Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2512.01917 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908684433489920 |
|---|---|
| author | Searcy, Jacob Dulal, Anish Bridgham, Scott Cordes, Ashley Aoki, Lillian Bohannan, Brendan Zhu, Qing Silva, Lucas C. R. |
| author_facet | Searcy, Jacob Dulal, Anish Bridgham, Scott Cordes, Ashley Aoki, Lillian Bohannan, Brendan Zhu, Qing Silva, Lucas C. R. |
| contents | Natural climate solutions (NCS) offer an approach to mitigating carbon dioxide (CO2) emissions. However, monitoring the carbon drawdown of ecosystems over large geographic areas remains challenging. Eddy-flux covariance towers provide ground truth for predictive 'upscaling' models derived from satellite products, but many satellites now produce measurements on spatial scales smaller than a flux tower's footprint. We introduce Footprint-Aware Regression (FAR), a first-of-its-kind, deep-learning framework that simultaneously predicts spatial footprints and pixel-level (30 m scale) estimates of carbon flux. FAR is trained on our AMERI-FAR25 dataset which combines 439 site years of tower data with corresponding Landsat scenes. Our model produces high-resolution predictions and achieves R2 = 0.78 when predicting monthly net ecosystem exchange on test sites from a variety of ecosystems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01917 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Footprint-Aware, High-Resolution Approach for Carbon Flux Prediction Across Diverse Ecosystems Searcy, Jacob Dulal, Anish Bridgham, Scott Cordes, Ashley Aoki, Lillian Bohannan, Brendan Zhu, Qing Silva, Lucas C. R. Machine Learning Natural climate solutions (NCS) offer an approach to mitigating carbon dioxide (CO2) emissions. However, monitoring the carbon drawdown of ecosystems over large geographic areas remains challenging. Eddy-flux covariance towers provide ground truth for predictive 'upscaling' models derived from satellite products, but many satellites now produce measurements on spatial scales smaller than a flux tower's footprint. We introduce Footprint-Aware Regression (FAR), a first-of-its-kind, deep-learning framework that simultaneously predicts spatial footprints and pixel-level (30 m scale) estimates of carbon flux. FAR is trained on our AMERI-FAR25 dataset which combines 439 site years of tower data with corresponding Landsat scenes. Our model produces high-resolution predictions and achieves R2 = 0.78 when predicting monthly net ecosystem exchange on test sites from a variety of ecosystems. |
| title | A Footprint-Aware, High-Resolution Approach for Carbon Flux Prediction Across Diverse Ecosystems |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2512.01917 |