Forests take up (sequester) carbon dioxide from the atmosphere as they grow, through the process of photosynthesis. The rate at which forests grow and sequester carbon is influenced by a range of factors such as climate, topography and soils as well as bushfire, pests and disease. When a forest is mature and older, the decay and growth rates equalise, and the forest no longer sequesters additional carbon, and may even have a net loss of carbon over time. However, it does remain a living store of carbon. Victoria has an internationally significant forest and woodland estate of some 7.4 million hectares, including some forest with a very high carbon storage capacity. Indeed, in the economically productive forest, carbon stocks are high by Regional, National, and even global standards. Within productive forest, variation across the estate is considerable, as has been described by a number of authors (Grierson et al. 1991; Kaye 2008a; Norris et al. 2010). In Victoria the forest type with the greatest capacity to store carbon is mountain ash forest (dominated by Eucalyptus regnans), with typical carbon densities of 246-372 t C ha^-1. Yet when weighted by area, foothill mixed species forests contribute significantly more in terms of potential carbon stock (Kaye 2008b). Temporal changes in Victorian native forest carbon balances have been considered in the context of forest management practices between 1945-2000 by Flinn et al. (2007). Over this period an estimated total of 110 million m³ (equivalent to about 2 million m³ yr^-1) of forest produce were removed from Victoria’s forests, with the bulk of this coming from eucalypt mixed species and ash (mountain and alpine) forests. Sawlogs, and to a lesser extent pulpwood, were the significant output over this period, with other products such as sleepers and fuelwood much more variable. The study also highlighted the high priority given to regeneration success in Victoria following timber harvesting over this period, which has ensured a strong and enduring recovery of carbon stores. Typically, carbon sequestration rates are reflected by mean annual increments of 5.5 m³ ha^-1 yr^-1 over 80 years for ash, and 3.85 m³ ha^-1 yr^-1 over 80 years for mixed species (Flinn et al. 2007).

With most past investigations of forest carbon associated with above-ground biomass (AGB) studies within the actively harvested, productive forest it has been difficult to develop a broader understanding of statewide carbon stocks. Victoria, through the Department’s (DSE and DEPI) LandCarbon project, was the first state to model the amount of carbon stored across this broader estate, including public forests, parks, and reserves, and to attempt to understand the impact of drivers of change on carbon stocks. LandCarbon, as reported by Norris et al. (2010), used event-based modelling to better understand (at the State scale) carbon flux, by accessing historical event data. Combining spatial data of forest management (using a geographical information system) with the National Carbon Accounting Toolbox and its carbon accounting model FullCAM (v3.13) an indicative set of carbon accounts were derived. Overall, Victoria’s public forests contained about 2,840 million tonnes of CO², equivalent to over 20 years of Victoria’s annual average emissions (Fairman and Law 2011). At $20-30 per tonne, this reflects an implicit value of carbon storage of $55-80 billion. In the event that the impact of forest management is a component of any future carbon markets, significant changes in stored carbon (both positive or negative) could be enormously important. In relation to understanding the possible impact on this storage simulations showed that fire events, particularly large-extent wildfires most heavily influenced carbon stocks. Harvesting related disturbance had a greater direct effect in reducing carbon stocks at particular locations than either wildfires or prescribed burning, yet the relative effects of wildfires was greater due to the spatial extent of these events. Even so, these forests are well adapted to disturbance and regrowth quickly replaces the biomass consumed (Adams 1996, Flinn et al. 2007). Consequently, in the longer term, modelling found that the disturbance effect was largely transient even if corresponding emissions were significant. The soil carbon pool was found to be relatively stable. The LandCarbon project highlighted that the estimation of forest carbon stocks, their change and impacts of disturbance was a large, complex and evolving task. It also showed that very small changes across the large stock of Victoria’s forest carbon can be highly significant compared with the State’s emissions. During the LandCarbon project a number of issues were identified with the construct, application and testing of the FullCAM model. This was demonstrated when FullCAM was compared to observed site-based estimates (eg. Roxburgh (2010), Hammersla (2010)) and found to systematically under-predict native vegetation biomass where vegetation carbon was observed to be greater than 100 t C ha^-1, corresponding to open, tall-open, and low-open eucalypt forest. Snowden et al. (2000) noted that in the estimation of biomass using allometrics there were few studies where total plot biomass had actually been measured to understand these uncertainties, with some exceptions, such as Stewart et al. (1979), Baker and Attiwell (1985), and Feller (1980). In relation to open eucalypt forest, Stewart et al. (1979) destructively sampled 31 trees (three species) from eastern Victoria, weighing the tree components to develop species-specific allometric equations for tree biomass. In 2010, to extend this limited sampling, the Comprehensive Carbon Assessment Program (CCAP) was used to validate above-ground carbon estimates of eucalypt dominated forest more broadly in Victoria, using plot-based destructive sampling and weighing of whole trees to determine AGB and develop species-specific allometric equations. A total of 337 trees across eight native forest plots, with a maximum diameter of 143 cm, at a height of 1.3m (D130), were harvested. Biomass equations for eleven eucalypt species were derived from these overstorey trees of variable age, height and diameter classes. The study was the most detailed of above-ground tree biomass available for open eucalypt forest, as reported in Bi et al. (2015). These equations reinforce that there is considerable scope to improve FullCAM to better reflect more productive forests.

