Sawn product potential from low-quality logs from mature eucalypt forests
Gary Waugh[1]
and Russell Washusen1
Summary
The potential for the production of solid wood products from low quality logs from the native forests of far-eastern Victoria was assessed. A total of 531 saw-logs obtained at the mill from bush-logs that were graded below existing sawlog specifications were evaluated for their potential to produce sawn structural and appearance (“natural feature grade”) products. As a result of careful log preparation, 82 logs were upgraded during cross-cutting to meet existing grades for saw-logs (D grade or higher). The remaining 449 logs which were below current specifications produced recoveries that suggest potential for production of solid wood in a specialised processing system. This was evident in some forest types and log classes, particularly when hear defect was less than 20% of log diameter. There may also be potential for development of a log grading system in saw-mill log yards that can better identify logs with capacity to produce recoveries similar to logs that meet current specifications.
Introduction
The potential use of low-quality roundwood for sawn products following logging and the rehabilitation of logged-over or fire-damaged forests of the East Gippsland Forest Management Area (Figure 1) has been the subject of several government and industry reports since the 1970’s (Forest Management Plan 1995). This area includes all public land allocated for wood production, east of Nowa Nowa to the NSW border. Some of this resource is currently utilised, but most is left in the forest, to be burnt during regeneration burns or left to decay. Remaining large defective logs are an impediment to future silvicultural operations, fire suppression and harvesting. The estimates of the volume of residue available for harvesting which includes logging residue and remaining defective trees, varies between approximately 0.6 to 1 million cubic metres per annum.
Figure 1
Eastern Victoria, showing location of the East Gippsland Forest Management Area
To date, most of the effort to secure this resource for further processing has concentrated on pulp and paper manufacture, either for on-shore processing, or exported as chips for off-shore processing.
During the late 1980’s and early 1990’s, the
Victorian government initiative, the ‘Value-Added Utilisation System’ (VAUS)
study was carried out with two major objectives:
(a) To trial different harvesting systems and intensities in East Gippsland forests where forest residue was available.
(b) To supply residue wood to industry so that different
value-added processing systems could be evaluated.
A major objective of the VAUS study was to explore the potential for the production of sawn products (Waugh 1991) at a time when most of the commercial forests were being lost to the forest industries and there was a need to develop alternate resources. Efforts focussed on the ability to thin regrowth forests for potential sawlog production and the improved utilisation of low-quality roundwood. The data collected even in the early stages on improved utilisation of roundwood gave results of sufficient encouragement for processing for structural products. Opportunities also existed for developing new products such as ‘Natural Feature Grade’ timber and to develop improved drying and secondary processing practices. Subsequently more detailed investigations have been conducted with the aim of providing better knowledge of sawn product potential (Ozarska et al 1999) and also providing information on expected residual wood availability using different processing options (Forest Management Plan 1995). This information is needed to assist both industry and the Victorian state government in their future planning for processing residual wood from East Gippsland.
Methods
Sampling Strategy
The forest strata
The sampling strategy was developed in consultation with the state government agency responsible for the management of the forest. Bush-log selection was weighted to reflect projected wood flows over the 1997-2002 period from five different forest strata (Table 1).
Other forest types, which are naturally low-yielding mainly due to environmental factors, were not considered as part of the trial. The management history and major species representation in the identified forest strata are:
(a) High-elevation mixed species forests (HEMS) are the wetter
and higher areas with the major forest species being Eucalyptus nitens, E. cypellocarpa, E. dalrympleana, E. obliqua
and E. dives. Most of the
high-yielding coupes (HEMS/H) had no previous logging history. However, some
forests had been partly subject to utilisation, or may have some fire history,
particularly on drier locations where E.
sieberi is the more prevalent species. These latter sites are usually
lower-yielding (HEMS/L), often composed of mixed forest of older
highly-defective trees and regrowth. Both of these strata provide a range of
high and low-quality logs, with the higher-quality logs being used to meet
existing sawmilling requirements.
(b) Much
of the low-elevation mixed-species forest (LEMS) is low-yielding, on drier
sites and only produces low-quality logs. Some sites have been subject to
periodic wild-fires and consequently carry considerable degrade (LEMS/F). Other
sites have not been satisfactorily regenerated, often due to the high
proportion of remaining over-wood suppressing any regeneration (LEMS/C). The
main species are E. sieberi, E. obliqua,
E. cypellocarpa and a number of stringybarks. Some areas of high-quality
forest, not previously utilised or burnt, do remain and were incorporated in
this trial (LEMS/H), as some low-quality logs are produced during harvesting
operations in this stratum.
