The report

This report attempts to lay the groundwork for an on-going effort to account for the carbon benefits of EU certified topical timber imports. Understanding the carbon benefits of certified sustainable tropical timber has the potential to drive demand and investment, potentially leading to increased carbon sequestration.

The reality is that calculating an accurate carbon footprint for certified timber is near impossible—there simply isn’t enough data. Certification frameworks don’t currently collect the data needed to facilitate carbon accounting, and there is a great deal of uncertainty involved in scaling existing one-off carbon assessments of timber operations to a global scale. Despite these difficulties, this report presents a rough model for carbon accounting using the data that exists, with the expectation that the model and results will evolve as more data becomes available. It will also outline the challenges and data gaps that need to be addressed to refine quantitative estimates.

The model draws on the best available data (where possible from peer reviewed studies) and builds on several assumptions. The analysis considers the potential impact of EU demand for certified timber on carbon pools in the tropical forest, the carbon emissions associated with harvesting, converting, and transporting certified tropical timber to the EU market, the carbon sequestered in certified tropical timber at point of delivery to the EU, and the potential for carbon emission reductions resulting from substitution of certified tropical timber for alternative non-wood products.

The results should be interpreted as approximate, produced with the goal of estimating emissions on an order of magnitude level. They show the likely contribution (both positive and negative) of various processes involved in supplying the EU with tropical timber to carbon emissions.

The report is written by Rupert Oliver, Forest Industries Intelligence Ltd

A note on units: Kg C = Kilo (10³) grams of carbon, Mg C = Mega (10⁶) grams of carbon, Gg C  = Giga (10⁹) grams of carbon, Tg C = Tera (10¹²) grams of carbon, Pg C = Peta (10¹⁵) grams of carbon. Mass of carbon is converted into CO2 equivalent by multiplying by 44/12 (since one atom of CO2 is composed on 1 carbon atom of mass 12 and two oxygen atoms each of mass 16). Hence 1 kg C = 3.667 kg CO2 eq.

The absence of data on carbon benefits in forest certification schemes necessitates pulling together a number of existing studies and data, and making significant assumptions in order to develop a model. Forest certification systems, which collect significant data on certified forestry operations unfortunately aren’t designed to provide a metric of carbon emissions from forest operations and supply chain activity. Neither FSC nor PEFC (the main forest certification standards) include any requirement for certificate holders to measure the carbon impact of operations, and neither framework has specific international requirements aimed at reducing carbon emissions (see Appendix 1 for details on current difficulties in extending certification data to carbon calculations).

Any carbon accounting requires proxy measures, and there are considerable data limitations and technical barriers to making even the most tentative assessment of certification on tropical forest carbon impacts. To develop our model we compile research on three specific areas of forestry: Long Term Impacts of Felling Cycle, Life Cycle Analysis (which includes felling operations, processing, and transport), and Substitution Benefits (the carbon advantages of sustainable timber over other materials).  Of note is the focus only on above-ground carbon stocks. Substantial stocks of carbon are locked in tropical soils as Soil Organic Carbon (SOC), and there is little understanding of the impact of different harvesting operations or conversion events on these stocks (estimates of the total carbon stored as SOC in the tropics vary from less than 200 Pg to 500–650 Pg).[1]

Long Term Impacts of Felling Cycle

Long-term impacts of felling cycle refers to the ability of forest carbon stocks to recover following timber harvesting. RIL (Reduced Impact Logging) and CL (Conventional Logging) operations are the main subject of most studies. For our purposes RIL serves as a proxy for certification. RIL and CL vary in the quantity of biomass extracted (and damaged), and the length of time between subsequent cuttings. How these factors contribute to the long-term health of a forest and its vulnerability to land-use changes  is likely to have a significant impact on the carbon emissions associated with supply of tropical timber products.

