An Agrigenomics Trifecta: Greenhouse Gas Drawdown, Food Security, and New Drugs

Abstract

An abundance of data, including decades of greenhouse gas (GHG) emission rates, atmospheric concentrations, and global average temperatures, is sufficient to allow a strictly empirical evaluation of the U.S. plan for controlling GHGs. This article presents an analysis, based solely on such data, that shows that the difference between atmospheric GHG levels that will be reached if current trends continue, and levels that would be achieved if the goals of the plan are met—even with worldwide implementation—is inconsequential. Further, the expected globally averaged temperature differences are well within measurement error. The results lend additional support to the argument that any mitigation strategy must include drawdown of atmospheric GHGs. Equally important, a particular drawdown strategy, agrigenomics, offers the opportunity for a revolutionary trifecta: climate change mitigation, food security, and medical advances.

Global warming is incontrovertible, and confidence is very high that the extreme events witnessed in recent decades are mainly anthropogenic in origin (IPCC 2021). The confident linkage of a changing climate to human activity, rather than to natural and unknown causes, is crucially important since it means that we can in principle reduce warming by limiting human-generated greenhouse gases (GHGs). Unfortunately, although the world has witnessed three decades of multinational meetings and agreements, the average global temperature continues to increase, as do its many and increasingly disruptive correlates.

One might take heart that in 2022, the United States Senate passed the Inflation Reduction Act (Science for Impacts, Resilience and Adaptation [SIRA]), which included some $391 billion for climate change mitigation and adaptation. It has been described as the most ambitious climate action ever taken by Congress and, sadly, it no doubt is.

EMISSION REDUCTION ALONE WILL NOT CONTROL GLOBAL WARMING

Current Policy

Recent estimates indicate that SIRA might reduce GHG emissions to a level that is 32%–40% below 2005 levels by the end of the decade (Jenkins et al. 2022). Before commenting on the consequences of such a reduction, it is important to note that the warming potential of GHGs emitted by the United States in 2005 was equivalent to that of ∼6.4 billion metric tonnes (Gt) of carbon dioxide (CO₂e), whereas by the end of 2019 emissions had declined to ∼5.8 Gt (Ritchie and Roser 2020).

A 40% goal, to be optimistic, would require reducing emissions by 0.36 Gt/yr over a 7-yr period. Since the United States currently accounts for only ∼12% of global emissions of CO₂e, much of the potential impact of the legislation, even if it proved effective, would be lost. I will, therefore, assume, with unwarranted optimism, that the rest of the world adopts an SIRA-like policy, and comment on the impact under that condition. The target for the average annual global emission reduction in CO₂e would then be ∼3 Gt/yr = 0.43 ppm/yr.¹

The SIRA-based plan immediately suggests three questions: What will the atmospheric concentration be if it is globally successful; How much of an impact will the reduction in CO₂e have, relative to business as usual (BAU), on the average global temperature; and What will the cost be per ppm of CO₂e reduction? The first question has been addressed, at least in terms of the U.S. contribution, by detailed and careful analysis of different decarbonization scenarios (REPEAT Project, repeatproject.org/policies/net-zero). The philosophy adopted here is not to forecast the reduction, but to ask for the implications of a prespecified reduction, as embodied in these questions. I believe informative approximate answers can be obtained based on reasonable assumptions and on purely empirical data (i.e., no theory).

The results presented here suggest that the impact of SIRA on climate change mitigation is essentially inconsequential and add urgency to introducing the kinds of action needed to mitigate climate change. Specifically, the final section of this article includes a discussion of agrigenomics as an additional and important component of a global policy that could, if well organized and funded at scale, substantially mitigate a problem that all available data indicate continues to grow worse at an alarming rate. Moreover, agrigenomics offers the possibility of a multiwin strategy, including major benefits to multiple economic sectors.

Analysis

There are several auxiliary quantities that are required to obtain answers: the annual BAU rate of increase in global CO₂e emission, defined as ξ₁; the annual rate of reduction, ξ₂, required by a specified reduction in CO₂e emission; an empirical relation based on historical data between atmospheric CO₂e and average global temperature, and the fraction of CO₂e emission that remains in the atmosphere, also based on historical data.

Useful estimates of ξ₁ and f, the fraction of CO₂e that remains in the atmosphere, can be made if changes in concentration vary linearly with time. As shown in Figure 1, the rate at which atmospheric CO₂e is increasing has satisfied that condition for at least the last four decades. The linear increase implicitly averages over a rate of accumulation minus a rate of loss for the various GHGs. Although the averaging time is small compared to the atmospheric residence time of most GHGs, especially CO₂, the GHG increase is more or less in a steady state for the relevant period.

