Thursday, June 5, 2008

Cigar Lighter Receptacle How To Replace

Gestione dei disastri: anatomia e tendenze




Terminology and definitions

As regards terminology, emergency management (disaster management ) is a minefield. Academics and practitioners have completed at least 7 schools of thought. [1] Since each of these schools have maintained a certain detachment from others, communication between disciplines was inhibited by the use of a diverse range of definitions of key terms such as 'danger' ( hazard ) , 'vulnerability' and 'risk'. In addition, persino il nome del campo è soggetto a variazioni da un ambiente accademico all'altro. Nella Gran Bretagna, per esempio, 'disaster management' riferisce a diversi aspetti della protezione civile, la gestione dei rischi e le operazioni di scopo umanitario, e il nome non significa soltanto la gestione di eventi estremi e dannosi.

Un effetto dello scopo straordinariamente largo della disaster management è che gli esperti non riescano a mettersi d'accordo sulla definizione dei termini. Malgrado alcuni tentativi di unire le due fazioni, rimane un grande schisma tra le scienze naturali, fisiche e di costruzione, da una parte, e le scienze sociali dall'altra. In un commento su questo fenomeno il grande sociologo E.L. Quarantelli ha notato nell'introduzione of his book What is a Disaster? that:

"If scholars in this area do not agree on whether a disaster is fundamentally a social construction or a physical event, the field has clear intellectual problems" (Quarantelli 1998, p. iv).

However, in a previous work he argued that the debate is healthy:

"taking care of what does the term 'disaster' means not getting distracted by an academic exercise with no results. It is instead a way to focus in a fundamental way on what should be considered important and significant ... " (Quarantelli1995, p. 225) [2]

The purpose of this report is not to investigate the meaning of a disaster or to add material to the debate on terminology. However, since the identity of any future initiative depends to a certain point the name used, it is worthwhile to conduct a quick review of terms and how they are interpreted.

Disaster management is preferred in Britain but the world is not accepted as an umbrella term. In its favor, is a descriptive term easily interpreted with emphasis on practical approach that is committed when the events should be managed. On the other hand, is too narrow, since it does not embrace the planning of interventions related disasters, a process that must necessarily precede management. Furthermore, the term does not embrace emergencies in the broadest sense, and some emergencies require special handling but do not become disasters. Finally, the term does not describe the broader approach of mitigation and preparedness and recovery after events.

Emergency preparedness (emergency preparedness) is the term commonly used in the U.S. and is used when a disaster management would be preferred in Britain. This term has the advantage of embracing all five phases of the 'disaster cycle' (mitigation, preparedness, emergency operations, rehabilitation and reconstruction) and a more size range extends from small emergency event that a major catastrophe. However, not emphasis on disaster management as a task preeminente.

Civil Defence is used in many parts of the world to cover all activities to prepare for disasters, manage and facilitate recovery in short expiry. The term is current in Europe, Russia, Canada, Latin America and the Caribbean. It has the advantage of being easily translated into French, English and Italian. In this sense it is a true umbrella term for its field of reference. However, it can not be easily applied to the management of humanitarian aid, unless it takes the form of mutual aid between adjacent states. Nevertheless, many of the countries which today are humanitarian operation theaters tomorrow will have their own services and civil protection systems. Eventually, there will be a convergence in this area between the rich world and the poor countries. The civil defense

was the ancestor of civil protection. She was born in precautions against air raids of World War II and the Cold War has developed into a system of measures to protect civilian populations against governments and armed aggression by foreign powers. The facts have shown that many civil defense systems were not only costly and ineffective, but also capable of being expropriated for subversive reasons. They could in fact be used to defend the government against the people who ruled, in support of coups and to suppress democratic rights. Civil protection is a system more open and participatory, and one that covers a much wider range of contingent events, including natural and technological disasters. However, the effect of armed renewed emphasis on aggression (ie terrorism), as is developing in the United States, is to refuse protection to the civil defense the previous calendar year. Most experts in this area, including many from North America, see this event as a retrograde step.

Homeland Security è un termine che non ha guadagnato molto spazio fuori degli Stati Uniti. Esso abbraccia tutte quelle attività che contribuiscono alla sicurezza domestica, e quindi di coinvolgere l'industria della sicurezza, un settore in forte crescita, nonchè le attività anti terrorismo e le preparazioni per i disastri naturali e tecnologici. Comunque, non è il termine giusto quando, ad esempio, l'impatto dei disastri naturali è più profondo di quello delle violazioni di sicurezza o gli oltraggi terroristici.

La pianificazione e gestione delle emergenze ( emergency planning and management ) è probabilmente una frase più accurata degli altri in senso descrittivo, ma è troppo lungo per essere stato adoperato largamente. Per di più, esso viene iterpretato in diversi modi da diversi gruppo e quindo non dispone di un consenso su esattamente quanto del campo copre.

Gestione delle crisi ( crisis management ), pianificazione delle contingenze ( contingency planning ) e gestione dei rischi ( risk management ) sono termini che possono essere usati largamente e liberamente, ma in genere vengono usati specificamente nel contesto del controllo delle crisi e abbassamento dei rischi industriali e commerciali. Quindi, sono di valore troppo ristretto per fungere come termini ombrello per l'intero campo.

In sintesi, il campo continuerà ad essere chiamato disaster management in questo documento perché questo termine descrive il gruppo fondamentale di attività associate con la protezione civile e la preparazione per le emergenze. Comunque, bisogna notare che per implicazione la 'disaster management' include anche la gestione del rischio, la riduzione della vulnerabilità e la pianificazione delle emergenze . Infatti, molto più che nella gestione delle aziende e degli organi di governo, nei disastri ciò che serve è l'applicazione di procedure formulate e sperimentate prima dell'evento. Qualsiasi altra cosa e si scende nell'improvvisazione, un fenomeno che deve essere ridotto al minimo.

I protagonisti della protezione civile

Avendo considerato il 'nome del gioco', bisogna definire i protagonisti ed i partecipanti nelle attività di protezione civile. A livello di autorità locale, cioè quello più basilare nell'ottica della gestione delle emergenze, ci sono essenzialmente due tipi di professionista. Il 'disaster manager' tradizionale è bianco, maschio, di media età e reduce di una carriera militare. Egli manca dei titoli accademici di specifica rilevanza a questo campo, sebbene abbia probabilmente preso alcuni corsi brevi nel settore. Un variante di questa figura è l'attuale o ex-membro dei servizi di emergenza che ha scelto una carriera nella gestione delle emergenze, o come un secondo compito o come una new direction in his career.

This figure, rather than stereotype, is gradually replaced by that of a younger person, male or female, of provvenienza variable, which has taken a degree or a master in this area and it is facing as a first career. While the first group is full of people who are poorly trained academically but have lots of experience, the top newcomers in the second group lacks experience, but to a certain extent able to compensate with their training (and eventually succeed in buying experience through their work). Of course, there's nothing wrong with disaster managers as a character from the ex-military or emergency services. What is needed is to achieve a balance between education, training and work experience that will ensure a sufficient level of professionalism when trying to reduce risk, create contingency plans and manage extreme events.

Disaster management in Europe

In many ways, emergency management is well developed in Europe, but there are some common defects, as illustrated by the following list:

- the lack of a robust and comprehensive system for encouraging common methods and exchanges between the different agencies that have responsibilities for emergency management;
- often, the emphasis on contingency planning is inadequate;
- no one learns enough to other countries and there are not enough links between the civil protection systems of different countries
- the civil protection systems currently in force in Europe carry a legacy of outdated methods and attitudes obligations;
- a culture of inflexible bureaucratic obstruction Journal segrettezza and reduces the chances of making the disaster management something more democratic and to increase public participation in preparing for emergencies;
- some you can not identify the source of wisdom and experience, and exploitation;
- it seems that some of the protagonists of European civil protection have become more involved in the management of their own reputations rather than opens themselves from disasters;
- with the rapid evolution of the field and the various forces and tendencies of centrism and subsidiary legislation in this field is often lacking or contraditoria.

These defects are examined by a lack of direction and encouragement from the European Union. In the field of civil protection the EU has committed its member states enough direction, coordination, risorese, and advice. In Brussels, the responsibility for the sector is divided among several directors, including those dealing with nuclear accidents, environmental and industrial hazards. Again, there is between the bureaucracies of the European Union or the European Commission needs a proper system for the exchange of knowledge, experiences, techniques and management methods. Yet there are ample signs that the European countries look to the EU and the EC for general guidance in this area, a need that is expected to grow when the 15 member states become 23.Malgrado these negative comments, there are some signs that the community European Union is taking note of the situation as regards the risks of disaster and that it wants to improve the preparations. It is commonly said that Europe has undergone major changes in civil protection in the aftermath of September 11, 2001. However, there was a delayed response that has slowly gained strength. In this we may distinguish between negative events that have "pushed" the system to change (Ie, the recent disasters) and the effect of "dragging" of technical and organizational innovations.

in the negative, the impact of disasters has become increasingly serious. Over the past five years the storms and floods, for example, have repeatedly caused serious damage in Europe in chaos. These disasters are becoming increasingly expensive and are charging more and more systems of insurance against damage and accidents and environmental management and regional systems of emergency management. In Europe, the recent floods of the rivers Rhine and Oder have not only caused great damage and loss but also getato ilm light on the need to increase international coordination of emergency management. The SARS outbreaks of foot and revealed the possibility that Europe will suffer in future disasters, major epidemics and disease. The inability of the British to manage well the recent outbreak of foot and could be repeated in other European countries.

also the ability to handle the terrorist Attacks may be less than the potential requirements. Few anti fire and hospital services that are able to manage well a biological terrorism attack. Despite decades of experience with terrorism in Europe (Irish and Basque separatism issues, the "leaden years", etc..), The public is still largely involved in the management of the new Community and national security threats posed by new emerging forms of terrorism.

