Model Scenarios
Forest owners can influence the long-term level of wood supply from their planted forests in two ways: the rate of harvesting (which can be expressed as the average age of clearfelling), and the level of new planting (assuming all clearfelled areas are replanted). Six scenarios based on different levels of these two factors have been modelled. Three scenarios are based on a target clearfell age and three are based on levels of new planting. The primary assumptions for these scenarios were outlined in the Overview and are described in more detail below.
Each wood supply region was modelled separately in developing these forecasts. The national forecasts are the sum of the 10 regional forecasts.
The forecast harvest volumes were constrained to follow actual harvest for the year ending 31 March 2000 and expected harvest for the year ending 31 March 2001.
Three scenarios based on target clearfell age have been modelled:
• base cut - target clearfell age for
radiata pine of 28 years
• early cut - target clearfell age for radiata pine of 25 years
• late cut - target clearfell age for radiata pine of 35 years.
These scenarios show the likely range of future harvest from the current forest estate. They assume that all harvested areas will be restocked but that no further new planting will occur. The target clearfell ages used in these three scenarios are identical to those used in the NEFD 1996 National and Regional Wood Supply Forecasts (Ministry of Forestry, 1996).
The base cut reflects current harvesting practice. The average clearfell age for radiata pine over the last five years is estimated as 27.8 years. The early and late cut scenarios provide forecast bounds within which the actual harvest is likely to fall unless there is a significant change from current forest management practice.
It was sometimes difficult for the models to achieve the target clearfell ages while at the same time meeting the non-declining yield requirements, particularly where the age class distribution within a region was highly uneven. This meant that the models had to allow harvesting at a wider range of ages. However, to encourage the models to harvest as closely as possible to the target clearfell ages, the FOLPI objective function was modified to include penalties (see Glossary) on harvesting at ages other than the target. The same scheme of penalties was used in all wood supply regions, and they were weighted to be low at ages close to the target clearfell age but very high for more distant ages.
The rotation ages for the species groups other than radiata pine were kept the same in each scenario. This was considered the most pragmatic approach given that the other species generally have a highly unbalanced age structure at the regional level and make up a relatively small proportion of the total resource. Table 10 summarises the target clearfell ages for each species group used in each of the clearfell age scenarios.
Table 10: Clearfell Ages for Wood Supply Scenarios
| Species group | Clearfell age (years) | ||
|---|---|---|---|
| Base cut target | Early cut target | Late cut target | |
| Radiata pine | 28 | 25 | 35 |
| Douglas-fir | 45 | 45 | 45 |
| Other softwoods | 40 | 40 | 40 |
| Hardwoods (short rotation) | 10 | 10 | 10 |
| Hardwoods (long rotation) 1 | 35 | 35 | 35 |
pNote:
1. The rotation age for the long rotation hardwoods in the Central North Island region
was set at 20 years to better reflect the current management regime for these species in
this region.
In most regions the average radiata clearfell age is relatively high for the initial two years, before moving towards the target. This is because the models maximise discounted volume production by cutting the oldest age classes first, leading to a higher average clearfell age for the initial periods of the model. The impact of this is partly reduced by the constraints that limit the annual harvest from areas aged more than sixty. After the initial two years there is a large increase in the projected harvest in most regions. This means that the proportion of the total volume being produced from the older age classes is significantly higher in the first two years of the models, where total harvest is constrained, than in the years that follow. As a result, the average clearfelling age quickly falls despite the fact that over-mature stands are still being harvested after the initial two years.
There was some difficulty in achieving the target clearfell ages as a result of the interaction between the age class distribution of a particular resource and the yield constraints imposed, particularly the non-declining yield constraint. Deviations from the target age are greatest where a large area is contained in a relatively narrow band of age classes and non-declining yield constraints are imposed, or where the area is evenly spread among all age classes up to an age which is significantly different from the target age being modelled. The average forecast clearfell age for the base cut is typically high initially as over-mature stands are harvested, before dropping to the target level. Any constraint that limits the volume which may be cut in a period to a level below what is available will force the average clearfell age to rise, as mature stands are held back.
The late cut scenario takes longer to stabilise at the target clearfell age for two main reasons. Firstly, an immediate change to an older clearfell age would make it impossible to maintain the current level of cut in most regions. This is most evident in the Central North Island. Secondly, it was necessary to maintain a higher harvest level in the medium term (by cutting stands younger than the target clearfell age) so that these stands would be available again in the next rotation, where there would otherwise be a shortfall in suitably aged stands to harvest. The alternative would have been to meet this shortfall by holding back stands well beyond the target clearfell age.