A review of research reveals mixed results on the question of whether RIL offers real carbon benefits compared to CL. A 2015 meta-analysis notes that modelling studies have implied that the carbon benefits of RIL may be magnified over 40-60 years owing to a reduction in residual damage, but that there is insufficient empirical data to confirm the hypothesis that RIL forests recover more quickly than CL forests.[2]

Other researchers are more confident on the significant long-term carbon storage benefits of RIL in tropical forests. A 2016 analysis by Sasaki et al that modelled long-term impacts of RIL compared to CL posits potentially large carbon benefits of RIL operations. They suggest that the negative consequences of “premature re-entry logging”—common in the lax regulatory environment of tropical regions—lead to the repeated loss of carbon stocks in the affected concession until all marketable trees are harvested.[3]

Sasaki et al created a model based on the assumption that RIL operations log only once at the beginning of the 40-year period, while in CL operations there are additional harvests in years 5 and 20 (see Figures X and Y below). Drawing on a wide range of studies, Sasaki et al estimate an average initial carbon stock in tropical production forest of 174.0 ± 11.6 Mg C ha⁻¹ (± is for confidence interval of 90%) prior to logging, and project that it declines to 121.8 Mg C one year after RIL logging (point P₁ in Figure X) but recovers to the pre-logging level or higher by the fortieth year if premature re-entry logging is prevented (point A). In contrast, under CL with two premature re-entries, carbon stocks continue to decline until all merchantable trees are harvested (point D). If premature re-entry logging is allowed, carbon stocks decline by 19.5% and 34.3% for first and second premature logging, respectively.

Sasaki et al estimate that preventing first and second premature re-entry logging across all production forest in tropical regions could reduce emissions by 279.9 and 493.7 Tg C year⁻¹ (1026.4–1810.4 Tg CO2 year⁻¹) respectively, in addition to removing (sequestering) −8.2 Tg C year−1 (−30.1 Tg CO2 year⁻¹) from the atmosphere (Figure Y).

These figures suggest that the emission reductions resulting from reduced impact logging in certified forest operations may be considerable.

In the absence of other research on the likely long-term effects of certified operations on forest carbon stocks at landscape level in the tropics, the Sasaki et al study provides the basis of assumptions in our calculations.

Operational and Lifecycle Costs

Calculating the carbon costs of the physical operations of cutting, transporting, and processing certified tropical timber requires an understanding of a number of processes. Both research on the carbon emissions of RIL and CL forestry practices as well as lifecycle assessments  (LCAs) of timber operations can help inform the development of a model for certified tropical timber emissions.

In practice there have been a relatively limited number of comparative assessments of carbon emissions of harvesting operations in certified and uncertified forest. These studies have largely focused on RIL vs. CL with ambiguous results.[4] A meta-analysis of the impacts of tropical selective logging on carbon storage and tree species richness came to the tentative conclusion that “RIL reduces the negative impacts of logging on tree damage but does not support suggestions that RIL reduces loss of aboveground biomass or tree species richness.”[5] It showed that “the volume of wood removed per hectare was by far the best predictor of changes in biomass in response to timber harvest,” suggesting that while RIL results in lower carbon emissions per hectare exploited, RIL and CL lead to the same level of emissions per unit of timber volume extracted. A contradicting study claims that improved logging and skidding practices have the potential to reduce harvesting emissions by 44% (selective logging emits 6% of tropical greenhouse gases annually).[6] Needless to say the data doesn’t clarify reality.

A study in Gabon comparing an FSC-certified concession with an uncertified CL concession on the basis of logging damage, above-ground biomass (AGB), and tree species diversity and composition showed that FSC resulted in less collateral damage per tree cut (averages of 9.1 and 20.9 other trees damaged per tree cut for FSC and CL respectively).[7] Like the meta-analysis, this implied significantly lower carbon emissions on a unit area basis. However, when expressed as impacts per timber volume extracted, the values did not differ between the two treatments.

