Deforestation Carbon Stock Tables
Methods
Species and stand tending regimes
MAF requested estimates of forest carbon stocks by stand age class for various tree species and stand tending regimes, for each geographical (growth modelling) region in New Zealand. The site productivity levels to be used were provided by MAF, based on specified values of 300Index and/or Site Index. The species and regimes modelled are shown in Table 1.
Table 1: Species and associated regimes for modelling deforestation carbon stock tables
| Species | Management regime | |||||
|---|---|---|---|---|---|---|
| Rotation length (years) |
Planting (sph) |
Pruning | comments | Thinning*** | comments | |
| Radiata Pine | 50 y | 1250 | 600 sph to 2.2m | all pruning events scheduled on target green crown of 4m | To 600 sph | Thinned to waste after 1st pruning |
| 350 sph to 4m | ||||||
| 250 sph to 5.8m | To 250 sph | Thinned to waste after 3rd pruning | ||||
| Radiata Pine | 50 y | 1250 | None | to 370 sph | Thinned to waste at MTH of 12 m | |
| Douglas-fir | 80 y | 2000 | None | to 400 sph | Thinned to waste at MTH of 16 m | |
| Other Softwoods |
50 y* | 1250 | Averaged results based on both radiata regimes described above, weighted by area and halved (50%) | |||
| Hardwoods | 25 y** | 1030 | None | none | ||
* maximum rotation age assumed to be the same as for radiata pine
** the hardwood growth model is currently restricted to a maximum rotation age of 25 years
***to smoothen out the effect of thinnings on the total carbon pools, thinning events were stretched over 3 years, representing average regimes rather then a specific stand (see text below)
Index300 and Site Index used
MAF supplied 300Index and Site Index values to be used for radiata pine in each of the nine Growth Modelling Regions for predicting stem total volume inside bark yields and stand carbon stocks by stand age class. These index values were originally calculated using plot data from a series of replicated genetic gain trials provided by the Stand Growth Modelling Research cooperative and are tabulated in the Radiata Pine Calculator (MacLaren and Knowles 2005). As these index values were calculated using an earlier version of the 300 Index Growth Model, they were recalculated for the same Permanent Sample Plot (PSP) data using the most recent version of the 300 Index Model (Version 2.1) and national volume function. The values represent GF from 0 to 25.
Table 2: Radiata pine 300Index and Site Index values for Growth Modelling Regions.
| Growth Modelling region | Includes | 300Index | Site index |
|---|---|---|---|
| Auckland | Northland and Auckland regions, Franklin, Hauraki and Thames Coromandel Districts | 29.3 | 32.1 |
| Waikato Taupo | Waikato local governmental region excluding Franklin, Hauraki and Thames Coromandel Districts | 29.0 | 34.8 |
| Bay of Plenty | Bay of Plenty Region | 27.4 | 33.1 |
| Gisborne – East Coast | Gisborne Region | 32 | 33.9 |
| Hawkes Bay | Hawkes Bay, Taranaki, Manawatu-Wanganui and Wellington Regions | 32.9 | 31.7 |
| Nelson | Nelson and Marlborough Regions | 24.1 | 29.1 |
| Canterbury | Canterbury and West Coast Regions | 20.2 | 23.8 |
| Otago | Otago Regions | 26.5 | 25.4 |
| Southland | Southland Region | 28.5 | 27.4 |
Models used
Volume and carbon yield tables were developed using a combination of various stem volume growth models and the biomass partitioning model C_Change (Beets et al. 1999).
The 300 Index Growth model (Kimberley et al. 2005) was used for radiata pine, using a new national volume equation (Kimberley and Beets 2007), in combination with C_Change.
For Douglas-fir, the 500 Index Growth model (Knowles 2005) as implemented in the Douglas-fir Calculator (MacLaren and Knowles 2005), was used in combination with C_change.
For Hardwoods the Eucalyptus nitens Growth model (Candy 2002) was used in combination with C_change.
Other Softwoods carbon yield tables were assumed to be 50% of the area-weighted average carbon stocks obtained for the two radiata pine regimes already described. A reduction of 50% is used in the 1995 NEFD Yield table for other softwoods.
Wood densities
The mean outer wood density of radiata pine for each of the nine regions are based on the former NZ Forest Service Conservancies which overlap or contain more than one of the nine growth model regions (Table 3) as tabulated in Cown et al. 1991. The outerwood density of rings 21-25 was input into a wood density model for predicting the density of annual growth sheaths (Beets and Kimberley 2007).
