Ms. Kirsten Corrigal,…

Ms. Kirsten Corrigal, Manager

Ministry of Natural Resources and Forestry

Policy Division, Crown Forests and Lands Policy Branch

Forest Legislation and Planning Section

70 Foster Drive, Suite 400

Sault Ste Marie, Ontario P6A 6V5

January 23, 2016

Re: Ontario’s Crown Forests: Opportunities to Enhance Carbon Storage? A Discussion Paper (EBR Registry Number 012-8685)

Dear Ms Corrigal:

Thank you for the opportunity to provide comments on Ontario’s “Opportunities to Enhance Carbon Storage” discussion paper. As I detail below, in my view this is a fundamentally irresponsible document (and I do not make such a statement lightly). The document states that current forest management results in a net removal of GHGs from the atmosphere, whereas as I detail below, the opposite is much more likely to be the case. Key facts and analyses are either based on guesses, or are incorrect. As a result, the document provides a misleading and damaging starting point for any discussion of climate friendly changes in sustainable forest management. The conclusions are fundamentally at odds with other studies that conclude that forest management results in net GHG emissions to the atmosphere. As a result of the incorrect analyses, the options provided ignore the most important options identified in the scientific literature. Ontario's forest management could easily be contributing to reducing GHG emissions, and helping with Ontario's overall goals, but not in the way envisioned in this document. The document has the potential to significantly delay the development of more GHG friendly policies in the Province's forestry sector.

OMNRF's forest carbon models are speculative and based on inaccurate forest ages

A key part of any discussion of the GHG implications of forest management in Ontario is the transition from a primary forest landscape to a secondary (managed) one. The great majority of boreal forest harvesting in Ontario is of primary forests, i.e., forests that have never been harvested. Primary landscapes in Ontario have larger areas of older forests than secondary ones; this means that they also have more timber (and more carbon). During the transition from a primary to a secondary landscape, companies inevitably find themselves shifting from a situation in which timber is abundant to one in which less timber is available (the so-called falldown effect). This same falldown effect applies to carbon. This means that any positive effects of forest management -- for example, carbon sequestration by young forests or long-term storage of carbon in wood products -- must be measured against the “debt” that is incurred during this transition. As an illustration of the importance of this idea, Fargione et al. (2008) consider this problem in detail for the specific case of converting primary ecosystems to agricultural biofuel production and calculated how long it takes to pay off the debt; that is, the time period involved from making climate change worse (through net emissions of carbon to atmosphere) to actually improving the situation (i.e., net removal of carbon from the atmosphere). They found that the magnitude of the debt imposed a relatively long period of net negative effects. For example, even for the conversion of native prairie (with its carbon rich soils) to ethanol producing corn, the period before payoff was some 93 years. For tropical rainforests, debt periods ranged from 86-423 years (even accounting for the carbon stored in wood products from the logging of the original forest). As a result, in assessing the effects of forest management in Ontario, one must be able to accurately measure carbon storage and fluxes in these original forests compared to the secondary ones. Unfortunately, as discussed below, the conclusions of the discussion paper (its Fig. 7) are based on the modelling that: 1) has not been parameterized for old forests and 2) relies on forest age information that is strikingly inaccurate. In a recent publication, two OMNRF scientists (Etheridge and Kayahara 2013) clearly identified the paramerization problem, noting that "science-based information to incorporate late successional forest stages into wood supply modeling is lacking in boreal Ontario" (p. 315). They noted that current forest management modelling is based on even-aged forests, and that data sets and analyses for older uneven-aged forests had not been used to validate model assumptions. Although expert judgement was used in developing the equations for these older forests, the paramerization is fundamentally speculative (and even bizarre). For example, in the model, once an aspen stand reaches 140 years of age, it is assumed to become a 60-year-old mixedwood stand; once that stand reaches 140 years, it is assumed to become a 60-year-old spruce-fir stand (Etheridge and Kayahara 2013). They conclude that the lack of empirical and science-based information for older age classes is a clear barrier to building robust model scenarios, yet such scenarios form the basis of the discussion document. Perhaps even worse, the forest ages that Ter-Mikaelian et al. (2013) rely upon are not actual stand ages (i.e., the time since the last major disturbance), but are averages ages of the current canopy trees in the stand (Etheridge and Kayahara 2013). Because boreal stands can go through multiple cohorts of such trees during their lifetime, such ages can be expected to sometimes severely underestimate actual stand ages. A good illustration of this problem is to compare the ages of primary forests between Ter-Mikaelian et al. (2013), who used current-tree ages, and Bergeron et al. (2001), who estimated actual stand ages from fire scars, ages of oldest trees, and fire records (from boreal forests in eastern Ontario and western Quebec). Here, I restrict the comparison to forests older than 100 years, which ensures that we are excluding secondary forests (commercial logging in the boreal forest started after c. 1920). It is not an exact comparison given that stand ages in the east of the province can be expected to be older on average than in the west (due to longer fire intervals), but the two study areas do overlap. The difference between the two studies is striking (Fig. 1). In Ter Mikaelian et al. (2013), stands older than 180 years are virtually nonexistent (1% of the forests >100 years old), whereas in Bergeron et al. (2001) they are common (52%). These two problems in combination -- speculative carbon parameterization and inaccurate stand ages -- mean that one can reasonably view the forest carbon estimates in Ter-Mikaelian et al. (2013) as little more than guesses. To see if they seem reasonable, we can look to another boreal study. Unfortunately, Ter Mikaelian et al. (2013) did not model the falldown effect, but instead compared "business as usual" management with areas that were not subjected to management, and hence presumably started to approximate the conditions of primary forest. After 50 years, managed forests had 92% as much carbon as the unmanaged forests (their Fig. 3). A study in central Finland took a very similar approach, but after 50 years managed forests had only 61-68% as much carbon as the unmanaged forest (Trivino et al. 2016; scenarios BAU, NTLR, NTSR vs SA). This comparison suggests that Ter Mikaelian et al. (2013) are seriously underestimating carbon accumulation as forests age. Their model apparently speculates that the volume (and carbon content) of aging stands is radically reduced in the decades after they achieve 100 years of age (see Fig. 2 in Etheridge and Kayahara 2013), which could be the source of the problem.

