New feed sources key to ambitious climate targets

Recent assessments from the Intergovernmental Panel on Climate Change (IPCC) conclude that net anthropogenic carbon emissions must be eliminated before the end of this century in order to limit increases in mean global surface temperatures to less than 2 \(^{\circ }\)C [1, 2]. However, 1 PgC has been suggested as a floor for annual agricultural emissions, even if significant progress is made to close yield gaps [3]. In particular, livestock production currently accounts for roughly one fifth of total anthropogenic greenhouse gas emissions, a footprint which must be expected to grow as meat demand scales with the size and affluence of global populations [4, 5]. Given a dwindling global carbon budget, a carbon-negative energy sector stands out as an obvious target for offsetting sustained or growing emissions from other sectors.

Among carbon-sequestering technologies, photosynthesis is simultaneously the cheapest and most efficient known solution. Reductions in atmospheric CO\(_2\) through increased biomass energy production coupled with carbon capture and underground storage (CCS) on a permanent or semi-permanent basis are theoretically possible and technologically foreseeable [6, 7]. However, large-scale biomass production for energy risks inflating food prices and land use change, creating a deep carbon debt which mitigates potential emissions benefits [8, 9]. Large-scale algaculture, an emerging technology, represents a potential source of biomass for energy production while avoiding these tradeoffs. More promisingly, however, microalgae can also be exploited as a source of animal feedstock, offsetting anticipated growth in demand for meat and dairy while allowing vast areas of agricultural land to be repurposed for biomass production or habitat restoration.

We use the FeliX model to assess the emissions and land use consequences of maximal land-based biomass production for energy [21, 22]. In the business-as-usual (BAU) scenario, future global population growth, dietary patterns, energy profiles, and agricultural yields are assumed to develop along historical trends and without perturbation from policy changes or transformational technologies. In the BioEnergy scenario, the expansion of biomass, wind, and solar energy is accelerated exogenously to match more aggressive climate action scenarios [23], simulating an accelerated transition to renewable energies. Together, these scenarios define a baseline against which the impacts of alternative technologies can be measured. The construction and calibration of the BAU scenario are presented and discussed in depth in this paper’s SI and on the model website [21].

Relative to this baseline, we assess the maximum theoretical emissions mitigation potential of microalgae alternately as a source of biomass for energy (Alg-Fuel) and as a feedstock (Alg-Feed). In these scenarios, we impose exogenously the construction of 25-50 Mha of algaculture at a constant rate between 2015 and 2060. The output of these systems (up to 75 tDM ha\(^{-1}\) year\(^{-1}\)) generates an annual total of up to 3.75 Gt DM [10]. On-shore microalgae production systems are assumed to occupy non-arable or permanently fallow land and to use recycled fresh or brackish water. Essential nutrients are supplied from municipal wastewater, agricultural runoff, or manure. Algae can be near carbon-neutral (modulo production and transportation costs), as CO\(_2\) is sourced initially from flue gases from the combustion of fossil fuels and subsequently from bioenergy production. Other potential sources include steel and cement manufacturing facilities, small industries, agriculture, and open-air respiration [11]. To minimize the infrastructure, transportation, and emissions costs, proximity to essential inputs should be a primary consideration in the construction of algae farms [11, 13].

In the Alg-Fuel scenario, algacultural output is burned directly for electricity or used to produce biodiesel. In the Alg-Feed scenario, this output is instead used to meet 40 % of global demand for feed. Extrapolating from FAO feed demand estimates and assuming conventional plant-to-animal conversion efficiencies, the maximum required annual output (40 % of feed) ranges from 0.5 GtDM in 2010 to 1.0 GtDM in 2100 [24]. Thus, this scenario requires up to 25 Mha, allowing for a 50 % reduction in areal productivity (32.5 tDM ha\(^{-1}\) year\(^{-1}\)) due to feed quality standards.

