Harnessing Deep Learning to Explore Microbial Community Structure and Carbon Storage Capacity in Mangrove Ecosystems: A Framework for Computationally

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2026 | Volume : 15 | 01 | Page :
    By

    Taranjeet Singh,

  • Manuraj modgil,

  1. Student, Department of computer science and engineering) BGIET, Sangrur, Punjab, India
  2. head of department, Department of computer science and engineering, BGIET, Sangrur, Punjab, India

Abstract

Mangrove ecosystems represent one of the most efficient natural carbon sinks on Earth, functioning as critical blue carbon habitats that sustain diverse microbial communities responsible for biogeochemical cycling and long-term carbon storage. Despite their global ecological significance, accurately quantifying and predicting carbon sequestration in mangrove systems remains challenging due to the complex interactions between microbial diversity, sediment chemistry, and environmental drivers. This study presents a comprehensive and sustainable artificial intelligence framework that integrates metagenomic sequencing, remote sensing data,
environmental metadata, and multimodal deep learning models to predict carbon sequestration potential in mangrove ecosystems. Specifically, we combine Convolutional Neural Networks (CNNs) for extracting spatial features from Sentinel-2 and Landsat imagery, Recurrent Neural Networks (RNNs) and Long Short-Term
Memory (LSTM) architectures for modeling microbial abundance trajectories and temporal carbon flux patterns, and an ensemble Multi-Layer Perceptron (MLP) for final carbon stock prediction. Microbial community data are processed using QIIME 2 workflows, including DADA2 denoising and SILVA-based taxonomic assignment, while environmental parameters such as salinity, pH, and soil depth are normalized and incorporated as predictive variables. The proposed hybrid architecture enables mechanistic integration of ecological, microbial, and spatial information, improving interpretability compared to traditional biomass-only models. Simulated experimental results demonstrate improved prediction stability and reduced generalization error using the ensemble approach. Furthermore, the study adopts Green AI principles by optimizing model size, leveraging transfer learning, and reporting computational resource usage to minimize carbon footprint. This integrative framework offers a scalable pathway for climate-resilient monitoring of mangrove blue carbon systems and supports sustainable AI development aligned with global climate mitigation goals.

Keywords: Mangrove Microbiome; Carbon Sequestration; Deep Learning; CNN; RNN; Sustainable AI; QIIME2

How to cite this article:
Taranjeet Singh, Manuraj modgil. Harnessing Deep Learning to Explore Microbial Community Structure and Carbon Storage Capacity in Mangrove Ecosystems: A Framework for Computationally. Research and Reviews : Journal of Computational Biology. 2026; 15(01):-.
How to cite this URL:
Taranjeet Singh, Manuraj modgil. Harnessing Deep Learning to Explore Microbial Community Structure and Carbon Storage Capacity in Mangrove Ecosystems: A Framework for Computationally. Research and Reviews : Journal of Computational Biology. 2026; 15(01):-. Available from: https://journals.stmjournals.com/rrjocb/article=2026/view=239297


References

1. Andreote, F.D., Jiménez, D.J., Chaves, D., Dias, A.C.F., Luvizotto, D.M., Dini-Andreote, F., et al. (2012). The Microbiome of Brazilian Mangrove Sediments as Revealed by Metagenomics. PLoS ONE 7(6): e38600. doi:10.1371/journal.pone.0038600. [turn0search0]
2. Holguin, G., Vázquez, P., & Bashan, Y. (2001). The role of sediment microorganisms in the productivity, conservation, and rehabilitation of mangrove ecosystems: an overview. Biology and Fertility of Soils, 33(4), 265–278. doi:10.1007/s003740000319. [turn0search1]
3. Alongi, D.M. (2012). Carbon sequestration in mangrove forests. Coral Reefs (or related review). doi:10.4155/cmt.12.20. [turn0search3]
4. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436–444. doi:10.1038/nature14539. [turn0search4]
5. Bolyen, E., et al. (2019). Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nature Biotechnology. doi:10.1038/s41587-019-0209-9. [turn0search5]
6. Schwartz, R., Dodge, J., Smith, N.A., & Etzioni, O. (2019). Green AI. arXiv:1907.10597. [turn0search14]
7. Li, Y., et al. (2019). Comparative metagenomics study reveals pollution impacts on mangrove sediments. Sci Rep. doi:10.1038/s41598-019-42260-4. [turn1search1]
8. Wang, L., et al. (2019). A review of remote sensing for mangrove forests: 1956–2018. Science of the Total Environment. [turn1search6]
9. Yuan, Q., et al. (2020). Deep learning in environmental remote sensing. (Review). [turn1search15]
10. Sasmito, S.D., et al. (2020). Mangrove blue carbon stocks and dynamics are controlled by hydrogeomorphic settings and land-use change. (Ecosphere).[turn1search20]
11. Kauffman, J.B., et al. (2020). Total ecosystem carbon stocks of mangroves across broad latitudinal gradients. (Ecosphere). [turn1search22]
12. Qian, L., et al. (2023). Vertically stratified methane, nitrogen and sulphur cycling in mangrove sediments. Microbiome. [turn1search25] 13. Zhang, Z.F., et al. (2023). Long-read assembled metagenomic approaches improve understanding of mangrove microbiomes. Microbiome. [turn1search13]
14. Rondon, M., et al. (2023). Remote sensing-based assessment of mangrove distribution and cover. Frontiers in Marine Science. [turn1search26]
15. Cheng, X., et al. (2023). Application of Deep Learning in Multitemporal Remote Sensing. MDPI Remote Sensing. [turn1search19]
16. Paheding, S., et al. (2024). Advancing horizons in remote sensing. (Springer). [turn1search27]
17. Choudhary, B. (2024). Blue carbon and the role of mangroves in carbon sequestration:review. [turn1search0]
18. Sasmito, S.D. (2020). Mangrove blue carbon stocks and dynamics. [turn1search20]
19. Puthusseri, R.M. (2021). Insights into response of mangrove sediment microbes to contamination. [turn1search9]
20. Qu, W., et al. (2018). Metagenomics investigation of agarlytic genes in mangroves. Frontiers in Microbiology. [turn1search21]
21. Gu, X., et al. (2023). Changes in mangrove blue carbon under elevated CO2. [turn1search24]
22. Zhang, R. (2025). Classification and Carbon-Stock Estimation of Mangroves using Remote Sensing. Remote Sensing. [turn1search18]
23. Sunkur, R., et al. (2024). Mangrove mapping and monitoring using remote sensing. Sci Rep. [turn1search14]
24. PNAS 2024 deep learning plant species mapping (PNAS article). [turn1search23]
25. Estaki, M., et al. (2020). QIIME 2 methodology overview. PubMed. [turn0search21]

Ahead of Print Subscription Original Research
Volume 15
01
Received 10/02/2026
Accepted 27/03/2026
Published 27/03/2026
Publication Time 45 Days
79

Login


My IP

PlumX Metrics