Most carbon accounting protocols are conservative, and currently assume that all the carbon in a forest is emitted when it is harvested. Timber harvesting does reduce the carbon stored in a forest. However, some carbon remains stored in roots and soil, some remains stored in the harvested wood products (HWPs), and some is released back into the atmosphere. There are few studies in Victoria that have considered these inconsistencies in any detail. Ximenes et. al. (2016) reported on a plot-based study, using destructive sampling and allometric equations that tracked the fate of carbon from representative native forests in New South Wales and Victoria. Victorian mountain ash forests near Toolangi were sampled to provide an insight into carbon balances where the forests were managed for multiple use (“production”) and conservation only. In relation to AGB it was found that trees with the largest diameters (ie. diameters of greater than 100 cm measured at D130) were not generally well represented in existing biomass equations. Consequently, the study found that biomass estimates of mature native forest stands that were not based on directly-weighed biomass (and that include trees with the largest DBH range) were generally not reliable and tend to overestimate biomass. The study also found that mountain ash had the highest commercial recovery of all the species, with all low-quality sawlogs being processed into pallets, and high-quality sawlogs being processed into HWPs such as flooring, mouldings, veneers, and furniture and structural timbers. Carbon storage estimates of these HWPs were based on the expected service life and rate of decay in landfill, using the latest research findings on the dynamics of the decomposition of HWPs in landfills. This research had demonstrated that C in landfill-HWPs (other than paper products) could be considered to be stored for the long-term. The overall conclusion of this exhaustive study was that the relative differences in the Green House Gas (GHG) balance of these native forests managed for “production” or “conservation” only did not warrant policies that aimed to halt native forest management for wood production, while also suggesting that the was scope to improve GHG outcomes through continued harvesting.

Other approaches to estimating AGB have been able to use data collected from destructive sampling for validation. The use of terrestrial laser scanning (TLS) is one such approach. Over the last decade methods have been developed for estimating AGB from terrestrial laser scanning (TLS) data where tree structures are laser scanned to produce cloud models. The TLS-derived cloud data can be used to build structures to estimate AGB, which can be used to address current uncertainties in allometric and Earth observation methods that quantify AGB (Demol et al. 2022). In Victoria, 65 of the 337 CCAP trees that were destructively sampled were scanned by TLS. Calders et al. (2015) reported on this approach in estimating AGB using TLS in eucalypt open forest. Using two of the destructively sampled CCAP plots and the data from the 65 sampled trees on these plots, TLS was able to estimate AGB to within 9.68% (69.02t cf. 75.7t). Existing allometrics for these species (at the time) under-estimated by 29.8%-36.57%. Additionally, the use of TLS data also enabled others aspects of AGB to be explored. For example, the TLS data for the CCAP plots showed, on average, that 80% of the AGB was found below 60% of the canopy (tree) height.

