Net Ecosystem Production and Actionable Negative Emissions Strategies by John DeCicco and Joonghyeok Heo, University of Michigan Energy Institute presented at the American Geophysical Union (AGU) Fall Meeting Session GC051 on "Negative Emissions: Staying Below 2°C" Tuesday, December 13, 2016

Thank you. It's great to have the opportunity to speak at this morning's session and I'll look forward to meeting many of you during the session and afterwards. The title of my talk is "Net Ecosystem Production and Actionable Negative Emissions Strategies." There are several points I want to make.

The first is to emphasize that carbon dioxide removal (CDR) is a matter of increasing a flow rate, namely the downward flow of carbon from the atmospheric stock to some form of fixed carbon in another stock. The rate-based nature of the issue is crucial when implementing any CDR mechanism. I'll next define what I mean by an actionable strategy. That will bring us to focus on the terrestrial biosphere, which is where the most actionable CDR mechanisms are now found, and I'll explain why a gain in net ecosystem production is a necessary condition for such mechanisms to be effective for mitigation. Biofuels, which have been promoted as a CO2 reduction strategy, are an example of a mechanism that engages the biosphere. However, the biofuel situation also reveals how badly things turn out when we neglect the rate-based nature of CO2 removal. The lesson is that, when moving from "Think Global" modeling to concrete action on negative emissions, we need to "Act Locally." To have confidence that the actions taken are helpful rather than harmful, it is essential to use scientific management that in turn requires investing in better tools for measuring real-world carbon flows in terrestrial systems.

A starting point for thinking about the issue is the stock-and-flow nature of the global carbon cycle. On the order of 200 PgC per year circulates between the atmosphere and the major carbon stocks of the terrestrial biosphere, oceans and geosphere. Mitigating the carbon part of the climate problem entails both decreasing the upward flow, that is, reducing CO2 emissions, and increasing the down flow, which is what negative emissions strategies are all about. Most recent discussions of carbon dioxide removal (CDR) emphasize its role in meeting long-term goals such as a 2°C limit. However, CDR mechanisms for increasing the terrestrial sink, through reforestation and other ways to recarbonize the biosphere, are very much a here-and-now option for which an experience base already exists.

In other words, they are actionable. By that I mean they are mechanisms that can be implemented today without technological breakthroughs. They should also be ecologically safe and economically affordable when implemented with known management strategies. Those are pre-conditions for a mechanism to be scalable, i.e., able to be expanded enough to contribute to mitigation at a level significant for national greenhouse gas inventories. An important thing to keep in mind is that meaningful action is ALWAYS possible; technology availability is never a limiting factor for environmental protection. One has to start somewhere; it is not possible to implement something at global scale overnight. In other words, actionable mechanisms epitomize the meaning of "think globally, act locally."

The terrestrial carbon cycle provides the CDR options that are now most actionable. To assess such options correctly, we must start with the basics of the terrestrial carbon cycle. The key metric on which we need to focus is net ecosystem production (NEP), a term that is probably well known to everyone here. NEP is the difference between net primary production (NPP), which is the net amount of carbon fixed by plants and made available to other organisms such as ourselves, and heterotrophic respiration, which is the CO2 released during consumption by organisms other than plants. NEP represents the amount of carbon that becomes available for sequestration or some form of utilization. 2

To explain this in more detail, here is a box model of a coupled bio- and fossil-fuel system that highlights the role of NEP. This excludes the oceans, just focusing on terrestrial systems, but the basic principles are the same. P stands for NPP and R stands for heterotrophic respiration. Carbon-based energy can be drawn from either the biosphere, shown as B, or fossil resources, shown as F. However, the amount of CO2 emitted, or E, does not vary with the source of the carbon used for energy. Emissions can be reduced through carbon capture and storage, shown as C. And there is the separate option of direct air capture, D. On the left side of the box, L represents carbon stock releases from land-use change, which is of course something to be avoided. A mass flow balance implies that, for any form of bio-based mitigation, there must be an increase in the difference between P and R, which is by definition net ecosystem production, NEP. This can be expressed as the condition that d(NEP)/dt must be greater than zero.

In short, one must fix carbon more rapidly than it is already being fixed by existing processes in the biosphere. In the case of bioenergy, feedstock production must effect an increase in P or decrease of R, or both. This is still true if feedstocks are produced on so-called "marginal" land. We can't just start removing large amounts of carbon from the biosphere; as an ecologist would put it, it's very risky to "starve the decomposers." Now, a gain in NEP is the necessary condition for bio-based mitigation, but it's not a sufficient condition. For that, we must evaluate process and indirect emissions, including leakages such as those due to land-use change. Even though it is often assumed, merely substituting biogenic carbon for fossil carbon is not a sufficient condition for reducing the net CO2 flow to the atmosphere. In fact, even so-called sustainable biomass production is not a sufficient condition.