More positively, the emergency management in Europe is gradually acquiring a greater sense of direction. A summit level, the scope and seriousness of the threats have been widely recognized. The scenarios of hazard, vulnerability, risk and emergency response have been completed. European bonds will eventually be established. In the meantime, risk management and contingency planning are gaining ground in professional way. For more., Security has become a large and important issue in the commercial world as it is in the life of the community and in government circles.

Disaster management in the rest World

Some of the problems and delays are common in Europe in the field of civil protection in the rest of the world. There is a lack of common standards, generally accepted, for emergency planning, disaster management, education and training. Traditionally, the field exists with minimal funds. Once finished, the emergencies are forgotten by the public and politicians that do not return. As a result, the field has been developed very occasionally. Major disasters serve as catalytic events that give a boost to legislation, planning and other applications for civil protection. These can be considered "windows of opportunity", recalling attention and resources to preparing for emergencies so that the level of interest from the public allows them (in fact, most of the civil protection laws following the disasters that cause a wave of legislative activity). Could it be that the opportunity to develop disaster management in all three ways (such as a particular academic subject, as a commercial business, as a function of government), depend to a certain point of the next disaster from occurring. However, if so, you can do very little except to recognize that opportunities can come without much notice.

So far the discussion has focused on household size and European, but the problem has another side. No matter, when the time about a dozen humanitarian emergencies in various states of development in different parts of the world. The European countries are involved with relief operations, aid officially shipped, and the work of the armed forces in maintaining peace, or unilaterally or through the UN-sponsored operations. These "complex emergencies", as they are called (ie, disasters are characterized by the failure of the military, social, economic and administrative) are likely to be more common in the future. In fact, the disorder seems likely to increase under the pressure of the growing unequal, and fanaticism, polarization, deprivation, vulnerability, and the prevalent use of force to solve the problems of the world.

In summary, the disaster management is a field that is growing rapidly throughout the world. The country has certainly advanced in America, which, at least at the federal level, offers a well-defined model of organization that has been used by many other nations. However, there are particular problems in applying a federal-state monolithic, given the different division of powers. No matter, if there are many parts of the world in which the field of disaster management is deprived of funds, which are poorly developed, and where under the care of organizations or institutions arrettrate or not yet well developed. So, there is a worldwide demand for training services and advice, although without the addition of large amounts of funding to provide these services. Will be some years before that governments realize that Nonsi can benefit from organization and training through appropriate measures to set up the event and the price too low.

The lesson of these findings is that, despite the growing wealth of opportunities, this field is constantly underrated and backed with insufficient funds. It is a really significant in both Europe and the rest of the world. The market for consultancy, training and education is therefore strong interest but weak financial support.

Summary Trends and opportunities in

nowadays we can observe the following trends in emergency management:

(a) by governments, there is a renewal of interest in the field, which, however, does not yet have enough experience or organization to bring the problem under controllo.Ai national, regional and local government bodies should seek help from professionals. You have to see, however, as these institutions are willing to pay for such services

(b) most likely natural disasters, especially floods, storms and snow will become more expensive and most impressive in the future than their impacts in the past. So they called for more demand to increase the level of preparedness and security, albeit in a desultory manner that varies with the disasters that happen. It seems that the political threat of a terrorist attack has put the shadow of natural disasters, but it is certainly only a temporary effect.

(c) The business and industrial sectors have started getting involved substantially in risk management and disaster recovery (although these terms do not mean exactly the same thing in the public service mean that the private sector. The field of security has grown rapidly, but it seems that has not yet fully understood his need planning, education and training.

(d) Terrorism is again a matter of greater importance, especially in regard to new forms unfamiliar that it might take. The years of terrorism linked to separatism have to put Europe; 'avvanguadia combat this problem, but the management of chemical incidents, biological, nuclear or radiological materials may require a radically different approach, along with new sources of knowledge. Europe looks to U.S. for direction, although again we must remember that what works in a federal country is not necessarily ideal for a monolithic state. Above all, we need to experience that combines the multi-disciplinary fields logistics, sociology, psychology, engineering, physics, chemistry, medical sciences, risk mapping and many other disciplines. The success of the initiatives will depend on how they can unify the various sources of knowledge.

(s) of insurance and reinsurance industries are seriously concerned about the question of insurance against disasters. In recent years, the losses in disasters have become excessive, but also the business opportunities have grown in size. Thus the proportion of financial loss in disasters repaid by the insurance has doubled in 10 years. There is, however, a critical lack of capital to underwrite the risks of disasters. Quindi, le compagnie dovrebbero cercare nuove fonti di esperienza per aiutarli a stimare i rischi di coinvolgimento in questo settore nel futuro. Purtroppo, molte compagnie di assicurazione non sono ancora particolarmente sensibili alla possibilità di collaborare con istituzioni accademiche, ed alcune hanno allestito dipartimenti di ricerca per conto proprio, nel caso della Compagnia di Riassicurazione di Monaco di Baviera, con grande successo.

(f) La domanda per la formazione in materia di disaster management è destinata a crescere. I principali clienti sono il settore della pubblica amministrazione, i servizi di emergenza, e le principali compagnie commerciali. I campo è ancora lontano da raggiungere lo stato di una vera e propria professione, but it's going in that direction. There is therefore a double demand for education and professional training and standards. However, since the field can not yet boast the status of a profession, it can not expect to receive remuneration at the level of that of engineers, doctors and other professionals raggionieri well established. Thus, the training must be of short duration and cost.



Notes [1] Geography and anthropology, sociology, social psychology, economics and actuarial sciences, geophysical sciences and construction, medicine and epidemiology, and studies of economic development (see Alexander, D. 1993. Natural Disasters ).


[2] Quarantelli, EL 1995. What is a disaster? International Journal of Mass Emergencies and Disasters 13 (3): 221-229; Quarantelli, EL (ed.) 1998. What is a Disaster? Perspectives on the Question . Routledge, London, 312 pp.

Tsp Buy Annuity Or Not

Dieci principi della protezione civile democratica



1. Civil protection must be a service provided specifically for the population, not only for the state in any of its forms.

2. It must respond to the expressed needs of security and expressed by ordinary people.

3. It must involve the people in participatory manner, in their own safety.

4. It must be organized primarily at the local level: the highest levels of government need to provide coordination, harmonization and support and should not replace the ability to respond to the crisis locally.

5. It must be fully demilitarized.

6. The service should achieve the highest level of professionalism possible.

7. Civil protection must use a generic emergency planning based on scenarios of danger and risk, and act to reduce vulnerability and to address the basic needs of the general population of the areas where the plans are implemented.

8. It should define the sustainable management of the emergency and the reduction of risk and working towards this end.

9. It should be compact with ecological sustainability and urban planning and regional belonging to the local area.

10. It must keep the public well informed of any risks and contingencies that may require you to take actions or measures.

How Big Was John Holmes

, 11 September 2001 terrorist attacks on the United States: impact on civil protection




The following article was written in 2002.

The terrorist attacks that occurred on the morning of Tuesday, September 11, 2001 are unprecedented in scale of operation, the boldness and the degree of coordination. It appears that as many as 6 attacks have been planned, of which 3 have hit targets in full, causing the loss of 3000 lives. At the time of writing, not clear if these outrages will be unique in modern history, or they will kick off a new phase of global instability. In any case, the attacks on New York and Washington DC are having a profound impact on civil protection. This article will examine some aspects of the attacks with reference to the organization of emergency planning and emergency response. Although thanks to the media, the events of that terrible day are well known to all, be the first to summarize the sequence events to better understand what elements are important to the future planning of civil protection.

attacks

at 08:45 on 11 September 2001, during a normal commercial flight, a Boeing 767 is hijacked and driven against the upper floors of the north tower of World Trade Center (WTC) in the financial district Peninsula New York in Manhattan. Eighteen minutes later, another 767 hijacked more strikes against the adjacent South Tower. Both towers had 110 floors high. At 10 and 10 minutes a Boeing 757 plunged into the headquarters of the U.S. military, the Pentagon and at the same time another 757 rushes in rural Pennsylvania, apparently mancando il bersaglio che i dirottatori volevano colpire a Washington o nello stato di Maryland. Almeno tre altri aerei con a bordo terroristi armati di coltelli venivano nel frattempo fermati a terra durante il groundstop , il successivo fermo generale dell’aviazione, sebbene la notizia della loro esistenza veniva resa pubblica soltanto dopo diversi giorni.

