FAS Public Interest Report
The Journal of the Federation of American Scientists
Summer 2003
Volume 56, Number 2
FAS Home | Download PDF | PIR Archive
Front Page
Nuclear Terror: Ambling Toward Apocalypse
21st-Century Physics: Grand Challenges
The Afghan Housing Crisis: Can New Technology Make a Difference?
Congress Permits Research on Smaller Nuclear Weapons
Molecular Manufacturing: Start Planning
Progress Report for FAS Learning Technology Initiatives
Challenging Conventional Wisdom on Arms Exports
How Well did TOPOFF2 Prepare Us?

The Afghan Housing Crisis: Can New Technology Make a Difference?

by Henry Kelly

Millions of Afghans facing the coming winter still live in temporary camps or camp in the shells of ruined buildings. Living in temporary housing, with little privacy and unhealthy conditions, many Afghans have not been able to escape the trauma of their protracted wartime environment. Longtime FAS member Art Rosenfeld asked whether the US science community had something useful to offer. The answer seems to be yes. But as might be guessed, finding the right technology to address the housing crisis may be the easiest part of the problem.

The Challenge:

Twenty-six years of almost continuous warfare, coupled with major earthquakes in the past decade, have damaged or destroyed much of the housing stock. Pressure on existing stock is growing rapidly as many of the six million Afghans that fled to Pakistan, Iran and other nations during the war begin to return. A population of 27 million is now struggling to accommodate the more than 1.8 million refugees who were estimated to have returned in 2002 alone. While funding from the US and other nations is woefully inadequate and unpredictable, some progress is being made. Funds typically go to non-governmental organizations (NGOs) with facilities in Afghanistan who struggle to help communities rebuild. In an effort to use local resources and building traditions, as well as to save funds and take advantage of available skills, most of these projects rely on time honored Afghan construction methods, using handmade mud bricks. Flat roofs are supported by wood beams covered by layers of branches, woven mats, and finally up to meter of clay. These structures can be built for less than $1000.

Though inexpensive to build, these traditional homes present major long-term risks. The most obvious problem is that the structures are death traps in earthquakes, and Afghanistan is one of the most active seismic regions of the world. Houses in Afghanistan should be designed to meet roughly the same standards as Los Angeles (4 m/s2 acceleration), but traditional methods founder at much lower levels. Mud walls are extremely brittle and fail when shaken and the enormous weight of walls and roofs cause disastrous injuries. More than 6000 people died in two earthquakes, four months apart, which shook the Afghanistan /Tajikistan border in 1998, even though they measured only 6.1 and 6.9 on the Richter scale.

A partially demolished house, common in Kabul.
Photo courtesy of www.esamskriti.com.

Traditional construction has also become more difficult because of the scarcity of wood. Many of the NGOs are forced to import wood from Pakistan and other nations since decades of nonexistent forest management have devastated Afghanistan's local timber supplies. Those not able to import wood are undoubtedly making do with inadequate, and dangerous, roof supports.

Wood shortages also underscore the energy crisis facing the nation. Traditional Afghan homes are heated with wood or charcoal. The difficulty in obtaining traditional fuels has forced many to turn to expensive kerosene or imported coal. Traditional mud homes have enormous thermal mass and can help keep the structures cool in the summer. Kabul has an altitude of 1800 m and nights are cool, but the winters are very cold (the average January temperature is 27F) and the mud walls provide little insulating value. These factors force a difficult choice between expensive fuel consumption and uncomfortable temperatures.

Kabul, Afghanistan.

Traditional heating and cooking systems also lead to terrible air quality inside the homes. While the mud homes are not airtight, the fires are not vented, leading to enormous buildups of combustion products. Lung and eye problems resulting from these pollutants have devastating effects, particularly on women who spend a larger fraction of their time indoors and close to the stoves.

The people building new homes in Afghanistan understand the terrible risks they're taking by putting people in unsafe traditional structures, but there appears to be no alternative.

Our Response:

The Federation of American Scientists is spearheading this effort to develop a low cost and energy efficient housing design. We began by talking with everyone we could find who knew something about conditions in Afghanistan and developed a set of performance goals. The result is summarized in Table 1. We sought a design that worked in Afghanistan but since most of the criteria are universal, we hoped to develop a solution that would be widely applicable worldwide - including the US.

We were helped by a number of Afghan engineers and scientists in the US and scholars who have worked in the Central Asia region. Hashem Akbari and Ashok Gadgil from the Lawrence Berkeley Laboratory, Les Norford from MIT, and Kirk Smith from UC Berkeley provided expertise in building technology, energy analysis, and interior air quality. Joe Colaco from the University of Houston (and CBM Engineers), brought expertise in engineering analysis and Roger Rasbach (of Rasbach Design) contributed architectural expertise and extensive knowledge of panel construction.

An attempt to find out how the US assistance program through the State Department and USAID tried to influence the technology of Afghan housing ran in circles. Their main interest is that funds be given to a group that knows how to erect homes in Afghanistan. The otherwise futile pursuit did, however, result in one of those wonderful moments of serendipity, when a State Department official, sotto voce, let us know that one NGO she'd worked with seemed particularly competent. This led to a contact with the humanitarian relief and development organization Shelter for Life International, Inc. (SFL), which has been providing housing in Afghanistan and elsewhere for many years, and their wonderful chief architect Harry van Burik.

