E-mail: [email protected]



Implementation Period : 2005 to 2015



Costs : The ten-year project has a budget of some US$175,000.

99

100 V OLUME 12: E XAMPLES OF N ATURAL D ISASTER M ITIGATION IN S MALL I SLAND D EVELOPING S TATES

SUMMARY Mauritius, which covers an area of 1,865 square kilometres, lies in the Indian Ocean some 900 kilometres off the east coast of Madagascar. A significant area of the island’s coastal zone, including its lagoons and the islets and coral reefs around the mainland, does not rise more than 32 metres above sea level. Areas lying within this range are either directly or indirectly susceptible to various types of natural or anthropogenic disasters. Such land is prone to seawater inundations and erosion from tsunamis and cyclones while the lagoons are prone to sedimentation and erosion. Furthermore, the lagoon ecosystems and the coastal regions are also at risk from algal blooms, invasions of poisonous jellyfish, oil spills and other pollution hazards. These areas are also prone to floods, landslides, droughts, heat waves or periods of cold weather, forest fires and other possible calamities, including volcanic eruptions. Satellite imagery coupled with a Geographic Information System (GIS) can be used to study the potential impact of disasters and, through proper modelling and planning, to develop appropriate mitigation measures regarding these disasters. Research efforts at the University of Mauritius have been targeted towards creating models based on various scenarios aimed at developing an integrated disaster-management and information system for handling a range of disasters, including cyclones, tsunamis and flood inundation, erosion and landslides, and oil spills.

BAC KG RO U N D A N D J U S T I F I C AT I O N Disasters are inevitable; it is not a question of if the next one will occur but when it will occur. In addition, recent years have seen increases in the frequency of disasters, the damage they cause and the cost of the response, especially in terms of relief and rehabilitation efforts. For example, during the first decade of the twentieth century, nine disasters were recorded whereas 4,000 disasters were recorded during the final decade of the century. As a result, 1990-1999 was designated as the International Decade for Natural Disaster Reduction (IDNDR) by the United Nations. This period also witnessed a three-fold increase in the number of large natural disasters compared to 1960-1969, and, in 1998, more major natural disasters occurred than in any other year on record. Between 1971 and 1995, an average of more than 130,000 people were killed each year owing to natural disasters while more than 140 million people suffered from injuries, homelessness and other adverse consequences. Indeed, during each year of the IDNDR, natural disasters resulted in the loss of life of some 128,000 people and adversely affected an average of another 136 million each year. Recent unfortunate events around the world have again brought into sharp focus the need to prepare societies to face the challenges posed by various disasters. It has become crucial for all nations to formulate policies and to develop infra-

Integrating Disaster-management Information: Mauritius

structure and control systems that minimize the threat to human lives and resources. Society must not only forge effective plans for disaster prevention, mitigation and management but also build the necessary infrastructure to ensure an effective response. Disasters can be classified in several ways. One distinction is that between man-made, or anthropogenic, disasters and natural disasters. Another classificaTable 1

101

tion relates to the main controlling factors leading to a disaster, which may be meteorological, geological, ecological, technological or even extraterrestrial. Descriptions and examples of various types of disasters that threaten societies and for which there is a need to plan preventive measures and disseminate information widely are presented in table 1. Mauritius is prone to many of the disaster types listed in this table.

Classification of disasters.

Disaster Type

Description

Examples

Geological

The concept of “terra firma” is often disproved when the Earth “shakes”’ or “shrugs”. It is necessary to understand and evaluate such instances in order to prepare preventive plans for any subsequent disasters, using such current techniques as remote sensing and Geographic Information Systems (GISs).

• Earthquakes and tsunamis; • Volcanoes; • Landslides and mud flows; • Dam failures and dam bursts; • Mine explosions; and mine collapses.

Medical

Medical disasters are preventable. The roles of physicians, pharmacists and patients are significant in the prevention or elimination of such disasters.

• Disease epidemics, e.g., HIV/AIDS; • Improper diagnostics; • Unlicensed/adulterated; medicines; • Incorrect medication.

Terrorism

Many countries are experiencing various types of terrorism, which is resulting in the loss of life and property. Preventive management methodologies must be sought to combat this type of disaster.

• Bomb blasts; • Extortion and kidnappings; • Hijackings; • Insurgency; • Riots.

Climate- and water-related

This type of disaster can either be brought about by avoidable attitudes (e.g., lack of concern) and actions of people or result from acts of nature, which can sometimes be unexpected and which are often unavoidable and cannot be controlled.

• Cyclones and hurricanes; • Tornadoes; • Heavy rains; • Flood and drainage problems; • Drought; • Famine; • Heat/cold waves; • Coastal erosion; • Lightning strikes; • Hailstorms; • Avalanches.

