About Forecaster Calculator


The Forecaster Calculator provides similar functionality to the Radiata Pine Calculator and the Douglas-fir Calculator. Growth and yield are predicted using the Forecaster desktop code-base, so outputs are consistent with the Forecaster desktop application.

The Forecaster Calculator is a simple forest simulation tool for radiata pine in New Zealand. For Douglas-fir there is no productivity surface and you will need a specific measurement event to calibrate the model. It simulates the tree growth and the log products produced at a single clearfell age on a per hectare basis.

Its target audience is primarily the small woodlot owner wishing to calculate a rough estimate of the volume and log product mix on a particular site at a particular age. Note that more accurate estimates of volume and log product mix will be produced by taking detailed plot based stem measurements pre-harvest, and inputting these measurements into a pre-harvest forest inventory tool (e.g. Cruiser or YTGen).

The Forecaster Calculator is built using the same set of models as the Forecaster desktop application which is widely used by the Forest Industry for yield table generation (i.e. log product volumes by age), regime evaluation and silvicultural scheduling.

The simplified interface of the Forecaster Calculator does not support the range of settings and options that the Forecaster desktop application provides. However, to drive Forecaster Calculator a series of “smart” defaults are used which are intended to reflect standard silvicultural regimes on normal farm/forestry sites, planted with commonly-used stock.

If you are interested in using Forecaster desktop, please see the Contact page...
The Forecaster Calculator is controlled in part by geographic location. The user is presented with a map of New Zealand and prompted to select a point. Forecaster Calculator has access to a number of geographic “surfaces” which predict some growth-related parameters based on location (latitude and longitude or easting and northing). These include:

  • Altitude (metres above sea level).
  • Site index. For radiata this is the average height of the 100 largest diameter stems per hectare at age 20 years. (Goulding 2005).
  • 300 Index. This is the stem volume mean annual increment (MAI) at age 30 years of a stand grown to a reference regime with final stocking of 300 stems/ha. (Kimberley et al. 2005).
  • Mean Annual Air Temperature. From Land Environments of New Zealand (LENZ) data.
Various “surfaces” have been developed by dividing the entire country into a grid with fixed size cells, and assigning values for these parameters to each cell. The values used by the Forecaster Calculator are calculated by averaging non-missing surface values within a circle (with an area of 30 hectares) of the selected location. Note that the site indices are applicable to tree stock with a genetic growth and form (GF) rating of 14.

Note that there is currently no surface for the 500 Index (which is required when simulating Douglas fir) so a value must be entered.
The models used are created using various national and regional defaults. The location of the site is used to determine the growth modelling region from the eight shown on the following surface:

Available regions are: Auckland Clays, North Island Sands, Central North Island, Hawkes Bay, Southern North Island, Nelson, Westland, Canterbury and Southland.

The stump height is set at 0.3 metres. This is the height above ground of the large-end diameter of the first log cut from the butt piece of each stem. The generic sweep and forking models are used, with parameters derived from the study by Snook (Snook 2013). No models are specified for carbon, wood density, stiffness, heartwood, BIX, MFA or spiral grain. The Blossim branch model is selected with parameters for large branch probability and scale described by Snook (op cit).

According to the growth modelling region a monthly growth adjustment (for within-year growth patterns) is set. The Blossim branch model is also informed of the growth modelling region and the regional drift factor for the 300 Index growth model is determined.

The radiata pine growth and yield models used are: 300 Index growth model, Mean-top-height / Age = 112, Tree volume table = 471, Tree taper table = 273, Tree felling breakage table = 1, DOS = DOS1999.
The silvicultural regime is specified by a list of commands, each of which comprises a condition, such as the age of the stand, and a series of one or more events such as plant, prune, thin or clear-fell. As the calculator requires only the basic parameters for these events the following defaults are assumed:

Planting – stock is GF 14.
Pruning – pruning is to a prune height with the larger, taller stems selected in preference, the ordering strictness set at 3 to allow for some spacing considerations in the stem selection. The number of stems to prune is specified.
Thinning – only waste thinning is supported. The residual stocking is specified and stems are selected from the smallest DBH and Height with an ordering strictness of 3.