Log selection
Log selection from each of the five forest strata was conducted to reflect the range of tree species, log diameter and heart defect. In the case of heart defect this included decay, insect attack and kino development. A second tier of stratification was used in the sampling strategy to ensure that the range in log diameter and heart defect was represented (Table 1).
Table 1. Target log selection strategy
based on projected log flows from the five forest strata
|
|
|
|
Forest
strata |
|
|
|
|||||
|
Log diameter (cm) |
End defect
(%) |
HEMS/H |
HEMS/L |
LEMS/C |
LEMS/F |
LEMS/H |
Total |
||||
|
<50 |
<20 |
30 |
20 |
30 |
35 |
15 |
130 |
||||
|
|
>20 |
45 |
30 |
40 |
25 |
20 |
160 |
||||
|
Log Total |
|
75 |
50 |
70 |
60 |
35 |
290 |
||||
|
>50 |
<20 |
20 |
12 |
15 |
12 |
8 |
67 |
||||
|
|
20-40 |
25 |
20 |
25 |
20 |
12 |
102 |
||||
|
|
>40 |
10 |
8 |
10 |
8 |
5 |
41 |
||||
|
Log Total |
|
55 |
40 |
50 |
40 |
25 |
210 |
||||
|
Overall Log Total |
|
130 |
90 |
120 |
100 |
60 |
500 |
||||
Logs were selected in long “bush-log” lengths and log cross-cutting strategies were implemented aiming to maximise “saw-log” grade wherever possible, but adhering to minimum length requirements. All of the saw-logs were graded according to Victorian government specifications which recognises four log grades ‘A to D’ as shown on the log grading card shown in Figure 2.
The study was confined to logs below existing 'D' class sawlog specification. The log grading system employed for assessing wood from state forests in Victoria is based on log diameter, heart defect and surface defects. By determining diameter and heart defect and knowing the log grade (ie "D" grade), the amount of surface defect is virtually "locked in". For example, a log length of 400 mm diameter with a 100 mm pipe need only have one defective face to fall below "D" grade sawlog specification. However, a 800 mm diameter log with the same percent pipe has to have four defective faces to be below "D" class sawlog quality. An additional constraint applied to this study was that a saw-log cross-section was required to have a minimum annulus of 12 cm as solid wood around a defective heart. Therefore a log of 400 mm diameter could have a maximum diameter heart defect of 160 mm (which is 40% of log diameter).
Figure 2. Image of card used for log grading for the East Gippsland log study
In most cases it was possible to identify species, or at least, assign the log to a group of species (such as stringybarks or gum-type eucalypts). While it was important to ensure that all species were represented in the sample, it was impossible to further replicate the study based on species due to the inability to manage species availability and limits on the size of the project.
The original plan was to avoid the need to sample in the forest, but rather to rely on existing log flows. This had to be modified due to the accepted operation of splitting logs (particularly those with large defective cores) at the log dumps prior to loading trucks. This practice tended to remove certain types of logs from sampling.
Once at the saw-mill the long-length bush logs were cross-cut to maximise grade recovery and all final length saw-logs were regraded. This resulted in several of the saw-logs being upgraded to ‘D’ grade or better.
Sawing trials were carried out over a number of years so that it was possible to sample across the broad range of wet and dry-weather coupes over the course of several logging seasons. This made it possible to broadly sample the range of forest strata and also avoid confining sampling to one site within each strata.
Sawing trials
Most of the sawing trials were carried out in East Gippsland sawmills in 1990 and again in February, October and December 1997. In all trials the following procedures were adopted:
a) All major log
characteristics were recorded and their location mapped on log diagram sheets.
Characteristics recorded included log form (sweep and crook) and defects such
as the size and condition of knots (green, dead or decayed branch stubs and
surface bumps that indicated over-grown branch stubs), the incidence of loose
and tight kino, insect attack, bole damage, log end-splits, decay and spiral
grain.
b) Following grading all logs
were end colour-coded, to assist with the identification of sawn and residue
wood.
c) The sawing strategy employed
at the head-saw and resaw was recorded in all trials. Products were sawn to
meet the product requirements of the respective mills and the dimensions
varied.
d) All product dimensions were
recorded and all products graded to ASA 2082-1979," Sawn Structural
Products from Eastern Australian Hardwoods" to one of the four engineering
and structural grades or rejected. During the first trial in 1990 only the
worst defect (the “grade-limiting” defect) was recorded for each product but in
all subsequent trials all defects were recorded on product grading sheets.
