Overview of the
Deep Red Myrtle Project
6 September 1999.Contents

1. Introduction 1
2. Definitions and focus 1
3. Management framework 2
4. Design of the resource review 3
5. What is deep red myrtle (DRM)? 4
6. Myrtle forest and the DRM resource 5
7. What is the loggable area of DRM-rich forest outside the
Corridor? 6
8. How much myrtle sawlog is recoverable from DRM-rich forest? 8
9. How much will it cost to harvest the recoverable volume? 9
Appendix 1. Digital colour analysis 11
Appendix 2. Stand structure from assessment plots 14
Appendix 3. GIS details 16
Appendix 4. Sources for recovery data 24
Appendix 5. Log sizes 25
Appendix 6. Example of recoverable volume estimation 28
Appendix 7. Example of harvest cost estimation 30.1
1. Introduction
Myrtle (Nothofagus cunninghamii) is the most widespread and abundant tree species in
Tasmanian rainforest. On most sites the timber is straw to pale pink in colour, but 'deep red
myrtle' occurs in some areas. Deep red myrtle (DRM) is preferred by veneer and sawnwood
buyers and attracts a premium price. In the 1997 Tasmanian Regional Forest Agreement
(RFA), Commonwealth and State Governments acknowledged the importance of DRM in
the marketplace and made special provision for its supply. Under RFA clause 55 in the
'Public land' section,
'The Parties agree that:
a) during the first 4 years of this Agreement, the State will review its resource estimates for deep
red myrtle available for supply to the furniture and craft industries, in terms of volume, quality
and economic accessibility, and will publish a report of the findings; and
b) the State will arrange for the review described at (a) above to be independently audited by an
auditor agreed to by the Parties, and for a report by that auditor to be published;
c) the further management of the Savage River Pipeline corridor [a myrtle-rich area in
northwest Tasmania traversed by the Savage River Pipeline Road] will be considered by the
State prior to the first five yearly review of this Agreement in the light of the report and audit
described at sub-clauses (a) and (b) above; and
d) if the resource review and audit confirm the availability, outside the Savage River Pipeline
corridor, of adequate resource of acceptable quality and economic accessibility, to maintain a
supply of at least 4,500 cubic metres per year of deep red myrtle, for the remainder of the term of
the Agreement, then harvesting and associated roading within the area will be further postponed
for that period; and
e) in the alternative, the area will be further considered by the State to ensure the availability of
deep red myrtle for the period.'
The DRM resource review required under RFA clause 55a is being carried out by Forestry
Tasmania (FT). Dr Bob Mesibov is the project officer, with supervision by a steering
committee consisting of Dr Hans Drielsma (General Manager, Forest Management; FT),
Paul Smith (Regional Forester, North; FT), Ian Whyte (Chief Executive, Forest Industries
Association of Tasmania Ltd) , Glenn Britton (Managing Director, Britton Bros. P/L)
and Mike Peterson (Senior Forester (Special Species Timber), FT).
This document explains how the DRM resource is being reviewed and summarises results
from the project's first year.
2. Definitions and focus
Throughout this overview we use 'DRM' (deep red myrtle) to refer to a particular kind of
myrtle timber. In section 5 we summarise what we know about DRM's properties.
No formal, plot-based assessment of the DRM resource has ever been carried out, and we
will not be attempting one now. Deep redness is a feature of myrtle timber, not of myrtle
forest. The only way we could objectively estimate the size of the standing DRM resource
would be to fall or take cores from a large number of myrtle trees in undisturbed forest. The
cost would be very high and we would be exposing large areas of forest to an increased risk
of death from myrtle wilt disease.
To locate DRM we are relying instead on the combined long-term experience of foresters,
loggers and sawmillers. All agree that DRM is largely restricted to high-fertility sites, which
in northwest Tasmania are underlain by Tertiary basalt and some other igneous rock types..2
DRM outside northwest Tasmania is rare, but uniformly pink timber has been recovered
from high-country rainforest in the Northeast, e.g. from Tertiary basalt near Weldborough.
We will therefore be using location and geology as indicators of DRM in myrtle forest. For the
purposes of this review, we define 'DRM-rich forest' as myrtle forest on igneous bedrock in
northwest and northeast Tasmania (see Appendix 3 for a more exact definition). We will be
studying the myrtle sawlog resource in this forest and no others. The consequences of
excluding other myrtle forests from the review are considered in section 3.
DRM-rich forest occurs on both private and public land in Tasmania. We are excluding
private DRM-rich forest from the resource review because the State has no direct control
over sawlog supply from private land, and because the commitment in the RFA for a review
was specifically for public land.
3. Management framework
Myrtle sawlogs are supplied to the timber industry from selective logging of old-growth
rainforest, from clearfall harvest of mixed eucalypt forest (old-growth eucalypt over
rainforest), from selective blackwood and Huon pine logging and from conversion of native
forest to plantation. The various harvests are managed differently and have somewhat
different yields.
Rainforest
Selective logging for myrtle in rainforest in State forest is constrained by policy and
silviculture.
Forestry Tasmania's draft rainforest policy requires that rainforest patches greater than 10 ha
which exhibit negligible unnatural disturbance will be retained as rainforest wherever
possible. Such patches can be selectively logged, but only in accordance with published
silvicultural prescriptions. The draft rainforest policy also restricts selective harvesting to
Special Management Zones for special species timber, i.e. to areas which are managed in the
long term as rainforest. All 185 coupes so far identified as potential near-term sources of
DRM in this review (see section 7) are designated 'STM' for special timbers management in
the coupe databases used by FT forest planners.
Silvicultural prescriptions 1 for selective rainforest harvests are based on trials carried out in
the 1970s and 1980s. The trials demonstrated that myrtle forest on fertile sites can be
successfully regenerated after logging. Myrtle seedtrees are left at a minimum spacing of
15-20 m, and non-sawlog trees (myrtle and other species) are retained wherever possible to
provide seed and shelter for regeneration. All patches of existing regeneration and advance
growth are also retained as they represent many valuable years of seedling establishment
and growth.
DRM-rich forest has been the principal source of large myrtle sawlogs in recent years
(Table 1). Loggers look for the largest stems in such forest because red heartwood increases
in extent (i.e., in its proportion of a stem's cross-section) and deepens in colour as a myrtle
gets older. Big logs also have larger volumes of heartwood free of rotten core, which is
common in myrtle. For these reasons, harvests from DRM-rich forest are relatively rich in
category 4 myrtle sawlogs, i.e. logs with a mid-diameter >75 cm (Table 1).
Other sources
Rainforest sawlogs from mixed eucalypt, blackwood, Huon pine and conversion harvests
are referred to as 'arisings'. As Table 1 indicates, myrtle arisings are relatively poor in large-diameter
logs. No investigation has been carried out of colour in myrtle arisings, but
industry advice to FT is that run-of-the-bush myrtle from areas outside DRM-rich forest is
1 Jennings, S. and Hickey, J. (1998) Rainforest Silviculture. Native Forest Silviculture Technical Bulletin No. 9.
Hobart: Forestry Tasmania..3
generally pale-coloured, regardless of log size.
There is another reason to ignore arisings in the DRM resource review. Myrtle sawlog
volume from areas outside DRM-rich forest has averaged 900 m 3 over the past six years
(Table 1) and is unlikely to increase substantially during the life of the RFA. In comparison,
38 ha of DRM-rich forest in a single coupe, BO201B, have so far yielded 570 m 3 of category 4
and utility grade myrtle sawlog, with another third of the area still to be logged. DRM-rich
forest is a far more concentrated and reliable source of DRM than the other sources of myrtle
sawlog scattered around Tasmania.
Table 1. Myrtle sawlog volumes (m 3 ) from public land, 1993/94 to 1998/99. Source: FT log
delivery docket records.
Cat. 4 1 Utility 1 Cat. 4 proportion
Selective 2 2251 1329 0.63
Arisings 2 1558 3855 0.29
1 Category 4 sawlogs are currently >75 cm in mid-diameter, utility sawlogs are <75 cm.
2 'Selective' loggings were in DRM-rich forest; 'arisings' come from a range of harvests (see text)
Conservation
Conservation is an important part of rainforest management. DRM-rich forest is included in
'callidendrous and thamnic rainforest on fertile sites', which is one of the forest communities
considered for reservation under the Tasmanian RFA. Old-growth forest of this type is
estimated to occupy 159 650 ha on all land tenures 1 . Existing and post-RFA reserves on
public land (preliminary 1997 boundaries) contained 93 870 ha, or 59% of the old-growth
total.
Conservation of DRM-rich forest in northwest Tasmania has been examined more closely as
part of this review (see Appendix 3). Roughly 17 200 ha of DRM-rich forest are in new and
existing reserves, and more than 90% of the reserved forest is old-growth. Almost all the
ca. 27 000 ha of non-reserved DRM-rich forest on public land in the Northwest is managed
by Forestry Tasmania. About 14 600 ha of this non-reserved forest are in coupes which have
been identified as potentially suitable for DRM production, both inside and outside the
Savage River Pipeline Corridor (see Appendix 3). Most of the remainder is informally
reserved.
Most of the DRM-rich forest on private land is in the Hampshire and Surrey Hills blocks
owned by North Forests. The company maintains private forest reserves which account for
ca. 800 ha of the DRM-rich forest on its Northwest properties (pers. comm., J.L. Wilson, North
Forests).
4. Design of the resource review
We want to answer the following question: During the life of the RFA, how much deep red myrtle
is available for supply to the furniture and craft industries from areas outside the Savage River
Pipeline Corridor, and at what cost is it available?
Our plan is to catalog the available forest coupes in which DRM is likely to occur. We will
then estimate how much myrtle sawlog can be recovered from each coupe, and how much it
will cost to harvest that volume. Given certain assumptions (discussed in later sections), we
1 From Table 1. Reservation levels of Forest Communities in the CAR Reserve System on Public Land.
In: Tasmanian Regional Forest Agreement Between The Commonwealth Of Australia & The State Of
Tasmania, November 1997. Attachment 6. The Comprehensive, Adequate and Representative Reserve System
on Public Land..4
believe that our area estimates for DRM-rich forest will be reasonably accurate. There will be
uncertainties about recovery estimates, but we hope to 'bracket' these uncertainties by
examining the likely sources of natural variation in yield per hectare.
We will be estimating recoverable volume, rather than standing volume, because past rainforest
assessments have shown that the two are statistically not well correlated. Good-looking trees
may contain no sawlog whatsoever, and good myrtle sawlogs have been recovered from
trees that look, from the outside, like decayed and defective rejects. Recoverable volume
estimates will therefore be based on yields from past and current logging operations.
