Lobelia Earth S.L. has made an advancement in carbon cycle research with a novel method to more accurately measure carbon in scattered tree populations. This development, detailed in a study (DOI: 10.34133/remotesensing.0359) published on November 21, 2024, in Journal of Remote Sensing, promises to revolutionize land management and climate adaptation strategies, offering fresh hope for more effective carbon sequestration tracking.
The core innovation of this study lies in the development of an Artificial Neural Network (ANN) model trained on over 400 individual tree crowns, incorporating both spectral signatures and crown area extracted from Pléiades high-resolution satellite imagery. By doing so, the researchers achieved more precise above-ground carbon (AGC) estimates, delivering an R² of 0.66 and a relative RMSE of 78.6%. This method significantly reduces the biases seen in previous technologies, particularly those that underestimated carbon stocks in dryland regions.
To create this model, the researchers constructed a comprehensive AGC reference database from on-the-ground tree measurements, converting them into biomass using species-specific allometric equations. Through deep learning models, they were able to segment individual tree crowns and extract spectral information from very high-resolution (VHR) imagery, which was then used to train and validate the ANN model. The result was a highly accurate model, with a tree-level RMSE of just 373.85 kg, confirming its robustness in predicting AGC from remote sensing data.
Martí Perpinyana-Vallès, the study's lead author, says "By integrating field data with advanced Earth observation techniques, our study provides a reliable method for estimating biomass at multiple scales. This innovation holds the potential to significantly improve our understanding of carbon sequestration dynamics and enhance land management practices globally."
The study utilized Pléiades Neo satellite imagery, known for its exceptional 30cm native resolution, which enabled unprecedented precision in Earth observation. This precision, combined with deep learning algorithms for crown extraction and ANN models for AGC prediction, allowed for the accurate geolocation of individual trees—addressing longstanding limitations in carbon stock estimation.
Looking ahead, the future applications of this technology are vast. It promises to improve global carbon cycle assessments, optimize land use, and enhance reforestation initiatives. Furthermore, it could provide essential data for climate change mitigation strategies, helping policymakers address pressing environmental challenges. As the method gains wider adoption, it has the potential to harmonize carbon estimation discrepancies, offering invaluable support for international climate agreements and global sustainability efforts.
###
References
DOI
10.34133/remotesensing.0359
Original Source URL
https://doi.org/10.34133/remotesensing.0359
Funding information
This work was partly conducted in the Jeunesse Sahelienne pour l’Action Climatique (JESAC)563project which is funded by Intermon Oxfam Spain, and under Industrial PhD grants AGAUR (2021564DI 121) and DIN2020-010982 financed by MCIN AEI 10.13039/501100011033 and by the European565Union ”NextGenerationEU/ PRTR”. Aitor Ameztegui is funded by a Serra-H´unter fellowship from566Generalitat de Catalunya. This work was supported by ESA Network of Resources Initiative.
About The Journal of Remote Sensing
The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.