Each of these bioenergy production scenarios is evaluated over a range of energy sector emissions mitigation with CCS (25–75 %) reduction on emissions from the energy sector) to identify the CCS threshold at which each energy mix achieves net-zero anthropogenic emissions.

Demographics in the FeliX model are calibrated to leading exogenous projections. Global population is projected to reach 10.5 billion—a nearly 50 % increase—by 2100 [26], while global GDP per capita is projected to increase fivefold over the same period [27]. This growth leads to increases in per capita demand for energy (+25 %) and vegetal (+20 %) and animal (+40 %) food calories, all of which pose significant, well-documented challenges for global human and natural systems [3]. Within this context, the potential consequences of algaculture for fuel and feed can be evaluated in terms of associated land use changes and carbon emissions pathways.

Energy profile and supply

Figure 1 displays the time series of primary energy production for the BAU, BioEnergy, Alg-Fuel, and Alg-Feed scenarios. Primary energy production in the BAU and BioEnergy scenarios is calibrated to baseline and aggressive climate action estimates, respectively, from comprehensive analyses [7, 23]. As such, these are not model results, but rather define the scenarios and represent benchmarks to frame discussion of the benefits and costs associated with each approach to energy production. Overall, the market share of fossil fuels falls from 95 % at the beginning of the century to 60 % in BAU and nearly 40 % in the BioEnergy scenario. These scenarios delimit a range of energy futures consistent with the GEA and fall between RCPs 4.5 and 6.0 [23, 28].

In the Alg-Fuel scenario, the projected annual output of 50 Mha of algaculture generates 65 EJ year\(^{-1}\) (Fig. 1c), primarily offsetting oil consumption. If algal biomass is instead used as feed (Alg-Feed), agro-biomass production expands to 446 EJ year\(^{-1}\) (Fig. 1d), 90 % of which is due to the productivity of former pasture and feed-crop land. This eliminates all dependence on oil and offsets demand for coal and gas, reducing fossil fuel consumption to under 20 % of primary energy supply.

Land use change

In the FeliX model as in leading estimates, population and GDP growth combine to generate an 80 % increase in total calorie demand by 2100 [3]. This rate of growth in agricultural demand outpaces yield growth projections [29], driving expansion of arable land and permanent pastures and meadows by midcentury (Fig. 2) and increasing land use change emissions. Given this growth, agriculture for food and–increasingly–feed is likely to take up the most productive arable land, limiting the feasibility of large-scale first- and second-generation biofuel production on prime agricultural land [30]. Agricultural residues generate up to 50 EJ year\(^{-1}\) in the BioEnergy scenario, less than a quarter of anticipated bioenergy demand.

Concurrent with agricultural expansion, the land needed to meet anticipated bioenergy demand grows to over 800 Mha in the BAU, BioEnergy, and Alg-Fuel scenarios. This must be expected to exacerbate competition for arable land, driving up food and energy prices and resulting in deforestation as biomass plantations replace natural and lightly-managed forests [30, 31].

Algaculture avoids such tradeoffs, simultaneously satisfying growing demand for animal products and biomass. Used as feed, microalgal biomass can free up to 40 % of pastures, meadows, and feedcrop land. This area—2.0 Bha in 2100 in the BAU scenario—is sufficient to produce annual biomass harvests equivalent to over 355 EJ. In this way, highly-productive algacultural technologies can be harnessed to offset the resource and environmental costs of rising demand for animal proteins.

Emissions

In the BAU scenario, annual emissions from the energy and land use sectors increase by 30 % between 2010 and 2100 despite growth in the total market share of renewable energies. This emissions pathway results in monotonically-increasing atmospheric CO\(_2\) concentrations and a rise in global temperatures nearing 3 \(\,^{\circ }\)C above preindustrial levels by 2100 (Figs. 3 and 4). Marginal increases in biofuel production in the BioEnergy and Alg-Fuel scenarios do reduce carbon emissions, but do not represent the kind of transformation required to avoid 2 \(\,^{\circ }\)C warming–in part, because they do not reduce (and may increase) emissions from land use change.

