Page 1 of 12 25/11/11

Performance Audit Report

Yield Forecasts – hardwood plantations

Issue The Auditor General’s Performance Audit “Sustaining Native Forest Operations: Forests NSW” of April

2009 included an action item (Recommendation 3), requiring Forests NSW to “Undertake and publicly

report the results of a review of yield estimates for hardwood plantations”

This report addresses the review of yield estimates for the post-1994 hardwood plantations.

Information for this study has been sourced from the analysis contributing to the annual valuation

process.

Background

Overview

The post-1994 (1994 and younger) plantation hardwood resource established on the North Coast of

NSW by Forests New South Wales was planted to supplement hardwood sawlog supplies from a

restructuring native forest resource.

A comprehensive review of North Coast timber supply was undertaken in 2003 following changes to the

Forests NSW estate announced by the NSW Government; however, this review did not include the post-

1994 hardwood plantations. As a result, yield estimates from these plantations are the subject of this

review. The report is underpinned by the analysis undertaken for the 2011 annual valuation process.

Current resource

The northern hardwood plantation resource occurs in a large number of relatively small blocks

predominantly on the coastal lowland between Newcastle and the Queensland border. Figure 1.1 shows

their location, along with the location of current processing facilities.

Page 2 of 12 25/11/11

Figure 1: Extent of hardwood plantation resource and location of processing facilities.

Planting began in 1994 under a mandate granted to Forests NSW to establish a new hardwood

plantation resource to replace a reduction in future supply caused by removal of native forest from

production.

In order to secure sufficient land, Forests NSW made arrangements to access privately held land.

Payment for the use of this land is either by way of an annual rental (annuity) or by sharing final harvest

proceeds in proportion to the contributions made by each party to the cost of growing and selling the

crop (crop share). Approximately 10,200 hectares of plantation is on land managed under these

arrangements. A further 18,300 hectares of post-1994 hardwood plantations occur on State Forest land.

Figure 2 shows the distribution of stocked plantation area by these broad tenure classes.

Page 3 of 12 25/11/11

Figure 2: Distribution of hardwood plantation resource by land tenure.

The plantations are widely scattered and many of those on private land occur in blocks of less than 50

hectares. Figure 3 shows the distribution of blocks by planted area.

Figure 3: Distribution of hardwood plantation block size.

Annuity; 6,644;

23%

Crop Share; 3,521; 13%State Forest;

18,287; 64%

2011 Land Base Breakdown by Tenure

0

20

40

60

80

100

120

140

0-2

0

20

-40

40

-60

60

-80

80

-10

0

10

0-1

20

12

0-1

40

14

0-1

60

16

0-1

80

18

0-2

00

20

0-2

20

22

0-2

40

24

0-2

60

26

0-2

80

28

0-3

00

30

0-3

20

34

0-3

60

36

0-3

80

38

0-4

00

42

0-4

40

>5

00

Nu

mb

er

of

Blo

cks

Block Area (Ha's)

Distribution of Block Sizes

Page 4 of 12 25/11/11

A key feature of this plantation resource is the variety of species planted (a total of 22), and the very

uneven and narrow age class distribution, with the majority of the area planted between 1996 and

2004. Figure 4 and Table 1 summarise the age class distribution for the major species planted.

Figure 4: Area by age class and species

0

1,000

2,000

3,000

4,000

5,000

6,000

Are

a (H

a's)

Year Planted

Age Class Distribution

Spotted Gum

Other

Flooded Gum

Dunns White Gum

Gympie Messmate

Blackbutt

Page 5 of 12 25/11/11

Table 1: Stocked area by year of planting at 30 June 2011

The earlier plantings are of variable quality, largely due to large establishment targets being achieved in

a short time-frame. The main species planted were Blackbutt, Dunn’s White Gum, Flooded Gum and

Spotted Gum. Gympie Messmate has also been identified separately in Table 1 due to its sawlog

qualities being similar to Blackbutt and the high demand for these products.

Silvicultural policy

The silvicultural objective for these plantations is to produce sawlogs of comparable size and quality to

those from the native forests to supply the local sawmilling industry. Rotations of about 30 years are

expected to be required, with a single intermediate yield of pulpwood for chip export from thinning in

stands within an economically viable zone of markets, and with sufficient stems to support a thinning

operation.

