Forest simulation models in France : main developments and challenges

1

COST ACTION FP0603 Forest models for research
and decision support in sustainable forest
management COUNTRY REPORTING TEMPLATE

• Forest simulation models in France main
  developments and challenges
• WG1
• J-D Bontemps, C Meredieu

1st Workshop and Management Committee
Meeting.Institute of Silviculture, BOKU.8-9 of
May 2008Vienna, Austria
2
Main features of French forests

• Forest cover (total/share) 15.7 million ha,
  28.6 of the territory (1,2)
• Growing stock, annual growth and cuts
• 2.4 billion m3, 103.1 million m3/yr, fellings
  around 61 million m3 /yr (1998-2002) including
  windfall volumes, felling rate 60 (1)
• Main species
• (by decreasing total growing stock) - sessile
  oak/Quercus petraea (Liebl.), pedunculate
  oak/Quercus robur (L.), common beech/Fagus
  sylvatica (L.), norway spruce/Picea abies
  (Karst.), maritime pine/Pinus pinaster (Ait.),
  silver fir/Abies alba (Mill.), Scots pine/Pinus
  sylvestris (L.) (1)
• Main non-wood products and services
• protection forests for erosion/landscapes (1/5
  of total area mainly in mountain regions French
  Alps Pyrennées), hunting activities (7-35 of
  income), recreation, water quality
  (unquantified), biodiversity (1 of total area)
• Main risks
• storm damage (regular dense high stands),
  fires (Mediterranean areas, between 10000 and
  70000 ha/yr)
• Management and silvicultural characteristics
• even-aged multi-purpose but production dominated
  forestry in plains (naturally regenerated
  hardwoods, planted softwoods), hardwoods
  rotation age 130 - 170 yr - high-dimension
  quality-wood targeted / softwoods rotation age
  40-80 yr, medium-dimension/quality wood
  targeted, short production cycles
• coniferous plantations as protection or
  production forests or pure/mixed
  coniferous-dominated close-to-nature forestry in
  mountain areas
• French Guyana ?

(1) IFN (2006). The French Forest, 141p. (2)
according to the FAO definition
3
Forest modelling approaches and trends

• Empirical models
• Main types of models developed (by order of
  recent importance)
• Distance-dependent tree (DDT) models
• Distance-independent tree (DIT) models
• Stand with DIT downscaling models
• Trends in modelling
• DDT model development (and gap model) for mixed
  and/or irregular stands
• Hybrid models explicit process incorporation
  in DDT models
• - light resource and competition
• - reproduction, regeneration mortality
• Coupling between empirical models
  quality/risk/economy/visualization modules
• Recent research is concentrating on
• Understanding of mixed-stands dynamics
• Knowledge integration (connexions of models to
  1- upstream (resource/climate/nutrition), 2-
  downstream (quality/risk/economy) environments

4
Forest modelling approaches and trends

• Mechanistic models
• Which exist ?
• - Castanea (Dufrêne et al, 2005, Ecological
  Modelling)
• for several species in pure-regular stands
• - GRAECO (Porté, 2001 Bosc et al., 2005 only
  French papers)
• for Martime pine in pure-regular stands
• - Hybrid models SAMSARA (DDT, Courbaud et al,
  2003) light resource
• Main features
• - ecosystem forest models with explicit
  connexion to the environment
• light, water, temperature
• - simulation of NPP
• - ongoing research on allocation processes

5
Modelling non-timber products and services

• Prediction of Dead Wood as an indicator of
  biodiversity
• Brin et al., 2008 FEM (in press)
• For Maritime pine stands, connecting with
  thinnings and clear cut
• Impact of species substitution on Carbon storage
• Vallet et al., 2008 FEM (in press)
• Substitution of a slow-growing hardwood species
  (Quercus petraea) by a fastgrowing
• conifer plantation (Pinus nigra subsp. laricio)