Figure 1.

Carbon dioxide equivalent (CO₂e) atmospheric concentration from 1980 to 2019 has increased linearly at a rate of 2.84 ppm/yr. (From the National Oceanic and Atmospheric Administration [NOAA] annual greenhouse gas index, gml.noaa.gov/aggi/aggi.html.)

In contrast to the implicitly averaged atmospheric concentration increase, annual emission prior to 2007 changes erratically and nonlinearly (Fig. 2; Ritchie and Roser 2020). Consequently, emission data prior to 2007 are not useful. For the 13-yr period from 2007 to 2019 inclusive, the slope of CO₂e emission versus year is reasonably linear, and a least-squares fit gives ξ₁ = 0.085 ppm/yr. As indicated above, ξ₂ = −0.43 ppm/yr.

Figure 2.

(A) Carbon dioxide equivalent (CO₂e) emitted annually, 2007–2019 inclusive. The slope ξ₁ = 0.085 ppm/yr. The fraction f, of CO₂e emitted that remains in the atmosphere, was estimated by dividing the total emitted CO₂e during the period shown by the increase in atmospheric concentration (Fig. 1) during the same period. (B) CO₂e emitted annually, 1990–2019 inclusive (Ritchie and Roser 2020).

The fraction, f, of emitted CO₂e that remains in the atmosphere was estimated by dividing the change in atmospheric concentration during the 2007–2019 period (40 ppm) by the sum of the annual emissions during that period (93 ppm) to obtain f = 0.43.

The Atmospheric Concentration Difference between SIRA and BAU Is Roughly 1%

Let c_n denote the carbon dioxide equivalent concentration n years after year 0 (see Box 1),

$${\mathrm{c}}_n={\displaystyle {\mathrm{c}}_0+nf\mathrm{\Delta }+\frac{nf\left(n+1\right){\xi }_j}{2};j=1,2,}$$ (1)

and δ, the difference between concentrations for SIRA and BAU,

BOX 1. DERIVATION OF EQUATION 1.

The simplest way to obtain Equation 1 is to write out expressions for the concentration in the first few years.

The atmospheric concentration at the end of 2020 for BAU (j = 1) and SIRA (j = 2) is

$${\mathrm{c}}_1={\mathrm{c}}_0+f\left(\mathrm{\Delta }+{\xi }_j\right)j=1,2,$$

where f(Δ + ξ_j) is the concentration increase in year 2020.

The amount emitted in 2021 is the amount emitted in 2020 + the change in emission in 2021,

$$\left(\mathrm{\Delta }+{\xi }_j\right)+{\xi }_j=\mathrm{\Delta }+2{\xi }_j.$$

Therefore,

$${\mathrm{c}}_2={\mathrm{c}}_1+f\left(\mathrm{\Delta }+2{\xi }_j\right)={\mathrm{c}}_0+f(\mathrm{\Delta }+{\xi }_j)+f(\mathrm{\Delta }+2{\xi }_j)=c_0+2f\mathrm{\Delta }+3f{\xi }_j.$$

The emission in 2022 is the emission in 2021 + the change in emission in 2021,

$$\mathrm{\Delta }+2{\xi }_j+{\xi }_j=\mathrm{\Delta }+3{\xi }_j,$$

$${\mathrm{c}}_3={\mathrm{c}}_2+f\left(\mathrm{\Delta }+3{\xi }_j\right)={\mathrm{c}}_0+2f\mathrm{\Delta }+3f{\xi }_j+f(\mathrm{\Delta }+3{\xi }_j)={\mathrm{c}}_0+f(3\mathrm{\Delta }+6{\xi }_j).$$

The relation between c_n and c₀ can now be seen to have the form,

$${\mathrm{c}}_n={\mathrm{c}}_0+nf\mathrm{\Delta }+f{r}_n{\xi }_j,$$

where r_n = 1, 3, 6, 10,… for n = 1, 2, 3, 4,… r_n will be recognized as the sum of the first n integers:

$$r_n=\sum _{k=1}^{n}k.$$

Therefore, r_n = n(n + 1)/2,

$${\mathrm{c}}_n={\displaystyle {\mathrm{c}}_0+nf\mathrm{\Delta }+\frac{fn\left(n+1\right){\xi }_j}{2},}$$

and the atmospheric concentration difference, δ, between BAO and SIRA after n years follows immediately:

$$\delta ={\displaystyle \frac{fn\left(n+1\right)\left({\xi }_1-{\xi }_2\right)}{2}.}$$

$$\delta ={\displaystyle \frac{fn\left(n+1\right)\left({\xi }_1-{\xi }_2\right)}{2},}$$ (2)

where Δ is the emission rate, and c₀ atmospheric concentration, in year 0.

