Cover Image
close this bookManaging Natural Disasters and the Environment (World Bank, 1991, 232 p.)
close this folderDevelopment: from vulnerability to resilience
View the documentCase study: housing reconstruction in Mexico City
View the documentLiving with floods: alternatives for riverine flood mitigation
View the documentCase study: creating job and income opportunities for refugees in Pakistan
View the documentManaging drought and locust invasions in Africa
View the documentDisasters and development in East Africa
View the documentThe link between reconstruction and development

Case study: housing reconstruction in Mexico City

Alcira Kreimer and Edward Echeverria

The earthquake that struck Mexico City in September 1985 took more than 5,000 lives and damaged the housing of about 180,000 families. RHP, the agency that was set up three weeks later to rebuild urban areas damaged by the earthquake, is a textbook example of successful reconstruction. By July 1987, only 14 months later, RHP had rebuilt 45,100 dwellings - an average of 3,220 dwellings a month. Today one of every seven families living in the city’s historic center has a new or rehabilitated RHP dwelling. This was one of the largest reconstruction programs since the recovery from World War II. Almost all of the federal and city development and management agencies contributed to reconstruction. More important, the beneficiaries - the earthquake victims - helped daily to expedite decisions and construction. More than 1,200 private companies participated in the program and more than 175,000 jobs were created, but by May 1987 RHP had begun reducing its staff and most personnel had returned to their former agencies. As Manuel Aguilera Gomez, RHP’s director general, wrote afterward: “We all learned to conciliate the desirable with the feasible. We learned to listen with care and interest to the sentiments of those affected by reconstruction. Little by little - in stages - the attitudes of the program beneficiaries changed from hostility, uncertainty, incredulity, suspicion, and doubt to hope and confidence.”

On September 19, 1985, at 7:19 a.m., Mexico City was struck by an earthquake that measured 8.1 on the Richter scale and lasted more than a minute and a half. The next day there were a number of lesser quakes, the strongest of which measured 7.8 degrees. The maximum horizontal acceleration was nearly 20 percent of gravity on a dominant two-second cycle. This ground movement resonant cycle coincided with the natural vibration period of the five- to 12-story buildings that predominate in the city’s dense historic center - making the earthquake one of the most destructive in the hemisphere’s history. Poorly built tenements housing low-income families in overcrowded conditions suffered the worst damage. They had already deteriorated from lack of maintenance and repair. Tenement rents had been frozen since World War II so there had been no incentive for rehabilitation. The catastrophe took more than 5,000 lives, caused 16,000 injuries, and damaged or destroyed 12,700 buildings - 65 percent of them residential. The housing of about 180,000 families was damaged and 50,000 people had to be temporarily rehoused. Also affected were 340 office buildings in which 145,000 government workers were employed, plus 1,200 small industrial workshops, 1,700 hotel rooms, 1,200 schools, and 2,000 hospital beds. The loss exceeded US$4 billion as calculated by the Ministry of Finance and the Economic Commission for Latin America and the Caribbean (CEPAL).

Housing reconstruction

The government of Mexico asked the World Bank for assistance for the reconstruction of hospitals, schools, and low-income housing and for research into revised building codes, zoning, and regulatory measures to reduce the city’s vulnerability to earthquakes. There were four rehousing programs:

· Popular Housing Reconstruction (RHP) - 48,000 dwelling units, benefiting 260,000 people, reconstructed onsite on expropriated sites.

· Phase II - 12,000 dwellings on nonexpropriated sites.

· Casa Propia - 8,000 dwellings rehabilitated for resident owners.

· Housing Foundation (FOVI) - 12,000 units of relocated housing.

What follows is a description only of the first of these, RHP - a success story in emergency construction and a model for community involvement.

Popular Housing Reconstruction

On October 14,1985 (just three weeks after the disaster), RHP (Popular Housing Reconstruction) was set up by presidential decree as an autonomous agency with a life of two years (see box by Manuel Aguilera). RHP had a mandate to:

· Rebuild and reorganize urban areas damaged by the earthquake, following the principles of urban renewal and social development.

· Define a policy of social development that preserves and protects the physical and social patterns of urban life, guarantees ownership of the dwellings to the beneficiaries, and provides needed urban services.

· Combat land speculation.

· Rationalize the building finance and investment that would be channeled to the program.

The program was to unfold in five stages:

Stage 1. October 1985 - March 1986. Damage assessment, planning and design.
Stage 2. April - December 1986. Intense construction and social organizing.
Stage 3. January - March 1987. Allocation of dwellings, legalization and registration of deeds.
Stage 4. April - September 1987. Completion of program.
Stage 5. October 1987 - April 1988. Diagnostic history, records, and closure.


In the first months after the earthquake, RHP updated an initial survey to estimate the number of people affected, their socioeconomic characteristics, and the physical condition of their dwellings. On the basis of this census, the victims were awarded certificates validating their eligibility for housing assistance. Early proposals for reconstruction focused on vacant land in outlying areas, including a site adjacent to the airport. But World Bank financing was contingent on rebuilding onsite with minimal relocation, a policy based on negative experiences the Bank had had with large-scale relocation in other disaster areas. Most families had lived in their neighborhoods for a generation or more and wanted to remain there, so the government adopted a policy of reconstruction onsite. 1

This decision required expropriation of privately held land and the provision of temporary shelter by families in the immediate vicinity. This called for both political and administrative skill and enormous sensitivity in dealing intimately, day in, day out, with 60,000 families for more than a year and a half.

On October 11,1985, the Expropriation Decree was published in the Official Gazette. Some greeted it with appreciation for its social justice; others condemned it as populist and demagogic for its violation of property rights. Errors and omissions needed correcting and individual cases were protested in the courts, but the decree itself was successfully administered with a taking of 4,312 lots or 200 hectares (500 acres).

The Expropriation Decree announced that meeting the collective needs of the people whose homes were destroyed by the earthquake was in the public interest; that the city government was to occupy the property immediately, authorize its upgrading and renewal, and sell the new housing to the people who had been living there; and that the city government was to pay compensation to the former owners within 10 years, according to the capacity of the Treasury.

By January 1986, the RHP was reorganized to create two departments, Construction and Administration, at the same level as the Office of the Director General. The most important change was the decentralization of the Construction Department into five zonal offices in charge of supervising and controlling construction and of building temporary shelters. To reinforce the core staff, RHP borrowed senior planners and engineers from the Ministries of Communications and Transport, the Secretariat of Urban Development and Ecology (SEDUE), the Federal Electricity Commission, the Urban Transportation Commission, and many agencies of the Federal District (DF) of Mexico City. The zonal offices were further decentralized into 12 construction and operations modules that managed all construction activities.

In January 1986 social and technical teams started to match socioeconomic survey data to data on the physical condition of each dwelling. In the process they organized the earthquake victims into community groups to review the whole program, site plans, and prototype apartment designs. Based on the census, the government issued certificates of residency establishing earthquake victims’ eligibility for benefits. These were issued regardless of who legally owned the building. Many families had abandoned their dwellings immediately after the earthquake, however, so it was difficult to track them down to document their rights to new dwellings. Regular meetings were held to maintain the quake victims’ social organization, to review and revise program plans, and finally to approve plans and sign documents for the construction of each apartment. For many groups, planning and redesign took as long as eight to nine months, construction only four. Only after agreement was reached were beneficiaries legally formed into a condominium association that agreed to vacate the building so reconstruction could start. Temporary shelters of corrugated aluminum or zinc were generally located within a block of the building site. Wherever possible, they were built on public lands - parks, sport centers, roadway median strips, service roads, and sometimes actually in adjacent streets.

The social teams organized the condominium associations into “renovation councils” for each reconstruction or rehabilitation site. Although they had no legal status they provided forums for people to speak out. The councils were formally installed with elected representatives authorized to negotiate with the RHP about individual needs. They also had to decide on the legal status of their housing association - whether to be a condominium, a cooperative, or a nonprofit organization.

Many people who had no previous experience in community action found themselves as spokespersons for their association. They became outspoken not only to the press but also to the RHP and city officials. The Director General and Director of Social Affairs spent long hours negotiating and responding to their concerns. The beneficiaries’ participation made the process far more rewarding for the federal government, the city, the community, and the beneficiaries.

During planning and design, five types of groups worked with the earthquake victims: political parties, university groups, technical support groups, and private voluntary and religious organizations. More than half the sites received support from one or more organizations, starting with the census survey. University groups began by evaluating damages and later acted as technical support. The political parties also played an important role.

By February 1986 the new “Personal Certificate of Rights” was developed to replace the original “Certificate of Residency,” to eliminate fraud and clarify other questions of need. RHP experienced great difficulty handling so much data and residents were reluctant to be interviewed again. There were two main problems. First, people were uncertain and mistrustful because, months after the earthquake, reconstruction had not begun. Second, there was discord about the size of dwellings. Many felt that the size of the dwelling should be proportionate to the titleholder’s ability to pay. But this was impractical, given the number of beneficiaries and the time constraints. By the end of February the Social Development Office issued a “Handbook of Social Procedures” explaining how disaster victims could get the replacement housing conceded to be their right.

The technical staff of architects, planners, and engineers labored over criteria to distinguish which damages were caused by the earthquake and which by physical deterioration of the dwellings. The detailed building survey showed that the earthquake had damaged or destroyed 59 percent of the buildings beyond repair or rehabilitation. A third had deteriorated because of neglect and some of these would have to be rebuilt. Rehousing on the expropriated sites was initially set at 46,700 dwellings, of which 23,200 (half) were to be rebuilt, 14,900 (32 percent) rehabilitated, and 8,600 (18 percent) improved by minor repairs. Exposing the structural elements revealed that many buildings slated for rehabilitation had to be replaced. So the number of reconstructed dwellings increased to 35,900.

Reducing urban and natural risks in Mexico City

Manuel Aguilera Gomez

The director of Mexico City’s incredible housing reconstruction effort after the earthquake of 1985 describes what happens when artificially induced ecological changes heighten a city’s natural vulnerability. He calls for a reorientation of urban development - a qualitative long-term change to close the gap between the city’s urbanization pattern and degree of vulnerability. He stresses the need for community development over the continuing development of a metropolitan program vast beyond reason.

In this last decade of the twentieth century, Mexico City - located more than 2,000 meters high, on an enormous lakebed - has become one of the largest metropolises in the world. Risk and disaster have always been part of life in the nation’s capital. Home to more than 15 million Mexicans (about 19 percent of the country’s population) who live in the Federal District and 17 surrounding municipalities, the city is a disjointed assembly of mixed urban habitats.

These habitats - located in the Valley of Mexico, surrounded by volcanos and mountain chains - differ in their subsoil, altitude, and degree of modernization. Their major infrastructure and service systems are not integrated, and a complex network of regulations and overlapping administrative jurisdictions operate in isolation, with no sense of metropolitan unity. The metropolitan area now covers nearly 1,500 square kilometers.

Economic, political, and cultural factors have all played a part in Mexico City’s current problems. The period of fastest urban growth in the valley was from about 1955 to 1982, when expansion averaged 3 percent yearly. This period coincided with a general agricultural crisis in Mexico and the adoption of a development model based on a strong concentration of industry and services. At present, nearly 30 million industrial plants are located in the Federal District alone, which covers 55 percent of the metropolitan area. Although rural migration to the city has fallen off in the last five years, it remains high, with about 300,000 new arrivals each year. Improvements in the Federal District’s demographic policy have reduced the annual rate to 1.2 percent, but the same is not true of outlying areas in the State of Mexico, such as Chalco, in the Northeast of the metropolitan region, where the population is growing more than 5 percent a year.

The city’s size, artificially induced ecological changes, and constant changes in the use and allocation of urban land have clearly heightened the city’s vulnerability. When urbanization pushes the rate of expansion beyond a certain point, risk and vulnerability increase and become a constant factor that must be considered in the design and implementation of urban development policies. The city now lives with permanent risks of earthquake, landslides, floods, severe traffic congestion, and interruptions of the water and power supply to certain critical regions.

We live in an urban complex that began to deteriorate ecologically more than 400 years ago, when the Spanish conquistadors began the prolonged drainage of Lake Texcoco. This was completed only at the start of this century, but the consequences of this action are still at the root of many of our problems. In our time we have seen a city rapidly modernizing but as a result it is a victim of bold architecture that sometimes defies the swampy and geological nature of the basin of Mexico.

Its residents are aware of overexposure to contingencies of varied nature and magnitude. But not until the 1985 earthquakes did we become fully aware of our profound vulnerability as an urban community. The painful experience of the 1985 earthquakes illustrates the vulnerability we live with daily and the community’s ability to organize quickly in response to devastation from natural disasters.

The earthquake on September 9, 1985, reached 8.1 on the Rich-tor scale. In parts of the city with soft and humid subsoils, or because of amplification, the intensity topped 10. The effects of the seismic waves caused the death of more than 5,000 people and damaged buildings more than eight stories high located in the epicenter, the central-eastern part of the city. Similarly, the duration of the earthquake destroyed older neighborhoods, multifamily dwellings, and such strategic service networks as the water supply, drainage, and main roads and avenues - thus delaying rescue operations.

But as a witness and actor in this period in the history of our capital, as head of the Popular Housing Reconstruction program, I can state that the emergency and subsequent reconstruction were times of unforgettable solidarity and human effort. Few countries and cities have gotten back on their feet in so short a time. Reconstruction was overwhelmingly successful. For example, 48,000 dwellings in the hardest hit areas were raised from the rubble - in the poor but well-serviced center of the city - within 18 months. Virtually all of the city’s infrastructure and service systems were restored in only three months.

The terrible lesson of the earthquakes has resulted in increased community awareness of not only the constant risks but also vulnerability caused by so-called over-urbanization. The first step in reducing the vulnerability of the Federal District was the redesign and implementation of an integrated civil protection system. The second was a reorientation of the urbanization process and the city’s development model.


The first task was to identify the most vulnerable zones and regions, especially the nature and scope of risks in the central and northeastern part of the city which, because of its soil and structures, is most likely to be affected by geological phenomena. The belt of gullies, depressions, and ravines that crosses the city from west to southeast has the highest risk of cave-ins and land faults, as well as flooding because of intense annual rainfall and inadequate storm drainage. The basic purpose of the civil protection system in the Federal District is to guarantee an organized, speedy, and efficient government and community response to any emergency and to coordinate joint efforts to restore normalcy in services and the rhythm of daily life.

The project to modernize civil protection has five strategic guidelines:

· Fostering solidarity.

· Local, decentralized responses to emergencies.

· Adequate training for each population group by zone and type of activity.

· Coordination of the public agencies and the community to give the rescue effort a sense of unity and balance.

· An international exchange of experiences to introduce Mexico City to advanced technologies for the prevention of, and response to, disasters.

The spirit behind these strategies is solidarity and grassroots participation. All of our equipment and trained personnel are useless in a catastrophe if we cannot incorporate the solidarity, civic-mindedness, and the responsibility demonstrated by the millions of anonymous citizens who saved Mexico City in 1985. Specialized teams of volunteers must work with groups in the community, contributing their resources and personnel to promote more collective security. By the end of 1989, neighborhood organizing efforts in the Federal District had attracted 16,000 volunteers to the civil protection system. The second phase will tap the imagination and talent of these volunteers for tasks of such magnitude and importance that we must all join in.


Mexico City has reached a stage in its development that requires medium- and long-term qualitative changes to close the gap between the city’s urbanization pattern and degree of vulnerability. We cannot say that Mexico City will stop growing or that its vast problems of pollution and vulnerability will cease to exist or continue to increase if we reduce urban growth. But the urbanization pattern must be reoriented to stabilize the pollution and vulnerability indices, so that we can gradually make overall improvement and restore environmental quality.