102 V OLUME 12: E XAMPLES OF N ATURAL D ISASTER M ITIGATION IN S MALL I SLAND D EVELOPING S TATES

Table 1 (continued)

Classification of disasters.

Disaster Type

Description

Examples

Biological

This type of disaster typically arises as a result of negligence in the maintenance of health and hygiene practices by those handling food and improper use of materials used to prepare food or quarantine checks at international borders.

• Biological warfare; • Disease epidemics in domestic animals and wildlife; • Food poisoning; • Pest and disease outbreaks in crops; • Invasions of non-native organisms.

Chemical, environmental, industrial and nuclear

Improper maintenance and operational practices in various industries can result in disasters.

• Chemical and industrial disasters; • Nuclear disasters; • Wars; • Pollution; • Deforestation; • Wildlife disasters.

Accident-related

Disasters can be termed as in-depth, intense or widespread accidents. These disasters, frequently caused by errors of judgment, can often be avoided.

• Airplane accidents; • Rail and road accidents; • Ship collisions; • Structural collapses; • Electricity and gas fires; • Festival- and crowd-related accidents; • Forest, urban and rural fires; • Explosions; • Computer crises, e.g., viruses; • Meteorites.

Another useful distinction that can be made between disasters relates to the duration of their impact and the time needed for forewarning. The timescales of disasters differ from a few seconds in the case of earthquakes, for example, to hours for cyclones, days for floods and months or even years for droughts. While natural disasters occur on all continents, susceptibility differs from one area to another. Table 2 provides global statistics on the frequency of natural disasters and the associated human losses for the period from 1947 to 1980, while details of their distribution are given in

table 3. For example, cyclones and hurricanes (which accounted for 40.8 per cent of fatalities), earthquakes (36.8 per cent) and floods (15.9 per cent) were the most devastating disasters while tsunamis accounted for just 0.4 per cent of the fatalities. During this period, 86.2 per cent of the fatalities occurred on the Asian continent. This data can be compared to data from more recent natural disasters. For example, the earthquake that occurred in Pakistan in October 2005 caused more than 30,000 fatalities, while the 9.0-magnitude earthquake of December 2004

Integrating Disaster-management Information: Mauritius

Table 2

Fatalities caused by natural disasters between 1947 and 1980 (from Shah, 1983).

Disaster Type

No. of events

Earthquakes Tsunamis Volcanic eruptions Floods Cyclones and hurricanes Tornadoes Severe storms Fogs Heat waves Avalanches Snowfalls and extreme cold Landslides Total

Table 3

103

Fatalities

180 7 18 333 210 119 73 3 25 12 46 33

450,048 4,519 9,430 194,435 498,516 7,648 22,977 3,550 7,470 5,025 13,197 5,493

1,059

1,222,308

Distribution of fatalities caused by natural disasters, 1947-1980 (from Shah, 1983).

Region

No. of fatalities

Asia Caribbean and Central America South America Europe Africa North America Oceania Total

near Sumatra, Indonesia, which triggered the Indian Ocean tsunami, is estimated to have killed at least 80,000 people, with more than 200,000 estimated to have died from the resulting tsunami and coastal run-up. As these two examples illustrate, more than 95 per cent of the death toll arising from natural disasters occurs in

1,054,090 50,676 49,275 28,694 23,540 11,531 4,502

Percentage of total 86.2 4.1 4.0 2.3 1.9 0.9 0.4

1,222,308

developing countries where there is a high incidence of poverty, vulnerability and an inadequate capacity for disaster prevention, mitigation and response.

THE YOKOHAMA MESSAGE AND DISASTER REDUCTION It is nevertheless possible to minimize potential risk by developing early-warning

104

V OLUME 12: E XAMPLES

OF

N ATURAL D ISASTER M ITIGATION

strategies, preparing and implementing development plans that provide resilience to such disasters and helping to rehabilitate communities hit by natural disasters. The 1994 Yokohama Message that derived from the World Conference on Natural Disaster Reduction noted that “Disaster prevention, mitigation and preparedness are better than disaster response in achieving the goals and objectives of the Decade. Disaster response alone is not sufficient, as it yields only temporary results at a very high cost. We have followed this limited approach for too long. This has been further demonstrated by the recent focus on response to complex emergencies, which, although compelling, should not divert from pursuing a comprehensive approach. Prevention contributes to lasting improvement in safety and is essential to integrated disaster management.” The Yokohama Message went on to state that “Regional and international cooperation will significantly enhance our ability to achieve real progress in mitigating disasters through the transfer of technology and the sharing of information and joint disaster prevention and mitigation activities. Bilateral and multilateral assistance and financial resources should be mobilized to support these efforts.” The Yokohama Message further noted that “The information, knowledge and some of the technology necessary to reduce the effects of natural disasters can be available in many cases at low cost and should be applied. Appropriate technology

IN

S MALL I SLAND D EVELOPING S TATES

and data, with the corresponding training, should be made available to all freely and in a timely manner, particularly to developing countries.”

