Table of contents :

Catastrophic events share characteristic nonlinear behaviors that are often generated by cross-scale interactions and feedbacks among system elements. These events result in surprises that cannot easily be predicted based on information obtained at a single scale. Progress on catastrophic events has focused on one of the following two areas: nonlinear dynamics through time without an explicit consideration of spatial connectivityor spatial connectivity and the spread of contagious processes without a consideration of cross-scale interactions and feedbacks^ref. These approaches rarely have ventured beyond traditional disciplinary boundaries. An interdisciplinary, conceptual, and general mathematical framework for understanding and forecasting nonlinear dynamics through time and across space has been proposed. Decisions that minimize the likelihood of catastrophic events must be based on cross-scale interactions, and such decisions will often be counterintuitive. Given the continuing challenges associated with global change, approaches that cross disciplinary boundaries to include interactions and feedbacks at multiple scales are needed to increase our ability to predict catastrophic events and develop strategies for minimizing their occurrence and impacts^ref.
Dilley and colleagues at the World Bank broke down the most of the globe into 8 million grid cells of about 25 km² each. They then mapped the risks of human and economic damage from 6 types of disaster, such as cyclones and landslides, on to each one and built up a picture of the world's most exposed places^ref. The world's most vulnerable countries include Bangladesh, Nepal, Burundi, Haiti and Taiwan. In these places, > 90% of people are at 'high risk' of death for 2 or more types of disaster. The researchers define high-risk areas as having the top 30% of risk compared with other areas of the world. Although many of these areas were already known to be in danger, the report provides a more sophisticated way to compare risks across countries and regions, allowing governments and aid agencies to prioritize their resources. Much of the damage and death that disasters cause is preventable: by building earthquake-proof structures, for example. But repeated hits lock many of the world's developing countries into a cycle that makes it difficult to fund changes, especially as much aid goes into immediate relief efforts. The World Bank plans to use its hotspot map to identify those countries most in need and help them implement a preventive, rather than reactive, approach to disasters. Its approach is already affecting homeowners in Turkey, who must weather frequent earthquakes. When providing aid, the World Bank requires them to buy insurance for their homes. This shifts the responsibility for safe buildings from the government to the individual and private-sector insurance companies. The World Bank also intends to encourage governments to invest in measures such as flood embankments and cyclone shelters by granting loans to countries who plan for disasters. Countries have already started to request money specifically for risk management, indicating that the message is getting through. We must stop making it more complicated than it is : if you want to reduce problems after disasters, you just have to protect people by giving them better housing, better education and better health services.

• volcanic eruptions :


History :
• the eruption of the volcano Krakatau in the Netherlands East Indies (Indonesia) in 1883 had worldwide impact. This was perceived in the 3 quite different types of global propagation that occurred after the eruption :
• a rapid pressure wave, noticeable only to measuring instruments
• followed a few hours later by the spread of the news of the event
• succeeded by a slowly expanding optical phenomenon that lasted for a couple of years.
Krakatau was the first natural catastrophe of global magnitude that was almost immediately recognized as such throughout the world, largely thanks to the recently installed worldwide telegraphic network^ref. It was associated to a tsunami in the Indian Ocean.
Assuming that most of the man-made energy, which is used for everything from residential lighting and heating to manufacture and transport, will end up in the form of heat, in 2001, the amount of heat energy produced by volcanoes was 1000 times less than the energy consumed by the USA. Over 2001 and 2002, these volcanoes kicked out about 5 x 10¹⁶ J/yr - enough to power New York city for a few months. Single eruptions of one or two volcanoes can make up a large part of the year's heat budget. Nyamuragira in the Democratic Republic of Congo and Mount Etna in Italy contributed about 40% of the volcanic energy total for 2001. When Mount St Helens northwest USA erupted in 18 May 1980 killing 57 people, it released > 10¹⁸ J of heat at once - about 20 times the total heat flow from all the volcanoes studied in 2001^ref : that blast removed the mountaintop, created the biggest landslide in recorded history, and deposited a blanket of ash as far away as Montana.
St Helens experienced a series of very small earthquakes that began on 23 September 2004 : these tiny shudders are thought to be caused by the underground movement of pressurized steam or gases released from rising magma. Since the 1980 eruption, a dome of lava nearly 300 metres high has built up within the crater at the mountain's summit. This forms a plug that seals off the vent to the magma chamber below. Now magma seems to be rising in the vent. It's possible that the hot gases it will release might blow debris and ash out of the dome. There have been no sightings of gas being released from the dome, for example, which would be expected if magma is rising. There could be landslides and flows of debris from the crater, or even flows of molten lava. The region around the volcano, in the Cascades mountain range, is mostly uninhabited.
Surveillance :
• remote sensing of seismic tremors
• geodetic monitoring : global positioning system (GPS) measurements of changes in ground level  from satellites
• satellite measurements of heat changes as rising magma warms up a volcano's crater is still an imperfect science that involves subjective judgements
• After spending more than a year assessing 169 active volcanoes in the USA and the Mariana Islands, experts have identified the volcanoes that pose the greatest threat to people and property. The US Geological Survey (USGS) report, issued on 29 April 2005, also pins down what steps should be taken to fill the gaps in the current monitoring systems for those volcanoes. We know that many of the most damaging effects of eruptions can be mitigated if proper monitoring is in place. Unlike experts on earthquakes and many other natural disasters, volcanologists can do a fairly good job of predicting when disaster will strike, provided they have studied the volcano in question. The flanks of a volcano may swell up because of the magma inside, or small tremors may be provoked, and these hints can be used to forecast volcanic activity. The eruptions of volcanoes that have been heavily studied for years, such as Italy's Mount Etna, can be reliably predicted to within hours or even minutes. But for volcanoes that have not been studied and are not being monitored, researchers have little hope of predicting an eruption. We know there are many under-monitored volcanoes in the USA. The USGS report ranks each volcano based on the characteristics of its eruption (whether it would spew fast- or slow-moving lava, for example), how often it tends to erupt, whether there are major developments in the area, how close the nearby airports are, and how many passengers fly over it each day. The results include a top-10 list of the most dangerous volcanoes : the score shown below is arrived at by looking at both the destructive capacity of each volcano, along with the people and property at risk.
• Kilauea, Hawaii : erupting now
• St Helens, Washington : erupting now
• Rainier, Washington
• Hood, Oregon
• Shasta, California
• South Sister, Oregon
• Lassen Volcanic Center, California
• Mauna Loa, Hawaii : signs of unrest
• Redoubt, Alaska
• Crater Lake, Oregon
The volcanoes highlighted as priorities include five volcanoes that are currently erupting or showing unrest, such as Mount St Helens in Washington; 13 very threatening volcanoes that are inadequately monitored, most of which are in the Cascade Range on the west coast of North America; and 19 volcanoes that pose a great threat to aviation safety but have no real-time sensors for volcanic activity, many of which are in Alaska. Guffanti notes that people often rush to monitor a volcano only after it has erupted and proved that it is dangerous. But this is a bad idea : i is difficult and risky to play catch-up with a volcano. In 2002, for example, hundreds of thousands of people were forced to evacuate and dozens of people died in the eruption of Goma volcano in the Democratic Republic of Congo. Scientists scrambled to improve monitoring equipment at that volcano only after the eruption took place. Japan, on the other hand, is known for its proactive approach to monitoring volcanoes and for applying the most advanced technology to that quest. We have a lot to gain by looking at what's going on in Japan.  In order to address the gaps in monitoring identified by its report, the USGS suggests setting up a National Volcano Early Warning System. Among other things, the system would enhance the monitoring of the most dangerous volcanoes, establish a watch office that would run 24 hours a day, seven days a week, and aid the sharing of data between observatories. "This is the most thorough and quantitative effort I've seen for focusing resources on the volcano hazards problem. The USGS plans to polish the warning-system proposal with input from other federal agencies, local officials and businesses. The report's authors hope that the framework will be phased in over the next 8-10 years.
Top 10 volcanoes by heat output 2001-2002 (x 10¹⁵ J/yr) :
• Kilauea, Hawaii : 13
• Nyamuragira, Democratic Republic of Congo : 9
• Etna, Italy : 6
• Fuego, Guatemala : 3
• Soufrière Hills, Montserrat, West Indies : 3
• Shiveluch, Kamchatka :2
• Erta Ale, Ethiopia : 2
• Ambrym, Vanuatu, South Pacific : 2
• Piton de la Fournaise, Reunion, Indian Ocean : 2
• Nyiragongo, Democratic Republic of Congo : 2
An eruption of Mount Vesuvius, near Naples in Italy, could kill as many as the Indonesian tsunami, but lessons learned from the 1997 eruption on the Caribbean island of Montserrat could reduce the damage. Vesuvius is one of the most serious problems facing Europe. The activity at the Soufrière Hills Volcano in Montserrat began in 1995, but the worst damage happened in 1997, when fast-flowing streams of hot gas and ash, called pyroclastic flows, struck several villages. The same kind of eruption devastated Pompeii in 79 AD. A pyroclastic flow on 26 December 1997, already hot enough to ignite furniture and other materials, became even more damaging as it picked up debris and hurled it at buildings^ref. This volcanic flow was not the "turbulent mess" that has often been assumed. The pressure of the hot gas was surprisingly low, and it caused much of its damage by entering buildings through open windows and doors. These streams became highly focused. Like gusts of wind, they come in one window and go out another. A person cowering in a corner might escape the scorching material untouched. People have assumed that pyroclastic flows would batter buildings down, but the Montserrat study shows that heat-resistant coverings on windows and doors could greatly reduce damage to buildings. The Neapolitan authorities plan a mass evacuation if Vesuvius threatens to blow, as one day it surely will,k but it could take at least 5 days to clear the area, and pyroclastic flows might strike sooner than that. Protective barriers for windows might save lives. The Montserrat data will be used to develop a computer model of an eruption of Vesuvius. A new study of the effects of such flows on the Roman city of Pompeii might aid their efforts^ref. The temperatures in Pompeii were estimated from the amount of magnetism left unmelted in rocks and building fragments. By mapping the temperatures throughout the remains, the researchers could see how the shapes and arrangements of buildings and streets set up turbulence that could have cooled the flow in some places. The findings paint a bleak picture. The town, 9 km southeast of Vesuvius, was smothered in about 2.5 m of ash even before the 300 °C pyroclastic flows struck, choking life and caving in roofs. And changes in the flow swirling over walls didn't seem to reduce the temperature below about 100 °C, so survivors of the ash would have burned to death. Civil engineers are interested in the temperature measurements at Pompeii because they offer clues about how soon emergency vehicles could drive into a town struck by pyroclastic flows without their tyres melting.
Volcanoes can cool the planet by keeping methane-producing bacteria at bay. Scientists already knew that a major volcanic eruption can cool the Earth for a couple of years, because the particles and chemicals thrown into the air make clouds that reflect sunlight back out into space. But the sulphur dioxide in volcanic plumes also affects the climate indirectly. Falling in acid rain, these fumes can feed sulphur-loving bacteria in wetlands, allowing them to out-compete methane-generating bacteria, and so reducing the amount of methane emitted into the atmosphere. About 50% the world's methane - a greenhouse gas stronger than carbon dioxide - comes from bacteria in peat bogs and rice paddy fields. The sulphurous effects of a volcanic eruption were simulated by adding the mineral sodium sulphate to a peat bog in northeast Scotland : the area was fertilized in 1998, and methane emissions were 40% lower 2 years later. The bog still contained elevated levels of sulphur;  it would take 5-10 years from time of fertilisation for methane emissions to return to normal^ref. The experiment was intended to recreate the acid-rain fallout from the enormous 1783 eruption of the volcano Laki in Iceland, which emitted > 120 million tonnes of sulphur dioxide, 10 times more than western European industry produces each year. Iceland has events like this every few hundred years, and there are volcanoes that continuously pump out sulphur for decades. But simulating Laki's impact on a small area of Scotland requires a fair amount of guesswork : we have really good estimates of how much sulphur came out of Laki, but we don't know exactly where it rained out as acid rain. Volcanic sulphur might have a bigger climatic impact than industrial emissions, simply because most wetlands are closer to volcanoes than to factories or power plants. Further research should pin down how much global cooling volcanoes can trigger in this way. But in pre-industrial times, when volcanoes were virtually the sole source of acid rain, these sudden fertilization events could have driven very rapid climate shifts. 50 million years ago, the warm greenhouse climate of the day was due, in large part, to methane from the extensive wetlands that covered the Earth : during that time, large volcanic eruptions could have been real agents of rapid climate change through this mechanism. Some scientists have suggested that fertilizing paddy fields with sulphur could slow down climate change. Industry and volcanoes may already produce enough acid rain to maximize the bug-boosting effect.
