Previous Page Table of Contents Next Page

CHAPTER 12 - HURRICANE HAZARDS

A. HURRICANE: THE PHENOMENON
B. HISTORICAL OCCURRENCE AND IMPACT ON THE AMERICAS: HURRICANE GILBERT
C. RISK ASSESSMENT AND DISASTER MITIGATION
D. COPING WITH HURRICANES IN SMALL TOWNS AND VILLAGES
REFERENCES

SUMMARY

This chapter describes the nature of hurricanes and their destructive capability. It outlines measures that can be taken to reduce the impact of a hurricane and, in particular, identifies appropriate mitigation measures for small towns and villages.

The destruction caused by hurricanes in the Caribbean and Central America is a force that has shaped history and will shape the future of the region. The danger arises from a combination of factors that characterize tropical cyclonic storms: rise in sea level, violent winds, and heavy rainfall. In the Greater Caribbean Basin from 1960 through 1988 (excluding the United States and U.S. territories) hurricanes caused more than 20,000 deaths, affected 6 million people, and destroyed property worth over US$9.5 billion (OFDA, 1989). The great bulk of this harm was done to the Caribbean island countries, whose small economies are least able to withstand such impacts.

Data on hurricane damage have been collected since the discovery of the Americas, and recent statistics show that mitigation measures have made a difference since the 1930s. While the ferocity of the storms has not abated over the years, and population has increased substantially in the area, the casualty rate has decreased as a result of the incorporation of mitigation measures and the increased effectiveness of preparedness activities. This improvement in saving lives has been countered by a marked increase in property damage. This is a clear indicator that structural mitigation measures are not keeping pace with the rapid increase in development in vulnerable areas.

A important feature of this chapter is its detailed discussion of hurricane hazard mitigation in small towns and villages. In this setting, largely beyond the reach of national mitigation activities, simple strategies are both essential and highly effective.

A. HURRICANE: THE PHENOMENON

1. HURRICANE DEVELOPMENT
2. TEMPORAL DISTRIBUTION OF HURRICANE OCCURRENCE IN THE CARIBBEAN
3. HAZARDOUS CHARACTERISTICS OF HURRICANES

"Tropical cyclone" is the scientific term for a closed meteorological circulation that develops over tropical waters. These large-scale non-frontal low-pressure systems occur throughout the world over zones referred to as "tropical cyclone basins" (NOAA, 1987). The name for them varies: in the Atlantic and northeast Pacific they are called "hurricanes" after the Mayan word for devil, in the northwest Pacific "typhoons," and in the South Pacific and Indian Ocean simply "cyclones." Of all tropical cyclone occurrences, 75 percent develop in the northern hemisphere, and of these, only one out of three are hurricanes in the northeast Pacific or northwest Atlantic (UNDRO, 1978). The storms of the northern hemisphere travel westward; those of the southern hemisphere move eastward.

In the Atlantic tropical cyclone basin, which includes the Atlantic Ocean, the Caribbean Sea, and the Gulf of Mexico, hurricanes originate mostly in the northern Atlantic and to a lesser degree in the Caribbean. The areas most at risk are the Caribbean island countries north of Trinidad (73 strikes by major hurricanes between 1900 and 1988), Mexico and the southeastern United States, Central America north of Panama, and to a limited extent the northern coast of South America (Tomblin, 1979). Hurricanes also originate in the northeast Pacific, where they can affect the west coast of Mexico. Most of South America is essentially at no risk, because the tropical southwestern Atlantic and the southeastern Pacific are devoid of these meteorological occurrences, but systems originating on the west coast of Africa can potentially strike the northernmost part of the continent; for example, in 1988 Hurricane Joan formed on the northwestern coast of Africa and struck the coast of Venezuela and Colombia before hitting eastern Nicaragua. Figure 12-1 shows the paths of the hurricanes originating in the Atlantic, the Pacific, and the Caribbean.

1. HURRICANE DEVELOPMENT

a. Birth: Tropical Depression
b. Growth: Tropical Storm and Hurricane
c. Death: Landfall or Dissipation

All of the embryonic tropical depressions that develop into hurricanes originate in similar meteorological conditions and exhibit the same life cycle. The distinct stages of hurricane development are defined by the "sustained velocity" of the system's winds-the wind velocity readings maintained for at least one minute near the center of the System. In the formative stages of a hurricane, the closed isobaric circulation is called a tropical depression. If the sustained velocity of the winds exceeds 63km/h (39 mph), it becomes a tropical storm. At this stage it is given a name and is considered a threat. When the winds exceed 119km/h (74 mph), the system becomes a hurricane, the most severe form of tropical storm. Decay occurs when the storm moves into nontropical waters or strikes a landmass. If it travels into a nontropical environment it is called a subtropical storm and subtropical depression; if landfall occurs. the winds decelerate and it becomes again a tropical storm and depression. Figure 12-2 summarizes this classification.

Figure 12-1 - OCCURRENCE OF TROPICAL STORMS AND CYCLONES IN THE WESTERN HEMISPHERE 1/

1/ Wind strength of Beaufort 8 and above

Source: Munchener Ruck. Mapa Mundial de los Riesgos de la Naturaleza. (Munich, Federal Republic of Germany, Munchener Ruckversicherungs: 1988)

a. Birth: Tropical Depression

Hurricanes are generated at latitudes of 8 to 15 degrees north and south of the Equator as a result of the normal release of heat and moisture on the surface of tropical oceans. They help maintain the atmospheric heat and moisture balance between tropical and non-tropical areas. If they did not exist, the equatorial oceans would accumulate heat continuously (Landsberg, 1960).

Hurricane formation requires a sea surface temperature of at least 27 degrees Celsius (81 degrees Fahrenheit). In the summer months, the sea temperatures in the Caribbean and Atlantic can reach 29 degrees (84 degrees), making them prime locations for inception. The surface water warms the air, which rises and then is blocked by warmer air coming from the easterly winds. The meeting of these two air masses creates an atmospheric inversion. At this stage, thunderstorms develop and the inversion may be broken, effectively lowering the atmospheric pressure.

b. Growth: Tropical Storm and Hurricane

The growth of the system occurs when pressure in the center of the storm drops well below 1000 millibars (mb) while the outer boundary pressure remains normal. When pressure drops, the trade winds are propelled in a spiral pattern by the earth's rotation. The strong torque forces created by the discrepancy in pressure generate wind velocities proportional to gradient of pressure. As the energy level increases, the air circulation pattern is inward towards the low pressure center and upward, in a counter-clockwise spiral in the northern hemisphere and clockwise in the southern hemisphere. The cycle perpetuates itself and the organized storm begins a translational movement with velocities of around 32 km/h during formation and up to 90km/h during the extra-tropical life.

The zone of highest precipitation, most violent winds, and rising sea level is adjacent to the outer wall of the "eye." The direction of the winds, however, is not towards the eye but is tangent to the eye wall about 50km from the geometric center (Mathur, 1987). The organized walls of clouds are composed of adjoining bands which can typically reach a total diameter of 450km (Earthscan No. 34-a, 1983). The central eye, unlike the rest of the storm, is characterized as an area of relatively low wind speeds and no cloud cover with an average diameter of 50-80km and a vertical circulation of up to 15km.