La capacità dei serbatoi di carburante dei 767 della Boeing di 90.770 litri, mentre quella dei 757 di 42.680 litri. Tutti i 4 aerei erano decollati con i serbatoi pieni pochi minuti prima di essere dirottati. Le 266 persone a bordo, compreso i 19 dirottatori, muoiono negli scontri e non ci sono superstiti. In tutti i tre edifici colpiti, il carburante dei velivoli alimenta deflagrazioni. Fireballs are injected into both towers of the WTC: the northern fire takes between 95 º and 103 º floor, the south between 82 º and 93 º. The fire at the Pentagon, the largest office building in the world, continues for several hours but was contained by the massive structure of the building, designed to withstand a military attack.

The WTC was built in the 60s, with a structure composed of steel beams covered in concrete. The tower has a strong central column in which there were lifts (198 in all), scales and pipes for the distribution of services. The supporting beams radiating from this central section connecting to the outside with the rest of the casing structure.
impacts and the flames destroy the fire, while the temperature in the upper floors of two buildings rose rapidly to about 800-1100 degrees. Within an hour it became the cement dust or soot. Before the beams deformed by the intense heat and then melts. Sixty minutes after the impact the south tower collapses. That follows the north at 1028, 103 minutes after the brunt of the first plane.

According to the designers, the towers were constructed to withstand the impact of a Boeing 707, the normal airliner in the 60s, and block the development of a fire for about two hours, the time "Project" for the total evacuation of the two tallest towers (the World Trade Center consisted of 7 buildings). It would, however, extremely difficult to design these buildings to withstand the impact, the explosion and the fireballs of premeditated and carefully planned attacks like those of September 11.

The overall figure of the dead is estimated to be 44 units in Pennsylvania, 189 in the Pentagon and World Trade Center in 2765, including hundreds of foreigners in the second case from about 60 countries. A few dozen of the victims were Italian. Given that at times of peak times, the host at 40.000 WTC workers and 80,000 visitors, the mortality rate in New York is reduced by evacuation which had, after all, successful. But dozens of people trapped on upper floors with no possibility of rescue by jumping from windows to avoid dying a slow end in the flames.

Of particular importance for civil protection, the death of 343 firefighters and 78 policemen, who rushed to the WTC just after being warned the fighting and, in many cases they climbed the fire escape with the intention of the towers to save people trapped, or to fight the flames.

The collapse could have been much more devastating impact if the aircraft had taken place, and the resulting fires developed in the lowest points of the towers. Thus, these could fall on the buildings adjacent turning as felled trees, while in reality it fell vertically on a rather narrow. At the beginning of the collapse, the displacement of the load on the upper floors at the lower of about 100,000 tons, and the resulting pile of rubble, which also contains the remains of as many as 3 other buildings (including the WTC-7, a 47-story skyscraper ), weighs about 1,200 million tonnes. The collapse generated seismic waves of earthquakes comparable to a maximum magnitude of 3.2.

Analysis

While almost all the individual elements of the attacks are comparable to some aspects of previous events (eg, the clash of a fighter plane against the Empire State Building), the sum of the effects, and thus the severity of the total impact, has no parallel in the history of terrorism. The magnitude of the attack, the degree of coordination and lack of notice fall outside of any other experience. Therefore, planning before the event could not draw enough inspiration from other events to describe the scenario. Despite this, many lessons can be learned by what happened. Here is a brief reflection on some key issues.

problems in New York

evacuation procedures. For years, the evacuation of tall buildings was the subject of dispute. The skyscrapers are high levels of vulnerability with regard to the risk of fire or structural failure. Some photographs taken during the evacuation of one of the towers of the WTC show narrow stairs (of width less than a meter) and crowded with people trying to get down, pinching the walls to pass the firefighters going up, loads of cylinders oxygen and other tools.

Having received conflicting instructions from colleagues, not all the occupants of the towers were going immediately to the outputs, especially those who were still saw the rain of debris from above and that they felt more secure inside and out. Eventually, many of those who reached the stairs were il buio, il fumo e la presenza di grandi quantità di acqua provenienti dalla rottura dei sistemi di spegnimento degli incendi.

Sebbene una quantità elevata di persone siano riuscite ad evacuare dalle torri (forse il 90%), molte persone hanno impiegato più di un’ora di arrivare al pianterreno dal 70º piano e oltre. Una volta arrivate, il fumo, la polvere e la pioggia di detriti che cadeva continuamente nella piazza del WTC rendeva molto pericoloso il tentativo di uscire. Più positivamente, come avrebbero previsto i sociologi che si occupano dei disastri, il panico sembra essere stato molto limitato: la stragrande maggioranza delle persone si sono comportate in modo calmo e razionale, anche in situazioni apocalittiche.

Scenari di pianificazione per gli edifici alti . Nei giorni successivi, alcuni esperti di ingegneria strutturale comunicano ai mass media la gravità degli incendi (cioè, pochi minuti dopo l’impatto) rendendosi conto che le torri dovevano inevitabilmente crollare. Le implicazioni di tali dichiarazioni per la pianificazione degli interventi di emergenza sono profonde. Sebbene sottostimare il rischio di crollo comprensibile sotto circostanze davvero eccezionali, a New York 248 pompieri e 95 soccorritori partirono da 5 distretti e da 34 compagnie dei Vigili del Fuoco newyorchese e poi morirono. Per di più, molti mezzi dei servizi di soccorso vengono schiacciati nel doppio crollo delle torri.

E’ normale basare contingency plans for catastrophic events more likely and less rather than on almost unthinkable with truly apocalyptic consequences. Yet disaster planning needs to "think the unthinkable": it should get used to the Disaster Manager to adapt its decisions and actions at the very unusual conditions. Taking into account that there is nothing more cruel and deceitful of hindsight, we must rethink, rather deeply, the management of major disasters that hit the tall buildings. As part of the municipal emergency plans, we must formulate detailed scenarios of response to such situations.

In the United States each year 16,000 to 20,000 tall buildings catch fire, killing between 80 and 90 people and wounding 800-900. Obviously, the problem is not restricted USI: for example, Sao Paulo in the 70s was the scene of two famous fires in skyscrapers, where dozens of people died trapped. So it's not a problem so unusual.

The most significant issue is the safety of emergency workers, the need to speed up the evacuations, and the need to protect the evacuees while leaving the building. The first part of the attached table provides a range of theoretical predictions of time to evacuate the WTC calculated using some nonlinear equations developed in a series of evacuation drills carried out on tall buildings in downtown Toronto in Canada in the 70s. The second part shows the real time of evacuation after the terrorist attack on the WTC in 1993 when a bomb caused a series of fires and the general diffusion of smoking in both towers. As the data, the total evacuation of these buildings required more than 2 hours, and maybe more than 3.

emergency management. The emergency management center in the district of Manhattan was to be located in the north tower of the WTC, where the Port Authority of New York-New Jersey had rented a suite of offices on the 2 º, 14 º and 19 º floors. The decision to house the main operations center in the WTC was an act of defiance against terrorism following the bomb in '93, He was also a strategic move, given the central role of the WTC in the area of \u200b\u200bhigh finance in New York. Obviously, under the circumstances become less suitable place overall. From this we can conclude that in complex urban environments as Manhattan would be required that different operational centers located in protected areas and connected in a network. Each of these should have the ability to become the main center if that were destined to be rendered unusable.

Other aspects of emergency management in New York have had more positive outcomes. Since the end of the peninsula of Manhattan surrounded by water on three sides, it was pretty easy to cordon off the affected area and control access. I battelli evacuavano i feriti attraverso il fiume Hudson ad un posto medico avanzato a Jersey City, sulla riva del New Jersey. Malgrado la lentezza del trasporto littorale, era molto più facile e sicuro soccorrere le vittime così, ad una certa distanza dal fumo e dalla confusione di Manhattan. Di 5284 feriti, il 7,9% aveva bisogno di una degenza in ospedale. Il centro traumatologico di primo livello più vicino al sito della sciagura raggiunse la sua quota di pazienti (circa 200) entro 2 ore dall’inizio della catastrofe, e in seguito i feriti sono stati distribuiti tra 83 delle 170 ospedali delle 5 borough e 3 contee della zona metropolitana di New York.

Alle ore 17,00 dell’11 settembre il soccorso medico aveva raggiunto la its full strength, but the second wave of casualties, anxiously awaited by doctors and nurses, there was not. Eight hours after the onset of the disaster there was no more need to practice triage.

Search and Rescue. At the site of the New York disaster billion and 200 million tons of debris had accumulated in a tangle very compact but quite unstable. It was extremely difficult to penetrate this to find people trapped, and most importantly, fires broke out all the time. The voids were filled with dust, mud or fire and the surrounding buildings threatened to collapse at any moment (in fact, the WTC-7, 47 floors collapsed towards 17.20 on the same day of the tragedy). Despite search and rescue operation involving up to 1,200 rescue workers at a time (the maximum number that could work in space available), very few people were found alive, and those only at the beginning of the rescue operation.

The enormous weight of the buildings collapsed, and excessive fragmentation took place during the collapses, meant that the proportion of voids was less than the figure of 15% found in other big collapses of buildings, such as in earthquakes. The scale of the massive accumulation and instability of the site required the use of heavy vehicles of the types that are usually not used in such circumstances for fear of crushing the victims still alive but trapped under the rubble. In addition, the precariousness of the site required a series of interruptions to the work of rescue workers were struggling while desperately to consolidate the walls still standing. However, with the passage of time, work became more regular and, by dividing the site into 4 zones of control, 90,000 tonnes of rubble were removed during the first week.