GHG emissions from forestry are likely to exceed the carbon stored in wood products

Unfortunately, it gets worse. Of course, in calculating the net benefits (or detriments) of the forestry sector from a climate change perspective, one must consider not only the carbon stored and sequestered in the forest, but also carbon storage in wood products and GHG emissions from wood harvesting and processing, etc. Following Ter Mikaelian et al. (2013), the discussion document assumes that wood storage plays a key role in making forestry a net sink with respect to GHGs. Indeed, and remarkably, the document states that "when carbon stored in harvested wood products is factored into carbon accounting, sustainably managed forests are always a carbon sink" (p. 8). Evidence suggests that this is incorrect. Ter Mikaelian et al. (2013) note that their calculations of the potential GHG benefits of wood products do not include emissions related to product life cycles and methane produced in landfills. However, this omission critically affects the net GHG impact of the industry. In their life cycle analysis of the U.S. forestry sector, Heath et al. (2010) calculated that methane emissions made up 27% of emissions for the forestry sector in the United States. Carbon storage in wood products replaced only 34-42% of total carbon emissions from the industry, and this was under the assumption that the forests themselves were in equilibrium with respect to their GHG emissions. As discussed above, the fall down effect in Ontario is presumably leading to a carbon debt. The study by Heath et al. (2010) suggests that, moreover, the sector is not actually paying off the debt at all; rather, it is making it worse. The statement in the document could reasonably be revised to say: "forest harvesting in Canada always results in a net release of GHG into the atmosphere"! Although Ter Mikaelian et al. (2013) acknowledged that their analyses did not account for emissions from several key sources, they downplayed the importance of a full life-cycle analyses by noting that life-cycle emissions tend to be more than compensated for by wood product displacement of more energy-intensive materials. Although they did not include such effects in their life-cycle analysis, Heath et al. (2010) examined this possibility and estimated that such savings accounted for only 13% of methane emissions.