Some coupes will be more expensive to log than others, and overall harvest cost will vary
with the sequence of coupe harvests and with the particular roading plans proposed for each
sequence. Costs are also affected by pulpwood sales. In general, it is cheaper to harvest
myrtle sawlog if myrtle pulpwood can be harvested at the same time, but there are
silvicultural limits on pulpwood removal. We will consider both high and low levels of
permissible pulpwood recovery in estimating costs.
The DRM resource review will not be a simple exercise, but we hope to make its
complexities understandable to general readers of the final report. Clearly, we will not be
able to conclude our review with a statement that there are X cubic metres of myrtle sawlog
available and it will cost Y dollars to harvest them. Instead, we plan to list a variety of
logging plans in a table. Next to each plan we will set out its associated costs and recoveries,
given specified (and varying) assumptions. Earlier sections of the final report will detail the
uncertainties referred to above, and will show how they are likely to affect the cost and
recovery estimates.
The summary table will also be a convenient starting point for discussion of the question
implied in RFA Clause 55d: Is enough myrtle sawlog volume available at an acceptable cost from
DRM-rich forest outside the Pipeline Corridor to allow a cut of 4500 m 3 /yr for the life of the RFA?
Most of the information going into the resource review has already been gathered in the
form of GIS files, assessment results and logging records. A limited amount of information
will come from harvests during the review, but these data will be important in establishing
current logging costs and in learning about recoveries from different forest types and areas.
Much of the time devoted to this review is being spent in compiling a list of available coupes
and evaluating what is known about the myrtle forest they contain. An 'exhaustive'
approach of this kind is justified because the total forest area involved is not large, and
because RFA funding has made possible a close study of a particularly valuable timber
resource.
5. What is deep red myrtle (DRM)?
There are two varieties of Tasmanian myrtle, the red and the white. The wood of the red variety is
mostly bright pink in colour, and often beautifully marked. It is close-grained and tough. It is an
attractive, sound, mild-working wood, easily seasoned, and is capable of a high polish…The Light
Myrtle is a variety of the same species, not as close-grained as the dark nor of so high a specific
gravity. In colour the wood is a brownish grey…
Lewin, D.W. (1906) The Eucalypti Hardwood Timbers of Tasmania and the Tasmanian Ornamental
and Softwood Timbers. Gray Brothers, Hobart (pp. 51-52)
…as a first step, myrtle might be sorted into the two classes commonly recognised in the trade as
"white" myrtle and "red" myrtle, and each class dried separately. Such a method of classification
has been adopted in the later tests made at the laboratory, the classes being more fully described as
follows:-
"White" Myrtle.-Straw coloured to pale pink and comparatively soft.
"Red" Myrtle.-Pink to red or red-brown and comparatively hard.
Greenhill, W.L. and Thomas, A.J. (1937) A Guide to the Seasoning of Australian Timbers. Part 2.
Division of Forest Products Technical Paper No. 22. C.S.I.R., Melbourne (p. 31).5
The timber industry has distinguished 'red' from 'white' myrtle for more than 100 years, but
these classes have been imposed on a continuous variation in colour from straw-brown to
blood-red. 'Deep red myrtle' is only a name for timber at the darker end of the 'red' myrtle
range. Nevertheless, it has long been recognised that myrtle colour correlates with wood
properties: the deeper the red, the heavier and harder the timber, and the longer to air-dry.
'Deep redness' is a signal to the sawmiller that special care needs to be taken in seasoning,
and to the wood user that machining will not be easy. 'Deep redness' is also a selling feature.
For many years Burnie Timber, which sourced its wood from DRM forest in northwest
Tasmania, marketed its myrtle under the name 'Australian Cherry'.
The cause of redness in myrtle is unknown. Nearly the full range of colour can sometimes be
found in a single, even-aged stand. DRM is relatively common on highly fertile soils, but
even 'good red myrtle country' on Tertiary basalt can carry older stands of trees with straw-to
pink-coloured heartwood.
M. Margaret Chattaway looked closely at red myrtle heartwood in the early 1950s 1 . Her
specimen was a section from a Tasmanian tree about 350 years old which was 105 cm in
diameter at 3 m from the ground. The heartwood margin was very irregular and the
heartwood proper was filled with fungal hyphae. Chattaway regarded it as 'probable that the
heartwood in this log…was of pathological origin. This is supported by the zone lines that can be
traced in the wood…and by the irregular outline of the heartwood. Other large logs of "red" myrtle
beech were seen in the felling area. In every one the outline of the red heartwood showed similar
marked irregularities.'
The association of myrtle redness with fungal invasion has not been further studied. It seems
unlikely that a 'reddening' fungus could also be responsible for the extensive decay which is
common in both 'red' and 'white' myrtle, because DRM stems are often sound to the core.
Early in the resource review we devised a digital analysis of myrtle colour, described in
Appendix 1. DRM has a characteristic colour signature when optically scanned. This
signature could be used (in conjunction with colour standards) to grade redness in wet or
dry timber samples.
6. Myrtle forest and the DRM resource
[Myrtle] is, however, very faulty in the log. Timber growing in the swamps in conjunction with
blackwood is usually considered as useless. To grow into sound timber, it apparently requires fair
soil at a fair elevation, say over 800 feet. Consequently, although there are most extensive forests of
this species, there are very few areas which are economical milling propositions.
Rodger, G.J. Forest Survey of Tasmania (p. 29). In Development and Migration Commission
(1929) Report on Afforestation and Reforestation of Tasmania. Tasmanian Technical Report No. 4.
Government Printer, Canberra.
Studies have shown that it is almost impossible to estimate ocularly the degree of defect in
standing myrtle trees. An apparently sound myrtle may be shown to be almost entirely rotted
when felled. Conversely, a myrtle which from ocular assessment would be judged to be rotten, can
be shown to be sound when felled. A study of the ability of forest assessors to predict internal
defect reliably from external appearance, showed that experience is of little value and a blind guess
was as good as the considered opinion of an experienced assessor in most cases (FORTECH, 1982).
Montgomery, P.J. (1985) A Critical Examination of Inventory of Rainforest in Northwest Tasmania
with Emphasis on the Assessment of Inner Stem Defect. Report Submitted in Partial Fulfilment of
the Requirements for the Graduate Diploma of Science, Department of Forestry, Australian National
University (p. 5). Canberra; 86 pp. [FORTECH (1982) Rainforest in the Northwest of Tasmania.
Study for H.C. Sleigh Ltd. and the Tasmanian Forestry Commission. Canberra; 151 pp.]
1 Chattaway, M.M. (1952) The sapwood-heartwood transition. Australian Forestry 16(1): 25-34..6
In the 1970s and 1980s, Forestry Tasmania carried out several assessments of the myrtle
resource in northwest Tasmanian rainforest. The aim in each case was to improve both the
accuracy and the precision of volume estimates. Assessments of gross bole volume and
pulpwood volume were generally reliable, but sawlog results were very disappointing.
(A summary of FT's rainforest resource studies will be included in the final report of the
DRM resource review.)
One difficulty faced by assessors was the natural variability of the forest. As an example,
consider the 60 plots (mostly 0.2 ha) measured in undisturbed 'M+' rainforest on Tertiary
basalt in the Northwest between 1973 and 1986. Sawlog volume assessed by eye ranged
50-fold from 8 to 450 m 3 /ha (mean 136 ± 97 s.d.).
A second difficulty was the high level of internal defect (fungal decay) in live myrtle.
Assessors attempted to correct for defect by felling sample trees, cross-cutting the stems
according to standard protocols, and estimating sound volume from defect measured on the
ends of the cross-cut pieces. Thirty-two of the 60 plots referred to above were included in a
defect analysis carried out in 1985/86. The ratio of recoverable sawlog to assessed sawlog
volume on the 32 plots averaged only 0.43, and ranged from zero to 1.25.
Internal defect on DRM-rich forest sites is lower, on average, than on low- to medium-fertility
sites, where DRM is very rare. Regardless of source, higher sawmill recoveries are
achieved from larger myrtle logs. In recognition of the dependence of recovery on log size,
the stumpage rate for Crown myrtle sawlog less than 75 cm mid-diameter (utility grade) is
only half that for logs >75 cm (category 4). Logs of the latter size, obviously, must come from
trees over 75 cm in diameter at breast height (dbhob). Ocular assessment of 'M+' forest on
Tertiary basalt shows that myrtles >75 cm dbhob contribute more than 85% of total sawlog
volume on most plots (i.e., sawlog volume uncorrected for defect).
As part of the DRM resource review we are attempting to recover useful information on
stem size and spatial distribution from the hundreds of rainforest plots visited by FT
assessors over the years. A preliminary analysis (Appendix 2) suggests that old-growth M+
rainforest is self-replacing, but on a spatial scale which allows for the development of even-aged
patches of mature trees.
7. What is the loggable area of DRM-rich forest outside the Corridor?
To find the available DRM-rich forest we used GIS tools and a selection procedure which
progressively discarded unavailable or unsuitable areas. In this section we describe the
procedure. See Appendix 3 for GIS details and numerical results.
Over the years FT has invested considerable effort in defining and stratifying rainforest on
aerial photographs. The most recent result 1 is a mapping which covers the whole State and
includes all photo-interpreted rainforest patches down to 3 ha in extent. Stratification by
height and vegetation sub-type has proved to be very difficult, and rainforest is now broadly
typed as either:
M+ Usually taller than 25 m, with a sparse understorey which often includes manferns
(Dicksonia antarctica ); usually growing on more fertile sites
M- From 8 m to usually less than 25 m, with a dense understorey which mingles with
the canopy trees resulting in a fine-textured appearance on aerial photographs;
usually growing on sites of low to moderate fertility.
1 Mapping prepared in 1996 for the RFA. See also the history reviewed in: Hickey, J., Davis, S.,
Wardman, R. and Harris, J. (1993) How much rainforest is in Tasmania? A better answer to a
difficult question; Tasforests 5:13-24. Mapping used in this project has minor updates to 1998..7
We used P.I. typing as follows:
1. The forest types rich in DRM (see section 2) are M+ growing on soils derived from igneous
bedrock. We selected all patches of such forest in northwest Tasmania using GIS tools.
2. We located all DRM-rich forest within post-RFA provisional coupes in State forest. Areas
outside these coupes are currently unavailable for logging and are not included in
harvest planning, whether for rainforest logging, eucalypt logging or plantation
development. Coupe boundaries have been drawn by forest planners to exclude riparian
reserves, very steep slopes and other special features.