CCAP was also involved in collaborative research to estimate AGB and below-ground biomass (BGB) in environmental plantings of native species in low-medium rainfall agricultural regions of Victoria. In a collaboration with CSIRO, 11 Victorian sites of ages 8 to 16 years were used to develop three different types of allometric estimations of AGB, with diameter at a height of 130cm or 10cm the explanatory variable, as follows:

• Generic universal growth habit - trees (D130 <100cm), tall shrubs (D10 <30cm), small shrubs (D10 <35cm)
• Generic genus and growth-habit – trees (various, D130), shrubs (various D10)
• Species-specific - eucalyptus trees (various, D130), acacia trees (various, D130), acacia shrubs (various, D10)

The species-specific AGB allometric equations were the most accurate for site-based predictions, while the generic equations were better applied for regional estimates. The collated empirical observations of AGB were used to improve yield curves used in FullCAM for use with environmental plantings. Methods and results are detailed in Paul et al. (2013). Three of the Victorian sites were also used to research the quantity of biomass stored below ground. The root biomass study excavated roots within 10m x 10m plots, with between 96 and 371 trees or shrubs harvested per site. Generic root allometric equations were developed with D130 or D10 used as an exploratory variable for non-eucalypts (D130 <40cm, D10 <46cm) and eucalypts (D130 <52cm, D10 <63cm). Methods and results are detailed in Paul et al. (2014). CCAP was also involved in researching the quantity of soil carbon, specifically to develop more cost-effective methods of determining soil organic carbon content. An approach using mid-infrared (MIR) spectroscopy to scan soils was found to be able to predict the proportion of soil organic carbon with different classes of decomposability in a cost-effective manner. Methods and results are detailed in Madhaven et al. (2017), England et al. (2016), and Paul et al. (2018). Overall, in Victoria, the research and study of AGB and BGB at the site-level using destructive sampling in open eucalypt forests, mountain ash forests, and mixed-species environmental plantings has provided a better basis to develop more accurate estimates of carbon storage and flux at this scale. Additionally, these studies have progressed modelling at the state-scale by providing the means to reduce uncertainty associated with the use of different carbon accounting approaches, such as FullCAM. Enhanced technologies, including TLS and MIR, also provide the possibility of improvement in the understanding and management of carbon storage and flux, which in a future carbon dominated economy is likely to be an increasing imperative.

The Impact

From a forestry perspective, concern over the increasing concentrations of greenhouse gases in the atmosphere largely centres on the quantity of carbon sequestered or accumulated in carbon sinks such as native or plantation forests. The extant and health of forests is an important consideration in determining their ability to trap carbon as part of photosynthesis, store carbon as part of their organic cellular structure, and give off carbon during respiration. Generally, trees accumulate or retain carbon until they age or die and start to decay. Providing forests are able to maintain a level of growth and regeneration they will sequester carbon for potentially hundreds of years.

In 2010, Victoria was the first state to model the amount of carbon stored across its public estate, including public forests, parks, and reserves, and to attempt to understand the impact of drivers of change on carbon stocks. As reported by Norris et al. (2010) and Fairman and Law (2011), the carbon mass in Victoria’s public forests is stable at about 2,840 million tonnes of CO2 (775M t C). State-scale simulations showed that carbon stocks were most heavily affected by wildfire events, with lesser impacts from fuel reduction burning and the harvesting of sawlogs. Although significant volumes of forest produce have been removed from Victoria’s native forests (up to 2 million m³/yr), the harvesting outcome reflects

• that a significant proportion of carbon has been sequestered in building timbers and furnishings in housing.
• the general success in regenerating harvested coupes and protecting the forest from devastating wildfire. (The 2000s have seen three major fire events burning in excess of one million hectares but the state-scale simulations took account of these events)

It is not surprising that the LandCarbon project simulations showed that carbon stocks were most heavily affected by wildfire events, with lesser impacts from fuel reduction burning and the harvesting of logs. The large-extent wildfires most heavily influenced carbon stocks because even though carbon emission was comparatively small the fires covered large areas. In a wildfire there is scorching and some death of trees and understorey, dependent on fire intensity and the species impacted. There is also some level of bark, leaf, branch and litter consumption by the fire, but most of the stored carbon remains on site in the form of either dead or living standing trees. Over time dead material will decay and emit carbon, and regeneration and resprouting trees will accumulate carbon. Native forests are well adapted to fire disturbance and regrowth quickly replaces the biomass consumed (Adams 1996). Consequently, with wildfire disturbance carbon storage is reduced initially but is restored over time dependant on the rate of forest recovery. Similarly, while fuel reduction burning initially results in some carbon emissions there is usually a more rapid recovery of carbon storage, due to lower fire intensities and less mortality. The area affected is usually much less than with wildfires.