6
Models for predicting risk of hazards

• Models for predicting damages (non-exhaustive
  list) are developed as generic modules that can
  be coupled to existing GY empirical models
  (mechanistic models currently excluded)
• - Wind Damage
• - Biomechanics tree biomechanics for
  predicting wind damage in a forest stand (Ancelin
  al, 2004, Forest Ecology and Management)
• - Mechanical resistance with static winching
  tests (Cucchi al, 2004 2005, Forest Ecology
  and Management)
• - Rock fall Risk
• - RockforNET quantifying downstream rockfall
  risk in protection forests (Stoffel al, 2005,
  2006, Forest Ecology and Management)
• - Fire Risk
• - Fire Paradox wildland fire management by
  the wise-use of fire and post fire dynamics
  (http//www-capsis.cirad.fr/models ? Fire paradox)

7
Simulators and information systems

• Most French GrowthYield models are - or in the
  process of being - implemented on the CAPSIS
  plateform on a free-license simulator (see
  http//www-capsis.cirad.fr/ and next slide)
• Therefore, they are available for use by forest
  managers, and suppose exchanges between users and
  modellers. Rather used for building realistic
  sivlvicultural guidelines than locally optimizing
  a resource
• No process-based model is of current use for
  decision support
• Illustrations for empirical models
• Fagacees, PNN, Sylvestris, PP3 are models
  for regular stand and were used a for drawing
  silvicutural guides from the French Forest
  Service (Jarret, ONF, 2004 Sardin, ONF,
  2005...)
• Mountain Samsara (DDT models) have been at
  use for Alps silvicultural guide
• Eucalypt encloses a GIS-connexion for
  simulating extended resource in an area
• Maritime Pine models are able to use FNI plots
  to provide ressource evaluation in a regional
  simulator Sylvogène

8
Research highlight
CAPSIS Project http//www-capsis.cirad.fr/ Deve
lopment of an integrative simulation plateform
aimed at - integrating forest production and
dynamics models with consideration for ergonomy
and tool interactivity - developing generic
simulation tools useable for all related
modelling approaches (virtual thinning,
graphs...) - favouring connexions between others
tools (data bases, GIS, others software) -
intended for forest modellers, managers and
educational purposes ? More than 60 ongoing
projects
9
Future challenges

• - Knowledge integration in CAPSIS plateform
  (ergonomy, universality)
• - Connexion of models to wood quality/genetic
  improvement/risk/economy modeling extensions
• - Connexion of empirical models to environment in
  the context of climate change either
  statistical or process-based
• - Landscape simulator with GIS-connexion to test
  the effect of spatial effects edges,
  recruitment, harvested areas
• - Regional/ National simulator to provide
  resource information
• - Structure-function modeling linking
  architecture and functioning quantitative/qualit
  ative assessment of resource
• - For process-based models NPP allocation to
  tree/stand compartments, spatial upscaling using
  satellite data for calibration

10
Innovative references

• On coupling with risk modules
• - Ancelin, Courbaud, Fourcaud (2004).
  Development of an individual tree-based
  mechanical model to predict wind damage within
  forest stands, For Ecol Manage, 203101-121.
• On connecting empirical models to environment
  (coupling with GY simulators)
• - Seynave, Gégout, Hervé et al (2005). Picea
  abies site index prediction by environmental
  factors and understorey vegetation a two-scale
  approach based on survey databases. Can J For
  Res, 351669-1678
• On dynamics of heterogeneous stands (Guyana
  tropical forest)
• Picard, Bar-Hen, Franc (2001). Modelling forest
  dynamics with a combined matrix/individual-based
  model. For Sci, 48643-652
• On coupling gene fluxes and forest dynamic model
  
• - Dreyfus Ph. et al. (2005). Couplage de modèles
  de flux de gènes et de modèles de dynamique
  forestière. Un dialogue pour la diversité
  génétique - Actes du 5ème colloque national BRG,
  Lyon,
• On hybrid modeling
• - Courbaud, Coligny de, Cordonnier (2003).
  Simulating radiation distribution in a
  heterogenous Norway spruce forest on a slope. Ann
  For Sci, 1161-18.
• - Courbaud, Coligny de, Goreaud (submitted). An
  individual model of competition for light allows
  to simulate coherently the development patterns
  of dense monospecific forest stands, Can J For
  Res.
• On structure-function modeling
• - Yan, Kang, De Reffye, Dingkuhn (2004). A
  Dynamic, Architectural Plant Model Simulating
  Resource-dependent Growth. Ann Bot, 93591 - 602.