Web resources :
• Volcano World

Methane information from US EPA
• Exploris : a European project which is studying the human risks of eruptions and how to mitigate them. As well as Vesuvius and Montserrat, Exploris is looking at hazards posed by Mount Teide in the Canary Islands, Sete Cidades in the Azores, and others.
• Volcanic Activity Report
• US Geological Survey (USGS)
• Cascades Volcano Observatory (CVO)
• Volcano Hazards Program
• Smithsonian's Global Volcanism Program
• Stromboli Online, hosted by Italian researchers Roberto Carniel and Marco Fulle and Swiss teacher Jürg Alean
• storm / tempest : an individual low-pressure disturbance, complete with winds, clouds, and precipitation. The name is associated with destructive or unpleasant weather
• wind :

Names of winds :
• abroholos : a squall frequent from May through August between Cabo de Sao Tome and Cabo Frio on the coast of Brazil.
• austru : a east or southeast wind in Rumania.  They are cold in winter and may be a local name for a foehn wind
• Bali wind : a strong east wind at the eastern end of Java.
• Barat : a heavy northwest squall in Manado Bay on the north coast of the island of Celebes, prevalent from December to February.
• Barber : a strong wind carrying damp snow or sleet and spray that freezes upon contact with objects, especially the beard and hair.
• Bayamo : a violent wind blowing from the land on the south coast of Cuba, especially near the Bight of Bayamo.
• Bentu de Soli : an east wind on the coast of Sardinia.
• Bora : a cold, northerly wind blowing from the Hungarian basin into the Adriatic Sea
• Borasco : a thunderstorm or violent squall, especially in the Mediterranean.
• Boreas : a ancient Greek name for north winds. The term may originally have meant "wind from the mountains" and thus the present term BORA
• Brickfielder : a wind from the desert in Southern Australia. Precedes the passage of a frontal zone of a low passing by. Has the same dusty character as the Harmattan
• Brisa, Briza : a northeast wind which blows on the coast of South America or an east wind which blows on Puerto Rico during the trade wind season. 2. The northeast monsoon in the Philippines.
• Brisote : the northeast trade wind when it is blowing stronger than usual on Cuba.
• Brubu : a name for a squall in the East Indies.
• bull's eye squall : a squall forming in fair weather, characteristic of the ocean off the coast of South Africa. It is named for the peculiar appearance of the small isolated cloud marking the top of the invisible vortex of the storm.
• Cape doctor : the strong southeast wind which blows on the South African coast
• Caver, Kaver : a gentle breeze in the Hebrides.
• Chubasco : a violent squall with thunder and lightning, encountered during the rainy season along the west coast of Central America.
• Churada : a severe rain squall in the Mariana Islands during the northeast monsoon. They occur from November to April or May, especially from January through March.
• Contrastes : winds a short distance apart blowing from opposite quadrants, frequent in the spring and fall in the western Mediterranean.
• Cordonazo : the "Lash of St. Francis." Name applied locally to southerly hurricane winds along the west coast of Mexico. It is associated with tropical cyclones in the southeastern North Pacific Ocean. These storms may occur from May to November, but ordinarily affect the coastal areas most severely near or after the Feast of St. Francis, October 4.
• Coromell : a night land breeze prevailing from November to May at La Paz, near the southern extremity of the Gulf of California.
• Diablo : Northern California version of Santa Ana winds. These winds occur below canyons in the East Bay hills (Diablo range) and in extreme cases can exceed 60 mph. They develop due to high pressure over Nevada and lower pressure along the central California coast
• Doctor : a cooling sea breeze in the Tropics. 2. See HARMATTAN. 3. The strong SE wind which blows on the south African coast. Usually called Cape Doctor
• Elephanta : a strong southerly or southeasterly wind which blows on the Malabar coast of India during the months of September and October and marks the end of the southwest monsoon.
• Etesian : a refreshing northerly summer wind of the Mediterranean, especially over the Aegean Sea.
• Euros : the Greek name for the rainy, stormy southeast wind
• Foehn : a warm dry wind on the lee side of a mountain range, whose temperature is increased as the wind descends down the slope. It is created when air flows downhill from a high elevation, raising the temperature by adiabatic compression. Classified as a katabatic wind. Examples include
• Chinook / snow eater : a type of foehn wind. Refers to the warm downslope wind in the Rocky Mountains that may occur after an intense cold spell when the temperature could rise by 20 to 40°F in a matter of minutes
• Santa Ana : a strong, hot, dry wind blowing out into San Pedro Channel from the southern California desert through Santa Ana Pass.
• Fremantle Doctor : a cooling seabreeze in Western Australia,often made note of during hot summer-time cricket matches
• Gregale : a strong northeast wind of the central Mediterranean.
• Haboob : a strong wind and sandstorm (or duststorm) in the northern and central Sudan, especially around Khartum, where the average number is about 24 per year. The name come from the Arabic word, "habb", meaning wind
• Harmattan : the dry, dusty trade wind blowing off the Sahara Desert across the Gulf of Guinea and the Cape Verde Islands. Sometimes called the DOCTOR, because of its supposed healthful properties.
• Knik wind : a strong southeast wind in the vicinity of Palmer, Alaska, most frequent in the winter.
• Kona storm : a storm over the Hawaiian Islands, characterized by strong southerly or southwesterly winds and heavy rains.
• Leste : a hot, dry, easterly wind of the Madeira and Canary Islands.
• Levanter : a strong easterly wind of the Mediterranean, especially in the Strait of Gibraltar, attended by cloudy, foggy, and sometimes rainy weather especially in winter.
• Levantera : a persistent east wind of the Adriatic, usually accompanied by cloudy weather.
• Levanto : a hot southeasterly wind which blows over the Canary Islands.
• Leveche : a warm wind in Spain, either a foehn or a hot southerly wind in advance of a low pressure area moving from the Sahara Desert. Called a SIROCCO in other parts of the Mediterranean area.
• Maestro : a northwesterly wind with fine weather which blows, especially in summer, in the Adriatic. It is most frequent on the western shore. This wind is also found on the coasts of Corsica and Sardinia.
• Maria : a fictional wind popularized in "Paint Your Wagon" (Lerner and Lowe, 1951) and by the Kingston Trio (1959), whose name may have originated with the 1941 book "Storm" by George R. Stewart. (Jan Null, Golden Gate Weather Services)
• Matanuska wind : a strong, gusty, northeast wind which occasionally occurs during the winter in the vicinity of Palmer, Alaska.
• Mistral / Cierzo : a cold, dry wind blowing from the north over the northwest coast of the Mediterranean Sea, particularly over the Gulf of Lions. See also FALL WIND.
• monsoon : a seasonal wind that brings rain to many places of the world, for example India and Southern Asia. The Indian monsoon, which waters India's agriculture, could run dry because of human impacts on the environment, a team of climate researchers has warned. Kirsten Zickfeld and her colleagues at the Potsdam Institute for Climate Impact Research say that the monsoon has two major settings: on, as at present, and off, when it produces very little rainfall. A switch-off would be catastrophic for India's main crop, rice, which depends on heavy monsoon rains. Even a minor change in monsoon timing or intensity can have a big impact. If the rains are delayed by just a few days, that affects the agricultural yields. The monsoon's disappearance would wreak havoc, probably requiring Indian farmers to completely change their crops and methods. Changes in land use and air pollution on the Indian continent are pushing conditions towards the off state. They don't know if or when it might happen, but they say there is cause for concern. The monsoon is driven by an air-pressure difference between the land and the Indian Ocean. Usually, the hot season creates low-pressure zones over the warm continent. Air rushes in from the higher-pressure zone over the water, bringing rain to the land. Anything that reduces this pressure difference - such as cooler land temperatures - can weaken the monsoon. And once the weakening exceeds a certain threshold, the climate switches into a new state in which moist air over the ocean is no longer carried inland^ref. In India and southeast Asia, several factors are causing less sunlight to warm the ground. There are more aerosols, because of industrial growth and greater vehicle use, which reflect light back into space. And clearing forests for farmland is replacing dark, light-absorbing treetops with lighter, more reflective soil.  "This raises a red flag", says Hans Joachim Schellnhuber, Zickfeld's co-worker and director of the Tyndall Centre for Climate Change Research in Norwich, UK. "If we continue to change land cover, and at the same time aerosols increase, we're moving towards the 'off' point. There are signs that the Chinese monsoon is weakening, perhaps for the same reasons. Their model is crude, so it can't predict when, or even if, current trends will trigger a change in the state of the monsoon. But their work's simplicity also has its advantages. Complicated computer models have suggested that the monsoon might change, but it has been hard to pick out the crucial causes. Global warming, caused by rising levels of atmospheric carbon dioxide, might make the monsoon more intense, increasing rainfall. That could be equally bad for the area, as illustrated by the 1,000 people killed in Mumbai last week in floods due to abnormally heavy monsoon rains. The worst case would be what Zickfeld's team calls a 'roller-coaster effect': drying of the monsoon, followed by the return of an even more intense wet monsoon as aerosol emissions are cleaned up but carbon dioxide goes on increasing. Such a series of changes would seriously challenge the adaptive capabilities of India's rural society.
• Nashi, N'aschi : a northeast wind which occurs in winter on the Iranian coast of the Persian Gulf, especially near the entrance to the gulf, and also on the Makran coast. It is probably associated with an outflow from the central Asiatic anticyclone which extends over the high land of Iran. It is similar in character but less severe than the BORA.
• Norte : a strong cold northeasterly wind which blows in Mexico and on the shores of the Gulf of Mexico. It results from an outbreak of cold air from the north. It is the Mexican extension of a norther.
• Nor'easter / Northeaster : a northeast wind, particularly a strong wind or gale; an unusually strong storm preceded by  northeast winds off the coast of New England
• Nor'wester : this is a very warm wind which can blow for days on end in the province of Canterbury New Zealand. The effect is especially felt in the city of Christchurch. The wind comes in from the Tasman Sea, drys as it rises over the Southern Alps, heats as it decends, crosses the Canterbury Plains, then blows through Christchurch
• (Blue) Norther : a cold strong northerly wind in the Southern Plains of the United States, especially in Texas, which results in a drastic drop in air temperatures
• Ostria : a warm southerly wind on the Bulgarian coast; considered a precursor of bad weather
• Pali : a local name for strong winds which blow through the Pali Pass above Honolulu, HI
• Pampero : a west or southwest wind in Southern Argentina. This wind (often violently) picks up during the passage of a cold front of an active low passing by
• Papagayo : a violet northeasterly fall wind on the Pacific coast of Nicaragua and Guatemala. It consists of the cold air mass of a norte which has overridden the mountains of Central America
• Shamal : a summer northwesterly wind blowing over Iraq and the Persian Gulf, often strong during the day, but decreasing at night.
• Sharki : a southeasterly wind which sometimes blows in the Persian Gulf.
• simoon : dry, hot, dust-laden wind, which blows in the Sahara, Palestine, Syria and the desert of Arabia.