Table 3: Basic density for ring group 21-25 years. Density figures are taken from Cown et al 1991, based on NZ Forest Service conservancies.
| Region | Growth Modelling Region | Basic Density for Ring Group 21 – 25 (kg/m3) | Based on NZ forest service conservancy |
|---|---|---|---|
| 1 | Auckland | 507 | Auckland |
| 2 | Waikato Taupo | 472 | Rotorua |
| 3 | Bay of Plenty | 472 | Rotorua |
| 4 | Gisborne-East Coast | 472 | Rotorua |
| 5 | Hawkes Bay | 442 | Wellington |
| 6 | Nelson | 471 | Nelson |
| 7 | Canterbury | 433 | Canterbury/Westland |
| 8 | Otago | 429 | Southland |
| 9 | Southland | 429 | Southland |
Wood density for Douglas fir was taken from the Douglas fir Calculator. Density was input by stand age class into the C_Change model.
The wood density of other softwoods was assumed to be the same as radiata pine in the same growth modelling region.
Wood density of hardwoods is based on Shelbourne et al 2002 representing density of Eucalyptus nitens as one of the more common eucalypt species likely to be deforested in New Zealand. Densities from a number of sites over New Zealand, were averaged to calculate a national density for Eucalyptus nitens. The effect of ring age on wood density variation from pith to bark were based on published studies (McKinley et al 2000).
Altitude and Latitude for modelling a regional radiata pine yield table
Altitude and latitude influence height growth and are used as additional input variables in the 300Index model. For each of the nine Growth Modelling Regions the geographical centre point was determined and this latitude was used as model input. The mean altitude for each Growth Modelling Region was obtained by averaging altitudes of forests in the region. The occurrence and distribution of exotic forests for the calculation of mean altitude were based on the Forest owner database Version 2004 (MAF website).
Age of pruning and thinning for radiata pine and Douglas-fir
For radiata pine the stand age when a specified pruning lift was undertaken varied by region, depending on the Site Index and 300Index (and to a lesser extent on altitude and latitude). The Radiata Pine Calculator was used to schedule pruning operations (stand age was rounded to nearest integer) to ensure that crown length averaged 4 m following each lift.
Radiata pine stands were scheduled for thinning when the stand was predicted to attain a specified Mean Top Height (MTH) using the Radiata Pine Calculator. For example, a target MTH of 12 m was specified for the radiata minimum tending regime (unpruned).
Thinning of Douglas-fir stands was scheduled using the Douglas-fir Calculator based on a MTH of 16 metres.
Further specifications for Douglas fir and other species
The Douglas-fir Calculator was used to develop a single national volume yield table for stands at least 7 years of age and older. In stands less than 7 years of age stem volumes were estimated using the following formulae based on Kimberley and Beets (2007)
AgeY = reference age
AgeX = current age
The national yield table for Douglas-fir was based on:
• a Site Index of 31.3 m
• a 500 Index of 18.4 m3/ha/year
• a latitude of 40 degrees
The Index500 and Site Index values are the default values in the Douglas-fir Calculator and were derived using all available PSP data for Douglas-fir in New Zealand. The same database was used to develop the 500 Index Growth Model (the growth model incorporated in the Douglas-fir Calculator).
The Eucalyptus nitens growth model was used to develop a national stem volume table for hardwoods. This model was developed using all available PSP data for Eucalyptus nitens in both New Zealand and Tasmania (Candy 2002). The national volume table assumed a site index of 28.6, a mean tree height (MH) of 13.3 m and a BA of 25.6 m2/ha at stand age 11 years.
For Other Softwoods, a national radiata pine volume table was developed from the area weighted average of the regional yield tables, and then reduced by 50%, to be consistent with the reduction of 50% that was used in the 1995 NEFD Yield table for other softwoods. Areas of forests within each region were taken from the NEFD 2006 and are shown in Table 4.
Table 4: Growth modelling region and area of exotic forests.
| Growth Modelling Region | Area (ha) | |
|---|---|---|
| 1 | Auckland | 240,831 |
| 2 | Waikato Taupo | 308,891 |
| 3 | Bay of Plenty | 210,690 |
| 4 | Gisborne-East Coast | 157,009 |
| 5 | Hawkes Bay | 349,639 |
| 6 | Nelson | 170,025 |
| 7 | Canterbury | 147,305 |
| 8 | Otago | 127,842 |
| 9 | Southland | 87,820 |
NEFD tables for comparison
The yield tables developed using the growth models were compared with the version of the NEFD yield tables supplied by MAF (Table 5). Graphs showing the modelled and NEFD yield curves by regime, grouped by Growth Modelling Region, are shown in an Appendix to this report.