Another study in East Kalimantan directly comparing the carbon emissions of operations on three FSC certified concessions with those on six non-FSC certified concessions concluded that FSC certified concessions did not have lower overall CO2 emissions from logging activity.[8] However it did find evidence of a range of improved practices using other field metrics. The authors comment that “one explanation of these results may be that FSC criteria and indicators, and associated RIL practices, were not designed to achieve overall emissions reductions. Also, commonly used field metrics are not reliable proxies for overall logging emissions performance. Furthermore, the simple distinction between certified and noncertified concessions does not fully represent the complex history of investments in improved logging practices.” They conclude by suggesting the introduction of a new term called RIL-C that encompasses practices that result in measurable emissions reductions, and call for certification standards to embrace RIL-C requirements.

Assessing the actual emissions associated with the extraction, processing, and distribution of wood products sourced from certified forest operations is typically derived from formal LCA studies (of which carbon footprint is just one aspect). LCAs quantify all material and energy flows and emissions associated with a defined “functional unit” (such as “1 m3 of Malaysian meranti plywood”) at a specified part of its life cycle. LCA data may be made available through stand-alone reports (ideally prepared independently in conformance to ISO standards—ISO14040/44), as verified datasets in large harmonized databases such as Gabi or SimaPro, or as Environmental Product Declarations (EPDs—independently verified and registered documents issued voluntarily by product manufacturers).

A review of the LCAs that might inform the current analysis (see table below) reveals significant limitations. To date there has been no attempt yet to apply a single harmonized LCA methodology to assess the carbon footprint of different wood products on a global or regional scale. There are no formal EPDs in circulation covering any tropical wood product and the GABI database contains no data specific to a tropical hardwood product. LCA reports on tropical wood products are scarce, and those that exist are narrow in scope (limited geographic coverage or few data points). None of the LCA studies related to tropical hardwood products accounted for the impact of tropical timber harvesting on forest carbon pools. LCAs that do consider forestry typically assume that harvesting is carbon neutral over the long term due to regrowth in the forest and only account for direct emissions associated with extraction machinery. They don’t tend to address the question of how to assign carbon debt or credits.

While tropical hardwoods are poorly covered by LCA analysis, there has been comprehensive analysis of American hardwood products as part of a large study by PE International (now Thinkstep) commissioned by the American Hardwood Export Council (AHEC) in 2012. There are strong parallels between the harvesting and production of tropical and American hardwoods—both involve selection harvesting of managed semi-natural forests covering large geographic areas, the transport distances to transport logs from forest to processing location are at least comparable, and the production processes are similar. Our model makes a number of assumptions to translate these LCAs to certified tropical timber.

Table: Summary of carbon footprint results for tropical hardwoods & comparable temperate products

Substitution benefits are situations where certified tropical timber replaces other industrial materials, a proportion of which may require more fossil-fuel energy to produce, be non-renewable, and not offer the carbon sequestration benefits of certified timber. Several studies have attempted to quantify the carbon mitigation benefits of using timber products in place of non-timber products.[10] The most recent, comprehensive, and relevant to the current analysis was undertaken by Leskinen et al for the European Forest Institute in 2018. It notes that the quantification of substitution benefits is not straightforward and involves many uncertainties. To overcome this complexity, Leskinen et al calculate an average “substitution factor” from a meta-analysis of 51 studies comparing the carbon footprint of utilizing wood-based products compared to non-wood products in a wide range of different applications (433 in total). The study indicated that wood products were linked to lower net emissions than substitute products.

Overall, the 51 reviewed studies suggest an average substitution effect of 1.2 kg C / kg C, which means that for each kilogram of C in wood products that substitute non-wood products, there occurs an average emission reduction of approximately 1.2 kg C.” This is more conservative than the substitution effect derived by Sathre & Gustavsson[11] in an earlier 2009 report which draws on a meta-analysis of 20 studies to conclude that in wood product, the displacement factors range from a low of -2.3 to a high of 15.0, with an average of 2.0 (and most lying in the 1.0 to 3.0 range).

