Mangrove ecosystems are increasingly recognized for the ecological services they provide. Over 35 percent of mangrove forest area has been lost in the past two decades, which exceeds the loss of tropical rain forests and coral reefs. In Mozambique, and in the Maputo bay in particular, mangroves have been affected by small-scale cutting, freshwater diversion and changes in water currents, which have lead to the degradation of large areas. To make informed choices for conservation and restoration of mangrove ecosystems, baseline data on forest composition, structure, biomass is needed. In this study we used field measures of forest composition, structure and environmental data to describe the current status of the mangroves in the Maputo Bay, Mozambique. We estimated mangrove structure and biomass using Shuttle Radar topography mission (SRTM) data and assessed the changes in forest area and distribution over ten years using Landsat images. Species zonation, height distributions and biomass were dependent on the distance to shore, freshwater input, but not on nutrient availability. Mangrove forest area increased in ten years, but the changes did not occur uniformly across the bay, with degraded mangrove areas spreading close to Maputo and healthy forest areas increasing on Inhaca Island and in the Incomati Estuary. We describe the best areas for mangrove restoration depending on the expected outcome. If the goal of the restoration is to colonize the largest area of lost or potential mangrove forest, then planting projects should be carried out in areas with the largest mangrove loss, such as in the Costa do sol and Espiritu Santo Estuary. If however, mangrove biomass, as opposed to area, is to be taken into account, we propose that areas with high biomass, such as Inhaca Island and the estuaries of the Inkomati and Espiritu Santo be prioritized for restoration projects.