Hurricane classification is based on the intensity of the storm, which reflects damage potential. The most commonly used categorization method is the one developed by H. Saffir and R.H. Simpson (Figure 12-3). The determination of a category level depends mostly on barometric pressure and sustained wind velocities. Levels of storm surge fluctuate greatly due to atmospheric and bathymetric conditions. Thus, the expected storm surge levels are general estimates of a typical hurricane occurrence.

Figure 12-2 CLASSIFICATION OF HURRICANE DEVELOPMENT

ENVIRONMENT

DEVELOPMENT

CRITERIA

Tropical

Depression

max sustained winds < or = 63 km/h (39 miles/h)

Tropical Storm

63 km/h < sustained winds < 119 km/h (74 miles/h)

Hurricane

sustained winds > or = 119km/h (74 miles/h)

Tropical Depression (dissipation)

max sustained winds < or = 63km/h (39 miles/h)

Nontropical

Subtropical Storm (dissipation)

63km/h < sustained winds < 119km/h (74 miles/h)

Subtropical Depression (dissipation)

max sustained winds < or = 63km/h (39 miles/h)

Source: Adapted from Neumann, C.J. et al Tropical Cyclones of the North Atlantic Ocean, 1871-1986 (Washington, D.C.: U.S. Department of Commerce, NOAA, 1987).

Figure 12-3 SAFFIR-SIMPSON HURRICANE SCALE (SSH)

Hurricane Category Number

Sustained Winds

Atmospheric Pressure in the Eye (millibars)

Storm

Surge

Damage

(km/h)

(miles/h)

(meters)

(feet)

Level

1

119 - 153

74 - 95

980

1.2 - 1.5

4.0 - 4.9

Low

2

154 - 177

96 - 110

965 - 979

1.8 - 2.4

5.9 - 7.9

Moderate

3

179 - 209

111 - 130

945 - 964

2.7 - 3.7

8.9 - 12.2

Extensive

4

211 - 249

131 - 155

920 - 944

4.0 - 5.5

13.0 - 18.0

Extreme

5

> 249

> 155

< 920

> 5.5

> 18.0

Catastrophic

Source: Adapted from Oliver, J., and Fairbridge, R. The Encyclopedia of Climatology (New York: Van Nostrand Reinhold Co., Inc., 1987).

c. Death: Landfall or Dissipation

Typically, a hurricane eventually dissipates over colder waters or land about ten days after the genesis of the system. If it travels into a non-tropical environment, it loses its energy source and falls into the dominant weather pattern it encounters. If, on the other hand, it hits land, the loss of energy in combination with the increased roughness of the terrain will cause it to dissipate rapidly (Frank, 1984). When it reaches land in populated areas, it becomes one of the most devastating of all natural phenomena.

2. TEMPORAL DISTRIBUTION OF HURRICANE OCCURRENCE IN THE CARIBBEAN

The official hurricane season in the Greater Caribbean region begins the first of June and lasts through November 30, with 84 percent of all hurricanes occurring during August and September (Frank, 1984). Figure 12-4 shows the seasonal character of hurricanes. The greatest risk in Mexico and the western Caribbean is at the beginning and end of the season, and in the eastern Caribbean during mid-season.

Every year over 100 tropical depressions or potential hurricanes are monitored, but an average of only ten reach tropical storm strength and six become hurricanes. These overall averages suggest that activity is uniform from year to year but historical records indicate a high degree of variance, with long periods of tranquillity and activity (Figure 12-5). The Atlantic basin has the widest seasonal variability. In 1907, for example, not a single tropical storm reached hurricane intensity, while in 1969, there were 12 hurricanes in the northern Atlantic (NOAA, 1987).

Because the cycles vary in periodicity and duration, prediction is difficult. Recent forecasting developments, connecting hurricane activity levels with El Niño and the Quasi-biennial Oscillation have made it possible to predict the variance in Atlantic seasonal hurricane activity with an accuracy of 40 to 50 percent (American Meteorological Society, 1988), but this degree of accuracy, while considered high by meteorological standards, is not good enough for planners trying to develop appropriate emergency response systems. There is no doubt that the quality of forecasting will continue to improve, but until that happens planners must rely on historical information to calculate the probability of occurrence in a given year. Simpson and Lawrence in 1971 used historical data to make these calculations for the entire east coast of the United States and Gulf of Mexico coast, using 80km (50 miles) segments (ESCAP/WMO, 1977).

3. HAZARDOUS CHARACTERISTICS OF HURRICANES

a. Winds
b. Rainfall
c. Storm Surge

a. Winds

Hurricane wind speeds can reach up to 250km/h (155mph) in the wall of the hurricane, and gusts can exceed 360km/h (224mph).The destructive power of wind increases with the square of its speed. Thus, a tripling of wind speed increases destructive power by a factor of nine. Topography plays an important role: wind speed is decreased at low elevations by physical obstacles and in sheltered areas, while it is increased over exposed hill crests (Davenport, 1985; see Figure 12-6). Another contributor to destruction is the upward vertical force that accompanies hurricanes; the higher the vertical extension of a hurricane, the greater the vertical pulling effect.

Destruction is caused either by the direct impact of the wind or by flying debris. The wind itself primarily damages agricultural crops. Entire forests have been flattened by forces that pulled the tree roots from the earth. Man-made fixed structures are also vulnerable. Tall buildings can shake or even collapse. The drastic barometric pressure differences in a hurricane can make well-enclosed structures explode and the suction can lift up roofs and entire buildings. But most of the destruction, death, and injury by wind is due to flying debris (ECLAC/UNEP, 1979), the impact force of which is directly related to its mass and the square of its velocity. The damage caused by a flying car to whatever it strikes will be greater than if the wind had acted alone. Improperly fastened roof sheets or tiles are the most common projectiles. Other frequent objects are antennas, telephone poles, trees, and detached building parts.

Figure 12-4 - NUMBER OF TROPICAL STORMS AND HURRICANES (open bar) AND HURRICANES (solid bar) OBSERVED ON EACH DAY, MAY 1-DECEMBER 31, 1886 THROUGH 1986, IN THE NORTH ATLANTIC OCEAN

Source: Neumann, C.J. et al. Tropical Cyclones of the North Atlantic Ocean, 1871-1986 (Washington, D.C.: U.S. Department of Commerce, NOM, 1987).

Figure 12-5 - ANNUAL DISTRIBUTION OF THE 845 RECORDED TROPICAL CYCLONES IN THE NORTH ATLANTIC REACHING AT LEAST TROPICAL STORM STRENGTH (open bar) AND THE 496 REACHING HURRICANE STRENGTH (solid bar), 1886 THROUGH 1986

Note: The average number of such storms is 8.4 and 4.9 respectively.

Source: Neumann, C.J. et al Tropical Cyclones of the North Atlantic Ocean, 1871-1986 (Washington, D.C.: U.S. Department of Commerce, NOAA, 1987).

