A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study

As a dominant terrestrial ecosystem, forests play a pivotal role, which is substantially challenged by climate extremes. At the same time, the practice of patient science to investigate and understand different intricate climate-driven phenomena is no longer an option. On the other hand, recent tech...

Full description

Saved in:
Bibliographic Details
Main Authors: Guilherme Cassales, Serajis Salekin, Nick Lim, Dean Meason, Albert Bifet, Bernhard Pfahringer, Eibe Frank
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Ecological Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1574954125000238
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850078916808540160
author Guilherme Cassales
Serajis Salekin
Nick Lim
Dean Meason
Albert Bifet
Bernhard Pfahringer
Eibe Frank
author_facet Guilherme Cassales
Serajis Salekin
Nick Lim
Dean Meason
Albert Bifet
Bernhard Pfahringer
Eibe Frank
author_sort Guilherme Cassales
collection DOAJ
description As a dominant terrestrial ecosystem, forests play a pivotal role, which is substantially challenged by climate extremes. At the same time, the practice of patient science to investigate and understand different intricate climate-driven phenomena is no longer an option. On the other hand, recent technological advancements enable scientists to simultaneously collect and analyse a large volume of complex data. High-resolution tree stem radius measurements and predictive simulation through machine learning algorithms offer powerful opportunities for understanding these dynamics. However, when these machine learning methods are applied without careful consideration of data quality, model biases, and other critical factors, their potential is often compromised. In this study, we aimed to evaluate four Deep Learning algorithms (namely CNN, LSTM, Transformer, and ETSFormer), using automatically measured and high temporal resolution tree stem radius data. We explore the complexities of handling voluminous and authentic datasets required by these algorithms. Initial experiments show that it is possible to achieve an MAE as small as 0.0026 mm on the full data. Furthermore, our study delves into the temporal resolution of data, demonstrating the feasibility of using reduced datasets without compromising accuracy levels. Our best result showed that a reduction of 97 % in collection events increases the MAE by only 6 % with the LSTM model, demonstrating that resource use optimisation can be achieved by slightly reducing the temporal resolution of data collection with marginal error increase. This also shows that LSTM can effectively capture longer-term and complex dependencies, which indicates promising future work with additional environmental data.
format Article
id doaj-art-4a0db7e45b1a44ef80cba641b2cd87ff
institution DOAJ
issn 1574-9541
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series Ecological Informatics
spelling doaj-art-4a0db7e45b1a44ef80cba641b2cd87ff2025-08-20T02:45:24ZengElsevierEcological Informatics1574-95412025-05-018610301410.1016/j.ecoinf.2025.103014A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case studyGuilherme Cassales0Serajis Salekin1Nick Lim2Dean Meason3Albert Bifet4Bernhard Pfahringer5Eibe Frank6TAIAO, University of Waikato, Hamilton, New Zealand; Corresponding author at: Te Ipu o Te Mahara AI Institute, Department of Computer Science – Tari Rorohiko, Private Bag 3105, Hamilton 3240, New Zealand.Scion (New Zealand Forest Research Ltd.), Rotorua 3046, New ZealandTAIAO, University of Waikato, Hamilton, New ZealandScion (New Zealand Forest Research Ltd.), Rotorua 3046, New ZealandTAIAO, University of Waikato, Hamilton, New ZealandTAIAO, University of Waikato, Hamilton, New ZealandTAIAO, University of Waikato, Hamilton, New ZealandAs a dominant terrestrial ecosystem, forests play a pivotal role, which is substantially challenged by climate extremes. At the same time, the practice of patient science to investigate and understand different intricate climate-driven phenomena is no longer an option. On the other hand, recent technological advancements enable scientists to simultaneously collect and analyse a large volume of complex data. High-resolution tree stem radius measurements and predictive simulation through machine learning algorithms offer powerful opportunities for understanding these dynamics. However, when these machine learning methods are applied without careful consideration of data quality, model biases, and other critical factors, their potential is often compromised. In this study, we aimed to evaluate four Deep Learning algorithms (namely CNN, LSTM, Transformer, and ETSFormer), using automatically measured and high temporal resolution tree stem radius data. We explore the complexities of handling voluminous and authentic datasets required by these algorithms. Initial experiments show that it is possible to achieve an MAE as small as 0.0026 mm on the full data. Furthermore, our study delves into the temporal resolution of data, demonstrating the feasibility of using reduced datasets without compromising accuracy levels. Our best result showed that a reduction of 97 % in collection events increases the MAE by only 6 % with the LSTM model, demonstrating that resource use optimisation can be achieved by slightly reducing the temporal resolution of data collection with marginal error increase. This also shows that LSTM can effectively capture longer-term and complex dependencies, which indicates promising future work with additional environmental data.http://www.sciencedirect.com/science/article/pii/S1574954125000238Point dendrometerTree stem radiusTime series forecastingDeep learningArtificial intelligenceCNN
spellingShingle Guilherme Cassales
Serajis Salekin
Nick Lim
Dean Meason
Albert Bifet
Bernhard Pfahringer
Eibe Frank
A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
Ecological Informatics
Point dendrometer
Tree stem radius
Time series forecasting
Deep learning
Artificial intelligence
CNN
title A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
title_full A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
title_fullStr A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
title_full_unstemmed A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
title_short A comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer: A case study
title_sort comparative study of four deep learning algorithms for predicting tree stem radius measured by dendrometer a case study
topic Point dendrometer
Tree stem radius
Time series forecasting
Deep learning
Artificial intelligence
CNN
url http://www.sciencedirect.com/science/article/pii/S1574954125000238
work_keys_str_mv AT guilhermecassales acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT serajissalekin acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT nicklim acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT deanmeason acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT albertbifet acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT bernhardpfahringer acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT eibefrank acomparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT guilhermecassales comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT serajissalekin comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT nicklim comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT deanmeason comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT albertbifet comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT bernhardpfahringer comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy
AT eibefrank comparativestudyoffourdeeplearningalgorithmsforpredictingtreestemradiusmeasuredbydendrometeracasestudy