Evaluation of different modeling approaches for total tree-height estimation in Mediterranean Region of Turkey

M. J. Diamantopoulou, R. Özçelik

Abstract


Efficient management of timber resources and wood utilization practices require accurate and versatile information about important characteristics of forest resources for evaluating the numerous management and utilization alternatives for timber resources. Tree height is considered one of the most useful variables along with stocking and diameter at breast height, in estimating forest stand wood volumes and productivity. Six nonlinear growth functions were fitted to tree height-diameter data of three major tree species in Western Mediterranean Region’s forests of Turkey. The generalized regression neural network (GRNN) technique has been applied for tree height prediction, as well, due to its ability to fit complex nonlinear models. The performance of the models was compared and evaluated. Further, equivalence tests of the selected models were conducted. Validation showed the appropriatness of all models to predict tree height. According to the model performance criteria, the six nonlinear growth functions were able to capture the height-diameter relationships and fitted the data almost equally well, while the constructed generalized regression neural network (GRNN) models were found to be superior to all nonlinear regression models, in terms of their predictive ability.

 


Full Text:

PDF

References


Anonymous. 2006. Forest Resources. Ankara: The General Directorate of Forests.

Arabatzis AA, Burkhart HE. 1992. An evaluation of sampling methods and model forms for estimating height-diameter relationships in loblolly pine plantations. For Sci 38, 192-198.

Bishop CM. 1995. Neural Networks for Pattern Recognition. Oxford University Press. 498 pp.

Calama R, Montero G. 2004. Interregional nonlinear heightdiameter model with random coefficient for Stone pine in Spain. Can J For Res - Rev Can Rech For 34, 150-163.

Castedo-Dorado F, Anta MB, Parresol BR, Gonzalez JGA. 2005. A stochastic height-diameter model for maritime pine ecoregions in Galicia (Northwestern Spain). Ann For Sci 62, 455-465. http://dx.doi.org/10.1051/forest:2005042

Chtioui Y, Panigrahi S, Francl L. 1999. A generalized regression neural network and its application for leaf wetness prediction to forecast plant disease. Chemometrics and Intelligent Laboratory Systems 48, 47-58. http://dx.doi.org/10.1016/S0169-7439(99)00006-4

Cigizoglu HK, Alp M. 2006. Generalized regression neural network in modeling river sediment yield. Adv Eng Softw 37, 63-68. http://dx.doi.org/10.1016/j.advengsoft.2005.05.002

Colbert KC, Larsen DR, Lootens JR. 2002. Height-diameter equations for thirteen midwetern bottomland hardwood species. North J Appl For 19, 171-176.

Corne SA, Carver SJ, Kunin WE, Lennon JJ, Van Hees WWS. 2004. Predicting forest attributes in southeast Alaska using artificial neural networks. For Sci 50, 259-276.

Curtis RO. 1967. Height-diameter and height-age euations for second-growth Douglas-fir. For Sci 13, 365-375.

Davison AC, Hinkley DV. 1997. Bootstrap methods and their application. Cambridge Univercity Press, Cambridge. 585 pp.

Diamantopoulou MJ. 2010. Filling gaps in diameter measurements on standing tree boles in the urban forest of Thessaloniki, Greece. Environ Modell Softw 25, 1857-1865. http://dx.doi.org/10.1016/j.envsoft.2010.04.020

Dolph KL. 1989. Height-diameter Equations for younggrowth red fir in California and southern Oregon. Pacific Soutwest Forest and Range Experiment Station, Reserach Note PSW-408.

Eerikainen K. 2003. Predicting the height-diameter pattern of planted Pinus kesiya stands in Zambia and Zimbabwe. For Ecol Manage 175, 355-366.

Efron B. 1979. Bootstrap methods: another look at the jackknife. The Annals of Statistics 7, 1-26. http://dx.doi.org/10.1214/aos/1176344552

Efron B, Tibshirani RJ. 1993. An introduction to the bootstrap. Chapman and Hall, NY. 436 pp.