For many years, urban growth meant that most investments went to expand such major infrastructure systems as the water supply, storm drainage, and the subway. Improvements were virtually the exclusive preserve of the modern microregions of the city, to the detriment of the poorer, deteriorated areas. True, many of the major infrastructure systems must continue to provide a centralized service that serves the urban whole. But most government action - whether to reduce pollution, improve the urban space, or diminish the uncertainty of foreseeable risks - should be a local, decentralized response to the priority demands of specific, localized social entities.

In the future, most investments in Mexico City must be made in mixed urban habitats. If they are to be integrated in a just and fair manner, community development must be stressed over the continuing growth of vast metropolitan programs. Aggregate demand must yield to priority demand - which should be met primarily by use of local resources.

Through group representation, communities must participate directly in managing everything that bears on their daily lives. This is the true meaning of democracy and the only real way for social justice to prevail in one of the largest cities in the world.

The technical staff, together with architectural consulting firms, selected three prototype designs for reconstruction. RHP had to take a pragmatic approach to unit size, rebuilding for an average family of 4.6 persons. Each dwelling unit has a minimum net area of 40 square meters, a kitchen/living/dining room with hot water, two bedrooms, a complete bathroom with shower, and a small laundry patio or balcony. The structures are all earthquake-resistant.

The architects and planners were ingenious in providing a diversity of building shapes and heights with a striking use of color and texture. Most buildings are three stories high, with grade beams of 20 x 40 centimeters, load-bearing walls, and reinforced concrete slabs. The structural beams on each floor are tied to the columns to provide a rigid structural frame. The common walls serve as shear walls to provide added rigidity. Many of the sites had such poor (highly compressible, saturated) soil conditions that as much as 1.2 meters had to be excavated and replaced with compacted crushed stone.


The bureaucratic procedures for plan approvals, construction, modification, and completion of each building were streamlined between RHP, the Low-Income Housing Fund (FONHAPO), the office of the Federal District (DDF), and the National Bank of Works and Public Services (BANOBRAS). Plans and construction contracts for more than 3,000 sites had to be approved, so it was imperative to reduce approval time to days instead of months. Close daily supervision by zonal offices, community groups, and future condominium owners kept the construction on schedule. A technical team from FONHAPO helped RHP administer the loans. The average cost of the dwellings on the first group of sites was US$4,030, so repayment was based on a down payment of 10 percent and the rest, US$3,630, was amortized over 5.5 to 8.5 years, depending on the purchaser’s ability to pay. Monthly payments were set at 20 percent to 30 percent of one minimum wage indexed to inflation. The payments were adjusted once a year using a computerized formula. A special municipal trust was activated to receive the monthly payments and maintain the loan portfolios. As construction expanded to hundreds of sites, construction management teams had to streamline their daily accounting and budgeting systems, which quickly reached a peak expenditure of US$1 million a day. Auditors from the National Chamber of Deputies periodically reviewed the financial accounts. Their reports were published for public commentary - which was continually favorable.


Under the emergency act to expedite procedures, in July 1986 RHP called for tenders on delivery of 12,000 toilets and 32,000 wash basins, kitchen sinks, and water heaters. The government and World Bank procurement guidelines on local and international competitive bidding were fully respected but early approval of documents and down payments ensured on-time delivery. Savings of about US$1.5 million were achieved by the bulk purchase, despite the difficulty of storing such quantities of goods.

By mid-1986, RHP had been asked to rehabilitate or restore more than 100 buildings considered to be of historic value, providing residential living space, security, and adequate sanitation within the program’s budget. Project designers from the National Institute of Anthropology of History helped identify buildings that were candidates for restoration and conversion to residential use. In many of these old “mesones,” densities before the earthquake were one family per room, so many families had to be relocated. The decision of whom to move was left to the group. Residents on the alternative site had accepted the move of those who were relocated. On acceptance, final sales prices and plans - including the date construction was to start, the date people were to move to temporary shelters, and the agreement to receive monthly rent during reconstruction - had to be ratified by all partners.

Help from social and community groups and universities was essential. Interdisciplinary groups of students of medicine, psychology, sociology, architecture, and engineering - coordinated by RHP social services - provided the much-needed social services in the temporary shelters, while RHP paid for utilities and provided security, fumigation, and maintenance and repairs to the sanitary installations. As a rule, the beneficiaries were willing to accept the rules of the reconstruction program.

Demand for temporary shelters increased and it was difficult to build more of them in the streets, so in May 1986 SEDUE acquired 28 hectares next to the airport on which to build 1,200 prefabricated dwellings. These were occupied initially by one family per room. When reconstruction was complete, they were modified back to one family per four rooms of about 48 square meters and sold to the beneficiaries. A total of 20,000 temporary dwellings were built. As a complementary measure, rent assistance was increased to significantly reduce the demand for temporary shelters.

By mid-December RHP technical and administrative teams were so immersed in construction management, with many sites operating around the clock, that they chose to work through their normal Christmas holiday to advance the program. In the first eight months of construction, RHP’s construction management team maintained its output at the rate of more than 2,600 dwellings a month - a remarkable achievement. By the end of December 1986 (the end of Stage 2), 21,200 dwellings had been completed and 10,437 of them were already occupied. Dwellings were finished and assigned so quickly that it was necessary to speed up removal of the temporary camps - first those on the streets, then those on expropriated lots.


In January 1987, RHP began to restore the streets and parks that had been occupied by the temporary shelters. Sidewalks were repaired and the area was cleaned up, gardened, and repaved. On February 6th the International Union of Architects announced that RHP had won an international architectural award, the Sir Robert Matthew Gold Medal award. The reconstruction program was considered by the Architect’s Union “the best piece of work on human settlements carried out internationally in the previous three years.”

During this period, the Legal Department was verifying the massive data the notaries needed to legalize the transfer of deeds to the new owners and to register them in the Public Registry of Properties. This involved deeds for permanent housing on 2,870 lots constituted as condominiums, along with 46,720 individual deeds for each dwelling or commercial outlet. In addition, the 1,200 prefabricated dwellings near the airport were divided into 11 lots for deed registration. The Social Development Department undertook the problem of verifying the technical and social historical data on each lot. Each site development lot required a complete technical-social file. There were external problems as well. Many of the lots were unregistered and the Cadastral Office of Deeds had incomplete or no data. The Public Register of Properties was not equipped to handle this enormous load, so RHP had to send its personnel and computer data equipment to help check out the background information of the register in the Treasury, the Legal Department of the City Government, and the Tribunals. By the end of March 1987, they had distributed 1,807 deeds for individual dwellings and 188 condominium deeds. Construction was virtually complete, but on March 31 the President of the Republic authorized that “closure” activities continue until the end of September 1987, that the Low Income Housing Loan Trust Fund (FIDERE) would recover the loans, and that the City Government would take over all of RHP’s rights and obligations on RHP’s demise.


By July 1987 the reconstruction of 45,100 dwellings was completed. It had taken 14 months, with an average 3,220 dwellings a month. During the peak period of February-April 1987, RHP was completing more than 120 dwellings a day. To process final payments to the more than 700 construction contractors, on the more than 10,000 contracts of supplies and services, RHP developed a computer program that indexed construction costs to inflation and completion schedules. Final estimates were reviewed by building control personnel who sent them to finance personnel for verification. Contractors who owed money to RHP completed payment by certified check. If payments were due the contractor, a final balance sheet was sent to the financial department for final payment. RHP would return bond security on the advanced payment, leaving the bond covering construction warranties.

By the end of September, the rent payment assistance program was closed. It had benefited 19,900 families with an average of 364,704 pesos (US$750) a family. Rent assistance, which lasted an average eight months per family, included payments for moving furniture and personal possessions twice. The loan portfolio for recovering the loans for 42,000 dwellings and commercial outlets and 2,745 lots (or 94 percent of the total) - together with the hardware and software used to administer the system - was handed over to the financial trust, FIDERE. By September 1987, the gross repayment (monthly payments, advance payments, and insurance) had reached about US$ 10 million.

Decentralizing Mexico’s health care facilities

Edward Echeverria

The earthquake of 1985 caused disproportionately heavy damage to Mexico’s health care facilities because they were concentrated in the capital city center. The Ministry of Health’s Centro Medico (3,000 beds) and the Central Hospitals of the Social Security Institute (IMSS, 2,600 beds) - which included important Mexico City hospitals - were virtually destroyed. Immediately after the quake, plans to rebuild these health care facilities followed the national strategy of decentralizing federal government functions to other states.

Health care reconstruction

The Government of Mexico (GOM) took an integrated approach to decentralization. Financing and investments were coordinated at the federal level, planning and programs at state and municipal levels. The World Bank had supported a policy of decentralization since 1985, helping the GOM in projects aimed at achieving spatial decentralization by developing alternative growth poles outside of Mexico City. The earthquake and reconstruction provided an opportunity to execute this policy.

IMSS, the second most important health care provider in Mexico, serves 40 percent of the population: workers covered by health insurance. In the last 20 years, IMSS has gained extensive experience in the design, construction, and operation of health care facilities throughout the country. IMSS’s technical design office, which had a deconcentration plan, organized and managed the replacement of 2,000 beds destroyed by the earthquake. It proposed to provide about 1,200 beds in six “second-level” zonal hospitals to serve an estimated 1.2 million people on the periphery of Metropolitan Mexico City. Each hospital would provide ambulatory and hospital services, including gynecology, obstetrics, pediatrics, general surgery, internal medicine, orthopedics, trauma, ENT (ear, nose and throat), and ophthalmology. These zonal hospitals would take care of 95 percent of the cases locally, eliminating the need to travel to the Centro Medico - which henceforth would provide specialized “third-level” services, with only 300 beds. Before the earthquake, about 40 percent of IMSS hospital beds were in the city center, more than two hours from most of the 7 to 10 million IMSS beneficiaries living in the metropolitan area.

The remaining 800 beds were to be built in five regional hospitals distributed countrywide according to need. Some were new nursing units and health care facilities added to existing hospitals so that the five regions - Ciudad Obregon, Vera Cruz, Leon, Puebla, and Merida - could become fully autonomous in providing all types of health care. This would reduce further the need to transfer second- and third-level-care patients for treatment in Mexico City. These actions would improve the level of health services and make them more accessible, at lower unit costs. Costs would continue to be recovered through user and employer fees in accordance with established practices.

Cost and schedules

Four of the hospitals on the periphery of Metropolitan Mexico were built on schedule and operating in 1989. Problems in site acquisition delayed the other two. They had to be relocated, which meant revising site and building plans. Their completion was scheduled for September 1990. The five regional hospitals were completed, equipped, and operating in 1989.

Procuring medical equipment (especially the CAT scan) required a long lead time. Bids for more than US$44 million worth of equipment were finally opened in December 1987. Despite large price increases the project had to absorb, costs remained within the projected range of US$50,000 to $55,000 per bed for equipment and $50,000 per bed for construction.


With the help of personnel from the General Archives of the Nation, RHP began analyzing the documentation for the whole program. The structure and content of the General Archives were designed so data could be retrieved for research. A building was rehabilitated to store: (1) construction contracts, bidding documents, licenses, and papers for temporary housing; (2) social-legal documents and certificates of rights; (3) accounts, budgets, and data on sources of finance; and (4) information on communications with the media. Security measures were set up so that, by law, the documents would be available for seven years for audits, revisions, and research.

By May 1987, RHP had begun reducing its staff. Most personnel returned to their former agencies. Borrowed equipment and furniture were returned and prefabricated offices and warehouses dismantled. RHP donated much of its office equipment and many computers and vehicles to the city government, the Phase II housing program, and other housing organizations.

From suspicion to hope

RHP finished 45,100 dwellings and 3,600 commercial workshops. It was one of the largest reconstruction programs since the recovery from World War II. Today one of every seven families living in the historic center has a new or rehabilitated RHP dwelling. Clearly, Mexico City could renovate a major portion of the city when both the human and financial resources were mobilized. Almost all of the federal and city development and management agencies contributed to reconstruction. More important, the beneficiaries - the earthquake victims - helped daily to expedite decisions and construction.

A total of 1,240 private companies participated in the program - 738 building contractors, 64 professional firms of supervisors, 184 suppliers, and 258 firms preparing studies and projects. From October 1985 to December 1987, more than 175,000 jobs were created - including 1,200 for the RHP agency itself and many more in construction and services.

Toward the end of reconstruction, Manuel Aguilera Gomez, RHP’s director general, wrote, “The earthquake revealed the nakedness of part of the city center. The solution was not to hide the poor in the suburbs, but rather to provide them with appropriate housing. To achieve this aim we all learned to conciliate the desirable with the feasible. We learned to listen with care and interest to the sentiments of those affected by reconstruction. Little by little - in stages - the attitudes of the program beneficiaries changed from hostility, uncertainty, incredulity, suspicion, and doubt to hope and confidence.”


1. Data revealed that 65 percent of the families had lived in the neighborhood for more than 20 years, 18 percent for 10-19 years, and 15 percent for nine years or less. More than 97 percent of them rented their dwelling and 70 percent of the dwellings occupied less than 40 square meters. Moreover, 80 percent of the heads of household - 87 percent of the men and 69 percent of the women - said they were working. These data were consistent with data on the main economic activities of the city center.

Living with floods: alternatives for riverine flood mitigation

Frederick C. Cuny

Most efforts designed to reduce the effects of floods have focused on such structural measures as the construction of dams or embankments (polders, levees, and the like). Many of these large-scale, capital-intensive projects have been questioned on both technical and environmental grounds. Many development experts question whether large-scale flood control projects are economically suitable for the least-developed countries, since they increase the country’s debt significantly for little economic return. And some flood control projects may be counterproductive. Embankments may foster unrealistic expectations that all flooding can be prevented and stimulate movement onto floodplains, thereby increasing total risk. In recent years, there has been increased interest in alternative strategies for protecting the floodplains, especially in rural areas. A key strategy has been to encourage people living in rural areas and in some small communities to adapt to floods and to capture their benefits for economic development. Traditional rural societies have developed many ways to adapt to floods and their consequences. These strategies can often be adopted or modified into a national “living with floods” strategy. Where this strategy has been applied, it has been cost-effective, easy to implement, and compatible with the environment. These measures can be applied before and after floods. More important, they can be incorporated in long-term development programs at little extra cost.

Flooding along the low-lying plains near rivers is the most widespread hazard in the world. Historically, people have avoided living in flood-prone areas but as populations grow and land becomes scarce, more and more people are forced to use these areas for activities that floods can harm. Many societies have developed a portfolio of counterflood actions designed to control flooding, prevent disaster, and, where possible, to harness the floods for such uses as irrigation, navigation, and aquaculture.

The technology to control floods exists, but in recent years concern has grown about the environmental consequences of flood control measures undertaken over large areas. Some experts have challenged the cost-benefit claims of flood control proponents, arguing that the benefits are slow to accrue and cannot always be accurately accounted for. Many have questioned whether sophisticated flood control works are appropriate for developing countries, given their high costs and the countries’ added debt burdens.

As nations threatened by widespread flooding plan their development strategies, they must understand that there are options - ways to live with floods and harness their benefits with limited flood control efforts. Many of these approaches are cost-effective and can be carried out by local communities with help from government and development agencies. They capitalize on people’s self-reliance and in the long term reduce costs and speed up protection. Countries that adopt them can adopt capital-intensive programs and to a large extent eliminate costly, never-ending maintenance operations.

The importance of indigenous responses cannot be overemphasized. In remote rural areas where government assistance may be delayed or virtually impossible to provide, these responses may determine how quickly and effectively a family recovers. It is important not only to understand them but to ensure that outside responses do not inhibit or discourage people from applying them.