Year

PlantedBlackbutt

Gympie

Messmate

Dunns

White Gum

Flooded

Gum

Spotted

GumOther Total

Pre94 65 0 0 0 0 2 67

1994 38 3 28 50 68 186

1995 52 51 94 9 22 227

1996 509 52 445 92 356 300 1,753

1997 1,122 45 1,140 198 878 131 3,513

1998 1,012 236 904 625 2,513 140 5,430

1999 544 105 839 362 161 105 2,117

2000 763 257 2,118 725 761 401 5,025

2001 475 63 1,592 294 915 111 3,449

2002 228 220 363 68 627 96 1,602

2003 117 22 526 1 520 27 1,213

2004 175 28 224 442 28 898

2005 112 51 142 50 7 362

2006 235 5 126 38 63 137 603

2007 270 51 6 10 337

2008 534 32 6 15 587

2009 1 45 81 127

2010 594 110 14 239 957

Total 6,844 1,228 8,612 2,548 7,301 1,919 28,452

Page 6 of 12 25/11/11

Project design Modelling undertaken as part of this review incorporated:

Updated definition of available area.

Yield information from existing growth models, scaled using actual harvest data from clearfelling

of older pre-1994 stands.

Predictions of the proportions of the final yield in logs by product size and quality, made using

stand-level equations to estimate product volumes from average stand diameter. These

equations also used data from pre-1994 clearfell operations

Yield scheduling using Remsoft® Woodstock planning system, with the objective of achieving

smoothed long-term sustainable harvest levels in the context of a broader long term hardwood

supply strategy for the North Coast.

Historically there has been limited measurement information on which to base the forecast of expected

yield per hectare from the hardwood plantation resource across the full range of site, species, stocking

and planned silviculture. The growth model developed to simulate the growth and yield of these species

is limited by the lack of data representing older ages.

This major review of modelled growth performance was undertaken to improve the reliability of

estimates and also forecast future volume from the hardwood plantation resource in the context of

supply from the surrounding native forest resource.

Model inputs

Yields

This review is based on yield tables developed from the currently available growth models, described in

Muhairwe (2003)1. As these models have limited application for older ages, the yield tables were scaled

using actual clearfell yield information generated by the harvesting of pre-1994 stands.

Stands were grouped by species, site productivity class (three Mean Annual Increment (MAI) based

classes of > 18 m3/ha/year, 12-18 m3/ha/year, and <12 m3/ha/year) and stocking (two classes of around

750 stems/ha and around 450 stems/ha at age 5) based on the then available inventory plot data.

Using the scaled yield data, thinning regimes were compiled for each combination of site class, stocking

class and species group.

Figure 5 shows that the resource is expected to be extremely variable in terms of future volume

production, reflecting the variation in species, site and silviculture.

1 “Growth and yield models for Eucalypt plantations (Future Forests, Planted Forests Division). Dr Charles K.

Muhairwe, March 2003 (Updated December 2003).

Page 7 of 12 25/11/11

Expected age 40 MAI values of Total Recoverable Volume (TRV) per hectare range from a low of 4.5 to a

high of about 14 m3/ha/year. The area-weighted average MAI of TRV at age 40 is around 9.2

m3/ha/year.

Figure 5: Distribution of age 40 MAI of TRV (m3/ha/year)

On average, mean annual total volume increment peaks at about age 30, but remains near this level

through to age 60 (see Figure 6). This means that an average harvest age of 30 years is the optimum for

maximising volume.

Figure 6: Profile of average Mean Annual Increment

0

1000

2000

3000

4000

5000

6000

7000

8000

6 7 8 9 10 11 12 13 15

Ne

t St

ock

ed

Are

a (h

a's)

MAI of TRV/ha at Age 40

0.01.02.03.04.05.06.07.08.09.0

10.0

5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59

TRV

MA

I p

er

he

ctar

e

Age (Years)

Area Weighted Average MAI Curve - 2011 Yield Curves

Page 8 of 12 25/11/11

Volume by Product

The yield table volumes were estimated in a range of log types, from high quality logs with small end

diameters (sed) of 35 cm and 25 cm to low quality logs of 25 cm sed, and pulpwood.

There is no directly comparable inventory reconciliation data available to verify these estimates.

Product class breakdown was determined by apportioning equations that were developed using product

specifications from the surrounding hardwood industries.