3. We excluded coupes with a total DRM-rich forest area less than 10 ha, which we regard as
the lower limit for practical management. This discount amounted to 2% of the area
identified within coupes.
4. We looked at the distribution of P.I. types within coupes. Coupes with dense young
myrtle patches scattered through old-growth would be very difficult to log without
greatly increasing the incidence of myrtle wilt in the retained myrtle regrowth. No such
coupes were noted. One coupe was excluded because all loggable DRM-rich forest was in
a narrow strip adjoining the coupe boundary, another because loggable patches were too
small and scattered.
5. We excluded coupes which will be managed for eucalypt production. Rainforest patches
within such coupes will either be converted to eucalypt (when small) or retained as
informal rainforest reserves (when large). We also excluded coupes which will be
converted to plantation. In designating coupes for conversion, planners have followed FT
policy guidelines which require undisturbed rainforest patches over 10 ha to be retained
wherever possible as informal reserves.
At this stage we had a large set of coupes potentially available for long-term sustainable
management of DRM-rich forest. We pruned this list to find the couped area available for
logging during the life of the RFA.
6. We excluded coupes which contain less than 10 ha of 'loggable' DRM-rich forest, as
defined by P.I. types (see Appendix 3). 'Non-loggable' DRM-rich forest contains myrtle
regrowth patches or is typed as fire-damaged, cut over or 'dead and dying'. These types
are left unlogged to avoid damaging myrtle regeneration.
7. Finally, we excluded the coupes which lie within the Savage River Pipeline Corridor.
The result of the selection procedure is a working list of 157 coupes in northwest Tasmania.
Within the coupes are approximately 8200 ha of loggable M+ forest (excluding pure myrtle
regrowth and fire-damaged, moribund and cut-over types) on igneous bedrock, of which
6850 ha are on Tertiary basalt (see summary tables for the 157 coupes in Appendix 3).
For the purposes of the resource review, 8200 ha is the gross loggable area of DRM-rich
forest in northwest Tasmania outside the Pipeline Corridor. There are several recognisable
sources of error in this figure, all of which are positive; in other words, 8200 ha is likely to be
an overestimate.
The most important source of error is photo-interpretation. M+ forest is easily recognised on
aerial photographs, but larger patches typed as M+ may contain loosely bounded 'inclusions'
of M-, which are likely to be highly defective myrtle, poor in DRM. In a selective logging
operation in M+ forest, M- patches would be avoided. The extent of such inclusions is hard
to estimate, but may be as high as 5% of the total M+ area.
A second source of error is bedrock mapping. Geologists have sometimes used aerial
photographs to 'fill in' their picture of unvisited areas, basing their judgments on field-tested
associations between forest vegetation and rock type. One such association is 'tall, open
myrtle forest in a relatively high, level area close to known Tertiary basalt indicates Tertiary
basalt '. Since M+ forest patches can also occur on non-igneous bedrocks, such as Cambrian.8
mudstones, it is possible that some of the DRM-rich forest we identified is actually on less
fertile substrates, and is therefore less likely to be rich in DRM. We suspect that this
geological mapping error would be smaller than P.I. mapping errors.
A third source of error is also geological: not all igneous bedrocks carry DRM-rich forest. A
case in point is coupe WH018A, which carries 29 ha of M+ on Tertiary basalt and 15 ha of
M+ on Devonian granitics. The coupe was visited in the 1999 summer, and it was clear that
while the myrtle on basalt was typical M+, the myrtle off basalt was more defective. If we
assume that non-basalt igneous substrates are only half as productive of myrtle sawlog as
basalt, the substrate proportions tabulated in Appendix 3 show that this difference is
equivalent to a ca. 10% discount on the gross loggable area of DRM-rich forest.
These considerations suggest that the gross loggable DRM-rich forest area within available
coupes might be up to 10-15% less than the 8200 ha estimated using GIS tools. It is highly
unlikely to be greater than 8200 ha.
As noted in section 2, there is also DRM-rich forest in northeast Tasmania. A selection
process (see Appendix 3) found 28 available coupes containing at least 10 ha each of mature
M+ forest on igneous bedrock. This resource totals ca. 1070 ha but has not been investigated
as intensively as northwest Tasmanian myrtle, and no per-hectare logging yields are
available. Northeast myrtle is being treated separately from northwest myrtle in the resource
review.
8. How much myrtle sawlog is recoverable from DRM-rich forest?
As of this writing, five trial coupes and 10 production coupes have been at least partially
harvested in DRM-rich forest (Appendix 4) and another 5-10 coupes may be logged during
the course of the resource review. Yields from these coupes will be used to estimate
recoverable sawlog and pulpwood volumes.
Myrtle pulpwood is unavoidably produced in the selective logging of rainforest. Pulpwood
logs arise as headlogs in sawlog trees and from 'duds', i.e. myrtles which look like sawlog
trees when standing, but which are found to have excessive levels of internal decay when
felled. Sales of heads and duds reduce the overall costs in logging DRM-rich forest, but an
upper limit to pulpwood recovery is set by the silvicultural requirement to leave enough
undamaged seedtrees at sufficiently close spacing. Pulpwood recovery within these limits is
determined by market demand. When pulpwood sales are easy to achieve, heavier selective
logging ('overstorey retention') also allows greater sawlog recoveries per hectare, perhaps
mainly because fallers are more willing to cut down a possible sawlog myrtle and take the
risk that the tree is a dud. Yields are available from light and heavy selective logging in the
past, but more data are needed.
When the Forest Practices Plan (FPP) for a coupe is prepared, an on-ground inspection is
carried out and a logging boundary is marked with strips of blue tape tied to vegetation.
This blue-taped operational boundary is invariably inside the provisional boundary of the
coupe, and often excludes features (such as small streams, special botanical features, etc.)
which are not visible on aerial photographs. The operational boundary may also leave out
portions of the DRM-rich forest identified using GIS, and take in adjoining M+ forest of good
quality.
We will be using operational areas, rather than actually cut-over areas, to calculate per-hectare
recoveries of myrtle sawlog. Operational areas will be determined from ground
survey or from adjustments recently made (at the time of logging) to the GIS-recorded
boundary of the coupe in question. For each coupe, per-hectare recoveries will simply be
volumes divided by operational area. Within the coupes logged so far, only a small area of
non-DRM-rich forest appears to have been cut.
Before we can project per-hectare figures to cover all remaining unlogged coupes, we need
to discount the gross loggable area of DRM-rich forest within those coupes - to 'blue-tape'.9
the coupes (on digital maps) in advance of an FPP inspection. This provisional 'blue-taping'
will be based on a study of boundary marking and logging tracks (mapped by ground
survey) in already logged and currently logged coupes. It will also take into account the
possibility that alternative harvesting methods may become economically attractive during
the life of the RFA.
One result of this 'blue-taping' exercise will be an estimate of the net loggable area of DRM-rich
forest outside the Pipeline Corridor, which we expect to be significantly smaller than the
8200 ha estimate derived from a GIS-based selection (section 7). More importantly, we will
be estimating the net loggable area of DRM-rich forest in each available coupe. From per-hectare
recovery data, we will then assign to each coupe a recoverable sawlog volume. This
assignment is critical in determining the economic accessibility of the DRM resource (see
section 9).
Net loggable area may exclude patches badly affected by myrtle wilt. Such patches will be
identified from time-series of aerial photos, on which brown, wilting crowns and grey,
leafless crowns stand out clearly against the dark green canopy of healthy myrtle. We may
also attempt a discount of net loggable area for losses due to wilt over the life of the RFA,
using remeasurements of permanent dieback transects established in unlogged forest in the
1980s. At this stage of the resource review, myrtle wilt is an important source of uncertainty.
A great deal has been learned about the disease and its distribution 1 , but we do not yet know
what proportion of wilt-affected stands are successfully regenerating and are therefore off-limits
to logging. We hope to gather more information from visits to wilt-affected coupes.
We also intend to refine our recovery estimates using log dimension data. Myrtle sawlogs
are currently graded by mid-diameter, with logs >75 cm attracting the highest stumpage.
Harvest cost is thus dependent on the diameter distribution of the sawlogs harvested. A
preliminary analysis has shown that log sizes are correlated with stand structure (Appendix
5), and that it should be possible to predict the proportion of large-diameter sawlogs to come
from individual coupes. This work will be continued during the resource review.
To show how the volume estimation described above will work, we offer in Appendix 6 an
example based on an imaginary coupe.
9. How much will it cost to harvest the recoverable volume?
Clause 55a of the 1997 Tasmanian RFA requires that the State review the DRM resource in
terms of its 'economic accessibility'. The measure of 'economic accessibility' chosen for this
review is 'dollars per cubic metre at mill door'. In practice, other criteria may be used in
deciding whether particular logging scenarios are economic, and there may be negotiations
in future between FT and industry regarding who pays what costs.
We have devised a simple protocol for estimating mill-door cost. The protocol ties each
coupe to an existing or future road network. In each network we will consider sawlog
volume, pulpwood volume, stumpage, roading costs (for new roads), logging cost, cartage
and road toll (for new and existing roads). Logging options within an individual network
will be costed separately. When all networks have been costed, we will tabulate the
aggregate costs for a range of practical logging scenarios. For example, the same volume and
cost per cubic metre might be realised by logging one-third of network A and two-thirds of
B, or vice-versa. Both scenarios would be tabulated. Impractical scenarios will be excluded
but reasons will be given for their exclusion. It would make little sense, for example, to
harvest a 10 ha patch of DRM-rich forest at the far end of a road, while leaving unlogged a
200 ha patch adjoining the near end of the road.
1 Packham, J.M. (1994) Studies on Myrtle Wilt. Unpublished Ph.D. thesis, University of Tasmania,
Hobart..10
How the costing protocol works is shown in Appendix 7 using a hypothetical example. In
what follows we consider some of the details.
A number of jobs will not be costed separately because they are part of FT's normal
operations, and are covered in stumpage and road toll. These jobs include planning tasks,
road location (when done by, or with assistance from FT), logging supervision, fire
protection and regeneration surveys. On the industry side, jobs such as Forest Practices Plan
preparation and logging supervision will not be costed. To further simplify the protocol, all
on-truck costs will be combined into a single 'logging cost per cubic metre', with variation
allowed for logging difficulty. For costing purposes, logging includes equipment floating,
tracking, log falling, dud falling, snigging, landing work and loading.
The costing variables considered in the protocol are itemised below. Dollar figures are 1999
estimates and will be adjusted as necessary after the next and succeeding summer loggings.