Problems in Washington, DC

An editorial comment published in authoritative newspaper The Washington Post a week after the disaster said that "an examination of the events of last Tuesday indicates that the District [Washington DC] was unprepared for emergency and therefore was unable to react and assist the public quickly and effectively. "This statement serious enough to require clarification and explanation.

communications and emergency response . The United States has a system for transmitting emergency messages to the public through radio and television that is regularly put to the test. On the morning of the disaster this tool was not activated in the District of Columbia. Nevertheless, the majority of citizens obtained information on what to do by the same media that would have sent messages from the government, but the interpretations of the situation made and received from journalists are not necessarily the official ones. In fact, in some cases it seems that an official policy of the emergency was not there. For example, the head of the Office of the Mayor, unable to communicate by telephone because of the overhead lines and cellular networks, used the mail to order the evacuation of federal agencies. But four minutes after the Chief of the city sent the countermand, always e-mail. The first message was sent in response to information, proved wrong, that three other planes were to fall on the capital and the second was the result of a decision not to suspend the operation of government during the crisis.

Aside from the obvious conclusion that e-mail is not a good way to spread an evacuation order, but the Internet was slowed down by overloaded and therefore lost its ability to transmit messages in real time. In any case, in fact non-essential workers were returning home on their own, creating smooth road all over the District of Columbia and blocking the movement of emergency vehicles. A system of traffic management (the result of planning "Year 2000", called "Y2K") was activated only after 3 hours and managed to loosen congestion.

According to the forecasts for any emergency, some satellite phones should be placed in key government officials in Washington. Unfortunately, these tools have remained closed in a closet until the next day. In addition, the Local Health District did not have radios capable of monitoring communications between hospitals and ambulances, and then, given the non-functioning of telephones, was not able to participate in the formulation of logistical decisions and to estimate the availability of health services.

emergency plans. The Metropolitan Police Department of the District of Columbia, which has a workforce of 3800 employees, a plan is not kept or anti-terrorist nor a procedure to inform the police and their commanders on the road where and how to respond to the crisis. The commanders were forced to improvise a plan to the minute. The police was not aware of the decision taken by the federal government to send his employees home, and then was caught by surprise by the influx of traffic on the roads.

Meanwhile, the houses of Parliament were not evacuated. Was it better that way, since deputies and senators were not trained in the evacuation and the plan was not updated regarding the location of emergency exits of the building. Although he feared an attack right at the Capitol.

On the other hand, however, the Metropolitan Washington immediately put in place its emergency plan and continued to function well throughout the crisis. Unfortunately, many commuters took that was not in operation and tried to return home on the roads, so the circulating half-empty trains.

The Arlington County (Virginia), peripheral to the city, launched its emergency plan 10 minutes after the start of the crisis. Luckily, however, that there were not many casualties, because the administration of the city of Washington declared that the Washington Hospital Center, the best trauma center in the area, could not accommodate a large number of victims.

Conclusion. At a time when the Secret Service realized that a third plane was heading to the White House (and then with a sudden change at the Pentagon), did not have a procedure to bring it down and at the fighters would cover 200 km before arriving in Washington. But apart from such a drastic measure as hypothetical and killing, was abundantly clear that the magnitude of a possible terrorist attack on the U.S. capital was widely underestimated in the plans, if there were. The scenarios were too modest, particularly about the likely level of chaos, and the existing plans were neither comprehensive nor sufficiently clear about the likely risks. In fact, an emergency simulation carried out during the month of June 2001 provided for a chemical attack in the open air in front of the Smithsonian National Museum, no damage and few casualties.

addition, the Agency for Civil Protection City of Washington had neither the funds nor the manpower to create and burn in a contingency plan of the size you need. Following the crisis, its parent organization, FEMA, churches (and usually won) $ 250 million a week to carry out relief work: infinitely superior to those amounts that were requested (and obtained) to finance the plan before the attacks.

According to the consensus that prevailed before Sept. 11, the city needed to plan well coded but largely generic, such as the plan "ThreatCon" ( Conditions threatening, that is, threatening conditions) of the U.S. military, which lists what to do a series of alert levels. The ThreatCon worked pretty well during emergencies in New York and Washington, but as a tool to deal with such disasters was not enough.

Conclusion

If the terrorist outrages of 11 September 2001 will remain a rather unique series of events in history, or into a new era of attacks, it is not known yet. The events of that day took place in the middle of town so rich in resources and sources of assistance are not missed. If anything, the opposite of congestion and the reactions of convergence towards the site of the attack were a major problem. As a necessary consequence, measures to quell the confusion were necessarily drastic. In this sense, terrorism leads, inevitably, a more authoritarian than other forms of disaster, with the exception of the war itself. Emergency planners need to better address this issue in order to incorporate a structure of civil protection in response to terrorism dominated by police and military forces.

The medical, psychological, economic and strategic disaster will go on for years, with profound impact not only on the victims' families and the prospects for peace in the world, but also on the community of the American Civil Protection. Hopefully, however, that the events of that terrible day, and all his seguito domestico e internazionale, stimoleranno un miglioramento della pianificazione di emergenza, con "regole di combattimento" più precise e scenari di risposta più dettagliati e accurati. Creare questi strumenti considerato un obbligo da molti operatori nel settore che vogliono rendere omaggio ai coraggiosi soccorritori che hanno perso la vita nel cuore di Manhattan.

Tabella nº 1. Tempi di evacuazione della World Trade Center, New York
Tempi teorici di evacuazione
______________________________________
Pauls (1980) equazione di flusso medio
10.000 persone per scala = 2 ore 14 minuti per uscire
20.000 persone per scala = 2 ore 41 minuti per uscire
______________________________________
Pauls e Jones (1980) equazione per il tempo totale di evacuazione
10.000 persone per scala = 1 ore 59 minuti
20.000 persone per scala = 3 ore 56 minuti
______________________________________
Tempi misurati di evacuazione
Quenemoen et alii (1996) evacuazione in seguito alla bomba e l’incendio del 1993
Tempi di evacuazione per 161 persone:
48% meno di 60 minuti
27% tra 60 e 120 minuti
25% pi di 120 minuti

L’11 settembre 2001, il crollo delle due torri termin le evacuazioni 62 e 103 minuti dopo l’inizio della crisi con circa il 90% delle persone evacuate.
______________________________________
Opere citate:
Pauls, J.L. 1980. Building evacuation research findings and recommendations. In D. Can(curatore) Fires and Human Behavior. John Wiley & Son, New York.
Pauls, J.L. e B.K. Jones 1980. Building evacuation: research methods and case studies. In D. Canter (curatore) Fires and Human Behavior. John Wiley & Son, New York.
Quenemoen, L.E., Y.M. Davia, J. Malilay, T. Sinks, E.k. Noji and S. Klitzman 1996. The World Trade Center bombing: injury prevention strategies for high-rise buildings. Disasters 20(2): 125-132.

Weed Leaf Nail Design

The method of scenarios in planning and emergency management



Estratto dal libro: Principles of Emergency Planning and Management David Alexander

In the context of disasters, a scenario is a hypothetical progression of circumstances and events designed to illustrate or reveal the consequences of any decision, action or impact. Usually the scenario is set up to provide a consistent and logical answer to the question "what if ...?" However, rather than aim for the future, sometimes using the method of scenarios to look back (a positive and constructive way to use hindsight) as a tool for debriefing to better learn the lessons of an event's happened before.

A scenario is a model of conditions and circumstances. It is usually designed to reveal the link between these two things (that is, how the conditions affect the circumstances and conditions as circumstances change). The method of the scenarios can be used, for example, to reveal the consequences of the impacts of disasters, or a certain range of mitigation works, or certain strategies of search and rescue of survivors. The scenarios may explain the vulnerability interacts with dangerousness to create special forms of risk and impact. The scenarios for losses due to disasters can be modified according to certain measures to reduce the risk and in relation to how they work during the disaster.

It is clearly essential to build the scenarios as chains of events that are plausible and reliable. One of best ways to do this is to use a systems approach (systems approach). The entries are specified in the form of a series of conditions and forces that cause changes in fundamental variables (such as a river flood wave interacts with the strength of a bank). Further conditions are specified within the compass of the system: some of these limitations, and other guides, the mechanisms of change in the situation as it evolves. Together, the conditions act as a forcing function, or motive of the events. The emission model is the scenario, as constructed, and its changes over time taken into account.

The scenarios are useful to contingency planning, given that the method can be used to clarify the conditions that need a plan (ie, the presence of dangerous phenomena and their likely effects), that the impact of preparation in terms of how it can reduce risk. In fact, detailed studies of hazard, vulnerability and risk exposure can be integrated into the development of the scenarios that are created to investigate the losses and injuries in disasters. As a result, they may form the basis of emergency planning and measures to mitigate the risk.

The first step in building a scenario is to specify the initial entries and the limitations (Boundary conditions). For example, you could develop a scenario for a magnitude 7 earthquake, which constitutes the danger of fires dl emission scenario and soul. The initial conditions are given the vulnerability of the built heritage, dall'inquadramento activity at the time of the earthquake, people living in the area at risk, and the ability of the emergency forces to respond quickly to disaster. The development scenario is based on the temporal progression of events. The vulnerability data indicate that a number of houses could collapse and a number of people could die. Under this scenario can be used to judge the probabile efficacia delle operazione di soccorso o per stimare la riduzione dei danni e degli infortuni avuta con un potenziamento dell'intervento immediatamente dopo il sisma.