Harvesting of sawlogs on the other hand involves significant loss of carbon, but depending on how the wood is used, the carbon it contains may remain stored in wood products for a considerable period. Harvesting also involves a regeneration operation that often uses slashburning to produce a seedbed. There is also the expenditure of fossil fuel carbon to operate harvesting, transport and processing equipment. Consequently, determining a carbon budget for harvested areas is quite involved when all factors are considered. Over time harvested areas do accumulate carbon as regeneration grows and develops in much the same way as a forest that has been killed by wildfire. The specifics will depend on many factors, not least being the forest type, the amount of forest that is harvested, what products are produced from the logs, and how successfully the forest is regenerated. Usually the area harvested is small in comparison to fire. Ximenes et. al. (2016) reported on a plot-based study that tracked the fate of carbon from representative native forests in New South Wales and Victoria. In Victoria, mountain ash forests near Toolangi were sampled to provide an insight into carbon balances where the forests were either managed for multiple use (“production”), or solely for conservation. The study found that the harvested mountain ash had the highest commercial recovery of all the species/sites sampled, with low-quality sawlogs being processed into pallets, and high-quality sawlogs being processed into wood products such as flooring, mouldings, veneers, and furniture and structural timbers. The overall conclusion of this exhaustive study was that the relative differences in the greenhouse gas balance of these native forests managed for “production” or “conservation” only were very much in favour of forests managed for production, especially when socio-economic factors were considered, while also suggesting that there was scope to improve greenhouse gas outcomes through continued harvesting. Additionally, energy considerations of timber as an alternative to raw materials such as steel, concrete and aluminium illustrates the significant carbon storage benefits of using timber alternatives.

References

Adams MA (1996). Distribution of eucalypts in Australian landscapes: landforms, soils, fire, and nutrition. In: Attiwill PM and Adams MA (Eds.) Nutrition of Eucalypts. CSIRO Publishing, Collingwood, Australia, 61-76.

Baker TG and Attiwill PM (1985). Above-ground nutrient distribution and cycling in Pinus radiata and Eucalyptus obliqua forests in southeastern Australia. Forest Ecology and Management 13: 41-52.

Bi H, Murphy S, Volkova L, Weston C, Fairman T, Li Y, Law R, Norris J, Lei X, Caccamo G (2015). Additive biomass equations based on complete weighing of sample trees for open eucalypt forest species in south-eastern Australia. Forest Ecology and Management 349: 106-121.

Calders K, Newnham G, Burt A, Murphy S, Raumonen P, Herold M, Culvenor D, Avitabile V, Disney M, Armston J, and Kaasalainen M (2015). Nondestructive estimates of above-ground biomass using terrestrial laser scanning. Methods in Ecology and Evolution, 6(2), 198-208.

Demol M, Verbeck H, Gielen B, Armston J, Burt A, Disney M, Duncanson L, Hackenberg J, Kukenbrink D, Lau A, Ploton P, Sewdien A, Stovall A, Takoudjou SM, Volkova L, Weston C, Wortel V, and Calders K (2022). Methods in Ecology and Evolution, 13, 1628-1639.

England JR, Paul KI, Cunningham SC, Madhavan DB, Baker TG, Read Z, Wilson BR, Cavagnaro TR, Lewis T, Perring MP, Herrmann T, Polglase PJ (2016). Previous land use and climate influence differences in soil organic carbon following reforestation of agricultural land with mixed-species plantings. Agriculture, Ecosystems and Environment 227, 61–72.

