Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas

This study addresses the task of forecasting Basal Area Increment trends in forest ecosystems, which is essential for conservation and biodiversity management, particularly in the context of climate change. Traditional forecasting techniques, such as Linear Mixed Models, Random Forest and standard A...

Full description

Saved in:
Bibliographic Details
Main Authors: P. Casas-Gómez, J.F. Torres, J.C. Linares, A. Troncoso, F. Martínez-Álvarez
Format: Article
Language:English
Published: Elsevier 2025-03-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S157495412400493X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832595383309565952
author P. Casas-Gómez
J.F. Torres
J.C. Linares
A. Troncoso
F. Martínez-Álvarez
author_facet P. Casas-Gómez
J.F. Torres
J.C. Linares
A. Troncoso
F. Martínez-Álvarez
author_sort P. Casas-Gómez
collection DOAJ
description This study addresses the task of forecasting Basal Area Increment trends in forest ecosystems, which is essential for conservation and biodiversity management, particularly in the context of climate change. Traditional forecasting techniques, such as Linear Mixed Models, Random Forest and standard Artificial Neural Networks, often fail to account for the time-dependent nature of tree growth and utilize simple architectures. To overcome these limitations, we introduce the use of two different Deep Learning models: the Long Short-Term Memory network and the Temporal Convolutional Neural Network, which capture the temporal dependencies of growth by incorporating lagged Basal Area Increment values. Our methodology includes rigorous hyperparameter tuning to optimize the Deep Learning models’ architecture. We evaluate the models’ performance across 15 species in the Himalayan region, individually and collectively, using temperature and precipitation data as predictors. The Deep Learning model significantly outperforms state-of-the-art techniques, achieving the lowest Root Mean Squared Error (7.407 for LSTM and 6.202 for TCNN), highest R2 (0.495 for LSTM and 0.585 for TCNN) and lowest Mean Absolute Percentage Error values (37.653 for LSTM and 34.296 for TCNN). These findings highlight the potential of Deep Learning networks to provide accurate and reliable Basal Area Increment forecasts, offering valuable insights for forest management and conservation efforts in the face of ongoing climate change.
format Article
id doaj-art-2ea267d8a32f4fb688c87d815d0cc07e
institution Kabale University
issn 1574-9541
language English
publishDate 2025-03-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-2ea267d8a32f4fb688c87d815d0cc07e2025-01-19T06:24:38ZengElsevierEcological Informatics1574-95412025-03-0185102951Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the HimalayasP. Casas-Gómez0J.F. Torres1J.C. Linares2A. Troncoso3F. Martínez-Álvarez4Data Science and Big Data Lab, Pablo de Olavide University, ES-41013 Seville, Spain; Corresponding author.Data Science and Big Data Lab, Pablo de Olavide University, ES-41013 Seville, SpainPhysical, Chemical and Natural Systems Department, Faculty of Experimental Sciences, Pablo de Olavide University, ES-41013 Seville, SpainData Science and Big Data Lab, Pablo de Olavide University, ES-41013 Seville, SpainData Science and Big Data Lab, Pablo de Olavide University, ES-41013 Seville, SpainThis study addresses the task of forecasting Basal Area Increment trends in forest ecosystems, which is essential for conservation and biodiversity management, particularly in the context of climate change. Traditional forecasting techniques, such as Linear Mixed Models, Random Forest and standard Artificial Neural Networks, often fail to account for the time-dependent nature of tree growth and utilize simple architectures. To overcome these limitations, we introduce the use of two different Deep Learning models: the Long Short-Term Memory network and the Temporal Convolutional Neural Network, which capture the temporal dependencies of growth by incorporating lagged Basal Area Increment values. Our methodology includes rigorous hyperparameter tuning to optimize the Deep Learning models’ architecture. We evaluate the models’ performance across 15 species in the Himalayan region, individually and collectively, using temperature and precipitation data as predictors. The Deep Learning model significantly outperforms state-of-the-art techniques, achieving the lowest Root Mean Squared Error (7.407 for LSTM and 6.202 for TCNN), highest R2 (0.495 for LSTM and 0.585 for TCNN) and lowest Mean Absolute Percentage Error values (37.653 for LSTM and 34.296 for TCNN). These findings highlight the potential of Deep Learning networks to provide accurate and reliable Basal Area Increment forecasts, offering valuable insights for forest management and conservation efforts in the face of ongoing climate change.http://www.sciencedirect.com/science/article/pii/S157495412400493XBAITime series forecastingDeep learningClimate
spellingShingle P. Casas-Gómez
J.F. Torres
J.C. Linares
A. Troncoso
F. Martínez-Álvarez
Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
Ecological Informatics
BAI
Time series forecasting
Deep learning
Climate
title Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
title_full Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
title_fullStr Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
title_full_unstemmed Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
title_short Forecasting basal area increment in forest ecosystems using deep learning: A multi-species analysis in the Himalayas
title_sort forecasting basal area increment in forest ecosystems using deep learning a multi species analysis in the himalayas
topic BAI
Time series forecasting
Deep learning
Climate
url http://www.sciencedirect.com/science/article/pii/S157495412400493X
work_keys_str_mv AT pcasasgomez forecastingbasalareaincrementinforestecosystemsusingdeeplearningamultispeciesanalysisinthehimalayas
AT jftorres forecastingbasalareaincrementinforestecosystemsusingdeeplearningamultispeciesanalysisinthehimalayas
AT jclinares forecastingbasalareaincrementinforestecosystemsusingdeeplearningamultispeciesanalysisinthehimalayas
AT atroncoso forecastingbasalareaincrementinforestecosystemsusingdeeplearningamultispeciesanalysisinthehimalayas
AT fmartinezalvarez forecastingbasalareaincrementinforestecosystemsusingdeeplearningamultispeciesanalysisinthehimalayas