• Sirocco : a warm wind of the Mediterranean area, either a foehn or a hot southerly wind in advance of a low pressure area moving from the Sahara or Arabian deserts. Called Leveche in Spain.
• Squamish : a strong and often violent wind occurring in many of the fjords of British Columbia. Squamishes occur in those fjords oriented in a northeast-southwest or east-west direction where cold polar air can be funneled westward. They are notable in Jervis, Toba, and Bute inlets and in Dean Channel and Portland Canal. Squamishes lose their strength when free of the confining fjords and are not noticeable 15 to 20 miles offshore.
• Suestado : a storm with southeast gales, caused by intense cyclonic activity off the coasts of Argentina and Uruguay, which affects the southern part of the coast of Brazil in the winter.
• Sumatra : a squall with violent thunder, lightning, and rain, which blows at night in the Malacca Straits, especially during the southwest monsoon. It is intensified by strong mountain breezes.
• Sundowner : warm downslope winds that periodically occur along a short segment of the Southern California coast in the vicinity of Santa Barbara. The name refers to their typical onset (on the populated coastal plain) in the late afternoon or early evening, though they can occur at any time of the day. In extreme cases, wind speeds can be of gale force or higher, and temperatures over the coastal plain and even at the coast itself can rise significantly above 37.8°C (100°F)
• Taku wind : a strong, gusty, east-northeast wind, occurring in the vicinity of Juneau, Alaska, between October and March. At the mouth of the Taku River, after which it is named, it sometimes attains hurricane force.
• Tehuantepecer : a violent squally wind from north or north-northeast in the Gulf of Tehuantepec (south of southern Mexico) in winter. It originates in the Gulf of Mexico as a norther which crosses the isthmus and blows through the gap between the Mexican and Guatamalan mountains. It may be felt up to 100 miles out to sea
• Tramontana : a northeasterly or northerly winter wind off the west coast of Italy. It is a fresh wind of the fine weather mistral type.
• Vardar / Vardarac : a cold fall wind blowing from the northwest down the Vardar valley in Greece to the Gulf of Salonica. It occurs when atmospheric pressure over eastern Europe is higher than over the Aegean Sea, as is often the case in winter
• warm braw : a foehn wind in the Schouten Islands north of New Guinea.
• white squall : a sudden, strong gust of wind coming up without warning, noted by whitecaps or white, broken water; usually seen in whirlwind form in clear weather in the tropics.
• Williwaw : a sudden blast of wind descending from a mountainous coast to the sea, in the Strait of Magellan or the Aleutian Islands.
• Willy-willy : a tropical cyclone (with winds > 33 knots) in Australia, especially in the southwest. More recent common usage is for dust-devils.
• Zephyros : the ancient Greek name for the west wind, which generally light and beneficial.  It has evolved into "zephyr" which denotes a soft gentle breeze
• Stevenson screen : a white box with ventilated sides that is used to house weather instruments and protect them from direct sunlight
• Beaufort wind speed scale : a scale of numbers representing different wind speeds and a description of their effects on land or sea. It was invented by Admiral Beaufort in the early 19th century.
• anemometer: an instrument that measures the speed or force of the wind.
• barometer : an instrument for determining the pressure of the atmosphere
• isobar : the line drawn on a weather map connecting points of equal barometric pressure
• closest point of approach : point where hurricane eye makes closest contact to shore without actually making landfall
• cold front : the leading edge of an advancing cold air mass that is underrunning and displacing the warmer air in its path. Generally, with the passage of a cold front, the temperature and humidity decrease, the pressure rises, and the wind shifts (usually from the southwest to the northwest in the Northern Hemisphere). Precipitation is generally at and/or behind the front, and with a fast-moving system, a squall line may develop ahead of the front. See occluded front and warm front
• convergence : wind movement that results in a horizontal net inflow of air into a particular region. Convergent winds at lower levels are associated with upward motion. Contrast with divergence
• depression : in meteorology, it is another name for an area of low pressure, a low, or trough. It also applies to a stage of tropical cyclone development and is known as a tropical depression to distinguish it from other synoptic features.
• filling : used in describing the history of a low-pressure system or an area of cyclonic circulation, it means an increase in the central pressure of the system. Although it usually describes the action of a pressure system on a constant pressure chart, it also means a surface low is decreasing in cyclonic circulation and losing its characteristics
• rain : precipitation in the form of liquid water droplets greater than 0.5 mm. If widely scattered, the drop size may be smaller. It is reported as "R" in an observation and on the METAR. The intensity of rain is based on rate of fall. "Very light" (R--) means that the scattered drops do not completely wet a surface. "Light" (R-) means it is greater than a trace and up to 0.10 inch an hour. "Moderate" (R) means the rate of fall is between 0.11 to 0.30 inch per hour. "Heavy" (R+) means over 0.30 inch per hour
• drought : a prolonged period with very little or no rain
• deepening : used in describing the history of a low-pressure system or an area of cyclonic circulation, it means a decrease in the central pressure of the system. Although it usually describes the action of a pressure system on a constant pressure chart, it also means a surface low is increasing in cyclonic circulation and acquiring more energy
• explosive deepening : a decrease in the minimum sea-level pressure of a tropical cyclone of 2.5 mb/hr for at least 12 hours or 5 mb/hr for at least 6 hours.
• rapid deepening : a decrease in the minimum sea-level pressure of a tropical cyclone of 1.75 mb/hr or 42 mb for 24 hours.
• disturbance : this has several applications. It can apply to a low or cyclone that is small in size and influence. It can also apply to an area that is exhibiting signs of cyclonic development. It may also apply to a stage of tropical cyclone development and is known as a tropical disturbance to distinguish it from other synoptic features
• front : the boundary between two dissimilar air masses
• occluded front : the front formed by a cold front overtaking a warm or stationary front and lifting the warm air above the earth's surface
• warm front : the leading edge of an advancing warm air mass that is replacing a retreating relatively colder air mass. Generally, with the passage of a warm front, the temperature and humidity increase, the pressure rises, and although the wind shifts (usually from the southwest to the northwest in the Northern Hemisphere), it is not as pronounced as with a cold frontal passage. Precipitation, in the form of rain, snow, or drizzle, is generally found ahead of the surface front, as well as convective showers and thunderstorms. Fog is common in the cold air ahead of the front. Although clearing usually occurs after passage, some conditions may produced fog in the warm air
• cold front : the leading edge of an advancing cold air mass that is underrunning and displacing the warmer air in its path. Generally, with the passage of a cold front, the temperature and humidity decrease, the pressure rises, and the wind shifts (usually from the southwest to the northwest in the Northern Hemisphere). Precipitation is generally at and/or behind the front, and with a fast-moving system, a squall line may develop ahead of the front. See occluded front and warm front.
• jet stream : relatively strong winds concentrated within a narrow current in the atmosphere.
• nautical mile : a unit of length used in marine navigation that is equal to a minute of arc of a great circle on a sphere. One international nautical mile is equivalent to 1,852 meters or 1.151 statue miles. Refer to a sea mile.
• knot : a unit for the measurement of speed in the nautical system. It is the nautical miles per hour.
• North Atlantic basin : the Atlantic Ocean north of the equator, the Caribbean Sea, and the Gulf of Mexico.
• satellite : used in reference to the manufactured objects that orbit the earth, either in a geostationary or a polar manner. Some of the information that is gathered by weather satellites, such as GOES9, includes upper air temperatures and humidity, recording the temperatures of cloud tops, land, and ocean, monitoring the movement of clouds to determine upper level wind speeds, tracing the movement of water vapor, monitoring the sun and solar activity, and relaying data from weather instruments around the world
• satellite pictures : pictures taken by a weather satellite, such as GOES-9, that reveal information, such as the flow of water vapor, the movement of frontal systems, and the development of a tropical system. Looping individual pictures aids meteorologists in forecasting. One way a picture can be taken is as a visible shot, which is best during times of visible light (daylight). Another way is as an IR (infrared) shot, which reveals cloud temperatures and can be used day or night
• spiral rainbands : bands of thunderstorms that spiral inward towards the center, where they wrap themselves around the eye
• squall : a sudden increase of wind speed by at least 18 miles per hour (16 knots) and rising to 25 miles per hour (22 knots) or more and lasting for at least one minute.
• storm surge : an abnormal rise in sea level accompanying a hurricane or other intense storm, and whose height is the difference between the observed level of the sea surface and the level that would have occurred in the absence of the cyclone. Storm surge is usually estimated by subtracting the normal or astronomic high tide from the observed storm tide. Note: waves on top of the storm surge will create an even greater high-water mark
• storm tide : the actual level of seawater resulting from the astronomic tide combined with the storm surge. If the storm comes ashore during astronomical low tide, the surge will be decreased by the amount of the low tide. If the storm makes landfall during astronomical high tide, the surge will be that much higher.
• synoptic scale : the size of migratory high and low pressure systems in the lower troposphere that cover a horizontal area of several hundred miles or more such as hurricanes. Contrasts with macroscale, mesoscale, and storms.
• trade winds : the wind system, occupying most of the tropics, which are northeasterly in the Northern Hemisphere and southeasterly in the Southern Hemisphere
• tropical disturbance : a discrete system of clouds, showers, and thunderstorms (organized convection) that originate in the tropics. Generally 100 to 300 miles in diameter and originating in the tropics or subtropics, disturbances have a nonfrontal migratory character, and maintain their identity for 24 hours or more. It may or may not be associated with a detectable perturbation of the wind field. An upper level of low pressure causes this to occur. Approximately 100 of these types of events occur annually during hurricane season.
• thunder : the sound that follows a flash of lightning and is caused by sudden expansion of the air in the path of the electrical discharge
• thunderstorm : a local storm produced by a cumulonimbus cloud, always with lightning and thunder, and usually accompanied by strong gusts of wind, heavy rain, and sometimes hail

Symptoms & signs : thunderstorms have often been linked to epidemics of tracheobronchial asthma, especially during the grass flowering season; however, the precise mechanisms explaining this phenomenon are unknown. Evidence of high respirable allergen loadings in the air associated with specific meteorologic events combined with an analysis of pollen physiology suggests that rupture of airborne pollen can occur. Strong downdrafts and dry, cold outflows distinguish thunderstorm rain from frontal rain. The weather system of a mature thunderstorm likely entrains grass pollen into the cloud base, where pollen rupture would be enhanced, then transports the respirable-sized fragments of pollen debris to ground level where outflows distribute them ahead of the rain. The conditions occurring at the onset of a thunderstorm might expose susceptible people to a rapid increase in concentrations of pollen allergens in the air that can readily deposit in the lower airways and initiate asthmatic reactions^ref.
• vortex : any circular or rotary flow in the atmosphere that possesses vorticity
• vorticity : the measurement of the rotation of a small air parcel. It has vorticity when the parcel spins as it moves along its path. Although the axis of the rotation can extend in any direction, meteorologists are primarily concerned with the rotational motion about an axis that is perpendicular to the earth's surface. If it does not spin, it is said to have zero vorticity. In the Northern Hemisphere, the vorticity is positive when the parcel has a counterclockwise, or cyclonic, rotation. It is negative when the parcel has clockwise, or anticyclonic, rotation.
• cyclone / low-pressure system : a storm of rotating and converging winds associated with atmospheric low pressure region or air mass. The cyclone winds spin around a common center with a relative pressure minimum, counter-clockwise in the northern hemisphere, or clockwise in the southern hemisphere. It usually has a diameter of 2000 to 3000 km. When developing, a cyclone typically consists of a warm front pushing northward and a cold front pushing southward with the center of low pressure (cyclone center) located at the junction of the 2 fronts. An area of a relative pressure minimum that has converging winds and rotates in the same direction as the earth.
• extratropical cyclone : cyclone in the middle and high latitudes often being 2000 km in diameter and usually containing a cold front that extends toward the equator for hundreds of kilometers. These cyclones forms outside the tropics, the center of storm is colder than the surrounding air, have fronts and the strongest winds in the upper atmosphere.