Lo scenario procede attraverso una serie di fasi, le quali vengono solitamente definite dagli eventi (ad esempio, può estendersi dall'isolamento delle vittime all'inizio dell'emergenza all'arrivo delle forze incaricate con la ricerca e il salvataggio dei superstiti, e poi all'inizio dei processi di ripristino). Alla fine di ogni fase si può "congelare il tempo" e trarre la somma degli eventi successi, delle azioni e le decisioni prese. Così si ottiene una "sezione trasversale" del disastro, un quadro del progresso fatto e dei problemi affrontati dai principali actors (disaster managers, relief workers, firefighters, volunteers, doctors, paramedics, etc.).. When you explicate the entire progression of events, and was once the final outcome, you can change the scenario by searching for answers to a series of questions like "what happens if we double the number of rescuers?" or "what will happen if the earthquake caused landslides that hinder the main roads into the disaster area?" The answers to these questions are based on a number of changes to the outcomes of the scenario.

is particularly important that the emergency planning construct scenarios in written form, if necessary with the help of relevant experts, for all major risks and dangers that come within its competence. These will serve to illustrate the situations that must be planned and will help to establish the parameters of the process of emergency planning at the local level.

In many cases the construction of scenarios a priori there are no means to verify the result, if a disaster does not happen like that in the near future (thus making the scenario a kind of prophecy!). So we look for the verisimilitudine the plausibility of the facts, conditions, mechanisms and outcomes represented by the scenario. While going on the progression of events must continually ask themselves: is this possible? " or: "Is likely to happen this way?" It is best to avoid scenarios or improbable events, and also, of course, the injury, the excesses of fantasy, irrationality, excessive detail. A good scenario writing takes the reader with clarity, transparency and logic through a series of events will demonstrate the framework for decisions, illuminate the reasons for actions taken, describe the outcomes, and justify the different events with a credible and logical explanation . The present exemplary scenario the large picture of the situation and not be encumbered by the invention of large amounts of detail. Its bases are collected data on hazards and risks, and also the trend di eventi simili che sono veramente accaduti in passato (quindi è essenziale studiare meticolosamente i dati che possono aiutare la formulazione dello scenario). Così, l'esito finale sarà interamente plausibile, sebbene non necessariamente immune al dibattito.

Come esempio consideriamo lo scenario costruito per una ripetizione del grande terremoto di Kanto, che avvenne in Giappone nel 1923, il quale è raccapricciante. Si aspetta tra 40.000 e 60.000 morti, fino a US$1.200 miliardi di danni al patrimonio edilizio, danni per $900 miliardi all'infrastruttura, e perdite di $1.000 miliardi dovute all'interruzione delle attività redditizie. Comunque, il vero esito dipende dalle assunzioni su cui è basato questo scenario: some researchers have considered exaggerated sums of money found here. However, it is clear that many important things are at stake: the final figure of losses could be equivalent to half of GDP and Japanese financial markets may force the world to suffer a serious downturn.

Of course, most of the scenarios are less grand than this, and so now we will consider a hypothetical example that has the merit of being very simple. To be brief, to exclude most of the details of the fund.

Around 16.00 a normal Friday afternoon in a case of May an earthquake measuring 6.7. The strong shaking lasts 33 seconds and generates horizontal accelerations the surface reaching 31% of gravity (0.31 g). Among the architectural heritage of the city hardest hit (50,000 inhabitants) the damage is serious but not universal. Most spectacular of all, a 200 m long section of a highway viaduct collapsed on a parking machine, crushing several cars and maybe some of their occupants. Secondly, the facade of brick and concrete of a department store in the mall of the city collapsed and blocked the main road with its rubble. Again, as there may be victims. In third place in a shed, in use at the time of the quake, with a steel frame collapsed in the industrial zone, while in the residential you had the partial collapse of a 4-story building occupied by the elderly. In three points of the city breaking gas pipes has resulted in fires that are violent but fortunately located. Elsewhere in the rupture of water pipes has led to a series of small floods and a severe loss of pressure of water supply in the city. The lack of mutual assistance indicates that neighboring communities have to deal with serious damage. Finally, in the area there is only one hospital with emergency room, and its functionality and efficiency are limited by severe structural damage and the absence of several members of the medical and support of some officials.

The scenario goes on the basis of a rational approach managing a situation that is currently full of unknown, uncertainties and limits on the capacity to act. The damage described above is clearly based on appraisals of vulnerability that are made in "times of peace" in mind with a certain magnitude of earthquakes and other seismic parameters of the specification. The description refers to a time point about half an hour after the earthquake: it is not complete at this time because you will not have a complete range of information on the disaster. For example, the number of deaths and injuries is not known. The further development of scenario depends on a number of assumptions of how it will continue the rescue operation, and shows how the situation than people who remain homeless and injured survivors and trapped. The end result is a graphic of the immediate need for massive resources, a problem that the planner will face a very serious emergency.

Pain In Tummy With A Feeling Of Bowel Movement

A quartet of disasters



I wrote these four essays in the mid-90s for an encyclopedia published in America in the twentieth century. I offer the translation of the English text as an example of the methodology for the analysis of catastrophic events, practiced with the aim to draw conclusions for the future.


EARTHQUAKE IN ARMENIA

brief summary. About 25,000 people have died in an earthquake of magnitude 6.9 that caused the collapse of many buildings in three jerry-built towns and 150 villages in northern Armenia. The event

. The main shock occurred at 11.41 on 7 December 1988, about 80 km south of the main chain of the Caucasus. The shaking lasted 30 seconds and reached 21% gravitational accelerator. Four minutes later came a aftershocks of magnitude 5.9, which caused further collapse of the buildings already damaged by the first knocked.

Officially, 24,944 people died, although some estimates were much higher. Of the 31,000 wounded, the hospital had 12,200.

Circa il 40% del territorio armeno fu interessato dal terremoto, e, dei 150 paesi danneggiati, 58 villaggi furono quasi interamente distrutti e 3 città furono gravemente danneggiate: Leninakan (pop. 290.000), Spitak (pop. 20.000) e Kirovakan (pop. non conosciuta). Ben 8 milioni di metri quadri degli alloggi (il 17% del totale armeno) furono distrutti o resi inagibili. Soltanto a Leninakan, 72 edifici alti sono crollati e 60 altri sono stati danneggiati oltre qualsiasi possibilità di ricupero. A Spitak, 9 km dall'epicentro, nessun alloggio rimase utilizzabile. In tutta la zona interessata dal sisma i senzatetto furono 514.000.

Mentre, i due impianti termonucleari dell'Armenia non ricevettero danni (un colpo di fortuna, dato la loro mancanza di adeguati sistemi di sicurezza), 130 fabbriche furono danneggiate o distrutte, un duro colpo all'economia armena. Le perdite economiche ammontavano a US$15 miliardi.

Le frane sismiche erano numerosissime, come erano i fenomeni di liquefazione. Alcuni villaggi molto isolati furono colpiti da valanghe detritiche, causando perdite di vite, e diverse strade e linee di ferrovia interrotte dai movimenti franosi. Come in altri eventi sismici di impatto molto disperso, i soccorsi raggiunsero i villaggi più remoti soltanto con massiccio ritardo.

Il comportamento sismico degli edifici . La cattiva qualità del suolo sotto Kirovakan, Leninakan e Spitak causò il crollo di diversi edifici, ma the type and quality of construction they recorded much more. The older buildings were built of stone, often of a volcanic rock that broke during the seismic shaking. These buildings were very heavy horizontal elements of mattoneria or concrete, and then the dynamic load of the earthquake caused the crack angle to cause the collapse to the outside walls and the free fall of internal plans. However, many older buildings made of stone have withstood the earthquake, and most buildings had collapsed less than 15 years.

The more modern buildings were of reinforced concrete, often with preformed panels are also made of concrete. But the weakness of the joints in the frames Structural often leads to the progressive and total collapse of the building. If not, the scales fell off and collapsed during the seismic shaking, thus preventing the evacuation of occupants.

altisismiche standards in Armenia was insufficient in that it underestimated the probability of strong shaking and because they were based on an insufficient amount of data. However, while the quake was not likely in itself, the seismicity of the area was well known, such as to indicate the appropriate measures to be taken. But the Armenian authorities had ignored a fundamental scientific work of 1975, which clarified the need for earthquake resistant construction.

The epidemiology of infortuni . Dal terremoto di Guatemala del 1976, a quello di Armenia del 1988, si è imparato molto sugli infortuni causati dagli eventi sismici, specialmente in rapporto alle inefficienze delle operazioni di soccorso, l'inutilità dei materiali inviati nella zona disastrata, e la predominanza di ferite da schiacciamento dovute al crollo degli edifici scossi dal sisma.

Mentre alcune vittime morirono di asfissia a causa della polvere proveniente dagli edifici sgretolati dai tremori, molti sono morti schiacciati sotto travi e macerie. La "sindrome dello schiaccamento" avviene quando la proteina mioglobulina viene rilasciata in quantità massiccia nel flusso del sangue dalla rottura del tessuto muscolare. Essa finisce per impedire il funzionamento the kidneys, requiring immediate dialysis to save the life of the patient. Similar phenomena occur when potassium is released into the blood stream from the crushed cell disruption. In Yerevan, the kidney dialysis unit sent by the United States allowed the execution of 25-35 daily dialysis.

