Biochar as a carbon dioxide removal strategy in integrated long-run climate scenarios

doi:10.21203/rs.3.rs-1805544/v1

Download PDF

Article

Biochar as a carbon dioxide removal strategy in integrated long-run climate scenarios

https://doi.org/10.21203/rs.3.rs-1805544/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 30 May, 2024

Read the published version in Environmental Research Letters →

Version 1

posted

You are reading this latest preprint version

Limiting global warming to under 2°C would require stringent mitigation and likely additional carbon dioxide removals (CDR) to compensate for hard-to-avoid emissions. The relatively low cost and potential co-benefits of biochar as a soil amendment has gained attention in this context. We use a global multisector model to analyze biochar deployment in the context of energy system uses of biomass with CDR under different carbon price trajectories. We find that biochar can create an annual sink of up to 2.8 GtCO₂, reducing global mean temperature increases by an additional 0.5-1.8% across scenarios by 2100 for a given carbon price path. In our scenarios, biochar’s deployment is dependent on potential crop yield gains and application rates, and the competition for resources with other CDR measures. We find that biochar can serve as a competitive CDR strategy, especially at lower carbon prices when bioenergy with carbon capture and storage is not economical.

biochar

carbon dioxide removal

BECCS

co-benefits

GCAM

Limiting global mean temperature change to less than 2°C by 2100 will likely require the deployment of carbon dioxide removals (CDR) to abate hard-to-avoid emissions^1–3. CDR strategies are generally divided into natural sink enhancements like afforestation, soil carbon sequestration (SCS), or accelerated weathering, and engineering solutions like direct air (DACCS) and bioenergy carbon capture and storage (BECCS). A CDR strategy that has gained attention is the application of biochar to soils^4–7. Biochar production is relatively low cost^8,9 and promises co-benefits in agricultural systems like increased nutrient and water retention¹⁰. Thus it potentially results in higher crop yields^11,12, a key benefit in a land-constrained world with a growing population and in comparison to other CDR measures in long-term climate change mitigation scenarios^9,13.

Biochar is high carbon content charred organic matter used as a soil amendment¹⁴. Biochar itself does not remove CO₂; its CDR potential comes from the CO₂ sequestered during photosynthesis. The biogenic CO₂ is transformed into a more stable carbon form via pyrolysis^15–17, a thermal process in the absence of oxygen that deconstructs bio-polymers into biochar (35%), combustible syngas (35%), and bio-oil (30% on a mass balance basis)¹⁸. The pyrolysis conditions (temperature level and retention time) and the feedstock quality (lignin content) define the biochar characteristics and ultimately its recalcitrant organic carbon content.

The ability to store carbon in a recalcitrant form gives biochar an advantage over CDR strategies with uncertainties regarding their permanence, such as afforestation or SCS.¹⁵ Biochar offers potential co-benefits within the agricultural system, including increased crop yields^12,19, decreased use of nitrogen fertilizers²⁰, and increased water retention in soils^7,21,22. Yet, biochar directly competes for resources with BECCS pathways (i.e., bioliquid or biopower production with CCS). The competition with BECCS is interesting due to the energy uses of bioenergy in addition to CDR. Biochar production requires energy, but generates an essentially carbon neutral combustible syngas^7,23. Biochar supply and demand have multiple interacting effects on cropland extension and influence the availability of land for biomass for energy production and other land-based CDR options. These complex dynamics within the energy-economy-land-climate system necessitate a comprehensive multisector framework on a global basis to capture potential land-use leakage effects.

Prior studies have assessed biochar’s potential as a climate change mitigation and CDR strategy, concluding that its mitigation potential is highest as a soil amendment versus a direct form of energy use^7,15,24. Researchers stressed that the potential impact on crop yields is a key driver to biochar economics and that additional analysis with integrated assessments models (IAMs) is needed to assess the complex dynamics across the global agricultural and energy system under climate change mitigation scenarios^15,24.

Global biochar deployment potential is estimated in the literature to range from 1 to 1.8 giga (10⁹) tons of carbon dioxide per year (GtCO₂/year)^5,17 to 4.91 GtCO₂/year⁶. Different sustainability criteria, affecting the amount of available feedstock and thus CDR that can be achieved, are the main driver for this range. The estimates are largely supply driven since there is uncertainty about the potential demand for biochar, both on the impacts on crop performance and critically on the amount of biochar that can be applied per unit of land before it becomes saturated¹⁹. When biochar is valued highly for its CDR potential, these application rates and limits are a key determinant of its potential scale.