• subtropical cyclone : a low pressure system that develops over subtropical waters that initially has a non-tropical circulation but in which some elements of tropical cyclone cloud structure are present. Subtropical cyclones can evolve into tropical cyclones
• subtropical depression : subtropical cyclone in which the maximum sustained surface wind speed (using the U.S. 1-minute average) is < 38 mph (33 knots)
• subtropical storm : a subtropical cyclone in which the maximum sustained surface wind speed (using the U.S. 1-minute average) is > 39 mph (34 knots).
• tropical cyclone : a general term for all cyclone circulations originating over tropical waters. Its characteristics include a warm-core, non-frontal pressure system of synoptic scale that originates over the tropical or subtropical waters and has a definite organized surface. Used to define wind circulations rotating around an atmosphere which include tropical depressions, tropical storms, and hurricanes. The strongest winds of this cyclone are near the Earth's center. The basic element of Lighthill's "sandwich model" of tropical cyclones is the existence of "ocean spray," a layer intermediate between air and sea made up of a cloud of droplets that can be viewed as a "third fluid." A mathematical model of the flow in the ocean spray based on a semiempirical turbulence theory has been proposed and it has been demonstrated that the availability of the ocean spray over the waves in the ocean can explain the tremendous acceleration of the wind as a consequence of the reduction of the turbulence intensity by droplets. This explanation complements the thermodynamic arguments proposed by Lighthill^ref.
• Fujiwhara effect : a binary interaction where tropical cyclones within a certain distance (300-750 nautical miles depending on the sizes of the cyclones) of each other begin to rotate about a common midpoint
• center : the vertical axis or core of a tropical cyclone. It is usually determined by cloud vorticity patterns, wind, and/or pressure distributions.
• eye : the center of a tropical storm or hurricane characterized by a roughly circular area of light winds and rain-free skies and the lowest pressure. An eye will usually develop when the maximum sustained wind speeds exceed 78 mph. It can range in size from as small as 5 miles to up to 60 miles (20-50 km) but the average size is 20 miles. In general, when the eye begins to shrink in size, the storm is intensifying.
• eye wall : an organized band of convection surrounding the eye, or center, of a tropical cyclone. It contains cumulonimbus clouds, severest thunderstorms, heaviest precipitation and strongest wind
• feeder bands : in tropical parlance, the lines or bands of thunderstorms that spiral into and around the center of a tropical system. Also known as outer convective bands, a typical hurricane may have three or more of these bands. They occur in advance of the main rain shield and are usually 40 to 80 miles apart. In thunderstorm development, they are the lines or bands of low-level clouds that move or feed into the updraft region of a thunderstorm
• probability of tropical cyclone conditions : the probability, in percent, that the cyclone center will pass within 50 miles to the right or 75 miles to the left of the listed location within the indicated time period when looking at the coast in the direction of the cyclone's movement.
• tropical storm (TS) : a tropical cyclone in which the maximum sustained surface wind speed (1 minute average) is within the range of 39 to 73 mph (34 to 63 knots). At this point, the system is given a name to identify and track it. In the Atlantic/Caribbean/Gulf of Mexico basin, the names start with "A" each season.
• tropical depression (TD) : a tropical cyclone in which the maximum sustained surface winds (1 minute average) are < 38 mph (33 knots). Characteristically having one or more closed isobars, it may form slowly from a tropical disturbance or an easterly wave, which has continued to organize. At this point, it gets a cyclone number, starting with "TD01" at the beginning of each storm season.
• hurricanes : a tropical cyclone in the Northern Hemisphere with substained winds > 74 mph (64 knots) in the North Atlantic Ocean, Caribbean Sea or Gulf of Mexico and Eastern Pacific. The word is believed to originate from the Caribbean Indian storm god "Huracan". These winds blow in a large spiral around a relatively calm center of extremely low pressure known as the eye. Around the rim of the eye, winds may gust to > 200 mph. The entire storm, which can be up to 340 (550) in diameter, dominates the ocean surface and lower atmosphere over tens of thousands of square miles. Hurricanes draw their energy from the warm surface water of the tropics (usually above 27°C) and latent heat of condensation, which explains why hurricanes dissipate rapidly once they move over cold water or large land masses. Most North Atlantic hurricanes start over the middle of the ocean, and move west towards the Caribbean. Some continue west through the Gulf of Mexico, but others move up the east coast of Florida. It is extremely difficult to say whether the storms will move over land or back out over the sea. However, forecasters can usually predict a hurricane's path about 3 days in advance, with enough accuracy to give residents time to evacuate if necessary. This means hurricanes cause far fewer deaths today compared with 50 years ago. Although forecasters are increasingly adept at predicting the path of a hurricane, they can still be taken by surprise when it comes to guessing its strength. 'Hurricane-hunter' aircrafts are equipped with wind, temperature and pressure gauges. These planes have a dual function: they help forecasters by observing the storm in real time and gain valuable readings with which researchers can construct models of hurricane behaviour. To get right into the heart of the storm, the planes deploy 'dropsondes', packages of meteorological instruments dropped by parachute, but it is difficult to get the instruments in the right place. Climatologists are insisting the battering of hurricances in USA in 2004 (Charley, Frances, Ivan, Jeanne) is pure chance. Global warming is probably having little or no effect on the number of hurricanes, and there have been no more hurricanes in 2004 than last. In 2003, however, most of the storms blew themselves out over the sea. In 2004, many of them have headed straight for populated areas, which is simply bad luck.

Immense waves capable of sinking the largest ships might not be freaks of nature, but a common result of hurricanes. That's the implication of new evidence that Hurricane Ivan, which whirled across the Gulf of Mexico in September 2004, probably generated waves > 40 metres high from crest to trough. The water pressures measured by their array of seafloor sensors, 60 to 90 m below the ocean off the northeast coast of the gulf, indicate the passage of waves nearly 30 m high. But the waves near the eye of the hurricane, which unfortunately passed over the sensors while they were not taking measurements, would probably have topped 40 m. These are the largest wave heights ever recorded with instruments in US waters : they're larger than we ever thought they would be. Ivan wasn't even a particularly large hurricane. Terrible walls of water are a staple of nautical lore. But oceanographers have only recently come to accept them, as simple statistics suggest that such extreme events should almost never happen. But there's no fundamental reason why ocean waves can't grow to immense sizes. Nobody knows what the upper limit might be. The largest waves to hit shore come from tsunamis - waves generated most often by seafloor movements during major earthquakes. But while at sea these waves are just a few centimetres high, only growing as they reach shallow waters. The Indian Ocean tsunami of December 2004 reached 30 metres above sea level in the hardest hit areas. There are eyewitness accounts of similarly enormous waves at sea. In 1995, the Queen Elizabeth 2 liner survived an encounter with one about 30 m high in the North Atlantic, and 6 years later a similar wave smashed windows on the cruise ship Bremen in the South Atlantic and nearly sank it. It is now widely suspected that such rogue waves, generated by wind and currents, might explain the mysterious, regular disappearance of large ships at sea. One, the German supertanker München, vanished in 1978. Such waves are generally ascribed to unusual circumstances, such as fast-moving storms. But the new findings suggest that waves even larger than these might romp across the oceans whenever a hurricane hits. The latest hurricane models predict such waves should occur very often. The findings are consistent with less accurate measurements made from ships. The study may spell bad news for ships and coastal defences. If hurricane activity increases, as some expect to happen along with climate change, such giant waves could become more frequent^ref.
Global warming may be contributing to making hurricanes stronger. The storms have been getting more destructive over the past 3 decades. Global warming might increase the effect of hurricanes further still in the coming years. Previous studies have looked at the effects of rising global average temperatures on tropical cyclones, commonly called hurricanes or typhoons. But these past studies have tended to focus on whether the events have become more frequent. No such trends have been clearly observed. But researchers looked instead at the question of whether cyclones have got more intense, that is, whether they hit harder and last longer.  Theories and computer simulations of climate indicate that warming should indeed generate an increase in storm intensity. Emanuel analysed records of tropical cyclones since the middle of the last century, and found that the amount of energy released in these events in both the North Atlantic and the North Pacific oceans increased markedly since the mid-1970s. Both the duration of the cyclones and the largest wind speeds they produced have increased by about half over the past 50 years. These increases in storm intensity are mirrored by increases in the average temperature at the surface of the tropical oceans, suggesting that this warming—some of which can be ascribed to global warming—is responsible for the greater power of the cyclones. As the human population in coastal regions gets ever denser, the damage and casualties produced by more intense storms could increase considerably in the future.
• hurricane eye : the relatively calm area near the center of the storm. In this area, winds are light and the sky is often partly covered by clouds
• landfall : the term used to describe where the hurricane eye actually passes over land, usually used to describe the continental States rather than islands in the Caribbean
• hurricane eye landfall : when the eye, or physical center of the hurricane, reaches the coastline from the hurricane's approach over water
• hurricane path or track : line of movement (propagation) of the eye through an area
• hurricane season : the portion of the year having a relatively high incidence of hurricanes. The hurricane season in the Atlantic, Caribbean, and Gulf of Mexico runs from June 1 to November 30. The hurricane season in the Eastern Pacific basin runs from May 15 to November 30. The hurricane season in the Central Pacific basin runs from June 1 to November 30
• swath : the width of the path of the hurricane. Usually this path area is about 125 miles wide with 75 miles to the right of the eye and 50 miles to the left of the eye
• intertropical convergence zone (ITCZ) : the axis dividing the southeast trades from the northeast trades, toward which the surface winds tend to converge The easterly trade winds of both hemispheres converge at an area near the equator called the "Intertropical Convergence Zone (ICTZ)", producing a narrow band of clouds and thunderstorms that encircle portions of the globe
• maximum envelope of water (MEOW) : describes the predicted areas inundated and amount of storm surge for a particular area during the landfall of a hurricane. Used in the SLOSH Model
• maximum envelope of wind (MEOW) : describes the predicted areas inundated and amount of wind for a particular area during the landfall of a hurricane. Used in the Inland Wind Model.
• typhoons : a hurricane that occurs in the Western Pacific Region of the Philippines or the China Sea. The word is believed to originate from the Chinese word "ty-fung". When there's a cyclone around, you want to know where it is headed. But that may be harder to predict than we thought. 2 simultaneous tropical cyclones can influence one another in ways that are subtle and difficult to forecast, even if they are nearly 2,000 km apart. Weather forecasters currently use computer methods to predict the course of a cyclone - a fierce storm system that gathers around a region of low atmospheric pressure. But when there is another cyclone in the vicinity, small differences in the initial meteorological conditions used in the model can significantly affect the prediction of the first cyclone's path and behaviour 2 days later. The distance over which cyclones can affect each other is surprising. Previous observations of twin cyclones had led experts to suspect that they could not affect one another if they were > 1,500 km apart. A cyclone-prediction model developed by the US Navy was used to forecast the trajectories of 2 real, coexisting cyclones, Katsana and Parma, that formed in the western Pacific Ocean in late October 2003. The predicted courses corresponded fairly closely to those of the real storms over a 2-day period. But a careful look at the results showed that the forecast for Parma depended on that for Katsana, despite the fact that the 2 storms were > 1,500 km apart for most of their lives. (Their closest approach was 1,333 km.) Parma did not, in turn, seem to affect Katsana, which was the stronger of the 2 cyclones. What's more, for cyclone Katsana, the researchers showed that the predictions would have been altered significantly if the model's initial conditions 500 km from the cyclone's centre were changed slightly, showing how surprisingly sensitive these storms can be.Although the forecasts in these cases were good, the tools highlighted the potential for initial errors to grow. 2 cyclones together is not uncommon - it happens about three times every 2 years in the western North Pacific Ocean, and about once every 3 years in the Atlantic. Despite the potential problems for cyclone forecasting, Peng and Reynolds say that their results have a positive spin too. The new technique they used for analysing cyclone interactions, called singular vector diagnostics, seems to give them a better handle on the problem than standard methods. It may tell us how much confidence we can have in our forecasts^ref
The Saffir-Simpson damage-potential scale is a 5-category wind speed / storm surge classification scale developed in the early 1970s by Herbert Saffir, a consulting engineer, and Robert Simpson, then Director of the National Hurricane Center, to measure the intensity of an Atlantic hurricane from 1 to 5. The strongest SUSTAINED hurricane wind speeds correspond to a strong F3 (Severe Tornado) or possibly a weak F4 (Devastating Tornado) value. The scale categorizes potential damage based on barometric pressure, wind speeds, and storm surge. Scale numbers are available to public safety officials when a hurricane is within 72 hours of landfall. Scale assessments are revised regularly as new observations are made. Public safety organizations are kept informed of new estimates of the hurricane's disaster potential. In practice, sustained surface wind speed (1-minute average) is the parameter that determines the category since storm surge is strongly dependent on the slope of the continental shelf. Whereas the highest wind gusts in Category 5 hurricanes correspond to moderate F4 tornado values, F5 tornado wind speeds are not reached in hurricanes.