3

To provide a concrete example of what this all means, we examined the expansion of biofuel production driven by the Renewable Fuel Standard (RFS). This policy has increased U.S. biofuel use from 4 billion gallons per year in 2005 to 15 billion gallons as of last year. In addition to backing by farm lobbies, policies to promote biofuels are promoted for reasons of CO2 reduction. As EPA Administrator Gina McCarthy put it, "A growing biofuel industry helps us act on climate and cut carbon pollution." Such beliefs rest on lifecycle modeling. Unfortunately, however, those models not explicitly evaluate NEP. But it can be directly evaluated using crop data, as we've done in our recent study published online by Climatic Change in August.

Here's what we found, examining the 8-year period from 2005 to 2013 following the establishment of the RFS. The upper arrows are carbon exchanges with the atmosphere, including CO2 uptake during crop growth and CO2 emissions from the biomass itself during processing and biofuel combustion. The lower arrows represent flows of fixed carbon, either as biomass exported for use as food and feed or fossil fuel input. Several things happened over this period. Because of the intervening economic slowdown and gains in vehicle efficiency, tailpipe emissions dropped significantly. Less carbon was exported as food and feed, and there was a significant decrease in fossil fuel input. However, more corn and soybeans were grown. Because corn is a highly productive crop, the rate of CO2 uptake on cropland increased by 20 teragrams, that is a million metric tons per year on a carbon mass basis, over the period.

4

What we need to do is to parse out how this substitution of biomass carbon for fossil carbon compares to the gain in carbon uptake on land. These results are shown here. Biofuels progressively displaced fossil fuels over the period, resulting in a cumulative release of 132 teragrams of biogenic carbon. The gains in carbon uptake are shown as the green curve. They fluctuate with the growing season; for example, years of low crop yield in 2006 and the drought year 2012 are shown as dips in carbon uptake curve. The difference between total biogenic emissions and uptake represents net emissions, that is, the portion of biomass carbon burned that was not balanced out by additional carbon uptake. It turns out that the increase in uptake was enough to offset only 37% of the biomass CO2 emissions over this period. In other words, most of the biomass carbon was not "neutralized." These datadriven results significantly constrain the net CO2 impact of the biofuel use. Once processing and indirect effects are considered, the implication is that biofuel use has significantly increased net CO2 emissions compared to fossil fuel use to date.

So, what went wrong? The lifecycle modeling that is so widely used treats all biomass flows as if they are already in equilibrium with the atmosphere. It effectively treats biomass as "free" carbon when substituted for fossil carbon, neglecting to measure NEP. To get things right, we can't sidestep to need to explicitly measure and manage any carbon that we appropriate from the biosphere in the hope of using it for mitigation, whether for so-called carbon neutral bioenergy or for negative emissions. As a recent EOS article put it, we need to invest in a "robust infrastructure for science and observation of the carbon cycle at multiple scales." This need is understood in the forest carbon community, but unfortunately not in the energy policy community. Keep in mind that once carbon has been removed from the atmosphere, we are as ahead of the game as we will ever be in terms of climate protection. And so for the foreseeable future, the priority should be enhancing the terrestrial sink by increasing rates of net carbon uptake in ways that are ecologically sound and verifiable.

5

Our work in progress includes developing a spatially explicit estimate of NEP at the U.S. continental scale. This effort builds on analyses we've done using the Cropland Data Layer, and involves working to consistently combine the data sets needed to cover diverse land-use categories. The goal is to develop baseline estimates of NEP, as needed for large-scale terrestrial carbon management. That is in turn essential for guiding any negative emissions strategies that utilize the biosphere.

To conclude, carbon dioxide removal is a hereand-now opportunity, as urgent to pursue today as any other mitigation measure. The terrestrial biosphere offers the CDR options opportunities that are now actionable, but science-based management is crucial. Moreover, the principle that CDR requires increasing the net rate of carbon uptake applies to any form of bioenergy, including BECCS. NEP is a key metric in this regard. There's no such thing as "free" carbon. Careful, measurementbased scientific management is essential. The research and policy priorities should be making the best use of the tools we have while developing better tools that will enable us to fix more carbon in ways that are ecologically sound and can be built to progressively larger scales. Thank you.

6