2 LandRCBM_split3pools
This documentation is work in progress. Potential discrepancies and omissions may exist for the time being. If you find any, please contact us here.
2.1 Overview
LandRCBM_split3pools translates forest landscape dynamics simulated from LandR compatible with the spadesCBM. The module provides the annual increments for the three above ground biomass pools: merchantable, foliage, and other. It does so for the yield tables which are used during the spinup event of spadesCBM, and for the annual cohort data which are used for the annual event of spadesCBM.
2.2 Inputs
The inputs of LandRCBM_split3pools are a combination of outputs from LandR, data to convert total above ground biomass into biomass of merchantable, foliage and other wood pools, and outputs in spadesCBM.
The default inputs are for the Western Boreal Regional Integrated Assessment (RIA) study area. Most inputs are expected to be provided by other modules (CBMdefault, Biomass_core, Biomass_yieldTables, Biomass_borealDataPrep, etc.).
| Name | Class | Description | Source |
|---|---|---|---|
| cohortData | Data table | Table with cohort-level attributes | Biomass_borealDataPrep |
| cbm_vars | list | List of 5 data tables: key, parameters, pools, flux, and state | CBM_core |
| pixelGroupMap | Spatraster | Pixel group map from LandR Groups pixels that share the same cohort composition | Biomass_borealDataPrep |
| rasterToMatch | SpatRaster | Template raster to use for simulations | User provided |
| standDT | Data table | Table of stand attributes. Stands can have 1 or more cohorts | CBM_dataPrep |
| studyArea | sfc | Polygon object that defines study area | User provided |
| table6 | Data table | Boudewyn table with proportion model parameters | National Forest Inventory |
| table7 | Data table | Boudewyn table with caps on proportion models | National Forest Inventory |
| tableMerchantability | Data table | Parameters for estimation of merchantable proportion of stemwood | Google Drive |
| yieldTablesCumulative | Data table | Yield tables intended to supply CBM spinup requirements | Biomass_yieldTables |
| yieldTablesId | Data table | Table spatially linking yield tables to pixels | Biomass_yieldTables |
2.3 Module functioning
2.3.1 Events
LandRCBM_split3pools has 7 events, 3 of which are plotting events. Saved plots can be found in the ~/LandRCBM/outputs/[simulation name]/figures/LandRCBM_split3pools directory.
2.3.1.1 splitInit
This initial event splits the total above ground biomass of the yield tables and the initial landscape into biomass of each of the above ground carbon pools (merchantable, foliage, and other).
2.3.1.2 postSpinupAdjustBiomass
After the CBM spinup, the biomass in the aboveground (merchantable, foliage, and other) and belowground (coarse roots and fine roots) live carbon pools will not match the observed data. This event re-initializes these pools to match observed initial aboveground biomass.
2.3.1.3 annualIncrements
In this event, cohortData from the current simulation year is compared with the previous year and the difference in biomass for each cohort is calculated and converted into increments of the three aboveground carbon pools.
2.3.1.4 prepareCBMvars
Each year, the object cbm_vars is updated to incorporate the calculated annual aboveground increments, handle dead organic matter cohorts (cohort without aboveground biomass), new cohorts, and disturbed cohorts.
2.3.1.5 summarizeAGBPools
Here, the above ground biomass in each above ground carbon pool is summarized for each species across the landscape at each time step.
2.3.1.6 plotYC
This event plots the yield curves for a number of randomly selected pixel groups. The yield curves describe how aboveground carbon pools change with stand age.