Despite sustained interest in biochar as a CDR strategy, there is a need to reduce uncertainties around its global potentials, taking into account supply and demand, and impacts on global agriculture and land availability in competition with other CDR measures^1,22. We use the Global Change Analysis Model (GCAM) to explore the system impacts of assumed parameters in an integrated multisector analysis of global land, energy, water, and emissions. We attempt to address questions and challenges raised in the literature^1,15,23,24 and explore the implications of uncertain parameters in a comprehensive and integrated manner. We construct scenarios by varying uncertain assumptions and evaluate the scale and impact of future biochar deployment. We do not reduce specific technological or physical uncertainties, nor make a priori assumptions about the economic or sustainable limits to biochar and CDR. Instead, we let the amounts respond economically to the scenario and policy assumptions about the feedstocks available (including land for dedicated feedstock production), and the demand and impact on crop productivity, and we then analyze their impacts.

Carbon prices are a key driver of carbon sequestration in GCAM. Here, we apply four carbon price paths: Reference (no carbon price), Low, Medium, and High (SM1). The High carbon price follows a trajectory that would lead to a global warming potential by 2100 of 2.6 W/m²; Medium is half of the High carbon price trajectory; and Low is half of the Medium. Additionally, we have varied application rates (10 or 20 tons per hectare, t/ha), included or excluded yield improvements, and allowed or restricted BECCS. Our analysis includes 4 scenarios without biochar (one per carbon price trajectory), and 32 scenarios with biochar deployment varying the other parameters (4 carbon price trajectories, 2 application rates, 2 yield options, and 2 BECCS options) (SM2).

Biochar Carbon Sequestration

Global biochar carbon sequestration increases more rapidly the higher the carbon price due to the economic incentives (Fig. 1). However, in the long-term biochar sequestration is affected by GCAM’s negative emissions constraint, which can make biochar less competitive. GCAM constrains the amount regions spend on subsidizing negative emissions to 1% of the region’s GDP (SM3). When BECCS is available there is a higher competition for biomass, and thus global biochar carbon sequestration peaks at a given year and later declines. The peak-year is reached by 2095, 2080 and 2065 in the Low, Medium, and High carbon price trajectories, respectively. Before the peak-year, biochar carbon sequestration is the largest in scenarios with an application rate of 20 t/ha and increased yields. When BECCS are unavailable there is less competition for biomass and biochar carbon sequestration continues to grow throughout the century.

Prior to the peak year of deployment, higher carbon prices lead to higher biochar deployment. In each scenario’s peak-year, doubling application rates increases sequestration by 35–89% in Low, 38–84% in Medium and 55–76% in High carbon price trajectories. Modeling yield improvements provides a more significant incentive for biochar deployment, thus there is higher carbon sequestration by biochar in scenarios with this component (465–639% in Low, 376–534% in Medium, and 264–318% in High carbon price trajectories in each peak-year).

Eliminating BECCS results in less competition for biomass resources and higher carbon sequestration from biochar. However, overall emissions without BECCS in the model are higher than those scenarios with BECCS (SM4). When BECCS is unavailable, the increase in biochar deployment is not enough to compensate for the lack of BECCS technology. Finally, our biochar carbon sequestration estimates are within the boundaries found in the literature^5,6,17 (Fig. 1a, grey horizontal lines).

Though biochar allows for higher carbon sequestration, the share of biochar carbon sequestration is smaller than most of the other sectors. Biochar carbon sequestration never exceeds 24% of total carbon sequestration in the second half of the century, with higher carbon sequestration from biochar in scenarios where BECCS are unavailable, as opposed to those where BECCS are allowed (Fig. 1b-c and SM5 for details). Similar to Woolf et al.²² findings, given its relatively low cost of deployment compared to other CDR options in the model, such as BECCS, the proportion of biochar carbon sequestration is much higher in earlier periods than later ones (SM5).

Figure 1| Global carbon sequestration from (a) biochar alone, and (b) from different sectors. The figure shows the 32 scenarios with biochar deployment, with the columns labeled REF, Low, Med, and High for carbon price paths. In figure (a), each facet and color represent a carbon price trajectory. Solid lines represent scenarios where BECCS is available, whereas dotted and shaded lines represent scenarios where BECCS is unavailable. Circle geoms represent scenarios with default yields and application rates of either 10 t/ha (white circle) or 20 t/ha (black circle). Increased yields are represented by triangle geoms, with 10 t/ha (white triangle) and 20 t/ha (black triangle). The grey horizontal lines on Panel (a) represent the biochar carbon sequestration potentials found in the literature: Griscom et al.⁵ estimated over 1 GtCO₂/yr, Woolf et al.¹⁷ estimated a sustainable potential of 1.8 GtCO₂/yr, and Roe et al.⁶ estimated a maximum technical potential of 4.91 GtCO₂/yr. In (b), each facet represents one scenario, each column represents a combination of carbon price and BECCS availability (e.g., “REF & BECCS”), and each row represents a combination of the application rate and yield improvements (e.g., “10 t/ha & Default Yield”). Note that there are 32 scenarios out of 36, since the remaining 4 scenarios do not have biochar deployment.

