Methodology
The forest estate optimisation model FOLPI was used for the wood supply modelling. FOLPI is routinely used for forest planning in New Zealand. The modelling guidelines were based on those used for the 1996 NEFD wood supply forecasts. All models were run until 2097, however only the projections for the first forty years are presented in this report.
The objective function used was to maximise the discounted wood volume harvested subject to specified constraints. Total clearfell volume was discounted at 5 percent in all regions, and clearfell age penalties were built into the objective function (Clearfell Age Scenarios, page 25). Five percent was selected because it approximated the current annual increment (as a proportion of total volume) in radiata pine regimes nationally at the base target clearfell age of 28 years. However, the effect of the discount rate is strongly masked by clearfell age penalties and other model constraints.
Each species group was modelled independently under separate yield constraints. Radiata pine clearfell yields were constrained to be non-declining in each region, ie, the yield in any one year was required to be greater than or equal to the yield in the previous year. Douglas-fir yields were also constrained to be non-declining where a significant resource existed (more than 10,000 hectares), otherwise harvests were held generally constant. Since the intention for other softwoods and long rotation hardwoods was to cut out the mature stands as rapidly as possible and replant into radiata pine, non-declining yield constraints were not imposed. Instead, yields for other softwoods were generally held constant while the mature stands were cut and then allowed to fluctuate within limits constrained by volume.
In regions where there was a significant area of forest over 60 years old the harvest for these old stands was spread out over a ten to fifteen year period.
Production thinning volumes were excluded from the yield constraints, as the thinning yields obtained are considered to be indicative at best, particularly over the longer term (see Production Thinning on page 23). This enabled non-declining clearfell yields to be attained, to which thinning yields may be added separately.
The constraints pertaining to the clearfelling yields in each region and for each species group are summarised in table 9.
Table 9: Clearfell Yield Constraints (all volumes are m3/year)
| Wood supply region | Radiata pine |
Douglas-fir | Other softwoods | Hardwoods |
|---|---|---|---|---|
| Northland | NDY | no resource | 2002-2026 < 10 000; | < 40 000 |
| 2027 on < 20 000 | ||||
| Auckland | NDY | no constraints | < 20 000 | 2002-2019 < 10 000; |
| 2020 on < 20 000 | ||||
| Central North Island | NDY | NDY | < 50 000 | 2002 > 200 000; |
| 2002-2030 NDY (long); | ||||
| < 500 000 (short) | ||||
| East Coast | NDY | < 100 000 | 2002-2006 < 80 000; | < 15 000 |
| 2007 on < 30 000 | ||||
| Hawkes Bay | NDY | < 75 000 | < 20 000 | < 20 000 |
| Southern North Island | NDY | < 80 000 | < 40 000 | 2002-2019 < 15 000; |
| 2020 on < 40 000 | ||||
| Nelson & Marlborough | NDY | NDY | < 50 000 | no constraints |
| West Coast | NDY | < 20 000 | 2002-2024 < 10 000; | < 20 000 |
| 2025 on < 45 000 | ||||
| Canterbury | NDY | 2002-2036 < 150 000; | < 100 000 | < 10 000 |
| 2037 on NDY | ||||
| Otago & Southland | NDY | NDY | > 30 000 | < 80 000 (long); |
| NDY (short) |
Notes:
1. NDY is an abbreviation for non-declining yield. For the definition of non-declining
yield see appendix 1.
2. Unless otherwise stated all constraints are imposed from 2002. For 2000 and 2001 each
crop type was constrained to the expected level of harvest as described in Description of
Data - Planting and Harvesting Data.
The level of production thinning was not managed or constrained in the models. Production thinning yields are generated in the model with the movement of areas in the production thinning crop types into age classes where a production thinning yield was specified in the yield table. The production thinning wood supply over the period modelled therefore reflects:
• the thinning yield assumed in the yield tables
• the initial crop typing of the stands into production thinning crop types (note that the hardwood and other softwood species group crop types did not include production thinning). This affects the short to medium term and should give a reasonable indication of current production thinning intentions.
• the allocation of restocked and newly planted areas into production thinned crop types. As this allocation was based on current proportions rather than known intentions (see Restocking on page 24 and New Planting Scenarios on page 26) the yields arising are indicative only.
The production thinning yields presented in this report were smoothed after modelling using a two-year rolling average.
The forecasts assumed that all areas were replanted in the year following clearfelling. The crop types used for the replanting varied according to species:
| Radiata pine | Each crop type clearfelled was replanted into
the same crop type. However, where crop types were separated on the basis of year of planting, the old crop types were replanted into the corresponding "new" crop types. |
|
| Douglas-fir | Each crop type clearfelled was replanted into the same crop type. Two aggregated NEFD silvicultural crop types were used for Douglas-fir: with production thinning and without production thinning. | |
| Other softwoods | All clearfelled areas were replanted into
radiata pine. The area of softwoods replanted in radiata pine was split between the four NEFD silvicultural crop types in proportion to the national planted forest estate as at 1 April 1999. |
|
| Short rotation hardwoods | All areas clearfelled were replanted back into short rotation hardwoods. | |
| Long rotation hardwoods | All areas clearfelled were replanted in radiata pine. The area of long rotation hardwoods replanted in radiata pine was split between the four NEFD silvicultural crop types in proportion to the national planted forest estate as at 1 April 1999. |