NEW TECHNOLOGIES Although there has been a dramatic increase in the magnitude and frequency of disasters over recent decades, there has also been a dramatic increase in the technical capabilities to mitigate them. Technological breakthroughs in information technology, telecommunications and geomatics have led to their increasing use in disaster management and mitigation (DMM). Such geomatic tools as remote sensing, Geographical Information Systems (GISs) and global positioning systems (GPSs) can be successfully deployed in DMM programmes only when detailed knowledge is obtained about the expected frequency, character and magnitude of hazard events in an area. GIS provides a powerful tool for collecting, storing, retrieving, transforming and displaying spatial data. It also allows scientists to combine different kinds of data in models that can be used during different stages of disaster management. At the University of Mauritius, remote sensing techniques have been combined using GIS and GPS tools into an integrated platform known as the disaster-management information system (DMIS). The DMIS has been developed as a strategy for disaster management and a comprehensive framework for monitoring, assessing and mitigating disasters. DMIS analyses can also identify gaps in current practices

Integrating Disaster-management Information: Mauritius

and recommend appropriate strategies for disaster management using these technologies. Preliminary results obtained from the use of these technologies within the DMIS have been obtained for Mauritius by studying zones that are potentially at risk from floods, landslides, oil spills, storm surges and tsunamis.

DESCRIPTION Mauritius is prone to many of the disaster types listed in table 1, such as cyclones, earthquakes, flooding, hurricanes and landslides. As a result, in 2005, the Faculty of Science at the University of Mauritius initiated a 10-year project designed to integrate disaster-management information to help the island better prepare itself in the event of a natural disaster. Research efforts at the University of Mauritius have been targeted towards creating models based on various scenarios aimed at developing an integrated disaster-management and information system for handling a range of disasters, including cyclones, tsunamis and flood inundation, erosion and landslides, and oil spills.

A N I N T E G R AT E D D I S A S T E R M A N A G E M E N T I N F O R M AT I O N SYSTEM FOR MAURITIUS Scientists at the University of Mauritius now have access to information-gathering and organizing technologies such as remote sensing and GISs, which have become an integrated, well-developed

105

and successful tool in disaster management. This has been made possible not only in Mauritius but elsewhere owing to the fact that the requirement for hazard mitigation and monitoring ranks high when new Earth observation satellites are being planned. Remote sensing and GISs provide a database from which the evidence of previous disasters can be interpreted and combined with other information to arrive at hazard maps that indicate which areas are potentially at risk. The use of remote-sensing data, such as satellite imagery and aerial photographs, allows scientists to map the variabilities of terrain properties, such as vegetation, surface water and geology, in both space and time. In addition, satellite images can provide useful environmental information on a wide range of scales, from entire continents to details covering a few metres. Many disasters, including cyclones, droughts, floods and volcanic eruptions display precursors that satellites can detect. Remote sensing also allows such events to be monitored while they are occurring, thus helping scientists to plan for and monitor an event. Over the past few decades, the importance of effective information management has been increasingly recognized by agencies involved in disaster management and several countries have set up DMISs according to their own specific needs. Applications of these systems include emergency-response planning, short-range early-warning systems and long-range mitigation and prevention

106

V OLUME 12: E XAMPLES

OF

N ATURAL D ISASTER M ITIGATION

planning. Mauritius, for example, has taken steps to improve communication networks and to promote the exchange and dissemination of information for disaster preparedness, including earlywarning and pre-disaster planning. It has also improved relevant training for personnel and access to early-warning, risk-assessment and disaster-management technologies. Directed by the Prime Minister’s office, the Government of Mauritius has also established a disaster emergency fund aimed at providing efficient and quick responses and relief. Mauritius, however, does not yet have an integrated DMIS platform for dealing with disasters. The ability of leaders and administrators to make sound disaster-management decisions, including the analysis of risks in order to decide on appropriate countermeasures, can be greatly enhanced by the integration of information from different sources. For example, to understand the full shortand long-term implications of floods and to plan accordingly require the analysis of combined data on meteorology, topography, soil characteristics, vegetation, hydrology, settlements, infrastructure, transportation, population, socioeconomics and material resources. This information comes from many different sources and it is difficult to bring it all together. The establishment of an integrated DMIS requires: • an outline of the information that disaster managers need during different phases of a disaster cycle;