The Central North Island graphs (see page 69) clearly depict this effect. In the late cut scenario the target clearfell age is not met until around 2035. Until then it gradually rises towards age 35 as crop maturity builds up, while the total harvest fluctuates at around the 2000 level. There is not enough area in stands close to the target clearfell age to sustain a non-declining radiata pine yield based on the specified 2000 and 2001 harvest levels until 2035.
In some regions the target clearfell age for the early cut scenario could not be achieved while maintaining the non-declining yield constraints for radiata pine. This occurred in regions where the resource was mature or over-mature relative to the target clearfell age; examples are: Northland, Auckland, Central North Island and Nelson & Marlborough regions. In these cases the average clearfell age achieved was higher than the target.
The early cut effect described is best illustrated in the Northland region (see page 49). In the base cut model the harvest rises almost immediately to the long-run sustainable yield plateau. A target clearfell age three years younger (age 25) than that modelled in the base cut scenario cannot be achieved while sustaining non-declining total clearfell volume. While the model placed greater weight on non-declining yield, in reality it is possible in these situations that the actual cut may initially rise above the sustainable level as older stands are liquidated, and then reduce.
Three levels of national new planting have been modelled:
• plant 20 000 ha/year - new planting of
20,000 hectares/year
• plant 40 000 ha/year - new planting of 40,000 hectares/year
• plant 60 000 ha/year - new planting of 60,000 hectares/year.
These levels of new planting are held constant over the forecast horizon, starting 1 April 2002, and all three scenarios assume a target clearfell age for radiata pine of 28 years.
The three rates of new planting modelled are intended to provide a best estimate of future medium-term new planting along with a lower and an upper bound. The figures used are based on the results of a survey of new land planting intentions for the period 2000 to 2010 undertaken by the Ministry of Agriculture and Forestry in 1999. Graph 4 shows historical rates of new planting.
Graph 4: New Land Planted in Production Forest Since
1970
New planting was allocated into crop types based on the proportion of each existing crop type by region as at 1 April 1999.
In the new planting scenarios, a long-run sustainable plateau is never reached because of the continual new planting. There is less fluctuation in rotation age than in the base cut, because the additional area allows the models to achieve a non-declining yield without deviating from the target clearfell age. To compensate for this tendency to achieve the target rotation age at the expense of maximising volume, a constraint was added to the models to force the new planting scenarios to cut at least as much as the base cut in the first 10-15 years.
The national new planting levels have been apportioned into the 10 wood supply regions as shown in table 11.
Table 11: Allocation of National New Planting (Hectares/Year) into Wood Supply Regions
| Wood supply region | Proportion of total (%) | Plant 20 000 ha/yr | Plant 40 000 ha/yr | Plant 60 000 ha/yr |
||||
|---|---|---|---|---|---|---|---|---|
| Northland | 10.0 | 2 000 | 4 000 | 6 000 | ||||
| Auckland | 2.0 | 400 | 800 | 1 200 | ||||
| Central North Island | 15.0 | 3 000 | 6 000 | 9 000 | ||||
| East Coast | 15.0 | 3 000 | 6 000 | 9 000 | ||||
| Hawkes Bay | 10.0 | 2 000 | 4 000 | 6 000 | ||||
| Southern North Island | 15.0 | 3 000 | 6 000 | 9 000 | ||||
| Nelson & Marlborough | 7.0 | 1 400 | 2 800 | 4 200 | ||||
| West Coast | 1.0 | 200 | 400 | 600 | ||||
| Canterbury | 7.5 | 1 500 | 3 000 | 4 500 | ||||
| Otago & Southland | 17.5 | 3 500 | 7 000 | 10 500 | ||||
| Total | 100.0 | 20 000 | 40 000 | 60 000 | ||||
In deciding on the allocation of national new planting into the ten wood supply regions, the following factors were considered:
• recent new planting levels in each
region;
• availability of suitable land, eg:
| (i) |
in Northland, the East Coast and areas of the Southern North Island (Wairarapa and inland Wanganui) there are large areas of underdeveloped land and marginal farm land, which may be better suited to planted production forestry; | |
| (ii) |
in Otago & Southland and in some areas of the Southern North Island there are large areas of pastoral land suitable for planted production forestry; | |
| (iii) |
the Auckland region has a developed and highly profitable agricultural and horticultural land use, particularly in South Auckland and the Waikato, leaving little available land for planted production forestry; |
• the physical and economic suitability of land for planted production forestry.
