Transcript Multi-objective forest planning risk management

COST ACTION FP0603: Forest models for research and decision
support in sustainable forest management
Forest simulation models in Switzerland:
main developments and challenges
WG1
1st Workshop and Management Committee Meeting.
Institute of Silviculture, BOKU.
8-9 of May 2008
Vienna, Austria
Main features of Swiss forests
(BAFU, Steckbrief Schweizer Wald, 2nd Swiss National Forest Inventory, 1999)


Forest cover (total, share): 1.2 Mio ha, 30%
Timber

growing stock: 418 Mio m3
annual growth: 7.4 Mio m3/year

cuts: 5.7 Mio. m3/year


Main species:


Norway spruce (48%), Beech (17%), Fir (15%), Larch (5%), Pine (3.5%)
Main non-wood products and services:



Protection against rockfall and avalanches
CO2 sequestration
Biodiversity
•

the herb layer and of the landscape
Recreation and scenic beauty
• (walking, nordic ski)
• (typical image of wooded pastures in Swiss mountains)


In wooded pastures: forage, milk, meat
Regulation of water-household
Main features of Swiss forests

Main risks…

… against which the forests protect
• avalanche formation
• rockfall

…to forests
• Reduced protection function due to too old stands
• Climate change
• Inadequate species in already dry regions due to climate change
• Increased pest abundance (e.g. bark beetle)
• Changes in landscape structure due to segregation of land use and climate
change: separation of closed forests and open grasslands, increased
aggregation of land cover, decline of dominant species (Norway spruce and
larch)

Management and silvicultural characteristics:


Small clear cuts
Single tree felling (Plenterwirtschaft)
Forest modelling approaches and trends
Markov chain m.
PNV
Selected
tree species
EFISCEN
BIOME-BGC,
CLM,
LPJ-(GUESS)
Treeline
Forece (gap.m.)
ForClim (gap m.)
SEIB DGVM
Plant hydraulic
model
ForLand
Treeline
dynamics/
land use
DisCForM
LandClim
TreeMig
MASSIMO
WoodPam
MASSIMO
improved
Cost/benefit
protection
forest m.
ForClim
improved
MEPHYSTO
Mountland
Forest modelling approaches and trends
Empirical models

Approaches

Several static models for distribution of
• Potential natural vegetation
• Tree species
• Timberline position (Gehrig-Fasel, 2005)


Application of EFISCEN
MASSIMO (Kaufmann, 2001)
• Individual based, stochastic growth model
• NFI derived


Recent research is concentrating in:



Markov-Chain models
Recalibration of MASSIMO with latest NFI data (2004-2007)
 Growth function, harvesting probabilities, regeneration, mortality
Trends in modelling

Impact of climate change in MASSIMO on
•
•
•

growth function,
tree species composition
and mortality
Long-term harvesting potential (30-100 years)
Forest modelling approaches and trends
Mechanistic models

Approaches
• Population dynamical models
• Gap, distribution based models
• Ecophysiological models
• Plant water household model
• Applications of biogeochemical models and DGVMSs
• BIOME-BGC, LPJ, CLM
• Various landscape models
• Integrated models
• With disturbances
• Cost/Benefit
• Starting: with socioeconomy

Trends in modelling
• Integrated models
• Merging of approaches
Modelling non-timber products and services

Static models, ForClim, TreeMig



WoodPaM




Species distributions after climate change
Species suitabilities
Forage production available for livestock
Diversity indexes at patch and landscape scales
Landscape aggregation index
Planned models within MOUNTLAND


Diversity indexes at patch and landscape scales
Landscape aggregation index
Models for predicting risk of hazards


Protection forest model
LANDCLIM


Fire-forest dyn. interaction
Mountland model (Davos), starting

Interaction between forest dynamics and avalanche (risk)
Future challenges
Describe the main challenges modelers and modelling face in your
country so that can respond effectively to management or
scientific questions/problems in your country

Management issues:


Prediction of tree species composition and stand structure in forested areas
under various scenarios of management (including silvopastoral management)
and climate change (warming, episodic events)
Scientific issues:



Heterogeneity due to topography
Shifting mosaics in natural and silvopastoral systems (grazing ecology and
forest dynamics)
Consequences of the hierarchical organization of ecosystems
Innovative references