Preventing floods is not a universal solution. In many cases flood control is not only feasible but more practical because urban areas, vital infrastructure, and critical communication and transportation networks must be protected. But encouraging communities to adapt to floods can be considered as an alternative or complement to capital-intensive structural flood control measures (such as embankments, dams, diversions, and river draining works). Knowing the difference between types of floods is essential for choosing the appropriate strategy.

Types of floods

There are four basic types of floods: flash floods, standing floods, sea surges, and riverine floods.

Flash floods occur as a result of the rapid accumulation of runoff waters from a rainstorm in a mountainous or hilly area. The water usually collects in relatively confined areas - such as gullies, wadis, or arroyos - then cascades until it reaches another stream or a wider, less restrictive area where the water spreads out and its velocity is reduced. The speed of the flood and the debris it carries are what make flash floods dangerous. (Flood velocity is determined by the steepness of the grade of the confined area.) Historically, people have avoided living in constricted areas, although some economic activities (such as sand or gravel extraction) have been carried out there. With rapid, unchecked urbanization, this has changed. Many poor people are forced to live in these areas - so the threat to human life is growing annually. For the most part, the best strategy in arroyos is still prevention. Downstream structural measures, such as diversions and check dams, provide some protection.

Standing floods occur when accumulated rainwater can neither drain off the surface rapidly nor be absorbed quickly into the soils or the water table. Standing floods usually cover relatively small areas. Many are caused by poorly designed transportation networks, such as roads and railways, that cut across natural drains and cause the water to back up behind roadbeds. Little can be done to prevent standing floods except engineering works to collect the water, in canals, and transport it to natural drainage zones or pump it into passing streams or onto lands on which it will do no damage. If the cause of the floods is a man-made restriction, bridges or culverts can often reduce the floods to an acceptable level.

Coastal flooding can occur as a result of storm surges (wind-blown masses of water caused by tropical cyclones - also known as hurricanes or typhoons - or storm-related high tides). Residents of coastal areas have adopted a variety of measures to reduce the impact of these floods. They have built communities on raised platforms, for example, or houses on stilts, and seawalls and other structures to absorb the force of storm surges. They have adopted such escape strategies as evacuation and community shelters as a response to both high winds and floods.

Riverine floods occur when a river overflows its normal streambed because of heavy rains anywhere in the river’s watershed. Normally, the area of flooding can be predicted, usually on the basis of topography and past flood history. The overflow area - the floodplain - may be a few hundred meters or dozens of kilometers wide. As a river gets closer to the sea, it usually passes through an area of alluvial buildup. This area, known as the river’s delta or fan, is normally fairly flat - and rivers tend to meander in wide lazy Ss as they slow down and empty into the ocean. In these deltas, flooding can be widespread; the extent of the flood and the height of the waters can be affected by rains far up the watershed and by ocean tides, which can slow the discharge of the river and raise water levels at high tide.

Of all floods, riverine floods are the most difficult to control and the ones for which a “living with floods” strategy is most feasible. To understand why, let us examine the ecology of riverine areas and the benefits of riverine flooding.

Riverine ecology

Riverine areas are important ecological zones. The world’s great rivers carry minerals and nutrients to the seas and form delicate, highly complex delta ecosystems that are crucial in maintaining the balance of nature. Most important, the marshlands and swamps provide a bridge between salt- and freshwater environments. Next, the tidal and saltflats provide a way to keep excessive salt penetration from moving inland. Then, on the coastal plains, landforms build up and human habitation becomes feasible and productive. As the rivers slow down, they deposit sediment which gradually creates sandbars in the streambeds. The sandbars divide and redirect the waters, eroding some areas and building others up. The sediment buildup offshore eventually extends the land mass outward into the ocean.

The alluvium the rivers deposit is the key to the deltas’ richness. This sediment is a mixture of all the topsoils the rivers pass through. The heavier, rocky materials settle out farther upstream. The suspended solids that reach the fan are fine and light and usually rich in nutrients and humus. In these rich soils, agriculture flourishes; the deltas of the world are also the world’s breadbaskets. Rich soil and plentiful water ensures that farmers can grow crops annually, if not semiannually. Even a small amount of land can produce enough crops for societies to prosper - and it is not unusual for these areas to be densely populated. Agriculture is not all that flourishes in this environment. In waters rich in nutrients, fish abound - in both the rivers and the discharge zone.

A riverine environment shapes the daily life of societies along its banks. Agriculture and aquaculture dominate the economy. The people’s basic diet depends on abundant sources of water and the flora and fauna within.

A riverine environment promises perennial floods. Usually, not a year will go by without some part of the delta flooding to some degree. To an outsider it might appear that floods are something to which riverbank societies must adapt to survive but most often the disaster would be if floods did not occur periodically. The benefits of floods far outweigh their negative effects. Many rural societies welcome flooding. In Bangladesh, the Barsha festivals celebrate the flood season and, when waters cover some areas, people take to their boats and venture out of their villages to trade and renew commercial and familial links.

The benefits of flooding

After a severe flood, it is often hard to remember that floods also have a positive side. What are some of the benefits?

· Floods deposit rich silts and replenish top-soil with nutrients vital to agriculture. After widespread flooding, there is almost always a bumper crop the next harvest season that partially makes up for the losses from flooding.

· Floodwaters carry nutrients that stimulate fish development and increase the number of fish. Floods may restock fish in isolated ponds, lakes, or streams that do not flow year round.

· Some ecologists suggest that floods improve the natural varieties of food grains, destroying the weaker strains over time and permitting the stronger strains to thrive.

· Water left standing in fields may help recharge shallow aquifers. Floodwaters may replenish water supplies in lakes and ponds.

· Floodwaters purge the rural (and sometimes urban) environment. This can have a major impact on public health. Some observers have noted that diarrheal diseases usually decline after widespread floods.

· Floods deposit sandbars that can be seeded to form barrier islands. These can be used to expand land area and as barriers against tropical storm surges.

· Floods deposit silts that can be “mined.” At a minimum, sand can be collected for construction. More important, the mud can be used for topsoil, construction, and landfill. Possibly they can be mined for important - often valuable - minerals.

In addition to the known benefits, there are many untapped benefits. Floods could be used as a source of energy, for example.

When is a flood disastrous?

Sometimes riverine floods are so big that their harm outweighs their benefits. The negative effects of flooding include death, property loss, a cumulative increase in personal and national debt, the increased incidence of certain diseases, soil erosion, sandcasting, the penetration of saltwater into soils and aquifers, the siltation of rivers and irrigation canals, and damage to and the destruction of public infrastructure, roads, railway beds, and other transportation.

If flooding usually occurs annually, what separates the normal, uneventful flood from a disastrous flood? Usually the magnitude of the flood. That magnitude can be measured as the number of people killed, the extent of the damages (physical and economic), the incidence of increased disease, and other factors. Usually floods classified as disastrous are defined by a combination of these criteria. But these criteria are subjective, representing mostly the views of urban or industrial observers. When do people who live in riverine areas call a flood disastrous?

In many societies, people distinguish between an inundation and a flood. In 1978, Unnayan, a Bengali rural development agency, worked with flood victims after massive floods in West Bengal, India. Cultivators they interviewed indicated that a flood has occurred when water remains in the field long enough to waterlog, when water strands or drowns livestock, when water currents scour the land, when floods deposit sand that cannot be economically removed, when floodwaters increase the salinity of fields, when water rises above the housing plains, when successive floods prevent immediate recovery, or when fish ponds are inundated and fry and fingerlings are swept away. For landless agricultural laborers, a destructive flood is one that reduces their job prospects or prevents them from going elsewhere in search of work. For nonfarm villagers, a flood is defined as water penetrating and damaging commercial buildings, water penetrating and damaging housing, or water preventing the transport of goods. Most rural people said that a flood was “when waters rose faster than one could take preventive measures,” a response which indicates that if more warning could be given, floods might not be so destructive. This response implies a degree of indigenous local flood preparedness (Sen 1978a).

Urban people could define floods similarly to nonfarm villagers, adding that when floods occur normal business is disrupted, schools are closed for excessively long periods, food and fuel are hard to get, and basic services (such as water supplies, electricity, and communications) do not function for long periods.

Governments usually classify floods according to the level of government infrastructure that has been damaged or destroyed, and the number of houses and communities affected. Infrastructure that could be damaged include roads, lifelines and critical facilities, public buildings, schools, and health facilities. Rural people may not need or expect extensive flood control measures. In urban areas, the public is likely to demand structural measures to prevent floodwaters penetrating the community.

As traditional rural society becomes more modem and urbanized, there will be substantially more nonfarm-, nonfishing-, and nonriver-based labor and a corresponding expansion of costly infrastructure that is not inherently flood-resistant. Even in rural areas, the number of villagers not directly involved in agriculture, who do not benefit from flooding, is likely to increase rapidly. So there is a steadily growing constituency for flood control. This constituency - because it is predominantly urban and influential and represents a substantial proportion of heavy investment in the country - is often more effective at advocating flood control than its opponents are at advocating the living-with-floods strategy.

Many governments have a problem weighing the net benefits of floods against the relative cost and benefits of trying to prevent or control them. The negative effects of flood control measures - especially environmental or ecological changes they may cause - are often hard to predict so they should not be taken lightly or rushed into immediately after a particularly damaging flood. Flood control works require intensive study, and proposals must be accepted by all sectors of society before they are implemented.

Adapting to floods

Over the centuries, many societies have developed complex adjustments to floods. These adjustments and their role in riverine ecology should be the starting point for understanding alternatives to costly flood control works.

Adaptations to floods are most visible in the built environment, especially in housing and buildings. Common adaptations include: building houses on stilts so floodwaters can pass underneath, building houses on plinths or platforms so they are raised above flood levels, and building escape areas under roofs.

Siting is important in flood avoidance. Some societies have taboos against building certain forms of structures in known floodplains. In India, for example, the ancient Hindu building code, the Vastu Shastra, mandates that structures in coastal areas should be circular with conical roofs - a good design to resist floods and high winds. Traditional architecture usually adapts to flooding in the selection of timber, the design of the house’s base or foundation, and the building’s orientation to the flow of floodwaters. Human settlements are often collective expressions of these adaptations - for example, villages built on artificial platforms or areas raised above flood level, canal villages, artificial islands, floating villages or settlements, and settlements built on flood control embankments. Even cities adapt to floods; the most famous is Venice, Italy. More commonly, sections of cities adapt to floods by building floating settlements or stilt housing in flood-prone areas.

That regions can come to grips with floods without massive flood control works is evident in some of Thailand’s lower delta. Houses there are raised on stilts, roads generally run parallel to the rivers, and, where they cross the watershed, massive culverts permit the water to flow through without backing up. Floods occur every year, but are only rarely damaging. The region has adapted to the environment.


Riverine economies must adapt to floods in every facet of the economy from agriculture to industry, from the selection of crops to the selection of tools, transport, and conveyances.

Most riverine economies are agriculture-based. The crops selected, such as rice, usually depend on large quantities of water. Integrated cropping of grains and such long-stem water perennials as jut or reeds provides additional income and serves as insurance against flood losses.

Some crop varieties show a natural resistance to floods. The longer-grain natural varieties of rice fared much better in the 1988 floods in Bangladesh than the hybrid high-yielding varieties (HYVs). Farmers whose paddy stems were knocked over and waterlogged in the floods simply broke the stems at the internodes and watched them regenerate in time to produce a crop only slightly less abundant than normal. HYVs had to be plowed under and replanted.

Cultivators in deltas have become familiar with backup strategies to counter flood losses: selecting flood-resistant varieties, most often, and planting alternative postflood crops. In Viet Nam’s Mekong delta, farmers keep a reserve stock of potato seeds to plant quickly after a flood, using the water hyacinth tailings that floodwaters usually deposit on their fields as a mulch for the seedlings. Rural economies often adjust naturally to a flood. After the 1988 floods in Bangladesh, farmers whose standing paddies were affected by the floods could still collect the stunted stems and sell them as fodder for livestock. So much fodder was lost that prices for the salvaged stalks were almost equal to prices for harvested paddies (UNDP 1989b).

Fishing is always important in riverine economies. Casual fishing in local streams gives families an additional source of protein and an alternative or supplemental income. More important for the general economy, many people in riverine economies normally engage in professional fishing or fishing-related enterprises. Lakes or ponds used for fishing rely on floods to periodically replenish natural stocks of fingerlings.

The most important forms of transport in riverine areas are boats, barges, and rafts - craft that are inherently flood-resistant. In floods, other types of transport usually come to a halt, but river people usually have enough boats to remain mobile. In the Philippines, people often use the floods to float bamboo and forest products, such as logs, downstream to markets that otherwise could not be reached economically. In Bangladesh, people who live some distance from the rivers put goods they have prepared during the year on rafts and take advantage of the floods to transport them to the principal streams for collection and sale in large markets.

People also use floods to replenish drinking and irrigation water supplies. Researchers working in India in 1982 noted that villagers often view flood warning as a signal to intercept and impound floodwater for use on crops, holding back as much water as possible to raise a dry season crop - particularly in areas where short floods are common (Schware 1982a).


Most riverine societies have not only developed ways to live with and use floods, but have developed responses for avoiding the unusually large events. It is important to identify these measures as they are a clue to where to begin developing a national flood preparedness strategy. As a flood season nears, most riverine people instinctively take certain precautions: moving valuables to higher places in their shelters, moving animals to higher grazing areas or farther from the river, repairing boats, building rafts, and replacing the parts of their houses that have weathered and deteriorated.

Traditional warning systems

Many riverine societies have developed warning systems to give people enough notice to evacuate their homes and find shelter from floods. These systems may appear to be loosely organized, but in practice have been effective for disseminating basic warnings over wide areas. Often a variety of long-established message networks function at the same time, providing mutually supporting messages that condition people’s responses. At one level, people observe such natural cues as old flood markers inscribed on trees or ants moving their eggs to higher ground. At another level, they set up a network of voluntary river watchmen and embankment patrols.

Important research on traditional warning systems has been carried out in Bengal, India. In 1980-82, the National Center for Atmospheric Research (in Boulder, Colorado), in a series of studies on flood warning messages, identified socioeconomic constraints on the effectiveness of the official warning system and compared it to the traditional “folk” system. The researchers concluded that the official machinery for disseminating warning messages in the Damodar Valley was jeopardized by mechanical shortcomings and a communication network that was not ideal for communicating to many people widely dispersed in remote villages. They saw grounds for reappraising the official approach to flood warnings. They found that some problems could be met by applying appropriate technologies; others required better coordination of information, to prevent unnecessary time lags; still others called for more awareness of the needs of people who might use the information system (Schware 1982a).

Folk systems adapt to modem times. In India, for example, villagers near police stations that are linked to national police radio networks often make a point of sending someone to the police station several times a day to get the latest radio reports of upstream news on flooding. If there are reports of flooding, the person returns and warns the village. Warnings can be disseminated in many ways. In Indochina, fireworks are often used to alert people that flood notices have been posted. Flags are used in some Philippine and Indonesian communities, among others. In Latin America the ringing of church bells is a common flood warning.