More promisingly, significant expansion of renewable energies combines with avoided land use change emissions in the Alg-Feed scenario to achieve an emissions pathway between RCPs 2.6 and 4.5 [28]. Even before additional potential emissions benefits from CCS, this strategy is expected to reduce global temperature change 0.7\(^{\circ }\)C relative to BAU by 2100. Figures 3 and 4 project potential emissions and climate mitigation in the Alg-Feed scenario from the superposition of CCS (25, 50, or 75 % energy sector emissions reduction). With the addition of this technology, net zero emissions are achieved when at least 25 % of emissions from the energy sector are sequestered. Higher capture rates transform the energy and land use sectors into a net carbon sinks. The dependence of total emissions abatement in the Alg-Feed scenario (relative to BAU) on essential model parameters and other scenario-specific assumptions is listed in Table 2.

Though we have concentrated to this point on maximizing the emissions impact of algal feedstock by offsetting as much feed demand as possible, projected atmospheric carbon concentrations in year 2100 are shown for a range of algal contributions to animal feed (10–40 %) and energy sector emissions mitigation due to CCS (0–75 %) in Table 3. Yellow cells fall on the threshold for global surface warming (2.0 \(\pm\) 0.2 \(^{\circ }\)C). Green cells indicate sub-2 \(^{\circ }\)C warming in year 2100 of the simulation, and red cells indicate supra-2 \(^{\circ }\)C change.

A range of energy sector emissions mitigation from CCS (at top) are applied to a range of algae production scenarios (listed at left as percentages of total feed consumption). Yellow shaded cells indicate warming of 2.0 ± 0.2 \(^{\circ }\)C. Red shaded cells indicate warming projections in excess of this threshold, while green shaded cells indicate sub-1.8 \(^{\circ }\)C warming

Microalgal feedstock does not need to be maximized in order to contribute to the decarbonization of the energy and land use sectors. On any scale, algaculture creates the possibility of freeing large areas of arable land for biomass production. While the magnitude of the resulting carbon sink varies, microalgal feedstock holds significantly more promise than marginal improvements including conventional methods of agricultural intensification and algal biofuels, and is much closer to reality than other potentially transformational alternatives. For these reasons, algaculture and the land use optimization it enables should be seriously considered as essential tools for climate mitigation up to and including a decline in atmospheric carbon concentrations to preindustrial levels before the end of the century.

The BAU scenario in this analysis forecasts increasing carbonization and warming of the atmosphere and oceans. This results in climate change, which will magnify demographic pressures on food and water security as well as forests and other ecosystems and functional biodiversity. In many ways, however, this baseline assumes the possibility of carrying on with business-as-usual through the end of the present century. Most importantly, it is possible that depletion of cheap fossil fuel reserves or growing appreciation of the threat posed by climate change will lead to greater-than-forecast demand for biomass. In the closed land system, failure to prepare for this transition will exacerbate competition for arable land, degrade food security, and accelerate deforestation and biodiversity loss.

In terms of emissions and temperature change, the BioEnergy and Alg-Fuel scenarios do not differ significantly from BAU. Ultimately, land scarcity precludes the possibility of establishing plantations and energy crops to produce enough biomass to substitute for a significant fraction of fossil fuel consumption [30, 32]. This leaves CCS (on scales that have not yet been proven feasible) or other so far unidentified technologies as alternatives for rapid, urgent emissions mitigation. Further, although the agricultural and plantation yields assumed in this analysis are conservative, harvesting of tree crowns and agricultural residues can deplete topsoils of vital nutrients, increasing dependence on fertilizers. In addition, an increase in demand for land-based biomass such as is foreseen in BioEnergy scenarios carries several hidden costs. In addition to land use change consequences, risks include disruption of ecosystems services including soil carbon sinks, biodiversity, and water cycles [7, 33, 34]. The direct and indirect costs of bioenergy production must be weighed against the benefits of this approach.