We emphasise that these 1999 estimates are not to be regarded as fixed elements of costing.
Sawlog volume. For each coupe an estimate will be made of myrtle sawlog in category 4,
utility and outspec grades. Volume uncertainties will be treated as explained in
Appendix 7.
Pulpwood volume. Low and high estimates will be made for each coupe.
Stumpage. FT's charge for the three grades of myrtle sawlog and for pulpwood.
Roading cost:
Road construction. Roads to clusters of coupes are either class 3 gravel roads costing
$30000/km, or class 4 dirt roads costing $8000/km. Roads within coupes are either class 4
dirt roads or temporary dirt tracks costing $5000/km, but may be class 3 gravel roads or
class 4 dirt roads if used to access another coupe. Costs include culverts, which will be
minimum standard structures and will in most cases be removed after logging.
Bridges, etc. Bridges are assumed to be all-timber structures designed for at least five
years use, and costs will be based on FT experience. A special structure may be needed
for temporary crossings of the Savage River pipeline.
Road maintenance and repair. Roads for one-off myrtle logging will not require ongoing
maintenance but they may need minor, between-season repairs to potholes, etc. Use of
the Savage River Pipeline Road will probably be on the understanding that the road will
be repaired at the end of each logging season.
Logging cost. To simplify the complicated task of costing the harvest within individual
coupes, the protocol uses three indicative on-truck costs: $24/m 3 for mainly short snigs in
relatively dense stands, $32/m 3 for mainly short snigs in relatively sparse stands and
$40/m 3 for mainly long snigs in relatively sparse stands.
Cartage. The cartage rate used in the protocol will be the appropriate '$/tonne-km'.
Road toll. This cost will vary considerably and may be zero where coupes are close to main
roads. At the time of writing, it is uncertain whether use of the Savage River Pipeline
Road will attract a fee payable to the road managers, Australian Bulk Minerals.
It has not yet been decided whether logging roads used for one-off myrtle logging will be
decommissioned and rehabilitated after use. We assume that decommissioning, if it occurs,
is outside the scope of costing for access. Another uncertainty concerns cartage. In future,
myrtle sawlog volume may be allocated to customers either truck by truck (the current
system) or coupe by coupe. In the hypothetical example given in Appendix 7, sawlogs are
carted to a single customer in Smithton..11
Appendix 1. Digital colour analysis (see section 5)
Red myrtle is redder than ordinary myrtle, and ordinary myrtle is paler than red myrtle, but
what do 'redder' and 'paler' mean?
It would be useful to have an objective, reliable means of defining myrtle colour. Ideally, a
DRM definition method should
· distinguish points along colour gradients,
· distinguish mixed or mottled colouring from uniform colouring,
· give quick, reproducible results,
· be applicable to either green or dry wood samples of any size, and
· be useable by anyone, including the colour-blind!
All of these advantages are potentially offered by digital colour analysis. This Appendix
summarises the basics of digital colour and presents preliminary results of one kind of
analysis on wood samples.
Digital colour
The colour system used in most computers, scanners and other common digital devices is
based on '24-bit RGB addition'. A colour is defined in this system as a three-part number: a
red (R) value, a green (G) value and a blue (B) value. Each value can range from 0 to 255, and
every combination of R, G and B values (such as R=123, G=46 and B=201) is a different
colour. The total number of colours in this system is 256x256x256, or 16,777,216.
The system is additive, like human colour vision. In other words, an absence of colour (R, G
and B all equal to zero) is black, and an equal mix of colours (such as R=128, G=128, B=128)
looks gray, with the brightest gray (R=255, G=255, B=255) looking white.
RGB addition is used to create the colours we see on a computer monitor. The printing of
colour on paper uses a different system, and the software programs which translate screen
colours into printed colours aren't always 100% successful: the coloured images in this
Appendix are not exactly like the originals!
RGB histograms and uniformity of colour
This myrtle veneer, shown here approximately full size, is from Gunn's Veneers in Somerset.
Although not very red, the veneer sample is relatively uniform in colour. It was scanned at
300 dpi with a Hewlett-Packard ScanJet 5100C colour scanner and saved as a 24-bit bitmap
file (ca. 750kb).
To the right of the image is its RGB histogram, as produced by the graphics program Paint
Shop Pro. The horizontal scale is marked from 0 to 255 in increments of 5. The vertical scale,
which is arbitrary, shows the relative number of pixels (picture units) in the image having
the indicated blue, green and red values. Each pixel in the image has a 3-number RGB
colour, and there are many different such colours in the image. What Paint Shop Pro does is
lump together all the colours, then analyse the spreads of R, G and B values in the mix..12
Notice that the peaks in the histogram are fairly sharp: not many different R, G and B values
are used in the image. 'R 209' in the histogram's title means that the average R value in the
image is 209.
'Brittons 3' (above) is a sample of very non-uniform myrtle veneer from Britton Bros. Notice
that all three peaks in the histogram are broad; the blue peak is almost flat because its
highest values are near the black (0) and white (255) ends of the value range. Digital colour
analysis is clearly able to distinguish uniform from non-uniform colouring, but it would be
important for standardised comparisons to fix an appropriate sample size. At any given
scanning resolution, the larger the sample the less uniform the timber.
Comparing colours
The top image is of another Brittons veneer sample. Sections of this multi-coloured veneer
were selected out from the scan as shown (image manipulation with Paint Shop Pro) and
separately analysed. Notice that the histogram for the right-hand section is almost a match
for that of the Gunn's veneer (preceding page). The real-life colours are also very close,
showing that RGB histograms can serve as timber colour 'fingerprints'.
Notice, too, that as myrtle colour deepens (right to left across the veneer) the red peak stays
sharp but moves to lower R values. Blue and green peaks also shift to lower values but drop
much more than red. In the 'deep red' veneer patch at far left, blue has almost disappeared
and green has a very broad peak at low G values. A typical colour here might be R=166,
G=40 and B=5, which in pure form looks like this:.13
Wet vs. dry timber colour
The first image below is another myrtle veneer sample from Brittons, scanned as dry wood.
The second image is the same veneer sample after wetting (the wet veneer was placed in a
transparent plastic sleeve to protect the scanner). The effect of wetting on the RGB histogram
is the same as that seen in the 'reddening' across the Brittons 1 sample, above: R stays sharp
and drops, G drops considerably and B almost disappears.
This illustrates an important feature of scanned, digital images. The red pigments in myrtle
do not change chemically when the timber is wet, but the timber's apparent redness does.
Digital colour technology is not spectroscopic chemical analysis, and cannot be used to
detect the presence/absence or relative quantity of colour-producing pigment in the wood.
What it detects is the appearance of the timber, just as our eyes do. Just as you cannot fairly
compare colouring in two pieces of myrtle, one dry and one wet, you cannot compare digital
analyses of two pieces of myrtle unless both are wet or dry.
Samples for analysis
Almost anything can be scanned: veneer sheets, scraps of wood, tree cores, and log ends.
Using a handheld scanner and a field computer, scanning could be done at logging sites.
Digital colour analysis can even be applied to sawdust, as shown below. The main image is a
sample of chainsaw waste from a particularly red bit of myrtle. Separate RGB analyses are
given for different parts of the image. The scan was at 1200 dpi..14
Appendix 2. Stand structure from assessment plots (see section 6)
Introduction
In the 1970s and 1980s, Forestry Commission personnel measured more than 350 Mature
Forest Inventory (MFI) plots in an effort to improve estimates of the rainforest sawlog and
pulpwood resource in northwest Tasmania. Not all of the results are relevant to the DRM
resource review. Many of the plots were on lower fertility sites, and some of the Tertiary
basalt plots were in M- forest (growing in poorly drained spots) or in cut-over M+.
Data from rainforest asssessment plots are being re-compiled for this study. For each of
195 plots we currently have accurate plot locations and a complete set of tree measurements
(Table 1).
Table 1. MFI rainforest plots used as data sources.
No. of plots:
Project Year On basalt Off basalt Total
Forwood M3 1973 25 9 34
Project 40 (in part) 1976 45 34 79
Prelogging 1980 6 0 6
Prelogging 1982 2 0 2
Rainforest 1985 23 14 37
Prelogging 1986 4 0 4
Rainforest 1986 21 12 33
Totals 126 69 195
To ensure consistency, myrtle sawlog volumes have been recalculated from overbark
diameters, stump heights and sawlog heights on all plots using the volume equation
adopted by the Forestry Commission in 1982 for South Arthur myrtle. The formula is:
V = a (1 - e -cH ) n D 2 ,
where V is the volume (m 3 ) from ground level to height H (m), D is dbhob (cm), and a, c and
n are the constants 0.00125338, 0.0380063 and 0.753122, respectively.
Thirty of the 195 plots were in forest which had been selectively logged in the past. Of the
remaining plots, 103 were on Tertiary basalt and 62 were on other bedrocks. In agreement
with ecological studies 1 , uncut plots on basalt were found to carry less leatherwood (Table
2). They also had roughly 20% more myrtle stems, myrtle basal area and assessed myrtle
sawlog than off-basalt plots:
Table 2. Mean statistics for undisturbed plots.
Myr stocking Myr basal area Myr sawlog Lea basal area
Basalt No. of plots (stems/ha) (m 2 /ha) (m 3 /ha) (m 2 /ha)
on 103 100 54 124 1.0
off 62 80 46 100 2.7
The 103 undisturbed, on-basalt plots include some with areas so small (e.g. 0.04 ha) that their
inclusion in an analysis would distort the results in stand-table analysis. In what follows we
exclude plots smaller than 0.2 ha. Among the remaining plots were some with fewer than 10
myrtle stems; these are also excluded to avoid distortions. The residual set of 82 plots are all
undisturbed, on basalt and at least 0.2 ha, and carried at least 10 myrtle stems >20 cm dbhob.
1 Jarman, S.J., Kantvilas, G. and Brown, M.J. (1991) Floristic and Ecological Studies in Tasmanian Rainforest.
Tasmanian NRCP Technical Report No. 3. Forestry Commission, Tasmania, and Department of the
Arts, Sport, the Environment, Tourism and Territories, Canberra..15
Stand structure
The combined myrtle stand table for the 82 study plots has a 'reverse J' shape which
suggests long-term steady-state regeneration (Fig. 1)1 . However, a sorting of the plots based
on k-means cluster analysis (Fig. 2) shows that 40 plots have a more pronounced 'reverse J'
shape when combined and the remaining 42 include older, pulse-regenerated myrtle stands
with relatively few younger trees. We are currently examining the distribution of the two
stand types at a range of spatial scales. It does not seem likely, from preliminary results, that
the older single-cohort patches derive from single events or episodes which regenerated
rainforest across a broad area in northwest Tasmania.