Biochar Demand and the Effects of Varying Assumptions

Biochar demand is highly sensitive to the assumed application rates and yield improvements on cropland. Longer-term biochar demand is also affected by the negative emissions constraint from GCAM (SM3). Eliminating BECCS reduces the competition for biomass and for negative emissions, and biochar demand increases throughout the century (Fig. 2a).

Total biochar demand in each peak-year ranges between 34–104, 66–776, 68–699, and 74–465 Mt in the Reference, Low, Medium, and High carbon price trajectories, respectively. In a Reference carbon price trajectory, irrespective of BECCS availability, doubling application rates increases biochar demand between 19–30%, and modeling yield improvements increases demand by 133–147%. When BECCS is available, doubling application rates increases biochar demand by 36–89, 38–84, 52–78% in the Low, Medium, and High carbon price trajectories, respectively, in each peak-year. When BECCS is available, modeling yield improvements increases biochar demand by 462–679, 388–541, 253–319% in the Low, Medium, and High carbon price trajectories, respectively, in each peak-year (SM6). When BECCS is unavailable, doubling application rates increases biochar demand by 110–116, 106–162, 115–127% in the Low, Medium, and High carbon price trajectories, respectively, in each peak-year. When BECCS is unavailable, modeling yield improvements increases biochar demand by 72–114, 56–98, 20–24%, in the same carbon price trajectories and peak-years (SM6).

Even though each carbon price scenario peaks at a different year, the behavior observed until this period is similar across scenarios. When BECCS is available, modeling yield improvements has a larger impact on biochar demand than doubling application rates, whereas when BECCS is unavailable, this relationship is reversed. The higher the carbon price, the faster biochar demand grows. When BECCS is available, the competition for biomass leads to a decrease in biochar demand with higher carbon prices since BECCS has a higher carbon capture rate. When BECCS is unavailable, biochar demand increases significantly. Even though biochar deployment allows for a decrease in CO₂ emissions and in climate related variables (SM4), biochar demand for biomass is comparable to other sectors, never surpassing 43% of total biomass consumption in any year, and never surpassing 15% from 2060 (SM7). While demand varies by region, the same general behavior is observed (Fig. 2b, SM8).

Figure 2| (a) Total biochar demand and (b) regional demand for biochar by GCAM region in Medium scenarios. Panel (a) shows the 32 scenarios with biochar deployment. Each facet and color represent a carbon price trajectory. Solid lines represent scenarios where BECCS is available, whereas dotted and shaded lines and shaded area represent scenarios where BECCS is unavailable. Circle geoms represent scenarios with default yields and application rates of either 10 t/ha (white circle) or 20 t/ha (black circle). Increased yields are represented by triangle geoms, with 10 t/ha (white triangle) and 20 t/ha (black triangle). In panel (b) each map represents one scenario in 2100, with the rest of the carbon price scenarios in SM8, as well as 2050 results.

Syngas Coproduct

The production of biochar yields syngas and liquid biofuels as coproducts, though the energy content of the biofuel is negligible in slow pyrolysis^7,23,24. We assume a net syngas coproduct of 0.02 GJ/kg of biochar produced⁷. The amount of syngas produced is directly proportional to the biochar production. The higher the carbon price, the faster the increase in biochar production, thus its syngas coproduct, until the negative emissions constraint is reached. If BECCS is unavailable, biochar and syngas increase throughout the century.

Annual syngas coproduct ranges between 0.7–2.1, 1.3–13.9, 1.3–12, and 1.5–9.2 EJ in the Reference, Low, Medium, and High carbon price trajectories, respectively, in each peak-year in scenarios with BECCS. The lower end represents scenarios with an application rate of 10 t/ha, default yields and BECCS availability, and the upper bound represents a scenario with 20 t/ha, increased yields and BECCS unavailability. When BECCS is unavailable, syngas coproduct increases to 2100 to a maximum of 18.1 EJ, with a similar pattern to the one observed in biochar production: larger effects for doubling application rates than for modeling yield improvements (Fig. 3a).

Global syngas coproduct is a small but important fraction of global natural gas production in the carbon price scenarios, reaching up to 14% of the total (Fig. 3b). We assumed that the syngas is upgraded and fed into the gas market along with other sources of gas, at which point is a carbonaceous fuel. On a total system basis however, the syngas is produced biogenically with the carbon uptake from growing the biomass feedstock accounted for. In the total system, the syngas is net zero carbon and does reduce the net carbon emissions from using gas in proportion to its percentage of total gas consumption as a first order.