Esteban LG, Fernández FG, De Palacios P, Romero RM, Cano NN. 2009. Artificial neural networks in wood identification: The case of two Juniperus species from the Canary Islands. IAWA Journal 30, 87-94.

Esteban LG, Fernández FG, De Palacios P. 2011. Prediction of plywood bonding quality using an artificial neural network. Holzforschung 65, 209-214. http://dx.doi.org/10.1515/hf.2011.003

Fang Z, Bailey RL. 1998. Height-diameter models for tropical forest on Hainan Island in southern China. For Ecol Manage 110, 315-327.

Fekedulegn D, Siurtain MPM, Colbert JJ. 1999. Parameter estimation of nonlinear growth models in forestry. Silva Fenn 33, 327-336.

Fernández FG, Esteban LG, De Palacios P, Navarro N, Conde M. 2008. Prediction of standard particleboard mechanical properties utilizing an artificial neural network and subsequent comparison with a multivariate regression model. Investigacion Agraria Sistemas y Recursos Forestales 17, 178-187.

Foody G, Cutler M, Mcmorrow J, Pelz D, Tangki H, Boyd D, Douglas I. 2001. Mapping the biomass of Bornean tropical rain forest from remotely sensed data. Global Ecology & Biogeography 10, 379-387. http://dx.doi.org/10.1046/j.1466-822X.2001.00248.x

Fox J. 1991. Regression Diagnostics: An introduction (Quantitative Applications in the Social Sciences). Sage Publications, Newbury Park, CA. 92 pp.

Huang S, Titus SJ. 1992. Comparison of nonlinear heightdiameter functions for major Alberta tree species. Can J For Res - Rev Can Rech For 22, 1297-1304.

Huang S, Price D, Titus SJ. 2000. Development of ecoregionbased height-diameter models for white spruce in boreal forests. For Ecol Manage 129, 125-141.

Hirsch R. 1991. Validation samples. Biometrics 47, 1193-1194. PMid:1742438

Larsen DR, Hann DW. 1987. Height-diameter equations for seventeen tree species in southwest Oregon. Oregon State University Forest Research Laboratory Corvallis Research Paper: 49.

Leahy K. 1994. The overfitting problem in perspective. AI Expert 9(IO), 35-36.

Li X, Yang P, Ren S. 2009. Study on Transpiration Model for fruit tree based on generalized regression neural network. Proc ICEC Int Conf on «Engineering Computation», Washington, DC, (UDA), doi 10.1109/ICEC.2009.71.

Liu C, Zhang L, DavIs CJ, Solomon DS, Brann TB, Caldwell LE. 2003. Comparison of neural networks and statistical methods in classification of ecological habitats using FIA data. For Sci 49, 619-631.

Loehle C. 1997. A hypothesis testing framework for evaluating ecosystem modl performance. Ecol Model 97, 153-165. http://dx.doi.org/10.1016/S0304-3800(96)01900-X

Lootens JR, Larsen DR, Shifley SR. 2007. Height-diameter equations for 12 upland species in the Missouri Ozrak highland. North J Appl For 24, 149-152.

Maier HR, Dandy GC. 2000. Neural networks for the prediction and forecasting of water resources variables: a review of modeling issues and applications. Environ Modell Softw 15, 101-124. http://dx.doi.org/10.1016/S1364-8152(99)00007-9

Meng SX, Huang S, Lieffers VJ, Nunifu T, Yang Y. 2008. Wind speed and crown class influence the height-diameter relationship of lodgepole pine: Nonlinear mixed effects modeling. For Ecol Manage 256, 570-577.

Norusis MJ. 2000. SPSS for Windows. SPSS Inc., Chicago, IL.

Olson D, Delen D. 2008. Advanced Data Mining Techniques. Spriger-Verlang. 180 pp.

Özçelik R, Diamantopoulou MJ, Wiant HV, Brooks JR. 2008. Comparative study of standard and modern methods for estimating tree bole volume of three species in Turkey. For Prod J 58, 73-81.