Table 5: NEFD yield tables supplied by MAF
| NEFD Region | Version |
|---|---|
| Northland | 1995 |
| Auckland | 1995 |
| Central North Island | 1995 |
| Hawkes Bay | 1995 |
| East Cape | 1995 |
| Southern North Island | 1995 |
| Nelson Marlborough | 1995 |
| West Coast | 1995 |
| Canterbury | 1995 |
| Otago | 1995 |
| Southland | 1995 |
Scaling factor
A scaling factor was applied to adjust the modelled volume yield tables to the NEFD 1995 yield tables as requested by MAF. A fixed scaling factor was derived from the existing NEFD information on the average clearfell yield for radiata pine, based on the 2006 NEFD report. The average recovered volume at 1st April 2006 was 486 m3/ha at age 28.2 years. Assuming a mean recovery of 85%, his results in 486/0.85 = 572 m3/ha of total stem volume.
The comparison of this total stem volume at age 28 (572 m3/ha) with the modelled weighted average volume of 781 m3/ha (are weighted by area for pruned and unpruned regimes in each regions) gives a scaling factor of 26.8% (or 0.732). This factor was used to scale the modelled volume yield tables for radiata pine. The same factor was applied to Douglas-fir, other softwoods and hardwoods - because insufficient data exists from the NEFD regarding species other than radiata pine.
Thinning events and their representation in the Yield tables
Thinning affects the carbon allocation significantly, especially in the year that thinning takes place. This is evident through a reduction in tree volume and live carbon stocks in that year although the growth of the remaining trees ensures that carbon stocks recover quickly in subsequent years. Because the thinned trees are transferred to the dead biomass pools at time of thinning, the reduction in total carbon stocks is much less pronounced than the reduction in live carbon. It should be noted that C_Change predicts a slight reduction in total carbon at the point of thinning because fine roots associated with the thinned trees are allocated immediately to the soil carbon pool. Soil carbon is not included in these yield tables (Beets pers. comm.). To avoid the occurrence of this short term drop in carbon stocks we interpolated the yield table data before the thinning event and the following year. This results in a smoothing of the yield curve.
In the case of a very heavy thinning with a relatively slow growing tree species (e.g. Douglas-fir) or even in some cases with radiata pine, there can be a real drop in the stand total carbon in subsequent years. This occurs whenever the loss in carbon due to the decay of the large pool of dead material (from thinnings) exceeds the ability of the residual trees to sequester carbon. The decay rates for dead biomass of all tree species used in these tables are based on radiata pine.
The specified heavy thinning regimes used in this report caused a marked slow down in carbon sequestration, and in some cases, a temporary loss in total carbon. This occurred because the modelled yield tables were chosen to be representative of actual regimes. This reduction in carbon sequestration rate, although reflecting the nominated regime, could lead to inconsistent results when applying the tables for deforestation purposes.
To avoid such short term thinning effects, the specified stocking reduction following thinning was spread out over three years so that the ‘average’ stocking across a large number of stands is modelled rather than the actual stocking of any single stand. The resulting yield table will therefore not representative any individual stand, but should more accurately reflect the average volume and carbon stocks across a large number of stands.
Second rotation stands
Young second rotation stands show higher carbon stocks then first rotation stands (afforestation), because dead woody litter and fine woody litter residues are carried over from the previous rotation. Residues from the previous rotation were modelled assuming the same thinning and pruning regimes were applied in both rotations, and assuming first rotation stands were clearfelled at ages given in Table 6:
Table 6: Rotation age of first rotation stands used for modelling carbon stocks in 2nd Rotation stands
| rotation age of previous stand | |
|---|---|
| Radiata pine: | 28 years |
| Douglas-fir | 45 years |
| Hardwoods | 25 years |
| Other softwoods | 35 years |
Extraction of 85% of the stem volume following clearfelling was assumed.
Averaging Carbon stocks of radiata regimes for regions
MAF requested the averaging of the modelled radiata regimes for each region. The resulting “average” carbon stocks were then scaled as described above.
We averaged the modelled carbon stocks based on the pruned and unpruned radiata regimes for each region to derive “average” carbon stocks by using an area weighted averaging to account for differences in area of unpruned and pruned regimes in the modelled regions. The areas by regime (unpruned /pruned) were taken from the NEFD 2006 for the nine modelled regions and are shown in Table 7.
Table 7: Areas (hectares) for weighted averaging of unpruned and pruned regimes for region
| Growth Modelling Region | Unpruned (ha) | Pruned (ha) |
|---|---|---|
| Auckland | 113,415 | 125,096 |
| Waikato Taupo | 117,038 | 175,461 |
| Bay of Plenty | 41,715 | 151,218 |
| Gisborne-East Coast | 37,607 | 118,177 |
| Hawkes Bay | 98,371 | 233,286 |
| Nelson | 70,403 | 86,317 |
| Canterbury | 53,095 | 64,032 |
| Otago | 17,627 | 76,864 |
| Southland | 22,402 | 23,674 |
Contact for Enquiries
Sustainable Land Management and Climate Change
MAF
Pastoral House
25 The Terrace
PO Box 2526, Wellington
Tel: 0800 CLIMATE (254 628)
Contact this person