2.3.2 Estimating proportion of stemwood, bark, branches, and foliage biomass
In Boudewyn et al. (2007), total aboveground biomass is divided into biomass of stemwood, stem bark, branches, and foliage using the following equations:
\(p_{stewmwood} = \frac{1}{1 + e^{a1 + a2 \times vol + a3 \times lvol} + e^{b1 + b2 \times vol + b3 \times lvol} + e^{c1 + c2 \times vol + c3 \times lvol}}\) \(p_{bark} = \frac{e^{a1 + a2 \times vol + a3 \times lvol}}{1 + e^{a1 + a2 \times vol + a3 \times lvol} + e^{b1 + b2 \times vol + b3 \times lvol} + e^{c1 + c2 \times vol + c3 \times lvol}}\) \(p_{branches} = \frac{e^{b1 + b2 \times vol + b3 \times lvol}}{1 + e^{a1 + a2 \times vol + a3 \times lvol} + e^{b1 + b2 \times vol + b3 \times lvol} + e^{c1 + c2 \times vol + c3 \times lvol}}\) \(p_{foliage} = \frac{e^{c1 + c2 \times vol + c3 \times lvol}}{1 + e^{a1 + a2 \times vol + a3 \times lvol} + e^{b1 + b2 \times vol + b3 \times lvol} + e^{c1 + c2 \times vol + c3 \times lvol}}\)
Where:
- \(p_{stewmwood}\), \(p_{bark}\), \(p_{branches}\), and \(p_{foliage}\) are the proportion of aboveground biomass that is stemwood, stem bark, branches, and foliage respectively. These proportions sum to 1.
- \(vol\) and \(lvol\) are the gross merchantable volume per hectare and the log-transformed gross merchantable volume.
- \(a1\), \(a2\), \(a3\), \(b1\), \(b2\), \(b3\), \(c1\), \(c2\), and \(c3\) are model parameters fit separately by jurisdiction, ecozone, and lead tree species.
LandRCBM_split3pools uses an alternative set of equations with aboveground biomass per hectare (in tonnes) and the log-transformed aboveground biomass as the independent variables instead of \(vol\) and \(lvol\). The parameters for this alternative set of equations were estimated using the same approach and data used to calibrate the original equations. These new parameters are available on the National Forest Inventory website.
2.3.3 Estimating merchantable proportion of stemwood
Merchantable wood is estimated in Boudewyn et al. (2007) with the following equations:
\(nonmerchfactor = k + a * b_m^b\)
\(nonmerchfactor = b_{nm}/b_m\)
\(b_{nm} = b_m + b_n\)
Where:
\(b_n\) is the stemwood biomass of live, nonmerchantable trees
\(b_m\) is the stemwood biomass of merchantable sized live trees
\(b_{nm}\) is the total stemwood biomass in live trees
\(k\), \(a\), and \(b\) are model parameters fit separately by jurisdiction, ecozone, and lead tree species (Table 4 in Boudewyn et al. (2007)).
Because it is not possible to analytically infer new parameters for the equation above, they had to be inferred through simulations resulting in a new exponential model with this form to estimate the mechantable proportion of stemwood (\(b_m/b_{nm}\)):
\(b_m/b_{nm} = k - e^{-a(b_{nm} - b)}\)
Where:
- \(k\), \(a\), and \(b\) are model parameters inferred separately from those in Boudewyn et al. (2007)
If the proportion of stemwood that is merchantable (\(b_m/b_{nm}\)) exceeds the simulated inferred \(cap\), then the value will default to the \(cap\) proportion like so:
\[b_m/b_{nm} = \begin{cases} b_m/b_{nm}, & b_m/b_{nm} > \text{cap} \\ \text{cap}, & b_m/b_{nm} \le \text{cap} \end{cases}\]
The merchantable proportion of stemwood biomass is calculated in this way for each species present in the simulation. Below is an example of some species along with a comparison to the estimates using the original Boudewyn et al. (2007) equation.
figure 1: comparison between merchantable proportion curves from the fitted function used in Biomass_yieldTables and the equations from Boudewyn et al. (2007) for a number of species
2.3.4 Outputs
| Name | Class | Description |
|---|---|---|
| aboveGroundBiomass | Data table | Above ground biomass in each pool for each pixel and cohort |
| cbm_vars | List | List of 5 data tables: key, parameters, pools, flux, and state |
| cohortDT | Data table | Table of cohort-level attributes |
| gcIncrements | Data table | Growth increment matrix. Provides increments for each pool in each pixel and cohort |
| gcMeta | Data table | Growth curve metadata |
| summaryAGB | Data table | Sum of carbon mass for each species and above ground pools at each time step |