Mangrove forests are known worldwide for the natural splendour and thriving biodiversity. These unique ecosystems provide critical functions for the coastal environments which they surround. Apart from their role in helping to maintain biodiversity they also provide stability and protective services against coastal erosion and storm surges (Menendez et al., 2020), have high capacity for carbon sequestration (Lovelock & Duarte, 2019). Moreover, they support livelihoods, ocean health and biodiversity (Dasgupta et al., 2022; Mallick et al., 2021,). Despite these critical roles, it is estimated that since the 1950s about 35% of global mangrove habitats have either been removed and degraded (Valiela et al., 2001). This trend is expected to continue due to anthropogenic influences and climate change (Akram et al., 2023).

Despite the many studies which have shown a global decline in the rate of mangrove loss (Bunting et al., 2022; Murray et al., 2022; Giri et al., 2011; Spalding et al., 2010), IUCN (2024) data indicates that half of all mangroves are still under threat. The United Nations Framework Convention on Climate Change (UNFCCC), the Convention on Biological Diversity (CBD) and the Ramsar Convention on Wetlands are among a number of international conventions that all advocate for protecting and preserving mangroves.

Earth observation (EO) for mangrove mapping

Detailed and up-to-date maps of mangrove distribution and extent can provide key information for measuring changes, particularly with respect to mangrove loss (Bunting et al., 2023; Giri, 2021). This can help identify hotspot areas and sites for focusing preservation and conservation efforts. Mangrove extent maps can also be used in conjunction with other key layers such as mangrove height, species type, condition and biomass to estimate carbon stocks (Hidayah et al., 2023; Hu et al., 2020). With recent advances in space technologies, Earth observation data is becoming more affordable and ubiquitous.

Challenges for small islands

Mangrove mapping in small island states comes with its own set of limitations and challenges. Limited supporting infrastructure, unpredictable weather conditions, cloud cover and funding availability are just a few. Mangrove forests on small islands are usually smaller in extent and typically have high spatial heterogeneity. Individual mangrove species are often highly mixed with each other or with other non-mangrove vegetation types. These conditions often lead to spectral contamination or mixed spectral returns and reduce the accuracy of mangrove delineation and classification products (Pham et al., 2019). As a result, the use of low and medium spatial resolution satellite imagery is not optimal at small island scales. Tropical and sub-tropical regions consistently experience cloudy skies, a phenomenon which is not conducive for optical satellite imaging. Small islands are also characterized by very dynamic tidal regimes, which can effectively influence the spectral characteristics and the resulting appearance of mangroves in satellite Earth observation data (Xia et al., 2018).

Furthermore, many small island nations are financially and technically challenged when it comes to accessing and using Earth observation technologies. This can be a severely limiting factor for the planning and implementation of mapping projects. In recent times, lower-income nations have been able to access space-based data more readily, helping them to manage their resources. Small island developing states (SIDS) typically require data at much smaller scales and finer resolutions, which have traditionally been out of reach due to the high costs. With the vast array of open-source imagery currently available, satellite imagery has been increasingly integrated into national mapping and monitoring frameworks (Avalon-Cullen et al., 2023; Soanes et al., 2021; Cherrington et al., 2020). The case study described in this article involves using high-resolution EO imagery for mapping mangrove species distribution and extent. It further demonstrates the capability of utilizing high-resolution data for mapping small-scale Earth features at higher accuracies.

Mangrove species mapping

Maps showing the distribution and extent of key mangrove species can be critical for mitigating the heightened threats from climate change, coastal development and other human activities (Sunkar et al., 2024; Twomey & Lovelock, 2024). These maps provide key metrics useful for understanding ecosystem dynamics and planning conservation strategies. Species-specific maps also provide information on the ecosystem response to stressors such as water pollution or salinity changes that may be caused by sea level rise (Zhao et al., 2020). They can further help to inform planning and adaptation towards the development of longer-term survival strategies. Figure 3 shows mangrove species maps for the mangrove forests of the Caroni Swamp region on the Caribbean Island of Trinidad.

These seasonal maps were generated by integrating optical data from Sentinel-2 imagery with radar backscatter data from Sentinel-1 (SAR) imagery and tree height data from a global high resolution Canopy Height Model (CHM) produced by the World Resources Institute (WRI) and Meta. Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Green Normalized Difference Vegetation Index (GNDVI) were also utilized to improve the detection of the two main mangrove species. Figure 4 is an illustration of the overall workflow for the methodology described above. A full description of the methodology is provided in (Ramsewak and Jagassar, 2024). An overall accuracy of 98.95 per cent was achieved for the combined optical-radar-CHM classified product when tested using an error matrix and independent GPS ground data.

Discussion

Increased availability of EO data due to recent technological advances has improved the capabilities for mapping mangrove forests, species and other dynamics at smaller scales. Data from Planet and Sentinel satellite constellations have been crucial for small island states, particularly due to various open-source programs that made the data available free of charge. Since mangrove species typically occur in small patches interspersed amongst each other as well as other types of vegetation and land cover types (Yang et al., 2022), higher resolution is especially useful for small-scale mapping on islands. Planned satellite launches later in 2025, such as the Copernicus Expansion Program (ESA, 2024), NASA’s Surface Biology and Geology (SBG) (Thompson et al., 2022) and NASA-ISRO Synthetic Aperture Radar (NISAR) missions will further enhance capabilities for mapping on small islands (NASA, 2024).

Conclusion

To conclude, advances in Earth observation technologies will continue to improve mapping products, predictive models and conservation frameworks to protect our terrestrial and marine ecosystems. High spatial and spectral resolution satellite imagery is becoming more affordable and accessible to lower-income small island nations. Combined with a planned increase in open-source radar data, these datasets can be used to effectively capture detailed ecosystem metrics such as mangrove species at higher accuracies, enabling more dynamic and timely monitoring of these critical ecosystems.