FIGURE 12-6 ISLAND TOPOGRAPHIC EFFECTS ON MEAN SURFACE WIND SPEEDS

Source: Davenport, A.G. Georgiou, P.M., and Surry, D. A Hurricane Wind Risk Study for the Eastern Caribbean, Jamaica and Belize with Special Consideration to the Influence of Topography. (London, Ontario, Canada: Boundary Layer Wind Tunnel Laboratory, The University of Western Ontario, 1985).

Building standards to withstand high wind velocities are prescribed in almost all countries that face a high risk. The codes recommend that structures maintain a certain level of strength in order to withstand the local average wind velocity pressure, calculated by averaging wind pressure over a period of ten minutes for the highest expected wind speed in 50 years. The Caribbean Uniform Building Code (CUBIC) under consideration by the Caribbean countries, prescribes the reference wind velocity pressure for each country. Figure 12-7 shows the relationship between wind speed, expressed in the codes in terms of meters per second rather than kilometers or miles per hour, and general property damage. Note the correlation between this and the SSH scale in Figure 12-3.

b. Rainfall

The rains that accompany hurricanes are extremely variable and hard to predict (ECLAC/UNEP, 1979). They can be heavy and last several days or can dissipate in hours. The local topography, humidity, and the forward speed of a hurricane in the incidence of precipitation are recognized as important, but attempts to determine the direct connection have so far proved futile.

Intense rainfall causes two types of destruction. The first is from seepage of water into buildings causing structural damage; if the rain is steady and persistent, structures may simply collapse from the weight of the absorbed water. The second, more widespread and common and much more damaging, is from inland flooding, which puts at risk all valleys along with their structures and critical transportation facilities, such as roads and bridges. Chapter 8 describes flooding in more detail.

Landslides, as secondary hazards, are often triggered by heavy precipitation. Areas with medium to steep slopes become oversaturated and failure occurs along the weakest zones. Thus, low-lying valley areas are not the only sites vulnerable to precipitation. Chapter 10 is devoted to this phenomenon.

c. Storm Surge

A storm surge is a temporary rise in sea level caused by the water being driven over land primarily by the on-shore hurricane force winds and only secondarily by the reduction in sea-level barometric pressure between the eye of the storm and the outer region. A rough relationship between atmospheric pressure and the storm surge level was shown in Figure 12-3. Another estimate is that for every drop of 100 millibars (mb) in barometric pressure, a 1m (3.28 feet) rise in water level is expected. The magnitude of the surge at a specific site is also a function of the radius of the maximum hurricane winds, the speed of the system's approach, and the foreshore bathymetry. It is here that the difficulty arises in predicting storm surge levels. Historical records indicate that the increase in mean sea level can be negligible or can be as much as 7.5 meters (24.6 feet) (ECLAC/UNEP, 1979). The most vulnerable coastal zones are those with the highest historical frequencies of landfalls. Regardless of its height, the great dome of water is often 150km (93 miles) wide and moves toward the coastline where the hurricane eye makes landfall.

Figure 12-7 RELATIONSHIP BETWEEN WIND SPEED AND GENERAL PROPERTY DAMAGE

Wind Speed

Damage

22-35m/sec

minor

36-45 m/sec

intermediate (loss of windows)

>45m/sec

structural

Source: ECLAC/UNEP. Natural Disasters Overview. Meeting of Government-Nominated Experts to Review the Draft Action Plan for the Wider Caribbean Region, Caracas, Venezuela, 28 January - 1 February (Caracas: ECLAC/UNEP, 1979).

Storm surges present the greatest threat to coastal communities. Ninety percent of hurricane fatalities are due to drowning caused by a storm surge. Severe flooding from a storm surge affects low-lying areas up to several kilometers inland. The height of the surge can be greater if man-made structures in bays and estuaries constrict water flow and compound the flooding. If heavy rain accompanies storm surge and the hurricane landfall occurs at a peak high tide, the consequences can be catastrophic. The excess water from the heavy rains inland creates fluvial flooding, and the simultaneous increase in sea level blocks the seaward flow of rivers, leaving nowhere for the water to go.

B. HISTORICAL OCCURRENCE AND IMPACT ON THE AMERICAS: HURRICANE GILBERT

1. JAMAICA
2. MEXICO

Hurricanes are by far the most frequent hazardous phenomena in the Caribbean. Tomblin (1981) states that in the last 250 years the West Indies has been devastated by 3 volcanic eruptions, 8 earthquakes, and 21 major hurricanes. If tropical storms are also taken into account, the Greater Caribbean area has suffered from hundreds of such events.

The economic and social consequences of this phenomenon are severe, especially in less developed countries, where a significant percentage of the GDP can be destroyed by a single event. Figure 12-8 lists the major hurricanes and tropical storms in the Americas and the damage they caused.

Without a comprehensive list of costs and casualties, the economic and social disruption caused by a disastrous event is hard to grasp. It is not the purpose of this chapter to provide all this information, which can be found in the great volume of literature on individual events. But a brief review of how one hurricane affected various sectors in Mexico and Jamaica will help planners to understand the complexities of the turmoil that such a natural event can cause.

Hurricane Gilbert struck the Caribbean and the Gulf Coast of Mexico in 1988, causing comprehensive damage in Mexico, Jamaica, Haiti, Guatemala, Honduras, Dominican Republic, Venezuela, Costa Rica, and Nicaragua. Arriving in Saint Lucia as a tropical depression, it resulted in damage estimated at US$2.5 million from the flooding and landslides caused by the heavy rain (Caribbean Disaster News No.15/16,1988).

The physical variations in this hurricane resulted in different types of damage. It was considered a "dry" hurricane when it struck Jamaica, discharging less precipitation than would be expected. Thus, most of the damage was due to wind force which blew away roofs. By the time it approached Mexico, however, it was accompanied by torrential rains, which caused massive flooding far inland.

Hurricane Gilbert began as a tropical wave on September 3, 1988, on the north coast of Africa. Six days later, the system was across the Atlantic and had evolved into Gilbert as a tropical storm. It struck Jamaica on September 12 as a Category 3 (SSH Scale) hurricane and traveled westward over the entire length of the island. Gaining strength as it moved northwest, it hit the Yucatan Peninsula, in Mexico, on September 14, as a Category 5 (SSH Scale) hurricane. By September 16 it had been weakened and finally dissipated after moving inland over the east coast of Mexico.

Sustained winds in Jamaica were measured at 223 km/h, and greater across high ridges. The barometric pressure was the lowest ever recorded in the Western Hemisphere at 888mb, 200km east-southeast of Jamaica. The barometric pressure when it hit Jamaica was 960mb. The forward speed was 31 km/h. The eye had a 56km diameter, but little storm surge occurred in Jamaica. Average rainfall registered from 250mm to 550mm. Serious flooding due to storm surge and heavy rains was not a problem. Landslides occurred at high elevations where most of the rainfall was concentrated.

By the time Hurricane Gilbert hit Mexico it had changed characteristics. In the Yucatan the storm surge reached 5 meters in height and rainfall averaged 400mm. By the time Gilbert struck the northern coast of Mexico, the winds had increased to 290km/h and the storm surge had reached 6 meters.