Biochar in Cropland

Cropland allocation to biochar grows in all scenarios, though scenarios that have BECCS available see less cropland allocated to biochar after mid-century than their BECCS unavailable counterparts (Fig. 4a), an outcome of the negative emissions constraint. In the Reference scenario, biochar cropland allocation never exceeds 18% by 2100. In the High carbon price scenarios, cropland with biochar increases to almost 60%. In the scenarios with BECCS available, the land allocated to biochar reaches a peak-year from the negative emissions constraint, and the conversion rate of land with biochar decreases. Eliminating BECCS deployment increases cropland with biochar significantly when the carbon price is high, given the large incentive to sequester carbon, and less alternatives to do so.

Historically, most of global cropland allocation is rainfed, and a small portion of it is irrigated. In scenarios without biochar, carbon prices give an incentive to intensify production with advanced technology. When biochar is introduced, the biochar-rainfed technology takes a larger share compared to the biochar-irrigated technology (Fig. 4b). This is a result of the assumption of greater yield improvements from biochar use on rainfed land than on irrigated land, which incentivizes for crops to be grown in rainfed plots (SM9 for cropland details, SM10 for other land impacts).

Water Withdrawal Reductions

Global water withdrawals grow with GDP and population in a Reference case without biochar from 3613 km³ in 2015 to 4539 km³ in 2100. All carbon price scenarios see higher global water withdrawals compared to the Reference case. Scenarios with yield improvements from biochar application have a small reduction of global water withdrawals compared to scenarios without. BECCS availability plays a different role depending on the carbon price: in High and Medium carbon price trajectories, eliminating BECCS leads to lower water consumption in the medium-term, followed by an increase to 2100 compared to scenarios with BECCS. This is because in the medium-term the elimination of BECCS reduces electricity generation demands for water; however, after the peak-year there is a higher demand for water for crops (specifically crops for refined liquids, SM11), which offsets the reductions of water demand in electricity. In the Low carbon price trajectory, the water demand for crops between scenarios with and without BECCS is not as different, so the lower total water consumption is explained by the reductions in electricity water demand (Fig. 5a, SM12).

Yield effects reduce water consumption for crops between 0.1-3% in 2100 (Fig. 5b). Although the percentage is small, global water withdrawals for crops represent almost 60% of the total in the base year (2015), so the absolute quantities are significant. This is caused by two factors. First, modeling higher yields means slightly less land allocated to cropland for the same crop production (detailed in the Methods), which means lower water consumption. Second, based on our assumptions, yield increases are higher for rainfed plots than irrigated plots. Since there is an economic incentive to deploy biochar, and biochar brings greater yields to rainfed plots, there is a shift to rainfed.

Our results suggest that biochar can play a significant role as a CDR option in long-term climate change mitigation. The total global carbon sequestration from biochar reaches up to 2.8 GtCO₂/yr, which is within literature estimates of feasible amounts of sequestration. While we did not explicitly consider specific sustainability limitations in our modeling, our results are based on an integrated analysis of land, energy, and water, and thus can be considered sustainable at least to the first order. In addition to its contribution to sequestration, biochar’s carbon-neutral syngas co-product contributes to decarbonization in the energy sectors to the extent it replaces natural gas.

Biochar does stand in direct competition for resources with BECCS but is more competitive at lower carbon prices. Across our scenarios, we find that biochar and BECCS are complementary technologies. Capping BECCS in favor of a higher biochar deployment results in a higher global warming potential and temperature increases by 2100. BECCS technologies in GCAM capture over 90% of the carbon embodied in the feedstock, compared to the 56% net capture rate of biochar. Yet, BECCS requires higher carbon prices to be economically viable, and is deployed later in the century while biochar sequesters carbon in earlier periods. Thus, from a global climate change mitigation perspective, mechanisms for regional biochar deployment should be incentivized rapidly.

The assumed application rate, an uncertain parameter from the literature in terms of usefulness and saturation points, is a critical component. Without a payment for carbon sequestration, biochar’s profitability is determined by crop yield improvements or reduction in inputs (i.e., water, fertilizer). We find that higher application rates mean more biochar would have to be purchased per unit of land, decreasing the land profit rate, and resulting in less land using biochar than when a lower application rate is assumed. The opposite occurs when carbon drawdown is valued. Then, a higher application rate implies a higher profit rate per unit of land and the share of land in biochar is higher than with a lower carbon value. This suggests that a carbon pricing regime drives biochar deployment and facilitates the activation of co-benefits attributed to biochar in cropland systems.

