Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
121 tokens/sec
GPT-4o
9 tokens/sec
Gemini 2.5 Pro Pro
47 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A community palm model (2405.09530v2)

Published 1 May 2024 in cs.CY, cs.CV, and cs.LG

Abstract: Palm oil production has been identified as one of the major drivers of deforestation for tropical countries. To meet supply chain objectives, commodity producers and other stakeholders need timely information of land cover dynamics in their supply shed. However, such data are difficult to obtain from suppliers who may lack digital geographic representations of their supply sheds and production locations. Here we present a "community model," a machine learning model trained on pooled data sourced from many different stakeholders, to produce a map of palm probability at global scale. An advantage of this method is the inclusion of varied inputs, the ability to easily update the model as new training data becomes available and run the model on any year that input imagery is available. Inclusion of diverse data sources into one probability map can help establish a shared understanding across stakeholders on the presence and absence of a land cover or commodity (in this case oil palm). The model predictors are annual composites built from publicly available satellite imagery provided by Sentinel-1, Sentinel-2, and ALOS-2, and terrain data from Jaxa (AW3D30) and Copernicus (GLO-30). We provide map outputs as the probability of palm in a given pixel, to reflect the uncertainty of the underlying state (palm or not palm). This version of this model provides global accuracy estimated to be 92% (at 0.5 probability threshold) on an independent test set. This model, and resulting oil palm probability map products are useful for accurately identifying the geographic footprint of palm cultivation. Used in conjunction with timely deforestation information, this palm model is useful for understanding the risk of continued oil palm plantation expansion in sensitive forest areas.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (37)
  1. Deforestation displaced: trade in forest-risk commodities and the prospects for a global forest transition. Environmental Research Letters, 14(5), 2019. doi:10.1088/1748-9326/ab0d41.
  2. Disentangling the numbers behind agriculture-driven tropical deforestation. Science, 377:eabm9267, 2022. doi:10.1126/science.abm9267.
  3. The impacts of oil palm on recent deforestation and biodiversity loss. PLoS ONE, 11(7):e0159668, 2016. doi:10.1371/journal.pone.0159668.
  4. European Commission. Regulation of the european parliament and of the council on the making available on the union market and the export from the union of certain commodities and products associated with deforestation and forest degradation and repealing regulation (eu) no 995/2010, 2023. URL https://data.consilium.europa.eu/doc/document/PE-82-2022-INIT/en/pdf. Viewed 20240228.
  5. USDA Foreign Agricultural Service. Commodity explorer, 2024. URL https://ipad.fas.usda.gov/cropexplorer/cropview/commodityView.aspx?cropid=4243000. Accessed 4/16/2024.
  6. Hannah Ritchie. Palm oil. Our World in Data, 2021. URL https://ourworldindata.org/palm-oil. Accessed 4/16/2024.
  7. FAO. Fra 2020 remote sensing survey: Vol. no. 186, 2022.
  8. What causes deforestation in indonesia? Environmental Research Letters, 14(2):024007, 2019. doi:10.1088/1748-9326/aaf6db.
  9. Slowing deforestation in indonesia follows declining oil palm expansion and lower oil prices. PLOS ONE, 17(3):e0266178, 2022. doi:10.1371/journal.pone.0266178.
  10. Biophysical and socioeconomic drivers of oil palm expansion in indonesia. Environmental Research Letters, 16(3):034048, 2021. doi:10.1088/1748-9326/abce83.
  11. More than meets the eye: Using sentinel-2 to map smallholder plantations in complex forest landscapes. Remote Sensing, 10(11):1693, 2018. doi:10.3390/rs10111693.
  12. Oil palm expansion and deforestation in southwest cameroon associated with proliferation of informal mills. Nature Communications, 10(1):114, 2019. doi:10.1038/s41467-018-07915-2.
  13. Oil palm (elaeis guineensis) mapping with details: Smallholder versus industrial plantations and their extent in riau, sumatra. Remote Sensing, 11(21):2590, 2019. doi:10.3390/rs11212590.