1. JAMAICA

a. Affected Population and Damage to Social Sectors
b. Impact on the Economy and Damage to Productive Sectors
c. Damage to Natural Resources

a. Affected Population and Damage to Social Sectors

Even though the loss of life was limited to 45 reported deaths, 500,000 people lost their homes when approximately 280,000 houses-almost 55 percent of the housing stock-were damaged. Of these, 14,000, or 5 percent, were totally destroyed and 64,000 were seriously damaged.

b. Impact on the Economy and Damage to Productive Sectors

The Planning Institute of Jamaica estimated the total direct damage at US$956 million. Nearly half was accounted for by losses from agriculture, tourism, and industry; 30 percent from housing, health, and education infrastructure; and 20 percent from economic infrastructure. The economic projections for 1988 had to be adjusted dramatically, to allow for expected losses of US$130 million in export earnings, and more than US$100 million in tourism earnings; therefore, instead of the expected 5 percent growth in GDP, a decline of 2 percent was projected. Other estimates were for increases in inflation (30 percent), government public expenditures (US$200 million), and public sector deficit (from 2.8 percent to 10.6 percent of GDP).

Figure 12-8 MAJOR TROPICAL STORMS AND HURRICANES OF THE ATLANTIC TROPICAL CYCLONE BASIN

REGION/COUNTRY

YEAR/MONTH

CASUALTIES

PEOPLE AFFECTED

DAMAGE THOUSANDS/US$

HURRICANE NAME

SOURCE

CARIBBEAN

Antigua




1792 00





Tomblin

1950 09

2


1,000

Dog

OFDA

1960 09

2



Donna

OFDA

1966 09





OFDA

Barbados




1780 00

4,326




Tomblin

1786 00





Tomblin

1831 00

2,000




Tomblin

1955 09

57



Janet

OFDA

Belize





1931 09

1,500


7,500


OFDA

1955 09

16


5,000

Janet

OFDA

1961 09

275


60,000


OFDA

1974 09


70,000

4,000

Carmen, Fifi

OFDA

1978 09

5

6,000

6,000

Greta

OFDA

Cuba













1768 00

1,000




Tomblin

1844 00





Tomblin

1846 00

500




Tomblin

1926 10

600




OFDA

1932 11

2,500




OFDA

1935 09

35

500



OFDA

1948 09

3


12,000


OFDA

1948 10

11

300

6,000


OFDA

1963 10

1,750




Tomblin

1966 09

5

156,000

18,000

Inez

OFDA

1968 10

0



Gladys

OFDA

1982 06

24

105,000

85,000


OFDA

1985 11

4

476,891


Kate

OFDA

Dominica





1806 00





Tomblin

1834 00

200




Tomblin

1963 09



2,600

Edith

OFDA

1979 08

40

70,000

44,650

David, Frederick

OFDA

1984 10

2

10,000

2,000

Klaus

OFDA

Dominican Republic






1930 09

2.000

6,000

40,000


OFDA

1963 10

400


60,000

Flora

OFDA

1964 08

7


1,000

Cleo

OFDA

1966 09

74

7,000

5,000

Inez

OFDA

1979 08

1,400

1,200,000

150.000

David, Frederick

OFDA

1987 09

3


23,700

Emily

OFDA

Grenada

1963 09

6



Flora

OFDA

Haiti










1909 11

150




OFDA

1915 08

1,600




OFDA

1935 10

2,150




OFDA

1954 10

410

250,000


Hazel

OFDA

1963 10

5,000


180,000

Flora

OFDA

1964 08

100

80.000

10,000

Cleo

OFDA

1966 09

480

67,000

20,000

Inez

OFDA

1979 08

8

1,110


David

OFDA

1980 08

300

330,000

40,000

Alien

OFDA

1988 09

54

870,000

91,286

Gilbert

OFDA

Jamaica















1722 00

400




Tomblin

1780 00

300




Tomblin

1786 00





Tomblin

1880 00

30




Tomblin

1903 08

65




OFDA

1912 11

142




OFDA

1917 09

57




OFDA

1933 10

10




OFDA

1935 10


2,000



OFDA

1944 08

32




OFDA

1951 08

154

20,000

56.000

Charlie

OFDA

1963 10

11


11,525

Flora

OFDA

1980 08

6

30,000

64,000

Alien

OFDA

1985 11

7


5,200

Kate

OFDA

1988 09

49

810,000

1,000.000

Gilbert

OFDA

St. Kitts/Nevis




1772 00





Tomblin

1792 00





Tomblin

1928 09





OFDA

1955 01





OFDA

Saint Lucia



1960 07




Abby

OFDA

1963 09

10


3,465

Edith

OFDA

1980 08

9

70,000

87,990

Alien

OFDA

St. Vincent




1898 00

300




Tomblin

1955 09

122



Janet

OFDA

1980 08


20,000

16,300

Alien

OFDA

1987 09


200

5,300

Emily

OFDA

Trinidad/Tobago


1933 06

13


3,000


OFDA

1963 09

24


30,000

Flora

OFDA

CENTRAL AMERICA

Costa Rica

1988 10

28

120,000


Joan

OFDA

El Salvador

1969 09

2

4,600

1,600

Francelia

OFDA

Guatemala

1969 09

269

10,200

15,000

Francelia

OFDA

Honduras



1969 09


8,000

19,000

Francelia

OFDA

1974 09

8,000

600,000

540,000

Fifi

OFDA

1978 09


2,000

1,000

Greta

OFDA

Nicaragua


1971 09

35

2,800

380

Edith

OFDA

1988 10

120

300,000

400,000

Joan

OFDA

Panama

1988 10

7

7,000

60,000

Joan

OFDA

NORTH AMERICA (EXCLUDING THE UNITED STATES)

Mexico













1951 08

50




OFDA

1955 09

300



Hilda

OFDA

1955 09

500


40,000

Janet

OFDA

1960 10

960




OFDA

1961 11

436



Tara

OFDA

1966 10

14

80,000

24,000

Inez

OFDA

1967 08

77

271,000

184,000

Katrina, Beulah

OFDA

1975 10

29



Olivia

OFDA

1976 10

600

175,000

100,000

Liza

OFDA

1977 09

10

50,000


Anita

OFDA

1982 09

225

50,000

30,000

Paul

OFDA

1983 10

135



Tico

OFDA

1988 09

240

100,000


Gilbert

OFDA

Sources: Tomblin, J. "Natural Disasters in the Caribbean: A Review of Hazards and Vulnerability," in Caribbean Disaster Preparedness Seminar, St. Lucia, June, 1979 (Washington, D.C.: OFDA/USAID, 1979); and Office of Foreign Disaster Assistance, U.S. Agency for International Development (OFDA/USAID). Disaster History: Significant Data on Major Disasters Worldwide, 1900-Present. July, 1989. (Washington, DC.- OFDA/USAID. 1089).

As expected, the economic activity most affected was agriculture, with the total destruction of banana and broiler production and of more than 50 percent of the coffee and coconut crops. Capital losses to the sector were estimated at J$0.7 billion. According to some calculations, the loss of revenue through 1992 will be US$214 million.