Recoveries of structural grade 4 and better, structural grade 3 and better were
calculated as a percentage of log volume using nominal green product dimensions
and log volumes calculated in accordance with Victorian government methods.
e) In addition, a simulated ‘Natural Feature Grade’ (NFG) was determined. This grade included boards that would meet the specifications for the NFG grade currently used by industry. However, it also included some pieces of a quality that would meet the CSIRO Criteria for Select Grade and better (products with few if any defects). A conversion chart was used to determine NFG (Figure 3) based on the structural grade and grade limiting defect. The strictest requirement (other than the complete exclusion of decay, distortion and wane) for NFG is placed on knot size and type. In case of the early sawing trials, where only the grade limiting defect was recorded it was assumed that knots were always the first defect graded and therefore the conversion could be applied. Overall, the conversion in determining NFG from structural grades (Figure 3) has a number of limitations:
·
NFG
criteria used the length restrictions that are applied in ASA 2082-1979. Higher
recovery could be expected with docking of defects and where board lengths were
reduced.
·
Tolerances
for distortion in the structural grading criteria were greater than those for
NFG. Therefore any recorded distortion was automatically rejected and also it
would be likely that some additional boards would be rejected for NFG.
·
Board
end splits would be docked out increasing the recovery of NFG.
Figure 3. Conversion chart for natural feature grade adapted from structural grade and grade limiting defect.
Results
The number of residual and graded logs and the respective mean recoveries are given in Table 2. In total 531 saw-logs were selected for processing and following grading at the respective saw-mills 15% were found to meet existing saw-log grades ranging from high quality B grade logs through to D grade. The number of residual logs processed was 449 which was fewer than the target of 500 set in the sampling strategy (Table 1). The actual number of residual logs processed for each strata and log category is shown in Table 3. The lower number of logs and the different number of logs for each log category was primarily because of the change in log characteristics once they were cut back to mill log lengths.
The recoveries of all products (Structural grade 4 and better), Structural grade 3 and better and NFG for each strata are given in Tables 4, 5 and 6 respectively.
Table 2. Mean recovery for different sawlog grades (strata combined).
|
Means |
Number of logs |
Recovery visual grade 3 (% of log volume) |
Recovery NFG (% of log volume) |
B grade |
1 |
26.2 |
27.3 |
|
C grade |
17 |
27.5 |
20.5 |
|
D grade |
64 |
19.2 |
13.4 |
|
Residual |
449 |
13.0 |
9.3 |
All logs
|
531 |
14.3 |
10.2 |
Table 3. Number of residual logs selected for sawing trials
|
|
|
|
Strata |
|
|
|
|
|
Log diameter (cm) |
End-defect
(%) |
HEMS/H |
HEMS/L |
LEMS/C |
LEMS/F |
LEMS/H |
Total |
|
<50 |
<20 |
31 |
35 |
55 |
54 |
12 |
187 |
|
|
>20 |
|
9 |
21 |
26 |
9 |
65 |
|
Total |
|
31 |
44 |
76 |
80 |
21 |
252 |
|
>50 |
<20 |
25 |
4 |
20 |
17 |
13 |
79 |
|
|
20-40 |
18 |
13 |
28 |
21 |
11 |
91 |
|
|
>40 |
4 |
3 |
6 |
5 |
9 |
27 |
|
Total |
|
47 |
20 |
54 |
43 |
33 |
197 |
|
Overall Total |
|
78 |
64 |
130 |
123 |
54 |
449 |
Table 4. Recovery of all structural and recovery products for different log categories and forest strata from residual logs only.
|
|
|
|
Strata |
|
|
|
|
|
Log diameter (cm) |
End-defect
(%) |
HEMS/H |
HEMS/L |
LEMS/C |
LEMS/F |
LEMS/H |
Total |
|
<50 |
<20 |
38.9 |
30.4 |
27.3 |
29.1 |
20.0 |
29.9 |
|
|
>20 |
|
16.6 |
12.0 |
16.4 |
20.7 |
15.6 |
|
Total |
|
38.9 |
27.5 |
23.1 |
25.0 |
20.3 |
26.2 |
|
>50 |
<20 |
37.1 |
33.4 |
30.3 |
28.1 |
26.7 |
31.5 |
|
|
20-40 |
22.4 |
25.5 |
21.2 |
22.7 |
20.2 |
22.3 |
|
|
>40 |
15.5 |
5.8 |
30.0 |
18.9 |
3.6 |
14.3 |
|
Total |
|
29.6 |
24.1 |
25.5 |
24.4 |
18.2 |
24.9 |
|
Overall Total |
|
33.3 |
26.5 |
24.1 |
24.8 |
19.0 |
25.6 |
Table 5. Recovery of structural products (visual grade 3 or better) for
different log classes and forest strata from residual logs only.