Complete Felling Cycle Calculations

Using the figures from the Sasaki et al model in our own calculations provides a basis for a very tentative estimate of the potential for certification to contribute to maintenance of tropical forest carbon stocks over the complete felling cycle, but requires several assumptions:

• The Sasaki model uses globally averaged figures, but the high variability in tropical forest carbon stocks and general lack of research into the carbon impacts of tropical forest operations on carbon stocks imply that actual emissions in tropical forests could vary greatly.
• The Sasaki model compares RIL and CL forestry practices, while we attempt to quantify the impacts of certification. Given the reality of FSC and PEFC certification standards implemented in the tropics, particularly the emphasis on maintaining forests’ long term productive capacity, the need to protect environmental values, and the requirement to adhere to management plans, we assume that certification prevents premature re-entry logging in the areas it covers just as RIL does.
• The Sasaki model assumes significant re-entry in its calculations for CL. Rising pressure on land and forest resources and the significant governance challenges in many tropical countries support the idea that in the absence of certification, there is likely to be considerable pressure to harvest at the highest levels of intensity.

Data in the IDH Timber Report (2019)[12] shows that 6.2% of production forest in the tropics is certified implying that certification may be effectively reducing emissions by up to 30.61 Tg C year^-1 (6.2% X 493.7 TgC year^-1), equivalent to around 110 Tg CO2 year⁻¹. Projecting these emissions reductions onto EU certified imports, suggests that between 50% (55 Tg CO2 year^-1) and 80% (88 Tg CO2 year^-1) of these reductions might be attributable to the EU tropical wood trade.

LCA Calculations

In order to calculate an estimated cradle to EU carbon footprint for EU certified tropical primary wood product imports we make a number of assumptions that draw on existing LCA data to create rough estimates for average footprints of each step of the process that brings tropical timber products from forests to EU ports. Where the data suggests a range of possible values our model errs on the side of undervaluing carbon benefits and overvaluing costs. Note that carbon pools are not considered in this calculation, and taken up in the complete felling cycle portion of this report.

Estimating the carbon cost of certified tropical timber’s lifecycle beyond the long-term effects of the felling cycle requires combining different studies and LCAs from both tropical and non-tropical forestry research. There are significant differences across studies and compiling them requires a number of assumptions:

• Each LCA shown in the research section has its own scope, and while most of the LCAs analyzed are ‘cradle to gate’ (and therefore include forestry operations), it is not always clear exactly which operations are included (e.g. just the final harvest or including other silvicultural interventions).
• Wood product LCAs are highly sensitive to geographic location, partly because of variation in transport modes and distances, but often even more critically due to variations in both the efficiency of processing and the extent of dependence on fossil fuels (as opposed to renewables) in the energy mix. The carbon footprint of wood product manufacturing in areas where there is high dependence on solar, biomass or hydro-electric can be orders of magnitude less than the carbon footprint of functionally equivalent products manufactured in areas where there is high dependence on coal, for example.
• Allocation of environmental impacts to the different co-products of wood processing activities may vary widely between studies. Allocation by economic value will produce very different results compared to allocation by unit mass or volume of final product.
• LCAs for specific wood products are highly sensitive to assumptions made about the use and treatment of waste. Much of the waste from wood manufacturing comprises offcuts, chips, and sawdust, which in some cases may be sent to landfill, diverted to other product streams, or burnt for biomass (which may generate carbon credits).
• Species and specification of final product greatly impact the carbon footprint of forest products with great variation in energy intensive processes like kilning. This is clearly illustrated by the AHEC-PE study on US hardwoods (see appendix 2 for a full review of how these may impact the model).

