Accurate estimation of biomass and carbon stocks in mangrove forests is a prerequisite for a better understanding of their role in climate regulation. Allometric equations remain appropriate tools in this context but are lacking for many mangrove species and sites across Africa. In this study, we destructively sampled 68 individual trees of the two dominant West African mangrove species (Rhizophora racemosa and Avicennia germinans) to (i) determine species-specific height-diameter allometry; (ii) evaluate biomass allocation to stem, branches, leaves and roots components; (iii) establish species-specific aboveground-, belowground- and total biomass allometric equations; and (iv) examine the accuracy of our best biomass model against existing equations for mangroves. Diameter at breast height (Dbh), total height (H), wood density and crown diameter were used as predictors in the models.Results showed that Dbh explained 53% and 62% of height variation for R. racemosa and A. germinans, respectively. Stems stored the highest biomass fractions (84.30% for R. racemosa and 52.80% for A. germinans), followed by branches, while the belowground compartment contributed to 19%–22% of the total biomass. Among the candidate biomass models, the models incorporating Dbh and height as a compound variable (Dbh2H) were the most suitable for estimating aboveground and total biomass, with 87–92% of explained variance. For the root components, wood density and crown diameter were additionally found to improve model performance for R. racemosa and A. germinans, respectively. Our study revealed that biomass in West African mangrove forests was more accurately predicted using the established equations than with the existing models.