Mapping drivers of tropical forest loss with satellite image time series and machine learning

19795 7 106507 2 323243 document 106507 Pišl_2024_Environ._Res._Lett._19_064053.pdf application/pdf 4740ecad4c2729a790b82cf700eeaab3 MD5 6327496 2024-06-10 07:28:04 https://pure.iiasa.ac.at/id/eprint/19795/1/Pi%C5%A1l_2024_Environ._Res._Lett._19_064053.pdf 19795 1 1 application/pdf text en public cc_by_4

Pišl_2024_Environ._Res._Lett._19_064053.pdf

published 106508 2 323260 document 106508 lightbox.jpg image/png ce1e0dde6f00e8bbd776470fea50df8d MD5 43195 2024-06-10 07:38:08 https://pure.iiasa.ac.at/id/eprint/19795/2/lightbox.jpg 19795 2 2 image/png other Thumbnails conversion from text to thumbnail_lightbox en public

lightbox.jpg

preview.jpg

medium.jpg

small.jpg

indexcodes.txt

http://eprints.org/relation/isVersionOf https://pure.iiasa.ac.at/id/document/106507 http://eprints.org/relation/isVolatileVersionOf https://pure.iiasa.ac.at/id/document/106507 http://eprints.org/relation/isIndexCodesVersionOf https://pure.iiasa.ac.at/id/document/106507 archive 353 disk0/00/01/97/95 2024-06-10 07:30:28 2024-10-17 14:11:09 2024-06-10 07:30:28 article show Pišl J. Rußwurm M. Haydn Hughes L. Lenczner G. See L. 8571 0000-0002-2665-7065 Dirk Wegner J. Tuia D. Mapping drivers of tropical forest loss with satellite image time series and machine learning pub ASA NODES deep learning; deforestation; earth observation; machine learning; remote sensing; time series; tropical forest The rates of tropical deforestation remain high, resulting in carbon emissions, biodiversity loss, and impacts on local communities. To design effective policies to tackle this, it is necessary to know what the drivers behind deforestation are. Since drivers vary in space and time, producing accurate spatially explicit maps with regular temporal updates is essential. Drivers can be recognized from satellite imagery but the scale of tropical deforestation makes it unfeasible to do so manually. Machine learning opens up possibilities for automating and scaling up this process. In this study, we developed and trained a deep learning model to classify the drivers of any forest loss—including deforestation—from satellite image time series. Our model architecture allows understanding of how the input time series is used to make a prediction, showing the model learns different patterns for recognizing each driver and highlighting the need for temporal data. We used our model to classify over 588 ′ 000 sites to produce a map detailing the drivers behind tropical forest loss. The results confirm that the majority of it is driven by agriculture, but also show significant regional differences. Such data is a crucial source of information to enable targeting specific drivers locally and can be updated in the future using free satellite data. 2024-06-01 published Institute of Physics Publishing (IOP) doi:10.1088/1748-9326/ad44b2 276 public Environmental Research Letters 19 6 e064053 TRUE 1748-9326 Geo-Wiki info:eu-repo/semantics/article