Finally, our results showed an indirect co-benefit related to water demands in crop production. Because the biochar application was shown in the literature to have higher relative improvement on yields from rainfed crops than from irrigated, biochar had a greater relative impact on the profitability of rainfed crops. In turn, this differential impact on profitability resulted in higher future shares for rainfed crops and lower shares for irrigated crops in the biochar scenarios. Less demand for irrigation meant less demand for water, lowering its cost in places where water is constrained.

Much remains unknown about the physical impacts of biochar such as long-term yield improvements and water demand reductions. The application rates, both minimum useful and the maximum tolerated, are unknown, as is the period over which biochar may need to be re-applied. These questions need to be addressed in empirical field trials, which can be used to recalibrate our model when results become available. Also, our study did not account for the logistical and infrastructural challenges related to a global production and application of biochar.

Our study showed the relevance of biochar as a CDR technology, particularly when carbon prices are low, and the importance of modeling biochar in an integrated framework that considers energy, land, water, and climate. Future work could focus on adding more CDR pathways into GCAM, and potentially other IAMs, to analyze climate mitigation in the full breath of CDR, particularly important given the different tradeoffs of the technologies. If CDR technologies become a facilitator for a well below 2°C target, then having a better understanding of their potential and linked tradeoffs becomes essential.

Overview

Our study considers a slow pyrolysis biochar production pathway maximizing biochar production rather than bioliquids and using lignocellulosic feedstocks based on supply and demand-functions across the integrated land and energy sectors. We develop 36 scenarios with varying assumptions regarding carbon prices, biochar application rates, crop yield improvements, and BECCS availability in GCAM version 5.3 (SM1 and SM2).

Model

GCAM is an integrated, global multi-sector model that links the economy, land, energy, water, and climate systems. The model has been developed for the past 30 years, and it has been used in numerous studies, including the development of scenarios for national and international assessments for the Intergovernmental Panel on Climate Change (IPCC) and the Shared Socioeconomic Pathways (SSP2)²⁵. GCAM is an open-source model and its documentation is available online (see: http://jgcri.github.io/gcam-doc/index.html).

The model represents 32 geopolitical regions, 235 water basins, and 384 land regions. GCAM runs from 1975 until 2100 in 5-year time steps, with 2015 as the last historical calibrated period. Model assumptions include population, labor productivity, technology characteristics and different policies. The outputs include emissions, prices, water and energy supplies and demand, agricultural production, land use, and atmospheric concentrations and temperature. The outputs are not predictions of the future; rather they are a way of analyzing a range of possibilities given different assumptions. GCAM ultimately allows users to explore what-if scenarios.

The land system in GCAM provides information about land use, land cover, carbon stocks and net emissions, the production of bioenergy, forest, and other crops in each of the 384 modeled regions. Demands for these products are driven by the size of the population, income levels and commodity prices. There are about a dozen land types modeled, including protected and commercial lands. Production of approximately twenty crops is modeled, with specific yields, water and fertilizer demands, and costs. Land allocation, crop production and consumption, prices and emissions are calibrated historically. For future periods there is a competition for land based on profitability. Economic land use decisions are based on a logit model of sharing based on relative profitability of using land for competing purposes, which means that the land allocated to any given type is based on how profitable that type is in comparison to other modeled types. The options with higher average profit will get a higher share then one with a lower average profit. The competition happens in a nested structure, where land uses first compete within their node, and then nodes compete among each other. On top of this, each crop can be irrigated or rainfed, and the production of the crop can be a high technology production or a low technology production, depending on the fertilizer application. Figure 1 shows the nesting structure for the land system in GCAM.

The energy system represents different sources of energy supply (both fossil and renewable), modes of energy transformation (such as refining), and energy service demands (such as transport, industrial energy use, and others). GCAM models the following resources: conventional and unconventional oil, natural gas, coal, uranium, wind, solar, geothermal, hydropower and biomass. The energy transformation sectors include electricity, refining, gas processing, hydrogen production and district services. GCAM reports the demands for and the supplies of energy forms as well as the emissions associated with that consumption. There are three main aggregate end-use sectors that demand energy services: buildings, industry, and transportation.

There are different GHGs and air pollutants tracked in GCAM, including carbon dioxide (CO₂), methane (CH4), nitrous oxide (N₂O), different hydrofluorocarbons (HFCs), carbon monoxide (CO), and others. Historical emissions for CO₂ are consistent with CDIAC, and for non-CO₂ with EDGAR 4.2²⁶. Future emissions are determined by the evolution of the drivers and technology mix. In the energy system emissions are driven by input (e.g., fuel) or output (e.g., service produced by a particular technology. In the land system emissions can be driven by output (e.g., crop production) or land area (e.g., forest fires).

