Comparison of Parametric and Nonparametric Methods for Estimating Size–Density Relationships in Old-Growth Japanese Cedar (Cryptomeria japonica D. Don) Plantations
Abstract
:1. Introduction
2. Materials and Methods
2.1. Data
2.1.1. National Forest Resource Inventory Permanent Plots
2.1.2. Long-Term Experimental Plots
2.2. Parametric and Nonparametric Methods
2.2.1. Segmented Regression Models
2.2.2. Penalized Spline
2.2.3. Random Forest for Regression
2.3. Parameter Estimation and Method Evaluation
3. Results and Discussion
3.1. Size–Density Relationships Estimated by the Segmented Regression Models
3.2. Comparison of Parametric and Nonparametric Methods
4. Conclusions
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Burkhart, H.E.; Tomé, M. Modeling Forest Trees and Stands; Springer: Dordrecht, The Netherlands, 2012; p. 457. [Google Scholar]
- Yang, S.; Burkhart, H.E. Estimation of carrying capacity in loblolly pine (Pinus taeda L.). For. Ecol. Manag. 2017, 385, 167–176. [Google Scholar] [CrossRef] [Green Version]
- VanderSchaaf, C.L.; Burkhart, H.E. Using segmented regression to estimate stages and phases of stand development. For. Sci. 2008, 54, 167–175. [Google Scholar]
- Cao, Q.V.; Dean, T.J. Using segmented regression to model the density—Size relationship in direct-seeded slash pine stands. For. Ecol. Manag. 2008, 255, 948–952. [Google Scholar] [CrossRef]
- VanderSchaaf, C.L.; Burkhart, H.E. Comparison of methods to estimate Reineke’s maximum size-density relationship species boundary line slope. For. Sci. 2007, 53, 435–442. [Google Scholar] [CrossRef]
- Yang, S.; Burkhart, H.E. Application of height-and diameter-based relative spacing for estimation of stand basal area. For. Sci. 2018, 64, 28–32. [Google Scholar] [CrossRef]
- Zhang, X.; Lu, L.; Cao, Q.V.; Duan, A.; Zhang, J. Climate-sensitive self-thinning trajectories of Chinese fir plantations in south China. Can. J. For. Res. 2018, 48, 1388–1397. [Google Scholar]
- Cao, Q.V.; Dean, T.J.; Baldwin, V.C. Modeling the size-density relationship in direct-seeded slash pine stands. For. Sci. 2000, 46, 317–321. [Google Scholar] [CrossRef]
- Hagihara, A. Deriving the mean mass-density trajectory by reconciling the competition-density effect law with the self-thinning law in even-aged pure stands. J. For. Res. 2014, 19, 125–133. [Google Scholar] [CrossRef]
- Reineke, L.H. Perfecting a stand-density index for even- aged forests. J. Agric. Res. 1933, 46, 627–638. [Google Scholar]
- Vospernik, S.; Sterba, H. Do competition-density rule and self-thinning rule agree? Ann. For. Sci. 2015, 72, 379–390. [Google Scholar] [CrossRef] [Green Version]
- Yoda, K.; Kira, T.; Ogawa, H.; Hozumi, K. Self-thinning in overcrowded pure stands under cultivated and natural conditions. (Intraspecific competition among higher plants XI). J. Inst. Polytech. Osaka City Univ. Ser. 1963, 14, 107–129. [Google Scholar]
- Kira, T.; Ogawa, H.; Sakazaki, N. Intraspecific competition among higher plants. I. Competition-yield-density interrelationship in regularly dispersed populations. J. Inst. Poly. Osaka City Univ. Ser. D 1953, 4, 1–16. [Google Scholar]
- Zhang, X.; Cao, Q.V.; Duan, A.; Zhang, J. Self-thinning trajectories of Chinese fir plantations in southern China. For. Sci. 2016, 62, 594–599. [Google Scholar] [CrossRef] [Green Version]
- Nunes, M.H.; Görgens, E.B. Artificial intelligence procedures for tree taper estimation within a complex vegetation mosaic in Brazil. PLoS ONE 2016, 11, 1–16. [Google Scholar] [CrossRef] [Green Version]