Those living on the hurricane-buffeted east coast of the USA, or those who sell them insurance, will probably take notice of a model that predicts the annual damage done by such storms. The model, based on wind patterns over the land and sea in July, can predict whether damages will be greater or less than average in the storm season from August to October. Previous models have been able to forecast the number of hurricanes forming at sea, based on variables such as ocean temperature, or fluctuations in air pressure over the North Atlantic Ocean. But even with a completely accurate forecast of hurricane formation it is difficult to say how many will make it to land. A model using data about winds in the troposphere, which extends about 10 kilometres up from the surface of the Earth, from 1950 to 2003 can say whether a year would see damages that were above or below average in 74% of cases. Persistent wind systems, dubbed 'steering winds', have been identified in 6 key regions. Together, these winds seem to determine whether the coming season will be a bad one for US residents. Near Bermuda, for example, unusually high pressures in July can cause onshore winds that consistently blow towards the US coast for the remainder of the summer. Hurricane systems tend to form in the tropical North Atlantic before drifting westwards.. If they encounter a steering wind pointing away from the coast they will swerve harmlessly back out to the ocean. But if the steering winds are pointing towards land, the storms are funnelled on to the shore. The researchers are not sure why the steering winds remain consistent for the entire summer. There's lots of day-to-day variation, but overall the winds do seem to persist for around three months. This will be really important for insurance companies : when faced with the possibility of a catastrophic event, insurers need to decide whether they will buy more coverage or not. Over time, insurers could save 30% on their returns and premiums using this model. In fact, they had a chance to do so in 2004, when a devastating sequence of 4 hurricanes cost insurance firms billions. The researchers used an early version of their model to forecast a worse-than-average season^ref
When hurricane Katrina hit New Orleans on 29 August 2005, the city thought it had escaped the worst. The category-5 hurricane had weakened slightly to category 4 before making landfall, and residents were confident that they had avoided disaster. But the following day, after the storm itself had passed, a 100-m section of the levees protecting the area from the flood waters was breached, along with at least 2 other smaller stretches, inundating some 80% of the city. By 31 August, estimates of the ultimate death toll were in the thousands, and all this with > 2 months to go until the end of hurricane season. Did experts know this might happen? Yes. New Orleans is protected by a series of flood walls called levees that help to hold back nearby Lake Pontchartrain, which in turn is connected to the Gulf of Mexico. Parts of the city sit several metres below sea level. And the system's 565 km of walls were built to withstand only category-3 hurricanes. So a direct strike from a severe storm has long been anticipated as one of the worst natural disasters that could befall the mainland USA^ref. Could something have been done to prevent this? Yes. The levees could have been higher. The New York Times has reported that the estimated cost of protecting against a category-5 hurricane, the highest on the scale, is $2.5 billion. The USA had had a really long run of good luck with hurricanes. Lots of building decisions were made thinking we would continue to have the benign conditions of the 1970s and 80s. The natural marshlands that protect New Orleans from surrounding waters could also have been protected from degradation. A 30-year restoration plan, called Coast 2050, was published in 1998, but it put the bill at a staggering $14 billion. Damages from the current flooding are expected to run to tens of billions of dollars. Part of the problem is that planners did not take into account the recent upswing in hurricane incidence. Unfortunately, 'Don't worry, be happy' is not a very good philosophy for dealing with this kind of thing. How exactly did the levees fail? It is still unclear why the 100-metre section of levee along the 17th Street canal was the one to break. It had recently been upgraded, and was constructed of concrete several feet thick, unlike the earthen structures elsewhere in the city. Experts point out that Lake Pontchartrain was sloshing around in the wake of the storm, which might have caused water to tip over the edge of nearby levees. This water may have eaten away at the foundations of the wall, ultimately causing it to topple. How long will it take to repair the damage? All of the 20-odd pumping stations surrounding the 17th Street canal, the main route by which water is normally pumped out of the city, have been knocked out by the flood. Keeping New Orleans free of water was a daily challenge even before Katrina struck. With almost the entire area between Lake Pontchartrain and the Mississippi River under water, clearing the flood will take at least a month. For now, helicopters and barges are dropping sandbags and concrete highway construction barriers into the largest levee break in an attempt to plug the hole. As this story went to press, the waters were slowly starting to recede. How many people will be affected? Nearby towns already fear death tolls in the hundreds, and the overall number is expected to be in the thousands. It is potentially the worst natural disaster on US soil since the 1906 San Francisco earthquake, which claimed up to 6,000 lives. Besides this, some 11,000 National Guard troops have been assigned to the region to distribute food supplies, rescue those stranded, and quell the looting that has sprung up in New Orleans. Is climate change to blame? It is impossible to say for certain. There is evidence that hurricanes are becoming more intense, but this may be due to natural variation. New Orleans was last hit by a hurricane in 1969, marking the end of a particularly violent couple of decades. This was followed by a relatively quiet patch in Atlantic hurricanes, lasting until 1995. Since then, storms have been heating up again. Hurricanes tend to be stronger when sea surface temperatures in the Atlantic are higher. But data on these temperatures only stretch back a couple of decades, since satellites began to be used to monitor the oceans. And computer models for climate change cannot predict small-scale, individual events such as hurricanes. Nevertheless, sea surface temperatures are predicted to rise by a few degrees by 2100, meaning that devastating hurricanes may become more frequent. Whether these will make landfall or veer out to sea, however, is not known. What can we expect from the rest of this hurricane season? The Atlantic hurricane season traditionally lasts until November, so there could be more in store. So far this year, the region has produced 11 tropical storms, four of which have become hurricanes. The final tally could be around 20 storms with 10 hurricanes. There's no reason to suggest it won't carry on as it has done. The past decade has seen a sudden switch to high activity. Monitoring such storms and evacuating people where necessary remains the best form of defence. Of course, not all hurricanes will home in on major cities to such devastating effect. Katrina probably picked the worst place to come ashore, with the possible exception of Miami. But this week's events may well be a wake-up call. When the armed forces are mobilized in the response to the horrifying conditions along the Gulf coast, there will be a very small number of soldiers who are unaware that they have returned from overseas harboring malaria. There is also an alarming proximity of dengue and yellow fever in Central America. With all the standing water left behind by Hurricane Katrina there will be lots of mosquitoes, and a very small human reservoir of active hosts is all that is required for a large vector population to infect a large susceptible population. It wasn't so long ago that these diseases were endemic along the Gulf coast. But whereas many may think that with such vast areas of standing water, mosquito populations will increase explosively, this is not necessarily the case. In many parts of the world, excessive rainfall or flooding often flushes out larval habitats and can lead to decreased numbers of adults emerging and biting. Mosquitoes have a great diversity of breeding places. Some prefer organically polluted waters (for example contaminated with feces, cadavers, or rotting vegetation); other species breed in domestic containers, drains, water-filled tree-holes, in flooded meadows, swamps and marshes; while still others prefer, or at least tolerate, brackish waters. Often mosquito breeding increases some time after flooding when the water recedes, either naturally or through pumping, as this tends to increase the number of smaller bodies of water, which are often more suitable larval habitats than larger ones. With different mosquito species colonising such ecologically different larval sites, it is difficult to predict which -- if any -- species will proliferate. The CDC believe that there might be mosquito problems and that the potential exists for outbreaks of WNV, St Louis encephalitis, and dengue. For this to happen, however, there must be infected hosts in the area, or later entering the area, which in the case of West Nile includes birds, especially corvids (incidentally, Louisiana is presently a hot spot for WNV), while St Louis encephalitis is also maintained in birds (mainly passeriformes and columbiformes). But there is little information on the effect hurricane Katrina has had, and might have, on local bird populations. In the USA and most of the rest of the world, dengue viruses have no animal reservoir hosts. I do not think that the hurricane will increase the likelihood of yellow fever, as this would require infected people to be in the area, and such people will be viremic only for a very short period -- about 3-4 days. I might point out that yellow fever is not endemic along the Gulf coast, but is endemic in many parts of South America. Nor do I believe there is much chance of malaria transmission from infected troops, because I imagine any soldier with malarious symptoms (such as high intermittent fever) returning from overseas would have been treated with antimalarial drugs. Moreover, infected people would have to be gametocyte carriers. That is because gametocytes are the forms of the parasite found in humans that have to be ingested by a female anopheline mosquito for the mosquito to become infected; then after some 10-14 days, the mosquito becomes infective and thus able to transmit malaria to humans. I understand that past experience has generally shown that there are not epidemics of mosquitoborne infections after hurricanes. For example, I believe that there were no major epidemics of vector-borne infections after the tsunami that hit Asia in 2004. The American Mosquito Control Association is well aware of the catastrophic damage done to parts of Louisiana, Mississippi, and Alabama and the very severe limitations this has imposed on local authorities who normally routinely control mosquitoes in populated areas. Whether or not we believe that vectorborne disease epidemics will appear, I believe control agencies should be as prepared as possible to deal with any emergencies. Mosquito experience with Hurricane Camille in 1969 showed that the hurricanes completely wiped out mosquito populations for at least a month. One can be sure the same thing is happening along the Gulf coast. Eastern North Carolina had the same predictions of mosquitoes emerging from the flood waters of hurricane Floyd in the fall of 1999. Although a considerable amount of aerial spaying was done immediately following the flood (thanks to federal resources), I am not convinced that this prevented any significant arboviral disease outbreaks. Much like the concerns about cholera, typhoid, hepatitis A, etc. during the earlier parts of the flooding, these fears were unfounded. During these large flood events, there is so much water that any microbe, insect or human is easily overwhelmed and often becomes hard to find; but that doesn't mean we shouldn't keep looking. There were no serious mosquito-borne outbreaks post Hurricane Floyd, but that probably there would not have even if  there had been no aerial spraying.  