Özçelik R, Brooks JR, Diamantopoulou MJ, Wiant HV. 2010. Estimating Breast Height Diameter and Volume from Stump Diameter for Three Economically Important Species in Turkey. Scand J Forest Res 25, 32-45. http://dx.doi.org/10.1080/02827580903280053

Parresol BR. 1992. Baldcypress height-diameter Equations and their prediction confidence intervals. Can J For Res - Rev Can Rech For 22, 1429-1434.

Parresol BR. 1993. Modeling multiplicative error variance: an example predicting tree diameter from stump dimensions in baldcypress. For Sci 39, 670-679.

Patterson D. 1996. Artificial Neural Networks: Theory and Applications. Prentice Hall, Singapure. 477 pp.

Peng C, Zhang L, Liu J. 2001. Developing and validating nonlinear height-diameter models for major tree species of Ontario's boreal forests. North J Appl For 18, 87-94.

Peng C, Zhang L, Zhou X, Dang Q, Huang S. 2004. Developing and evaluating tree height-diameter models at three geographic scales for black spruce in Ontario. North J Appl For 21, 83-92.

Pienaar LV, Turnbull KJ. 1973. The Chapman-Richards generalization of von Beralanffy's growth model for basal area growth and yield in even-aged stands. For Sci 19, 2-22.

Ratkowsky DA. 1990. Handbook of nonlinear regression models. Marcel Dekker, N.Y. 241 pp.

Robinson AP, Froese RE. 2004. Model validation using equivalence tests. Ecol Model 176, 349-358. http://dx.doi.org/10.1016/j.ecolmodel.2004.01.013

Robinson AP, Duursma RA, Marshall JD. 2005. A regressionbased equivalence test for model validation: shifting the burden of proof. Tree Physiol 25, 903-913. http://dx.doi.org/10.1093/treephys/25.7.903 PMid:15870057

SAS InstItute Inc., 2002. SAS/ETS User's Guide, 9th ver. SAS Institute Inc, Cary, NC.

Sharma M, Zhang SY. 2004. Height-diameter models using stand characteristics for Pinus banksiana and Picea mariana. Scand J Forest Res 19, 442-451. http://dx.doi.org/10.1080/02827580410030163

Sharma M, Patron J. 2007. Height-diameter equations for boreal tree species in Ontario using a mixed-effects modeling approach. For Ecol Manage 249, 187-198.

Soares P, Tome M. 2002. Height-diameter equation for first rotation eucalypt plantations in Portugal. For Ecol Manage 166, 99-109.

Specht DF. 1991. A Generalized Regression Neural Network. IEEE Transactions on Neural Networks 2, 568-576. http://dx.doi.org/10.1109/72.97934 PMid:18282872

Swingler K. 2001. Applying Neural Networks. A practical Guide. Morgan Kaufman Publishers Inc, Great Britain. 303 pp.

Temesgen H, Gadow K. 2004. Generalized height-diameter models: an application for major tree species in complex stands of interior British Colombia. Eur J For Res 123, 45-51. http://dx.doi.org/10.1007/s10342-004-0020-z

White H. 1980. A Heteroskedastic consistent covariance matrix and a direct test of heteroskedasticity. Econometrica 48, 817-838. http://dx.doi.org/10.2307/1912934

Yuancai L, Parresol BR. 2001. Remarks on height-diameter modeling. Res. Note. SRS-10. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 6p.

Zeide B. 1989. Accuracy of equations describing diameter growth. Can J For Res - Rev Can Rech For 19, 1283-1286.

Zeide B. 1993. Analysis of growth Equations. For Sci 39, 594-616.

Zhang L. 1997. Cross-validation of non-linear growth functions for modeling tree height-diameter relationship. Annals of Botany 79, 251-257. http://dx.doi.org/10.1006/anbo.1996.0334

Zhang L, Peng C, Huang S, Zhou X. 2002. Evaluation of ecoregion-based tree height-diameter models for jack pine in Ontario. The Forestry Chronicle 78, 530-538.




DOI: 10.5424/fs/2012213-02338

Webpage: www.inia.es/Forestsystems