On the other hand, we have demonstrated the theoretical and technological potential of microalgal feedstock to relieve land scarcity, allowing arable land to be leveraged to produce cleaner energy while addressing the threats of climate change, deforestation, eutrophication, and food and water scarcity. In recognition of the fact that these systems can be engineered to produce biomass without generating a commensurate burden on critical ecosystems cycles and services, algacultural systems have already been established at some experimental farms to overcome biomass shortages in dry seasons.

The AEF currently produces algal biomass for feed at a cost of $1,840 per dry ton (cf. Additional file 1: Table S2). Prices as low as $500 per dry ton are generally seen as feasible, but only if CO\(_2\) (57 % of AEF costs) can be sourced at no expense. At these prices, algae could have supplied 40 % of global feedstock in 2013 at a cost of 250–920 billion US$(2013). As a comparison, the global gross production value of livestock in the same year was 1,262 billion US$(2013), and fossil fuel subsidies totaled 550 billion US$(2013) [27, 35]. This estimate does not account for the value of algal co-products, nor does it include the land value created by the transformation of low-productivity pastures into plantations.

Algal biomass is already a viable alternative to fishmeal ($1,880 ton\(^{-1}\)) and should eventually compete with soymeal (average $375 ton\(^{-1}\) over the last year) [36]. Critically, growth in demand for conventional protein sources has driven up the prices of both commodities (soy: 88 %, fishmeal: 165 % over the last decade), and this long-term trend can be expected to continue. On the other hand, research and investment in algacultural pilot programs will lead to higher productivities and lower costs for the cultivation, harvesting, and processing of algae at industrial scales. Looking forward, microalgae production systems represent an ideal transition technology from fossil fuels to bioenergy. For governments looking for shortcuts to sustainable development, algal feedstock manages to satisfy the competing imperatives of food security and climate mitigation by reducing resource burdens while commodifying CO\(_2\). On large scales, this establishes the conditions for cascading greenhouse emissions savings and a return to preindustrial atmospheric carbon concentrations.

To be sure, there exist a number of challenges to engineering and operating algacultural systems on the scale envisioned in this analysis. Systems must be engineered which are robust against contaminant or mutant strains of algae, zooplankton, and viruses and other pathogens, which represent threats to stable and highly-productive monocultural systems [20]. As potential sites for colocation with carbon sources are exhausted, carbon capture and transportation infrastructures will need to be expanded at the same rate as algacultural production systems in order to maintain high productivities and low costs. Large scale carbon capture for algae production and permanent sequestration will likely increase energy consumption and costs, as will energy-intensive methods of algal biomass processing. For freshwater systems as well as those that fertilize with wastewater, the feasibility of recycling water through successive harvests in open and closed systems must be studied. Finally, the supply of nutrients to large scale algae production would necessitate an international market for manure and other forms or recycled nutrients from animal and human sources. Additional research should also be done to match leakage points in global nitrogen and phosphorous cycles with algacultural production systems, thereby minimizing the nutrient loads of these systems as well as the deleterious effects of agricultural runoff.

Apart from obstacles to large scale algae production, biodiversity loss–already a problem in the BAU and BioEnergy scenarios—would likely be exacerbated by even greater conversion of pasture and rangeland to energy crop plantations [37]. The elimination of pasture will require higher livestock densities, though rotational grazing in silvipastoral systems can mitigate animal crowding while fertilizing plantations, and the replacement of low quality feedstocks with algae may reduce reliance on antibiotics. Finally, afforestation and reforestation on the scale envisioned here would lower terrestrial albedo in these areas, potentially blunting the cooling impact of greater carbon sequestration [38]. Despite these unresolved questions, this analysis demonstrates that algacultural feedstock at any scale represents a promising and simultaneous solution to food security and climate change, and that these systems merit greater attention and closer scrutiny than they have thus far received.