Table:  Assumptions to estimate total cradle to EU carbon footprint of EU imports of certified tropical primary wood products

The estimated LCA for tropical timber is projected onto EU tropical timber consumption to create a cradle to EU estimated footprint. The calculations below attempt to estimate in very broad terms (orders of magnitude) the direct emissions and carbon sequestration associated with the processing and delivery of tropical hardwoods to the EU market.

Table: Estimated cradle to EU carbon footprint of EU certified tropical primary wood product imports

This rough calculation shows that the total carbon emissions associated with delivery of all certified tropical timber products at port of import in the EU may be in the order of 322-563 Gg CO₂ eq. At this stage of the life cycle of certified product, this is offset by carbon sequestered in the certified products of 380-637 Gg CO₂ eq. The total cradle to EU carbon footprint including stored (biogenic) carbon is in the region of -58 to -75 Gg CO₂ eq. If all tropical wood imported into the EU had been certified in 2018, the carbon footprint would have been -359 Gg CO₂, an additional net carbon storage of 284-301 Gg CO₂ eq.

Substitution Calculations

Calculating substitution benefits requires generalizing variables including wood species and processing, product being substituted, comparative lifespans, and end of life of management.

Drawing on the more conservative substitution factor derived by Leskinen et al and the estimate of total carbon content of certified tropical timber calculated in the IDH Timber Report (2019), we’re able to calculate that potential substitution benefits of certified tropical timber in the EU range from 456 Gg CO₂ eq. to 764 Gg CO₂ eq. in 2018.[13]

Net Carbon Calculations

The table below summarizes the estimates of carbon impacts of EU demand for certified tropical timber including the potential impact on carbon pools in the tropical forest, emissions during harvesting, conversion, and transport to the EU, carbon sequestration in certified tropical timber at point of delivery to the EU, and potential carbon emission reductions due to substitution of certified tropical timber for alternative non-wood products. For the numerous reasons cited, the level of confidence surrounding all these estimates must be considered very low.

Table: Summary of estimated carbon impacts linked to EU demand for certified tropical timber

The analysis gives some idea of the scale of emissions reductions that might be associated with certification. As we’ve made clear throughout, in order to come up with a rough estimate we were required to draw from a number of different studies and stretch assumptions to their limits. As a result, the actual numbers cannot be relied on for accuracy at this stage, but do provide an increased understanding of potential impacts associated with certification, and a roadmap to increasing the accuracy of future calculations. In order to drive demand for certified timber in the EU and globally, this must be an iterative exercise.

The numbers suggest that certified timber may contribute net carbon benefits to the EU in the realm of -456 to -764 Gg CO₂ eq. year^-1. Growing imports of certified timber would scale these benefits. This report also make one fact imminently clear—that the potential impact of certified forest operations on forest carbon pools will likely dwarf the carbon emissions and sequestration associated with certified forest products in operations, processing, and trade. Maintaining the long-term health of forest areas may be the single most important factor in certified forest operations’ carbon benefits. In this study we do not consider the potential longer-term consequences of CL as a precursor to deforestation involving even higher levels of emissions. As Sasaki et al note, “CL is likely to destroy all merchantable trees, and therefore put tropical production forest at risk of being cleared for industrial agricultures as logging option is no longer feasible economically.”

Subsequent attempts to allocate credits for avoided deforestation and forest degradation must build on this study and be based on more detailed data at national and sub-national level. New studies will require improving data on: specific types of forest that are certified, impact of different rotations and harvesting intensities on both production volume and carbon emissions, emissions from both above-ground biomass and soils associated with forest degradation and deforestation, risk of degradation and/or deforestation in the absence of certification, percentage of land converted to different uses and associated emissions, and the direction of trade of the products from certified forests are all paramount to improving carbon calculations.