Other productive sectors were also affected seriously. Manufacturing suffered J$600 million (in 1989 dollars) in losses, mainly from a decline of 12 percent in its exports. Tourism lost US$90 million in foreign exchange, with 5 percent fewer visitor arrivals in the third quarter of 1988 than during the same time period in 1987. Loss of electricity decreased bauxite production by 14.2 percent for that quarter compared to the third quarter of the previous year, and alumina exports declined by 21 percent.

c. Damage to Natural Resources

The coastal resources of Jamaica suffered extensive damage from hurricane forces. It is estimated that 50 percent of the beaches were seriously eroded, with the northeast coast being the most affected. An estimated 60 percent of all the trees in the mangrove areas were lost, 50 percent of the oyster culture was unsalvageable, and other non-measurable harm occurred to coral reefs and the water quality of the island (Bacon, 1989).

2. MEXICO

a. Affected Population and Damage to Social Sectors
b. Impact on the Economy and Damage to Productive Sectors
c. Damage to Natural Resources

a. Affected Population and Damage to Social Sectors

The Government of Mexico reported that the hurricane caused 200 deaths and approximately 200,000 homeless. In the state of Nuevo Leon, where the Monterrey area suffered from extensive flooding, 100 people died and 30,000 housing units were destroyed.

b. Impact on the Economy and Damage to Productive Sectors

The tourism industry suffered the greatest damage.

The tourist areas of the state of Quintana Roo, for example, suffered US$100 million in direct damage and lost an estimated US$90 million in revenues. The Inter-American Development Bank, after evaluating the damage to infrastructure in this sector, lent US$41.5 million for reconstruction.

c. Damage to Natural Resources

The impact across the Yucatan Peninsula in terms of damage to wildlife, beaches, and coral reefs was much higher than on the coasts of Jamaica. Extensive reduction in beaches and coral reefs was reported, and large numbers of birds lost their lives.

C. RISK ASSESSMENT AND DISASTER MITIGATION

1. DETERMINING THE RISK POSED BY HURRICANES
2. MITIGATING AGAINST HURRICANE RISK

1. DETERMINING THE RISK POSED BY HURRICANES

The risk posed by hurricanes to a particular country is a function of the likelihood that a hurricane of a certain intensity will strike it and of the vulnerability of the country to the impact of such a hurricane. Vulnerability is a complex concept, which has physical, social, economic and political dimensions. It includes such things as the ability of structures to withstand the forces of a hazardous event, the extent to which a community possesses the means to organize itself to prepare for and deal with emergencies, the extent to which a country's economy depends on a single product or service that is easily affected by the disaster, and the degree of centralization of public decision-making (Wilches-Chaux, 1989).

Population centers and economic activities in the region are highly vulnerable to disruption and damage from the effects of extreme weather. They are largely concentrated in coastal plains and low-lying areas subject to storm surges and landborne flooding. High demands placed on existing lifeline infrastructure, combined with inadequate funds for the expansion and maintenance of these vital systems, have increased their susceptibility to breakdowns. Uncontrolled growth in urban centers degrades the physical environment and its natural protective capabilities. Building sites safe from natural hazards, pollution, and accidents have become inaccessible to the urban poor, who are left to build their shelters on steep hillsides or in flood-prone areas (Bender, 1989). Agriculture, particularly the cultivation of bananas for export, is often practiced without the necessary conservation measures corresponding to the soil, slope, and rainfall characteristics of the area.

Communities, countries, or regions differ greatly in vulnerability, and hence in the effects they may suffer from hurricanes of similar strength. The very size of a country is a critical determinant of vulnerability: small island nations can be affected over their entire area, and major infrastructure and economic activities may be crippled by a single event. Scarce resources that were earmarked for development projects have to be diverted to relief and reconstruction, setting back economic growth.

To assess future risks, planners must study historical trends and correlate them with probable future changes. The main cause of increasing vulnerability is the population movement to high-risk areas. Most cities in the West Indies are in low coastal zones threatened by storm surge (Tomblin, 1979), and they continue to grow.

The economic sectors most affected by hurricanes are agriculture and tourism. Together, these represent a major portion of the economy for the countries in the Caribbean. Particularly for island countries, agriculture is the most vulnerable activity (ECLAC/UNEP, 1979). Hurricanes have disastrous effects on banana crops in particular. During Hurricane Alien, in August of 1980, Saint Lucia suffered US$36.5 million in damage, with 97 percent of the banana plantations destroyed. In St. Vincent 95 percent, and in Dominica 75 percent, of the banana plantations were ruined (Earthscan No. 34a, 1983). Damage to the tourism industry is more difficult to quantify since it includes many other economically identifiable sectors such as transportation and hotel services.

Crop statistics rarely account for long-term losses. The increased salinity in the soil due to the storm surge can have detrimental effects on production in subsequent years. For example, Hurricane Fifi decreased production in Honduras by 20 percent the year it occurred, but in the following year production was down by 50 percent. How much of this reduction was due to the increase in salinity is unclear, but it is known that salt destroys vegetation slowly.

2. MITIGATING AGAINST HURRICANE RISK

a. Reduction of Risk at the International Level
b. Reduction of Risk at the National Level
c. Reduction of Risk at the Local Level

Once the risk posed by hurricanes is understood, specific mitigation measures can be taken to reduce the risk to communities, infrastructure, and economic activities. Human and economic losses can be greatly reduced through well-organized efforts to implement appropriate preventive measures, in public awareness and in issuing timely warnings. Thanks to these measures, countries in the region have experienced a drastic reduction in the number of deaths caused by hurricanes.

Mitigation measures are most cost-effective when implemented as part of the original plan or construction of vulnerable structures. Typical examples are the application of building standards designed for hurricane-force winds, the avoidance of areas that can be affected by storm surge or flooding, and the planting of windbreaks to protect wind-sensitive crops. Retrofitting buildings or other projects to make them hurricane-resistant is more costly and sometimes impossible. Once a project is located in a flood-prone area, it may not be feasible to move it to safer ground.

The overall record on mitigation of hurricane risk in the Caribbean and Central America is not very encouraging. Cases abound of new investments in the public or productive sectors that were exposed to significant hazard risk because of inappropriate design or location, and even of projects that were rebuilt in the same way on the same site after having been destroyed a first time. Other cases can be cited of schools and hospitals funded with bilateral aid that were built to design standards suitable for the donor country but incapable of resisting hurricane-strength winds prevalent in the recipient country.

The tourism sector in the Caribbean is notorious for its apparent disregard of the risk of hurricanes and associated hazards. A hotel complex built with insufficient setback from the high-water mark not only risks being damaged by wave action and storm surge, but interferes with the normal processes of beach formation and dune stabilization, thus reducing the effectiveness of a natural system of protection against wave action. After the first serious damage is incurred the owners of the hotel will most likely decide to rebuild on the same site and invest in a seawall, rather than consider moving the structure to a recommended setback.

a. Reduction of Risk at the International Level

In the past three decades the technological capacity to monitor hurricanes has improved dramatically, and along with it the casualty rate has declined. New technology permits the identification of a tropical depression and on-time monitoring as the hurricane develops. The greatest advance has occurred in the United States, but developing countries benefit greatly because of the effective warning mechanism. The computer models also generate vast quantities of information useful for planners in developing nations.

