Evaluation of spatial predictions of site index obtained by parametric and nonparametric methods—A case study of lodgepole pine productivity

We demonstrate the potential of using least-squares regression, generalized additive model, tree-based model, and neural network model on layers of environmental data grids for mapping site index in a case study. Grids of numerical environmental variables represented layered data, and a sparse site index plot network was located in the grids. Site index data were based on stem analysis (observed height at the index age of 50 years) of 431 lodgepole pine trees in 88 sample plots. The plots were established in a 17,460 km2 boreal mixedwood forest of Alberta, Canada dominated by mature and over-mature stands. The generalized additive model presented a better fit and better adaptability to extreme data (i.e., mature stands) than the least squares nonlinear and other nonparametric techniques, such as the tree-based model and neural network model. Among the four models tested, nonlinear regression is of the data modeling culture, which assumes a stochastic data to relate productivity to environmental variables, and such models are optimized for estimation. Other three models belong to the algorithm modeling culture, which treat the relationship between productivity and independent variables as an unknown black box and try to find a function between them; therefore, these models are more suitable for prediction purpose. Implications for biophysical site index modelling with extreme data are discussed.

[1]  K. Klinka,et al.  Initial quantitative characterization of soil nutrient regimes. II. Relationships among soils, vegetation, and site index , 1987 .

[2]  E. Hogg,et al.  Climate and the southern limit of the western Canadian boreal forest , 1994 .

[3]  G. Gertner,et al.  Modeling red pine tree survival with an artificial neural network , 1991 .

[4]  Daniel W. McKenney,et al.  Spatial models of site index based on climate and soil properties for two boreal tree species in Ontario, Canada , 2003 .

[5]  George Z. Gertner,et al.  Using a Parallel Distributed Processing System to Model Individual Tree Mortality , 1991 .

[6]  Edgar Robichaud,et al.  The effect of site quality on the timing of stand breakup, tree longevity, and the maximum attainable height of black spruce , 1993 .

[7]  C. Tarnócai,et al.  Soil landscapes of Canada procedures manual and user's handbook , 1991 .

[8]  Oscar García,et al.  Evaluating forest Growth Models , 1997 .

[9]  Annika Kangas,et al.  Estimating individual tree growth with nonparametric methods , 2003 .

[10]  T. Kira,et al.  PRIMARY PRODUCTION AND TURNOVER OF ORGANIC MATTER IN DIFFERENT FOREST ECOSYSTEMS OF THE WESTERN PACIFIC , 1967 .

[11]  Russell C. Eberhart,et al.  Neural network PC tools: a practical guide , 1990 .

[12]  Shongming Huang.  A Growth and yield projection system for natural and regenerated stands within an ecologically based, enhanced forest management framework : yield tables for seed-origin natural and regenerated lodgepole pine stands / , 2001 .

[13]  A. R. Gibson,et al.  PREDICTING PINUS RADIATA SITE INDEX FROM ENVIRONMENTAL VARIABLES , 1984 .

[14]  D. Lindgren,et al.  Effects of temperature on the site productivity of Pinus sylvestris and lodgepole pine in Finland and Sweden , 1998 .

[15]  T. Yee,et al.  Generalized additive models in plant ecology , 1991 .

[16]  F. Smith,et al.  Age-related changes in production and below-ground carbon allocation in Pinus contorta forests , 1999 .

[17]  J. Régnière,et al.  Biophysical Site Indices for Shade Tolerant and Intolerant Boreal Species , 2001, Forest Science.

[18]  David A. Belsley,et al.  Regression Analysis and its Application: A Data-Oriented Approach.@@@Applied Linear Regression.@@@Regression Diagnostics: Identifying Influential Data and Sources of Collinearity , 1981 .

[19]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001, Statistical Science.

[20]  R. Allen,et al.  Evapotranspiration and Irrigation Water Requirements , 1990 .

[21]  Robert Tibshirani,et al.  An Introduction to the Bootstrap , 1994 .

[22]  R. Hebda,et al.  Modeling Tree-Ring Growth Responses to Climatic Variables Using Artificial Neural Networks , 2000, Forest Science.

[23]  B. Mackey,et al.  Primary databases for forest ecosystem management-examples from Ontario and possibilities for Canada: NatGRID , 1996, Environmental monitoring and assessment.

[24]  Michael D. Morgan,et al.  Meteorology: The Atmosphere and the Science of Weather , 1989 .

[25]  Gordon B. Bonan,et al.  A computer model of the solar radiation, soil moisture, and soil thermal regimes in boreal forests , 1989 .

[26]  J. H. Archibald,et al.  Field Guide to Ecosites of West-Central Alberta , 2002 .

[27]  J. P. Kimmins,et al.  Disturbances and the sustainability of long-term site productivity in lodgepole pine forests in the central interior of British Columbia—an ecosystem modeling approach , 2003 .

[28]  Michel Verleysen,et al.  Fast approximation of the bootstrap for model selection , 2003, ESANN.

[29]  R. Tibshirani,et al.  Generalized Additive Models , 1991 .

[30]  Tong Tang,et al.  Proceedings of the European Symposium on Artificial Neural Networks , 2006 .

[31]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001 .

[32]  J. R. Koehler,et al.  Modern Applied Statistics with S-Plus. , 1996 .

[33]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1993 .

[34]  I. Corns,et al.  Vegetational indicators as independent variables in forest growth prediction in West-Central Alberta, Canada , 1984 .

[35]  N. Draper,et al.  Applied Regression Analysis. , 1967 .