Last week I wrote about the "social value of an offset" (SVO). SVO provides a mathematical formula for determining the value of temporary or carbon removal projects relative to permanent ones. This let us quantify the intuition that temporary carbon storage has value, even if it is less value than permanent storage.
This week I want to look at a very specific scenario: How much does it really cost to store carbon in trees, given both their temporary nature, and their "non-additionality" (the odds that someone might have planted them anyway, regardless of credits)?
I will approach this through the lens of this recent paper, Prospects for Land Sector Carbon Dioxide Removal in the United States, by Resources for the Future (RFF). In forests, the rate of carbon uptake is age-dependent, slowing down as trees mature. RFF reports that in the US, market forces led to the replanting of about 21 million acres of forest on private land between 1987 and 2017, giving the country large numbers of relatively young, growing trees. Regrowth thanks to this reforestation removed about 900 MT CO2/year in 1990. As these forests age, this rate drops, and RFF estimates it has fallen to about 800 MT CO2/year today.
The RFF researchers then model a 'business as usual' scenario to predict future planting and management trends, and determine that the rate of carbon withdrawal into trees will drop to 600 MT CO2/year by 2060. This is bad for those of us who want to offset some US emissions with tree growth. To increase the rate of CDR provided by trees, we have to first make up for this baseline slowdown that comes naturally as forests age.
The first of RFF's proposed interventions is a 30 million acre afforestation plan (3 million acres/yr over 10 years), with a scale and structure of the 1985 USA "Conservation Reserve Program" which paid farmers rental fees to remove marginal agricultural land in ecologically sensitive areas. RFF's second proposed intervention continues the 3 million acre/yr removal over 30 years, for a total of 90 million acres removed. Rents for this land are assumed to be paid over a 40 year period, at which point the modeling horizon ends (and presumably the federal program does too).
The researchers find that over the simulation period, the 30 million acre, 10-year plan sequesters an additional 3.6 GT CO2 over its life, while the 90 million acre, 30-year plan sequesters 8.5 GT CO2. The authors are skeptical that the program can scale much beyond 90 million acres - appropriate land simply does not exist to surpass this. This second case thus represents the best performance possible, where the US can more or less maintain its existing CDR levels from afforestation to about 2050, at which point the maturity of trees will result in ever-slowing future uptakes, as shown in the graph below.
The cost of RFF's programs are estimated at between $20-40/ton CO2 removed (roughly in line with prior estimates). This cost, if taken at face value, is much cheaper than alternative forms of carbon dioxide removal. But RFF's analysis doesn't consider the permanence of the solution, or make explicit exclusions for the non-additionality of the baseline. What's the value of these trees when compared against the notion that farmers might have made these decisions anyway (for example, to protect their watersheds)? What's the value of the trees if they are cut down as soon as the subsidies end?
The Conservation Reserve Program as fortunately been well-studied, we already have pretty good answers to these questions. Historically, as the rental payments programs have expired, about 50% of land immediately returns to agriculture, re-emitting the carbon they previously stored. This is not a certainty - our future society could always opt to keep paying those rents to keep the forests intact, even if their carbon reduction days are over. But an infinite horizon of payments adds quite a bit of cost! The simplest approach that aligns with the RFF paper is to consider the trees to be a temporary store of carbon, to be used until more durable stores are available. The SVO methodology provides a way to calculate this, and would value such a temporary store at about 1/3 the value of a permanent store, which in turn triples the equivalent social cost of tree planting to between $60-120/ton.
The additionally associated with the Conservation Reserve Program has also been reviewed. Private landowners who didn't participate in the program (the "control" group) set aside about 20% as much land as those who did participate. One can argue that the program had an 80% additionally - every $1 spent provides about 80¢ of public benefit above what would have happened naturally. This brings that carbon dioxide reduction cost of the tree planting program to $75-150/ton.
This is not bad! Even the high estimate of $150/ton is still less than the $190/ton social cost of carbon, such that every dollar spent on such a program generates at least an additional 25¢ in social value. The investment worth considering, especially as part of a political package that brings rural stakeholders into the climate coalition.
But tree planting does not look is substantially cheaper than the long term goals for other forms of CDR. As a tool that can be deployed now, tree planting has value even though it is temporary. In the US in particular, its impact will be limited, both by cost and ability to scale. There is plenty of room for tree planting to make an impact in tropical nations, where the density of trees will be higher, and the cost of land lower. In the developed world, tree planting is helpful, but probably not as helpful as other forms of emission abatement, and in the long term potentially even more expensive than industrial forms of capture.
Comments