  14. Combining radar and optical imagery to map oil palm plantations in sumatra, indonesia, using google earth engine. Remote Sensing, 12(7):1220, 2020. doi:10.3390/rs12071220.
  15. Annual oil palm plantation maps in malaysia and indonesia from 2001 to 2016. Earth System Science Data, 12(2):847–867, 2020. doi:10.5194/essd-12-847-2020.
  16. Differentiating oil palm plantations from natural forest to improve land cover mapping in ghana. Remote Sensing Applications: Society and Environment, 30, 2023. doi:10.1016/j.rsase.2023.100968.
  17. Deep learning based oil palm tree detection and counting for high-resolution remote sensing images. Remote Sensing, 9(1):22, 2017. doi:10.3390/rs9010022.
  18. High-resolution global map of smallholder and industrial closed-canopy oil palm plantations. Earth System Science Data, 13:1211–1231, 2021. doi:10.5194/essd-13-1211-2021.
  19. A map of the extent and year of detection of oil palm plantations in indonesia, malaysia and thailand. Scientific Data, 8(1):96, 2021. doi:10.1038/s41597-021-00867-1.
  20. Rise and fall of forest loss and industrial plantations in borneo (2000–2017). Conservation Letters, 12(3):e12622, 2018. doi:10.1111/conl.12622.
  21. Tensorflow: Large-scale machine learning on heterogeneous systems, 2015. URL https://zenodo.org/records/10798587. Software available from tensorflow.org.
  22. A global palm oil map for the year 2017 using multi-sensor sar imagery, 2019. URL https://nikal.eventsair.com/NikalWebsitePortal/living-planet-symposium-2019/lps19/Agenda/AgendaItemDetail?id=d5c77654-1c5e-45af-92b6-cebdcde7eca4. Living Planet Symposium. ESA/ESRIN, Frascati, Italy.
  23. Comprehensive quality assessment of optical satellite imagery using weakly supervised video learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pages 2125–2135, 2023.
  24. Angular-based radiometric slope correction for sentinel-1 on google earth engine. Remote Sensing, 12(11):1867, 2020. doi:10.3390/rs12111867.
  25. Initial validation of the 30 m-mesh global digital surface model generated by alos prism. In The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, ISPRS, volume XLI-B4, pages 157–162, 2016. URL https://isprs-archives.copernicus.org/articles/XLI-B4/157/2016/isprs-archives-XLI-B4-157-2016.pdf.
  26. Validation of ’aw3d’ global dsm generated from alos prism. In ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, volume III-4, pages 25–31, 2016. URL https://isprs-annals.copernicus.org/articles/III-4/25/2016/isprs-annals-III-4-25-2016.pdf.
  27. An ecoregion-based approach to protecting half the terrestrial realm. BioScience, 67(6):534–545, 2017. doi:10.1093/biosci/bix014.
  28. Using landsat and nighttime lights for supervised pixel-based image classification of urban land cover. Remote Sensing of Environment, 205:253–275, 2018. doi:10.1016/j.rse.2017.11.026.
  29. Eeager: A neural network model for finding beaver complexes in satellite and aerial imagery. Journal of Geophysical Research: Biogeosciences, 128:e2022JG007196, 2023. doi:10.1029/2022JG007196.
  30. A leisurely look at the bootstrap, the jackknife, and cross-validation. The American Statistician, 37(1):36–48, 1983. doi:10.2307/2685844.
  31. Tom Fawcett. An introduction to roc analysis. Pattern Recognition Letters, 27(8):861–874, 2006. ISSN 0167-8655. doi:10.1016/j.patrec.2005.10.010.
  32. Towards a common validation sample set for global land-cover mapping. International Journal of Remote Sensing, 35(13):4795–4814, 2014. doi:10.1080/01431161.2014.930202.
  33. Global map of forest cover 2020 - version 1, 2023. URL http://data.europa.eu/89h/10d1b337-b7d1-4938-a048-686c8185b290. [Dataset].
  34. A family of bivariate distributions generated by the bivariate bernoulli distribution. Journal of the American Statistical Association, 80(390):332–338, 1985. doi:10.2307/2287890.
  35. Multivariate bernoulli distribution. Bernoulli, 19(4):1465–1483, 2013. URL http://www.jstor.org/stable/23525760.
  36. Nobuyuki Otsu. A threshold selection method from gray-level histograms. IEEE Transactions on Systems, Man, and Cybernetics, SMC-9(1), January 1979. URL link.
  37. Tile2vec: Unsupervised representation learning for spatially distributed data. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 3967–3974, 2019. doi:10.1609/aaai.v33i01.33013967.

Summary

We haven't generated a summary for this paper yet.