Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions
As the climate emergency escalates, the role of forests in carbon sequestration is paramount. This paper proposes a framework that integrates local capacities, multi-source remote sensing data, and meta-learning to enhance forest carbon assessment methodologies in data-scarce regions. By integrating...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/289 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587612842360832 |
|---|---|
| author | Kamaldeen Mohammed Daniel Kpienbaareh Jinfei Wang David Goldblum Isaac Luginaah Esther Lupafya Laifolo Dakishoni |
| author_facet | Kamaldeen Mohammed Daniel Kpienbaareh Jinfei Wang David Goldblum Isaac Luginaah Esther Lupafya Laifolo Dakishoni |
| author_sort | Kamaldeen Mohammed |
| collection | DOAJ |
| description | As the climate emergency escalates, the role of forests in carbon sequestration is paramount. This paper proposes a framework that integrates local capacities, multi-source remote sensing data, and meta-learning to enhance forest carbon assessment methodologies in data-scarce regions. By integrating multi-source optical and radar remote sensing data alongside community forest inventories, we applied a meta-modelling approach using stacked generalization ensemble to estimate forest above-ground carbon (AGC). We also conducted a Kruskal–Wallis test to determine significant differences in AGC among different tree species. The Kruskal–Wallis test (<i>p</i> = 1.37 × 10<sup>−13</sup>) and Dunn post-hoc analysis revealed significant differences in carbon stock potential among tree species, with <i>Afzelia quanzensis</i> (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mover accent="true"><mrow><mi>x</mi></mrow><mo>~</mo></mover><mtext> </mtext></mrow></semantics></math></inline-formula>= 12 kg/ha, P-holm-adj. = 0.05) and the locally known species <i>M’buta</i> (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mover accent="true"><mrow><mi>x</mi></mrow><mo>~</mo></mover></mrow></semantics></math></inline-formula> = 6 kg/ha, P-holm-adj. = 5.45 × 10<sup>−9</sup>) exhibiting a significantly higher median AGC. Our results further showed that combining optical and radar remote sensing data substantially improved prediction accuracy compared to single-source remote sensing data. To improve forest carbon assessment, we employed stacked generalization, combining multiple machine learning algorithms to leverage their complementary strengths and address individual limitations. This ensemble approach yielded more robust estimates than conventional methods. Notably, a stacking ensemble of support vector machines and random forest achieved the highest accuracy (R<sup>2</sup> = 0.84, RMSE = 1.36), followed by an ensemble of all base learners (R<sup>2</sup> = 0.83, RMSE = 1.39). Additionally, our results demonstrate that factors such as the diversity of base learners and the sensitivity of meta-leaners to optimization can influence stacking performance. |
| format | Article |
| id | doaj-art-fb0e35f7445e49de8d591b7d732d25ef |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-fb0e35f7445e49de8d591b7d732d25ef2025-01-24T13:48:01ZengMDPI AGRemote Sensing2072-42922025-01-0117228910.3390/rs17020289Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor RegionsKamaldeen Mohammed0Daniel Kpienbaareh1Jinfei Wang2David Goldblum3Isaac Luginaah4Esther Lupafya5Laifolo Dakishoni6Department of Geography and Environment, Western University, 1151 Richmond Street, London, ON N6A 3K7, CanadaDepartment of Geography, Geology and the Environment, Illinois State University, Normal, IL 61790-400, USADepartment of Geography and Environment, Western University, 1151 Richmond Street, London, ON N6A 3K7, CanadaDepartment of Geography and Tourism, Brock University, 1812 Sir Isaac Brock Way, St. Catharines, ON L2S 3A1, CanadaDepartment of Geography and Environment, Western University, 1151 Richmond Street, London, ON N6A 3K7, CanadaSoils, Food and Healthy Communities (SFHC), Ekwendeni P.O. Box 36, MalawiSoils, Food and Healthy Communities (SFHC), Ekwendeni P.O. Box 36, MalawiAs the climate emergency escalates, the role of forests in carbon sequestration is paramount. This paper proposes a framework that integrates local capacities, multi-source remote sensing data, and meta-learning to enhance forest carbon assessment methodologies in data-scarce regions. By integrating multi-source optical and radar remote sensing data alongside community forest inventories, we applied a meta-modelling approach using stacked generalization ensemble to estimate forest above-ground carbon (AGC). We also conducted a Kruskal–Wallis test to determine significant differences in AGC among different tree species. The Kruskal–Wallis test (<i>p</i> = 1.37 × 10<sup>−13</sup>) and Dunn post-hoc analysis revealed significant differences in carbon stock potential among tree species, with <i>Afzelia quanzensis</i> (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mover accent="true"><mrow><mi>x</mi></mrow><mo>~</mo></mover><mtext> </mtext></mrow></semantics></math></inline-formula>= 12 kg/ha, P-holm-adj. = 0.05) and the locally known species <i>M’buta</i> (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mover accent="true"><mrow><mi>x</mi></mrow><mo>~</mo></mover></mrow></semantics></math></inline-formula> = 6 kg/ha, P-holm-adj. = 5.45 × 10<sup>−9</sup>) exhibiting a significantly higher median AGC. Our results further showed that combining optical and radar remote sensing data substantially improved prediction accuracy compared to single-source remote sensing data. To improve forest carbon assessment, we employed stacked generalization, combining multiple machine learning algorithms to leverage their complementary strengths and address individual limitations. This ensemble approach yielded more robust estimates than conventional methods. Notably, a stacking ensemble of support vector machines and random forest achieved the highest accuracy (R<sup>2</sup> = 0.84, RMSE = 1.36), followed by an ensemble of all base learners (R<sup>2</sup> = 0.83, RMSE = 1.39). Additionally, our results demonstrate that factors such as the diversity of base learners and the sensitivity of meta-leaners to optimization can influence stacking performance.https://www.mdpi.com/2072-4292/17/2/289forestabove-ground carbonparticipatory GISremote sensingstacking |
| spellingShingle | Kamaldeen Mohammed Daniel Kpienbaareh Jinfei Wang David Goldblum Isaac Luginaah Esther Lupafya Laifolo Dakishoni Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions Remote Sensing forest above-ground carbon participatory GIS remote sensing stacking |
| title | Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions |
| title_full | Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions |
| title_fullStr | Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions |
| title_full_unstemmed | Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions |
| title_short | Synthesizing Local Capacities, Multi-Source Remote Sensing and Meta-Learning to Optimize Forest Carbon Assessment in Data-Poor Regions |
| title_sort | synthesizing local capacities multi source remote sensing and meta learning to optimize forest carbon assessment in data poor regions |
| topic | forest above-ground carbon participatory GIS remote sensing stacking |
| url | https://www.mdpi.com/2072-4292/17/2/289 |
| work_keys_str_mv | AT kamaldeenmohammed synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT danielkpienbaareh synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT jinfeiwang synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT davidgoldblum synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT isaacluginaah synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT estherlupafya synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions AT laifolodakishoni synthesizinglocalcapacitiesmultisourceremotesensingandmetalearningtooptimizeforestcarbonassessmentindatapoorregions |