Comparison of height-diameter models based on geographically weighted regressions and linear mixed modelling applied to large scale forest inventory data

María Quirós Segovia, Sonia Condés Ruiz, Karel Drápela

Abstract


Aim of the study: The main objective of this study was to test Geographically Weighted Regression (GWR) for developing height-diameter curves for forests on a large scale and to compare it with Linear Mixed Models (LMM).

Area of study: Monospecific stands of Pinus halepensis Mill. located in the region of Murcia (Southeast Spain).

Materials and Methods: The dataset consisted of 230 sample plots (2582 trees) from the Third Spanish National Forest Inventory (SNFI) randomly split into training data (152 plots) and validation data (78 plots). Two different methodologies were used for modelling local (Petterson) and generalized height-diameter relationships (Cañadas I): GWR, with different bandwidths, and linear mixed models. Finally, the quality of the estimated models was compared throughout statistical analysis.

Main results: In general, both LMM and GWR provide better prediction capability when applied to a generalized height-diameter function than when applied to a local one, with R2 values increasing from around 0.6 to 0.7 in the model validation. Bias and RMSE were also lower for the generalized function. However, error analysis showed that there were no large differences between these two methodologies, evidencing that GWR provides results which are as good as the more frequently used LMM methodology, at least when no additional measurements are available for calibrating.

Research highlights: GWR is a type of spatial analysis for exploring spatially heterogeneous processes. GWR can model spatial variation in tree height-diameter relationship and its regression quality is comparable to LMM. The advantage of GWR over LMM is the possibility to determine the spatial location of every parameter without additional measurements.

Keywords: Spatial analysis; Pinus halepensis Mill; forest modelling.

Abbreviations: GWR (Geographically Weighted Regression); LMM (Linear Mixed Model); SNFI (Spanish National Forest Inventory).

Keywords


Spatial analysis; Pinus halepensis Mill; forest modelling.

Full Text:

PDF HTML XML

References


References

Adamec Z, 2014. Využití moderních regresních metod pro modelování výškové křivky. Doctoral thesis. Mendelova Univerzita v Brně. Czech Republic.

Adamec Z, Drápela K, 2015. Generalized additive models as an alternative approach to the modelling of the tree height-diameter relationship. J For Sci 61(6): 235-243. http://dx.doi.org/10.17221/14/2015-JFS

Anselin L, Griffith DA, 1988. Do spatial effects really matter in regression analysis. Papers in Regional Science Association 65: 11–34. http://dx.doi.org/10.1111/j.1435-5597.1988.tb01155.x

Binbin L, Harris P, Charlton M, Brunsdon C, Nakaya T, Gollini I, 2015. R-CRAN Package "GWModel". Available at: http://cran.r-project.org/web/packages/GWmodel/GWmodel.pdf

Brunsdon C, Fotheringham S, Charlton M, 1996. Geographically Weighted Regression: A method for explore spatial non-stationarity. Geogr Anal 28 (4): 281–298. http://dx.doi.org/10.1111/j.1538-4632.1996.tb00936.x

Cabanillas A, 2010. Bases para la gestión de masas naturales de (Pinus halepensis Mill.) en el Valle del Ebro. Doctoral Thesis. Dept. de Ingeniería Forestal, Univ. Politécnica de Madrid, Spain.

Calama R, Montero G, 2004. Interregional nonlinear height-diameter model with random coefficients for stone pine in Spain. Can J For Res 34:150-163. http://dx.doi.org/10.1139/x03-199

Cañadas N, García C, Montero G, 1999. Relación altura-diámetro para Pinus pinea L. en el Sistema central. Actas del Congreso de Ordenación y Gestión Sostenible de Montes, Santiago de Compostela, Spain. I: 139:153.

