A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau

Accurate assessments of grassland above-ground biomass (AGB) are crucial for the sustainable utilization and protection of grassland resources and the eco-environment. In this study, a random forest (RF) model combined with the forward feature selection (FFS) and leave-location-out cross-validation...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hui Yu, Yufeng Wu, Liting Niu, Yafan Chai, Qisheng Feng, Wei Wang, Tiangang Liang
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	Alpine grassland Above-ground biomass Random forest Cross-validation Overfitting Ecology QH540-549.5
Acceso en línea:	https://doaj.org/article/d1a70b3efee3483494c24b3188492737
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:d1a70b3efee3483494c24b3188492737
record_format	dspace
spelling	oai:doaj.org-article:d1a70b3efee3483494c24b31884927372021-12-01T04:46:30ZA method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau1470-160X10.1016/j.ecolind.2021.107450https://doaj.org/article/d1a70b3efee3483494c24b31884927372021-06-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S1470160X21001151https://doaj.org/toc/1470-160XAccurate assessments of grassland above-ground biomass (AGB) are crucial for the sustainable utilization and protection of grassland resources and the eco-environment. In this study, a random forest (RF) model combined with the forward feature selection (FFS) and leave-location-out cross-validation (LLO-CV) methods was trained to predict the dry weight (DW) of grassland AGB based on multiple factors. The final model exhibited a performance of R2 = 0.66, root mean square error (RMSE) of 503.86 kg DW/ha and mean absolute error (MAE) of 376.51 kg DW/ha. The spatial distribution of grassland AGB increased from northwest to southeast over the entire Tibetan Plateau (TP) from 2001 to 2018. Grassland AGB increased more than it decreased (70.6% vs 29.4%, respectively) during the study period. Using a combination of FFS and LLO-CV, spatial overfitting was reduced, and the predictive accuracy of the RF was improved, thus enhancing the ability to predict the AGB in unknown locations from training data. This study proposes a robust methodology with which to improve the transferability of machine learning algorithms to predict grassland AGB in unknown locations.Hui YuYufeng WuLiting NiuYafan ChaiQisheng FengWei WangTiangang LiangElsevierarticleAlpine grasslandAbove-ground biomassRandom forestCross-validationOverfittingEcologyQH540-549.5ENEcological Indicators, Vol 125, Iss , Pp 107450- (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Alpine grassland Above-ground biomass Random forest Cross-validation Overfitting Ecology QH540-549.5
spellingShingle	Alpine grassland Above-ground biomass Random forest Cross-validation Overfitting Ecology QH540-549.5 Hui Yu Yufeng Wu Liting Niu Yafan Chai Qisheng Feng Wei Wang Tiangang Liang A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
description	Accurate assessments of grassland above-ground biomass (AGB) are crucial for the sustainable utilization and protection of grassland resources and the eco-environment. In this study, a random forest (RF) model combined with the forward feature selection (FFS) and leave-location-out cross-validation (LLO-CV) methods was trained to predict the dry weight (DW) of grassland AGB based on multiple factors. The final model exhibited a performance of R2 = 0.66, root mean square error (RMSE) of 503.86 kg DW/ha and mean absolute error (MAE) of 376.51 kg DW/ha. The spatial distribution of grassland AGB increased from northwest to southeast over the entire Tibetan Plateau (TP) from 2001 to 2018. Grassland AGB increased more than it decreased (70.6% vs 29.4%, respectively) during the study period. Using a combination of FFS and LLO-CV, spatial overfitting was reduced, and the predictive accuracy of the RF was improved, thus enhancing the ability to predict the AGB in unknown locations from training data. This study proposes a robust methodology with which to improve the transferability of machine learning algorithms to predict grassland AGB in unknown locations.
format	article
author	Hui Yu Yufeng Wu Liting Niu Yafan Chai Qisheng Feng Wei Wang Tiangang Liang
author_facet	Hui Yu Yufeng Wu Liting Niu Yafan Chai Qisheng Feng Wei Wang Tiangang Liang
author_sort	Hui Yu
title	A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
title_short	A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
title_full	A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
title_fullStr	A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
title_full_unstemmed	A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau
title_sort	method to avoid spatial overfitting in estimation of grassland above-ground biomass on the tibetan plateau
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/d1a70b3efee3483494c24b3188492737
work_keys_str_mv	AT huiyu amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT yufengwu amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT litingniu amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT yafanchai amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT qishengfeng amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT weiwang amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT tiangangliang amethodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT huiyu methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT yufengwu methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT litingniu methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT yafanchai methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT qishengfeng methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT weiwang methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau AT tiangangliang methodtoavoidspatialoverfittinginestimationofgrasslandabovegroundbiomassonthetibetanplateau
_version_	1718405794505949184

A method to avoid spatial overfitting in estimation of grassland above-ground biomass on the Tibetan Plateau

Ejemplares similares