1 Read, J. and Hill, R.S. (1985) Dynamics of Nothofagus-dominated rainforest on mainland Australia and
lowland Tasmania. Vegetatio 63: 67-78.
Fig. 1. Stand table for 82 plots (1860 stems)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
20- 29
30- 39
40- 49
50- 59
60- 69
70- 79
80- 89
90- 99
100- 109
110- 119
120- 129
130- 139
140- 149
150- 159
160- 169
170- 179
180- 189
190- 199
200+
Diameter class (cm)
Proportion of stems
Fig. 2. Stand tables for 40 'J' plots (1056 stems)
and 42 'pulse' plots (804 stems)
0.00
0.05
0.10
0.15
0.20
0.25
20- 29
40- 49
60- 69
80- 89
100- 109
120- 129
140- 149
160- 169
180- 189
200+
Diameter class (cm)
Proportion of stems
J
pulse.16
Appendix 3. GIS details (see section 7)
To take advantage of updated information, DRM-rich forest coupes have been selected
several times since the resource review began last year. The latest selection is described
below.
'Northwest Tasmania' was defined as the rectangle bounded on the W and E by AMG lines
280000 and 440000, and on the S and N by 5310000 and 5500000. This rectangle contains
roughly 1 800 000 ha of land running from the west coast to just east of the Forth River
mouth, and from Bass Strait south to Farm Cove on Macquarie Harbour.
'1. The forest types rich in DRM (see section 2) are M+ growing on soils derived from igneous
bedrock. We selected all patches of such forest in northwest Tasmania using GIS tools.'
Relevant P.I. types ('NewPItype') were selected from the latest P.I. cover in the FT GIS
library. Public land P.I. types are largely current but private land types are out of date (see
below). After deleting polygons less than 1 ha, totals were OK 117 195 ha, E 10 147 ha, MR
5673 ha and FCD 12 593 ha.
OK (apparently undisturbed old-growth M+ forest)
M+., M+.E+3f., M+.E-3f., M+.E1f., M+.E2(P)., M+.E2f., M+.K., M+.K.E2f., M+.Tb., M+.Tb.E2f., M+.Tc.Tb.,
M+.Ts., M+.Tw., M+.Tw.E2f., M+.Tw.dd E2f., M+.Vz., M+.W., M+.dd E2f., M+/+3., M+/2., M+/2.Tw.,
om M+.
E (apparently undisturbed old-growth eucalypt with M+ understorey)
E+3b.ER.M+., E+3b.M+., E+3bc.M+., E+3d.ER1f.dd M+., E+3d.ER3(P).M+.,
E+3d.ER3f.Mr2/M+., E+3d.M+., E+3d.M+.Mr2., E+3d.Mr1/M+., E+3d.Mr2.M+.,
E+3d.Mr2/M+., E1b.M+., E1c.M+., E1d.M+., E2(P).M+., E2a.M+., E2b.M+., E2b.M+.Mr1., E2b.M+.Mr2.,
E2b.M+.Mr2.Tb., E2b.Mr2/M+.ER., E2c.ER3d.M+., E2c.ER3f.Mr1/M+., E2c.M+., E2c.M+.Mr2.,
E2c.Mr1/M+., E2c.Mr2.M+., E2c.Mr2/M+., E2d.ER2d.Mr1/M+., E2d.ER3d.M+., E2d.ER3f.M+.,
E2d.ER4d.M+., E2d.M+., E2d.M+.ER2(P)., E2d.M+.Mr2., E2d.M+.Tb., E2d.Mr1.M+., E2d.Mr1/M+.,
E2d.Mr2/M+., E3f.M+.Tw., E4c.M+.
MR (apparently undisturbed M+ with myrtle regrowth, or dense myrtle regrowth)
M+.Mr1., M+.Mr1.E2f., M+.Mr2., M+.Mr2.E2f., M+.Mr2.K., M+.Mr2.Tb., Mr1.M+.,
Mr1.M+.Tw., Mr1/M+., Mr2., Mr2.E+3f., Mr2.E2f., Mr2.K., Mr2.M+., Mr2.M+.E2f., Mr2.M+.Tb., Mr2.S.,
Mr2.S.E+3f., Mr2.S.T., Mr2.S.T.E2f., Mr2.S.T/2., Mr2.S.Tw., Mr2.T., Mr2.T.S., Mr2.Tw.S.E2f., Mr2.Vz.,
Mr2/M+.
FCD (fire-damaged, cut-over or dead and dying M+ forest)
co M+., co M+.E1f., co M+.E2f., co M+.K., co M+.Mr1., co M+.Mr2., co M+.Mr2.E2f.,
co M+.Mr2.K., co M+.Mr2.W., co M+.Tb., co M+.Tb.E2f., co M+.W., co M+/2.,
co Mr1.M+., co Mr1/M+., co Mr2.M+., co Mr2.M+.K., co Mr2/M+., co W.M+.,
co dd M+.K., co dd M+.TK., co fd M+., co fd M+.E2f., dd M+., dd M+.K., dd M+.fd E2f.,
fd M+., fd M+.fd Mr1., fd Mr2.M+.
Northwest Tasmanian bedrock types were selected from a GIS cover obtained by FT from
Mineral Resources Tasmania in December 1998. After deleting polygons less than 1 ha,
northwest Tertiary basalt (Tb) totalled 254 703 ha and other igneous rocks (Oi) 196 666 ha.
Cba Cambrian; boninitic lavas
Cbb Cambrian; low-Ti tholeiitic and boninitic lavas
Cbt Cambrian; low-Ti tholeiitic lavas
Cbtg Cambrian; gabbroic rocks associated with low Ti-lavas
Cda Cambrian; dominantly andesitic volcanics and intrusives
Cdb Cambrian; dominantly shoshonitic, basaltic to andesitic volcanics
Cdl Cambrian; dolerite of probable Cambrian age
Cdtl Cambrian; felsic lavas within Tyndall Group
Cdv Cambrian; dominantly felsic to intermediate volcanic rocks
Cdvt Cambrian; upper, dominantly volcanoclastic sequences of Tyndall Group
Cg Cambrian; coarse-grained basic rocks
Cgr Cambrian; granite
Cwb Cambrian; basalt
Cwmb Cambrian; basalt of the Mainwaring River area
Dd Devonian; dolerite dykes.17
Dg Devonian; undifferentiated granitic rocks
Dga Devonian; undifferentiated alkali-feldspar granite/granite/adamellite
Dgaa Devonian; dominantly adamellite/ granite and associated dykes
Dgaas Devonian; dominantly adamellite/granite (S-type)
Dgaf Devonian; alkali-feldspar granite
Dgafs Devonian; dominantly alkali-feldspar granite (S-type)
Dgas Devonian; undifferentiated alkali-feldspar-granite/granite/adamellite (S-type)
Jd Jurassic; dolerite
Ld Precambrian; mafic and felsic dykes
Lob Precambrian; basalt
Lsb Precambrian; tholeiitic basalt
Lvb Precambrian; tholeiitic basalts
Tb Tertiary; basalt
The four relevant P.I. groups cover ca . 57 400 ha on Tertiary basalt and other igneous rock
types in northwest Tasmania. A breakdown of type groups by bedrock groups (in hectares)
is:
OK E MR FCD Total
Tb 27500 1098 3375 8465 40438
Oi 14785 1018 494 731 17028
Total 42285 2116 3869 9196 57466
Existing and post-RFA reserves account for ca. 17 200 ha of the total (less patches <1 ha):
OK E MR FCD Total
Tb 6580 114 235 889 7817
Oi 8930 341 60 47 9378
Total 15510 455 295 936 17195
More than 13 000 ha of DRM-rich forest was typed on private land, mainly owned by North
Forests. The P.I. typing dates from the mid-1980s and does not reflect the extensive clearings
for plantation which have been carried out in the past 15 years. In the mid-1980s, roughly
half the private DRM-rich forest was cut-over or regrowth.
Almost all of the remaining DRM-rich forest is managed by Forestry Tasmania.
'2. We located all DRM-rich forest within post-RFA provisional coupes in State forest. Areas outside
these coupes are currently unavailable for logging and are not included in harvest planning,
whether for rainforest logging, eucalypt logging or plantation development. Coupe boundaries
have been drawn by forest planners to exclude riparian reserves, very steep slopes and other special
features.
'3. We excluded coupes with a total DRM-rich forest area less than 10 ha, which we regard as the
lower limit for practical management. This discount amounted to 2% of the area identified within
coupes.'
We considered only the coupes designated NFT (native forest), STM (special timbers
management) and BWC (blackwood working circle) in the former Circular Head and
Murchison Forest Districts. The area of DRM-rich forest within coupes in Mersey District
(eastern edge of northwest Tasmania) is not significant. Coupe data were current as of 28
April 1999, but were updated by excluding coupes and parts of coupes within all areas
proposed as reserves under the RFA ('Pten' in the current FT GIS tenure cover).
We found 384 provisional coupes containing DRM-rich forest (total ca. 17 200 ha), of which
281 coupes contained at least 10 ha of DRM-rich forest types (total ca. 16 800 ha)..18
'4. We looked at the distribution of P.I. types within coupes. Coupes with dense young myrtle patches
('MR' types) scattered through old-growth would be very difficult to log without greatly
increasing the incidence of myrtle wilt in the retained myrtle regrowth. No such coupes were
noted. One coupe was excluded because all loggable DRM-rich forets was in a narrow strip
adjoining the coupe boundary, another because loggable patches were too small and scattered.'
Coupe Problem Geology Type areas (ha)
OK E MR FCD Total
BF012C DRM type only on boundary Oi 10 4 - - 14
LG012E Small, scattered patches Oi 10 - - - 10
'5. We excluded coupes which will be managed for eucalypt production. Rainforest patches within
such coupes will either be converted to eucalypt (when small) or retained as informal rainforest
reserves (when large). We also excluded coupes which will be converted to plantation. In
designating coupes for conversion, planners have followed FT policy guidelines which require
undisturbed rainforest patches over 10 ha to be retained wherever possible as informal reserves.'
Using information current as of 28 April 1999, the 34 coupes listed below were excluded for
management reasons. An example of an informal reserve is ca. 25 ha of M+ forest on Tertiary
basalt within the recently logged DP021C. The OK total is misleading in some cases, where
ground inspection has found evidence of earlier logging.