The installation of embankments sometimes undermines these “folk” systems. People believe that the embankments prevent floods, so that there is no reason to maintain the traditional system. Where embankments provide only limited protection from flooding or where they are routinely breached or cut, the abandonment of folk warning systems has left communities with no suitable warning system. 1

Human responses to flood warnings

How people perceive and respond to flood warnings depends on available options in terms of a particular site, their economic needs, their health, and available information (Schware 1982). Sometimes people can only move to the nearest higher ground - for example, to rooftops, embankments, roads, railway lines, or even trees. Economic factors also come into play. Villagers tend to leave their homes only as a last resort if they believe that thieves in boats will steal their belongings. They cannot afford to leave behind their few tools and household possessions. Often they build rafts of bamboo or banana stems, tie them to a nearby tree, and stay on the rafts until flood-waters subside. Moving to roads or embankments is the next most common evacuation measure. Generally people avoid community shelters. Few can accommodate many people with adequate food and sanitation. Studies of flood shelters in India found that villagers overwhelmingly shun this option unless absolutely necessary.

Knowing about these typical responses is important for two reasons. First, it is doubtful that the longer lead times made possible by high-tech warning systems (radar, remote sensing, and the like) will significantly affect evacuation behavior in the near term. Second, the costly construction of dedicated flood shelters will not significantly reduce deaths and may be a waste of resources.

Traditional recovery strategies

Recovery begins with people’s efforts to salvage what remains of crops, tools, and personal belongings. If the family has livestock, a first task is to acquire fodder. Approaches to flood recovery depend on the type of crop, season, and so forth. For example, if it is the growing season, certain varieties of wheat can be cut where they have been bent by floodwaters and they will regenerate. Hybrid grains may not regenerate if waterlogged, but will often continue to grow and can be used for hay.

Flooding in riverine environments often increases crop production, especially when natural varieties are grown. After the massive 1988 floods in Bangladesh, the forecast was losses of 40 percent of the normal harvest - but the country actually produced 10 percent more rice than normal. Other staples also showed higher yields (USAID 1989).

Some strategies for economic recovery place a hardship on families - especially migration to find alternative sources of work. Traditionally, workers went to nearby communities to find seasonal work until they could accumulate enough resources to restart their own enterprises. But recent population increases may have saturated the postflood job market, forcing more people to go to large urban areas for wages.

Traditional mitigation measures

Living with the constant threat of floods forces people to take measures to mitigate losses should a damaging flood occur. Typical measures include:

· Adjusting planting and harvesting cycles. The most common method of flood mitigation is to adjust the crop cycle so that crops are already harvested by the time damaging floods are most likely. There are traditional guides for this practice and more recently farmers’ almanacs have been used to suggest planting dates (Brammer 1975).

· Seed banking. Farmers routinely hold back some seeds for replanting in case a flood early in the season destroys emerging crops.

· Famine crops. Many subsistence farmers set aside part of their land to grow flood-resistant crops that can be used for food if floods destroy the normal harvest (Campbell 1984).

· Mixing crops. Many farmers mix crop varieties (for example, HYVs with natural long-stem varieties) to reduce potential losses.

· Multiple fields. In some areas, farmers trade land to get fields at various elevations - with the objective of having at least one field that is less vulnerable to flooding. In Indonesia and the Philippines, village elders on some islands reapportion land annually to help farmers spread their risk.

· Alternative on-farm production. Many cultivators reduce their vulnerability by investing in other income-producing enterprises - most often, livestock, poultry (especially ducks), fishing, and the rental of draft animals. This is a much-favored mitigation strategy because the assets are movable.

· Informal “insurance,” cash pools, and the like. In Peru and elsewhere, farmers contribute cash to a common fund, or pool, as insurance against flood losses. If a cultivator loses his crop, he may borrow from the pool.

· Cooperatives (especially thrifts). Formal savings and loan coops, a recent innovation, provide cash to farmers in the event of a loss. Development agencies working with coops often encourage members to take collective mitigation measures.

Adapting traditional responses

The beginning point for any flood mitigation program should be to identify such traditional local responses to floods and incorporate them into the national strategy. The most important policy issue to be addressed in developing a national flood strategy is the tradeoff between structural and nonstructural measures. It may be technically feasible to control floods, but is the economic and ecological effect worth it? Do structural measures attract more people or economic activities onto floodplains where they should not locate? Does that produce long-term environmental problems? Will water tables or soils be affected?

Second, a government must decide whether it favors a centralized or decentralized planning and implementation model. Take Bangladesh, for example, where the area of potential flooding is staggering. As much as two-thirds of the land surface was covered by water in recent floods and as many as half of the people were said to be affected. With so much vulnerability, it is impossible for the government to meet all needs, so flood preparedness and mitigation measures must be selective and must rely on self-help and local initiative. What makes sense in Bangladesh is a community-based approach heavily oriented toward the adaptation of traditional responses.

To adapt traditional responses to official strategy means reorienting the planning process, to plan flood mitigation measures from the perspective of the communities most likely to be affected, and to involve villagers in local plans. Planning should be initiated at the village level and priority should be given to activities that stimulate self-reliance, promote cooperation and community involvement, and contribute to community development. Community development groups can play a major role in planning. Efforts should be made to involve local community development agencies and NGOs in the planning process, to improve coordination between these agencies and the government.


Flood mitigation proposals bring conflicting agendas into focus. Some cultivators may want their fields to be flooded and will take measures to ensure that they are. Not everyone appreciates flood control embankments. Breaches are often intentionally cut in levees - sometimes by people living between the levee and the river, in a desperate effort to lower the water on their side of the embankment; sometimes by cultivators who want to bring water, soil, or nutrients to their fields. People who benefit from floods - such as fishermen, brickmakers, and sand vendors - are also likely to oppose flood control.

Opposition also arises when land is expropriated for embankments. The average holding in many riverine areas is less than one hectare. An embankment could cover much of a plot and, in the process, thousands of small cultivators could be displaced or their land holdings reduced so much they are no longer profitable. In a country where demand for land is high, the political costs of embankments must be weighed carefully. In densely populated riverine areas, the best policy may be to limit embankments, permit controlled flooding, and emphasize community-based emergency preparedness and mitigation measures.

In any flood mitigation program, there are likely to be strong advocates of flood control and equally strong voices for “living with floods.” The voices for flood control are likely to be stronger. The constituency for flood control is primarily those who will be harmed the most by flooding or those who will benefit from flood control works. That includes threatened community groups, such as farmers living in low-lying, poorly-drained areas, nonfarm workers in villages, urban dwellers, and people whose livelihoods or income are affected. It also includes such members of the technical community as engineers, construction firms, builders and operators of infrastructure, and government ministries (such as agriculture, power, roads, and railways) whose projects or facilities are threatened. Other members of the constituency for flood control works include large landowners who will benefit from increased land area or irrigation and donors with investments in infrastructure, critical facilities, and projects in flood-prone areas.

Advocates for the “living with floods” approach are less powerful. They include groups with little representation in government, such as small subsistence cultivators, fishermen, and small craftsmen. Their support is normally from environmentalists, ecologists, and development workers - groups that traditionally lack political clout. It is this constituency that must ultimately be mobilized to support any nonstructural mitigation or preparedness efforts.

Flood mitigation efforts compete with other development projects for scarce resources. In any country - no matter what level of development - it is difficult to get disaster mitigation on the national agenda. It can only be justified if mitigation benefits long-term economic and community development. Thus disaster mitigation policy should be part of and should support general economic and social development.


One of the most important decisions to be made is where to place national responsibility for flood strategy. Designation of the responsible agency often determines how the program will be carried out. If an irrigation ministry is given the task, for example, there is likely to be more emphasis on structural flood control measures, such as embankments or engineering works. If the task is given to a high-level committee, the approach chosen may well reside with the committee’s technical staff. If responsibility is assigned to a relief authority, the focus will usually be only on immediate response and short-term disaster relief measures.

The lead agency should thus be carefully considered. Agencies with experience on this issue (such as UNDRO and USAID) in most cases recommend that a single agency be given responsibility for all pre- and postdisaster functions. If this is the course chosen, responsibility should be vested in a line ministry strong in operational capabilities (communications, equipment, and transport).

An alternative is to split responsibilities between ministries. The usual manner is to delegate responsibility for prevention and mitigation to a ministry with full-time responsibility for planning, construction, and maintenance of flood prevention and mitigation projects; then to delegate the relief and response function to a specialized relief or social service agency. The two then develop preparedness plans jointly and policy and coordination are handled by a high-level, interministerial committee chaired by a top government official.

This approach has some strengths. Comprehensive emergency management requires calling on specialties found in many different types of agencies. It is rare, and probably not even desirable, for all functions to be in one agency; the costs would be enormous. Recognizing the contributions each agency can make and developing the ability to coordinate and draw resources from them in an emergency is generally the most effective way to approach the matter.

Whatever agency is chosen, it is important that flood preparedness and mitigation efforts not be carried out in a vacuum. In most countries multiple hazards exist for which there are other disaster preparedness efforts. Many riverine areas are also subject to cyclones, for example. Efforts should therefore build upon any work already done for cyclone preparedness and mitigation; favor activities that support both cyclone and flood preparedness; and favor measures that build awareness of, and stimulate more effective response to, all major hazards.


In developing a national program, planners should keep in mind the types of local responses that can be improved with outside assistance. These include:

· Issuing warnings at the village level.

· Organizing working parties to repair embankments quickly and preempt breaches. This requires stockpiling materials locally in predetermined locations and developing ways to summon and mobilize the required number of workers on short notice.

· Organizing working parties to sandbag critical facilities. This also requires stockpiling supplies and mobilizing workers.

· Stockpiling and prepositioning tools and equipment to facilitate the relief and support of those who have been evacuated.

· Making transport arrangements so evacuation and relief can be speedy.

· Supporting agricultural recovery strategies, stockpiling information about practices to be followed, and prepositioning essential inputs that cannot be supplied by normal or postflood sources.

· Developing plans to provide emergency supplies of food and clean drinking water.


It is important that donors coordinate their support. A donors’ aid group can be formed for this purpose. The group should meet formally at least every two years to propose new projects and review progress. When any project is developed, the aid group as a whole can act as a project review committee, or at least a group of its members can serve on such a committee. An aid group serves two purposes. It shows that the donors treat the matter seriously and jointly support activities, and it serves as a bridge between old and new focuses when they change. Collectively, the group has more clout when it comes to taking corrective measures. The aid group could be made up of major donors, such as bilateral organizations, the UNDP, or the World Bank. But other interested groups - such as World Food Programme, UNICEF, World Health Organization, and the League of Red Cross and Red Crescent Societies - could also participate. To coordinate the donors’ inputs, a full-time disaster preparedness project officer should be appointed and supported with appropriate technical advice, as needed.

Several principles should guide project funding and planning. First, wherever possible, the donors should jointly finance projects. This signals - and ensures - that donor interest remains high. Second, priority should be given to projects that involve two or more ministries or departments. Ultimately, disaster response requires interministerial cooperation and coordination so it should begin in the preparedness phase. Finally, funding should involve local counterpart funds, where possible.

Program planning

Measures to prepare communities to minimize flood dangers involve flood mitigation and preparedness activities. Mitigation efforts aim to reduce a flood’s impact and prevent it from becoming a disaster. Some mitigation measures are structural, such as building platforms for villages or local embankments to control or diminish flooding. Most structural measures are individual actions such as reinforcing houses to make them more disaster-resistant or building houses on raised plinths above expected flood levels. Nonstructural measures, such as changing agricultural cropping patterns to reduce losses, involve changes in institutions, regulations, behavior, knowledge, and attitudes. For the most part, small-scale, self-help, and community-based measures are the main thrust of mitigation strategies.


The assumption underlying disaster preparedness is that a flood will strike a community and that the community - the people and institutions - should be prepared to deal with it. The focus is on saving lives, reducing property loss, and structuring the emergency response to be timely and to lay the groundwork for rapid recovery. Disaster preparedness for floods involve three kinds of activity: warning, evacuation, and flood-fighting.


Floods are one of the few forecastable natural hazards for which adequate warning can be given. The objective is to issue warnings as early as possible so people can protect their homes, belongings, and livelihoods, and then, if necessary, evacuate to a safe area. The essential elements of warning systems are vulnerability mapping, communications systems that reach the appropriate authorities in threatened communities, and methods of message dissemination that stimulate people to take effective safety measures.

Warning systems are crucial to “living with flood” strategies because people must have time to take action. National warnings systems often fail because of:

· Problems disseminating warnings locally, below the district level.
· Unclear “messages.”
· No viable response options by the time the message reaches the target communities.

In recent years, exceptional strides have been made in improving warning and evacuation systems, based on behavioral studies and analyses of interpersonal communications. In rural areas, a national warning system is only as effective as the extent to which it builds on the traditional warning system and extends communications to the villages. Warnings based on public radio broadcasts have not been effective in rural areas. National systems are usually more effective at warning urban communities, government institutions, public utilities, large industrial and commercial facilities, and emergency authorities. The centerpiece of rural systems should be local people who observe, interpret, and issue alerts - not the meteorological office. Emphasis should be given to:

· Improving locally generated warnings.

· Getting credible, respected local leaders to issue clear instructions about where to go.

· Arranging for the protection of livestock and movable assets.

· Providing or expanding nearby evacuation sites that meet rural people’s need for shelter, food, and the storage of livestock and personal effects.

· Using village-level evacuation transport appropriate to the environment.


Evacuation measures may include providing the transport needed to leave a threatened area quickly; rescuing stranded people; providing temporary shelter, food, water, and basic comforts; and then, when the emergency is over, helping people return to their homes. A successful evacuation, especially in riverine areas, requires detailed planning, identifying escape routes and shelter sites, training shelter managers, organizing rescue and transport teams, and prepositioning food and temporary relief items.


Flood-fighting involves measures taken during a flood to prevent or reduce damage to communities, public infrastructure, or critical facilities that might be caused by the overtopping or breaching of embankments, water rising too high, or water threatening areas because of changes in the course of streams, downstream blockage; and so forth. Flood-fighting usually involves building temporary embankments or small-scale plodders - principally by sandbagging - and making speedy repairs to flood control works. Flood-fighting requires stockpiling and prepositioning tools, materials, and equipment, and planning how to mobilize and support large work forces on short notice.

The most important - and most overlooked - function of preparedness is to structure the overall emergency response. This is usually carried out by means of an emergency action plan, standing orders for key ministries, or standard operating procedures (SOPs). Ideally, a preparedness plan is a series of subplans, including a warning plan, an evacuation plan, a search-and-rescue plan, an assessment plan, and an emergency response plan.

The whole premise of preparedness planning is that an emergency is not the time to be deciding what to do. During an emergency, information is incomplete, conflicting, and rarely accurate. The best time to make decisions is when they can be addressed rationally, without the pressures of time and urgency that force premature or incorrect decisions. The objective of disaster preparedness is to move decisionmaking forward - out of the emergency - and to predetermine as many of the decisions and responses as possible. Once the decisions are made, detailed plans can follow.


Many resources are essential to rapid recovery from floods. It is important to identify them and to increase people’s access to them in normal development projects. If planners can identify the usual postflood agricultural strategies that farmers use to mitigate their losses, for example, the government can be encouraged to provide the tools or inputs that will be needed locally to build up onsite reserves that could be tapped in an emergency to accelerate recovery and reduce costs. If alternative varieties of crops are grown, the seeds, fertilizers, and pesticides that increase yields can be provided in outlets near floodplains - and reserves can be kept on hand in case they are needed. Similarly, if certain implements are needed to salvage waterlogged crops, it may be possible to provide them through cooperatives, and to promote the practice of holding some in reserve at the coops for postflood use.

Distribution of relief supplies is more effective when people rely as little as possible on external mechanized transport and when local transport is emphasized - especially (in the riverine environment) local, or country, boats.

Concepts about what is needed and important in emergencies should be reevaluated, especially for shelter. It is almost as costly to deliver tents and plastic sheeting, for example, as to deliver galvanized iron roof sheets - a commodity of far more value to local people as it can be used first as a shelter and later in the replacement house.

