The paper is based on an on-going 3-year study in the wetland communities of Kampala. The study uses participatory methods and aims to contribute to (i) the development of low-income wetland communities, (ii) to prepare these communities to become less dependent on wetlands without receding into poverty, and (iii) the better management of the wetlands. The communities in direct dependence and intimate interaction with Nakivubo wetlands are mainly poor, live and work under hazardous conditions, and their activities pose a threat to the ecological function of the wetlands. Yet these wetlands are important for filtering the city’s waste and storm water before it flows into Lake Victoria’s Murchison Bay, which is Kampala’s source of piped water. Government approaches to the problem of wetland encroachment have largely failed because they are confrontational, and are not consistent or participatory. The study has in the first year conducted a series of activities including stakeholder analysis, resource analysis, livelihood analysis, a questionnaire survey and action planning. Preliminary data show that wetland dependency is very high among the poor nearby communities. They practice cultivation, brick-making and harvesting of wetland vegetation. However, these activities are under threat because wetland resources are dwindling due to increasing population and over-use. Livelihoods are threatened not only by the decreasing productivity of the wetland, but also by the ever-present government threat to evict wetland encroachers to restore its ecology. The study therefore works with communities to prepare for less dependence on wetlands so that they do not suddenly recede into worse poverty if they are evicted. They formulate strategies to enhance alternative livelihood, and for management of the wetland. Action plans have been formulated to address the situation through a newly created association.