CO₂ emissions from the energy sector are determined endogenously based on fossil fuel consumption and global emission factors, which are consistent with the CDIAC global inventory²⁷. CO₂ emissions from land-use/land-use change (LULUC) are tracked separately and depend on total emissions from land, carbon stocks, land area and average carbon density of a specific land area.

Bioenergy in GCAM

Bioenergy in GCAM interacts with two main systems: land and energy. Biomass is grown in the land system and transformed in the energy system into other products for consumption by the end-use sectors.

There are several bioenergy crops grown in the land system, each with specific yields by region, as well as residues that can be used as bioenergy (from different crops and forest). Municipal solid wastes are also tracked. There are dedicated lignocellulosic biomass crops (which is further split between grasses and woody crops), and there is also corn for ethanol, oil crop and palm fruit for biodiesel, and sugar crop for ethanol. Land for biomass production competes with other crops and other land types based on profitability.

Biomass is demanded by different energy transformation sectors, the main ones being electricity generation and refining. In the electricity sector biomass can be coupled with carbon capture and storage (CCS), which works as a negative emissions technology in the model (since the biomass sequestered carbon initially). In the refining sector, GCAM models different biofuel conversion pathways, including biodiesel (from soybean, oil palm or Jatropha), cellulosic ethanol, corn ethanol, sugar cane ethanol and Fischer-Tropsch biofuels. As in the electricity sector, some biofuel refining technologies can have CCS.

Carbon emissions accounting for biomass is based on two different stores of carbon: terrestrial carbon, and that stored in the biomass. Terrestrial carbon is the key driver for any emissions of CO₂ from LUC, with higher emissions happening when a forest is converted into cropland, for example, and lower emissions when land is converted to a type with higher terrestrial carbon. The carbon in the harvested biomass is relevant for the emissions that happen in the energy system, since they are released when the biomass is transformed (even with CCS some of the carbon content of biomass is released). For more information refer to the land documentation (https://www.pnnl.gov/main/publications/external/technical_reports/PNNL-21025.pdf).

Modeling Biochar in GCAM

There are two components for including biochar into GCAM: modeling its supply and modeling its demand. The branch of GCAM version 5.3 used for this project is available at: https://github.com/CandeBergero/GCAM_V5.3_biochar. The following section explains each of them.

Biochar Supply

Biochar is produced in the energy system in GCAM by pyrolysis, a thermochemical conversion process in the absence of oxygen. Slow pyrolysis yields the highest proportion of char compared to syngas and bio-oil, so this is the process we decide to model^7,14,28. There is a broad range for biochar yields in the literature based on the feedstocks used, the kiln temperature and the residence time, among other factors. For our study, we assume that all factors are aligned to achieve the maximum amount of biochar yield possible. Based on Schmidt et al.²⁹ this means a slow pyrolysis process at temperatures between 450°C (to achieve low H/C ratios) and 700°C (to reduce the fraction of syngas coproduct).

Biochar is produced by slow pyrolysis of lignocellulosic biomass feedstocks, following numeric parameters from Roberts et al.⁷ These feedstocks are produced in GCAM from bioenergy crops as well as crop and forestry residues. The cost of the slow pyrolysis facility is 45.93 USD per ton of feedstock (in 2007 USD). The production of 1 ton of biochar takes in 3.65 tons of biomass feedstock (switchgrass). The energy inputs are 211.2 MJ of natural gas per ton of biochar. The process yields as co-products 20,095 MJ of syngas and a small portion of bio-oil. Based on Brown et al.¹⁸, the bio-oil co-product from slow pyrolysis is so low in energy content that we are not modeling it⁷. We model the syngas co-product as natural gas, which can be traded between regions. We are assuming that part of the syngas co-product is used internally to decrease the energy demand of the pyrolysis plant, resulting in a net secondary output of 19,883.81 MJ of syngas per ton of biochar. Table 1 lists the assumptions we are using in our model with its respective source.

Table 1

Biochar Supply Assumptions
Metric	Value	Units	Source
Non-energy cost	45.93	2007 USD per ton of feedstock	Roberts et al. ⁷
Biomass input	3.65	Tons of switchgrass per ton of biochar
Natural gas input*	211.19	MJ per dry ton of biochar
Syngas co-product*	20,095	MJ per dry ton of biochar
Net syngas co-product	19,883.81	MJ per dry ton of biochar
*Note that we assume that the natural gas input to the pyrolysis facility is met by a fraction of the syngas co-product, which means there is no modeled energy input for biochar production.