I might add that the Vector Control Research Centre, Pondicherry, India reported that the Asian tsunami created new mosquito larval habitats in the Nicobar and Andaman Islands, when sea water was left behind post flooding. Such saltwater habitats were colonized by a local malaria vector, Anopheles sundaicus. This mosquito, which is widely distributed in Asia, lays her eggs in both fresh and saline waters. It is stated that there are indications that this mosquito increase has had a significant impact on human health, with a 14-fold increase in malaria infection in some regions^{ref1, ref2}. Vibrio vulnificus might have been picked up by people with open wounds who were forced to wade through polluted floodwaters for a long time : 7 cases and 5 deaths have been recorded to Sep 8. It is possible you could have a significant outbreak of West Nile virus (WNV) but we won't likely know for weeks. When hurricane Camille barreled into the eastern USA in 1969, heavy rains and the storm surge wiped out mosquitoes for about a month. Katrina is expected to have a similar effect. Once the mosquitoes' breeding cycle is restored, and eggs begin to hatch in standing water, residents in the Gulf area could be at greater risk for both WNV and St. Louis encephalitis (SLE), another mosquito-borne illness. Widespread spraying of the area might curtail an outbreak. Whether toxic chemicals and oil in the water might hamper mosquito breeding is not known. Neither WNV nor St. Louis encephalitis is passed from person to person. Flooding has probably flushed out mosquito larval habitats, and it is only when water recedes that smaller and discrete water bodies will be formed that are likely to be more attractive to mosquito breeding. Along the Gulf coast, the primary vectors of SLE are Culex pipiens and Culex quinquefasciatus, mosquitoes that often breed in waters that are contaminated with feces or other organic matter, habitats which could become plentiful. Vectors of WNV comprise several mosquito species, including the Cx. pipiens complex. Widespread insecticidal spraying is a possible course of action. This could involve larviciding aquatic waters suspected, or better still known, to be larval breeding places, or adulticiding. Both can be done from ground-based equipment (or boats) or by aerial applications. Depending on where adult mosquitoes are resting, thermal fogs or aerosols can be directed at out of door resting places, such as at vegetation, or indoors (houses and other buildings) where one might find adult females of the Cx. pipiens complex. To spray or not to spray can be a difficult question. It is likely to be costly, and the effects of it may be inconclusive. I wonder what effect the hurricane has had on the local bird populations, remembering that passeriformes and columbiformes are reservoir hosts of SLE, while corvids are reservoir hosts for WNV. It is helpful to refer to Komar et al., "Experimental Infection of North American Birds with the New York 1999 Strain of West Nile Virus, in which it is concluded, with respect to West Nile virus, that "An analysis of our data shows that passerine birds, charadriiform birds, and at least 2 species of raptors (American Kestrel and Great Horned Owl) are more competent than species evaluated from the following orders: Anseriformes, Columbiformes, Galliformes, Gruiformes, Piciformes and Psittaciformes. Indeed, many birds of the latter orders were found to be incompetent for transmission"^ref. Table 10, "West Nile virus reservoir competence index values derived for 25 species of birds"^ref, is particularly instructive. Larval mosquitoes that get flushed into large bodies of deep water become food for assorted predators or die from the increased water salinity.  The adult mosquitoes of the Culex species that are most often associated with the amplification and transmission of WNV typically shelter in low, enclosed areas (e.g., storm sewers, crawl spaces, sheds, hollow logs, culverts, etc.), and they would likely be drowned by the flood waters that came with the storm. It is infected adult mosquitoes (not birds) that are actually the local reservoirs of WNV, because these mosquitoes remain infected until the end of their lives.  Thus, if these older adult mosquitoes are the local source of WNV infection, and they are mostly wiped out by the storm, the local transmission cycle would be broken. The birds that serve as temporary reservoirs of WNV actually include many members of the order Passeriformes, not just crows. Furthermore, birds are not really thought to be reservoirs of the virus; they are amplifying hosts. When birds are infected with WNV, they remain infected during a small window (approximately 4-5 days) during which individuals of some species either remain healthy or become sick or die.  During that window of time, the infected birds may have sufficient viremia to infect mosquitoes. However, by the end of that period, the bird's immune system has either eliminated the virus from the bloodstream, or the birds have died from the virus.  In either case, they would no longer be infectious to mosquitoes. If the hurricane kills most of the local Culex mosquito population, it may take time for new Culex to recolonize the area.  By that time, local birds would no longer be infectious to mosquitoes.  Therefore, it seems unlikely that there would be much WNV circulating in mosquitoes in the hardest-hit areas of the Gulf Coast during the weeks following the hurricane. I agree that adult Culex mosquitoes such as the Culex pipiens complex that  can be vectors of WNV often shelter in many types of man-made constructions, including cellars, but I am not convinced flooding would cause their death by drowning. I would imagine that many would have flown away before their  resting sites became submerged in water. But this is speculation. Regarding birds. Most publications regard birds as reservoir hosts of WNV despite the fact  that -- in common with most rodents and birds infected with an arbovirus -- viremia is short-lived, usually  a matter of a few days.  If you go to most text books and publication WNV is said to have birds as reservoir hosts. Similarly most websites  cite  birds as reservoirs, including the CDC websites. I agree that infective mosquitoes will remain infective until they die, but the duration their infective life may be very short. However, there is evidence that WNV can survive in some hibernating mosquitoes. Ticks are much longer-lived than mosquitoes and certain species seem to be capable of transmitting  WN virus amongst bird populations,  and in such situations ticks as well as birds could be considered reservoir hosts. Yes, many other bird species besides corvids are infected, or have the potential to be infected, with WNV and some species appear to be more important in the transmission cycle than corvids, but this will depend on their local distribution. It is difficult to guess exactly what the repercussions will be in the weeks and months post hurricane Katrina. There are many published papers showing that flaviviruses, such as WNV, Japanese encephalitis virus (JEV), St. Louis encephalitis virus (SLEV), tick-borne encephalitis virus (TBEV), etc. (and other arboviruses), can be isolated from healthy non-viremic birds, and indeed, from other species, such as rodents and bats. This is well known to many arbovirologists globally. Therefore, birds can carry the virus at levels that appear to be below the threshold for easy detection by conventional methods. Whether or not these apparently non-viremic birds reactivate the virus to higher titres under stress or for some other reason has not yet been formally established. In this sense, birds and other species could be acting as reservoirs. Alternatively, in theory at least, only one hungry/thirsty infected mosquito may be needed to survive for the virus to start an epidemic^ref. Many human flaviviral diseases, such as WNV, JEV and SLE etc., require a 2-host life cycle (i.e. bird-mosquito and vise versa) in which humans are secondary victims. The cycle between birds and mosquitoes is considered to be dependent mutually on the 2 species. Among these 2 hosts, which one serves as the reservoir depends on the disease, and sometimes, on one's point of view. Theoretically, only one hungry/thirsty infected mosquito may be required to survive for the virus to start an epidemic. After a major hurricane, such as Katrina, field evidence showed that there were few adult female mosquitoes and an eerie absence of birds. For normally ubiquitous mosquitoes to recover their full population, it may take a month or so, as mentioned by many entomologists. But for the warm-blooded and brainy birds to return to their normal habitats, I wonder how long it will take. I am not an expert in ornithology. Colleagues from this field are welcomed to share their views with us. I overheard that Montserrat Island was
totally silent without birds and bird song right after hurricane Hugo's visit in 1992 (August). With or without a lessened population of birds (migratory or terrestrial) and fewer surviving infective adult female mosquitoes, one would not expect an unusual surge of outbreaks of endemic flaviviral infections in the hurricane stricken area, such as New Orleans and Louisiana more broadly. 2 Air Force planes were scheduled to spray insecticide over about 150 000 acres (about 234 square miles) east and south of New Orleans on Monday night. An area about the same size is to be sprayed tonight north of the city. Mosquito numbers started exploding about 3 days ago. One Louisiana mosquito-control district reported that the number of trapped mosquitoes has increased 800% over pre-Katrina levelsMany of the newly hatched mosquitoes are "floodwater" species that don't normally pose a serious disease threat. Mosquitoes that usually live in tree canopies have been forced to the ground by storm damage, and populations of salt-marsh mosquitoes are booming. They're very aggressive biters, and they can get to such numbers that the power linemen can't work -- even with repellent on. They can't function because there are so many mosquitoes. It's unclear how the number of human WNV cases will be affected. We've never had a hurricane hit when there was a lot of WNV activity happening at the time, so we're not quite sure what's going to happen. Areas hit hardest by Katrina might see little increase in the number of WNV cases. In those places, WNV-infected mosquitoes were either killed or blown away by the storm. And many of the wild birds that serve as a reservoir for the virus are gone as well. It's outside the hardest-hit areas that we would expect the WNV that's in the area to continue. Mosquito populations are exploding on the Gulf Coast at a time when the risk of human exposure to the blood-suckers is very high. Rescue and recovery teams spend all day outside, and some of them camp in tents at night. Some displaced residents are living outside. In storm-damaged houses that remain habitable, lack of air conditioning due to ongoing power outages has forced occupants outside to escape the heat.
During 29 Aug - 11 Sep 2005, surveillance identified 22 new cases of Vibrio illness with 5 deaths in persons who had resided in 2 states. These illnesses were caused by V. vulnificus, V. parahaemolyticus, and nontoxigenic Vibrio cholerae. These organisms are acquired from the environment and are unlikely to cause outbreaks from person-to-person transmission. No cases of toxigenic V. cholerae serogroups O1 or O139, the causative agents of cholera, were identified. This report summarizes the investigation by state and local health departments and CDC, describes 3 illustrative cases, and provides background information on Vibrio illnesses. Results of the investigation underscore the need for heightened clinical awareness, appropriate culturing of specimens from patients, and empiric treatment of illnesses (particularly those associated with wound infections) caused by Vibrio species. No confirmed cases of illness have been identified with onset after 5 Sep 2005; additional Vibrio cases are under investigation. A case of post-hurricane Vibrio infection was defined as clinical illness in a person who had resided in a state struck by Hurricane Katrina (i.e., Alabama, Louisiana, or Mississippi) with illness onset and reporting during 29 Aug-11 Sep 2005, where Vibrio species was isolated from a wound, blood, or stool culture. Among cases, a wound-associated Vibrio case was defined as an illness that likely resulted from infection of a wound or abrasion acquired before or during immersion in floodwaters. Wound-Associated Illnesses: 18 wound-associated Vibrio cases were reported, in residents of Mississippi (7) and Louisiana (5); in persons displaced from Louisiana to Texas (2), Arkansas (2), and Arizona (1); and in a person displaced from Mississippi to Florida (1). Speciation was performed in clinical laboratories for 17 of the wound-associated cases; 14 (82%) were V. vulnificus, and 3 (18%) were V. parahaemolyticus. 5 (28%) patients with wound-associated Vibrio infections died; 3 deaths were associated with V. vulnificus infection, and 2 were associated with V. parahaemolyticus infection. Age of patients with wound-associated illnesses ranged from 31 to 89 years (median: 73 years). 