IN

S MALL I SLAND D EVELOPING S TATES

• a way of sharing experiences with countries that have recently created their own DMISs; and • making recommendations for the integration of national, regional and international efforts. In addition, there are two essential preliminary stages before a DMIS can be established: • the purpose of the system must be defined by identifying the users and the end products; and • existing databases must be identified and integrated into the DMIS system. To effectively reduce the impact of a natural disaster, a comprehensive strategy for disaster management is required, often referred to as the “disaster-management cycle”. The information requirements of disaster managers necessitate two distinct but closely related categories of activities. Pre-disaster activities involve analysis and research, risk assessment, prevention, mitigation and preparedness while post-disaster activities include response and relief efforts, rehabilitation and reconstruction. Accordingly, there are two categories of disaster-related data: pre-disaster baseline data and post-disaster real-time data on the impact of a disaster and the resources available to combat its effects. A DMIS must therefore take into account preparedness, planning, mitigation, response and recovery, and national DMIS databases should include: • hazard assessment maps;

Integrating Disaster-management Information: Mauritius

• vulnerability assessment maps; • demographic distribution maps; • maps of infrastructure and critical facilities; • maps of logistics and transportation routes; • human and material response resources; and • communication facilities. Nowadays, the hazard and vulnerability assessments and mapping components of a DMIS, typically based on GIS information, are the cornerstone of preparedness planning as well as planning and implementation of a mitigation programme. The development of these databases in-country must take place from the bottom up, that is, from the database of the lowest administrative unit in the country, for example a sub-district, to the district, provincial and then to the national database. Pending the establishment of such a network in Mauritius, scientists at the University of Mauritius have used information and communication technologies and geomatic technologies as first steps towards the creation of an integrated DMIS platform for the country. At present, the system is based on a series of four case studies focusing on inundation maps for flood-prone areas at risk from erosion, potential landslides, oil-spill preparedness strategies and contingency plans, and tsunamis. Case Study: Flood Map In 2002, an assessment of the flood-prone

107

areas of Mauritius was carried out by GIBB Ltd., consultants for the Water Resources Unit. Most of the flood-prone areas assessed were selected from some 200 sites already identified by the National Drainage Committee (NDC) in collaboration with local municipalities and district councils as well as from listings held by the Ministry of Housing, Lands and the Environment. The consultants also carried out additional reconnaissance surveys to identify other sites, bringing the total to 315. For each site, an assessment of the problems, causes and damage/risk factors was made. Although the study was not exhaustive, it did cover all the island municipalities and district councils. However, what was still lacking was a comprehensive map showing the risk potential for the entire island. A flooding-potentiality map indicates areas of Mauritius that are potentially at risk of flooding during heavy rainfall, cyclones and torrential rain. In Mauritius, some 60 per cent of the annual rainfall occurs during the summer season (October to April) when torrential rains and cyclones can occur. Flood maps typically take into account information from several maps, including a rainfall map, a slope map, a land cover map, a river drainage map, a soil map and a lineament map, which pinpoints the locations of geological faults and fractures. Scientists at the University of Mauritius input rainfall and slope maps into their preliminary modelling studies. The rainfall map was derived using the long-term (1961-1990) mean of rainfall for February, the wettest month

108 V OLUME 12: E XAMPLES OF N ATURAL D ISASTER M ITIGATION IN S MALL I SLAND D EVELOPING S TATES

(when the island is the most at risk of flooding) using data from 193 rainfall stations located throughout the island. A preliminary weighting of values ranging from one to ten has been given to the rainfall map: the higher the rainfall, the higher the weight assigned. The reclassified values that have been adopted are shown in table 4. The slope map (in degrees) has been produced from a 75-by-75-metre digital elevation model of Mauritius. Again, a weighting (using values from one to ten) has been given to the various slopes. In Table 4

Table 5

this case, the steepest slopes have the lowest weighting while a higher weighting is attributed to more level areas because flatter, less sloping areas are more prone to flooding. The reclassified values adopted are presented in table 5. When laid over one another, the reclassified values of these two maps can be multiplied to obtain a flooding-potentiality map with values ranging from one to 100, where the higher values indicate a higher risk of flooding. In this study, the flood risk probability of any geographical location on the island was then catego-

Reclassified rainfall values used for input into the rainfall map. Rainfall range (mm)

Reclassified value (weighting)

125 - 216 216 - 250 250 - 282 282 - 311 311 - 340 340 - 372 372 - 406 406 - 443 443 - 480 480 - 535

1 2 3 4 5 6 7 8 9 10

Reclassified slope values used for input into the slope map for the flooding-potentiality map. Class of slope (degrees)

Reclassified value (weighting)

0 - 0.5 0.5 - 1 1 - 1.5 1.5 - 3 3-7 7 - 13 13 - 21 21 - 31 31 - 43 43 - 70

10 9 8 7 6 5 4 3 2 1

Integrating Disaster-management Information: Mauritius

rized into six classes (table 6). When plotted out, a map showing the risk categories of all regions of the island is generated (fig. 1). As it stands, this flooding-potentiality map has two main limitations. First, the

Table 6

109

slope and rainfall weightings have not been finalized yet and research is under way to finalize them. Second, other maps, including those concerned with geology, soil, land cover, lineament, river drainage and other drainage systems (for example in towns and villages), have not yet been

Risk categories from the flooding-potentiality map. Values range from 1 to 100 and are obtained by multiplying together all possibilities from the reclassified values for rainfall and slope presented in tables 4 and 5, respectively. For example, the weighting for high rainfall (10) multiplied by the weighting for low slope (10) produces a value of 100, i.e., a situation of extreme risk.

