Nonlinear models and machine learning algorithms for modeling height-diameter relationships in Pinus lawsonii Roezl
DOI:
https://doi.org/10.5377/ribcc.v10i19.19639Keywords:
R software, coefficient of determination, Akaike, Bayesian, errorAbstract
Background: Measuring tree characteristics, particularly height, in forest inventories is a costly and time-consuming task, especially in large areas. For this reason, efforts have been directed to generate mathematical models to estimate height from the diameter of pine trees. Aim: the objective of this research was to compare statistical models and machine learning algorithms to estimate total height based on normal diameter for Pinus lawsonii trees. Methodology: The study was carried out in the TESVB forest property, for which the height and diameter of 295 trees of different diameter categories were measured. Fifteen non-linear models were fitted from the data. Additionally, four machine learning algorithms were tested. The analysis was carried out with the R software. Result: With the non-linear models, it was observed that most of the models explained 66% (R2) of the variance in height based on normal diameter, however, all of them failed to meet the assumptions of normality and homoscedasticity. In the case of machine learning algorithms, the percentage of variance explained in total height from normal diameter for the training data ranged between 61 and 84 %. Conclusion: Considering the failure to meet assumptions, random forests are recommended to predict total height, and acceptable predictions with biological consistency were obtained considering the height measured in the field of Pinus lawsonii trees.
Downloads
315
References
Betancourt, G. A. (2005). Las máquinas de soporte vectorial (SVMs). Scientia et Technica, 27, 67–72.
Breiman, L. (2001). Random forests. Machine Learning, 45, 5–32. https://doi.org/10.1201/9780429469275-8
Castillo-Gallegos, E., Jarillo-Rodríguez, J., & Escobar-Hernández, R. (2018). Relación altura-diámetro en tres especies cultivadas en una plantación forestal comercial en el Este tropical de México. Revista Chapingo, Serie Ciencias Forestales y Del Ambiente, 24(1), 33–48. https://doi.org/10.5154/r.rchscfa.2017.05.033
Chau, J. (2024). gslnls: GSL Multi-Start Nonlinear Least-Squares Fitting_. R package version 1.3.2, <https://CRAN.R-project.org/package=gslnls>. (p. 37). https://doi.org/10.32614/CRAN.package.gslnls
Che, S., Tan, X., Xiang, C., Sun, J., Hu, X., Zhang, X., Duan, A., & Zhang, J. (2018). Stand basal area modelling for Chinese fir plantations using an artificial neural network model. Journal of Forestry Research, 30(5), 1641–1649. https://doi.org/10.1007/s11676-018-0711-9
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
Clutter, J. L., Fortson, J. C., Pienaar, L. V., Brister, G. H., & Bailey, R. L. (1983). Timber Management: A quantitative approach. (Primera ed). ohn Wiley & Sons.
Corral Rivas, S., Alma Mireya Silva, A., & Quiñonez-Barraza, G. (2019). A generalized nonlinear height-diameter model with mixed-effects for seven Pinus species in Durango, Mexico. Revista Mexicana de Ciencias Forestales, 10(53), 1–32. https://doi.org/https://doi.org/10.29298/rmcf.v10i53.500
Darmatasia, & Arymurthy, A. M. (2017). Predicting the status of water pumps using data mining approach. 2016 International Workshop on Big Data and Information Security, IWBIS 2016, 57–63. https://doi.org/10.1109/IWBIS.2016.7872890
Ercanli, İ., Senyurt, M., & Bolat, F. (2022). a Major Challenge To Machine Learning Models: Compati̇ble Predi̇cti̇ons Wi̇th Biological Realism in Forestry: a Case Study of individual tree volume. Proceeding of the “3rd International Conference on Environment and Forest Conservation (ICEFC), 1–31.
Flores-Morales, E. A., Aguirre-Calderón, O. A., Quiñónez-Barraza, G., González-Tagle, M. A., & Jiménez-Pérez, J. (2019). Normal diameter, height and volume estimation of Pinus pseudostrobus Lindl. From the diameter of the stump. Revista Mexicana de Ciencias Forestales, 10(55), 154–170. https://doi.org/10.29298/rmcf.v10i55.547
García Cuevas, X. A., Hernández Ramos, J., Hernández Ramos, A., Quiñonez Barraza, G., Tamarit Urías, J. C., & García Espinoza, G. G. (2017). Predicción del diámetro normal, altura y volumen a partir del diámetro del tocón en especies tropicales. Revista Mexicana de Ciencias Forestales, 8(43), 89–116. https://doi.org/10.29298/rmcf.v8i43.67
Guerra-De la Cruz, V., Islas-Gutiérrez, F., Flores-Ayala, E., Acosta-Mireles, M., Buendía-Rodríguez, E., Carrillo-Anzures, F., Tamarit-Urías, J. C., & Pineda-Ojeda, T. (2018). Modelos locales altura-diámetro para Pinus montezumae Lamb. y Pinus teocote Schiede ex Schltdl. en Nanacamilpa, Tlaxcala. Revista Mexicana de Ciencias Forestales, 10(51), 133–156. https://doi.org/https://doi.org/10.29298/rmcf.v10i51.407
H2O.ai. (2022). h2o: R Interface for H2O R package (version 3.42.0.2.). https://github.com/h2oai/h2o-3
Hernandez-Ramos, J., Aviles-Castillo, A., Jesus Garcia-Magana, J., Hernandez-Ramos, A., Garcia-Cuevas, X., & Flores-Lopez, C. (2020). Local and generalized height-diameter equations for Pinus patula Schl. et Cham. in Veracruz, Mexico. Ecosistemas Y Recursos Agropecuarios, 7(3), 1–11. https://doi.org/10.19136/era.a7n3.2457
INEGI. (2010). Compendio de información geográáca municipal 2010. Instituto Nacional de Estadistica y Geografia, 10. http://mapserver.inegi.org.mx/mgn2k/.
