1. 3-PG The Use of Physiological Principles in Predicting Forest Growth
Landsberg and Waring 1997
-The model is meant to be an easy-to-use tool for forest managers in simple forested ecosystems and can be run as an Excel spreadsheet with minimal training.

2. Uniform foliage distribution across stand
Objective
3-PG is a simple process-based model that simulates forest growth based on an understanding of physiological principles that predict tree growth
Key Assumptions
Uniform foliage distribution across stand
Single species (but see 3-PGmix, Forrester and Tang 2016)
Constant ratio of NPP to GPP (a constant fraction of GPP is respired)
Precipitation is evenly distributed over a month
Temporal & Spatial Scale
Monthly or annual time step
Typically stand-level, but can be scaled up to model forest growth across landscapes (3-PG(S); Coops et al. 1998)
-Very simply, the model calculates PAR absorbed by forest canopies and converts it to biomass production.

3. Input Drivers Key Outputs Incoming shortwave radiation (Фs)
Monthly precipitation (total)
Monthly mean daytime vapor pressure deficit (D)
Number of frost days per month
Initial stem, foliage, root biomass
Initial stand age, density
Soil type, maximum available soil water
Site fertility rating (nutrition)
Latitude
-Takes incoming solar radiation and converts this to “utilizable aPAR” which is basically saying that in the conversion of radiation to GPP, aPAR is modified by factors like high VPD, soil drought, sub-freezing temperatures, and stand age.
Figure 1. Simplified flow diagram of 3-PG model.
ФAPAR = Utilized portion of absorbed PAR.
f = A modifier. The conversion of radiation to GPP is modified by the effects of nutrition, soil drought, VPD, sub-freezing temperatures and stand age
(figure from
Key Outputs
Net primary production (NPP)
Stem, foliage, root biomass
Leaf area index (LAI)
Stand transpiration
Water use efficiency

4. Testing & Validation
“…no other models with comparable simplicity could be found that have been so rigorously tested in terms of observed-predicted comparisons…in so many different types of forests.” (Forrester and Tang 2016) A B
-This model has been tested and validated in stands from a wide variety of climates and with a wide range of species.
-This example is from a Loblolly pine stand in North Carolina and shows good agreement between observed and modeled values for two of the model outputs (LAI and DBH).
Figure 2. Observed and modeled A) LAI and B) DBH for fertilized (circles) and control (squares) stands of Loblolly pine in North Carolina. (Landsberg et al. 2000)

5. Examples of model use
Figure 3. Impact of climatic modifiers on photosynthesis for lodgepole pine. a) Soil water storage, b) Evaporative demand (VPD), c) Temperature, d) Frost. 1 = optimum conditions for photosynthesis and 0 = unfavorable. (Coops and Waring 2011)
-This is an example that is part of a study aimed at determining future presence of Lodgepole pine in the northwest.
-The first step of the study was to determine the extent to which specific climate/site conditions supported (or inhibited) NPP across its regional range, and that is what is shown in this figure.
-They then entered these variables into a decision-tree model for presence/absence predictions under current climate. Finally, they projected climate conditions through the end of the 21st century using a climate model (CGCM2). (They projected that by the end of the 21st century LP’s range will have significantly contracted)

6. Examples of model use: Applying 3-PG to mixed species stands
Figure 5. Using 3-PGmix to model how two species respond to thinning. C. sclerophylla thinned at year 10 and 15. (Forrester and Tang 2016).
-3PGmix: modifying 3PG so it can be used to quantify growth for individual species in mixed-species forests, and how they respond to disturbances (e.g. thinning).
-subtropical forests in China (mixed coniferous-deciduous forests)
-changes to the model: 1) a new light absorption model that includes multiple canopy layers, 2) considers vertical gradients in canopy microclimate, 3) accounts for proportions of each species, 4) includes parameters specific to deciduous species (e.g. leaf loss month).
Figure 4. Observed and modeled foliage mass (WF) and basal area for monocultures (c, e) and mixed species stands (d, f). Solid lines are 1:1, dashed lines are fitted to the data (Forrester and Tang 2016).

7. 1 2 Model Test Run: Parameterization
Species-specific parameterization necessary prior to model execution:
Biomass partitioning
Modifiers (temperature, frost, soil water, atmospheric CO2, site fertility, stand age)
Self-thinning
Canopy structure and processes (quantum efficiency, NPP/GPP, conductance) 2
Climate data, solar radiation inputs for period of interest
Pinus contorta values from Coops and Waring 2011.

8. Model Test Run Run Types: Single site Sensitivity analysis Multi-site
-sensitivity analysis: stand growth is simulated for a variety of site or climatic factors (e.g. available soil water, atmospheric CO2, site fertility).
-multi-site run: stand growth at a series of sites is simulated.

9. Model Test Run: Results
Run type: single site
Run type: sensitivity analysis