Risk category

Range of values

Extreme Very high High Moderate Fair Low

90 - 100 80 - 90 70 - 80 50 - 70 20 - 50 10 - 20

Figure 1 Floodingpotentiality map using rainfall and slope as inputs.

Flooding-potentiality Map Risk category Extreme Very high High Moderate Fair Low

1:300,000 0 3.75 7.5

15

110

V OLUME 12: E XAMPLES

OF

N ATURAL D ISASTER M ITIGATION

incorporated. Only when this information has been incorporated into a single data set will the flooding-potentiality map reach a sufficient level of accuracy to allow detailed analyses to be carried out. Together with the capacity of the GIS, the model should allow scientists to answer such questions as “Where will flooding happen?”, “Why is flooding happening?”, “What can be done to prevent flooding from happening?”, and “What will happen if something is modified?”. To achieve this goal, additional research and analysis are required to produce the other maps in the desired GIS format and for the corresponding weightings to be finalized. Nevertheless, scientists at the University of Mauritius have demonstrated that a GIS can be an effective tool for creating a flooding-potentiality map using relatively simple algebra techniques for mapping rather than complex models based on physical measurements. Case Study: Landslidepotentiality Map Landslide-potentiality maps indicate areas that are potentially at risk of landslides during periods of heavy or torrential rainfall and cyclones. To date, Mauritius lacks a comprehensive map that shows the risk potentiality for landslides across the whole island. The present study aims to produce such a landslide-potentiality map using a GIS. The inputs to a landslide-potentiality map are typically information from a rainfall map, a slope map, a land cover map, a geological map and a soil map. In this preliminary study, rainfall and slope maps were

IN

S MALL I SLAND D EVELOPING S TATES

used as inputs. Information from other maps, which are currently being processed, will eventually be added to the model landslide-potentiality map. As in the flooding-potentiality case study, the long-term (1961-1990) mean February rainfall was selected because this is when the risk of landslides is greatest. Again, the rainfall map was reclassified according to table 4. However, in this case, steeper slopes were given higher weightings (ranging from one to ten) based on the definitions shown in table 7 because the steepest areas are more prone to landslides. When multiplied together, these two reclassified values – rainfall (table 4) and slope (table 7) – create the figures for the landslide-potentiality map, with values ranging from one to 100, where the higher values indicate a higher risk of landslides. In this study, the probability of landslide risk for any geographical location on Mauritius was divided into six categories as indicated in table 8 and illustrated in figure 2. This landslide-potentiality map has the same two main limitations as the flooding-potentiality map and is also being refined and updated as new information is generated. Case Study: Oil-spill Sensitivity Map Since the direct impact of oil spills in the marine environment is generally widespread and long term, such spills can have devastating consequences on coastal and marine habitats, wildlife and human health

Integrating Disaster-management Information: Mauritius

Table 7

111

Reclassified slope values used for input into the slope map for the landslide-potentiality map. Class of slope (degrees)

Reclassified value (weighting)

0 - 0.5 0.5 - 1 1 - 1.5 1.5 - 3 3-7 7 - 13 13 - 21 21 - 31 31 - 43 43 - 70

1 2 3 4 5 6 7 8 9 10

removing oil spills. In this context, the relevant authorities have developed atlases that provide a means of determining the most sensitive marine and coastal areas that might be adversely affected should such a pollution incident occur.

Landslide-potentiality Map Risk category Extreme Very high High Moderate Fair Low

1:300,000 0 3.75 7.5

15

Figure 2 Landslide-potentiality map using rainfall and slope as inputs.

as well as economic activities such as fishing and recreation. As a result, the Government of Mauritius, through the Environmental Protection Act of 2002, mandated prescriptions of procedures for cleaning up and

With the aim of developing a nation-wide sensitivity mapping system for Mauritius, various atlases have been incorporated into a menu-driven computer system integrated with GIS and database technology. These data also provide valuable resources and logistical information that are crucial for decision-making during a marine pollution incident.

The environmental sensitivity index (ESI) technique developed by the United States National Oceanic and Atmospheric Administration was used to organize

112

V OLUME 12: E XAMPLES

Table 8

OF

N ATURAL D ISASTER M ITIGATION

IN

S MALL I SLAND D EVELOPING S TATES

Risk categories from the landslide-potentiality map. Values range from one to 100 and are obtained by multiplying together all possibilities from the reclassified values for rainfall and slope found in tables 4 and 7, respectively.