- Scolforo, H.F.; McTague, J.P.; Raimundo, M.R.; Weiskittel, A.; Carrero, O.; Scolforo, J.R.S. Comparison of taper functions applied to eucalypts of varying genetics in Brazil: Application and evaluation of the penalized mixed spline approach. Can. J. For. Res. 2018, 48, 568–580. [Google Scholar] [CrossRef] [Green Version]
- Silva, C.A.; Klauberg, C.; Hudak, A.T.; Vierling, L.A.; Jaafar, W.S.W.M.; Mohan, M.; Garcia, M.; Ferraz, A.; Cardil, A.; Saatchi, S. Predicting stem total and assortment volumes in an industrial Pinus taeda L. forest plantation using airborne laser scanning data and random forest. Forests 2017, 8, 254. [Google Scholar] [CrossRef] [Green Version]
- Socha, J.; Netzel, P.; Cywicka, D. Stem taper approximation using a regression set and a neural network model. Forests 2019, 11, 79. [Google Scholar] [CrossRef] [Green Version]
- Nishizono, T.; Tanaka, K. Does the Relationship between Quadratic Mean Diameter and Stem Density in Old Thinned and Unthinned Cryptomeria japonica Forests Deviate from a Power Function? J. For. Plann. 2012, 18, 63–76. [Google Scholar]
- Ogawa, K. Time-trajectory of mean phytomass and density during a course of self-thinning in a sugi (Cryptomeria japonica D. Don) plantation. For. Ecol. Manag. 2005, 214, 104–110. [Google Scholar] [CrossRef]
- Ogawa, K.; Hagihara, A. Self-thinning and size variation in a sugi (Cryptomeria japonica D. Don) plantation. For. Ecol. Manag. 2003, 174, 413–421. [Google Scholar] [CrossRef]
- Yang, Y.C. Studies on the growth and yield of Crytomeria japonica in the Experimental Forest of National Taiwan University. Bull. Exp. For. Natl. Taiwan Univ. 1975, 116, 49, (In Chinese with English Summary). [Google Scholar]
- Chiou, C.R. A study on the allocation of the plots-An illustration of Tai-Pi working circle. Q. J. Chin. For. 1996, 30, 105–116. [Google Scholar]
- National Taiwan University Experimental Forest. Meteorological data in the Experimental Forest of College of Bioresources and Agriculture of National Taiwan University from 1941 to 2005, Spec. Bull NO.26, 1st ed.; National Taiwan University Experimental Forest: Nantou, Taiwan, 2009; pp. 4–6. (In Chinese) [Google Scholar]
- Eilers, P.H.C.; Marx, B.D. Flexible smoothing with B-splines and penalties. Stat. Sci. 1996, 11, 89–102. [Google Scholar] [CrossRef]
- Ruppert, D. Selecting the number of knots for penalized splines. J. Comput. Graph. Stat. 2002, 11, 735–757. [Google Scholar] [CrossRef]
- Breiman, L. Random Forests—Random Features. Tech. Rep. 1999, 567, 29. [Google Scholar]
- Williams, G. Data Mining with Rattle and R: The Art of Excavating Data for Knowledge Discovery; Springer: New York, NY, USA, 2011; p. 382. [Google Scholar] [CrossRef]
- Zhang, J.; Oliver, W.W.; Powers, R.F. Reevaluating the self-thinning boundary line for ponderosa pine (Pinus ponderosa) forests. Can. J. For. Res. 2013, 43, 963–971. [Google Scholar] [CrossRef]
- Fukumoto, K.; Nishizono, T.; Kitahara, F.; Hosoda, K. Evaluation of Individual Distance-Independent Diameter Growth Models for Japanese Cedar (Cryptomeria japonica) Trees under Multiple Thinning Treatments. Forests 2020, 11, 344. [Google Scholar] [CrossRef] [Green Version]
- Horowitz, J.L. Semiparametric Models. In International Encyclopedia of Social & Behavioral Sciences, 2nd ed.; Elsevier Inc.: Amsterdam, The Netherlands, 2015; pp. 598–603. [Google Scholar] [CrossRef]
- Jeong, J.H.; Resop, J.P.; Mueller, N.D.; Fleisher, D.H.; Yun, K.; Butler, E.E.; Timlin, D.J.; Shim, K.M.; Gerber, J.S.; Reddy, V.R.; et al. Random forests for global and regional crop yield predictions. PLoS ONE 2016, 11, e0156571. [Google Scholar] [CrossRef]
- Hazelton, M.L.; Sciences, F.; North, P.; Zealand, N. Nonparametric Regression. Int. Encycl. Soc. Behav. Sci. 2015, 16, 867–877. [Google Scholar] [CrossRef]