Outline for developing a national program

A national program incorporating the recommendations suggested in this chapter should begin on two fronts simultaneously: at the community level, working from the bottom up, and at the national level, working from the top down.


Plans for community actions should originate at the grassroots level. District and national plans can be developed later to support the array of community plans.

(1) Identify traditional mitigation and preparedness measures. Begin by identifying the full range of local practices, evaluating their effectiveness, and determining ways they can be incorporated or expanded in national programs. Remember that people’s responses to floods are shaped by their perception of risks balanced by their perception of benefits. To motivate people to take action, proposals must address these perceptions. A starting point is thus to:

· Study perceptions of risk among different income and occupational groups in different areas, both rural and urban.

· Determine who is at risk. Not all communities are equally vulnerable to a flood. It is not possible or desirable to evacuate everyone or to extend protection to those who do not really need it, so vulnerability and risk mapping should be carried out. In the long term, detailed maps can be developed, using geographic information systems (GISs). In the short term, gross approximations can be made by examining recent flood experiences.

· Study the range of local responses, evaluating their effectiveness and determining ways they could be improved through government interventions. For example, yields of some alternative crops could be increased by timely provision of fertilizers, which could be stockpiled in strategic locations.

(2) Initiate village-level activities. Many activities and practices can be introduced and encouraged at the village level, including:

· Village-based warning and evacuation systems.

· Small-scale protective structures (such as village plinths and evacuation platforms).

· Protective measures for housing (such as water-resistant mud construction and treatment of structural timbers).

· Specific flood season agricultural practices.

· Planting bamboos or fast-growing trees that can be used for disaster-related purposes (such as rafts and components for temporary shelters).

(3) Promote economic development strategies that reduce vulnerability. This is one of the least-explored options. Measures that can be combined with other activities have a better chance of acceptance and implementation. For example, agricultural adjustments or innovations have a better chance of succeeding if they produce greater yields or profits than if they simply reduce a family’s exposure to risk - a risk that may not materialize in a given period. An inventory should be made of development programs and strategies that could have implications for disaster mitigation or preparedness.


While actions are being initiated at the community level, the following national and district-level program should also be under way:

· Evaluation of national and district-level emergency responses to past floods, and identification of lessons and areas where improvements are needed.

· Vulnerability and risk mapping.

· Development of warning systems.

· Development of evacuation and sheltering systems.

· Improvement of communications systems.

· Development of model plans that can be adapted to community needs and can serve as a basis for initiating mitigation and preparedness activities at the village level.


1. People intentionally cut embankments for many reasons. A UNDP study of the 1988 Bangladesh floods found that squatters who live between an embankment and the river cut the embankment when floods come hoping to lower the flood level. Some farmers behind an embankment cut it to flood their fields, water their crops, or bring nutrients to their fields. Some people want to flood ponds to recharge them and replenish fish stock.

Case study: creating job and income opportunities for refugees in Pakistan

Alcira Kreimer and Martha Preece

This simple, well-targeted project benefited about 30 percent of the Afghan refugee families by increasing job and income opportunities and providing specialized training in forestry and environmental protection to thousand of immigrants. The project succeeded in developing sound economic and environmental prevention measures to address the problem of degradation of natural resources. It demonstrates the importance of developing simple, down-to-earth solutions to environmental problems, and shows that sound development activities can alleviate the physical and economic damage caused by the massive influx of refugees. By involving the migrants in environmental protection and prevention efforts, the project has helped the region break out of the cycle of poverty, environmental degradation, and disaster vulnerability.

In the last ten years the number of people fleeing from wars, persecution, and natural disasters has escalated. Fifteen million people have been uprooted, and the number of people forced out of their own countries grew more than 13 percent between 1988 and 1989 (McCallin 1990). About 90 percent of those migrants were rural people moving into rural areas in countries already hard-pressed to meet their own people’s needs. These massive population movements have placed an extraordinary burden on developing countries’ physical and economic assets, ultimately damaging supplies of such natural resources as fuelwood, pastureland, and water. Without sound environmental policies, rapid migration and population growth may significantly deplete forests and ecologically sensitive rangelands, threatening the sustainability of development.

The uncontrolled alteration of environmental systems may increase vulnerability to extreme events. Deforestation and poor land management have already accelerated soil erosion and water runoff in many areas, increasing the threat of landslides, floods, and drought. Environmental mismanagement coupled with massive migration has made large areas more disaster-prone and has spread the impact of natural hazards to both man-made and natural environments.

The problem in Pakistan

The Islamic Republic of Pakistan covers about 197 million acres in four provinces: Baluchistan, the Northwest Frontier Province (NWFP), Punjab, and Sind. In the past 20 years, the country’s population (about 106 million) has increased about 3.1 percent a year. More than 72 percent of Pakistan’s inhabitants live in rural areas, and an estimated 65 million or more people live below the poverty line (US$188 per capita in 1986 prices).

Since 1979 as many as 3 million Afghans have arrived in Pakistan. The government has settled most of these people in 325 villages - more than 60 percent in NWFP and Baluchistan. It is said that one of every seven inhabitants of NWFP, and one of every four inhabitants of Baluchistan, is an Afghan immigrant (World Bank 1983). The extraordinary burden of these unplanned population movements hampers these areas’ ability to provide basic services and administer assistance.

The massive influx of immigrants has increased the demand for food, fuelwood, and shelter, accelerating environmental degradation and compounding the problems that have resulted from centuries of overexploiting pasturelands and uncontrolled deforestation. Degradation of the natural environment has exacerbated problems of soil erosion and water runoff, increasing vulnerability to floods. In the last 20 years, Afghans and their livestock have put added pressure on rangeland vegetation and their demand for fuelwood has accelerated the rate of deforestation (World Bank 1985).

The Baluchistan region. In the last few years, extensive environmental degradation in watershed regions has threatened fragile ecological balances, drying springs and reducing levels of groundwater. This has especially been true in the low-rainfall areas of Baluchistan, where soil erosion, deforestation, and depletion of the rangeland threaten irreversible degradation. In northern Baluchistan, near the Afghan border, accelerated population growth severely strains the ecological balance of the few relatively fertile valleys that get runoff water from the eroded mountain ranges. Here Afghan immigrants live in 16 major camps and smaller temporary settlements. Their presence and their livestock herds have contributed to deforestation and put added pressure on land and water resources. Compounding the problem, extensive sand dunes have developed because of wind erosion of the largely denuded mountains, imposing severe hardship on the local population and the region’s economy. Infrastructure and villages are in continual danger of being buried under sand. Since the mid-1950s, about 35 villages, 50 irrigation channels, many fields, and other residential and agricultural installations have been ruined and subsequently abandoned.

Rapid deforestation has also taken its toll on the eroded hills and the depleted juniper forest. Unregulated felling of trees has virtually eliminated the natural forest and severely eroded the mountain slopes, leading to sedimentation of reservoirs and irrigation systems. The destruction of vegetative cover and erosion of the hillsides have increased the number and power of runoff torrents that cannot be controlled by traditional irrigation structures. The needs of Afghan immigrants exacerbate these problems and threaten further irreversible degradation.

The Northwest Frontier Province. This region’s environmental problems are aggravated by the massive influx of immigrants that have settled in the plains, where available land can no longer support the rapidly growing population. The mismanagement of forests and intensive cultivation of hillsides with traditional techniques have increased the area’s susceptibility to landslides and flashfloods. The depletion of forests and the degradation of rangeland is a particular problem in the Hazara region, where a large proportion of the total population is Afghan immigrants. The enormous need for fuelwood has caused widespread deforestation and thus soil erosion, desertification, and the siltation of watercourses and reservoirs. Trees are felled at an alarming rate with virtually no replanting, so hillsides are almost devoid of vegetative cover. As a result the ability of the soil to permit infiltration and retain moisture has been reduced, accentuating problems of erosion and exacerbating the danger of flooding and landslides. Hillside stabilization is a major concern in the Kagan Valley, where the Forestry Department has been trying to prevent avalanches.

The Punjab region. Here soil erosion is the heaviest. Deforestation, particularly at higher elevations, is so dramatic that the damage is often irreversible. Landslides commonly destroy houses and infrastructure. In the rainy season, destructive flash floods and hillslides severely threaten the survival of some towns. The effects of population pressure and the mismanagement of natural resources are compounded by traditional rights to land and resources - particularly in tribal areas, where tradition conflicts with sound resource management practices. (Each adult male is allowed to fell three primary-category trees a year - and one more for each funeral. These hereditary privileges have multiplied with the growth of the local population.) Efforts by the Forestry Department to preserve the ecosystem and protect investments have been largely unsuccessful. They have failed to create incentives to influence local communities’ willingness to implement soil conservation treatments on their land.

The Income Generating Project for refugee areas

Since 1983, the Bank has been working closely with the UN High Commission for Refugees to identify, prepare, appraise, and supervise the Income Generating Project for Refugee Areas in Pakistan. Financed by donor grants, the project was administered by the Bank.

The Income Generating Project addressed some of the problems created by the protracted residence in Pakistan of some 3 million Afghan refugees. Its objectives were to create job and income opportunities for refugees and local residents, to improve the rural environment and repair the environmental damage caused partly by the influx of immigrants and their livestock, and to create viable economic resources in these areas. In the long term the project aims to restore the disturbed ecological balance and ease the pressure of population and cattle on natural resources.

The project consisted of many small-scale, labor-intensive subprojects in three sectors: forestry and watershed management; irrigation and flood control; and road construction, upgrading, and rehabilitation. In Phase I, 52 subprojects were undertaken in NWFP and Baluchistan between 1984 and 1987. The second phase (162 small subprojects) began in 1987 in NWFP and Baluchistan and one district in Punjab.

Afforestation and watershed management subprojects. Forests cover only 3.7 percent of the land in Pakistan, and only 1.2 percent of total forests are commercially productive. The forestry subprojects aimed to rehabilitate forests and watersheds that had deteriorated as a result of the added demand by immigrants for fuelwood supplies and livestock grazing areas. All subprojects focused on preventing soil loss by surface or gully erosion, and reducing rainfall runoff in seriously denuded watersheds to prevent and mitigate flood damage. Efforts were also made to protect the remaining forests from illicit use and afforestation, especially in the NWFP. There the focus was on restoring ground and tree cover; reducing soil erosion from overgrazing, cutting, and burning; and preventing surface runoff - to produce stabler stream flows. This subproject emphasized community participation in rural tree planting programs and natural forest management. The watershed management subprojects (mainly in Baluchistan) aimed to promote soil conservation practices, improve fuelwood production, and increase groundwater infiltration.

Irrigation and flood protection subprojects. These 106 subprojects (24 in Phase I, and 82 in Phase II) were to build spurs, bunds, and walls to protect disaster-prone areas from floods in NWFP, Baluchistan, and Punjab. In addition to minimizing flood hazards in villages and refugee camps, the projects are expected to improve agricultural output by protecting arable lands from excesive surface runoff and erosion.

Technical assistance was provided to help the pilot extension program in forestry and watershed management, to monitor the work of Afghans and locals, and to advise technical staff. Immigrants were trained in afforestation and environmental management and protection techniques, including tree planting, flood protection, drainage, and soil conservation. Pilot schemes were developed to reduce fuelwood consumption at the household level.

Sustainable environmental management

This project has been an excellent vehicle for involving the refugee community itself in alleviating some of the damage refugees have done to Pakistan’s environment and infrastructure. It illustrates how to integrate strategic work on conservation of natural resources, disaster prevention and mitigation, and the generation of jobs and income.

Managing drought and locust invasions in Africa

Thomas R. Odhiambo

The two most critical African natural disasters - drought and locust outbreaks - are least understood in terms of the four activities considered essential to long-term disaster reduction: hazard prediction, risk assessment, disaster preparedness, and disaster management. In finding a solution to the drought and desertification that plague Africa’s Sudano-Sahelian belt, planners would do well to study indigenous adaptations to the unpredictable Sahelian ecosystem. They should study them, translate them into scientific terms, validate them, and try to implement them to stabilize the Sahelian and other drought-prone ecosystems. The key to solving the problem of locust swarms may lie in the work being done by the International Centre of Insect Physiology and Ecology. ICIPE hopes by studying the behavior of locusts to find a way to dampen the locusts’ swarming cycle and intensify their solitary cycle so that locust populations can maintain a relatively sedentary, grasshopper-like lifestyle. ICIPE hopes that solitary locusts can continue to exist as natural herbivores in Africa’s semiarid breeding grounds, continuing their participation in the dynamics of the savannah ecosystem without periodically breaking into locust plagues.

A.J.W. Taylor (1978) described disasters as “catastrophic events that (a) interfere with everyday life, disrupt communities, and often cause extensive loss of life and property, (b) overtax local resources, and (c) create problems that continue for longer than those that arise from the normal vicissitudes of life.” In this sense, disasters need not be sudden, unexpected, or devastating. Otherwise, disasters such as drought, famine, and epidemics would be excluded. All disasters - industrial or natural - vary in the degree of vulnerability to which victims are subjected and the special attention they require. But disasters caused by the forces of nature are perhaps “the most readily accepted, even if with despair and resignation. They represent the unleashed fury of the natural world against which mankind is quite helpless” (Taylor 1978). This feeling of helplessness has given new impetus to a search for long-range solutions to the management of two pervasive natural disasters in Africa: drought and locusts.

“The true dimensions of poverty are existential rather than economic,” wrote Albert Tevoedjre (1978). To Walter Weisskopf (1989), misery reflected impotence:

The medieval world outlook believes in a supernatural, but comprehensible order. The Newtonian model believed in an order routed in nature and comprehensible through reason. Having abandoned a belief in providence, grace and other worldly rewards for religion’s virtue as well as the deterministic belief in nature and reason, the Heisenbergian paradigm seems to deprive us of all protection against the threat of the unknowable future and of the unknowable reality. (Quoted in Giarini and Stahel 1989.)

The point about such impotence is that people’s traditional, communal ability to cope with recurring setbacks can no longer provide a safety net for victims. They have become more vulnerable, and natural disasters have become an ultimate crisis of destitution. Consequently, more than any other community, Africa’s scientific community today has the responsibility of creating a new perception of natural disasters.

Developing long-range solutions to the two most difficult natural disasters in Africa - drought and locust outbreaks - requires an information-intensive, management-aware strategy. In 1988, the UN Secretary-General appointed an international ad hoc group of experts to recommend a scientific framework for launching the International Decade for Natural Disaster Reduction (IDNDR) for the 1990s. That group considered four kinds of activity essential to a solution-oriented long-range program:

· Hazard prediction through monitoring, early warning systems, and the like.

· Risk assessment - for example, by mapping hazard levels.

· Disaster preparedness, especially by training key personnel, educating the general public, and appropriately controlling land use and construction.

· Disaster management, by developing schemes for local evacuation, establishing lines of emergency supply, and mobilizing civil defense groups.

Drought and locust outbreaks are poorly understood in terms of these four activities. As the group of experts stated in its report:

Drought is a complex environmental phenomenon, including long-term climatological changes and wide-scale ocean/atmosphere interactions as well as ecological deterioration of human origin. The management of drought as a natural disaster has many factors in common with management of disasters of more sudden onset. Drought predisposes the environment to several rapid-onset natural hazards, including locust infestations and, in many instances, flash floods. Drought alone causes large agro-ecological damage and seriously disrupts socioeconomic life. Over this century, droughts have tended to intensify as a result of accelerating deforestation and large-scale soil erosion, especially in Africa; their management globally has become a matter of urgent action.

