A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland
International audience This study compares the predictive accuracy of eight state-of-the-art modelling techniques for 12 landforms types in a cold environment. The methods used are Random Forest (RF), Artificial Neural Networks (ANN), Generalized Boosting Methods (GBM), Generalized Linear Models (GL...
Main Authors: | , , , |
---|---|
Other Authors: | , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
HAL CCSD
2008
|
Subjects: | |
Online Access: | https://hal.science/halsde-00377979 |
id |
ftunigrenoble:oai:HAL:halsde-00377979v1 |
---|---|
record_format |
openpolar |
spelling |
ftunigrenoble:oai:HAL:halsde-00377979v1 2024-05-12T07:59:48+00:00 A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland Marmion, M. Hjort, J. Thuiller, W. Luoto, M. Thule Institute University of Oulu Department of Geography Oulu Department of Geosciences and Geography Helsinki Falculty of Science Helsinki Helsingin yliopisto = Helsingfors universitet = University of Helsinki-Helsingin yliopisto = Helsingfors universitet = University of Helsinki Laboratoire d'Ecologie Alpine (LECA) Université Joseph Fourier - Grenoble 1 (UJF)-Université Savoie Mont Blanc (USMB Université de Savoie Université de Chambéry )-Centre National de la Recherche Scientifique (CNRS) 2008 https://hal.science/halsde-00377979 en eng HAL CCSD Wiley halsde-00377979 https://hal.science/halsde-00377979 ISSN: 0197-9337 EISSN: 1096-9837 Earth Surface Processes and Landforms https://hal.science/halsde-00377979 Earth Surface Processes and Landforms, 2008, 33 (14), pp.2241-2254 spatial modelling statistical techniques AUC Kappa predictive performance Adaptive regression splines artificial neural-networks sub-arctic finland species distributions spatial autocorrelation logistic-regression climate-change gis permafrost scale [SDE.BE]Environmental Sciences/Biodiversity and Ecology [SDE.MCG]Environmental Sciences/Global Changes [SDV.BID]Life Sciences [q-bio]/Biodiversity [SDV.EE]Life Sciences [q-bio]/Ecology environment info:eu-repo/semantics/article Journal articles 2008 ftunigrenoble 2024-04-18T03:02:15Z International audience This study compares the predictive accuracy of eight state-of-the-art modelling techniques for 12 landforms types in a cold environment. The methods used are Random Forest (RF), Artificial Neural Networks (ANN), Generalized Boosting Methods (GBM), Generalized Linear Models (GLM), Generalized Additive Models (GAM), Multivariate Adaptive Regression Splines (MARS), Classification Tree Analysis (CTA) unit Mixture Discriminant Analysis (MDA). The spatial distributions of 12 periglacial landforms types were recorded in sub-Arctic landscape of northern Finland in 2032 grid squares at a resolution of 25 ha. First, three topographic variables were implemented into the eight modelling techniques (simple model), and then six other variables were added (three soil and three vegetation variables; complex model) to reflect the environmental conditions of each grid square. The predictive accuracy was measured by two methods: the area under the curve (AUC) of a receiver operating characteristic (ROC) plot, and the Kappa index (K), based on spatially independent model evaluation data. The mean AUC values of the simple models varied between 0.709 and 0.796, whereas the AUC values of the complex model ranged from 0.725 to 0.825. For both simple and complex models GAM, GLM, ANN and GBM provided the highest predictive performances based on both AUC and kappa values. The results encourage further applications of the novel modelling methods in geomorphology. Copyright (D 2008 John Wiley & Sons, Ltd. Article in Journal/Newspaper Arctic Climate change Northern Finland permafrost Lapland Université Grenoble Alpes: HAL Arctic Gam ENVELOPE(-57.955,-57.955,-61.923,-61.923) |
institution |
Open Polar |
collection |
Université Grenoble Alpes: HAL |
op_collection_id |
ftunigrenoble |
language |
English |
topic |
spatial modelling statistical techniques AUC Kappa predictive performance Adaptive regression splines artificial neural-networks sub-arctic finland species distributions spatial autocorrelation logistic-regression climate-change gis permafrost scale [SDE.BE]Environmental Sciences/Biodiversity and Ecology [SDE.MCG]Environmental Sciences/Global Changes [SDV.BID]Life Sciences [q-bio]/Biodiversity [SDV.EE]Life Sciences [q-bio]/Ecology environment |