|
|
|
|
strata |
|
|
|
|
|
Log diameter (cm) |
End-defect
(%) |
HEMS/H |
HEMS/L |
LEMS/C |
LEMS/F |
LEMS/H |
Total |
|
<50 |
<20 |
18.0 |
11.6 |
13.6 |
15.8 |
4.3 |
14.0 |
|
|
>20 |
|
5.6 |
4.3 |
6.8 |
13.4 |
6.7 |
|
Total |
|
18.0 |
10.4 |
11.0 |
12.9 |
8.2 |
12.1 |
|
>50 |
<20 |
21.8 |
12.7 |
12.5 |
16.3 |
14.4 |
16.6 |
|
|
20-40 |
13.5 |
13.9 |
13.3 |
14.8 |
11.7 |
13.6 |
|
|
>40 |
8.2 |
2.2 |
22.2 |
12.5 |
1.6 |
9.2 |
|
Total |
|
17.5 |
11.9 |
14.0 |
15.1 |
10.0 |
14.2 |
|
Overall Total |
|
17.7 |
10.8 |
12.3 |
13.7 |
9.3 |
13.0 |
Table 6. Derived recovery of natural feature grade for different log
classes and forest strata from residual logs only.
|
|
|
|
strata |
|
|
|
|
|
Log diameter (cm) |
End-defect (%) |
HEMS/H |
HEMS/L |
LEMS/C |
LEMS/F |
LEMS/H |
Total |
|
<50 |
<20 |
9.8 |
9.9 |
5.8 |
10.2 |
3.6 |
8.3 |
|
|
>20 |
|
5.2 |
2.8 |
5.3 |
7.7 |
4.8 |
|
Total |
|
9.8 |
8.9 |
5.0 |
8.6 |
5.3 |
7.4 |
|
>50 |
<20 |
20.0 |
5.0 |
11.7 |
12.5 |
11.0 |
14.1 |
|
|
20-40 |
12.4 |
13.2 |
10.0 |
10.2 |
11.2 |
11.1 |
|
|
>40 |
8.4 |
1.2 |
17.2 |
7.0 |
0.7 |
6.7 |
|
Total |
|
16.1 |
9.8 |
11.4 |
10.7 |
8.3 |
11.7 |
Overall Total
|
|
13.6 |
9.2 |
7.7 |
9.3 |
7.1 |
9.3 |
Discussion
During sawing of most East Gippsland eucalypts, about 10% of log volume is usually output as ‘recovery’ products that fail to meet target product specifications. They either have dimensions that fail to meet structural specifications (such as fence palings) or fail to meet grade requirements but are suitable for use in low-grade solid wood products such as pallets. For ‘C’ grade logs or better, overall sawn recovery is generally around 40-45% with target products (that have structural dimensions and grade) being of the order of 30-35% of total log volume. The expected total and target product recovery of ‘D’ class logs is 30-35% and 20-25% respectively. These recovery figures are similar to those reported for the recovery for visual grade 3 given in Table 2 for sawlogs that were graded better than ‘residual’ grade. They are also similar to the recoveries reported in Tables 4 and 5 for some log classes in some of the forest strata.
The results clearly show that the residual sawlogs of lower than ‘D’ grade can be processed to obtain sawn products to meet existing market requirements. Without considering the possibility of developing new grading specifications for residual logs, or applying existing local grading rules to better identify those logs that have potential for production of high quality timber, the results in Table 2 show that 15% of logs classified as residual logs meet existing sawlog specifications. The opportunity to closely examine logs in a log yard and apply a few simple rules for cross-cutting can up grade a significant number of logs.
The study also shows quite clearly that there is
still valuable product to be obtained from the remaining logs that fail to meet
the existing saw log specifications. This is particularly evident in some
forest strata, such as HEMS/H and HEMS/L, when logs were less than 50cm
diameter.