A sample of three Armenian villages, a total of 8,500 residents, showed a mortality of 49.5%, and a traumatic morbidity of 14.6%. But apart from the city level, the worst took place in some unusual buildings that collapsed with a mortality rate of over 90% of the occupants. Thus 205 of 215 workers died in a factory, and 285 of 302 children in a school. Typically, the plates constituent plans of such buildings disintegrated into fragments so small as not to give space to survive in the rubble.

Under the rubble lay about 40,000 people, of whom only 15,000 have been pulled out alive, most other survivors using only rudimentary tools. While some victim was rescued less than 19 days after the earthquake, 90% and saved lives of people trapped were freed within 24 hours after the main shock.

Logistics . The operations of search and rescue and relief efforts in general, were chaotic, inefficient and slow to begin. The weather alternated between snow fog, with night temperatures going down to-20EC. In fact, the situation weather, coupled with the fact that almost all hospitals in the area affected by the earthquake were not working anymore, slowing down relief efforts in particular. In addition, 80% of doctors and nurses in the area died in the earthquake. Then, the wounded had to be taken to Yerevan and Tbilisi. The lack of an adequate system of air traffic management, along with bad weather, had resulted in two aerial combat, with a total of 85 dead.

The problem of drugs. In the early days of the disaster are missed analgesics, antibiotics, anesthetics, intravenous fluids, syringes, stretchers, and equipment to transport the wounded.

Approximately 60 hours after the strike began rescue operation in which as many as 70 nations donavano money, manpower, food and other aid. Within a week after the disaster, the main airport in Yerevan operated 150 flights per day and 32 stores of donated material. Over 5,000 tonnes of medicines and medical supplies have arrived from abroad (25-30% by value of all donations), of which only 30% result, while 8% had expired and 11% were type useless. Most of the rest consisted of a total of 238 medicinal names in 21 different languages, making it impossible to identify the substance in question. The translations were rare in Russian and in Armenian non-existent. So at the end of the month you had to destroy 20% of medicines. In addition, a strategy efficiente per gestire i rifornimenti medici si è avuta soltanto 2 settimane dopo il disastro. Nel frattempo i produttori di medicinali hanno apparentemente colto l'occasione per scaricare i loro avanzi sulla zona disastrata.

Il contesto . Il terremoto armeno avvenne in mezzo ad un processo di trasformazione dell'Unione Sovietica. Fu il primo evento del genere in cui le autorità sovietiche accettarono una grande quantità di auiti dall'estero.

Tuttavia, la gravità del disastro non riuscì a cancellare le preesistenti preoccupazioni politiche: alcuni soccorritori azeri sono stati aggrediti dagli armeni, alcuni politici locali usavano la situazione per diffondere l'ultranazionalismo, le tensioni etniche grew strongly, and finally even the coprifuovo and the presence of 13,000 Soviet troops managed to contain the situation.

FIRE IN MEDITERRANEAN CALIFORNIA

brief summary. In the autumn of 1993 some spot fires in southern California burned 79,000 acres and 1,814 homes, especially in Malibu. Three people died in the flames. The event

. At the end of October 1993, a series of fires burned a large section of the Mediterranean's southern and most urbanized of the State of California. In six counties, including Los Angeles, 14 fire completely destroyed 731 houses and 66,800 hectares of heathland. Economic losses were estimated at more than $ 1 billion.

beginning of November 1993 nine other fires have killed three people, injured 111 and destroyed 1,084 homes, leaving more than 30,000 people homeless. In Malibu, 350 houses were destroyed, with damage amounting to $ 375 million. The injuries complained of smoke poisoning, burns and lung damage caused by breathing superheated air.

The Malibu fire began east of the city at 11.45 on November 2, 1993. In 4 hours, the flames traveled 19 km along the deep valleys carved into the coastal hills until it reaches the Pacific Ocean. The fire leapt in the air and 15-20 m beachfront palm trees burst spectacularly into flames.

Gusts a kind of Santa Ana wind threw hot and dry air at 90 mph and pushed the fire towards the urban area of \u200b\u200bthe coast, necessitating the evacuation of schools, factories, homes and a university. Many residents, worried about the fate of their homes, refused to evacuate. Therefore, rather than to fight the flames, firefighters in the middle of a residential area of \u200b\u200bEmerald Bay was used to save the lives of those who refused to go away.

Logistics . 7,000 firefighters were in the area. Undertook the bulldozers to remove combustible materials from the heath around Malibu, and it was great use of a different number of Bell-412 helicopters, capable of carrying 1,400 litri di acqua, e di aerei C-130, i quali possono trasportare 11.000 litri di sostanze chimiche utilizzate per sopprimere le fiamme. Tuttavia, i rischi erano troppo grossi per poter volare di notte.

La strategia per combattere l'incendio consisteva nella creazione di una serie di confini, mantenuti contro le fiamme da linee di pompieri insieme ai loro 250 mezzi. Si allestiva un posto di comando sul lungomare. Ma diverse volte, le linee sono state spezzate dalle fiamme, e di fronte a questo pericolo, si è dovuto ritirare anche il posto di comando. A Las Flores Creek, le fiamme passavano nel vento sopra i pompieri, i quali sono rimasti quasi circondati.

Gli incendi selvatici . Sebbene la protezione di risorse ad alto valore richieda l'impiego di un intervento sempre diretto e vigoroso, ogni giorno si dovrebbe formulare una nuova strategia per combattere l'incendio in base alle condizioni meteorologiche.

Un cosiddetto incendio selvatico brucia fuori controllo e minaccia le persone, gli insediamenti o altre risorse. Tramite barriere naturali o costruite, un incendio confinato viene limitato ad una determitata area, mentre il dilagarsi di un incendio contenuto viene impedito da una linea di controllo. Se quest'ultimo avvolge tutta la zona in fiamme, si ha un incendio diretto. L'incendio di Malibu fu prima contenuto e poi diretto.

Quando bruciano i rami e le foglie di un albero si ha un incendio di corona, che può divampare in rapporto alle fiamme in superficie (incendio dipendente) o propagarsi direttamente da un'albero all'altro (incendio scorrevole). La maggior parte degli incendi nella macchia mediterranea sono incendi dipendenti di corona e incendi di superficie, e possono causare incendi nel terreno ("ground fires") se il materiale combustibile giace sul terreno in uno strato spesso e consistente. I back-burns ("incendi ripassanti") possono avvenire se il materiale combustibile non viene interamente consumato dal primo passaggio delle fiamme, e se il vento cambia direzione tale da rispingere l'incendio indietro. I venti forti possono causare gli incendi a chiazza quando essi gettano frasche brucianti davanti alla linea del fuoco. Nella macchia, la temperatura dell'incendio può raggiungere 1100EC.

A Malibu si ebbe un incendio di superficie con elementi di incendio di corona che si autopropagava da un albero ad un altro. La radiazione e la circolazione dell'aria stimulava l'incendio a propagarsi sulle colline, non soltanto a progredire verso il mare spinto dal vento. Come spesso accade, le irregolarità del vento e della distribuzione territoriale di materiali combustibili spingevano il fronte dell'incendio in forma irregolare caratterizzata dal frequente scoppio di incendi a chiazza.

L'erosione accelerata e altri effetti ambientali . Dieci giorni dopo l'incendio a Malibu, la caduta di 4 cm di pioggia in 24 ore diede luogo ad un episodio di erosione accelerata nella zona bruciata, compresa una serie di colate of mud houses and took away a part of the coastal road.

When vegetation burns the Mediterranean, it distills some natural chemicals that waterproof the soil with a hydrophobic layer. At the same time the destruction of green vegetation by erosion, which is enhanced by the increased flow of water on the barrier layer. The result is a series of mudslides that come with the first rains after the fire.

The high frequency of fires, the power of deadly landslides and the continued urbanization of new areas determines the decline of populations of species of rare birds. California fires of 1993 have left a fragmented habitat so che alcune specie di avifauna rara probabilmente non riusciranno a superare la scossa.

Il rischio di incendio in California . Ogni anno circa 3.000 incendi scoppiano nelle macchie della California meridionale. Molti sono di origine dolosa, ma meno del 20% delle persone colpevoli vengono soperte.

Come altrove, in California il rischio di incendio raggiunge un massimo nelle aree tra le zone di nuova urbanizzazione e il terreno ancora in stato naturale. Sono delle aree di massima crescita demografica e quindi di grande frequenza degli incendi, i quali causano danni ingenti e costosi nel giro di poche ore. Circa un terzo del costo consiste nelle spese per combattere l'incendio.

Buone risposte al rischio . The devastating fires of 1970 and 1977 had resulted in the creation of the California Firescope, a structure based on a centralized Incident Command System (ICS) and a multi-agency Coordination System (MACS), with a center with a sophisticated system Fire Information. The emergency intervention is based on weather forecasts and infrared surveillance of areas of potential fire. In the case of the rapid development of a large fire, the key to effective intervention is the system of mutual assistance between the various regional and local authorities, which guarantees a certain level of timely workforce against the flames.