These catastrophic phenomena now demand a thoroughgoing scientific explanation, leading to solution-oriented technologies that may well incorporate elements of social invention. What Africa needs is to master science and technology, not merely identify with it. This requires that problem-solving become the principal aim of our scientific endeavor focused on drought and locusts. We must become dissatisfied with the existing state of knowledge about these natural processes and events, mistrusting the obvious and searching for the underlying order in natural processes (Sindermann 1985).

Drought in Africa

The Food and Agriculture Organization (FAO) of the United Nations predicts that unless Africa takes corrective action, rainfed croplands on the continent will become 30 percent less productive by the end of the 1990s, mostly because of soil depletion and erosion (World Resources Institute 1989). Part of Africa’s unrealizable potential for agricultural productivity is the 40 percent of the continent that is either extremely arid or subject to cyclical drought.

The Sahara, the largest desert in the world, receives less than 100 millimeters of precipitation a year (Conservation for Development Centre of IUCN 1986). Other arid areas, such as the Kalahari Desert in the south, also suffer chronic low precipitation, usually with high temperatures. Drought, on the other hand, is a temporary feature, experienced only when rainfall deviates appreciably from normal levels (of about 200 millimeters a year). Drought can occur in virtually any rainfall or temperature regime (Lockwood 1988). The Sahel - the zone across the continent south of the Sahara Desert, with annual rainfall of 200 millimeters - has experienced a north-south migration in its 600,000 years of human occupation. The Sahel has now stabilized between 13 degrees north and 19 degrees north, but the research of A.M. Lezine and her colleagues shows that the zone has shifted substantially in the past 20,000 years, a period that has been studied in some detail. Rock paintings show a succession of scenes from game-rich savannah to cattle-herding pastoralists to desert panoramas. Geomorphological and palaeontological evidence in the guise of fossil river systems and drainage networks (for instance, the Wadi Howar system in the Sudan) contains fossil remains of crocodiles, hippos, and river bivalves - in a region that now receives merely 25 millimeters of precipitation a year. Pollen analyses indicate that fossil lakes in the Sahel (for example, the Selima Oasis) once sustained savannah grasslands in areas now hyperarid. These studies conjure an image, some 18,000 years ago, of savannah grassland with scattered acacia trees that extended only as far north as 10 degrees north, which is further south than its current geographical limit. In 8500 before the present (BP), moisture-demanding types of vegetation extended northward rapidly, eventually reaching 400-500 kilometers north of their current agro-ecological limit. Reversal of this northward shift began in about 6100 BP and intensified in about 4500 BP; Sahelian vegetation as we know it today became established in its present zonal limits (13 degrees north and 19 degrees north) about 2,000 years ago.

Despite this volume of evidence, and the more recent episodes of drought in the Sahel during this century - including the great drought of 1984 that caused about 100,000 deaths - we cannot yet predict drought (Conservation for Development Centre of IUCN 1986). We believe that arid lands are rapidly spreading in Africa, often aided by the long cyclical drought periods, based on observations on the fringes of the Sahel and similar semiarid and arid areas such as the Kalahari and Namib Deserts, Somalia, and North Eastern Kenya (UNEP 1985). This aridification, and the rapid deforestation of much of Africa, is leading to three critical phenomena:

· Desertification of the Sahel, through bushfires, overgrazing, increased harvesting of fuelwood, and the spread of cultivation to marginal lands.

· Sahelization of the savannah grasslands.

· Savannahization of the forests.

Desertification has now reached 88 percent in Sudano-Sahelian Africa, 80 percent in southern Africa, and 83 percent in Mediterranean Africa. We cannot yet accurately quantify in time and space the changes associated with desertification, sahelization, and savannahization. Nor are we certain about the factors that determine the onset of drought, although some associations are becoming apparent.

Past studies suggest that drought in Africa south of the Sahara is correlated with (1) drought in the Central America-Caribbean Atlantic region and (2) warm sea-surface temperatures and rainfall on the Pacific coast of Latin America (associated with the so-called El Niffect) - in other words, associating African drought with interactions between the ocean and the atmosphere (Rind and others 1989). Similarly, rainfall in the western Sahel is affected by anomalies in sea-surface temperatures in both the Pacific and Indian oceans. Not that these examples reflect a direct causal relationship. Rather, the droughts over much of the Sahel and North Africa are associated with the reduced (but not delayed) northward extension of the Inter-Tropical Convergence Zone (ITCZ), the band of wet weather that circles the world at the point where the trade winds blowing from the Northern and Southern Hemispheres meet (Rind and others 1989).

For all our ignorance of what causes cyclical drought in Africa, its impact on the human population is profound. The two-decade drought from 1966-67 to 1987 was the worst for 150 years but others almost as bad have occurred every 20 to 25 years or so. Under these circumstances, it is meaningless to speak of protecting the environment. Rather, a way should be found to fit human activities into the reality of the natural systems.

Of the world’s 40 million nomadic pastoralists - whose livestock-based economies survive on deserts or semiarid savannahs - 25 million live in Africa, most now relegated to marginal lands (Bass 1990). Anthropological studies reveal that traditional pastoralists’ knowledge of the environment in which they live and work is highly complex and organized - their plant and animal breeding and husbandry the result of experiential fact-finding that cannot be ignored by modem scientists. According to a Dutch study, quoted by the Independent Commission on International Humanitarian Issues (1985), “traditional herders produce as much protein per hectare as do ranches in areas of equal rainfall in the United States and Australia; the Sahelian herders, however, accomplish this with infinitely less mechanical energy inputs, relying for the most part on manpower.” This husbandry, based on traditional knowledge, requires that the Sahelian herder not stay in one place - lest overgrazing and desertification set in. Settling herders in semiarid areas is therefore often unwise:

Sedentarizing nomads ignores the fact that they employ their marginal resources better than anyone else could. Nomads have lower birthrates than their settled neighbors. Their family herds support more people on the land than do commercial ranches. Nomads make few demands on the state, and they practice the kind of self-sufficiency that any enlightened government should want to encourage (Bass 1990).

Indigenous Sahelian social systems were intimately linked to their agro-ecosystems, until externally imposed changes in recent times led to environmentally destructive change, such as the charcoal production associated with trans-Saharan trade and the expansion of transportation networks (Gritzner 1988). Indigenous adaptations to the unpredictable Sahelian ecosystem - including integrated agro-sylvo-pastoral practices - are not only highly sophisticated but have inherent scientific validity (Gritzner 1988). We would do well to understand them, translate them into scientific terms, validate them, and then try to implement them to stabilize the Sahelian (and other drought-prone) ecosystems.

Desert locust swarms

The Sahel is full of other surprises. One recurrent problem is occasional massive outbreaks of locust swarms. In October 1988, desert locusts (Schistocerca gregaria) bred on an unprecedented scale along much of the Sahel, so that in early October several swarms of them reached the Atlantic coast of West Africa - from Guinea-Bissau in the south to Mauritania in the north - then invaded the Cape Verde Islands on October 5-6 and 12. Soon thereafter, the locusts were sighted over the Atlantic Ocean and began reaching the eastern West Indies from October 14, stretching from St. Croix in the north to the coastal areas of Guyana and Suriname in the south. They made this intrepid landfall 4,500 kilometers from the West African mainland with the assistance of the tropical storm Joan (Rainey 1989).

There are at least five important locust species (and their ecogeographical subspecies) in Africa: the desert locust, the tropical migratory locust (Locusta migratoria migratorioides) found everywhere in Africa and beyond, the red locust (Nomadacris septemfasciata) found in Eastern-Central Africa, the brown locust (Locustana pardalina) found in Southern-Central Africa, and the Senegalese grasshopper (Oedaleus senegalensis) in West Africa and the Atlantic Ocean islands. Undoubtedly, the desert locust is the most intractable, widespread, and perennially destructive of all these pestiferous locusts. There are at least 200 other grasshopper species resident in the Sahel that live normally as herbivores in this savannah ecosystem without swarming as marauding migrants.

In normal years - during the frequent drier periods when the desert locust goes into recession - the species is widely distributed over a wide belt of arid and semiarid lands (including the African Sahel), covering some 16 million square kilometers. Under this recession, the desert locusts live as inconspicuous, solitary grassland herbivores in these semiarid areas. Major locust swarms develop as a result of a rapid increase in locust numbers among the erstwhile recession population when widespread, heavy, prolonged rains occur after a long stretch of drought, in several scattered breeding areas. These wet periods permit two or three generations of the locust to develop in each breeding area, multiplying rapidly - sometimes by a factor of 105. These aggregate, match as hoppers, then swarm as adults, becoming highly mobile and traveling up to 1,000 kilometers a week, assisted by winds. Such gregarious locust swarms arise only after favorable rains, occurring over several seasons, succeed a long stretch of drought years (COPR 1982). The locusts may well invade a much larger area, covering some 29 million square kilometers, taking in all of the Sahel, North Africa, the Mediterranean region, the Middle East, and West Asia.

The old hypothesis that locusts migrate from areas that have become untenable is not correct. The flying locust swarms tend to be carried toward areas where rain has fallen, following the Inter-Tropical Convergence Zone (COPR 1982). The late Reginald Rainey (1989), who made a life-long study of locust migration, concluded, after 40 years studying locust swarming phenomena, that migration is intrinsically adaptive - that locusts use migration as a mechanism to exploit seasonal changes in the spatial distribution of environmental resources. The desert locust simply exploits the necessary environmental resources by becoming airborne, and latching onto the geographically patterned global and local wind systems, which eventually take them to their convergence zones and their rains. The locusts then exploit the ephemeral vegetation, which develops quickly after unusually good rains over an area normally occupied by arid thornbush. This larger invasion area is often even more fragile than the locust’s usual recession breeding area, so the locust invasion wreaks devastation.

The change from inconspicuous, solitary, grasshopper-like individuals (in the drought years) to highly mobile, gregarious, marauding locust swarms (early in the wet years) is associated with a major change in locust behavior, and fairly accurate forecasting systems exist for the appearance of matching hopper bands and locust swarms. The onset of recession - the petering out of locust plagues - is not so simple to predict. We have few clues as to what causes this behavioral change. In 1965-66, desert locust infestations worldwide were at their lowest level in 27 years, and an equally significant 18 years of locust silence followed. Rainey thought that the decline of desert locust infestations in the mid-1960s was associated with “a marked change in the global wind circulation, which had reverted sharply to a type of regime which had prevailed before the 1890s” when desert locust swarms had also been rare. The hypothesis is that these prolonged locust recessions are associated with a shift of the desert zones toward the equator, with equatorial rains (and the ITCZ) more concentrated than before close to the equator (Rainey 1989). It is recognized that these changes do not necessarily entail a direct causal relationship.

The challenge of managing locust invasions

It is a major challenge for the world scientific community - particularly Africans, who have lived with periodic marauding locust swarms for at least 600,000 years, since humans first settled in the Sahel - to begin to manage this natural disaster effectively. It is the belief of the International Centre of Insect Physiology and Ecology (ICIPE) that long-range desert locust management technologies will arise only when scientists know the behavior of locusts intimately and can interrupt mechanisms that regulate such behavior.

Attempts to change the physiology and behavior of the locust between the solitary and gregarious phases are at the heart of ICIPE’s innovative attempts to ground the locust - to make it remain sedentary, behaving as a solitary, grasshopper-like population - and to keep it permanently nonswarming, nonmigratory, and nonmarauding. These efforts, never before tried, are one approach to sustainable management of the desert locust. Other approaches include use of the locust’s natural enemies (including parasitoids and pathogens), the selective use of chemical locusticides, and a more refined locust forecasting and monitoring system. It is ICIPE’s conviction that interrupting the gregarization, the sexual maturation, and the oviposition pheromone systems (which together bring about the rapid and synchronized development of coherent, highly dense locust swarms), together with promoting the solitarization and sexual maturation-inhibition pheromone system (which intensify solitarization), should help locust populations maintain a relatively steady sedentary, grasshopper-like lifestyle. By accomplishing this goal, ICIPE hopes it will help the pheromone-maintained solitary locusts continue to exist as natural herbivores in Africa’s semiarid breeding grounds, continuing their important role as a participant in the dynamics of the savannah ecosystems, without periodically breaking out into locust plagues.

Disasters and development in East Africa

Daniel D.C. Don Nanjira

Despite countless disasters in East Africa, no data are available about their frequency, complexity, or magnitude. The Intergovernmental Authority on Drought and Development (IGADD) was formed in 1986 to strengthen the disaster preparedness capabilities of six East African states (Djibouti, Ethiopia, Kenya, Somalia, the Sudan, and Uganda). But IGADD should be strengthened and its mandate expanded to include other nations in the region and other issues besides drought and desertification. Millions have already suffered from disasters in East Africa and things will get worse unless policies change and corrective measures are taken by the East African states and the international community to strengthen disaster preparedness and to make the area self-reliant. The countries of East Africa must coordinate regional research efforts and implement regional strategies to conserve the soil and the environment and to develop agricultural self-sufficiency.

The Sudano-Sahelian belt

In 1986, Djibouti, Ethiopia, Kenya, Somalia, Sudan, and Uganda (part of the Sudan subregion) formed an Intergovernmental Authority on Drought and Desertification, better known as IGADD. IGADD was based at Djibouti. This paper is about that group of IGADD countries - plus Burundi, Mozambique, Rwanda, Tanzania. These countries are part of the Sudano-Sahelian region, which the UN’s 1977 Conference on Desertification defined as “the belt extending across Africa South of the Sahara and North of the Equator from the Atlantic Ocean on the West to the Indian Ocean on the East.” The Sudano-Sahelian belt is a zone of arid and semiarid land bordering the southern edges of the great Sahara Desert. The belt extends from the Atlantic coast almost 2,600 miles across Africa between the latitudes of 10 and 20 degrees north. Broadly speaking, it stretches 3,500 miles, from Mauritania to Somalia, and includes Djibouti. These 2 million square miles are one of the least developed regions in the world. All life in the region depends heavily on sparse and variable rainfall.

IGADD was an attempt to strengthen national and East African disaster preparedness capabilities to deal systematically with the common problems of drought and desertification, and to improve disaster prevention, long overdue in East Africa. These governments wanted to pool resources, coordinate their recovery and development efforts, implement a common strategy to combat drought and desertification, and develop and promote the funding and implementation of cooperative subregional projects.

The IGADD projects have so far concentrated on a food security and early warning system, the development of interregional communications, animal health, and agricultural research and related manpower development. IGADD is basically a subregional development organization with a total land surface of 5.2 million square kilometers - 23 percent of Sub-Saharan Africa. The subregion’s arid and semiarid lowlands receive less than 400 millimeters of rainfall a year. Farmlands cover more than 36 million hectares, or 7 percent of the IGADD’s total surface. Forests occupy 94 million hectares (about 19 percent) of the land surface, and permanent pastures cover 139 million hectares, or about 28 percent. The rest of the land - 41 percent - is unproductive. IGADD’s inland water surfaces cover 31 million hectares, but woodlands and grasslands are the backbone of the subregion.

IGADD countries have a population of about 100 million, about 26 percent of Sub-Saharan Africa’s entire population. Population growth in East Africa is among the world’s highest - 3 percent by World Bank estimates. Between 1972 and 1987, the population grew from 63.4 million to 103.1 million. Kenya, recently said to have the world’s highest population growth rate at 4.1 percent, is now said to have a growth rate of 3.5 percent a year. If this rate is maintained, IGADD’s population will jump from 100 million in 1987 to 168 million in 2000. This is an alarming trend. About 76 percent of the IGADD countries’ population is believed to inhabit rural areas, including 65 million sedentary farmers and about 11 million nomads and seminomads. About 40 percent of IGADD’s 25 million town-dwellers live in cities, the largest of which by 1987 estimates were Addis Ababa (1.6 million), Nairobi (1.5 million), and Khartoum (1.1 million).