Risk category

Range of values

Extreme Very high High Moderate Fair Low

80 - 100 60 - 80 45 - 60 30 - 45 15 - 30 1 - 15

information for such criteria as shoreline sensitivity, biological resources, exposure to wave and tidal energy, and human-use resources. This ranking system provides planners and those charged with responding to oil spills with a tool for establishing priorities for protection and clean-up operations in the event of a spill or in a planning and training situation. The shoreline habitats delineated for Mauritius are presented in order of increasing sensitivity to spilled oil as described in a 1987 report produced for the Ministry of Housing, Lands and the Environment. Factors such as their exposure to wave and tidal energy, biological productivity and ease of clean-up of the inter-tidal habitat determined the relative sensitivity of the shoreline (table 9). Examples of shorelines ranked as “one” include steep and exposed rocky cliffs, where oil cannot penetrate into the rock and will quickly be washed off by the action of waves and tides. Shorelines ranked as “ten” include protected, vegetated wetlands, such as mangrove swamps and saltwater marshes. Oil in these areas is likely to remain for a long period, penetrate deep into the substrate and cause damage

to many kinds of plants and animals. More recently, the inter-tidal habitats of Mauritius, which cover some 170 kilometres of shoreline, were mapped during ground surveys conducted from June 2003 to January 2004. Observations were taken from 09:00 to 14:30 each day. The inter-tidal habitats were then delineated directly onto 1:25,000 scale Mauritian geological topographic maps. Data were collected using a GPS receiver with an accuracy of three metres. Additional information from historical sources, maps and aerial photographs was also utilized. These data were used to develop layers representing different shoreline types (table 9) that were then compiled into GIS data sets. A table of contents interface organizes and controls the drawing properties of the GIS data layers. A GIS also enables the incorporation of different symbols to represent important environmental and human resources that could be affected by an oil spill. Such map elements include birds (sea birds, shore birds and wading birds), public beaches, hotels and special areas desig-

Integrating Disaster-management Information: Mauritius

Table 9

113

Shoreline types and their corresponding sensitivity as categorized in a 1987 report produced for the Ministry of Housing, Lands and the Environment. A ranking of “one” represents shorelines that are least susceptible to damage by oiling. A ranking of “ten” represents the locations most likely to be seriously damaged.

Sensitivity index 1 2 3 4 5 6 7 8 9 10

nated by legislation such as fishing resources and nature reserves. These areas are indicated on the map specifically to aid and direct response efforts. As well as allowing the user to modify maps to make them clearer, a GIS also allows the inclusion of several key components when producing a map. These include a title, a scale bar, a legend and the north arrow. GISs can also integrate tables, text files, digital photographs and video imagery into digital maps. Although many maps with greater accuracy can be produced, scientists at the University of Mauritius focused on 19 maps that, together, cover the whole coast of Mauritius plus some outlying islands. An example of one of the maps is presented in figure 3.

Description Exposed cliffs Low-lying basalt, beach rock and sea walls Exposed boulder and cobble beaches Sandy beaches Mixed sand and gravel beaches Exposed sand flats Sheltered rocky shores Coral reefs Sheltered tidal flats Marshes and mangroves

C a s e S t u d y : Ts u n a m i Inundation Map Waves of the December 2004 Indonesian tsunami that reached the shores of Mauritius were between 10 and 40 centimetres high. Even though this proved not to be significant, many coastal areas of Mauritius are still vulnerable to tsunami inundation. The Second International Coordination Meeting for the Development of a Tsunami Warning and Mitigation System for the Indian Ocean, held at Grand Baie, Mauritius, in April 2005, recommended a local risk management framework that has as one of its guiding principles the assessment of the risk from tsunamis. One of the elements of the proposed hazard and risk modelling was to develop maps that would detail the maximum run-up of any tsunami waves and their impact. Tsunami coastal run-up, or inland penetration,

114 V OLUME 12: E XAMPLES OF N ATURAL D ISASTER M ITIGATION IN S MALL I SLAND D EVELOPING S TATES

0 0.25 0.5 1

1.5

2 Kilometres

MAP 3

Anse La Raie Sans Souci Butte á l’Herbe Bassin Paquet Pte aux Roches

Legend Sea Birds

Sheltered Rocky Shore

Wading Birds

Sea Wall

Shore Birds

Sand Beach

Shrimp

Mixed Sand/Gravel

Fish Pond

Marsh

Fish

Mangroves

Fishing Reserve

Low Basalt

Nature Reserve

Gambions

Fisheries Post

Beach Rock

Police Station

Exposed Tidal Flats

Hotels

Exposed Cliffs

Reefs

Exposed Boulders

Sheltered Tidal Flat

Built-up Areas

Grand-Gaube

Bassin Bernard

Ile D’ Ambre

Figure 3 Map 3 from the Coastal Sensitivity Atlas of Mauritius for Oil Spill Response.