Computer models that estimate tracking, landfall, and potential damage were first implemented in 1968 by the U.S. National Hurricane Center (NHC). At this point there are five operational track guidance models: Beta and Advection Model (BAM), Climatology and Persistance (CLIPER), a Statistical-dynamical model (NHC90), Quasi-Lagrangian model (QLM) and the barotropic VICBAR. They vary in capacity and methodology and occasionally result in conflicting predictions, though fewer than formerly. The NHC evaluates incoming data on all tropical storms and hurricanes in the Atlantic and eastern Pacific tropical cyclone basin and issues an official track and intensity forecast consisting of center positions and maximum one-minute wind speeds for 0, 12, 24, 48, and 72 hours.

The NHC has also developed a hurricane surge model named Sea, Lake and Overland Surges (SLOSH) to simulate the effects of hurricanes as they approach land. Its predecessor SPLASH, used in the 1960s, was useful for modeling hurricane effects along smooth coastlines, but SLOSH adds to this a capability to gauge flooding in inland areas. These results can be used in planning evacuation routes.

A computerized model that assesses the long-term vulnerability of coastal areas to tropical cyclones has also been developed. This model, the National Hurricane Center Risk Analysis Program (HURISK), uses historical information on 852 hurricanes since 1886. The file contains storm positions, maximum sustained winds, and central pressures (unavailable for early years) at six-hour intervals. When the user provides a location and the radius of interest, the model determines hurricane occurrences, dates, storm headings, maximum winds, and forward speeds. Vulnerability studies begin when the median occurrence date, direction distribution, distribution of maximum winds, probability of at least x number of hurricanes passing over n consecutive years, and gamma distribution of speeds are determined. Planners can use these objective return period calculations to evaluate an otherwise subjective situation.

b. Reduction of Risk at the National Level

One of the most important steps a country can take to mitigate the impact of hurricanes is to incorporate risk assessment and mitigation measure design into development planning. The design of basic mitigation measures begins with the compilation of all historical records of former hurricane activity in the country, to determine the frequency and severity of past occurrences. Reliable meteorological data for each event, ranging from technical studies to newspaper reports, should be gathered. With all the information in place, a study of (1) the distribution of occurrences for months of a year, (2) frequencies of wind strengths and direction, (3) frequencies of storm surges of various heights along different coastal sections, and (4) frequencies of river flooding and their spatial distribution should be undertaken. The statistical analysis should provide quantitative support for planning strategies.

The design of mitigation measures should follow the statistical analysis and consider long-term effects. Both non-structural and structural mitigation measures should be considered, taking into account the difficulties of implementation.

Non-structural measures consist of policies and development practices that are designed to avoid risk, such as land use guidelines, forecasting and warning, and public awareness and education. Much credit for the reduction of casualties from hurricanes in the Caribbean should be given to the Pan Caribbean Disaster Preparedness and Prevention Project (PCDPPP), which has worked effectively with national governments on motivating the population to take preventive measures, such as strengthening roof tie-downs, and on establishing forecasting and warning measures.

Structural mitigation measures include the development of building codes to control building design, methods, and materials. The construction of breakwaters, diversion channels, and storm surge gates and the establishment of tree lines are a few examples of mitigation from a public works standpoint.

c. Reduction of Risk at the Local Level

The effectiveness of national emergency preparedness offices of countries in the region is often seriously limited because of inadequate institutional support and a shortage of technical and financial resources. In the smaller Caribbean islands, these offices are mostly one-person operations, with the person in charge responsible for many other non-emergency matters. It would be unrealistic to expect them to be able to act effectively at the local level in cases of area-wide emergencies, such as those caused by hurricanes. It is therefore essential to enhance the capacity of the population in small towns and villages to prepare for and respond to emergencies by their own means.

From 1986 through 1989, the OAS/Natural Hazards Project has worked with several Eastern Caribbean countries to evaluate the vulnerability of small towns and villages to natural hazards, and train local disaster managers and community leaders in organizing risk assessments and mitigation in their communities. These activities have resulted in the preparation of a training manual with accompanying video for use by local leaders. This effort has focused on lifeline networks-transportation, communications, water, electricity, sanitation-and critical facilities related to the welfare of the inhabitants, such as hospitals and health centers, schools, police and fire stations, community facilities, and emergency shelters.

The remainder of this chapter is dedicated to a summary overview of the process by which the leadership in a small town or village can introduce effective hazard mitigation.

D. COPING WITH HURRICANES IN SMALL TOWNS AND VILLAGES

1. Inventory of Lifeline Networks and Critical Facilities
2. Learning the Operation of Lifelines and Facilities and Their Potential for Disruption by Hurricane
3. Checking the Vulnerability of the Lifelines and Facilities through Field Inspection and Investigation
4. Establishing a Positive Working Relationship with the Agencies and Companies that Manage the Infrastructure and Services of the Community
5. Developing an Understanding of the Total Risk to the Community
6. Formulating a Mitigation Strategy

The degree to which local communities can survive damage and disruption from severe storms and hurricanes also depends to a large extent on how well the basic services and infrastructure, the common goods of the community, stand up to the wind and rain accompanying these storms. Whereas individual families bear full responsibility for preparing their own shelter to withstand the effects of storms, they have a much more limited role in ensuring that their common services are safeguarded, yet one that cannot be neglected.

Non-governmental agencies involved in low income housing construction and upgrading have developed practical and low cost measures for increasing the resistance of self-built houses to hurricane force winds. Typical of efforts of this nature is the work performed by the Construction Resource and Development Centre (CRDC) in Jamaica, which produced educational materials and organized workshops on house and roof reconstruction following Hurricane Gilbert.

The principal responsibility for introducing an awareness and concern in the community regarding the risk posed by hurricanes to the common good rests with the community leadership and local-or district-disaster coordinator, if such a function exists. It involves a lengthy process of identifying the issues, mobilizing resources from within the community and from outside, and building support for common action.

Such a process consists of six steps: (1) making an inventory of lifeline networks and critical facilities; (2) learning the operation of these and their potential for disruption by a hurricane; (3) checking the vulnerability of the lifelines and critical facilities through field inspection and investigation; (4) establishing a positive working relationship with the agencies and companies that manage the infrastructure and services of the community; (5) developing an understanding of the total risk to the community; (6) formulating and implementing a mitigation strategy.

1. Inventory of Lifeline Networks and Critical Facilities

Lifeline networks and critical facilities are those elements in the economic and social infrastructure that provide essential goods and services to the population in towns and villages. Their proper functioning is a direct concern of the community, since disruption affects the entire population.

The community leadership should gradually build up an inventory of these elements by locating them in a first instance on a large-scale map (1:5,000 or 1:2,500) of the community. The base maps can be obtained from the town and country departments or physical planning offices. The road network should indicate the road hierarchy (highway, principal access to settlement, local streets) and the location of bridges and other civil works such as major road cuts and retaining walls. Similar treatment should be given to the electricity and telephone networks and the water system. Residential areas and areas of economic activity should also be identified.