Most of the IGADD countries’ economies are agricultural, except Djibouti, whose economy is based on nomadic pastoral and related services. Agriculture contributes most to the gross national product (GNP) in Ethiopia and Somalia (48 percent). Services prevail in the Sudan (51 percent) and Kenya (48 percent). In Uganda and Djibouti, small-scale agricultural farming predominates. All told, there are about 12 million farming families - but no IGADD nation is self-sufficient in cereal.

Livestock in IGADD nations is dominated by smallholder production. About 2.3 million pastoral families have about 96 million tropical livestock units. Foreign exchange shortages are serious and health conditions are generally poor. Life expectancy ranges from 44 years in Somalia to 52 in Kenya - by far the lowest of any region in the world. Djibouti and Uganda fall below dietary standards set by WHO.

The East African disaster belt

The so-called Third World countries seem to suffer more from disasters than the developed nations. Why this is so is not clear. It may be because of political turmoil and economic exploitation in developing areas. It may be because their location subjects them to the geological and climatic forces of nature that occur mostly in the tropics. It may be because of the unholy alliance between disasters and development; because of the poverty syndrome; because Third World countries lack the resources to respond to disasters, which often destroy the infrastructure crucial to their socioeconomic development; because poverty, inadequate health care, and chronic food shortages make them susceptible to epidemics.

To the natural disasters so common in Africa have been added those of humans living collectively. Population growth has outpaced agricultural production. Arable land is scarce and expensive so low-income families that cannot afford land and decent housing are forced to move to shanties or other vulnerable housing in marginal areas such as the slopes of steep hills, along riverbanks and in flood-prone areas, in artificially reclaimed areas, active volcanic zones, and unsafe houses not designed to withstand extreme events.

East African economies are vulnerable to perilous events and East Africa has been prone to disasters from time immemorial. But the past 30 years have brought catastrophe of unprecedented proportions. East Africa’s natural hazard belt is characterized by heavy rains, floods, landslides, drought, and desertification. East Africa’s economic vulnerability to these disasters is aggravated by armed conflicts that swell the ranks of displaced persons and refugees. The gravest toll from natural disasters is the result of drought, desertification, and to some extent of pest infestation.

The East African economies are small economies, mainly agricultural in nature, situated in a natural hazard zone. These economies are highly specialized. Each country normally exports one or two primary commodities. So the economies are highly vulnerable to the vagaries of international trade, inflation, fluctuating prices for oil and other commodities, and other external shocks the countries cannot control. These countries face severe balance of payments and foreign exchange problems.

The East African economies are among the least developed in the world by any measure: by per capita GDP; proportion of the labor force outside the agricultural sector; and per capita social facilities, energy use, and road mileage. The East African peoples are among the world’s most dependent upon imported manpower for jobs requiring specialized training and skill; imports of fuel for energy and basic food for consumption; capital imports for economic modernization; and external determinants for economic growth. They also depend for their livelihood on the weather and a healthy ecological balance - and both let them down.

East Africa is part of Sub-Saharan Africa, the only region in the world where per capita food production has declined. Famine, hunger, and malnutrition are the result not only of drought, disease, and flooding, but of the ever-widening gap between the rate of agricultural (especially food) production and the rapid population growth rate, especially in the last 30 years. Hunger and famine accelerate the exodus of rural people fleeing drought in search of water and food for themselves and their livestock. They also increase the need for food imports. Imports of food aid to meet emergency needs divert resources from longer-term planning. Regular supplies of emergency food retard development.

Africa south of the Sahara has the highest population growth in the developing world. It also has the most developing countries in the world, some of which - Burundi, Ethiopia, Rwanda, and Somalia - are also among the poorest. Africa’s soils are high in iron and aluminum (laterite) and are mostly infertile. The iron and aluminum compounds become hard on exposure to the sun and air. Sand and laterite erode easily and hold little water. This, plus erratic, deficient rainfall, results in severe shortages of water and animal feed. The resulting devastation of crops and livestock has fed rural migration, rapid urbanization, rapid urban population growth, and the bottlenecks that accompany it. Meeting Africa’s unmet needs requires systematic water control (irrigation) and research and development into how to improve seeds and produce higher yielding crops.

African agriculture has been heavily shaped by policies designed to integrate colonies into metropolitan trading networks, often to protect populations of settlers. Government policy-makers have paid too much attention to urban issues at the expense of rural economies, have focused too much on armaments and too little on investment, too much on centralization and too little on decentralization and popular participation in government activities. There are no clear policies or government machinery for disaster management or for creating public awareness of what needs to be done in case of disaster.

The lack of national mechanisms for disaster prevention and mitigation increases East Africa’s susceptibility to disaster. It is difficult to analyze the economic effects of disasters on the subregion because there is so little data on the nature, extent, and effect of disasters and emergencies, especially in the past. Although drought and desertification take the heaviest economic toll, about 90 percent of human deaths and property damage from disasters are attributed to water and wind. There is a close relationship between disasters, development, and environmental degradation and the resulting economic retardation and poverty. Official and public complacency and ignorance about disasters increases the region’s vulnerability to them.

Disasters, development, and the environment

Desertification worldwide is spreading at the alarming rate of 6 million hectares a year, and erosion is damaging Africa’s soil 26 times faster than it was 30 years ago. The Sahara Desert is said to be advancing southward at an alarming rate - an estimated 125 miles a year. There is a real danger that Africa may become permanently plagued by drought and desertification. The spread of the Sahara over an area 40 times the size of Switzerland has involved the destruction of at least 300,000 hectares of arable Sahelian land and forests. The reforestation rate in Africa is only 1:30, which means that 1.3 million hectares of dense forest and 2.3 million hectares of open forest are destroyed annually in Africa, and only 93,000 hectares a year are replanted. The grim realities of hunger, malnutrition, starvation, and death hit small farmers, the rural landless, the urban unemployed, and other marginal groups the hardest.

Bad weather accelerates the loss of topsoil, nutrients, and humus and the dryness that results from water running off a compacted ground surface. The sectors most damaged by drought are agriculture, forestry, fisheries, livestock, meteorology and hydrology, food and nutrition (different food crops), industry, resources, transport, education, and health. Deforestation contributes to soil erosion and flood damage in East Africa. Floods have caused heavy casualties and economic losses in Kenya, Tanzania, Uganda, and the Sudan. Deforestation and soil erosion are the subregion’s most pressing environmental problems, and subsistence agriculture and heavy use of fuelwood are largely to blame. In drier areas, deforestation leads to desertification, which, once started, is irreversibly catastrophic. The tragedy of the Sahel illustrates the grave environmental and economic consequences of desertification and drought. Reduced agricultural production causes food crises that ruin human health, divert foreign exchange earnings from development to food imports, and create a dependence on food donations that are often used as political weapons. Prolonged food shortages accentuate social inequities, lower morale, and cause political and social instability. The grave effects of disasters and environmental degradation are felt most in the following sectors of the East African economy:

· Agriculture, East Africa’s main economic sector, foreign exchange earner, and source of employment.

· Tourism, a leading foreign exchange earner for Kenya and some other East African countries (for which conservation of nature and wildlife is essential).

· Industry, which is developing and needs more investment.

· Water supplies and management (needed for irrigation, sanitation, and nutrition).

· The environment, land use, forestry, and fisheries development.

· Livestock.

· Infrastructure (transport and communications, storage facilities, logistics, warehousing, and distribution).

· Energy, especially alternative renewable energies.

Changing the priorities of policymakers

To be effective, policies designed to deal with disasters in East Africa must also address problems of development and the environment, as the three are intertwined. Food and agricultural policies are most important, not just to East Africa but to the whole continent. For 30 years, agricultural production has deteriorated as urbanization and the population have grown, so per capita food consumption has declined. Post-harvest losses have also increased, as has Africa’s dependence on food imports. The resulting drain on foreign exchange earnings has retarded development in all African economies. Africa has a food problem largely because agriculture is not given the priority it deserves by policymakers. Enough resources should be allocated to promote agricultural and food productivity, which would improve welfare, especially in rural areas where most Africans live. Policies must:

· Give incentives to rural development and rural small-scale farmers and cooperatives.
· Involve rural women and youth in agricultural development.
· Improve the living conditions and real incomes of farmers and ordinary people.
· Encourage self-sufficiency in livestock and fish production.
· Reduce food waste.
· Diversify agricultural development.
· Strengthen food security.

Food security and early warning systems go together. To increase food production it is essential to develop improved seeds and aim at incentive producer prices. As part of early warning systems, it is essential to help national governments monitor crop conditions and food supplies and be alert to adverse trends.

To control drought and desertification, policies should protect vegetation, trees, and shrubs to prevent erosion, provide water catchment, maintain biological diversity, and produce fuelwood and fodder. About 95 percent of wood consumed in Africa is for fuelwood - Africa’s main energy source. To sustain this energy source, it is essential to launch reforestation programs, improve stoves, teach people more efficient ways to produce charcoal, conserve and rehabilitate rangelands, reduce and control overgrazing, make livestock production more efficient, and provide animal health centers, public information and education programs, and training facilities for farmers and pastoralists.

Water resources should be developed and better managed. These resources should be regulated, improved, and distributed evenly. The East African nations should provide research and training, should organize an integrated approach to lake and river basin development, should promote irrigation schemes, and design water points, boreholes, wells, and small reservoirs that resist pollution by barring inappropriate animal and human access.

Improving interregional cooperation will require improving transport and road access to drought-stricken areas so strategic food reserves and aid can be mobilized in emergencies; opening up land-locked areas by linking them to harbors; and improving telecommunication and power transmission between states.

There should be research and training to improve land use, soil conservation, and crop yields. Institutes of applied agricultural, livestock, and forest research should give priority to containing ecological (environmental) degradation, developing energy resources (especially new and renewable energies), developing environmental manpower, and improving communications between researchers, planners, and implementers.

Policymakers must address the problems of refugees and displaced persons (of which there are at least 2 million in East Africa alone), population growth, family planning, and land tenure. But most of all they must promote:

· Sustainable agricultural production that does not deplete the natural environment.

· Protection of crops and strategic food reserves, especially through pest control and improved storage and processing (drying and preserving) facilities to minimize post-harvest food losses.

· Intensified rural production to conserve natural resources and attain food self-sufficiency.

· Redevelopment of exhausted farmland, especially the rehabilitation of run-down irrigation systems.

· Improvement of village water management, including small-scale irrigation and water harvesting.

· Pastoral development.

· Disaster insurance, especially for tourism and wildlife.

· Afforestation.

· Development of drought-resistant livestock, trees, and crops.

· Increased productivity on arid and semiarid land.

· The formulation and implementation of national plans, programs, and projects for rehabilitation, recovery, and long-term development.

Emergency measures to relieve disaster victims should be short-term (last only three months), and rehabilitation and recovery programs medium-term (only three to 18 months). Their end should mark the beginning of the long-term development period (18 months and beyond) during which the focus is (East) Africa’s environmentally balanced socioeconomic development. All disaster-related projects should be either national or subregional. Subregional development policies endorsed by the national governments of cooperating states should be implemented as national projects coordinated to attain common strategic objectives. Subregional projects should harmonize national efforts toward the common objectives of subregional policies and national schemes.

Many policies are wrong-headed and should be changed. Food aid, for example, should become a tool for development. Food aid must be accepted in emergencies but policy should be to promote development, not to depend on relief or emergency food aid. Development strategies are wrong that stress the production of cash crops for export at the expense of food production for domestic consumption. Because of inappropriate pricing policies for crops, it has been difficult to supply enough food to urban populations and to create adequate food reserves. Other government policies produce inequalities in income distribution; neglect rural areas in overcentralized development efforts; neglect environmental concerns and ignore disaster mitigation in planning national development; fail to promote growth with equity (public investment that eliminates growth distortions and supports the poor); and fail to promote crop diversification, to reduce the economic losses from disaster.

Effective disaster management in East Africa should aim to:

· Provide for financing disaster preparedness, prevention, and mitigation.
· Strengthen national strategies and mechanisms for disaster management.
· Involve local people in national development and disaster management activities.
· Provide public information and education programs for the public, schools, and workplaces.
· Develop a coordinated approach to resolving East Africa’s environmental, development, and disaster problems.

Ill-conceived policies and programs aggravate rather than mitigate disasters. It is also important to coordinate action and support from the international community.

Action plans

Any action plan for dealing with disasters should:

· Look to the year 2000 and beyond.

· Develop strategies (measures) for all levels, from local to global.

· Establish a network of national disaster relief coordinators.

· Provide for community and regional training.

· Provide a regional plan for a network of regional training and information systems that cover medical surveillance, water stocking, drought surveillance, the provision of adequate storage facilities and grain stocks, and the like.

· Establish within the regional network a joint stocking facility, joint training facilities, joint transport of goods and services, joint health services, a regional program of feeder roads (currently vulnerable), and such emergency services as a joint flying doctor service.

· Jointly solicit international and bilateral technical, financial, and other support.

· Reactivate IGADD, which is too weak as it is. The IGADD ambassadors in Rome and elsewhere should be mobilized and asked to formulate practical proposals on ways to combat disasters. Research should be done on improving rural infrastructure, including road transport; an early warning system for food crops, locusts, climate, and natural hazards; environmental hazards; and public information and education programs.

How donors can help

The primary responsibility for development and for dealing with disasters and environmental hazards rests with the East African governments themselves. But these nations’ needs are too enormous to be satisfied without external financial and technical assistance. Financing institutions such as the World Bank could provide technical and financial assistance for long-term development projects, human resources development, and insurance coverage for disasters. FAO, IFAD, the OPEC Fund, UNIDO, WFP, WHO, and other agencies could provide assistance to help poor farmers; develop infrastructure; increase agricultural productivity; carry out research and development to strengthen disaster management capabilities; tackle disaster insurance problems; develop industrialization, support the recovery and rehabilitation of water resources, renewable energy resources, and food production, processing, and storage facilities; support relief supplies and food aid for development projects; develop information and early warning systems for crops and food security; and facilitate access to remote sensing of meteorological conditions.

HABITAT, UNDP, UNDRO, UNEP, UNHCR, UNICEF, and WHO can help with the preparation of national development plans and disaster preparedness activities; training, research, and other disaster management activities; technical assistance for developing national emergency policies and promoting public awareness for disasters and emergencies. Other antidisaster institutions such as the International Decade for Natural Disaster Reduction could help provide disaster management skills for East Africa. Workshops and seminars on disaster problems should be encouraged, and promoted on a larger, more systematic, scale.