constitutes the final phase of the tsunami effect, and flooding can extend 300 or more metres inland, covering large areas with water and debris. This effect is particularly dangerous for coastal communities. An inundation map is therefore necessary to identify the extreme flooding zone and elements at risk. Normally, the approach for inundation mapping is to estimate the worst-case scenario based on historical tsunami data. Since no historical tsunami records exist for Mauritius, the only approach is to map the inundation based on the maximum anticipated run-up height (the maximum height above sea level of the waves once they are on shore). Some tsunamis have reached a run-up height of 30 metres.

A series of maps showing areas of Mauritius that are under the threat of being inundated by seawater during a tsunami has been prepared. The concept behind the maps is that the tsunami wave can potentially reach up to 32 metres in elevation inland independent of the horizontal distance that it travels. The energy of a tsunami-generated wave moving inland decreases with its elevation. Therefore, the impact of the crashing wave is inversely proportional to the height of the wave. For example, when considering a 20-metre-high tsunami wave, considerably more damage is caused in the area lying between 0 and 8 metres above sea level than in the area between 16 and 24 metres above sea level. Table 10 compares the elevation above sea level with the degree of risk from tsunamis as

Integrating Disaster-management Information: Mauritius

115

Table 10 Risk category based on height above sea level of a tsunami. Risk category

Height above sea level (m)

Extremely high Very high High Moderate Fair Low No risk

Area at risk (km2)

0-2 2-4 4-8 8 - 16 16 - 24 24 - 32 > 32

27.9 14.0 29.0 51.5 68.3 66.6 Remainder

N

Elevation height and associated risk

W

E S

0 - 2 m (Extreme risk) 0 - 4 m (Very high risk)

Sea Mauritius

4 - 8 m (High risk) 8 - 16 m (Moderate risk) 16 - 24 m (Fair risk)

Sea 24 - 32 m (Low risk)

Grand Baie

No risk

0 0.450.9 1:50,000

1.8

Kilometres

Figure 4 Risk zones for an area in the east of Mauritius plotted on a 1:25,000 scale map. The shaded areas of the inset indicate regions that could be affected by a tsunami.

well as providing an idea of the area of land at risk, while figure 4 shows the risk zones for an area in northern Mauritius. Figure 4 (inset) provides a broad overview of the regions that could be affected by a tsunami. Geographically, the northern part of the island is likely

to be the most affected, whereas the southern part will be the least affected. This is because, in the southern part of Mauritius, the land rises quickly away from the coast while the northern part is the flattest and lowest area. Towns and villages located in each at-risk zone of

116

V OLUME 12: E XAMPLES

OF

N ATURAL D ISASTER M ITIGATION

the island, as highlighted in figure 4, can be inventoried and classified, permitting the strategic deployment of rescue forces in a systematic manner depending on the risk associated with the area.

P AT E N T I N G A N D C O M M E R C I A L I Z AT I O N There are no plans to patent or commercialize these risk maps.

PA R T N E R S H I P S The risk-potentiality maps were developed by a group of researchers from the Faculty of Science at the University of Mauritius together with colleagues from the Ministry of Housing, Land and the Environment.

REPLICABILITY Mauritius benefits from its small size and, in the case of risk-potentiality maps for flooding and landslides, an extensive series of weather stations located throughout the country. Given current remote-sensing and GIS technologies, however, there is no reason why such detailed risk maps cannot be prepared for any small island developing State or particular at-risk regions in other countries.

IN

S MALL I SLAND D EVELOPING S TATES

P O L I C Y I M P L I C AT I O N S The initial maps will eventually be presented to the Government of Mauritius and policy-makers will be lobbied to adopt and use the integrated disastermanagement information system for making decisions.

I M PA C T To date, the risk-potentiality maps described earlier are still at the preliminary stage and are being refined by adding further layers of information. Only once this has been done and the maps have been accepted by national policy-makers will any impact of the project be seen. This is still several years in the future.

LESSONS LEARNED Disasters pose a major threat and a tragic interruption to the development process of a country, especially to small island developing States such as Mauritius. They lead to significant loss of life as well as to the destruction of infrastructure and business activities and the disruption of social networks. The current system does not allow leaders and administrators to make sound disaster-management decisions owing to a lack of cross-sectoral integration of information.