Yield of Thinnings

Thinning operations to date, covering 157 hectares, yielded an average of 94 tonnes/ha. Assuming that

density of the timber is between 1.1 and 1.2 tonnes per green cubic metre, this is equivalent to a

volume 78 to 85 m3/ha. For modelling purposes, an area-weighted average yield of 65 m3/ha of

pulpwood was assumed for all thinning yields.

Page 9 of 12 25/11/11

Results The results of the Woodstock modelling are shown in Figure 7. The solid blue line demonstrates a

sustainable supply of total recoverable volume (sawlog and pulp) can be produced from the ‘post 94

estate’ in the context of the strategy to deliver a sustainable supply from native forest and hardwood

plantations across the region. The basis of this strategy is to supply a constant harvest level of

approximately 300,000 m3 for 70 years. This level commences in 2024 to coincide with the transition

point set by current wood supply agreements and the corresponding reduction in supply from native

forests. By way of comparison, the dotted green line shows the modelled wood supply level for the first

rotation of the current crop only.

Figure 7: Results from Woodstock modelling demonstrating sustainable levels than can be

maintained in the context of broader regional supply (solid blue line) compared with

liquidation of current hardwood resource only (dotted green line).

Figure 8 demonstrates how the non-declining supply level from post-1994 contributes to the combined

North Coast supply.

The average age of clearfelling (see Figure 9) gradually increases from around 15 years in the early

periods, when all yield expected to be pulp, through to a peak of 44 years in 2045. This is consistent

with operations found in the current pre-1994 plantation resource. This graph highlights the rotation

lengths that will be needed to achieve the modelled sustainable harvest levels.

Page 10 of 12 25/11/11

Figure 8: Combined North Coast Wood Supply

Figure 9 : Average age of clearfelling

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

20

10

20

12

20

14

20

16

20

18

20

20

20

22

20

24

20

26

20

28

20

30

20

32

20

34

20

36

20

38

20

40

20

42

20

44

20

46

20

48

20

50

20

52

20

54

Vo

lum

e (m

3)

Combined North Coast Total Recoverable Volume Trend 2011 Post94

Smoothed

2011 Combined NC Woodflow

0

5

10

15

20

25

30

35

40

45

50

20

11

20

13

20

15

20

17

20

19

20

21

20

23

20

25

20

27

20

29

20

31

20

33

20

35

20

37

20

39

20

41

20

43

20

45

20

47

20

49

Age

(Yrs

)

Average Clearfell Age

Page 11 of 12 25/11/11

The reliability of the modelled predictions was checked in the following way:

1. Comparison of younger age predictions was made with inventory measurements from post-

1994 hardwood plantations. This demonstrate a reasonable relationship for the species

currently being sold (predominately Blackbutt and Flooded Gum).

2. Comparison of predicted volume at older ages was made with inventory information from the

pre-1994 plantation resource, collected in 1997.The plots aged between 30 and 40 years at the

time of inventory, show a total standing volume of around 190 m3/ha. Utilising the weak

correlation between stocking (number of stems per hectare) and Total Standing Volume (TSV)

which is exhibited by many plantation species, adjusting to allow for differences in stocking

produced an observed TRV estimate of about 310 m3/ha, compared to the modelled average of

300 m3/ha TSV at age 35.

Page 12 of 12 25/11/11

Conclusion The results of the modelling reflect an improvement in the yield estimates associated with the post-1994 hardwood plantations. There are significant management challenges to achieving a sustainable harvest level from the post-

1994 hardwood plantation resource. These include: the narrow age class distribution; the variety of

species; the widely scattered nature of the resource, often occurring in small, isolated blocks; the

varying quality of the resource; and the range of land tenures. However, modelling wood flow in the

context of the broader long term hardwood supply strategy for the North Coast, demonstrates that a

sustainable supply can be achieved from this resource.

Achievement of forecast yields depends on actual thinning occurring at the simulated ages to the

simulated stockings. A delay in thinning, or removal of additional trees, will reduce yields at rotation

age.

Rotation lengths of up to 45 years will also be required to achieve the modelled volumes.

Notwithstanding the uncertainties of longer term growth trends for species not well represented in the

inventory data, the observed trends for the average growth trajectory of the current yield tables appear

consistent with available inventory data.