Coupe Geology Type areas (ha)
OK E MR FCD Total
DP017A Tb - - - 14 14
DP021C Tb 28 46 - - 74
DP023D Tb - 54 - - 54
FD035A Tb - - - 10 10
FD041C Tb 18 - - - 18
FY016A Tb 1 - - 10 11
HE007D Tb 5 - - 37 42
HL044A Tb 3 11 - - 14
LG007A Oi 1 - - 30 31
LG007F Oi - - - 23 23
ML055F Tb 10 - - - 10
NH011A Tb 5 8 - - 13
NH028A Tb 6 16 - 6 28
OO068B Tb 30 - - - 30
PU027D Tb 16 - - - 16
PU037A Tb - 6 - 12 18
PU039B Tb 16 - - 4 20
PU040D Tb 8 - - 18 26
PU042C Tb 14 - - - 14
PU043F Tb - - - 16 16
PW011A Oi 1 - 10 - 11
Tb 14 - 2 31 47
PW012C Tb - - 15 67 82
RD024A Tb 14 - - - 14
SU020A Oi 14 7 - - 21
SU020B Oi 26 26 - - 52
SU022B Oi 16 6 - - 22
SU022C Oi 28 25 - - 53
SU023A Oi 49 34 - - 83
SU024A Oi 15 11 - - 26
SU034B Oi 6 9 - - 15
SU037A Oi 12 3 - - 15
SU055A Oi 24 22 - - 46
WH014A Tb - - 12 - 12
WH016D Tb - 5 16 28 49
Totals 387 303 55 306 1051.19
'6. We excluded coupes which contain less than 10 ha of 'loggable' DRM-rich forest, as defined by
P.I. types (see Appendix 3). 'Non-loggable' DRM-rich forest contains myrtle regrowth patches or
is typed as fire-damaged, cut over or 'dead and dying'. These types are left unlogged to avoid
damaging myrtle regeneration.'
The 26 coupes and their contents are tabulated below. 'Loggable' types are OK and E, 'non-loggable'
are MR and FCD. Although MR types such as M+.MR2. can be logged if care is
taken to avoid damage to regrowth, the difficulty of such logging increases considerably
with MR proportion. We take the position that a coupe rich in MR is only worth logging if it
contains at least 10 ha of OK and E types as well.
Coupe Geology Type areas (ha)
OK E MR FCD Total
BO101B Tb 7 - - 11 18
BO103A Tb 5 2 14 21
BO104A Tb 1 - 11 15 27
BO106A Tb 6 - 4 - 10
BO205A Oi 7 - 4 - 11
BO218A Oi 1 - 15 - 16
BO239A Tb 3 - - 12 15
BO251A Oi 2 - - 25 30
LG023A Oi - - - 2 2
Tb - - - 38 38
LG024A Tb - - - 32 32
LG024B Tb - - - 87 87
LG024C Tb - - - 16 16
LG025A Tb - - - 41 41
ML017A Oi 8 - - 13 21
ML067A Oi 10* - - - 10* *rounded up
NH005A Tb 6 - - 42 48
NH009B Tb - - - 26 26
NH018E Tb - - - 90 90
NH018F Tb - - - 114 114
RD018B Tb - - - 11 11
WH001A Tb - - 29 - 29
WH001B Tb 7 - 101 38 146
WH001C Tb 7 - 8 44 59
WH016F Tb - - 16 - 16
WH016G Tb - - 5 14 19
WH016H Tb - - 34 10 44
Totals 70 0 228 695 994
'7. Finally, we excluded the coupes which lie within the Savage River Pipeline Corridor.'
The 62 Corridor coupes are listed below.
Coupe Geology Type areas (ha)
OK E MR FCD Total
FY056A Tb 308 - - - 308
MB070A Tb 226 - - - 226
MB070C Tb 70 - - - 70
MB071A Tb 79 - - - 79
MB071B Tb 22 - - - 22
MB071C Tb 15 44 - - 59
MB075A (part) Tb 20 - - - 20
MB075B Tb 34 6 - - 40
MB075C Tb 18 6 - - 24
MB075D Tb 68 - - - 68
MB076A Tb 246 - - - 246
MB077A Tb 54 - - - 54
MB077B Tb 81 - - - 81
MB077C Tb 65 - - - 65
MB077D Tb 35 - - - 35
ML097A Tb 33 - - - 33
ML097B Tb 16 - - - 16
ML097C Tb 57 - - - 57.20
Coupe Geology Type areas (ha)
OK E MR FCD Total
ML097D Tb 74 - - - 74
ML097E Tb 47 - - - 47
ML097F Tb 112 - - - 112
ML097G Tb 71 - - - 71
ML097H Tb 104 - - - 104
ML097I Tb 97 - - - 97
ML097J Tb 64 - - - 64
ML097K Tb 119 - - - 119
ML097L Tb 64 - - - 64
ML097M Tb 136 - - - 136
ML097N Tb 68 - - - 68
ML097O Tb 55 - - - 55
ML098A Tb 60 - - - 60
ML099A Tb 11 - - - 11
ML100A Tb 36 - - - 36
ML106A Tb 50 - - - 50
ML106B Tb 17 - - - 17
ML106C Tb 177 - 8 - 185
ML107A Tb 25 - - - 25
ML107B Tb 161 - - - 161
ML108A Tb 102 - - - 102
ML109A Tb 168 - - - 168
ML109B Tb 45 - - - 45
ML109C Tb 134 - - - 134
ML110A Oi 21 - - - 21
Tb 1 - - - 1
ML110B Oi 10 - - - 10
Tb 34 - - - 34
ML110C Oi 23 - - - 23
ML111A Oi 6 - - - 6
Tb 63 - - - 63
ML112A Tb 56 - - - 56
ML112B Tb 36 - - - 36
ML112C Oi 2 - - - 2
Tb 39 - - - 39
ML113A Tb 17 - - - 17
ML116A Tb 151 - - - 151
ML116C Tb 38 - - - 38
ML117A Tb 335 - - - 335
ML118A Tb 404 - - - 404
ML118B Tb 41 - - - 41
ML119B Oi 5 - - - 5
Tb 156 - - - 156
ML119C Tb 52 - - - 52
ML120A Tb 151 - 30 - 181
ML122A Oi 11 - - - 11
Tb 194 - - - 194
ML122B Oi 21 - - - 21
Tb 3 - - - 3
ML122E Oi 79 - - - 79
Tb 60 - - - 60
ML123A Tb 126 - - - 126
Totals 5578 56 39 0 5672
'The result of the selection procedure is a working list of 157 coupes in northwest Tasmania.'
The 157 coupes are listed below. Coupes in bold type are currently being logged or have
been logged since the most recent P.I. typing update.
Coupe Geology Type areas (ha)
OK E MR FCD Total
BF002B Oi 14 - - - 14
BF004E Oi 12 - - - 12
BF005C Oi 16 - - - 16
BF012A Oi 10 4 - - 14
BF012B Oi 9 9 - - 19.21
Coupe Geology Type areas (ha)
OK E MR FCD Total
BO100C Tb 25 - 6 34 64
BO100D Tb 43 - - 21 64
BO100E Tb 15 - 3 8 26
BO101A Tb 88 - 6 44 137
BO102A Tb 41 - - 17 58
BO109A Tb 15 - 3 - 18
BO109B Tb 61 - 9 3 73
BO109D Tb 12 - 2 - 14
BO114A Tb 15 - 12 58 86
BO114B Tb 26 - - 17 42
BO201A Tb 69 - 35 99 202
BO201B Tb 38 - - - 38
BO202A Tb 15 - 3 48 65
BO206A Oi 11 - 26 - 37
BO213A Oi 76 - - - 76
BO213B Oi 34 - - - 34
BO214A Oi 27 - - - 27
BO217A Oi 17 - 8 - 25
BO240B Oi 24 - 1 - 25
BO244A Oi 32 - 6 4 42
BO244B Oi 29 - - 19 48
BO244C Oi 45 - - - 45
BO250A Oi 33 - - - 33
BO250B Oi 37 - - - 37
BO251B Oi 15 - - - 15
FR039A Oi 49 38 - - 87
FY002D Tb 16 - - - 16
FY002E Tb 22 - - - 22
FY002F Tb 41 - - - 41
FY014A Tb 40 - - - 40
FY014B Tb 28 3 - - 31
FY014C Tb 42 - - - 42
FY014D Tb 49 - - - 49
FY014E Tb 9 3 - - 12
FY018A Tb 112 2 - - 114
FY027B Tb 23 - - - 23
FY027C Tb 58 - - - 58
FY027D Tb 171 - - - 171
FY029A Tb 54 - - - 54
FY030A Tb 92 - - - 92
FY031A Tb 24 - - - 24
FY031B Tb 71 - - - 71
FY033A Tb 41 - - - 41
FY033B Tb 44 - - - 44
FY033C Tb 55 - - - 55
FY033D Tb 122 - - - 122
FY033E Tb 74 - - - 74
FY033F Tb 30 - - - 30
FY034A Tb 58 - - - 58
FY034B Tb 72 - 3 - 75
FY034C Tb 82 - 4 - 86
FY035A Tb 122 - - - 122
FY035B Tb 54 - - - 54
FY035C Tb 135 - - - 135
FY035D Tb 139 - - - 139
FY035E Tb 78 - - - 78
FY035G Tb 154 - - - 154
FY036A Tb 66 - - - 66
FY055A Tb 52 - - - 52
FY055B Tb 77 - - 2 79
FY055C Tb 104 - - - 104
FY055D Tb 46 - - - 46
HE011A Tb 51 - - 8 58
HK011A Tb 23 - - - 23
LG011P Oi 29 - - - 29
LG013B Oi 26 - - - 26
LG014A Oi 51 - - - 51
LG014B Oi 73 - - - 73.22
Coupe Geology Type areas (ha)
OK E MR FCD Total
LG014D Oi 20 - - - 20
LG023B Oi 11 - - 42 53
MB066A Tb 12 - - - 12
MB067A Tb 97 4 - - 101
MB067B Tb 68 - - - 68
MB067C Tb 118 - - - 118
MB068A Tb 14 - - - 14
MB068C Tb 52 - - - 52
MB068D Tb 78 - - - 78
MB068E Tb 19 4 - - 23
MB068F Tb 63 - - - 63
MB069A Tb 44 - - - 44
MB069B Tb 64 - - 5 70
MB075A (part) Tb 11 - - - 11
MD001C Oi 96 - 6 - 102
MD003A Oi 42 - - - 42
ML021A Oi 12 - - - 12
ML022A Oi 15 - - - 15
ML023A Oi 26 - - - 26
ML031A Oi 13 - - - 13
ML032A Oi 22 - - - 22
ML068A Tb 52 - - - 52
ML068D Tb 39 - - - 39
NH019D Tb 21 - - 20 41
NH019H Tb 22 - 26 47
NH019I Tb 55 - - 18 74
NW003B Tb 17 - - - 17
NW003D Tb 17 - - - 17
NW006A Tb 49 - - - 49
NW006B Tb 52 - - - 52
NW006C Tb 115 - - - 115
NW006D Tb 15 - - - 15
NW006E Tb 14 - - - 14
NW007A Tb 32 - - - 32
NW007B Tb 57 - - - 57
NW007D Tb 130 - - - 130
NW007E Tb 44 - - - 44
NW008A Tb 20 - - - 20
NW008B Tb 56 - - - 56
NW011A Tb 52 - - - 52
NW011D Tb 29 - - - 29
NW011E Tb 105 - - - 105
OO068A Tb 53 - - - 53
PW018M Tb 137 - 12 - 148
RD018C Tb 59 - - - 59
RD019A Tb 94 11 - - 105
RD019B Tb 13 - - - 13
RD022A Tb 80 - - - 80
RD023B Tb 15 - - 18 33
RD023C Tb 58 - - - 58
RD023D Tb 85 - - - 85
RD024C Tb 34 11 - - 46
RD025B Tb 12 6 - - 18
RD025C Tb 56 4 - - 60
RN002C Oi 14 - - - 14
SU017A Oi 18 - - - 18
SU019A Oi 31 - - - 31
SU019B Oi 66 - - - 66
SU019C Oi 45 4 - - 49
SU035E Oi 9 3 - - 12
SU055B Oi 75 22 - - 97
WD006A Tb 45 - - - 45
WD006B Tb 20 - - - 20
WD013A Tb 50 - - - 50
WD014A Tb 49 - - - 49
WD015A Tb 38 - - - 38
WH001D Tb 53 - 12 1 67
WH001E Tb 110 - 82 26 218.23