Various sources can be tapped to obtain this information. Water, electricity, and telecommunication companies can be called upon to draw their networks on the maps for the area in question. The local representative of the ministry of public works or physical planning office can assist with the identification of the road network and the location of public facilities housing important services.

2. Learning the Operation of Lifelines and Facilities and Their Potential for Disruption by Hurricane

Community leaders should periodically organize brief sessions in which the engineers or managers responsible for the different lifelines and facilities can explain the workings of their systems to selected residents who may be involved in disaster preparedness and response. The maps that were prepared earlier should be helpful during these sessions, while at the same time particular details can be reviewed and updated. The focus of these sessions should be:

- Identification of the different elements that make up the system, their interaction, and their interdependency.

- How the different elements function, what can go wrong, and what the normal repair and maintenance procedures are.

- How each of the elements of the system can be affected by the forces accompanying a hurricane.

- What the consequences of a hurricane could be for the functioning of the system and for the users.

WHAT ARE THE LIFELINE NETWORKS:

Road network, with roads, bridges, road cuts and retaining walls, elevated roads, drainage works.

Water system, with surface intakes, wells, pipelines, treatment plants, pumping stations, storage tanks or reservoirs, water mains, and distribution network.

Electricity system, with generating plant, transmission lines, substations, transformers, and distribution network.

Telecommunication, with ground station, exchanges, microwave transmission towers, aerial and underground cables, and open line distribution network.

Sanitation system, with collector network, treatment plant and sewage fallout; public washrooms and toilet facilities; solid waste collection and disposal.

WHAT ARE THE CRITICAL FACILITIES:

Hospitals, health centers, schools, churches.

Fire stations, police stations, community centers, shelters, and other public buildings that house vital functions that play a role in emergencies.

3. Checking the Vulnerability of the Lifelines and Facilities through Field Inspection and Investigation

The vulnerability of buildings and infrastructure elements will be determined first of all by their location with respect to hazard-prone areas. Storm surges and wave action can inflict severe damage in waterfront and low-lying coastal areas; heavy rains accompanying the hurricanes can cause flash flooding or riverine flooding along the river banks and in low-lying areas; rain can also cause land slippages and mudslides on steep slopes and unstable roadcuts; and structures in exposed areas such as ridges and bluffs are particularly vulnerable to wind damage.

Hazard-prone areas should be systematically identified and located on the lifeline and critical facilities map, to show where lifeline networks and critical facilities may be especially vulnerable.

The next step consists of a visual inspection and observation of all important infrastructure elements and critical facilities. Details of location and construction that may affect vulnerability should be noted and recorded on a sheet, together with a brief description of the possible damage that may occur.

4. Establishing a Positive Working Relationship with the Agencies and Companies that Manage the Infrastructure and Services of the Community

Once the community leadership has collected a fair amount of information, a series of consultations should be organized with the engineers or managers responsible for each of the lifeline and critical facilities of the settlement, or with their local representatives, and further elaboration of the information collected thus far should take place.

Such consultations provide an opportunity for the community leadership to learn about the maintenance and emergency repair policies practiced in their settlements by the different agencies and utility companies, to get to know the officers responsible for carrying out emergency repairs, and to find out how to contact them under normal circumstances as well as in emergencies.

Good contacts between agency representatives and community leadership are of great help in exploring the coincidence of interest between the residents on the one hand and the service agencies and companies on the other. Through effectively managed participation by the residents in such tasks as monitoring the state of repair of the infrastructure or keeping drains clear, the community can receive better services at a lower cost to the agencies responsible. The actual hiring of workers or small firms from the settlement to execute some of the agencies' tasks should be encouraged wherever possible.

LEARNING FROM PAST DISASTERS

Very valuable information about the vulnerability of small towns and villages can be obtained by inquiring into the local hurricane damage history. This is done through interviews with older residents in the community, retired public works officials familiar with the area, and other informants; by searching in old newspapers, and documents; and other means that may be appropriate in the particular setting.

The information should be organized by event, and within each event by infrastructure element that was affected. Damage that resulted from that particular impact should be briefly described. An effort should be made to collect at least the following data:

a. The EVENT:

- date of occurrence
- duration
- areas affected
- measures of strength (wind speed, height of flood waters)
- other characteristics that distinguish the event from others

b. The particular ELEMENT that was affected:

- class and type of element

- physical characteristics

- any information on what may have made the element vulnerable at that time-for example, poor state of repair or accumulated debris

c. The DAMAGE that was caused:

- quantitative and qualitative description of direct physical damage
- description of indirect damage, such as loss of function, interruption of service, loss of Jobs

5. Developing an Understanding of the Total Risk to the Community

To be meaningful, the view of the risk posed by hurricanes to a settlement should include the perspective of the population and its economic activities. In such an integrated view, vulnerability is obviously more than the sum of the technical deficiencies experienced by structures or equipment in the face of excessive natural forces. The traditional sectoral organization of the public system provides a poor basis for an integrated vulnerability analysis, since it tends to overlook the dependency and interaction between different infrastructure systems, which are often major determinants of the vulnerability of a society.

The different pieces of information collected so far will have to be put together to create an understanding of the total risk to which the settlement can be subject, and of the variations of this risk within the settlement according to the location and vulnerability of specific elements of the infrastructure. The following techniques have proved helpful in this exercise.

- Creating a visual image

All the information collected earlier is compiled on the large-scale base map of the settlement, either directly on the same map, on acetate overlays, or a few different copies. The final number of maps depends on the scale of the base map and the complexity of the information.

INVOLVING THE COMMUNITY IN VULNERABILITY REDUCTION

In St. Kitts and Nevis, the Ministry of Education, the Ministry of Public Works, and the Disaster Preparedness Office organized local residents to repair the schools with materials donated by USAID. The school children benefited from safer, more operable buildings, while the community as a whole benefited from having safer hurricane shelters, a function which school buildings across the island automatically acquire during the hurricane season.

EXAMPLES FOR COMMUNITY ACTION

Contributions that local communities can make to reducing the vulnerability of their basic services are typically non-structural, and are built around routine monitoring and maintenance. Some examples:

- Avoid throwing garbage, especially large objects such as tires, tree branches, and appliances, into gullies and rivers. These tend to accumulate near bridges and culverts, forming obstacles to normal water flow.

- Do not remove natural vegetation from river and gully banks, and from cut slopes, in order to avoid accelerated erosion of the banks.

- Keep roadside drainage clear of silt and other objects; pay special attention to crossover culverts.

- Do not remove sand and stones from beaches.

- Keep overhanging branches away from electricity and telephone lines.

- Do not tamper with electricity/telephone poles; report any visible signs of deterioration of the poles or their stays.

- Report any visible signs of deterioration to public buildings, paying special attention to roofs and windows.

- Do not interfere with water intakes; report excessive silting or obstructions.

The maps will highlight where hazardous events can strike, who suffers the risks, what functions are threatened, where direct damage can be experienced, and what the level of risk is.