Un'incendio catastrophic stimulates certain measures, how to broaden the access routes to areas most at risk, buy more equipment (like air Canadian CL-215 "super scooper", which is able to lift 6,000 liters of water in just 12 seconds), and improve planning and coordination of emergency operations. Also, because burning some houses and spare others, the fire can also demonstrate the utility of standards for reducing inflammability of building materials.

good response to risk . In California, as elsewhere, the use of appropriate fire regulations varies from place to place as the tendency to follow them. After the great fire of 1991 in Oakland, 25% of homes not rebuilt were the measures required by new legislation to prevent the spread of the flames.

Repeated public opinion polls in California showed a progressive increase in the sensitivity of the fire problem, but without a tendency to take sufficient measures to mitigate the risk.

Despite a great plan for insurance against fire damage, insurance companies have done little to encourage their customers to reduce the risk of fire. In summary, some researchers argue that the offer of generous loans and repayments has almost rewarded the owners of homes burned for failing to mitigate the risk.

avalanches in the Alps EUROPEAN

Executive Summary . Diverse valanghe di neve uccisero 150 persone nel Dipartimento francese di Savoia, nell'alta valle del Rodano in Svizzera e nel Tirolo austriaco. In ciascun paese i disastri diedero luogo a nuovi programmi di mitigazione del rischio di valanga.

Gli eventi di febbraio 1970 . Nelle Alpi europee i disastri causati dalle valanghe di neve avvengono in base allo spessore della coltre di neve, alle condizioni meteorologiche e all'uso del terreno in fondo ai versanti interessati dall'instabilità. Avvengono due tipi di catastrofe. Nel primo, le condizioni meteorologiche favorevoli alle valanghe durano a lungo, dando luogo ad un alto rischio che continua per lunghi periodi. Ad esempio, nell'inverno del 1950-1, 650 persone morirono in una serie di valanghe alpine che complessivamente distrussero 2.500 edifici.

Nel secondo tipo, disastri provocati da singoli eventi che interessano soprattutto gruppi di sciatori o alpinisti avvengono negli anni di rischio generalmente basso. Così, in Svizzera nel 1960 un gruppo di 88 lavoratori morirono sotto una valanga proveniente da un ghiacciaio.

L'anno 1970 offre un buon esempio del primo tipo, una stagione disastrosa, che induceva ad una rivalutazione generale delle precauzioni contro le valanghe alpine.

Nel febbraio del 1970 le temperature nelle Alpi oscillavano continuamente intorno allo zero, con notevoli inalzamenti quotidiani. Nelle aree ad alta quota e sui versanti piuttosto ripidi, questa situazione provocava growing instability in dense layers of snow earlier by a heavy snowfall.

On February 10, 42 people died and 60 were injured by an avalanche in Val d'Isere in the French department of Savoie. It was a movement of powder snow, with a 45-90 front off me when to stop a thickness of 3.65 m. Invaded a youth hostel, where thirty young people died asphyxiated or crushed, while another dozen people was swept away from outside the building. The hostel was built only 7 years before his site was not considered a place of particular risk and was also protected on top by a series of concrete blocks, which, however, does not help stop the avalanche.

From the French authorities, the rapid recognition of a potential danger for the immediate closure of Val d'Isère. A helicopter was able to evacuate the wounded once before worsening weather conditions such as to block any possibility of flying. The first convoy of cars coming out of the valley was hit by a new avalanche, while a similar event elsewhere killed a child. As soon as possible, the authorities evacuated 5,000 people from hotels and residences in the area.

Following the disaster took issue much on the apparent reluctance of French authorities to invest in structural measures for mitigation of avalanches (tunnels, fences, roofs, etc.). When you invest a lot in the promotion of winter sports. Having recently moved from Chambéry to Grenoble, the National Institute for Research on Snow had just lost his representative in Val d'Isere.

Later that month, in Switzerland 11 people were killed and 19 injured by an avalanche that had not seen such a site in that more than half a century. Despite that 500 troops and 12 dogs to search for survivors, it was not easy to evacuate the victims amid the snow and the continued risk of further avalanches.

February 25, a village on Mount Cenis in the French Alps was hit by an avalanche 10 times greater than that of Val d'Isère. Seven lives were lost. The movement was immediately preceded by a pressure wave that demolished several buildings. It took 13 hours to extract the victims from the snow. The same day an avalanche killed two people in the Tyrol in the middle of the main street of St Leonhard Pitztal am. In the Austrian provinces of Tyrol and Vorarlberg avalanches were so widespread as to prevent the evacuation of 14,000 people.

Meanwhile, the Swiss made extensive use of explosive material thrown from helicopters to trigger avalanches before the snow accumulation can be dangerous. In Italy, meanwhile, seven soldiers died in an avalanche during their exercises.

The events of April 1970. On April 15, 1970, when temperatures were unusually alte per la stagione, una valanga a lastre di ghiaccio colpì un sanatorio nell'Alta Savoia francese. Si dovettero impegnare 21 ruspe per setacciare le macerie, da cui furono estratte 72 salme, 56 delle quali bambini. Alcuni ulteriori movimenti di neve, ghiaccio e macerie ferirono 13 soccorritori.

Il rischio valanga nelle Alpi . Nelle Alpi europee, come in altre catene montuose, le cifre degli infortuni per valanga variano parecchio da un anno all'altro. In Svizzera, ad esempio, la media è di 17.480 eventi e 24 morti, ma questo può variare di diversi ordini di magnitudine.

Le popolazioni indigene continuano a diminuire nelle aree alpine, mentre aumentano quelle saltuarie, le quali non sempre conoscono bene i rischi. In solo 30 anni alcuni centri sciistici nelle Alpi sono cresciuti di 20-50 volte rispetto alle loro dimensioni nel 1945. Intanto, le popolazioni di vacanzieri aumentano proprio nel periodo di massimo pericolo di valanga.

Nell'inverno del 1970 i rischi variavano con le condizioni meteorologiche. In febbraio avvenivano valanghe di neve polverosa associate con il freddo. Erano di una rapidità tale da causare onde di pressione capaci di travolgere persone ed edifici. Invece, le valanghe di aprile erano composte di lastre di ghiaccio, le quali generavano pressioni di impatto in rapporto con la loro più alta densità. Gli eventi di febbraio erano la conseguenza di tempeste, quelli di aprile di scioglimenti.

Il salvataggio . Si riscontravano grande difficoltà di estrarre le vittime dalla neve delle valanghe di febbraio 1970. I cani, che in Svizzera trovano il 50% delle persone sepolte sotto la neve, non riuscirono a individuare una persona intrappolata a più di un metro sotto la superficie. La neve più profonda richiede le sonde con pali, sonar oppure sensori ad infrarosso, ma la ricerca diventa assai difficile quando una forte nevicata è in atto.

La mitigazione . Le valanghe del 1970 hanno dato luogo ad un potenziamento delle previsioni del tempo che causa le valanghe. Ma gli eventi di aprile 1970 non furono di un tipo particolarmente prevedibile, ed inoltre una previsione è valida soltanto se viene collegata con un programma di structural and non-structural mitigation, including the planning of human settlements and the evacuation of populations during periods of highest risk.

For many years the French and Swiss have collected and compiled data on avalanche risk charts at scales ranging from 1:10.000 to 1:50.000. During the 70 municipalities of Alpine France undertook to plan their urban growth on the basis of risk mapping of flood by prohibiting new construction in areas "red" and limiting the areas 'blue'. However, some similar initiatives in Switzerland have been hampered by the division of resposabilità for urban zoning between the governments of the cantons and the federal state. This situation brought a remarkable heterogeneity of planning legislation against the danger of avalanches.

Despite this, the Swiss have made extensive use of structural methods nonstrutturali and mitigation of avalanches, using a good combination of measures. Instead, the Italians, for the most part, have used expensive structural measures, which were contrabilanciate by massive deforestation and a major development of sports facilities.

In summary, the Alpine winter disaster of 1970 represented an excellent window of opportunity "to introduce certain mitigation measures. However, the expansion of tourism and mountain sports has helped keep fairly constant number of disasters due to avalanches, although their amount is decreased due to better forecasting and planning against the Avalanche.

Aberfan TO LANDSLIDE IN WALES

brief summary. A pile of debris from a coal mine collapsed on Aberfan, a mining town in South Wales. The landslide destroyed 18 homes and an elementary school and killed 116 children and 18 adults. The event

. At 09:15 of October 21, 1966, landslide 117,000 m of debris from a coal mine and piled in a heap of height 67 meters. In a series of waves, the water-saturated debris flow traveled at 15-30 km / hr along a slope E. slope of 12.5 Approximately 75,000 m of debris stood at the foot of the heap, but 42,000 ft, traveled another 100 meters and came in the middle of the adjacent village of Aberfan (pop. 4,000), forming a deposit often 7-9 meters. The debris flow

demolished two houses and crushed their occupants. Later destroyed a school full of children and 18 homes in the country. Of the 144 people affected by the flow is not saved. Among the dead were five teachers and 109 school children aged between 7 and 10 years. Seven other children died and 29 other children and six adults were injured, some seriously. Several houses were flooded by water coming from the lava flows and broken un'acquedotto.

Conditions antestante . Since the '20s the debris from the mine had been stored in 7 piles of height between 17 and 67 m. In 1962 he began a process of refining of coal that left pieces of scrap were later dumped in seventh pile of ashes which made up 39% and 6-8% water. Consequently, this combination had a low density and was saturated with water from a natural spring located mainly in the mass of debris. No report was ever made to geological sites of the seven piles.