Kangas, A., & Haara, A. (2012). Comparison of nonspatial and spatial approaches with parametric and nonparametric methods in prediction of tree height. European Journal of Forest Research, 131(6), 1771–1782. https://doi.org/10.1007/s10342-012-0631-8
Krogh, A. (2008). What are artificial neural networks? Nature Biotechnology, 26(2), 195–197. https://doi.org/10.1038/nbt1386
Li, Y., Li, M., Li, C., & Liu, Z. (2020). Forest aboveground biomass estimation using Landsat 8 and Sentinel-1A data with machine learning algorithms. Scientific Reports, 10(1), 1–12. https://doi.org/10.1038/s41598-020-67024-3
Liaw, A., & Wiener, M. (2002). Classification and Regression by randomForest. R News, 2(3), 18–22. https://cran.r-project.org/doc/Rnews/
Lima, E. D. S., Souza, Z. M. De, Oliveira, S. R. D. M., Montanari, R., & Farhate, C. V. V. (2022). Random forest model to predict the height of Eucalyptus. Engenharia Agrícola, v.42, special issue, e20210153 https://doi.org/10.1590/1809-4430-Eng.Agric.v42nepe20210153/2022
López-Villegas, M. F., Santiago-García, W., Quiñonez-Barraza, G., Suárez-Mota, M. E., Santiago-Juárez, W., & Santiago-García, E. (2017). Ecuaciones globales y locales de altura-diámetro de 12 especies de interés comercial en bosques manejados. Revista Mexicana de Agroecosistemas, 4(2), 113–126.
Mayrinck, R. C., Roque, V. G. R., Filho, A. C. F., Filho, E. M., Arias-King, F., & Ribeiro, A. (2019). Height and volume functions for Pinus lawsonii, Pinus leiophylla, Pinus oocarpa and Pinus pringlei plantations in Guarei, São Paulo, Brazil. Southern Forests, 81(4), 325–334. https://doi.org/10.2989/20702620.2019.1636196
Meyer, D., Dimitriadou, E., Hornik, K., Weingessel, A., & F., L. (2023). e1071: Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), TU Wien. R News, 1–67. https://cran.r-project.org/package=e1071
Nunes-Miranda, E., Groenner-Barbosa, B. H., Godinho-Silva, S. H., Ussi-Monti, C. A., Phin-Tng, D. Y., & Rezende-Gomide, L. (2022). Variable selection for estimating individual tree height using genetic algorithm and random forest. Forest Ecology and Management, 504(119828), 1–13. https://doi.org/10.1016/j.foreco.2021.119828
Özçelik, R., Diamantopoulou, M. J., Crecente-campo, F., & Eler, U. (2013). Estimating Crimean juniper tree height using nonlinear regression and artificial neural network models. Forest Ecology and Management, 306, 52–60. https://doi.org/10.1016/j.foreco.2013.06.009
Quiñonez-Barraza, G., & García-Espinoza, G. G. (2018). ¿Cómo corregir la heterocedasticidad y autocorrelación de residuales en modelos de ahusamiento y crecimiento en altura? How to correct the heteroscedasticity and autocorrelation of residuals in taper and height growth models?. Revista Mexicana de Ciencias Forestales, 9(49), 28–59. https://doi.org/https://doi.org/10.29298/rmcf.v9i49.151
R Core Team. (2024). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
Șahin, A. (2024). Analyzing regression models and multi-layer artificial neural network models for estimating taper and tree volume in Crimean pine forests. IForest, 17, 36–44. https://doi.org/10.3832/ifor4449-017
Sharma, R. P. (2009). Modelling height-diameter relationship for Chir pine trees. Banko Janakari, 19(2), 3–9. https://doi.org/DOI: 10.3126/banko.v19i2.2978
Temesgen, H., Zhang, C. H., & Zhao, X. H. (2014). Modelling tree height-diameter relationships in multi-species and multi-layered forests: A large observational study from Northeast China. Forest Ecology and Management, 316, 78–89. https://doi.org/10.1016/j.foreco.2013.07.035
Zdenek, A., & Drápela, K. (2016). Comparison of parametric and nonparametric methods for modeling height-diameter relationships. IForest, 10, 1–8. https://doi.org/10.3832/ifor1928-009
Downloads
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2024 Rev. iberoam. bioecon. cambio clim.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright © Rev. iberoam. bioecon. climate change (Graduate School and UNAN-León, School of Agricultural and Veterinary Sciences / Department of Agroecology / Center for Research in Bioeconomy and Climate Cahnge (CRByCC).