However, Tables 4 and 5 show that the average total recovery of 25.6% and structural product recovery of 13% are both 5-10% lower than sawmillers expect to obtain from ‘D’ class logs. This would make processing residual logs for sawn products a very doubtful economic proposition. ‘D’ grade sawlogs currently make up a substantial proportion of the total sawlog volume for East Gippsland sawmillers (Forest Management Plan 1995) and are processed together with higher quality sawlogs for the same products. Even though stumpage prices are lower for ‘D’ grade sawlogs, harvesting, handling and sawmilling costs are similar to higher quality logs. This results in ‘D’ grade sawlogs being a marginal proposition for most sawmillers.
Alternatives which could be put in place to make processing of residual logs for sawn products are:
1.
Development of commercially
viable processing systems for sawing lower quality logs. This could mean the
development of markedly different processing systems, probably substantially
larger in scale than existing hardwood sawmills in the region and more closely
integrating sawing, chipping and residue handling systems. This system may
include the recovery of high-quality appearance-grade timber similar to that
currently produced as natural feature grade. Unlike most sawmilling systems,
where perhaps 80-90% of wood sawn at the headsaw proceeds to resawing, in such
systems, possibly only 30-40% of a log would be resawn, with 40-50% of wood
flow being directly from the headsaw as solid residue which is chipped for
further processing for fibre products. In this system as much as 20% of the
volume may be lost through pipe and decay, which would also be segregated at
the headsaw for further processing. Resawing would lead to further solid
residue, so in terms of end-product, the volume of wood chipped for fibre
products would be greater than sawn production. Segregation by diameter would
be necessary to direct logs through optimal breakdown systems for relatively
clean small logs and large logs carrying considerable heart defect.
2.
Identification of residual
logs with the potential to provide higher recovery of sawn products. Tables 4 and 5 clearly
show that there are readily identifiable groups of logs where product
recoveries are of the same order as ‘D’ class logs. This shows that external
defect indicators are not always reliable. Residual logs from high-quality,
high-elevation forests (HEMS/H) and logs from all strata with heart defects of
less than 20% of log diameter had recovery figures close to the average for ‘D’
class sawlogs. Further analysis of the external defect data may show that there
are some key indicators that can be used to enable a better selection of
residual logs for improved sawn product recovery.
The introduction of a ‘natural feature’ appearance grade does provide some opportunity to process logs for appearance products. Natural feature grade allows increased levels of cross-grain, insect attack and kino, along with larger green knots than is accepted in other appearance products. As these features are in abundance in products sawn from East Gippsland species, the resource lends itself well to this product. However, as can be seen in Table 6, the overall proportion of wood that would meet the visual criteria for appearance products, including natural feature grade, is very low across all forest strata and log types. Attempting to saw appearance dimensions directly from a log would result in excessive rejection of pieces due to other unacceptable defects such as wane, decay and dead knots. The only technique which would enable appearance products to be manufactured would be by selecting structural dimensioned pieces which met visual requirements for appearance products and resawing down to the thinner appearance dimensions.
Conclusion
Opportunities
do exist for producing sawn products from long-length logs rejected in the
forest as failing to meet minimum sawlog specifications. Care in cross-cutting,
along with further grading in the log yard where it is easier to look closely
at defects can lead to upgrading some logs to existing sawlog specifications
where they can be processed through conventional sawing systems. Opportunities
may also exist to produce sawn products from larger-scale specialised sawing
systems for lower-quality logs. In such systems it should be possible to make
better use of residual logs from high-quality, high elevation forests and all
logs which show heart defect of less than 20% of log diameter. Opportunities
may exist to obtain appearance products from this resource, but the low
proportion of the wood that would meet visual criteria would make it very
difficult to obtain these products during the initial sawing process. Higher
appearance product recovery may be best achieved as a secondary process by
resawing structural dimensioned products that meet the visual requirements for
appearance products.
References
Forest Management Plan – East Gippsland Forest Management Area . Department of Conservation and Natural Resources, Victoria. 1995
Ozarska, B., Thompson, R., Lee, M., Northway, R.,
Ilic, J., Molenaar, S., and Turville, G. 1999. The use of Australian Hardwoods
for High Value-Added Wood Products. CSIRO Forestry and Forest Products Client
Reports 554-562
Waugh,
G. 1991. Adding value to old-growth eucalypt forests. Second ANZIF Conference,
Christchurch.