Mangrove forests are the dominant vegetation growing on low wooded islands, which occur in the Caribbean, Indian and Pacific Oceans. In the northern Great Barrier Reef, we map remarkable, undocumented mangrove forest extension on ten low wooded islands in the Howick Group that collectively equates to an area of 667, 000 m^2 (66.7 ha). We combine extensive field survey with canopy height models derived from RPA imagery and allometric scaling to quantify above ground biomass in both old (pre-1973) and new (post-1973) forest areas. Forest expansion added ~ 10,233 tonnes of new biomass since the early 1970s. Such substantial expansion of mangrove forest has occurred within a short time span in response to changing environmental controls. These include sea-level rise, sediment transport and deposition, cyclone impact and the development of associated reef flat sedimentary landforms, including unconsolidated and lithified shingle ridges, which influence reef flat hydrodynamics. Our observations highlight the globally dynamic response of mangrove distribution and forest structure to environmental change and provide timely new estimates from understudied reef island settings.

Rapid in-situ mangrove community surveys were conducted at seven of the ten reef sites visited in the Howick Islands from the 1^st – 11^th June 2021 (Table S1). Surveys collected information on species composition, structure and health status of individual trees; and plot-scale vegetation metrics for development of setting-specific biomass scaling models. Fifteen shore-normal belt transect survey sites were planned using the 1973 mangrove forest maps combined with recent satellite imagery to ensure a diverse range of mangrove age classes (i.e. pre- and post- 1973 establishment) and species assemblages were assessed.

Transect start points were located using a real-time Kinematic GPS with satellite RTX correction (Trimble R10; Trimble Inc. USA) with expected horizontal and vertical uncertainties of ~2cm and ~5cm respectively. Transect orientation was measured with a hand-held compass. See supplementary Figure S1 for the location of in-situ mangrove biomass surveys.

Transects ranged in length from 54 – 343 m, depending on the forest width, site accessibility and time constraints (typically 5-6 hours per island). All trees encountered within a 2 m width of each transect that were ≥ 2 m height (i.e. 1 m either side of a central line marked by measuring tape) were measured for vertical height, maximum branch length (i.e. inclusive of lateral biomass growth), and tree diameter at breast height (cm; DBH). Tree height and branch length measurements were taken to the nearest 0.1 m using a 12 m telescopic measurement pole (Senshin Industry, Japan). Scrub mangroves (1-2 m tall) were measured for height (to nearest 0.01 m), and basal area diameter measured at 0.3 m height above ground. Mangrove seedlings or juveniles <1 m tall were were excluded from the survey since their biomass was low. Species identification and status (healthy; stressed; dead) were recorded for each individual.