Bugmann, H.K.M., 1996. A simplified forest model to study species composition
along climate gradients. Ecology, 77: 2055-2074.
Gillet F. (in press). Modelling vegetation dynamics in heterogeneous pasturewoodland landscapes. Ecological Modelling.
Kaufmann, E., 2001. Prognosis and management scenarios. In: P. Brassel and H.
Lischke (Editors), Swiss National Forest Inventory: Methods and Models of the
Second Assessment. Swiss Federal Research Institute WSL, Birmensdorf, pp. 336.
Lischke, H., Löffler, T.J. and Fischlin, A., 1998. Aggregation of individual trees and
patches in forest succession models - Capturing variability with height structured
random dispersions. Theor. Popul. Biol., 54: 213-226.
Lischke, H., Zimmermann, N.E., Bolliger, J., Rickebusch, S. and Löffler, T.J., 2006.
TreeMig: A forest-landscape model for simulating spatio-temporal patterns from
stand to landscape scale. Ecol. Model., 199: 409-420.
Rickebusch, S., Lischke, H., Bugmann, H., Guisan, A. and Zimmermann, N.E.,
2007. Understanding the low-temperature limitations to forest growth through
calibration of a forest dynamics model with tree-ring data. For. Ecol. Manage.,
246: 251-263.
Schumacher, S., Bugmann, H. and Mladenoff, D.J., 2004. Improving the
formulation of tree growth and succession in a spatially explicit landscape model.
Ecol. Model., 180: 175-194.
Zweifel, R., Zimmermann, L. and Newbery, D.M., 2005. Modeling tree water deficit
from microclimate: an approach to quantifying drought stress. Tree Physiol., 25:
147-156.
MASSIMO (Kaufmann 2001):
(Management Scenario-Simulation Model)
Model type
NFI 1 (1983-85) & 2 (1993-95)
Evaluation

Growth-function (non-linear
regression function estimating
individual basal-area increment)
Validation data

Forest Inventory Liechtenstein
600
400

200
Calibration data
Accuracy: -5.44%
0

Empirical, stochastic & dynamic,
individual-based, distance
independent model
4 Modules: Growth, mortality,
harvesting, and regeneration
Number of trees

-20
-10
0
10
Predicted basal area increment minus observed [cm]
Thürig et al. (2005)
TREEMIG: a spatio-temporal forest
model (Lischke et al. 2006, www.wsl.ch/projects/TreeMig/treemig.html
Local forest dynamics
Seed production
Implemented in space
Seed dispersal
-
Density regulation
WOODPAM: (Gillet, in press)
Vegetation dynamics in pasture-woodland landscapes under climate
change- towards a modeling tool for active adaptive management of
silvopastoral systems

Goal



Geographic area and scale




To develop a decision tool for active adaptive
management of silvopastoral systems
Spatially explicit dynamic mosaic model
suitable to simulate various scenarios of
climate change and land use
Jura, Alps
Extent: local landscapes (up to several km2)
Grain: 625 m2 (25 m x 25 m square cells)
Modeling approach



3 hierarchical levels (cell, management unit,
landscape) and 6 submodels (wood, herb,
cattle, soil, management, climate)
Coupling of population, community and
ecosystem processes
Focus on vegetation-cattle interactions under
climate and management constraints
Warming?
Storms?
Fires?
ForClim Improvement :
Bridging the gap between forest growth and forest succession
models
Goal:


Build upon a climate-sensitive forest
succession model to Increase local
precision,
thus bridging the gap between forest
growth (local precision) and forest
succession (wide range of applicability)
models
Approach:



Systematic model evaluation against
empirical data (yield trials etc.) and
systematic model-model comparisons
Model improvements (growth,
regeneration)
Model applications to study climate
change impacts on protection forests in
the Alps & other European mountain
ranges
Geographic


area and scale:
Alps, other European mountain ranges
(TBD)
Stand scale assessments
Left:
Simulated (filled bars) vs. measured (semi-transparent bars)
stand structure at the site Niederhünigen after 54 simulation
years.
Right:
Simulated equilibrium species basal area for the Swiss sites
Grande Dixence (cold), Adelboden (cool-wet) and the
eastern German site Schwerin (dry and warm)
Protection forest model
The protection forest model
combines a markov chain approach
for simulating forest
dynamics with risk assessment and cost-benefit
analysis.
integrates ecological, technical and
economic aspects of
protection forest
management.
can be used to comparatively
evaluate the long-term
effects of management
strategies (e.g., thinning,
planting, salvage harvesting,
construction of defensive
structures).
Risk reduction
= benefits
Management
Costs
Gap model ForClim
(Bugmann, 1996)

Concept of individualistic,
cyclical succession on small
patches
(H. Gleason)

Quantitative description of
tree population dynamics:
“gap“ models
(D. Botkin, H. Shugart)
Landscape model LandClim
(Schumacher et al, 2004)



Spatially explicit
(grid cells, ca. 30x30 m)
Dynamic
Modeling of succession



N
Dynamics of cohorts of trees:
establishment, growth, mortality
based on biomass and tree
number per cohort
Modeling of ‘disturbances’




DIVERSITY
Fire
Windthrow
Management
1km
Modeled processes sensitive to
climate
Schumacher et al. (2004, 2006)
Improved landscape model MEPHYSTO:
Merging empirical, ecophysiological and spatio-temporal population
dynamics forest models
Goal:
Spatial forest model
stand-size grain
 to be applied on large areas
 for assessment of, e.g.,
climate change or
management effects on
forest functions

Model
approach:
Combination
of
• large scale
ecophysiological,
• forest growth,
• tree species migration
models
Dynamic,
spatio-temporal,
process based
Focus on natural processes
Management included via
scenarios