A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.

Accurate and spatially-explicit maps of tropical forest carbon stocks are needed to implement carbon offset mechanisms such as REDD+ (Reduced Deforestation and Degradation Plus). The Random Forest machine learning algorithm may aid carbon mapping applications using remotely-sensed data. However, Ran...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Joseph Mascaro, Gregory P Asner, David E Knapp, Ty Kennedy-Bowdoin, Roberta E Martin, Christopher Anderson, Mark Higgins, K Dana Chadwick
Formato: article
Lenguaje:EN
Publicado: Public Library of Science (PLoS) 2014
Materias:
R
Q
Acceso en línea:https://doaj.org/article/08103648c6a04a1eac14471b6a484809
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:08103648c6a04a1eac14471b6a484809
record_format dspace
spelling oai:doaj.org-article:08103648c6a04a1eac14471b6a4848092021-11-18T08:35:26ZA tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.1932-620310.1371/journal.pone.0085993https://doaj.org/article/08103648c6a04a1eac14471b6a4848092014-01-01T00:00:00Zhttps://www.ncbi.nlm.nih.gov/pmc/articles/pmid/24489686/?tool=EBIhttps://doaj.org/toc/1932-6203Accurate and spatially-explicit maps of tropical forest carbon stocks are needed to implement carbon offset mechanisms such as REDD+ (Reduced Deforestation and Degradation Plus). The Random Forest machine learning algorithm may aid carbon mapping applications using remotely-sensed data. However, Random Forest has never been compared to traditional and potentially more reliable techniques such as regionally stratified sampling and upscaling, and it has rarely been employed with spatial data. Here, we evaluated the performance of Random Forest in upscaling airborne LiDAR (Light Detection and Ranging)-based carbon estimates compared to the stratification approach over a 16-million hectare focal area of the Western Amazon. We considered two runs of Random Forest, both with and without spatial contextual modeling by including--in the latter case--x, and y position directly in the model. In each case, we set aside 8 million hectares (i.e., half of the focal area) for validation; this rigorous test of Random Forest went above and beyond the internal validation normally compiled by the algorithm (i.e., called "out-of-bag"), which proved insufficient for this spatial application. In this heterogeneous region of Northern Peru, the model with spatial context was the best preforming run of Random Forest, and explained 59% of LiDAR-based carbon estimates within the validation area, compared to 37% for stratification or 43% by Random Forest without spatial context. With the 60% improvement in explained variation, RMSE against validation LiDAR samples improved from 33 to 26 Mg C ha(-1) when using Random Forest with spatial context. Our results suggest that spatial context should be considered when using Random Forest, and that doing so may result in substantially improved carbon stock modeling for purposes of climate change mitigation.Joseph MascaroGregory P AsnerDavid E KnappTy Kennedy-BowdoinRoberta E MartinChristopher AndersonMark HigginsK Dana ChadwickPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 9, Iss 1, p e85993 (2014)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Joseph Mascaro
Gregory P Asner
David E Knapp
Ty Kennedy-Bowdoin
Roberta E Martin
Christopher Anderson
Mark Higgins
K Dana Chadwick
A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
description Accurate and spatially-explicit maps of tropical forest carbon stocks are needed to implement carbon offset mechanisms such as REDD+ (Reduced Deforestation and Degradation Plus). The Random Forest machine learning algorithm may aid carbon mapping applications using remotely-sensed data. However, Random Forest has never been compared to traditional and potentially more reliable techniques such as regionally stratified sampling and upscaling, and it has rarely been employed with spatial data. Here, we evaluated the performance of Random Forest in upscaling airborne LiDAR (Light Detection and Ranging)-based carbon estimates compared to the stratification approach over a 16-million hectare focal area of the Western Amazon. We considered two runs of Random Forest, both with and without spatial contextual modeling by including--in the latter case--x, and y position directly in the model. In each case, we set aside 8 million hectares (i.e., half of the focal area) for validation; this rigorous test of Random Forest went above and beyond the internal validation normally compiled by the algorithm (i.e., called "out-of-bag"), which proved insufficient for this spatial application. In this heterogeneous region of Northern Peru, the model with spatial context was the best preforming run of Random Forest, and explained 59% of LiDAR-based carbon estimates within the validation area, compared to 37% for stratification or 43% by Random Forest without spatial context. With the 60% improvement in explained variation, RMSE against validation LiDAR samples improved from 33 to 26 Mg C ha(-1) when using Random Forest with spatial context. Our results suggest that spatial context should be considered when using Random Forest, and that doing so may result in substantially improved carbon stock modeling for purposes of climate change mitigation.
format article
author Joseph Mascaro
Gregory P Asner
David E Knapp
Ty Kennedy-Bowdoin
Roberta E Martin
Christopher Anderson
Mark Higgins
K Dana Chadwick
author_facet Joseph Mascaro
Gregory P Asner
David E Knapp
Ty Kennedy-Bowdoin
Roberta E Martin
Christopher Anderson
Mark Higgins
K Dana Chadwick
author_sort Joseph Mascaro
title A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
title_short A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
title_full A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
title_fullStr A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
title_full_unstemmed A tale of two "forests": random forest machine learning AIDS tropical forest carbon mapping.
title_sort tale of two "forests": random forest machine learning aids tropical forest carbon mapping.
publisher Public Library of Science (PLoS)
publishDate 2014
url https://doaj.org/article/08103648c6a04a1eac14471b6a484809
work_keys_str_mv AT josephmascaro ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT gregorypasner ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT davideknapp ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT tykennedybowdoin ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT robertaemartin ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT christopheranderson ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT markhiggins ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT kdanachadwick ataleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT josephmascaro taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT gregorypasner taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT davideknapp taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT tykennedybowdoin taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT robertaemartin taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT christopheranderson taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT markhiggins taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
AT kdanachadwick taleoftwoforestsrandomforestmachinelearningaidstropicalforestcarbonmapping
_version_ 1718421542987104256