- Creating impact scenarios

With the help of the maps, much can be learned about the risk to which the community is subject by formulating realistic scenarios of the impact of a hurricane on the settlement and simulating the consequences for population, lifelines, and critical facilities.

These scenarios can be reviewed with various groups in the community. Discussion of the different scenarios creates the perfect background against which to start thinking about what the community can do to reduce the risk, which is after all the purpose of the exercise.

6. Formulating a Mitigation Strategy

The formulation of a strategy to introduce appropriate mitigation measures that respond to the community's priorities is the culmination of all the efforts expended on the vulnerability analysis and risk assessment.

It is important that the community leadership focus on identifying realistic mitigation measures and proposing a simple implementation strategy. The common pitfall of identifying measures that require substantial funding should be avoided by concentrating on non-structural measures. Typical of the measures that should be emphasized are those that can be integrated into routine maintenance or upgrading of infrastructure; the avoidance of environmental degradation that can decrease the natural protective capacity of resources such as sand dunes, mangroves, and other natural vegetative coverage; and prevention by means of proper planning and design of new investments.

It is also important to establish the role of the different governmental levels and agencies in the country in the implementation of a mitigation strategy. The range of actions under the control of a small community is obviously quite limited, and depends on the degree of autonomy of the local government, the level of resources it controls, and the expertise it is able to mobilize.

REFERENCES

American Meteorological Society. Proceedings of the 17th Conference on Hurricane and Tropical Meteorology, April 7-10, 1987 (Miami, Florida: American Meteorological Society, 1988).

Bacon, P. Assessment of the Economic Impacts of Hurricane Gilbert on Coastal and Marine Resources in Jamaica. UNEP Regional Seas Reports and Studies, no. 110 (Kingston, Jamaica, 1989).

Bender, S. "Disaster Prevention and Mitigation in Latin America and the Caribbean" in Colloquium on Disasters, Sustainability and Development: A Look to the 1990's. The Environment Department and the Human Resources Development Division of the World Bank (Washington, D.C.: The World Bank, 1989).

Cambers, G. UNESCO Regional Office for Science and Technology for Latin America and the Caribbean. An Overview of Coastal Zone Management in Six East Caribbean Islands: Grenada, St. Vincent, St. Lucia, Dominica, St. Kitts, Antigua, East Caribbean Erosion Coasts and Beaches in the Caribbean Islands (Montevideo: COMAR-COSALC, 1985).

Caribbean Disaster News (St. John's, Antigua: UNDRO/PCDPPP). June 1989: "Telecommunications: The Experience of Hurricane Gilbert." September/December 1988, No. 15/16: "Lessons from Recent Events: Hurricane Gilbert," "Gilbert in the Caribbean," "Gilbert Smashes Jamaica," and "Hurricane Joan."

Collymore, J. Planning Hurricane Mitigation for Caribbean Agriculture (unpublished paper) (Blacksburg, Virginia: Virginia Polytechnic Institute, 1987).

Commonwealth Science Council. Coastal Zone Management of the Caribbean Region: A Status Report. Environmental Planning Programme Coastal Zone Management of Tropical Islands (1987).

Davenport, A.G., Georgiou, P.N., and Surry, D. A Hurricane Wind Risk Study in the Eastern Caribbean, Jamaica and Belize with Special Consideration to the Influence of the Topography (London, Ontario, Canada: The University of Western Ontario, 1985).

Earthscan Press Briefing Document no. 34a (Washington, D.C.: Earthscan Washington Bureau, 1983).

ECLAC/UNEP. Natural Disasters Overview. Meeting of Government-Nominated Experts to Review the Draft Action Plan for the Wider Caribbean Region, Caracas, Venezuela, 28 January -1 February, 1979 (Caracas: ECLAC/UNEP, 1979).

ESCAP/WMO. Guidelines for Disaster Prevention and Preparedness in Tropical Cyclone Areas (Geneva and Bangkok: ESCAP/WHO/LRLS, 1977).

Frank, N. Testimony before U.S. House of Representatives on August 9, 1984 in Congressional Record (Washington D.C., 1984).

Goldberg, S., Sim, A., and Kohler, R. "An Updated Fine-Grid Version of the Operational Barotropic Hurricane Track Prediction Model" in Proceedings of the 17th Conference on Hurricane and Tropical Meteorology, April 7-10, 1987 (Miami, Florida: American Meteorological Society, 1988).

Inter-American Development Bank Newsletter, June, 1989. "Emergency Plan for the Yucatan" (Washington, D.C.: Inter-American Development Bank, 1989).

Jarrell, J. Topic 9.3 Probability Forecasting in Rapporteur Report of WMO/CAS International Workshop on Tropical Cyclones.

Landsberg, H. "Do Tropical Storms Play a Role in Weather Balance of the Northern Hemisphere?" in Journal of Geophysical Research, vol. 65, no. 4 (1960).

Mathur, M. "Development of the NMC'S High Resolution Hurricane Model" in Proceedings of the 17th Conference on Hurricane and Tropical Meteorology, April 7-10, 1987 (Miami, Florida: American Meteorological Society, 1987).

Maul, G. Implication of Sea Level Rise in the Wider Caribbean Region (Washington, D.C.: The World Bank, 1988).

National Oceanographic and Atmospheric Agency. National Hurricane Operations Plan (Washington D.C.: U.S. Department of Commerce, NOAA, 1977).

Storm Surge and Hurricane Safety with North Atlantic Tracking Chart (Washington, D.C.: U.S. Department of Commerce, NOAA).

Neumann, C.J., et al. Tropical Cyclones of the North Atlantic Ocean, 1871-1986 (Washington, D.C.: U.S. Department of Commerce, NOAA, 1987).

Office of Foreign Disaster Assistance, U.S. Agency for International Development. Disaster History: Significant Data on Major Disasters Worldwide, 1900-Present (Washington, D.C.: OFDA/USAID, 1989).

Planning Institute of Jamaica. Economic and Social Assessment of the Impact of Hurricane Gilbert (Kingston: Planning Institute of Jamaica, 1989).

Ralph M. Field Associates. Natural Hazard Mitigation: A Recommended Strategy for Jamaica (February, 1984).

Tomblin, J. "Natural Disasters in the Caribbean: A Review of Hazards and Vulnerability" in Caribbean Disaster Preparedness Seminar, Saint Lucia, June, 1979 (Washington, D.C.: OFDA/USAID, 1979).

- "Earthquakes, Volcanoes and Hurricanes: A Review of Natural Hazards and Vulnerability in the West Indies" in Ambio, vol. 10, no. 6 (1981).

United Nations Disaster Relief Office. Disaster Prevention and Mitigation: A Compendium of Current Knowledge, vol. 4, Meteorological Aspects (New York, United Nations, 1978).

UNDRO News. April, 1989. "Hurricanes Gilbert and Joan: Implications for the Caribbean Housing Sector" (Geneva: UNDRO, 1989).

Wilches-Chaux, G. Desastres, Ecologismo y Formación Profesional (Popayán, Colombia: Servicio Nacional de Aprendizaje, 1989).

Previous Page Top of Page Next Page