In tropical developing countries, reducing emissions from deforestation and forest degradation (REDD+) is becoming an important mechanism for conserving forests and protecting biodiversity. A key prerequisite for any successful REDD+ project, however, is obtaining baseline estimates of carbon in forest ecosystems. Using available published data, we provide here a new and more reliable estimate of carbon in Bangladesh forest ecosystems, along with their geo-spatial distribution. Our study reveals great variability in carbon density in different forests and higher carbon stock in the mangrove ecosystems, followed by in hill forests and in inland Sal (Shorea robusta) forests in the country. Due to its coverage, degraded nature, and diverse stakeholder engagement, the hill forests of Bangladesh can be used to obtain maximum REDD+ benefits. Further research on carbon and biodiversity in under-represented forest ecosystems using a commonly accepted protocol is essential for the establishment of successful REDD+ projects and for the protection of the country’s degraded forests and for addressing declining levels of biodiversity.