- Kublin, E.; Breidenbach, J.; Ka, G. A flexible stem taper and volume prediction method based on mixed-effects B-spline regression. Eur. J. For. Res. 2013, 132, 983–997. [Google Scholar] [CrossRef]
Sources | # of Plots | # of Obs. | Stand Variable | Average | Min. | Max. | Std. |
---|---|---|---|---|---|---|---|
Long-term experimental plots | 15 | 460 | Age | 36 | 2 | 105 | 24 |
DBH | 26.6 | 1.1 | 56.1 | 10.4 | |||
Dq | 27.4 | 1.1 | 58.8 | 10.6 | |||
TPH | 1269 | 190 | 3657 | 693 | |||
BA | 60.5 | 0.2 | 170.2 | 26.2 | |||
NFRI permanent plots | 222 | 515 | Age | 43 | 8 | 93 | 14 |
DBH | 25.9 | 11.9 | 55.6 | 8.2 | |||
Dq | 27.2 | 12.1 | 57.4 | 7.6 | |||
TPH | 1316 | 340 | 4400 | 604 | |||
BA | 79.5 | 13.3 | 197.6 | 35 |
Coefficient | Full Model | Reduced Model | ||
---|---|---|---|---|
Estimate | p-Value | Estimate | p-Value | |
β1 | 7.850 | <0.0001 * | 7.949 | <0.0001 * |
β2 | −1.267 | 0.079 | −0.445 | <0.0001 * |
β3 | −1.579 | 0.008 * | ||
β4 | −1.293 | <0.0001 * | ||
α1 | 2.389 | <0.0001 * | 1.833 | <0.0001 * |
α2 | 3.027 | 0.225 | ||
α3 | 3.211 | <0.0001 * |
Method | Fitting Data | Validation Data | ||
---|---|---|---|---|
MB | RMSE | MB | RMSE | |
Segmented Regression (Full) | 0.000 | 0.079 | 0.098 | 0.153 |
Segmented Regression (Reduced) | 0.000 | 0.083 | 0.083 | 0.151 |
Penalized Spline | 0.000 | 0.075 | 0.103 | 0.158 |
Random Forest for Regression | −0.001 | 0.040 | 0.110 | 0.190 |
Fitting Data | ||||||||
---|---|---|---|---|---|---|---|---|
Method | Stage I | Stage II | ||||||
Phase I | Phase II | Phase III | ||||||
MB | RMSE | MB | RMSE | MB | RMSE | MB | RMSE | |
Segmented Regression (Full) | −0.002 | 0.059 | 0.000 | 0.032 | −0.058 | 0.060 | 0.002 | 0.105 |
Segmented Regression (Reduced) | −0.061 | 0.073 | 0.080 | 0.039 | −0.032 | 0.062 | −0.016 | 0.106 |
Penalized Spline | −0.004 | 0.054 | 0.005 | 0.032 | −0.019 | 0.060 | 0.004 | 0.098 |
Random Forest for Regression | 0.010 | 0.021 | −0.003 | 0.018 | 0.002 | 0.032 | −0.002 | 0.052 |
Validation Data | ||||||||
Method | Stage I | Stage II | ||||||
Phase I | Phase II | Phase III | ||||||
MB | RMSE | MB | RMSE | MB | RMSE | MB | RMSE | |
Segmented Regression (Full) | - | - | −0.015 | 0.141 | −0.058 | 0.158 | 0.179 | 0.154 |
Segmented Regression (Reduced) | - | - | 0.049 | 0.144 | −0.033 | 0.158 | 0.143 | 0.150 |
Penalized Spline | - | - | −0.013 | 0.141 | −0.019 | 0.159 | 0.192 | 0.164 |
Random Forest for Regression | - | - | −0.003 | 0.140 | −0.021 | 0.175 | 0.202 | 0.212 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Chiou, C.-R.; Cheng, C.-P.; Yang, S.-I. Comparison of Parametric and Nonparametric Methods for Estimating Size–Density Relationships in Old-Growth Japanese Cedar (Cryptomeria japonica D. Don) Plantations. Forests 2020, 11, 625. https://doi.org/10.3390/f11060625
Chiou C-R, Cheng C-P, Yang S-I. Comparison of Parametric and Nonparametric Methods for Estimating Size–Density Relationships in Old-Growth Japanese Cedar (Cryptomeria japonica D. Don) Plantations. Forests. 2020; 11(6):625. https://doi.org/10.3390/f11060625
Chicago/Turabian StyleChiou, Chyi-Rong, Ching-Peng Cheng, and Sheng-I Yang. 2020. "Comparison of Parametric and Nonparametric Methods for Estimating Size–Density Relationships in Old-Growth Japanese Cedar (Cryptomeria japonica D. Don) Plantations" Forests 11, no. 6: 625. https://doi.org/10.3390/f11060625
APA StyleChiou, C. -R., Cheng, C. -P., & Yang, S. -I. (2020). Comparison of Parametric and Nonparametric Methods for Estimating Size–Density Relationships in Old-Growth Japanese Cedar (Cryptomeria japonica D. Don) Plantations. Forests, 11(6), 625. https://doi.org/10.3390/f11060625