15 (83%) were male. The majority of patients were hospitalized; admission dates ranged from 29 Aug to 5 Sep 2005. Not all patients were initially hospitalized because of their wounds. An underlying condition that might have increased risk for severe Vibrio illness was reported in 13 (72 percent) of the patients with wounds; these conditions included heart disease (7 patients), diabetes mellitus (4), renal disease (3), alcoholism (3), liver disease (2), peptic ulcer disease (1), immunodeficiency (1), and malignancy (1). Non-Wound-Associated Illnesses: 4 persons were reported with non-wound-associated Vibrio infections (2 in Mississippi, 1 in Louisiana, and 1 displaced from Louisiana to Arizona). Information on the Vibrio species and clinical illness was available for 2 of these patients; the species were nontoxigenic V. cholerae isolated from patients with gastroenteritis. One of the infections occurred in a boy aged 2 months with diarrhea whose stool culture yielded both Salmonella group C2 and V. cholerae non- O1, non-O139. He was hospitalized for 2 days in Mississippi. The other V. cholerae non-O1, non-O139 isolate was from a stool specimen from an adult who was not hospitalized. No deaths were associated with the non-wound cases. After natural disasters such as Hurricane Katrina, the risk for illness related to infectious diseases is a public health concern. The findings in this report describe illnesses caused by Vibrio species, including wound infections resulting from post-hurricane exposure of wounds to flood waters. These findings underscore the need for prompt recognition and management of Vibrio wound infections by healthcare providers. When the number of illnesses from infectious diseases increases after a natural disaster, they usually are caused by infectious agents normally present in the community or local environment (Blake PA: Communicable disease control. In: Gregg MB, ed. The public health consequences of disasters. Atlanta, GA: US Department of Health and Human Services, CDC; 1989:7-12). Nationwide, an average of 412 cases of noncholeragenic Vibrio (all Vibrio species other than toxigenic V. cholerae O1 or O139) illnesses were reported each year during 2000-2004, including an average of 146 cases reported from the 5 Gulf Coast states^ref. The most frequently reported Vibrio species are V. parahaemolyticus, V. vulnificus, and nontoxigenic V. cholerae. Vibrio illnesses in the USA are seasonal and peak during the summer (Figure 2). During 2000-2004, in the month of September, an average of 14 (range: 11-18) noncholeragenic Vibrio infections were reported from Gulf Coast states; an average of 7 cases (range: 4-8) were wound-associated. Except for toxigenic V. cholerae O1 or O139, Vibrio illnesses are not nationally notifiable in the USA, and the actual number of noncholeragenic Vibrio illnesses is likely greater than the number reported. Cholera is a severe diarrheal illness caused by V. cholerae serogroups O1 or O139, which produce cholera toxin (i.e., toxigenic V. cholerae O1 or O139). A small endemic focus of toxigenic V. cholerae O1 exists in the Gulf of Mexico^ref. During 2000-2004, a total of 16 cases of cholera were reported in the USA, and 13 (81%) of these infections were acquired during overseas travel or by consumption of imported seafood. Only 3 (19%) infections were acquired in the Gulf Coast states, all in the year 2000. Therefore, the risk for acquiring cholera associated with Hurricane Katrina is extremely low. Since 2000, at least 7 noncholeragenic Vibrio species (V. vulnificus, V. parahaemolyticus, nontoxigenic V. cholerae, V. alginolyticus, V. fluvialis, V. mimicus, and V. hollisae) have been reported as causing illness each year in the USA. Although these organisms and those that cause cholera are grouped together under the genus Vibrio, they cause distinctly different illnesses. In the USA, noncholeragenic Vibrio usually are either foodborne, (e.g., resulting from eating raw or undercooked shellfish, particularly oysters, or other contaminated foods) or wound-associated (e.g., resulting from exposure to seawater or brackish waters where the organism naturally occurs). The incubation period for noncholeragenic Vibrio infection usually is 12-72 hours but can be as long as 1 week^ref. Noncholeragenic Vibrio illnesses are not transmitted easily from person to person. Outbreaks, which are rare, usually are the result of consuming contaminated shellfish. The most frequently reported post-hurricane Vibrio illnesses were V. vulnificus and V. parahaemolyticus wound infections. These cases represent an increase over the normal reported incidence of Vibrio wound infections in Gulf Coast states and are consistent with exposure after hurricane landfall. Although precise exposure histories are not yet available for all patients, the infections caused by V. vulnificus likely resulted from wounds exposed to flood waters among persons with medical conditions that predisposed them to Vibrio infections. No evidence has been found of increased Vibrio gastrointestinal illness. V. vulnificus wound infections can begin as redness and welling at the site of the wound and rapidly progress in patients at high risk to cause systemic illness, including sepsis. Whether acquired through wound infection or ingestion, V. vulnificus typically causes a severe and life-threatening illness characterized by fever and chills, decreased blood pressure (septic shock), and blood-tinged blistering skin lesions (hemorrhagic bullae). Persons with chronic liver disease or immunocompromising conditions are particularly at risk for severe V. vulnificus infections^ref (Tuttle J, Kellerman S, Tauxe RV: The risks of raw shellfish: what every transplant patient should know. J Transpl Coord 1994; 4:60-3). V. parahaemolyticus typically causes gastroenteritis after consumption of contaminated shellfish. Less frequently, V. parahaemolyticus causes wound infections that are generally less severe than V. vulnificus wound infections. However, in persons with liver disease or immunocompromising conditions, V. parahaemolyticus wound infections can lead to death. Nontoxigenic V. cholerae causes primarily gastroenteritis, but unlike toxigenic V. cholerae O1 or O139, nontoxigenic V. cholerae do not cause epidemics. Illness caused by this organism ranges in severity from mild diarrhea to severe watery diarrhea. Fever and bloody diarrhea are not typically observed. Immunocompromised persons and persons with liver disease can experience a more severe illness, including fever, chills, and septic shock. This organism has rarely been reported to cause wound infections. Vibrio infections are diagnosed by culture of wound, blood, or stool specimens. For stool specimens, a selective medium of thiosulfate-citrate-bile salts-sucrose agar (TCBS) is recommended. If clinical suspicion of enteric Vibrio infection exists, the microbiology laboratory should be notified so that TCBS media will be used. Clinical laboratories should send all Vibrio isolates to state public health laboratories for confirmation. CDC continues to work with local and state public health officials to investigate post-Katrina Vibrio illnesses. Persons working in hurricane-damaged areas, especially in areas with standing brackish water, should wear boots and other protective gear to prevent wounds and to prevent exposure of broken skin to contaminated water. To prevent Vibrio infections, persons with open wounds or broken skin should avoid contact with brackish water or seawater, especially if they have preexisting liver disease or other immunocompromising conditions. Injury prevention is especially important for persons in these high-risk populations. Healthy persons are at much lower risk for Vibrio infection. In areas where flood waters have receded and surfaces are dry, Vibrio should not be a concern because the organism is killed rapidly by drying (In: Mitscherlich E, Marth EH, eds. Microbial survival in the environment. New York, NY: Springer-Verlag; 1984:515-34). To reduce the risk for Vibrio wound infection, persons should wash all wounds that have been exposed to sea or brackish waters with soap and clean water thoroughly as soon as possible and seek medical care for any wound that appears infected. Clinicians should be vigilant for Vibrio infection in hurricane evacuee populations, particularly in patients with infected wounds and especially if the patients are in a high-risk group. If V. vulnificus is suspected, antimicrobial therapy should be initiated immediately; prompt treatment can improve survival. Antimicrobials effective against Vibrio infections include doxycycline, 3rd-generation cephalosporins (e.g., ceftazidime), fluoroquinolones, and aminoglycosides (Daniels NA, Evans MC, Griffin PM: Noncholera Vibrios. In: Scheld WM, Craig WA, Hughes JM, eds. Emerging infections 4. Washington, DC: ASM Press; 2000:137-47). Wound infections also should be treated with aggressive attention to the wound site; amputation of the infected limb is sometimes necessary^ref. 3 documented V. cholerae (all non-O1, non-O139) infection related to the hurricane occurred^ref. Vibrio cholerae is not endemic in Louisiana, but more pathogenic non-cholera Vibrios are, and they have already killed and will continue to do so. The ARBOR diseases will return soon, principally West Nile virus -- worse than in Mississippi -- and spraying will be required later when the floodwaters recede. Current mild diarrheal diseases are viral and secondary to poor sanitation and endemic RNA enteric viruses. More serious dysenteric disease outbreaks could follow among those who are not evacuating flooded areas and who are consuming contaminated food and water. Dysenteric diseases should not be a problem in well-run shelters. Baton Rouge will probably have a hepatitis A outbreak in 4-6 weeks so get vaccinated for HAV now, because there are inadequate stocks of IgG and HAV vaccine to respond to an outbreak. As we enter flu season, viral URIs will become a problem among the elderly in shelters and may result in community acquired pneumonias. Since many TB⁺ homeless persons and ex-prisoners may have been sheltered with the elderly, infants, and the immunosuppressed, MDR-TB could be transmitted to these susceptible populations. Reactivation of non-MDR TB in the elderly is usually more of a problem, especially among older immigrants from countries where TB is endemic (Viet Nam, Mexico, etc.) and primary infections were acquired during childhood. The elderly often baby sit the infants, who are highly susceptible to TB. TB should not be as much of a problem as viral URIs and secondary pneumonias. In the past, there have been outbreaks in southern Louisiana of V. cholerae "El Tor" of a specific phage type common from Apalachicola Bay to Galveston Bay, and usually south of Interstate highway 10 (I-10). We used it in a trial involving frog legs: see Sang, F.C.; Hugh-Jones, M.E.; Hagstad, H.V. 1987. A Research Note: Viability of Vibrio cholerae 01 on Frog Legs under Frozen and Refrigerated Conditions and Low Dose Radiation Treatment. Journal of Food Protection, 50:662-664. We obtained the culture from the NO PH Laboratory which had recovered it from some prior outbreak. It was never very virulent in the various outbreaks, and its recovery depended more on clinician awareness than anything else. I haven't seen it reported for some time now. Those outbreaks followed a lack of rain and thus presumably salt-water intrusion into the oyster bays and blue crab trapping areas. It was largely south of the I-10, presumably because of the Cajun propensity to under cook their crabs. V. cholerae non-O1 strains are still recovered occasionally. But, in reality, Jim is absolutely correct about the present status of cholera in Louisiana.
In the wake of Hurricane Katrina, experts are hoping that the natural marshland that buffers the Gulf coast can be restored. The lack of a healthy ecology in the area increased the damage from the August storm, and should serve as a lesson for other areas that need protection. It has long been known that the sandy barrier islands and marshy bayous of the Louisiana coast are capable of acting as a wave-deflecting and energy-absorbing one-two punch. These natural features are in decline however; more than 60 km² of land erode each year. And there is little mud and silt filtering into the area to replace the eroded material: modern agriculture holds on to silt, as do dams, so the muddy Mississippi is no longer quite so murky. Frequent low-intensity storms have battered the area over time, washing away much of the remaining land. Several of the barrier islands off the Louisiana coast have been washed away altogether by Katrina. Researchers are not sure that things would be different in New Orleans had the delta still had its Delaware-sized swathe of wetlands. But the communities along the Gulf coast would have been much better off. Smaller storms on the same areas 50 years ago: you've got people who as kids never saw water in their yard, and now there's water in the yard and covering the road. Without a doubt, our coastal landscape used to protect us from storms. Similar protection from natural landscapes have been noted elsewhere. Healthy bands of coral reefs or mangroves, for example, saved some areas from the damage wreaked by the tsunami in the Indian Ocean in December 2004; degraded areas fared worse. Experts caution that coastal protection in that area is vital. Bangladesh is losing its coasts much as we are, and they have typhoons equivalent to our hurricanes. A 1998 document called Coast 2050, written by state officials, called for significant investment in wetlands restoration, not only for the good of fisheries and endangered species, but for buffering hurricane storm surges. The state's Department of Natural Resources says it will act on the top priorities of the plan, depending on funds awaiting approval or appropriation in Congress in 2005. Funding is likely to be far short of the $14-billion estimated cost of the whole plan; observers say they are more likely to get $1.9 billion over 10 years. Ecosystem services are a hard sell in Congress, let's face it. It is not clear whether the disaster will focus attention on the area and liberate funds, or if money will be spent instead on immediate rebuilding. We need sustained funding for our chronic erosion problem : the scientists are really frustrated that it has been ignored. I think in many ways you need a catastrophe for the world to sit up.
