Agritrop
Home

Simulating realistic spatial structure for forest stands: a mimetic point process

Goreaud François, Loussier B., Ngo Bieng Marie-Ange, Allain Régis. 2004. Simulating realistic spatial structure for forest stands: a mimetic point process. In : Interdisciplinary Spatial Statistics Workshop 2004, 2-3 December 2004, Paris, France. s.l. : s.n., pp. 1-22.

Paper without proceedings
[img]
Preview
Published version - Anglais
Use under authorization by the author or CIRAD.
document_562310.pdf

Télécharger (629kB) | Preview

Abstract : The recent development of individual based and spatially explicit models raises the question of how to simulate relevant initial states for these models, and more particularly how to simulate realistic spatial structures. This is especially true for forest stand modelling. In this paper, we first review briefly the main simple steps in the simulation of forest stands with realistic spatial structure : (i) spatial structure analysis and ecological interpretation, (ii) expert simulation through classical point processes and ad hoc values of the parameters, and (iii) precise fitting of a spatial structure model using L(r) and L12(r) functions as criterion. We illustrate the advantages and limits of these steps on a real forest stand in mixed Oak - Scots pine forest near Orléans (France). We then propose a specific method - a mimetic point process - in order to simulate spatial structures similar to real patterns. This point process is based on a generalised Gibbs process, whose global cost function directly depends on the difference between the real measured values of L(r) and L12(r) functions, and the corresponding values computed on the simulated pattern at each iteration of the algorithm. We applied this method on a mixed Oak - Scots pine forest stand, and thus obtained an improved simulated pattern, that can be used as initial state for individual based forest growth models. We finally discuss the limits of this method, and its possible use to facilitate the valorisation of individual based models in the community of forest management. (Résumé d'auteur)

Mots-clés Agrovoc : forêt mélangée, Quercus petraea, Pinus sylvestris, Modèle de simulation, Distribution spatiale, Aménagement forestier

Mots-clés géographiques Agrovoc : France

Classification Agris : U10 - Mathematical and statistical methods
K01 - Forestry - General aspects

Auteurs et affiliations

  • Goreaud François, CEMAGREF (FRA)
  • Loussier B., CEMAGREF (FRA)
  • Ngo Bieng Marie-Ange, CEMAGREF (FRA)
  • Allain Régis, CEMAGREF (FRA)

Autres liens de la publication

Source : Cirad - Agritrop (https://agritrop.cirad.fr/562310/)

View Item (staff only) View Item (staff only)

[ Page générée et mise en cache le 2019-10-04 ]