Coupe Geology Type areas (ha)
OK E MR FCD Total
WH001F Tb 80 - 64 - 143
WH001G Tb 12 - 33 - 46
WH001H Tb 112 - - - 112
WH002A Tb 41 - - - 41
WH004B Tb 85 - - - 85
WH016E Tb 26 - 13 28 66
WH017B Tb 37 - 47 - 84
WH017C Tb 57 - 16 - 73
WH018A Oi 15 - - - 15
Tb 29 - - - 29
WH019A Oi 32 - 2 8 42
Tb 39 - - - 39
WH020B Oi 49 - - - 49
Tb 4 - - - 4
WH021A Tb 15 - - - 15
WH022A Tb 60 - 6 - 66
WH022B Tb 23 - 4 - 27
WH026A Tb 98 - 15 29 142
WH028C Tb 57 - 7 - 64
Summary for the 157 coupes in the Northwest:
Type areas (ha)
OK E MR FCD Total
Totals Oi 1281 80 43 73 1476
Tb 6469 49 397 528 7422
All igneous 7749 128 440 601 8918
Type areas (ha)
OK+(M+.MR) E pure MR FCD Total
Totals Oi 1290 80 34 73 1476
Tb 6801 49 64 528 7422
All igneous 8091 128 97 601 8918
'As noted in section 5, there is also DRM-rich forest in northeast Tasmania. A selection process (see
Appendix 3) found 28 available coupes containing at least 10 ha each of mature M+ forest on igneous
bedrock.'
M+ forest occurs in 41 of the 44 Northeast coupes which have been designated 'STM'
(special timbers management). Four of the 41 coupes (CC150A, MO136C, MO136D and
RR161A) contain only pure regrowth (MR types). Two more coupes (BS108F and MO123B)
contain less than 10 ha of OK type. The remaining 35 coupes and their types breakdown are
shown below. Ig rocks are Tertiary basalt and Devonian granitics, while NI rocks are
Cambrian to Lower Devonian mudstones and siltstones. In this table, OK types are M+ with
or without MR and wattle, while MR types are myrtle regrowth without mature myrtle.
Coupe Type areas (ha)
Ig OK Ig MR NI OK NI MR Total
BS103J 6 - 26 - 32
BS106A 30 - - - 30
CC136B - - 24 - 24
CC147B 25 - - - 25
CC147C 41 - 8 - 50
CC148A 17 3 - - 20
CC148B 28 - - - 28
CC148C 60 - - - 60
CC148D 44 - - - 44
CD110A 47 - - - 47
CD112F 14 - - - 14
CD116D 10 - - - 10
CD117B 19 - 5 - 24
CD117C 105 - 1 - 106.24
Coupe Type areas (ha)
Ig OK Ig MR NI OK NI MR Total
CD117D 34 - - - 34
CD117E 137 - - - 137
FL122D 52 - - - 52
FL122E 22 - 15 - 36
MO145D 21 - - - 21
RR101C 44 - - - 44
RR106A 37 13 - - 51
RR110A 12 30 - - 41
RR111A 16 19 - - 34
RR129B 43 - - - 43
RR129C 36 - - - 36
RR129D 27 - - - 27
RR130A 89 - - - 89
RR146A - - 29 22 51
RR149B - - 24 7 31
RR151A 38 - - - 38
RR158A - - 28 - 28
RR159C 13 - - - 13
RR162B - - 11 - 11
RR174A - - 33 - 33
RS144D 11 - - - 11
Totals 1079 65 203 29 1376
Appendix 4. Sources for recovery data (see section 8)
Volume data are available from each of the DRM-rich forest coupes listed below.
Operational areas have been or will be determined for the 11 coupes in bold type.
Coupe Logged Area (ha)1 Remarks
MB067A 1980 2 Plantation clearing. Pre-logging assessment and myrtle
defect measurement.
MB069B 1982 12 Selective logging trial. Pre-logging assessment.
RD025B 1982 5 Selective logging trial. Pre-logging assessment.
RD023B 1983 - 1985 35 Selective logging trials. Pre-logging assessment.
RD018B 1986 9 Selective logging trial. Pre-logging assessment.
NH009A 1994/95 ? Light selective logging; previously cut-over.
NH016B 1993/94 ? Light selective logging; previously cut-over.
OO068A 1995/96 na 2 Light selective logging.
BO201A 1995 - 1997 ? Very light selective logging.
NH019H 1996/97 ? Light selective logging; previously cut-over.
BO109B 1996 - date na 3 Light selective logging.
NW006C 1997/98 na 2 Very light selective logging.
NW006D 1997/98 na 2 Very light selective logging.
BO201B 1997 - date na 3 Light selective logging.
PU039B 1998 -1999 21 Plantation clearing.
1 Approximate area of DRM-rich forest logged.
2 Area not yet determined.
3 Logging not yet completed..25
Appendix 5. Log sizes (see section 8)
We compiled myrtle log size data for the loggings shown in Table 1. All relevant harvests to
1998 were included except those from NH009A, NW006C, NW006D and PU039B. Deliveries
from the last three coupes were all of short logs (6m or less) and are not directly comparable
with deliveries including logs up to 13m. The myrtle harvest from NH009A was
insignificant.
For this analysis we lumped category 4 and utility sawlogs together as 'sawlog' and all
others (outspec, dry, defective, old-style category 2) as 'outspec'.
Table 1. Loggings used as sources for data on myrtle logs delivered to sawmills.
No. of
myrtle
Coupe Area Year(s) Customer(s) logs
RD025B Upper Rapid (Pipeline) 1982/83 Brittons 72
RD023B Upper Rapid (Pipeline) 1983/84 Brittons 190
RD023B Upper Rapid (Pipeline) 1984/85 Brittons 137
RD018B Upper Rapid (Holder) 1985/86 Brittons 234
NH016B Newhaven 1993/94 Brittons 217
OO068A Oonah 1995/96 Corinna 216
BO201A Middlesex 1995/1996, 1996/97 Brittons, Corinna 436
NH019H Newhaven 1996/97 Brittons 137
BO109B Que 1996/97, 1997/98 Brittons, Corinna 471
BO201B Middlesex 1997/98 Brittons, Corinna 67
Total 2177
Stand structure seems to be reflected in mid-diameter profiles (sawlog + outspec). The three
Bulgobac block coupes (Fig. 1) carry some unusually large old myrtles, and Bulgobac logs
peaked in diameter in the 70-79 and 80-89 cm classes. RD018B and OO068A included
patches of older myrtle regrowth; these appear as small-diameter log peaks in the delivery
profiles (Fig. 2). The Upper Rapid (Pipeline) and Newhaven coupes were perhaps more
typical of old-growth M+ forest, and gave broad peaks in the 50-59, 60-69 and 70-79 cm
classes (Fig. 3). Figure 4 shows diameter profiles vs. volumes for all non-outspec logs in the
three kinds of M+ forest.
Myrtle sawlogs are currently categorised by mid-diameter. Logs > 75 cm in mid-diameter
are 'category 4' and logs < 75 cm are 'utility'. Since myrtle redness is known to increase with
age, category 4 logs would be expected to be significantly richer in DRM than utility logs.
Table 2 shows how volume proportions vary with mid-diameter specification in the loggings
we studied.
Table 2. How sawlog proportions vary with specification in three classes of M+ forest.
Mid-diameter % cat. 4 / % utility
minimum for cat. 4
sawlog (cm) 'Regrowth-rich' 'Typical' 'Old'
45 96 / 4 99 / 1 99+ / <1
50 86 / 14 97 / 3 99 / 1
55 70 / 30 92 / 8 99 / 1
60 58 / 42 84 / 16 96 / 4
65 43 / 57 76 / 24 92 / 8
70 34 / 66 63 / 37 84 / 16
75 (current) 26 / 74 52 / 48 73 / 27.26
Fig. 1. Diameter profiles from Bulgobac loggings
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
20- 29
30- 39
40- 49
50- 59
60- 69
70- 79
80- 89
90- 99
100- 109
110- 119
120- 129
130- 139
140- 149
Mid-diameter (cm)
Proportion of total logs
BO201A
BO109B
BO201B
Fig. 2. Diameter profiles from regrowth-rich M+
loggings
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
20- 29
30- 39
40- 49
50- 59
60- 69
70- 79
80- 89
90- 99
100- 109
110- 119
120- 129
130- 139
140- 149
Mid-diameter (cm)
Proportion of total logs
OO068A
RD018B.27
Fig. 3. Diameter profiles from Upper Rapid and
Newhaven M+ loggings
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
20- 29
30- 39
40- 49
50- 59
60- 69
70- 79
80- 89
90- 99
100- 109
110- 119
120- 129
130- 139
140- 149
Mid-diameter (cm)
Proportion of total logs
RD025B
RD023B
RD023B
NH019H
NH016B
Fig. 4. Generalised diameter profiles
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
30- 34
40- 44
50- 54
60- 64
70- 74
80- 84
90- 94
100- 104
110- 114
120- 124
130- 134
Mid-diameter (cm)Proportion of sawlog volume
regrowth
typical
old.28
Appendix 6. Example of recoverable volume estimation (see section 8)
The map opposite shows an imaginary coupe (light shading) bounded on three sides by
streams. GIS analysis (see Appendix 3) has located a patch of M+ type on Tertiary basalt in
the area (dark shading), of which 80 ha are within the coupe.
