Charles A. Setchell is Senior Shelter, Settlements, and Hazard Mitigation Advisor in USAID’s Office of US Foreign Disaster Assistance.

In the field of disaster relief and recovery we face a growing challenge of urbanization. This was the focus of a recent panel discussion at the Brookings Institution. Urban areas account for perhaps one percent of the world’s land mass but these areas are now home to more than half of humanity — roughly 3.5 billion people. Because of their scale, complexity, and concentrations of impoverished people living in hazard-prone slums, responding to crises and natural disasters in urban areas poses significant challenges to humanitarian agencies, which often have their institutional genesis and past experience rooted in the refugee crises of rural areas.

Looking ahead, more than 90 percent of total global population growth will be in the cities of developing regions, where resources and institutional capacities are limited. This level of growth will be so significant that the equivalent of a city of Bangalore, nearly six million people, will emerge during every month of every year for the next 20 years. Cities in developing regions will be the dominant form of global human settlement, and slums may well represent the dominant form of global housing design.

An important lesson is to integrate local context in our efforts. Local resources, institutions, expertise, and wisdom exists even in severely damaged human settlements, and should help form the basis for understanding the capacities, resources, opportunities, and disaster impacts that will guide response and recovery activities. Shelter needs should be responded to with a focus on its role within a settlement, not just “four walls and a roof.” In urban areas disasters compel a change in the unit of analysis from household to neighborhood. In Haiti, for example, USAID and other agencies have embraced this “neighborhood approach” as an operational means of working through — and out of — the rubble pile, and initial results are quite promising.

Connected to that, we should not overlook rubble piles and debris fields. They cover land, and clearing land is a precursor activity to housing and larger economic recovery. In Haiti, entire communities effectively contracted in size after the earthquake because so much rubble was covering land, reducing options and opportunities to shelter people. While moving rubble is neither quick nor easy, the cost of NOT removing rubble can be much greater. A debris management plan that could guide operations is an important tool for disaster recovery. However, most at-risk areas around the world lack such plans.

Disaster Risk Reduction activities should include emergency master planning as part of disaster response and recovery efforts. Land management is the Mother Lode of DRR precisely because it is a direct means of creating markets for land, housing, and other needed uses within a DRR-based framework.

Finally, we should take into account hosting, the shelter provided to disaster survivors by family and friends. This socially-defined, self-selected, and culturally-appropriate sheltering of people is often occurring before humanitarian actors arrive and — importantly – continues long after they leave. What is needed is basic support of various kinds to ensure that hosting doesn’t strain relations and pocketbooks. Not only has this form of assistance resulted in the provision of humanitarian shelter for thousands of disaster-affected families, it also often evolves into permanent housing solutions for those families — at a cost far below housing reconstruction efforts.

Please visit The Brookings Institution to view the panel discussion.