spellingShingle |
spatial modelling statistical techniques AUC Kappa predictive performance Adaptive regression splines artificial neural-networks sub-arctic finland species distributions spatial autocorrelation logistic-regression climate-change gis permafrost scale [SDE.BE]Environmental Sciences/Biodiversity and Ecology [SDE.MCG]Environmental Sciences/Global Changes [SDV.BID]Life Sciences [q-bio]/Biodiversity [SDV.EE]Life Sciences [q-bio]/Ecology environment Marmion, M. Hjort, J. Thuiller, W. Luoto, M. A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
topic_facet |
spatial modelling statistical techniques AUC Kappa predictive performance Adaptive regression splines artificial neural-networks sub-arctic finland species distributions spatial autocorrelation logistic-regression climate-change gis permafrost scale [SDE.BE]Environmental Sciences/Biodiversity and Ecology [SDE.MCG]Environmental Sciences/Global Changes [SDV.BID]Life Sciences [q-bio]/Biodiversity [SDV.EE]Life Sciences [q-bio]/Ecology environment |
description |
International audience This study compares the predictive accuracy of eight state-of-the-art modelling techniques for 12 landforms types in a cold environment. The methods used are Random Forest (RF), Artificial Neural Networks (ANN), Generalized Boosting Methods (GBM), Generalized Linear Models (GLM), Generalized Additive Models (GAM), Multivariate Adaptive Regression Splines (MARS), Classification Tree Analysis (CTA) unit Mixture Discriminant Analysis (MDA). The spatial distributions of 12 periglacial landforms types were recorded in sub-Arctic landscape of northern Finland in 2032 grid squares at a resolution of 25 ha. First, three topographic variables were implemented into the eight modelling techniques (simple model), and then six other variables were added (three soil and three vegetation variables; complex model) to reflect the environmental conditions of each grid square. The predictive accuracy was measured by two methods: the area under the curve (AUC) of a receiver operating characteristic (ROC) plot, and the Kappa index (K), based on spatially independent model evaluation data. The mean AUC values of the simple models varied between 0.709 and 0.796, whereas the AUC values of the complex model ranged from 0.725 to 0.825. For both simple and complex models GAM, GLM, ANN and GBM provided the highest predictive performances based on both AUC and kappa values. The results encourage further applications of the novel modelling methods in geomorphology. Copyright (D 2008 John Wiley & Sons, Ltd. |
author2 |
Thule Institute University of Oulu Department of Geography Oulu Department of Geosciences and Geography Helsinki Falculty of Science Helsinki Helsingin yliopisto = Helsingfors universitet = University of Helsinki-Helsingin yliopisto = Helsingfors universitet = University of Helsinki Laboratoire d'Ecologie Alpine (LECA) Université Joseph Fourier - Grenoble 1 (UJF)-Université Savoie Mont Blanc (USMB Université de Savoie Université de Chambéry )-Centre National de la Recherche Scientifique (CNRS) |
format |
Article in Journal/Newspaper |
author |
Marmion, M. Hjort, J. Thuiller, W. Luoto, M. |
author_facet |
Marmion, M. Hjort, J. Thuiller, W. Luoto, M. |
author_sort |
Marmion, M. |
title |
A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
title_short |
A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
title_full |
A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
title_fullStr |
A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
title_full_unstemmed |
A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland |
title_sort |
comparison of predictive methods in modelling the distribution of periglacial landforms in finnish lapland |
publisher |
HAL CCSD |
publishDate |
2008 |
url |
https://hal.science/halsde-00377979 |
long_lat |
ENVELOPE(-57.955,-57.955,-61.923,-61.923) |
geographic |
Arctic Gam |
geographic_facet |
Arctic Gam |
genre |
Arctic Climate change Northern Finland permafrost Lapland |
genre_facet |
Arctic Climate change Northern Finland permafrost Lapland |
op_source |
ISSN: 0197-9337 EISSN: 1096-9837 Earth Surface Processes and Landforms https://hal.science/halsde-00377979 Earth Surface Processes and Landforms, 2008, 33 (14), pp.2241-2254 |
op_relation |
halsde-00377979 https://hal.science/halsde-00377979 |
_version_ |
1798841422097416192 |