We couldn't get serious about design work, of course, without some source of funding. This is the part of any project that no one likes to talk about but lies at the heart of whether anything actually gets done. However important, the project didn't fit into any funder's bailiwick. The California Energy Commission (CEC) was willing to support design work as long as it had clear value for low-cost, seismically resistant, energy efficient housing in California but wanted to cost share with someone. The Department of Energy (DoE), the obvious co-sponsor, has a great staff who completely understood the problem, but their offices have been in a state of constant reorganization and budget tightening. In addition, their mission is energy, not seismic safety or other construction issues. After much negotiation they found funds that could be spent to design energy efficient homes for Native Americans. The combination of DoE and CEC funds allows us at least to start serious analysis of alternative designs and pay for some limited testing of components and integrated building systems.

The Engineering design

So we were off and running with a semi-impossible set of design specifications and an enthusiastic alliance of a relief & development organization, Afghans, university professors, engineers, architects, Native Americans, and people who represented many dimensions in this space.

This is one of the 5,000 houses SFL built in the Baghlan Province, Northern Afghanistan.
Photo courtesy of Shelter for Life.

Searching for technology concepts in home construction is not an easy task. The industry remains astoundingly isolated from the management and technical innovations that have transformed most other major business sectors. Survivors keep capital investments and core staff as low as possible. Even the largest homebuilders have no research or engineering staff. The construction process is often a game where minimum-bid subcontractors try to minimize costs regardless of the work created for trade that follows them. Lack of precision means that virtually everything - from drywall to cabinet work, must be hand fitted on site.

Regulatory efforts to increase seismic or wind safety or to increase energy conservation are fiercely resisted and difficult to enforce since the recommended approaches require adding material and labor costs. This is a natural reaction given the industry's lack of capacity for integrating analyses of cost, safety, and efficiency. We, of course, were looking for solutions that were easier and cheaper than conventional methods - techniques where it would be easier, not harder, to meet the performance specifications.

Table 1
Performance Specifications
  • architecturally attractive to the Afghans (traditional home designs)
  • good interior air quality
  • seismically stable (design for seismic zone 4)
  • comfortable in extreme temperatures with minimal use of external energy sources (e.g. highly energy efficient)
  • inexpensive to build (goal is less than $1000 per home)
  • inexpensive to maintain (no deterioration from moisture, heavy use)
  • secure in high winds
  • meets fire-protection standards
  • minimal imports
  • minimal (preferably no) use of wood
  • compatible with sustainable businesses in Afghanistan that can be started with modest capital investment (e.g. <$500K)
  • employ people with locally available skills
  • minimal or no proprietary technology
  • reproducible in other markets (including the US)

After reviewing a number of concepts, some of them stupendously bad ideas that had been inflicted on other USAID recipients, we are investigating a few versions of a single simple design involving styrene panels coated with a cementitious material that provide both structural strength and a finished coating. These are all variants of Structurally Insulated Panels (SIPs) that are used increasingly in US modular homes. The systems typically create a panel from a sheet of styrene foam insulation sandwiched between two layers of plywood or similar materials. Finish covering is applied to the exterior and interior after the panels are installed. The sandwich has good structural as well as insulating properties, can be assembled much more quickly than conventional homes, and ensures that dimensions are accurate, greatly streamlining all other tasks. These systems eliminate the thermal "short circuits" created, for example, by 2x4 studs that penetrate standard insulated walls. These "short circuits" mean that the actual heat flows through standard walls are 30% higher than would be predicted using the properties of the insulation alone - a figure that is even higher if the insulation is improperly installed.

We are exploring even simpler systems that simply erect a styrene shell and coat it with a cementitious material that could serve as a final exterior and interior surface. The only materials that would need to be imported to Afghanistan would be styrene pellets and any material that would need to be added to concrete. Styrene pellets are a worldwide commodity made in India and Pakistan. Imported by truck, they can be expanded into sheets in simple facilities (total cost less than $200,000) and increased in volume by a factor of 25.

Preliminary energy analysis by Norford, based on a housing design prepared by SFL and estimated costs, show that an ideal balance between low cost and high energy efficiency can be realized. We will compare at least four separate methods for coating the walls and roofs that have been used in the US and abroad. Colaco will conduct detailed structural simulations that will lead to tests for components of the preferred systems. Norford and Smith will help design simple methods for ensuring interior air quality even when electricity isn't reliable, or even available. We expect that there will be several iterations to optimize both structural and energy designs - something that is almost never done in construction. Typically the energy guys are called to provide heating and cooling after basic design decisions are unalterable. It's likely that structures for Native Americans in California will differ in some details from systems optimized for Afghanistan.

Next steps

One or more small structures will be built in California to validate the construction methods and record the work in a way that can help communicate the methods to Afghans and others. The structures will be tested on a shake table in California that can validate calculations on earthquake performance. We hope that members of the Afghan Urban and Housing Department will participate in the testing.

We plan to begin actual testing late this summer. The diverse backgrounds of the team members has put us all on a steep learning curve that, if nothing else, gives us new respect for disciplines we knew little about. So, is it possible to develop a revolutionary construction method that could be cheaper, safer, and more efficient in markets worldwide even in the face of a US deadlock over energy policy? We're giving it our best shot. Watch this space.