What can we conclude from this exercise?

The initial assumption that these pilot projects could draw/relay on the availability and applicability of UN derived urban indicators to develop a GIS for monitoring and analyzing urban indicators at the sub-city level was too ambitious and unrealistic for two reasons:

1. UN urban indicators, though providing a useful starting point, were intended for country comparisons rather than sub-city level analysis. Since urban problems are contextually based and solutions have to be developed locally, sub-city level analysis of indicators requires the generation of data that is situation specific. This is an important lesson from the Beira experience.

2. Unlike cities in other countries, data on key UN urban indicators were found to be either incomplete or non-existent. This was not surprising given that Mozambique only recently emerged from a long protracted civil war, and therefore faces major challenges on many fronts including reconstituting critical databases for urban policy analysis and institutional capacity building.

The conditions of a poor data environment presented both constraints and opportunities. Specifically, the need to generate new data forced an early decision concerning the choice of indicators and the methods of analysis.In this respect, the training modules developed both reflect the desire and need to familiarize the userwith basic techniques for selecting appropriate indicators, data gathering methods, mapping as well as techniques for integrating disparate data sources to analyze and monitor urban indicators.

Though the methods and spatial techniques applied in this project were dictated by the unique set of circumstances that exist in Beira, they offer useful lessons for other areas characterized by a poor data environment or where local authorities have limited capacity to monitor growth and change. For example, the emerging colonias settlements along the US-Mexican border are a case in point. The Office of Housing and Urban Development defines colonias as "rural communities and neighborhoods located within 150 miles of the US-Mexican border" (http://www.hud.gov ). Their conditions mirror those of informal settlements on the urban fringe in developing countries in that they lack adequate infrastructure and other basic facilities or develop spontaneously without jurisdictional sanction and viable livelihood systems.

For regional and local authorities attempting to respond to this growth phenomenon, the various methods used in the Beira context offer possibilities. Remotely sensed data (air photography and satellite imagery), can be used to estimate the pace of colonias growth as well as to assess the threat of encroachment to ecologically sensitive areas and farm land. Second, an integration of remotely sensed data and resultant GIS coverages can be used both as a predictive and planning tool for infrastructure improvement, assessing housing conditions, and demand estimation of related basic facilities, for example: water access, electricity, sanitation, schools and recreation.

Project funding provided by the US Department of Housing and Urban Development (HUD through the University Consortium on Geographic Information Science (UCGIS) with software support from the Environmental Systems research institute (ESRI).