Hurricane Rita , with its winds of 270 km/hr, battered Texas on Sep 2005 : up to 1 million people were evacuated from Houston and Galveston County. Many oil rigs and refineries around the coast have been abandoned, threatening to send US oil prices soaring. Meanwhile, NASA has closed the Johnson Space Center in Houston, and transferred International Space Station operations to Russian mission control in Korolev, near Moscow. After Hurricane Katrina, which devastated New Orleans and its environs on 29 August, the 6-man team of specialist forecasters at the National Hurricane Center in Miami, Florida, are under more pressure than ever to predict where Rita will make landfall. They're working harder than the normal 40 hours a week, and doing extra shifts. Their shifts are 8-9 hours, and because we provide 24-hour coverage many of them work overnight. How do they come up with the forecast? When the specialists start their shifts they're briefed by their predecessors, and then start collecting data from ships and buoys in the Gulf of Mexico. They also collate information from aircraft reconnaissance flights, and get updated satellite images every 30 minutes. All of this gives them a wind speed and direction, sea heights and atmospheric pressure. They also have a conference call with other meteorologists around the Gulf to discuss the data that will go into computer models that predict the hurricane's path. The specialist uses at least a dozen hurricane models, and sometimes results diverge. The models can take a system north or west, every which way. The skill of the specialist is to pick the most likely outcome. From start to finish it's about a 3-hour process to produce a forecast. The latest track has it hitting the coast of central Texas, about 160 km below Houston or so. We're predicting landfall by 7 am local time (11 am GMT) on Sat 24 Sep 2005. Typically our forecasts are very accurate 3 days in advance, although right now the cone of uncertainty covers most of the Texas coast. Is there any chance the storm could peter out before it reaches the coast? Rita looks like a category 5 at the moment. Although storms don't tend to maintain that intensity for long, it's definitely going to strike land as a hurricane. Even if it goes down to a category 1, with winds > 118 km/hr, that's certainly strong enough to do damage. Katrina was a category 1 when it hit southeastern Florida, and it uprooted trees and damaged cars and houses here. How certain can you be? Forecasting the tracks of hurricanes has become a lot more accurate over the past few years. For Katrina we got it almost exactly right. Direction forecasting has been improved by better computing power, more data from ships and buoys to put into the models, and advances made in those models by researchers. Intensity forecasting is another matter. At the moment it's difficult to predict what category hurricanes will be when they hit land. We're trying to improve that by getting the National Oceanic and Atmospheric Administration and Air Force airplanes that fly over the storms to make observations, to take pressure readings and also drop sounders into the hurricane to get measurements from inside. Has this hurricane season seen record activity? I think this is the fourth most active Atlantic season on record. The hurricane season ends on 30 November, but activity will peak around 20 October, so it's more than likely there will be 2 or 3 more named systems by the end of it. Is it unusual for Texas to be in the firing line? It's been a while since a major hurricane reached Texas. The last one was Hurricane Carla, which hit in 1961. They're certainly infrequent. [Hurricane Carla was a category-4 storm that caused > 30 deaths. Arguably the worst US natural disaster was from a hurricane that hit Galveston, Texas, in 1900, killing at least 8,000 people.] What do the forecasters do when the hurricane season ends? We issue forecasts for all Atlantic and eastern Pacific tropical storms year round, but after November they tend to be marine warnings for shipping. Late in Aug 2005, a Houston medical school finally got a US$50-million insurance payout for damages it sustained during a tropical storm 4 years ago. The money, ironically, comes just in time for the Texas Medical Center to face up to another storm: Hurricane Rita is due to strike this weekend. At least this time they'll be better prepared recalling lessons they learned in 2001 from the tropical storm nicknamed Allison. That storm flooded the multi-institutional centre, drowning research animals in basement labs and causing priceless tissue and cell samples to thaw and be destroyed. All told, Allison wreaked $2 billion in losses. The affected institutions are only just starting to recover this money. Since that disaster, the medical centre has spent millions of dollars to prevent future storm damage. Flood gates have been installed and auxiliary power generators moved to higher floors. The measures may well come into play during the evening of Friday 23 September and the following morning, when Rita is expected to slam into the Texas coast. By Friday morning, most of the low-lying parts of Houston had been evacuated. Traffic jams snarled the interstates leading north from the city, away from the oncoming storm. At the Baylor College of Medicine, part of the Texas Medical Center, researchers and students had cleared out. Susan Berget, vice-president of academic planning, was one of several officials who planned to hunker down on the campus to ride out the storm and assess the damage afterwards. Following the 2001 storm, Berget learned firsthand how ill-prepared the US Federal Emergency Management Agency (FEMA) was for research emergencies. They had never thought of how to deal with a research loss. To them, transgenic mice is a foreign concept. Houston scientists have long talked of the consequences of tropical storms, but before Allison they had taken little action. When oncologist Kent Osborne moved from San Antonio to Baylor in 1999, taking with him the world's largest bank of breast-cancer tumour specimens, he was concerned about flooding. The collection contained about 100,000 specimens and was later valued between $100 million and $300 million. When Baylor proposed putting the bank's 30 massive freezers in the flood-vulnerable basement, we put up a fight. But the freezers weighed so much they couldn't go on higher floors. They stayed down below until Allison's flood waters tossed them around like toys. Research losses are usually only partly covered by insurance. After a natural disaster, FEMA pays 75% of certain losses; under such terms, it contributed $4.5 million for Baylor to buy 2,000 breast-cancer specimens to help rebuild its bank. Allison prompted other Houston universities to make major changes. We have nothing that can be destroyed by water on our lower floors. Their preparations will undoubtedly be tested this weekend by Rita. In New Orleans, the lessons of Allison apparently failed to sink in. After Hurricane Katrina flooded the city on 30 August, the medical centres at Louisiana State and Tulane universities lost their auxiliary generators. The backup power supplies had been located on the ground floors and were easily knocked out by flooding; countless research samples were lost. Allison was a wakeup call Hurricane Rita's approach toward the Texas coast has raised the specter of a long-ago killer storm, which struck before meteorologists were able to track hurricanes with radar and weather satellites. On September 8, 1900, a powerful hurricane buried the thriving port city of Galveston in a storm surge of almost 16 feet (5 m). Conservative estimates put the death toll at 6,000, but 8,000 or more probably died. The unnamed hurricane that nearly scraped Galveston off the map 105 years ago was very similar to Hurricane Rita. The 1900 hurricane is thought to have packed winds of 130-140 miles/hour (210-255 km/hr), which would make it a Category 4 hurricane on today's Saffir-Simpson scale. As of 5 a.m. today, Hurricane Rita's strongest winds were blowing at 140 miles/hour. Over the past 2 days, millions of residents along the Gulf Coast have been streaming inland to avoid Rita, which is expected to make landfall Saturday near the Texas-Louisiana border. Unlike today's evacuees, however, Galveston residents in 1900 had no idea what was in store for them as the storm drew near their island city of 37,000. Ida Smith Austin, who survived the hurricane, said she knew bad weather was expected, but she and her husband didn't alter their plans. "A storm had been predicted for Friday night of September 7, but so little impression did it make on my mind that a most beautiful and well attended moonlight fete was given at our home Oak Lawn that night," Austin wrote in a letter dated November 6, 1900. The following day, as the hurricane drew nearer, the storm surge began to cover Galveston. "In a few minutes, we heard the lapping of the salt water against the sidewalk, and then it slowly crept into the yard," Austin wrote. "In an incredibly short time, the water surged over the gallery driven by a furiously blowing wind." The storm surge crushed buildings and pushed them into a huge pile. The debris became like a giant bulldozer blade, knocking down more buildings as the surge moved across the island. After the storm, Galveston residents were determined to rebuild their city and prevent a recurrence of the tragedy. The buildings that survived were raised, and sand from the Gulf of Mexico was pumped onto the island to lift it 8 feet (2.5 meters) above sea level. A 17-foot-tall (5-m-tall) seawall also was built to protect Galveston from storm surges. Galveston leaders a century ago probably thought this dramatic effort would protect the city from whatever the Gulf of Mexico could throw at it. But they didn't envision a storm like Rita. Depending on where Rita makes landfall, hurricane forecasters fear the storm could bring a surge of more than 20 feet (6 m) into Galveston Bay, which would easily overtop the city's seawall.
Cold War nuclear-bomb testing may have a beneficial legacy: helping forensic experts to estimate more accurately the age of dead bodies. Radioactive remains from the many blasts detonated during the 1950s and 60s are present in the tooth enamel of people all over the world, allowing examiners to work out when the teeth were formed. The method, developed by researchers at the Karolinska Institute in Stockholm, Sweden, has already helped Swedish police to identify 6 of that country's victims from 2004 Asian tsunami. And it may help with the task of putting names to the victims of Hurricane Katrina. The technique works because a burst of nuclear testing in the 1950s, chiefly by the USA and Russia, released huge amounts of the radioactive isotope carbon-14 into the atmosphere. This was subsequently taken up by crops and other plants, thus entering the food chain. This bomb carbon is not dangerous, but it is traceable. Since the signing of the Partial Test Ban Treaty in 1963, which banned above-ground nuclear detonations, atmospheric levels of C-14 have been dropping at a known rate. And because tooth enamel is not replenished after a certain age, the carbon in that enamel bears the hallmarks of the time when it was made. Someone in middle age, whose teeth were formed in the 1960s, will for example have higher levels of C-14 in their enamel than a teenager, whose teeth were made when there was less carbon-14 around. The dating method can judge a person's age to within 1.6 years^ref. This is an improvement on existing methods, such as examining skeletal remains, which is thought to be accurate to only 5-10 years. What's more, the new method gives an actual birth date for the body, rather than simply an age at death. This could be helpful in situations when it is unclear how long the victim in question has been deceased. An accurate estimate of a body's birth date could be a great help in identifying victims of disasters such as the Asian tsunami or the genocide in the former Yugoslavia. The alternative method of DNA profiling requires exhaustive analysis of tissue samples, many of which simply no longer exist. Teeth are far easier to preserve for later examination. Investigators struggling to identify bodies left behind by Hurricane Katrina may also adopt the method. Victims of drowning can sometimes be hard to identify thanks to water damage, and the flooding may have destroyed some records traditionally used for identification. In situations like New Orleans they will be having huge problems. In many cases dental records will have been washed away. As the flurry of Cold War testing recedes into the past, atmospheric carbon-14 levels will eventually tail off to a level where the method will no longer work, but that is several decades away. (1963 nuclear testing treaty)
Possible links between hurricane formation and global warming are a contentious issue in climate policy. And last week's Hurricane Katrina in the United States has fanned the flames. The depth of the divide between supporters and sceptics became apparent in January, when US meteorologist Chris Landsea resigned from the Intergovernmental Panel on Climate Change (IPCC). He was protesting against statements made by his panel colleague, Kevin Trenberth, who had supported a link between warming and storms in a press conference. Trenberth, head of the climate analysis section at the National Center for Atmospheric Research in Boulder, Colorado, tried to set the record straight in an article published in Science in June^ref.  Owing to large natural variability, trends in hurricane frequency are indeed difficult to attribute to climate change, but human influences on the environment probably affect hurricane intensity and rainfall. Trenberth's view is supported by the most recent and solid analysis of hurricane destructiveness over the past 30 years, by leading US hurricane researcher Kerry Emanuel of the Massachusetts Institute of Technology in Cambridge, Massachusetts. In his Nature paper^ref, Emanuel concludes that "future warming may lead to an upward trend in tropical cyclone destructive potential, and ... a substantial increase in hurricane-related losses in the twenty-first century." The main cause of all this may be increased sea surface temperatures; these are the most important variable affecting hurricane formation. Extra warmth means there is more energy available to a storm, and more water is likely to be sucked up into the clouds. North Atlantic surface temperatures have been significantly above average for at least a decade, a trend many scientists agree must be associated with global warming. In early August, the Gulf of Mexico, was a striking 2-3 °C warmer than it usually is at this time of year. Katrina sucked out so much heat energy from the Gulf that water temperatures dropped dramatically, in some regions from 30 °C to 26 °C, after the storm had had passed. Sea temperatures are likely to remain high until October 2005. The IPCC came to the tentative conclusion that hurricanes might be on the rise in their most recent assessment report, published in 2001. It stated: "There is some evidence that regional frequencies of tropical cyclones may change.... There is also evidence that the peak intensity may increase by 5% to 10% and precipitation rates may increase by 20% to 30%. But the panel adds that the certainty of these statements is low: lower than our understanding of air temperature changes, for example. There is a need for much more work in this area to provide more robust results. Naturally, the bulk of the media are less reserved. "The hurricane that struck Louisiana was nicknamed Katrina by the National Weather Service. Its real name is global warming. Such statements are not pure invention, but most scientists are made uncomfortable by this specific kind of attribution. Even in the presence of a statistically robust trend, it is unscientific to attribute a discrete atmospheric event to climate change. No matter what one's opinion is on global warming and hurricanes, you shouldn't score cheap points by turning a scientific question into a political one. Attempts to use hurricanes to justify energy polic