Spectral analysis for monitoring mangrove restoration: A case study in the Vietnamese Southern Coastline

Mangrove restoration efforts have been ongoing, but with varying levels of success, requiring spatial and temporal monitoring to better understand the stocks and drivers of success. Here, we used multi-spectral remote sensing and spatial regression techniques to examine mangrove distribution and restoration potential in the Vietnamese Southern Coastal (VSC) region from 1988 to 2023, an area where multiple episodes of mangrove restoration have been attempted over the past decades. Our results show that 51.5% of the mangrove area has recovered from previous losses, while 48.5% has been lost during the 1988-2023 period. Significant gains were observed between 2018 and 2023, accounting for 77.8% of the total restoration. However, over 40,000 ha of mangroves were lost during each decade between 1988 and 2018, primarily due to land-use changes. Regression analyses estimated a sustainable mangrove cover increase of 9.9% (23,407 ha) and persistence of 22.5% (52,936 ha), mainly in protected areas and low-impact zones. Conversely, 9.8% (23,056 ha) of mangroves in erosion-prone and human-disturbed regions face continued decline. Our study demonstrated the effectiveness of integrating long-term Normalised Difference Vegetation Index time-series analysis with spatial regression to monitor mangrove ecosystems. These techniques offered a scalable framework for global mangrove monitoring and restoration planning, supporting evidence-based conservation policies.