Inspection of the map and aerial photos allows us to locate a 'too-steep' area (25 ha), a badly
wilted area (5 ha) and a possible M+ extension (5 ha). Ground inspection confirms that the
M+ extension is apparently good-quality myrtle. Trees along and near the internal creek
appear highly defective and stand over light horizontal scrub on wet ground (wet M- forest,
10 ha). The badly wilted area is found to carry a good stocking of myrtle seedlings and is
therefore off-limits to logging.
We can now draw a proposed operational boundary (dashed line) which encloses 45 ha.
The ground inspection also reveals that the forest within the proposed boundary is largely
'typical' DRM-rich forest (Appendix 5), while the too-steep area is largely 'regrowth-rich'
We estimate recoverable volumes by using the table below. The figures in this table are for
illustrative purposes only and are NOT based on results of the studies described in section 8.
Per-hectare volumes (m3 or tonnes/ha):
Typical Regrowth-rich
Cat. 4 sawlog (low pulpwood) 10 - 15 5 - 10
Cat. 4 sawlog (high pulpwood) 12 - 18 8 - 12
utility sawlog (low pulpwood) 12 - 18 15 - 20
utility sawlog (high pulpwood) 15 - 22 20 - 25
outspec sawlog (low pulpwood) 2 - 5 5 - 10
outspec sawlog (high pulpwood) 5 - 10 8 - 12
pulpwood (low recovery) 15 - 35 25 - 40
pulpwood (high recovery) 25 - 50 30 - 45
Estimated recoverable volumes (m 3 or tonnes) are therefore:
Op. area Steep area Op. area + steep area
Cat. 4 sawlog (low pulpwood) 450 - 675 125 - 250 575 - 925
Cat. 4 sawlog (high pulpwood) 540 - 810 200 - 300 740 - 1110
utility sawlog (low pulpwood) 540 - 810 375 - 500 915 - 1310
utility sawlog (high pulpwood) 675 - 990 500 - 625 1175 - 1615
outspec sawlog (low pulpwood) 90 - 225 125 - 250 215 - 475
outspec sawlog (high pulpwood) 225 - 450 200 - 300 425 - 750
pulpwood (low recovery) 675 - 1575 625 - 1000 1300 - 2575
pulpwood (high recovery) 1125 - 2250 750 - 1125 1875 - 3375
The operational area volumes are the base estimates. Steep area estimates are additional
volumes potentially available if use of steep-country harvest technology for selective logging
is justified by high prices for DRM..29
Too steep for
conventional logging
Wet M- forest Badly wilted but
regenerating well
Apparently good-quality
tall M+ forest, loggable.30
Appendix 7. Example of harvest cost estimation (see section 9)
The map opposite (not to scale) shows roading to an imaginary cluster of five coupes east of
the Savage River Pipeline Road. Each coupe carries a net loggable total of 2100 m 3 of myrtle
sawlog: 900 m 3 of category 4, 900 m 3 of utility and 300 m 3 of outspec grades. For simplicity's
sake, no pulpwood is to be harvested in this example. Coupes 1, 4 and 5 are easy logging
with an on-truck cost of $24 per m 3 . Coupes 2 and 3 are harder logging (timber fairly
scattered) with an on-truck cost of $32 per m 3 . Cartage is costed in the example at $11/m 3
and FT road toll at $6/m 3 . Logging, cartage and road toll costs are all directly dependent on
total volume.
Class 3 gravel roads accessing coupes are marked 'R' and cost $30 000/km, while class 4 dirt
roads within coupes are marked 'C' and cost $8000/km. Note that 'C4' through coupe 4 is
class 3 if coupe 5 is to be logged, but class 4 if not. Bridge 'B' is 10m long, all-timber
construction and costs $25 000. A $5000 special pipeline crossing is marked 'X'. Pipeline
Road repair is costed at $1000/km. It is assumed that the whole cluster can be logged over at
most two seasons, at no more than three coupes per season.
This roading plan is for illustrative purposes only. In reality, much of the roading between
coupes in the Pipeline Road area could be class 4, and within-coupe roads could be
temporary dirt tracks. Actual roading plans would be based on conditions between and
within the real coupes concerned.
With the five coupes shown there are 31 possible logging options for the cluster. If we
assume that coupe 5 would only be logged if coupe 4 was also logged, there are 23 options:
1, 2, 3, 4, 1+2, 1+3, 1+4, 2+3, 2+4, 3+4, 4+5, 1+2+3, 1+2+4, 1+3+4, 1+4+5, 2+3+4, 2+4+5, 3+4+5,
1+2+3+4, 1+2+4+5, 1+3+4+5, 2+3+4+5, and 1+2+3+4+5.
The accompanying table shows component and total costs for each option (all costs in
dollars). Total costs range from $109 to $159 per cubic metre at mill door and average $122.
Note that the spread of costs at the cheaper end is small.
The table is taken from a Microsoft Excel spreadsheet. Uncertainties can be built into the
costing either by using a probability extension for Excel (such as the software program
Crystal Ball), or more laboriously by recalculating costs for a range of inputs, e.g. 1500, 2100
and 2700 m 3 total sawlog per coupe in the example given..31
Logging Volume Roading costs: Logging Road Total Cost lowest
option (m3) Stumpage Roads X B Repair cost Cartage toll cost per m3 cost
1 2100 85500 55700 5000 0 5000 50400 23100 12600 237300 113.00 7
2 2100 85500 100700 5000 25000 5000 67200 23100 12600 324100 154.33 21
3 2100 85500 108200 5000 25000 5000 67200 23100 12600 331600 157.90 22
4 2100 85500 93200 5000 25000 5000 50400 23100 12600 299800 142.76 20
12 4200 171000 126400 5000 25000 5000 117600 46200 25200 521400 124.14 16
13 4200 171000 133900 5000 25000 5000 117600 46200 25200 528900 125.93 17-a
14 4200 171000 118900 5000 25000 5000 100800 46200 25200 497100 118.36 14
23 4200 171000 126400 5000 25000 5000 134400 46200 25200 538200 128.14 19
24 4200 171000 133900 5000 25000 5000 117600 46200 25200 528900 125.93 17-b
34 4200 171000 141400 5000 25000 5000 117600 46200 25200 536400 127.71 18
45 4200 171000 120200 5000 25000 5000 100800 46200 25200 498400 118.67 15
123 6300 256500 152100 5000 25000 5000 184800 69300 37800 735500 116.75 12
124 6300 256500 159600 5000 25000 5000 168000 69300 37800 726200 115.27 8
134 6300 256500 167100 5000 25000 5000 168000 69300 37800 733700 116.46 10
145 6300 256500 145900 5000 25000 5000 151200 69300 37800 695700 110.43 2
234 6300 256500 159600 5000 25000 5000 184800 69300 37800 743000 117.94 13
245 6300 256500 160900 5000 25000 5000 168000 69300 37800 727500 115.48 9
345 6300 256500 168400 5000 25000 5000 168000 69300 37800 735000 116.67 11
1234 8400 342000 185300 5000 25000 10000 235200 92400 50400 945300 112.54 5
1245 8400 342000 186600 5000 25000 10000 218400 92400 50400 929800 110.69 3
1345 8400 342000 194100 5000 25000 10000 218400 92400 50400 937300 111.58 4
2345 8400 342000 186600 5000 25000 10000 235200 92400 50400 946600 112.69 6
12345 10500 427500 212300 5000 25000 10000 285600 115500 63000 1143900 108.94 1
Average 121.84
Logging Volume Roading costs: Logging Road Total Cost lowest
option (m3) Stumpage Roads X B Repair cost Cartage toll cost per m3 cost
1 2100 85500 55700 5000 0 5000 50400 23100 12600 237300 113.00 7
2 2100 85500 100700 5000 25000 5000 67200 23100 12600 324100 154.33 21
3 2100 85500 108200 5000 25000 5000 67200 23100 12600 331600 157.90 22
4 2100 85500 93200 5000 25000 5000 50400 23100 12600 299800 142.76 20
12 4200 171000 126400 5000 25000 5000 117600 46200 25200 521400 124.14 16
13 4200 171000 133900 5000 25000 5000 117600 46200 25200 528900 125.93 17-a
14 4200 171000 118900 5000 25000 5000 100800 46200 25200 497100 118.36 14
23 4200 171000 126400 5000 25000 5000 134400 46200 25200 538200 128.14 19
24 4200 171000 133900 5000 25000 5000 117600 46200 25200 528900 125.93 17-b
34 4200 171000 141400 5000 25000 5000 117600 46200 25200 536400 127.71 18
45 4200 171000 120200 5000 25000 5000 100800 46200 25200 498400 118.67 15
123 6300 256500 152100 5000 25000 5000 184800 69300 37800 735500 116.75 12
124 6300 256500 159600 5000 25000 5000 168000 69300 37800 726200 115.27 8
134 6300 256500 167100 5000 25000 5000 168000 69300 37800 733700 116.46 10
145 6300 256500 145900 5000 25000 5000 151200 69300 37800 695700 110.43 2
234 6300 256500 159600 5000 25000 5000 184800 69300 37800 743000 117.94 13
245 6300 256500 160900 5000 25000 5000 168000 69300 37800 727500 115.48 9
345 6300 256500 168400 5000 25000 5000 168000 69300 37800 735000 116.67 11
1234 8400 342000 185300 5000 25000 10000 235200 92400 50400 945300 112.54 5
1245 8400 342000 186600 5000 25000 10000 218400 92400 50400 929800 110.69 3
1345 8400 342000 194100 5000 25000 10000 218400 92400 50400 937300 111.58 4
2345 8400 342000 186600 5000 25000 10000 235200 92400 50400 946600 112.69 6
12345 10500 427500 212300 5000 25000 10000 285600 115500 63000 1143900 108.94 1
Average 121.84