From Equation 2, we find that SIRA, if globally applied, will decrease the atmospheric concentration relative to BAU by 6.2 ppm. To place this in perspective, it is ∼1.2% of the 2019 atmospheric CO₂e (500 ppm).

The importance of CO₂ to the total GHG concentration cannot be overstated. CO₂ not only consists of some 72% of annual GHG emissions, it does not react chemically, has no natural sinks, and therefore has a long and complicated atmospheric residence (Fig. 3; Ritchie and Roser 2020). Under SIRA, CO₂ would reach ∼431 ppm, or ∼3.4 trillion tonnes (Tt)—that is, an excess of about (431 − 280) × 7.8 = 1.2 TtCO₂ above preindustrial levels. Even if emission stopped immediately, the excess atmospheric CO₂ in 500 years would be in the neighborhood of 330 Gt.

Figure 3.

Approximately 28% of a pulse of CO₂ is still in the atmosphere after 500 years. Removal follows a multiexponential decay with three components, whose driving processes are not well identified. (Amplitudes and time constants data are from Shine et al. 2005.)

Temperature Differences Are Inconsequential

A linear least-squares fit of temperature to concentration (Fig. 4) has a slope of 0.0068C·CO₂e⁻¹ (read 0.0068°C/ppm of CO₂e). This gives a temperature deviation between BAU and SIRA of ∼0.04°C, which is within measurement uncertainty. There should be no need to emphasize, but I will nevertheless remark, that if the United States acts alone, ξ₂ would have to be multiplied by 0.12 used for estimating the share of U.S. CO₂e, and the temperature deviation from BAU would be essentially meaningless.

Figure 4.

Global average temperature has been increasing approximately linearly with carbon dioxide equivalent (CO₂e) concentration for more than four decades. (Temperature data are from NASA/GISS, climate.nasa.gov/vital-signs/global-temperature.)

Costs

An estimate of cost does not involve a complex economic analysis: consistency requires that it be obtained by multiplying the SIRA budget using the CO₂e scaling factor. In particular, $391 billion/0.12 gives $3.26 × 10¹² or (dividing by 6.2 ppm) $526 billion/ppm of CO₂e emission reduction relative to BOA. That translates to ∼$76/tonne of CO₂e, which is within the range of estimates of carbon abatement costs. It is important to be clear about what this money is buying—it is providing jobs in an important economic sector, and it is appropriately part of the Economic Recovery Act; it is not, however, doing anything to mitigate global warming.

We can hope that other wealthy nations will do their part in reducing GHGs, but hope is not policy, and for a plan such as SIRA, the impact would be far too small even if the world cooperated. Moreover, although SIRA sounds bold, it is not—and if, as some say, it is a step in the right direction, it is a political step, not a technological step. In fact, to the extent that it gives false hope and thereby reduces pressure for consequential action, it may be worse than doing nothing.

The reasons for the disappointing lack of consequence are not difficult to identify. One is that, unlike the majority of nations for which GHG emissions have increased since 2003, the U.S. GHG emissions have declined. Consequently, using 2003 as a point of reference can be misleading; it would seem more straightforward to speak in terms of a reduction below current levels. In addition, exuberance over an apparitional reduction in emission rate can blur what should be a clear vision of a planetary reality, that is, the continued rapid emission and accumulation of CO₂e.

THE PATH FORWARD MUST INCLUDE ATMOSPHERIC GHG REMOVAL

To be clear, I am not saying that current strategies using alternative energy sources, green industries, economic stimuli, and emission constraints should be abandoned. On the contrary, as argued previously (DeLisi 2019; DeLisi et al. 2020), they must be included as part of an integrated strategy, but one that includes a strong GHG drawdown component. In particular, the results indicate that under a reasonable set of circumstances, neither strategy alone effectively reduces atmospheric GHGs, but together the impact is substantial. They also indicate an unusually strong effect of a decadal delay in the start of an effective strategy (details are in DeLisi 2019).