These data demands may appear daunting, but there should be opportunities to link improving data on forest certification with comprehensive and standardized national greenhouse gas inventories and REDD monitoring systems now being implemented under the terms of the UNFCCC. The certification frameworks themselves could also play a significant role in improving data quality in this area if they establish specific requirements for low carbon impact forestry and explicitly require regular monitoring of changes in the carbon stock of certified forest areas. Developing a single harmonized LCA methodology to assess the carbon footprint of the full range of certified wood products imported from the tropics could play a key role in growing markets for certified sustainable lumber. Any analysis should seek to assess the carbon footprint of a representative sample of certified suppliers, with respect to regions of origin and product coverage, and include a detailed commentary on the methodologies by which a specific forest carbon debt or credit may be allocated to certified forest products on a unit volume or tonnage basis. More accurate systems of carbon accounting will drive demand and financing for certified timber, and help preserve the long-term environmental and social health of forest regions.

FSC’s international Principles and Criteria and generic indicators reference the need to protect and maintain “environmental values,” which include “ecosystem functions (including carbon sequestration and storage).” However, in contrast to most other environmental values such as conservation of biodiversity, rare and threatened species, habitats, and water resources, there is no direct reference in the FSC international standards to protection and enhancement of carbon stocks. FSC consulted with members on the possible introduction of a voluntary FSC Carbon Footprint Procedure in 2017, but based on the outcome of the consultation the process was put on hold until further notice.

The PEFC international standard is more explicit on carbon impacts: the first Criterion of the PEFC “Benchmark standard” is “Maintenance or appropriate enhancement of forest resources and their contribution to the global carbon cycle.” This criterion is reinforced by several indicators including: a requirement that the “quantity and quality of the forest resources and the capacity of the forest to store and sequester carbon shall be safeguarded in the medium and long term by balancing harvesting and growth rates, using appropriate silvicultural measures and preferring techniques that minimize adverse impacts on forest resources;” and a requirement that “climate positive practices in management operations, such as greenhouse gas emission reductions and efficient use of resources shall be encouraged.” The PEFC criteria don’t require that carbon impacts be monitored as part of the certification process, although they do leave the door open to national certification systems within the framework introducing such a requirement. PEFC’s reach is limited most tropical countries—it’s only been implemented on a large scale in Malaysia through the MTCS (Malaysia Timber Certification Scheme). The MTCS has specific indicators to “implement guidelines for reduced/low impact logging to minimize damage to residual stand” and require that “log extraction operations minimize product wastage, degradation, and foregone revenue opportunities,” with references to national guidelines for reduced impact logging (RIL) as a method of verification. There is no specific requirement to directly monitor these emissions.

The data on certified forest area that is available from FSC and PEFC cannot be easily disaggregated according to forest type and production capacity. Certified forest area data combined with other data sources can provide a reasonably robust estimate of annual timber production volume. However translating timber volume to carbon emissions estimates is difficult due to the huge variation in carbon content of different tropical forest eco-systems—the range of above-ground carbon per area of tropical ecosystems varies from a few tonnes per hectare to over 400 Mg C ha⁻¹.[14]

The AHEC-PE study on US hardwoods is sufficiently comprehensive to allow a detailed sensitivity analysis of various product parameters including wood density, drying time and lumber thickness. These are parameters that vary widely for hardwoods sourced from managed semi-natural forests and which (as shown by the AHEC-PE study) have a very large effect on carbon footprint figures.

The table below provides the results of a comparative cradle to gate of kiln dried sawn hardwood of three US hardwood species which differ primarily in the length of time required in the kiln to reach 7% moisture content, the standard specification for US hardwood. Kilning time is also significantly affected by lumber dimension – with thicker lumber requiring considerably longer than thinner lumber of the same species. Alterations in just these two parameters result in the carbon emissions associated with American hardwood sawn lumber to vary from as little as 473 kg CO2 eq. per tonne to 1967 kg CO2 eq. per tonne. If biogenic carbon storage (carbon stored in the wood product) is taken into account, these products vary from providing a significant carbon credit of over 1100 kg CO2 eq. for every tonne of product, to being a net emitter of 380 kg CO2 eq. for every tonne of product.