Biochar Demand

There is some debate regarding the benefits of applying biochar, although there is a consensus that, for diverse reasons that are out of the scope of this research, biochar could bring benefits to crop production^13,15,23. Given these benefits, we are modeling the demand of biochar in the land system, specifically for crop production. We have created two novel crop technology options that demand the biochar created in the energy system, analogous to how fertilizer is modeled in GCAM (created in the energy system; demanded in the land system for crop production). The novel technologies include all the same assumptions as the rainfed and the irrigated high technology options, with the caveat that these novel technologies demand biochar. With this approach, the novel biochar technologies compete with the previous existing technologies based on profitability. So, for each region, basin, crop, and year there is a competition between six technologies. Figure 2 shows the example for corn in any region and basin.

One important assumption regarding biochar demand is related to its application rates, which vary widely in the literature (from 10 to 100 tons per hectare^11,19). For our analysis, we are assuming application rates of 10 and 20 tons of biochar per hectare, and we are limiting this application to one time only in the years modeled. This means that from 2020 (first modeled year) to 2100 (last modeled year) biochar is applied at either 10 or 20 tons per hectare, one time per hectare.

Another important parameter that could favor biochar demand relates to its impacts on crop yields. We performed a Scopus search for biochar crop yield improvements and found 145 publications relevant for analysis. Based on this search, we have assumed average yield improvements that vary based on the climate zone of each of the 32 GCAM regions (temperate or tropical) and based on the watering practice per land type (irrigated or rainfed). The average yield improvements of applying biochar that we are using for our study are 12% and 19% increase for tropical irrigated and rainfed land, respectively; and 10% and 15% increase for temperate irrigated and rainfed land, respectively. To avoid confounding effects of biochar application to cropland, we are not modeling any further benefits in terms of water consumption or fertilizer use.

Based on Schmidt et al.²⁹, we assume that 70% of the carbon content in the biochar contributes to long-term carbon sequestration in soil on a centennial scale (or at least until 2100, which is where our analysis stops). The 70% rate is conservative, as ranges in the literature consider up to 97% sequestration^24,29. Based on our assumptions, the total carbon content in the biomass input to the slow pyrolysis process is split as follows: 56% is permanently stored in the biochar, 24% is released into the atmosphere, and about 20% is stored in the syngas coproduct. These estimates are consistent with ranges in the literature^7,24. We are modeling two main advantages for applying biochar in the soils: increased yields and carbon sequestration. Table 2 lists all the assumptions for biochar demand.

Table 2

Biochar Demand Assumptions
Metric	Value	Units	Source
Application rate	10 20	Tons of biochar per hectare	^11,19
Yield Improvements	12 (tropical irrigated) 19 (tropical rainfed) 10 (temperate irrigated) 15 (temperate rainfed)	Percentage	Refer to SM15
Carbon Sequestered	70	Percentage of Carbon in the Biochar Produced	¹⁶

Even though biochar application could reduce agricultural non-CO₂ emissions, such as nitrous oxide and methane³⁰, we are only modeling CO₂ impacts of biochar application given model capabilities: GCAM does not endogenously estimate non-CO₂ emissions in biomass feedstocks, given that these non-CO₂ emissions are calculated based on Marginal Abatement Curves and calibrated based on EDGAR v4.2²⁶. Additionally, we are not directly modeling the nutrient inputs from biochar into the soil²⁹, although this could be captured in the modeled increased yields. We are also not directly modeling the potential uses of the bio-oil co-product in the bio-economy nor its geological sequestration, as suggested by Schmidt et al.²⁹, which would bring additional incentives for biochar production. Finally, we are also not modeling biochar’s positive or negative priming effects since this exceeds GCAM capabilities and is not a well understood phenomenon¹⁶.

Scenario Design

We have designed 36 scenarios that combine a range of different carbon prices, different application rates, the inclusion of yield improvements from biochar application, and the availability of BECCS. The selection of these four parameters reflects the uncertainty of biochar deployment mentioned in the previous sections, as well as some external drivers that directly affect the competition for biomass. Given the uncertainties for both biochar production and consumption, by combining these parameters our goal is to identify ranges of possibilities, rather than diagnosing future biochar deployment. Table 3 lists the scenario components modeled in our study (SM2 introduces the scenario matrix).