Charlton M, Fotheringham S, 2009. Geographically Weighted Regression. White Paper for Science Foundation, Ireland: 1-14. http://www.geos.ed.ac.uk/~gisteac/fspat/gwr/gwr_arcgis/GWR_WhitePaper.pdf

Chen G, Zhao K, McDermid G, Hay G, 2012. The influence of sampling density on geographically weighted regression: a case study using forest canopy height and optical data. Int J Remote Sens IX (33): 2909-2924. http://dx.doi.org/10.1080/01431161.2011.624130

Condés S, Sterba H, 2008. Comparing an individual tree growth model for Pinus halepensis Mill. in the Spanish region of Murcia with yield tables gained from the same area. Eur J Forest Res 127: 253-261. http://dx.doi.org/10.1007/s10342-007-0201-7

Danlin Y, Yehua DW, 2013. Geographically Weighted Regression. http://slideplayer.com/slide/3428243/

Diéguez-Aranda U, Rojo Alboreca A, Castedo-Dorado F, Álvarez González JG, Barrio-Anta M, Crecente-Campo F, González González JM, Pérez-Cruzado C, Rodríguez Soalleiro R et al., 2009. Herramientas selvícolas para la gestión forestal sostenible en Galicia. Dirección Xeral de Montes, Consellería do Medio Rural, Xunta de Galicia, Spain. 259 pp.

Drápela K, 2011. Regresní modely a možnosti jejich využití v lesnictví. Habilitační práce. Mendelova univerzita v Brně. Czech Republic.

Fortin MJ, Dale M, 2005. Spatial Analysis: A Guide for ecologist. First ed. Cambridge: Cambridge University Press, UK.

Fotheringham S, Brunsdon C, Charlton M, 2002. Geographically Weighted Regression: The Analysis of Spatially Varying Relationships, John Wiley & Sons, New York, NY, 269 pp.

Gaffrey D, 1988. Forstamts-und bestandesindividuelles Sortimentierungsprogramm als Mittel zur Planung, Aushaltung und Simulation. Diplomarbeit Forscliche Fakultät, Univ. Göttingen. Germany.

Gollini I, Binbin L, Charlton M, Brunsdon C, Harris P, 2015. GWmodel: An R Package for Exploring Spatial Heterogeneity Using Geographically Weighted Models. J Stat Sof 63 (17): 1-50. http://dx.doi.org/10.18637/jss.v063.i17

Goodchild MF, Janelle DG, 2004. Spatially integrated social science. New York, NY: Oxford University Press.

Hart HMF, 1928. Stamtal en dunning; een orienteerend onderzoek naar de beste plantwijdte en dunning swijze woor den djati. Veenman & Zonen. Wageningen. Netherlands.

Jetz W, Ahbek CR, Ichstein JWL, 2005. Local and global approaches to spatial data analysis in ecology. Global Ecol Biogeogr 14: 97–98. http://dx.doi.org/10.1111/j.1466-822X.2004.00129.x

Juárez de Galíndez M, Pece MG, Gaillard de Benítez C, Sanguedolce J, Mariot V, Mazzuco R, 2007. Ecuaciones altura-diámetro en Prosopis nigra (Griseb) Hieron (algarrobo negro) en Santiago del Estero, Argentina. Foresta Veracruzana 9 (1): 9-14. Recursos Genéticos Forestales. Xalapa, México.

Lappi J, 1997. A longitudinal analysis of height/diameter curves. Forest Sci 43:555-570.

Li H, Reynolds JF, 1994. A simulation experiment to quantify spatial heterogeneity in categorical maps. Ecology 75: 2446–2455. http://dx.doi.org/10.2307/1940898

Lindstrom MJ, Bates DM, 1990. Nonlinear mixed effects for repeated measures data. Biometrics 46: 673–687. http://dx.doi.org/10.2307/2532087

Littell RC, Milliken GA, Stroup WW, Wolfinger RD, 1996. SAS System for Mixed Models. SAS Institute Inc, Cary, NC.

MAGRAMA, 2015. Ministerio de Agricultura, Alimentación y Medio Ambiente. Ministra: Isabel García Tejerina. http://www.magrama.gob.es

Matthews S, Yang TC, 2012. Mapping the results of local statistics: Using geographically weighted regression. Demographic Res 26(6): 151-166. http://dx.doi.org/10.4054/DemRes.2012.26.6

Mehtätalo L, 2004. A longitudinal height-diameter model for Norway spruce in Finland. Can J Forest Res 34(1): 131-140. http://dx.doi.org/10.1139/x03-207

Meng SX, Huang S, 2009. Improved calibration of nonlinear mixed-effects models demonstrated on a height growth function. Forest Sci 55(3): 238-248.