In the 70s Britain produced and released into open some 50 million tons of rock material of the type that has crumbled to Aberfan. In 1939 about 180,000 tons of debris from a mine near Aberfan had slid for a distance of 400 m across the main street of the area. But in 1966 the villagers took care of most of the risk of flooding, given that several times during wet periods the water flowed in the streets and entered houses. In fact, despite the advent of small movements in 1944 and 1963, the engineers did not anticipate that happening in coarse materials flows as well as those of the heaps of debris from the mines.

The reaction to the disaster. The disaster was followed immediately by a massive "response to convergence", where the labor force and help arrived at Aberfan in excess. Accumulating a fund for the victims, from 1,750,000 in donations, not only from all over the UK but also from 40 other nations. Despite a series of disputes between the trustees and the survivors of the disaster, the money was well used to compensate those who denounced the material losses, to create a memorial to the victims, and to build a community center for the country. However it could not mitigate the fact that a very supportive community had lost almost an entire generation of his children.

A question of responsibility. At first, neither the Welsh Office of the British government nor the National Board for Coal (NCB), admitted responsibility for the disaster. The villagers of Aberfan were fighting with all their strength to prosecute the guilty, but was found a thick resistance from the authorities. In addition, the NCB refused to raze the mounds of debris, which continued to constitute a danger to the country is a symbol of disgrace. Finally, public opinion forced the government to allocate funds for the removal of all the piles 7. The court of inquiry found 9 dell'NCB officers guilty of gross negligence, but the law was not that they could be prosecuted, and in fact even the dell'NCB Director had resigned. In summary, despite clear indications to the contrary, for many years before the disaster, the NCB repeatedly assured the people of Aberfan that there was danger of a collapse in heaps of debris. Following the disaster denied any responsibility for it was not forced by the court to admit his guilt. But the court failed to prosecute persons who with their actions had caused the disaster.

A notable aspect of the disaster in Aberfan is the total lack of plans, regulations, and standards expertise to manage the process of unloading around the mines. As usually happens in such cases, immediately after the disaster the laws were updated to remedy. But by then the eminent engineer Prof. Alan Bishop of Imperial College, London University, had shown by calculation and physical models which, under the circumstances, Aberfan era un disastro inevitabile.

Aspetti umani della tragedia . Dato un certo consenso al fatto che i minatori pagano un prezzo sproporzionato per la creazione del benessere della società, gli abitanti del Regno Unito e di altri paesi provavano grande compassione e simpatia verso Aberfan. Essendo profondamente scossi dal disastro, molti abitanti, poco abituati a stare al centro dell'attenzione, non erano felici dell'invadenza del mondo esterno. Però, la solidarietà del paese ha retto e poche persone e famiglie si sono traslocate al di fuori della zona. Ma tale era la scossa psicologica che molti dei sopravissuti di Aberfan negavano per anni che la tragedia fosse successa. Come nella colata detritica di Buffalo Creek, Pennsylvania (Also from a mine), which killed 118 people in 1972, among the victims of Aberfan was noted a "survivor syndrome", consisting of an inner terror, a tendency to relive the tragedy and a state of "psychological paralysis "or a" temporary collapse of ego. "

Finally, the deadly mudslides caused by mining activities continue to happen, as the Italian case is one example of the Val Was, 1985, in which 264 people died.

LANDSLIDE IN ANCONA

brief summary. The landslide of 1982 interested in 341 acres of the northern city of Ancona. It destroyed 476 buildings and left 3,661 homeless.
The event
. During the night of December 13, 1982 a landslide interested in the slope of the hillock at the north side of the city of Ancona and Marche on the coast of the Adriatic Sea. 294 residential buildings were damaged, consisting of 1,025 apartments and 182 other buildings, including two hospitals, a gym and the Faculty of Medicine, University of Ancona, some churches, a cemetery, several factories, a few farms, a barracks of the forces of 'order, and some workshops and industrial applications. The victims were 3,661, of whom one patient died when he had to be evacuated from the intensive care unit of one of the hospitals affected by the landslide. The town of

Borghetto had to be entirely demolished, and some parts of the suburbs of Ancona Posatora, Pinocchio, Palombella and Towers suffered serious damage. Severe impacts were also found by the road through the area and the Adriatic railway, which passes over the foot of the landslide. In addition, light pipes, gas and electricity were cut. The cost of the damage was estimated as £ 1 trillion.

causes. The move came after several weeks of rain. While the seismic shaking could have contributed to the destabilization of the slope, was not a trigger for the movement of 1982. The causes included the long-term effect in increasing coastal erosion the steepness of the slope, and mismanagement of the infiltration of rain and drainage of the slope in areas of recent and current urban growth. The main cause was the high short-term pore pressure of soil moisture, especially at the top of the slope.

History . At the point the slope of the landslide Montagnolo reaches 251 m above sea level, is off my 1400 has a slope that varies from 9 to 18E. The area was affected by problems of stability since 1770, when postal links were interrupted frequently by surface movements on the old road through the area. At the beginning of the twentieth century, a landslide destroyed 16 hectares of the tannery Barden and licked the rail at the base of the slope. An engineer called by the State Railway to make a geological survey concluded that the slope of the exceptional measures needed to ensure its stability. Barden located upstream of the landslide, the old house is placed upstream spectacularly revolved around 3E, constituting a clear testimony to the instability of the area.

In the 70s the General Regulatory Plan of Ancona took a large urban sprawl to the north of the city. In 1970 an employee of the Geological Survey of state he predicts accurately the mechanism and extent of the landslide (with 12 years in advance) and suggested a series of engineering measures to consolidate the area at risk. However, very few of the proposed works were built. Instead, the urbanization of the area continued with the construction of houses and buildings in the middle of the unstable area and the faculty of medicine at the foot of the old landslide Barden. Further geotechnical report, published in the Bulletin of the Italian Geological Society in 1974, concluded that the area was essentially stable, although, strangely, the data presented in that work showed exactly the opposite.

The mechanism of movement. The slope of a lithology Montagnolo there is a little complicated, which is composed of clay cracked blue board, with varying degrees of compactness, and a slightly consolidated sand. The geological investigations following the disaster concluded that the landslide took place on a cutting plane located up to 120 m below the surface of the slope. However, data from drilling and inclinometers did not confirm this hypothesis, and is more likely that the extraordinary size of the movement is a progressive movement of plates of land due to a change in the type "shear deformation" at a depth no greater than 40 -45 m, with involvement of the rest of the slope because of the shift of its overall balance.

teleosservazione The infrared shows the presence of two substantial patches of soil moisture, one just below the crown of the head of the landslide detachment, and the other near di una trincea di subsidenza che si apriva in mezzo della frana a causa dei movimenti distensionali franosi e della presenza di una faglia normale che attraversa il versante in direzione nord-sud.

Il comportamento degli edifici . La grande frana di Ancona ha fornito molta informazione sulla reazione di edifici di vari tipi alle forze e ai movimenti indotti da un grosso e complesso movimento franoso. Le strutture più vecchie erano di mattoni rossi cotti, di modeste dimensioni e quasi prive di fondazioni. Invece, le strutture più moderne erano per la maggior parte di cemento armato, sebbene nella zona vi fosse qualche struttura in acciaio. La spaccatura dovuta alla frana tendeva ad esser unidirezionale e a risultare da una pressione sostenuta causata dalla subsidenza, dal rigonfiamento o dalla spinta proveniente dal movimento franoso. Dove la subsidenza differenziale ha interessato le fondazioni, un edificio con struttura portante in cemento armato poteva inclinarsi fino a 6,5E senza crollare, ma nella maggior parte dei casi un'inclinazione di 2-3E era sufficiente per compromettere la stabilità dell'edifico irrimediabilmente. Se i pannelli della facciata fossero ben ancorati alla struttura portante, un edificio in acciaio poteva resistere la conpressione senza crollare fino ad un metro di deformazione.

In genere, gli edifici in cemento armato funzionavano meglio di tutti gli altri sotto il carico della frana, dato che il grado di integrazione delle loro strutture permetteva loro di resistere massive forces. The risk was highest where buildings were located through the steep, landslide, and then where they suffered from severe differential movements, which might tilt, swivel, compression or distension of their structures.

Conclusion. In summary, the Ancona landslide disaster was another highly anticipated and in fact often provided with ample time in advance. While interpretations may vary, the data showed very clearly that the presence of a high degree of risk of instability were freely available throughout the period of urban expansion antestante the disaster. In the years before and after the event, almost every aspect of the disaster was politicizzato, con l'effetto che si stentava e si stenta ancora a separare la realtà oggettiva dalle speculazioni e dalle opinioni prive di fondamenta. Con il senno di poi, la questione principale ancora da risolvere è quanto il disastro risultava da una trasformazione ambientale basata su un'ignoranza del rischio di frana e quanto invece essa risultava da una speculazione che deliberatamente ignorava il rischio. La catastrofe, con il suo elevato costo in termini di denaro pubblico, dimostra il pericolo, non soltanto di una mancanza di controllo ambientale dei processi di urbanizzazione, ma anche dei rischi inerenti a un controllo in cui il colpevole è anche il giudice e la giuria del risultante processo.