Fairman T, and Law R (2011). What does the National Carbon Accounting Toolbox tell us about the current carbon stocks of Victoria’s ecological vegetation classes? Ecological Society of Australia 2011 Annual Conference – Ecology in changing landscapes, Hobart, Tasmania.

Feller MC (1980). Biomass and nutrient distributionin two eucalypt forest ecosystems. Australian Journal of Ecology 5: 309-333.

Flinn DW, Squire RO, and Wareing K (2007). Forest Management in Australia: Implications for Carbon Budgets. In: NCAS Technical Report no. 32, 102-145 (Eds. RJ Raison and RO Squire), Australian Greenhouse Office, Canberra.

Grierson PF, Adams MA and Attiwill PM (1992). Estimates of carbon storage in the above-ground biomass of Victoria’s forests. Australian Journal of Botany 40: 631-640.

Hammersla W (2010). Verification of the Australian National Carbon Accounting System and the robustness of Australian allometrics. Honours Thesis. The University of Melbourne, Victoria.

Kaye WJ (2008a). Estimation of tree above-ground biomass across native forests of Victoria. MSc Thesis. The University of Melbourne, Victoria.

Kaye WJ (2008b). Modelling of carbon sequestration in native forest ecosystems in Victoria. MSc Thesis. The University of Melbourne, Victoria.

Madhavan DB, Baldock JA, Read Z, Murphy SC, Cunningham SC, Perring MP, Herrmann T, Lewis T, Cavagnaro TR, England JR, Paul KI, Weston CJ, Baker TG (2017). Rapid prediction of particulate, humus and resistant fractions of soil organic carbon in reforested soils using infrared spectroscopy. Journal of Environmental Management 193, 290–291.

Norris J, Arnold S and Fairman T (2010). An indicative estimate of carbon stock on Victoria’s publicly managed land using the FullCAM carbon accounting model. Australian Forestry 73, 209-19.

Paul KI, Roxburgh SH, England JR, Ritson P, Hobbs T, Brooksbank K, Raison J, Larmour JS, Murphy S, Norris J, Neumann C, Lewis T, Jonson J, Carter JL, McArthur G, Barton C, and Rose B (2013). Development and testing of allometric equations for estimating above-ground biomass of mixed-species environmental plantings. Forest Ecology and Management. 310, 483-494.

Paul KI, Roxburgh SH, England JR, Brooksbank K, Larmour JS, Ritson P, Wildy D, Sudmeyer R, Raison RJ, Hobbs T, Murphy S, Sochacki S, McArthur G, Barton C, Jonson J, Theivyanathan S, Carter JL (2014). Root biomass of carbon plantings in agricultural landscapes of southern Australia: Development and testing of allometrics. Forest Ecology and Management. 318, 216-227.

Paul KI, England JR, Baker TG, Cunningham SC, Perring MP, Polglase PJ, Wilson B, Cavagnaro TR, Lewis T, Read Z, Madhavan DB, Herrmann T (2018). Using measured stocks of biomass and litter carbon to constrain modelled estimates of sequestration of soil organic carbon under contrasting mixed-species environmental plantings. Science of the Total Environment 615, 348–359.

Roxburgh S, England J, Paul K (2010). Developing capability to predict biomass carbon in biodiverse plantings and native forest ecosystems. Report by CSIRO for Department of Sustainability and Environment, Victoria, 53.

Snowdon P, Eamus D, Gibbons P, Khanna P, Heith H, Raison J, Kirschbaum M (2000). Synthesis of Allometrics, Review of Root Biomass and Design of future Woody Biomass Sampling Strategies. NCAS Technical Report 17. Australian Greenhouse Office, Canberra.

Stewart HTL, Flinn DW, and Aeberli BC (1979). Above-ground biomass of a mixed eucalypt forest in eastern Victoria. Australian Journal of Botany 27: 725-740.

Ximenes F, Bi H, Cameron N, Coburn R, Maclean M, Sargeant D, Mo M, Roxburgh S, Ryan M, Williams J, and Boer K (2016). Carbon stocks and flows in native forests and harvested wood products in SE Australia. Report to FWPA PN 285-1112.