Integrating Disaster-management Information: Mauritius

FUTURE PLANS To deal with and to manage a disaster, one of the most important preparations that can be made is the development of a properly structured information system for disaster management. A remote-sensing and GIS database can be used to create elaborate and effective disaster-management information systems (DMISs), and an integrated approach, using scientific and technological advances, should be adopted to mitigate and to manage natural hazards. Moreover, there should be a national policy for disaster management. Disaster managers and emergency planners need, therefore, to develop and apply appropriate and effective disastermanagement plans and measures. The risk-potentiality maps are still being refined and improved. For example, the flood and landslide maps will be overlaid with other maps, including those concerned with geology, soil, land cover, lineament, river drainage and other drainage systems (for example, in towns and villages). The collection of data for these maps is ongoing. Mauritius does not yet have an integrated DMIS for dealing with disasters. The first maps of the risk of tsunami inundation for Mauritius and the neighbouring island of Rodrigues will shortly be finalized in collaboration with Government officials and will form part of a DMIS although no decision as to where this DMIS will be located has yet been taken. The creation of an integrated DMIS calls for a well-planned and coordinated set of

117

activities at various levels. The results of the use of information and communication technologies and geomatic technologies described above can be envisaged as the first steps towards the creation of an integrated DMIS for Mauritius.

P U B L I C AT I O N S Aworer, L. and Rughooputh, S.D.D.V. (2006). Status of beach erosion in Mauritius. In: 21st Colloquium on African Geology, Maputo, Mozambique, July 2006. Gunlach, E.R. and Murday, M. (1987). Coastal Sensitivity Atlas of Mauritius for Oil Spill Response. Technical report, Ministry of Housing, Land and the Environment, Mauritius. Le Roux, J.J., Sumner, P.D. and Rughooputh, S.D.D.V. (2005). Erosion modelling and soil loss prediction under changing land use for a catchment on Mauritius. South African Geographical Journal, 87:29-140. Nigel, R. and Rughooputh, S.D.D.V. (2006a). Flooding and flooding potentiality maps for Mauritius. In: 21st Colloquium on African Geology, Maputo, Mozambique, July 2006. _____. (2006b). Landslide and landslide potentiality maps for Mauritius. In: 21st Colloquium on African Geology, Maputo, Mozambique, July 2006. Nigel, R., Runghen, H. and Rughooputh, S.D.D.V. (2006). Tsunami preparedness for Mauritius. In: 21st Colloquium on African Geology, Maputo, Mozambique, July 2006. Rughooputh, S.D.D.V. and Peerally, B. N. K. (2003). Tracking of cyclones within

118

V OLUME 12: E XAMPLES

OF

N ATURAL D ISASTER M ITIGATION

the proximity of Mauritius. In: First National Ocean Science Forum, National Global Ocean Observing System Coordination Committee, 18-19 July 2003, Réduit, Mauritius, pp. 40-41. Rughooputh, S.D.D.V., Badal, R., Nigel, R., Runghen, H., Motah, B. and Bissessur, D. (2006). Tsunami preparedness map for Mauritius. Prepared for the Prime Minister’s Office, Government of Mauritius. Runghen, H., Bhuruth, M., Rughooputh, S.D.D.V., Rughooputh, H.C.S. (2003). Oil spill information system for Mauritius: Oil spill shoreline sensitivity mapping and analysis, Industrial Technology, 2003 Institute of Electrical and Electronics Engineers International Conference, vol. 1:450-455. Runghen, H., Bhuruth, M. and Rughooputh, S.D.D.V. (2005). A digital oil spill sensitivity atlas for Mauritius using GIS. GISdevelopment.net, vol. 1, Issue 21 [Online] Available from: www.gisdevelopment.net/application/ environment/conservation/env_con001.htm. [Accessed 29 May 2007]. Shah, B.V. (1983). Is the environment becoming more hazardous? A global survey, 1947 to 1980. Disasters, 7:202-209. Water Resources Unit. (2002). Flood-prone Areas of Mauritius: Study of the Land Drainage System of Mauritius. Ministry of Public Utilities, Mauritius.

IN

S MALL I SLAND D EVELOPING S TATES

Case Study Prepared by: Soonil D.D.V. Rughooputh Address: University of Mauritius, Réduit, Mauritius Tel.: (+230) 454 1041 ext. 1481 Fax: (+230) 454 9642 E-mail: [email protected]

Project Participants: M. Bhuruth, Faculty of Science, University of Mauritius: Mathematical modelling. K. Gooniah, a graduate student in a Master of Science programme, Faculty of Science, University of Mauritius: Tsunami study. F. Nathire, Ministry of Housing, Land and the Environment, Mauritius: Cartographer. R. Nigel, a graduate student in a doctoral programme, Faculty of Science, University of Mauritius: GIS catchment modelling. S.D.D.V. Rughooputh, Faculty of Science, University of Mauritius: Environmental scientist. H. Runghen, a graduate student in a doctoral programme, Faculty of Science, University of Mauritius: GIS oil-spill modelling.