Mangrove aboveground carbon inventory (AGC) mapping is crucial for understanding carbon modifications in coastal ecosystems over time. AGC contains biomass from shrubs, bushes, and different aboveground elements components which might be extremely weak to human actions like aquaculture enlargement and land-use conversion. In response to Sasmito et al. (2019), modifications in mangrove land cowl equivalent to conversion to aquaculture ponds, oil palm plantations, or pasture can result in an 82% discount in mangrove AGC. Regardless that AGC contributes solely about 20% of whole ecosystem carbon in mangroves, it’s nonetheless a vital indicator. Why? As a result of the very first thing we see in mangrove is the “tree” itself and as soon as we lose that, there’s a scientific signal of breakdown on different carbon swimming pools.
Historically, AGC is measured by discipline sampling by calculating the diameter at breast peak (DBH), making use of allometric equations (like these from Chave or Kauffman), after which changing biomass to carbon utilizing a set conversion issue. Whereas efficient, this methodology is exhausting, time-consuming, and spatially restricted. Due to this fact, many researchers are shifting towards digitalization and automation of AGC monitoring an strategy aligned with international local weather efforts like REDD+ (Decreasing Emissions from Deforestation and Degradation). And it may be solely resolved with “Distant Sensing”-based measurement.
The primary recognized analysis to estimate mangrove AGC utilizing distant sensing got here from Simard et al. (2006), performed within the Everglades Nationwide Park, USA. They used SRTM (Shuttle Radar Topography Mission) and airborne LiDAR to estimate cover peak after which derived AGC from that. Whereas the accuracy was nonetheless restricted on the time (R² = 0.62 after conversion), their work was pioneering and foundational. After that analysis, Simard continued to co-author at the least 4 different main research in the identical discipline.
By 2013, using machine studying entered the ecosystem. A notable instance is Jachowski et al. (2013), who examined 18 machine studying algorithms utilizing WEKA, an open-source modeling toolkit. Their finest mannequin achieved a Root Imply Sq. Error (RMSE) of 25.01 Mg C ha⁻¹ with an R² of 0.66.
Within the 20s, analysis on AGC mangrove mapping gained extra well-liked, with Vietnam turning into as one of the vital productive nations on this discipline. Researchers more and more leveraged open-source satellite tv for pc knowledge (like Sentinel-2, SRTM, and Landsat) mixed with machine studying strategies. From my evaluate of latest papers on Scopus and Net of Science, it seems that machine studying fashions can scale back RMSE by 3–4 Mg C ha⁻¹ in comparison with conventional strategies. Nevertheless, utilizing just one machine studying mannequin can result in over- or underestimation, which is why ensemble modeling is gaining recognition.
Readers, Think about if sooner or later there’s a platform like Windy.com however for mangrove carbon. Stakeholders might visualize AGC predictions, mannequin comparisons, and real-time updates utilizing freely obtainable satellite tv for pc knowledge. Each day observations from SRTM or Sentinel-2 could possibly be validated with LiDAR or discipline knowledge on a seasonal or annual foundation. Even internet cameras or drone footage could possibly be built-in for ground-level insights. Past monitoring, this instrument might help Excessive Carbon Inventory (HCS) and Excessive Conservation Worth (HCV) mapping, making it helpful for each conservation and carbon offset functions. Finally, the imaginative and prescient is to construct an automatic AGC mapping instrument an utility the place customers merely enter fundamental parameters and obtain a mapped AGC estimate while not having to code or run advanced scripts. This could simplify the entry to AGC estimation and help policymakers in growing Forest Reference Emission Ranges (FREL) or Forest Reference Ranges (FRL) extra effectively.
AGC is likely to be only one a part of the mangrove ecosystem carbon pool, however it’s the most seen and weak. With the best mixture of distant sensing, machine studying, and digital platforms, we will revolutionize how we monitor and shield mangrove carbon and finally help international local weather resilience.
References
Jachowski NRA, Quak MSY, Friess DA, Duangnamon D, Webb EL, Ziegler AD. 2013. Mangrove biomass estimation in Southwest Thailand utilizing machine studying. Utilized Geography. 45:311-321.
Sasmito SD, Taillardat P, Clendenning JN, Cameron C, Friess DA, Murdiyarso D, utley . 2 19. e t o land‐use and land‐ over hange on mangrove blue carbon: A scientific evaluate. International Change Biology. 25(12):4291-4302.
Simard M, Zhang Okay, Rivera-Monroy VH, Ross MS, Ruiz PL, Castañeda-Moya E, Twilley RR, Rodriguez E. 2006. Mapping Peak and Biomass of Mangrove Forests in Everglades Nationwide Park with SRTM Elevation Knowledge. photogramm eng distant sensing. 72(3):299–311.