Annex 1

Table 1 East Africa’s disaster-proneness and economic vulnerability


(square miles)



Population density per square kilometer

Date of independence

Year of UN membership

Economy/main agricultural commodities

Per capita GNP
(US$) *

Type of disaster

Cost of disaster as percentage of GNP



Tropical volcanic soils, irregular rains





Subsistence agriculture/coffee, beans, groundnuts, sweet potatoes


Drought, famine, refugees, civil strife, displaced persons, floods, pests, epidemics




Arid volcanic rocks, torrid, high tropical monsoons





Trade services/livestock, vegetables, sheep, goats, asses, cattle, camels, fishing


Drought, pests, epidemics, civil strife, floods, refugees




Tropical plateau, semidesert



Ancient times


Agriculture/tobacco, barley, maize, potatoes, beans, sugarcane, groundnuts, coffee, cotton, livestock products


Drought, desertification, famine, pests, civil strife, epidemics, earthquakes, floods, refugees, displaced persons




Equatorial tropical forests





Fruits, sugarcane, cotton, cotton seeds, forestry, sisal, maize, millet, sorghum, cashew nuts, coffee, pineapples, tea, coconuts, pyrethrum, tobacco


Drought, desertification, refugees, floods, pests, earthquakes, epidemics




Tropical forests, dry and hot





Cassava, cotton seed, cashew nuts, groundnuts, maize, fruits, livestock products


Floods, civil strife, cyclones, epidemics, famine




Tropical, wet and dry, marshy, volcanos





Agriculture/cassava, beans, tea, coffee, peas, livestock, maize, sorghum


Drought, civil strife, famine




Savannah plains, semideserts





Pastoral/agriculture (irrigated and plantations), maize


Drought, desertification, pests, epidemics, oil spills, refugees, displaced persons, civil strife




Flat tropical plains, semideserts





Agriculture/sugarcane, forests, cotton seed, groundnuts, millet, sesame seeds


Desertification, famine, floods, epidemics, pests




Tropical rainforests, woodlands





Agriculture/rice, sisal, sesame, cotton seed, maize, millet


Desertification, famine, drought, epidemics, floods




Equatorial tropical forests, plateau





Agriculture/coffee, maize, bananas, beans, groundnuts, millet, sweet potatoes


Epidemics, drought, floods, famine, civil strife, refugees, displaced persons, desertification


*Nine nations are less-developed countries; Kenya is low-income.

The link between reconstruction and development

Jelena Pantelic

Reconstruction after an earthquake should improve the residents’ standard of living. Local social and cultural values and resources should be incorporated into reconstruction and development efforts.

It is a fallacy that disasters do not choose their victims, but strike all people alike. Quite the opposite seems to be true. Disasters often particularly affect the most vulnerable segments of the population, people who can afford to occupy only the dangerous flood-prone valleys or the edges of ravines and landslide areas, or who live in substandard homes or work in unsafe buildings. Poverty lies at the root of disaster vulnerability just as it lies at the root of most problems developing societies face today. So efforts to reduce disaster vulnerability are inseparable from general development efforts - especially after a disaster, when reconstruction becomes a primary medium of development policy.

This view of the link between reconstruction and development has been gaining ground, replacing the old notion of competition for resources - in which recovery was seen as diverting funds from development efforts. Reconstruction after disaster is viewed today as a process that can effectively unify development and recovery goals - by improving the disaster resistance of physical structures, improving the standard of living, generating new jobs and creating new skills, and integrating them with the community’s social and cultural values and resources.

Rebuilding physical structures

To rebuild damaged or destroyed physical structures has always been the main goal of communities affected by earthquakes. Rather than simply restore buildings to pre-earthquake conditions that may have contributed to their vulnerability in the first place, reconstruction should improve a structure’s quality, especially its earthquake-resistance. Reconstruction of physical structures after earthquakes should both reduce seismic hazards and upgrade the standard of living.


To allow development to continue after future disasters, the first step is to improve structural performance during an earthquake. This usually begins with regulation - adopting new or revising existing building codes. Mexico City’s Emergency Building Code, for example, was in place just five weeks after Mexico’s 1985 earthquake (Esteva, forthcoming). But rigorous regulations on design and practice mean little without proper enforcement, which is one of the weak links in improving earthquake-resistance in physical structures. Inadequate implementation of the building codes, rather than their absence, was one of the principal causes of deadly building failures in the 1988 earthquake in Armenia.

Throughout history reconstruction after earthquakes has inspired new construction technologies. The famous wooden frame called gaiola became a standard component of masonry construction in Lisbon after the catastrophic earthquake of 1755 (Tobriner 1980). More recently, reconstruction programs in Guatemala successfully promoted the use of lamina (light-weight aluminum sheeting) as roofing material instead of the traditional heavy ceramic tiles that had proved lethal in the 1976 earthquake (Bates 1979).

Construction techniques and materials are important but so is the location of a building and the quality of soil on which a structure is built. Land-use planning should regulate development in vulnerable sites. Relocating entire settlements to safer sites is one of the oldest, most radical land-use measures, but in the long term it rarely succeeds. The residents of Antigua, Guatemala, for example, reluctantly abandoned their city after Spanish authorities officially relocated it in 1779 - only to return several years later (Tobriner 1980). Similarly, the town of Gediz in Turkey, which was ruined in the 1970 earthquake and rebuilt on another site, exists 12 years later in two thriving neighboring locations, Old and New Gediz (Aysan and Oliver 1987). Yugoslav authorities seriously considered relocating Skopje after a 1963 earthquake, but opted instead for a restrictive land-use strategy by rezoning the seismically hazardous Vardar valley as open space (Davis 1978).

Introducing new building regulations and technologies requires a well-organized education and training program if earthquake mitigation measures are to be effective. Training technical personnel and the population at large in earthquake-resistant construction has become more common in the past 20 years. In Guatemala after the 1976 earthquake, for example, some programs trained residents how to distinguish between safe and vulnerable locations for their homes and how to improve the traditional building process, rather than providing them with complete houses (Davis 1978, Cuny 1983). Similarly, in Nepal, which is now recovering from the effects of the 1988 earthquake, several dozen demonstration houses have been built in strategic locations to show safe building practices to people from remote areas (Fujiwara and others 1989, Kreimer 1989). In the last decade, earthquake education courses for professional architects, engineers, and planners have proliferated in the United States - and the public at large has not been neglected. Specialized earthquake preparedness projects in California disseminate information about the threat of earthquakes and provide guidelines for self-help improvement of the earthquake performance of homes (BAREPP 1990).


Substandard physical structures, especially in low-income residential areas, are common in developing countries. Dwellings often lack kitchens, bathrooms, and water, as well as access to sewers, electric power, paved roads, and health and education facilities. More often than not these residential quarters are overcrowded and poorly maintained and are rarely owner-occupied. This makes them particularly susceptible to earthquake damage, as was shown in earthquakes in Mexico City in 1985, the Italian countryside in 1980, and the medieval cities on the southern coast of Yugoslavia in 1979. So physically upgrading structures after earthquakes to improve the occupants’ standard of living has become an important goal of reconstruction. In Mexico City, for example - in a remarkably successful large-scale attempt to reconstruct residential buildings - both the buildings’ earthquake-resistance and the occupants’ standard of living were improved. Before the earthquake, the average size of an apartment was 22.25 square meters, shared by an average 4.37 occupants, with 63 percent of households sharing bathrooms and 30 percent sharing kitchens with other families. The rebuilt dwellings were an average 40 square meters and contained two bedrooms, a living and dining area, a bathroom, kitchenette, and washing area (Puertos 1987, Stolarski 1987, RHP 1987).

Increasing the size and quality of dwelling units alone will not improve the occupants’ standard of living, if that is done at the expense of more vital necessities. Kreimer (1980), for example, shows that access to jobs and services is more important to the low-income population in developing countries than the quality of the structure they occupy. In Managua, after the 1972 earthquake, long journeys to work reduced the attractiveness of the low-income residents’ new, well-equipped housing units built on the city’s outskirts (Bolton 1988). By contrast, residential neighborhoods in Mexico City were rebuilt on the same site near the city center - a vibrant focus of the formal and informal economy - which enabled people to maintain their position in the life of the community.

Social aspects of reconstruction

Local social and cultural values must be considered in development and reconstruction programs. According to Weitz (1986) - who calls for “massive representation” of local communities in development projects - a “major reason for the recurrent failures of past development efforts is the neglect to involve value systems in development planning and implementation.” Similarly, Goulet (1978) asserts that the values of the society itself must calibrate the terms of its development. Analysts of recovery programs after earthquakes and other disasters link the failure of many reconstruction programs to a lack of respect for the social and cultural values of the affected communities (Aysan and Oliver 1987). Two objectives of reconstruction should be to strengthen the local community by using its resources and to incorporate the cultural values of the community into the reconstruction process.


Many people participate in planning and implementation of reconstruction after an earthquake, but the victims are often left out of the process. In the short term, reconstruction is thus deprived of local skills, experience, manpower, institutions, and sometimes significant funds for rebuilding. In the long run, the community is robbed of the invigorating experience of rebuilding and runs the risk that its lifestyle may be changed beyond recognition. Outside decisionmakers often bring ready-made solutions, foreign technologies, and inappropriate lifestyles to communities whose residents are excluded from meaningful participation in reconstruction. Turkish authorities, for example, provided prefabricated housing with modem amenities to the victims of the Gediz 1970 and Lice 1975 earthquakes, housing that proved to be sorely inappropriate for the traditional rural and nomadic culture and religious beliefs of the affected population (Aysan and Oliver 1987, Cavanagh and Johnson 1976).

Involving the local community and resources in reconstruction planning and implementation can materially strengthen a community hit by an earthquake. For example, in Renovacion Habitacional Popular (RHP) - the groundbreaking housing reconstruction program that provided almost 50,000 housing units in Mexico City - decisionmaking involved representatives of all concerned groups from the local communities in which reconstruction was taking place, including neighborhood associations, tenement groups, and church organizations. The result of negotiations, a formally signed social agreement, became a blueprint for implementing reconstruction, in which the local community continued to participate actively (RHP 1987, SEDUE 1987).

Assistance from outside the community - although often welcome and sometimes necessary as a “catalyst” for successful reconstruction - must not take on a leading role. Foreign aid can expedite development but “cannot in and of itself develop a poor nation” (Weitz 1986). The same applies to reconstruction. Local participation brings in the special knowledge, work experience, and technical and organizational skills needed for successful recovery. Local involvement in reconstruction provides a framework for social and economic community development. In Mexico City, for instance, many funds were funneled back into the communities affected by the earthquake. Renovacion Habitacional Popular created about 115,000 new jobs in the construction industry, which were filled almost exclusively by locally recruited workers (RHP 1987). And the community’s economic base was strengthened in the long term by providing production shops and commercial spaces for local businesses.

Two objectives in Mexico City were to strengthen the alliance between the public and the private sector and to reinforce existing institutions rather than establish new institutions. RHP was organized as a task force, combining experts from the public and private sector. Members of this group were on “loan” from their parent organizations as long as reconstruction lasted. Then RHP was dissolved, preventing the institutionalization and bureaucratization of the reconstruction program. The public sector provided institutional support, a timely flow of information and funds, and shortcuts in bureaucratic procedures, while the private sector contributed practical experience in finance, design, construction, and management.

Urban growth and natural hazards

Michael Cohen

There is a great need for research and development on the urban environment, particularly on the relationship between global and urban environmental problems. If global warming does occur and sea levels really do increase, think what the implications are for cities such as Bombay, Lagos, Rio, Dacca, Manila, Shanghai, and New York. What will happen to the quality of the aquifer? If the water supply for such huge populations is ruined, what will happen to economic activity? If half of the GDP is generated in the cities, how will that affect economic growth? What are a country’s economic prospects without its major urban centers? Or look at it the other way: How do cities affect global environmental problems? Cities pollute - and eat up scarce environmental resources. We don’t know as much as we need to know but one thing is clear. We must define urban policy and take action in urban areas. Four aspects of urban growth demand special attention in discussions of natural hazards: demographic growth, constraints on urban productivity, spatial growth, and weak local institutional capabilities.

Demographic growth. In 1960, only one city - Shanghai - had a population of 10 million; in the year 2000, 17 cities are expected to have populations above 10 million. And secondary cities in most developing countries are growing faster than many larger cities. The urban population, which was about 1.3 billion people in 1988, will expand by about 600 million people by the year 2000. Urban growth continues, the numbers are huge, and no effort to restrain urban growth - such as limiting migration - has worked. It is impossible to keep people in rural areas and in any case growth often occurs in the cities themselves, from natural increase. Cities impose tremendous pressures on environmental resources, especially air and water. What is the carrying capacity of large cities? We do not know, as cities are growing to unprecedented levels.

Constraints on urban productivity. Cities are crucial to economic productivity. Almost half of the gross domestic product (GDP) in most of the Bank’s member countries is generated in urban areas. This is less true in rural African countries, but certainly applies in such countries as Argentina, Brazil, and Mexico. About 80 percent of future growth is expected to come from the urban economy.

But in most developing countries, cities are not as productive as they could be. There are serious constraints on productivity, the most obvious being deficiencies of infrastructure, such as traffic problems in places like Cairo and Mexico City and shortages of water in Karachi and Bombay. Economic policies make daily economic activity far too complicated and complex regulatory frameworks hinder rather than regulate growth. But worst of all constraints on productivity are the weak institutional frameworks for managing cities. The result is extensive urban poverty. By World Bank estimates, about one quarter of city dwellers - roughly 330 million people - live below the urban poverty line. And a substantial portion of the 600 million to be added to the urban population will be urban poor.

Spatial growth. Cities are expanding rapidly into peripheral areas. Areas that once were agricultural land are now inhabited by increasingly dense populations. Employment tends to decentralize; areas that were once residential, and before that agricultural, are now becoming employment centers. Areas that are now residential were once zoned for industry and still may be industrial to some extent. Land used for mixed purposes creates hazards, introducing industrial pollution and hazardous waste to residential neighborhoods, for example.

With spatial growth in some areas comes densification and centralization in others, which increase rather than decrease the risks associated with natural disasters. Pollution is heavy, air quality is deteriorating, marginal lines are being settled, and environmental risks in cities are increasing rapidly. But among data on the developing countries, you will find surprisingly little information on the urban environment.

Weakened local institutional capabilities. In most countries local governments are weaker now than they were 20 years ago - particularly in infrastructure. Administrative pressures to centralize government have starved local governments of revenues, autonomy, and technical capabilities. Urban infrastructure in most parts of the world is in a crisis, with poor maintenance undermining the value of development investments. Institutional capabilities are especially poor in zoning and the enforcement of environmental regulations. Even data on air and water pollution in Mexico City and most African and Asian cities are not available.

Our phenomenal ignorance of the escalating economic, social, and political stakes of urban growth makes risk assessment exceedingly difficult. The likelihood of environmental disasters is probably increasing, and the prospects of their affecting cities is certainly increasing - because human activity is becoming densified and residential, and industrial activities coexist in the same neighborhoods in interactions about which regulations and policy are not clear. We must work together to identify ways to reduce these risks, but there is much to be done.


Physical structures identify neighborhoods, embody a way of life, and express the cultural values of the community. One important measure of a reconstruction program’s success is the extent to which a community can preserve its cultural identity and lifestyle. Urban ambiance, the historical heritage, and traditional architectural values are frequently destroyed by an earthquake, but often reconstruction programs sacrifice these values, too, damaging the social fabric of the community. Before the 1963 earthquake, for example, Skopje was a closely knit city with a strong medieval Ottoman heritage. Reconstruction converted it to a low-density, thin, linear city, 24 kilometers long - changing forever the lifestyle of its citizens (Davis 1975).

More efforts were made to maintain the cultural identity of communities recovering from earthquakes and other disasters in the 1970s. The cultural values of the victims in Guatemala, for example, were integrated into its reconstruction programs (Thompson and Thompson 1977). Of course, some strategies to preserve a place’s cultural identity may cause hardships for its residents. Venzone, for example - a historical town damaged in the 1976 Fruiuli earthquakes - embarked on a slow process of rehabilitating its historical buildings and preserving its unique ambiance at the expense of its residents, who were displaced to prefabricated dwellings outside of town (Hogg 1980). Mexico City’s local residents, however, were not displaced but camped near the reconstruction sites, observing and helping in the rebuilding of their new homes. The concept of the traditional neighborhood and the spirit of the “vecindad” were skillfully translated into the design for new condominiums, and the historical “vecindades” - the subjects of special architectural conservation projects - received their original residents as soon as they were rebuilt.