Milne BT, 1991. Heterogeneity as a multi-scale characteristic of landscape. In: J Kolasa and TA Pickett, Ecological Heterogeneity: 69-84. Springer-Verlag, New York.

Nanos N, Calama R, Montero G, Gil L, 2004. Geostatistical prediction of height/diameter models. Forest Ecol Manag 195: 221-235. http://dx.doi.org/10.1016/j.foreco.2004.02.031

Páscoa F, 1987. Estructura, crescimento e produçao em povoamentos de pinheiro bravo. Um modelo de simulaçao. Tesis doctoral. Universidade Técnica de Lisboa. Portugal.

Petterson H, 1955. Barrskogens volymproduktion. Meddelanden från Statens skogsforskningsinstitut 45 (1A): 1-391.

R Development Core Team, 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.r-project.org/

Robinson GK, 1991. That BLUP is a Good Thing: The Estimation of Random Effects. Statistical Sci 6(1): 15-32. http://dx.doi.org/10.1214/ss/1177011926

Salas C, Ene L, Gregoire T.G, Naesset E, Gobakken T, 2010. Modelling tree diameter from airborne laser scanning derived variables: A comparison of spatial statistical models. Remote Sens Environ 144: 1277–1285. http://dx.doi.org/10.1016/j.rse.2010.01.020

Schabenberger O, Pierce FJ, 2002. Contemporary Statistical Models for the Plant and Soil Sciences. CRC Press, Boca Raton, FL, 738.

Sloboda VB, Gaffrey D, Matsumura N, 1993. Regionale und lokale Systeme von Höhenkurven für gleichaltrige Waldbestände. Allg Forst-u J-Ztg 164: 225-228.

Tardanico R, 2006. GWR&2D Mapping. Florida International University. http://www2.fiu.edu/~tardanic/

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

Turner MG, 1989. Landscape ecology: the effect of pattern on process. Ann Rev Ecol Syst 20: 171–197. http://dx.doi.org/10.1146/annurev.es.20.110189.001131

Vonesh EF, Chiinchilli VM, 1997: Linear and nonlinear models for the analysis of repeated measurements. Boca Raton, Chapman & Hall/CRC: 560.

Wang Q, Ni J, Tenhunen J, 2005. Application of a geographically weighted regression to estimate net primary production of Chinese forest ecosystem. Global Ecol Biogeogr 14: 379– 393. http://dx.doi.org/10.1111/j.1466-822X.2005.00153.x

Watson DF, Philip GM, 1985. A Refinement of Inverse Distance Weighted Interpolation. Geoprocessing 2:315-327.

Wiens JA, 1976. Population responses to patchy environments. Ann Rev Ecol Syst 7: 81–120. http://dx.doi.org/10.1146/annurev.es.07.110176.000501

Yang Y, Huang S, 2011: Comparison of different methods for fitting nonlinear mixed forest models and for making predictions. Can J Forest Res 41: 1671–1686. http://dx.doi.org/10.1139/x11-071

Zhang L, Bi H, Cheng P, Davis CJ, 2004. Modelling spatial variation in tree diameter-height relationships. Forest Ecol Manag 189: 317-329. http://dx.doi.org/10.1016/j.foreco.2003.09.004

Zhang L, Gove JH, Heath LS, 2005. Spatial residual analysis of six modelling techniques. Ecol Model 186: 154-177. http://dx.doi.org/10.1016/j.ecolmodel.2005.01.007

Zhang L, Ma Z, Guo L, 2008. Spatially assessing model errors of four regression techniques for three types of forest stands. Oxford Journals 81(2): 209 – 225. http://dx.doi.org/10.1093/forestry/cpn014

Zhang L, Ma Z, Guo L, 2009. An evaluation of spatial autocorrelation and heterogeneity in the residuals of six regression models. Forest Sci 55: 533–548.

Zhang L, Shi H, 2004. Local Modelling of Tree Growth by Geographically Weighted Regression. Forest Sci 50 (2): 225-244.




DOI: 10.5424/fs/2016253-09787

Webpage: www.inia.es/Forestsystems