Has OMNRF abandoned an evidence-based approach?

In summary, the fundamental analyses upon which the discussion paper is based have numerous serious problems, and the evidence suggests that its major assumption -- that status quo harvesting is a net positive from a climate change perspective -- is most likely in error. This means that the whole thrust of the document is likely erroneous because of its focus on minor changes to current practices. The document states that its purpose is to "start a dialogue with Ontarians on how climate change mitigation opportunities can be optimized within Ontario’s CURRENT approach to sustainable forest management" (my caps). Instead, it should be opening a dialog about how best to how alter and supplant current forest management so that our actions can actually mitigate global climate change, rather than make it worse. The literature is full of interesting ideas in this regard; for example, increased set asides, longer rotation periods, the replacement of clearcutting with partial harvesting methods, or a landscape-level mix of such approaches coupled with status quo harvesting. Remarkably, important benefits that could be derived from the unmanaged forest and remaining primary forests are not even considered in the discussion document. Trivino et al. (2016) found that maximization of timber revenues (the business as usual approach) was incompatible with optimizing GHG mitigation, but noted that minor reductions in profits could be strategically leveraged to attain significant advances in both carbon storage and biodiversity management. It is interesting to ask: how can such a large organization with such an excellent scientific staff produce such a flawed document? One possibility is that the organization has expended so much time and effort in designing a sustainable forest management system that it is reluctant, or perhaps incapable, of imagining a world in which the system does not form the basis of suitable management. Alternatively, it may relate to a culture where science “informs” policy, but where the ultimate product is created by non-scientists who bring science in as one (among many) considerations. This approach will not work in a climate change context because of the intense focus on verification. Either way, it appears that a major shake up and a significantly more innovative approach is required if OMNRF is actually going to help mitigate climate change. Our forests are at the forefront of climate change impacts, so positive actions from the organization are desperately needed.

Yours sincerely,

Jay R. Malcolm

Faculty of Forestry

University of Toronto

33 Willcocks St.

Toronto, ON M5S 3B3

416-978-0142

jay.malcolm@utoronto.ca

Literature Cited

Bergeron, Y., Gauthier, S., Kafka, V., Lefort, P., and Lesieur, D. 2001. Natural fire frequency for the eastern Canadian boreal forest: consequences for sustainable forestry. Canadian Journal of Forest Research 31:384 391.

Etheridge, D. A. and Kayahara, G. J. 2013. Challenges and implications of incorporating multi cohort management in northeastern Ontario, Canada: A case study. Forestry Chronicle 89:315-326.

Fargione, J., Hill, J., Tilman, D., Polasky, S., and Hawthorne, P. 2008. Land clearing and the biofuel carbon debt. Science 319:1235-1237.

Heath, L. S., Maltby, V., Miner, R., Skog, K. E., Smith, J. E., Unwin, J., and Upton, B. 2010. Greenhouse gas and carbon profile of the U.S. forest products industry value chain. Environmental Science and Technology 44:3999-4005.

Ter-Mikaelian, M. T., Colombo, S. J., and Chen, J. 2013. Effects of harvesting on spatial and temporal diversity of carbon stocks in a boreal forest landscape. Ecology and Evolution 3:3738 3750.

Trivino, M., Pohjanmies, T., Mazziotta, A., Juutinen, A., Podkopaev, D., Le Tortorec, E., and Mönkkönen, M. 2016. Optimizing management to enhance multifunctionality in a boreal forest landscape. Journal of Applied Ecology (Online):1-9.

[Original Comment ID: 207883]