To run a regime in Forecaster Calculator, a measurement of the crop is required as the start-point for the growth projections. If a measurement event is not supplied then one is generated using the 300 Index growth model by using the supplied site index and 300 index. The date of this generated measurement event is calculated from the plant year and crop age at the first silvicultural event.
Eight log product definitions are used in the Forecaster Calculator’s cutting strategy when simulating Radiata. These align with the set in the Radiata Pine Calculator spreadsheet.

Grade Pruned S1 S2 S3 L1 L2 L3 Pulp
Pruned TRUE
Length min (m) 4.2 4 4 4 4 4 4 4
Length max (m) 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5
SED min (mm) 300 400 300 200 400 300 200 100
Branch max (cm) 0 6 6 6 12 12 12 15
Price ($) 126 88 80 64 65 56 52 40
The Cutting strategy has a cutting cost of $1.00. This ensures a long log is preferred to two shorts of equal total value. The bucking mode is set to maximum value, that is, each stem piece is cut to produce the mix of logs that will yield the maximum value. This is a reproducible strategy that aligns with the predictions made by most pre-harvest inventory systems such as Cruiser or YTGen.
Structural (Framing) Regime
On an average site (site index 30, 300 index 26), this regime was constructed as follows:

Plant 1000 s/ha;
Thin at age 5.5 to 800 s/ha;
Thin at age 10 to 500 s/ha;
Clearfell at age 30.

The results in terms of the mix of log product volumes are shown below. Differences in the percent of TRV by log grade are listed.

Log Grade Forecaster
Calculator
Volume (m3/ha)
% of TRV PRad Spreadsheet
Calculator
Volume (m3/ha)
% of TRV Difference
S1 6.8 1 54.3 7 -6
S2 330.3 40 154.9 19 21
S3 193.1 23 212.0 26 -3
L1 0 0 29 4 -4
L2 127.1 15 86.5 10 5
L3 80 10 91.7 11 -1
Pulp 90.3 11 200.8 24 -13
TRV 827.6 100 829.0 100
TSV 979 976
TSV = total standing volume inside-bark at age 30 (m3/ha).
TRV = total recovered volume at age 30 (m3/ha).

The TSV values are very close as expected. The percentage recovered yield predicted by Forecaster Calculator was 85%, which is the average value suggested in the spreadsheet. The grade mix is similar apart from S2 and Pulp where Forecaster Calculator predicted more S2 volume and less pulp.

Clear-wood (Pruned) Regime
On the same site a pruned regime was constructed as follows:

Plant 1000 s/ha;
Prune 450 to 2.4m at age 5;
Thin at age 5.5 to 700 s/ha;
Prune 400 to 4.6m at age 7;
Prune 350 to 6m at age 10;
Thin at age 10 to 350 s/ha;
Clearfell at age 30.

Results are shown below.

Log Grade Forecaster
Calculator
Volume (m3/ha)
% of TRV PRad Spreadsheet
Calculator
Volume (m3/ha)
% of TRV Difference
Pruned 146.3 21 180.8 26 -5
S1 6.5 1 6.5 1 0
S2 230.5 33 72.6 10 23
S3 84 12 136.9 19 -7
L1 0 0 32.3 5 -5
L2 68.2 10 92.5 13 -3
L3 81.8 12 64.2 9 3
Pulp 76 11 118.8 17 -6
TRV 693.3 100 705.0 100
TSV 827.2 828.9
TSV = total standing volume inside-bark at age 30 (m3/ha).
TRV = total recovered volume at age 30 (m3/ha).

Forecaster Calculator predicted 84% recovery for this regime and, again, estimated more S2 volume than the spreadsheet.

Discussion
No attempt was made to tune the log mix proportions to any pre-conceived set. However such adjustment would be possible in Forecaster desktop by varying the parameters of the “quality” models, that is, branching, sweep and forking. For example, a higher probability of large branches will tend to reduce the proportion of S grades. Adjustments can also be made by varying the relative price /m3 of the log grades.
Kimberley, M.O.; West, G.G.; Dean, M.G.; Knowles, L.R. 2005: The 300 Index — a volume productivity index for radiata pine. New Zealand Journal of Forestry 50: 13–18.

Goulding, C.J. 2005: Measurement of trees, NZIF Forestry Handbook 4th Edition. NZ Institute of Forestry

Snook, J. 2013: Recommendations to Calibrate Branch Size, Sweep and Forking within FFR Forecaster. Scion Report.