This level of variation, which applies to a relatively limited range of timber species all derived from similar forest environments, is likely to be at least as high for tropical hardwoods. In the absence of even the most rudimentary data on the volume, species range, moisture content, and size specification of the certified tropical hardwoods imported into the EU, it would be inappropriate at this stage to make any claim regarding the carbon footprint of certified tropical timber products at point of delivery in the EU.

Table: Comparative cradle to gate carbon footprint of kiln dried sawn hardwood of a fast drying, medium drying and slow drying US hardwood by lumber thickness & process (kg CO2 eq. per tonne of product). Source: PE-AHEC

References

[1] Grace et al, 2014, Glob Chang Biol. 2014 Oct; 20(10): 3238–3255. Perturbations in the carbon budget of the tropics   https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4261894/#b202

[2] Martin, P.A., et al. Impacts of tropical selective logging on carbon storage and tree species richness: A meta-analysis, Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.010

[3] Sasaki N, Asner GP, Pan Y, Knorr W,Durst PB, Ma HO, Abe I, Lowe AJ,Koh LP and Putz FE, 2016, Sustainable Management of Tropical Forests Can Reduce Carbon Emissions and Stabilize Timber Production, Production. Front. Environ. Sci. 4:50., https://doi.org/10.3389/fenvs.2016.00050

[4] Ellis et al (2019) Reduced-impact logging for climate change mitigation (RIL-C) can halve selective logging emissions from tropical forests. Forest Ecology and Management. Volume 438, 15 April 2019, Pages 255-266. https://doi.org/10.1016/j.foreco.2019.02.004

[5] Martin, P.A., et al. Impacts of tropical selective logging on carbon storage and tree species richness: A meta-analysis, Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.010

[6] Ellis et al (2019) Reduced-impact logging for climate change mitigation (RIL-C) can halve selective logging emissions from tropical forests. Forest Ecology and Management. Volume 438, 15 April 2019, Pages 255-266. https://doi.org/10.1016/j.foreco.2019.02.004

[7] Certified and Uncertified Logging Concessions Compared in Gabon: Changes in Stand Structure, Tree Species, and Biomass, V. P. Medjibe et al, Environmental Management (2013) 51:524–540

[8] Griscom, B., P. Ellis & F.E. Putz. 2014. Carbon emissions performance of commercial logging in East Kalimantan, Indonesia. Global Change Biology, 20:3. doi: 10.1111/gcb.12386

[9] GWP stands for Global Warming Potential

[10] Including: Leskinen et al, 2018, Substitution effects of wood-based products in climate change mitigation. From Science to Policy 7. European Forest Institute; Sathre & Gustavsson, 2009, State of the art review of energy and climate effects of wood product substitution,

[11] Roger Sathre and Leif Gustavsson, Växjö University, Sweden; Sathre and O’Connor, 2010, Synthesis of research on wood products and greenhouse gas impacts, October 2010, FP Innovations Technical Report No. TR-19R

[12] IDH, 2019, Unlocking sustainable tropical timber market growth through data, assess at https://www.idhsustainabletrade.com/uploaded/2019/11/IDH-Unlocking-sust-tropical-timber-market-growth-through-data.pdf

[13] Both Leskinen et al (2018) and Sathre & Gustavsson (2009) state that the basis for the substitution or displacement factor is the carbon stored in the wood product rather than the carbon footprint of the wood product (which also takes account of emissions involved to supply of the wood product). This estimate follows this approach and is calculated by multiplying carbon stored in certified tropical timber products by 1.2, the factor derived by Leskinen et al.

[14] Ziegler AD, Phelps J, Yuen JQ, et al. Carbon outcomes of major land-cover transitions in SE Asia: great uncertainties and REDD plus policy implications. Global Change Biology. 2012;18:3087–3099. [PubMed] [Google Scholar]