Prediction of tree crown width in natural mixed forests using deep learning algorithm
Crown width (CW) is one of the most important tree metrics, but obtaining CW data is laborious and time-consuming, particularly in natural forests. The Deep Learning (DL) algorithm has been proposed as an alternative to traditional regression, but its performance in predicting CW in natural mixed fo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2023-01-01
|
| Series: | Forest Ecosystems |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2197562023000404 |
| _version_ | 1797382928086335488 |
|---|---|
| author | Yangping Qin Biyun Wu Xiangdong Lei Linyan Feng |
| author_facet | Yangping Qin Biyun Wu Xiangdong Lei Linyan Feng |
| author_sort | Yangping Qin |
| collection | DOAJ |
| description | Crown width (CW) is one of the most important tree metrics, but obtaining CW data is laborious and time-consuming, particularly in natural forests. The Deep Learning (DL) algorithm has been proposed as an alternative to traditional regression, but its performance in predicting CW in natural mixed forests is unclear. The aims of this study were to develop DL models for predicting tree CW of natural spruce-fir-broadleaf mixed forests in north-eastern China, to analyse the contribution of tree size, tree species, site quality, stand structure, and competition to tree CW prediction, and to compare DL models with nonlinear mixed effects (NLME) models for their reliability. An amount of total 10,086 individual trees in 192 subplots were employed in this study. The results indicated that all deep neural network (DNN) models were free of overfitting and statistically stable within 10-fold cross-validation, and the best DNN model could explain 69% of the CW variation with no significant heteroskedasticity. In addition to diameter at breast height, stand structure, tree species, and competition showed significant effects on CW. The NLME model (R2 = 0.63) outperformed the DNN model (R2 = 0.54) in predicting CW when the six input variables were consistent, but the results were the opposite when the DNN model (R2 = 0.69) included all 22 input variables. These results demonstrated the great potential of DL in tree CW prediction. |
| first_indexed | 2024-03-08T21:12:55Z |
| format | Article |
| id | doaj.art-fa3d903454e747dfa64d8079dcb6cc4a |
| institution | Directory Open Access Journal |
| issn | 2197-5620 |
| language | English |
| last_indexed | 2024-03-08T21:12:55Z |
| publishDate | 2023-01-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Forest Ecosystems |
| spelling | doaj.art-fa3d903454e747dfa64d8079dcb6cc4a2023-12-22T05:32:31ZengKeAi Communications Co., Ltd.Forest Ecosystems2197-56202023-01-0110100109Prediction of tree crown width in natural mixed forests using deep learning algorithmYangping Qin0Biyun Wu1Xiangdong Lei2Linyan Feng3Key Laboratory of Forest Management and Growth Modelling, National Forestry and Grassland Administration, Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, 100091, China; Southwest Survey and Planning Institute, National Forestry and Grassland Administration, Kunming, 650031, ChinaKey Laboratory of Forest Management and Growth Modelling, National Forestry and Grassland Administration, Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, 100091, ChinaKey Laboratory of Forest Management and Growth Modelling, National Forestry and Grassland Administration, Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, 100091, China; Corresponding author.Key Laboratory of Forest Management and Growth Modelling, National Forestry and Grassland Administration, Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing, 100091, ChinaCrown width (CW) is one of the most important tree metrics, but obtaining CW data is laborious and time-consuming, particularly in natural forests. The Deep Learning (DL) algorithm has been proposed as an alternative to traditional regression, but its performance in predicting CW in natural mixed forests is unclear. The aims of this study were to develop DL models for predicting tree CW of natural spruce-fir-broadleaf mixed forests in north-eastern China, to analyse the contribution of tree size, tree species, site quality, stand structure, and competition to tree CW prediction, and to compare DL models with nonlinear mixed effects (NLME) models for their reliability. An amount of total 10,086 individual trees in 192 subplots were employed in this study. The results indicated that all deep neural network (DNN) models were free of overfitting and statistically stable within 10-fold cross-validation, and the best DNN model could explain 69% of the CW variation with no significant heteroskedasticity. In addition to diameter at breast height, stand structure, tree species, and competition showed significant effects on CW. The NLME model (R2 = 0.63) outperformed the DNN model (R2 = 0.54) in predicting CW when the six input variables were consistent, but the results were the opposite when the DNN model (R2 = 0.69) included all 22 input variables. These results demonstrated the great potential of DL in tree CW prediction.http://www.sciencedirect.com/science/article/pii/S2197562023000404Mixed forestsDeep neural networksCrown widthStand structureCompetition |
| spellingShingle | Yangping Qin Biyun Wu Xiangdong Lei Linyan Feng Prediction of tree crown width in natural mixed forests using deep learning algorithm Forest Ecosystems Mixed forests Deep neural networks Crown width Stand structure Competition |
| title | Prediction of tree crown width in natural mixed forests using deep learning algorithm |
| title_full | Prediction of tree crown width in natural mixed forests using deep learning algorithm |
| title_fullStr | Prediction of tree crown width in natural mixed forests using deep learning algorithm |
| title_full_unstemmed | Prediction of tree crown width in natural mixed forests using deep learning algorithm |
| title_short | Prediction of tree crown width in natural mixed forests using deep learning algorithm |
| title_sort | prediction of tree crown width in natural mixed forests using deep learning algorithm |
| topic | Mixed forests Deep neural networks Crown width Stand structure Competition |
| url | http://www.sciencedirect.com/science/article/pii/S2197562023000404 |
| work_keys_str_mv | AT yangpingqin predictionoftreecrownwidthinnaturalmixedforestsusingdeeplearningalgorithm AT biyunwu predictionoftreecrownwidthinnaturalmixedforestsusingdeeplearningalgorithm AT xiangdonglei predictionoftreecrownwidthinnaturalmixedforestsusingdeeplearningalgorithm AT linyanfeng predictionoftreecrownwidthinnaturalmixedforestsusingdeeplearningalgorithm |