Growth Model Users Group Growth Model Run-off II

1
Growth Model Users GroupGrowth Model Run-off II
2
History

• GMRO I was conducted in 2002.
• The objective was to test the effect of growth
  model and user combinations on the projections of
  three predominately Douglas-fir stands.
• 7 models CRYPTOS, DFSIM, FPS, FVS, ORGANON, SPS,
  and TADAM were tested, using 11 different
  versions.
• 16 users contributed results.

3
History

• Some observations on the results
• Region effects are very large. FVS in
  particular can more than double volume between
  regions.
• There was significant variation among users
  running the same model and version.
• All models maintain a remarkably consistent
  relationship between stand height and volume
  (Eichorns Rule).
• CRYPTOS (an out-of-growing-region model)
  projections were very similar to northern growth
  models.

4
GMRO II

• A proposal to examine common Douglas-fir growth
  model performance for the following scenarios
• Commercial thinning
• Nitrogen fertilization
• Thinning and fertilization

5
Why Do This?

• GMUG Poll indicated that most users rely on their
  growth models for both biologic and economic
  evaluation of treatment effects.
• There is a long history of thinning studies in
  Douglas-fir and the effects of thinning are
  thought to be understood.
• Yet, thinning effects in many models require
  modifiers to component growth equations,
  indicating that the empirical models are not
  sufficient as-is to predict post-thinning growth.
• N fertilization also has a long history in the
  PNW. Region-wide response has been well
  documented in the literature.
• However, analyses of response at the tree level
  (required to incorporate fertilization into tree
  list models) has not been well documented. The
  translation of increased nutrient availability to
  growth in tree dimensions is not well modeled
  empirically.
• Both of these treatments leave a lot of room for
  the modeler to define response relationships.
  Their decisions may have important impacts on
  growth yield.

6
Questions

• What models should be tested?
• How many stands should we project (variation over
  site quality, density)?
• What growth reporting intervals should we use
  and how long should the projection run?
• Who should run the projections (users, modelers,
  both)?
• What should be compared (volume, height, Dq,
  basal area, crown ratio, trees per acre, diameter
  distributions)?

7
Questions

• Treatments
• Thinning
• What d/D?
• What thinning intensity (single, multiple)?
• Should we specify the thinned trees or leave it
  to the person running the model?
• Fertilization
• What dosage (200 lbs N/acre)?
• How many applications?
• What re-application interval?

8
Incentive?