A Strategy that Includes Agrigenomics Offers the Possibility of a Revolutionary Trifecta

GHG drawdown is an engineering problem and numerous solutions have been proposed, some of which have been under consideration for decades (National Academies of Sciences, Engineering and Medicine 2019). Until recently, however, the approach that we believe has the greatest potential, agrigenomics, has received very little attention. To a first approximation, agrigenomics can be thought of as traditional land management on steroids. But as the contributions to this collection demonstrate, its advantages are more than just quantitative: it enables possibilities that were previously difficult even to imagine. Perhaps most importantly, it offers the possibility of a triple win, including new approaches to food security in response to a changing climate, and the possibility of discovering new generations of promising compounds for pharmaceutical development. And, of course, substantial economic benefits derive from all three.

The biological approach that has received the most attention is bioenergy with carbon capture and sequestration (BECCS). The idea is to plant and then remove trees after a period of rapid growth, followed by burning to provide energy, capturing and sequestering the released carbon, and then continuing the sequence with another round of seedlings. Trees of course require land, and, unfortunately, the land requirement for even a moderate reduction in emission, let alone a reduction in the concentration of atmospheric GHGs, is essentially unattainable. Other biomass energy/carbon removal methods are also being explored (DeLisi et al. 2020), but they all face serious scaling difficulties.

I mention BECCS even though it is not genomically based, because of its visibility and because agrigenomic advances can lower the threshold for its use by substantially improving land use. Efficient land use, of course, has numerous benefits beyond just boosting bioenergy and carbon removal efforts. For instance, agrigenomic advances can enable the cultivation of crops in difficult environments through genetic modification to increase drought and salinity tolerance (Caine et al. 2019). And of course, an increase in photosynthetic efficiency, such as by the creation of photorespiration pathways that are new to nature (Scheffen et al. 2021; Erb et al. 2023), could have an enormous impact. FuturaGene has already engineered eucalyptus trees with a 30% increase in yield (May et al. 2022). Here again, the idea would be to grow and replace high-yield trees, and to thereby increase CO₂ removal. Moreover, the possibility of moving beyond eucalyptus to tress whose wood is high on the Janka scale,² for example, at or above the hardness of Brazilian ebony, could provide a source of building material that would substantially reduce the carbon footprint of cement production.

Another promising area of research is the use of mitochondrial editing to modify the active site of rubisco in C3 plants, potentially increasing the photosynthetic efficiency and carbon-to-oxygen ratio. Finally, toxicity and other damage that accompanies the use of excess fertilizer can be reduced by increasing the efficiency with which microbes fix atmospheric nitrogen.

The third arm of the trifecta is medical. Until the relatively recent ascendency of structural chemistry, living organisms, and especially plants, were the main source of drugs. Currently, some 11% of the drugs classified as essential by the World Health Organization originated in plants (Veeresham 2012) including digitalis, ergotamine, quinine, and salicylates (Sen and Samanta 2015).

Indeed chemical methods have not so much displaced natural products as the source of drugs, as they have accelerated the opportunities for exploiting plants in the search for new medicines. Thus, sequencing enables rapid, inexpensive elucidation of structure and the structures of enzymes involved in their synthesis, design enables increased potency, and low-cost synthesis enables amplification of availability. An example is the vinca alkaloids that are widely used to treat various cancers (Sears and Boger 2015).

Considering the diversity and magnitude of the plant kingdom, it would be surprising if there were not a treasure trove of untapped opportunities that could be realized with accelerated efforts in agrigenomics. These include the potential for the discovery of enzymes crucial to metabolism, of new pathways for secondary metabolites, and transgenic plants with specified traits of medical benefit.

It should be evident from these remarks and from the other articles in this collection that scientific and technological opportunities for climate modulation abound. There are also, of course, multiple possible roadblocks. Most are too complex to be discussed here, but two in particular must be mentioned if this monograph is to have even the semblance of completeness. Regulatory and other related social issues were discussed in depth at our 2020 conference on Climate Change and Systems Biology and are summarized in DeLisi et al. (2020) and Giddings (2023).

With respect to organization, although there is plenty going on scientifically, a challenge as substantial as climate change is not likely to be met by scattered efforts here and there; it is best met by a well-thought-through cohesive plan. Highly complex challenges encountered in the past—the Manhattan Project, the moonshot, the Human Genome Project—were all met because science and technology were supported by superb large-scale organizational leadership. Climate change, with its mix of politics, economics, ethics, cultural diversity, and technology, is, in my opinion, considerably more complicated than any technological challenge faced by previous generations. It is past the time that some nation, or group of billionaires willing to pool resources, step forward with a plan to kick-start a historic effort. If that happens a parade will follow, if for no other reason than fear of being left behind economically.