Table 3

Scenario Components
Component	Variations	Acronym	Description
Carbon Price	No carbon price	REF	Carbon prices trajectories. We are modeling four trajectories: a no carbon price (“REF”), and three carbon price pathways. The High case is consistent with a 2.6 W/m² global warming potential by 2100 in GCAM v5.3. The Medium trajectory has prices that are half of the High case. The Low case has a carbon price trajectory that is half of the Medium trajectory. For more details refer to SM1 and SM2
	High	High
	Medium	Med
	Low	Low
Application Rates	0 tons/hectare	0	The application rates are assumed as either 10 (“10”) or 20 (“20”) tons per hectare, applied only once during the modeled years (2020–2100). There is an application rate of 0 tons per hectare for the 4 scenarios that do not have biochar.
	10 tons/hectare	10
	20 tons/hectare	20
Yield Impacts	Default GCAM increase	DY	In GCAM historical yields depend on production and land allocation. Future agriculture productivity increases are assumed based on FAO’s estimates (“DY”). We have increased those estimates following the explanation on section 2.3 to reflect the benefits of the biochar applied to the soil (“IY).
Yield Impacts	Biochar application yield increase	IY
BECCS	BECCS Available	BECCS	GCAM uses BECCS in three energy transformation sectors: electricity generation, refining, and hydrogen production. We have decided to either allow BECCS (“BECCS”) or not (“noBECCS'') for these transformation pathways.
BECCS	BECCS Unavailable	noBECCS

Acknowledgments

This research is based on work supported by the Bioenergy Technologies Office (BETO) of the United States Department of Energy (DOE). The Pacific Northwest National Laboratory is operated for DOE by Battelle Memorial Institute under contract DE-AC05-76RL01830.

P.L. and Y.W. were funded by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. The views and opinions expressed in this paper are those of the authors alone, and do not necessarily represent the views of the DOE or the U.S. Government.

Author Contributions Statement

C.B., M.Wi., and M.We. performed the experiments. C.B., M.Wi., and P.L. designed the study.

Y.W. provided data. C.B., M.Wi., M.We., and P.L. wrote the paper with inputs from the rest.

Competing Interests Statement

The authors declare no competing interests.

Data Availability

The version of GCAM 5.3 used for this study is available at https://github.com/CandeBergero/GCAM_V5.3_biochar. The output data from the scenarios is available at request to the authors.

Amonette, J. E. et al. Integrated biochar research: A roadmap. Journal of Soil and Water Conservation 76, 24A-29A (2021).
Fuhrman, J., McJeon, H., Doney, S. C., Shobe, W. & Clarens, A. F. From Zero to Hero?: Why Integrated Assessment Modeling of Negative Emissions Technologies Is Hard and How We Can Do Better. Front. Clim. 0, (2019).
Holz, C., Siegel, L. S., Johnston, E., Jones, A. P. & Sterman, J. Ratcheting ambition to limit warming to 1.5\hspace0.167em°C–trade-offs between emission reductions and carbon dioxide removal. Environ. Res. Lett. 13, 064028 (2018).
Wu, P. et al. A scientometric review of biochar research in the past 20 years (1998–2018). Biochar 1, 23–43 (2019).
Griscom, B. W. et al. Natural climate solutions. PNAS 114, 11645–11650 (2017).
Roe, S. et al. Contribution of the land sector to a 1.5 °C world. Nat. Clim. Chang. 9, 817–828 (2019).
Roberts, K. G., Gloy, B. A., Joseph, S., Scott, N. R. & Lehmann, J. Life Cycle Assessment of Biochar Systems: Estimating the Energetic, Economic, and Climate Change Potential. Environmental Science & Technology 44, 827–833 (2010).
de Coninck, H. et al. Strengthening and Implementing the Global Response. https://www.ipcc.ch/sr15/chapter/chapter-4/ (2018).
Fuss, S. et al. Negative emissions—Part 2: Costs, potentials and side effects. Environmental Research Letters 13, (2018).
Ippolito, J. A., Laird, D. A. & Busscher, W. J. Environmental Benefits of Biochar. J. Environ. Qual. 41, 967–972 (2012).
Ye, L. et al. Biochar effects on crop yields with and without fertilizer: A meta-analysis of field studies using separate controls. Soil Use and Management 36, 2–18 (2020).
Jeffery, S. et al. Biochar boosts tropical but not temperate crop yields. Environ. Res. Lett. 12, 053001 (2017).
Suruchi, Bhadwal et al. Special Report on Climate Change and Land — IPCC site, Chapter 4. https://www.ipcc.ch/srccl/ (2019).
Kumar, A., Saini, K. & Bhaskar, T. Hydochar and biochar: Production, physicochemical properties and techno-economic analysis. Bioresource Technology 310, 123442 (2020).
Smith, P. Soil carbon sequestration and biochar as negative emission technologies. Global Change Biology 22, 1315–1324 (2016).
Wang, J., Xiong, Z. & Kuzyakov, Y. Biochar stability in soil: meta‐analysis of decomposition and priming effects. GCB Bioenergy 8, 512–523 (2016).
Woolf, D., Amonette, J. E., Street-Perrott, F. A., Lehmann, J. & Joseph, S. Sustainable biochar to mitigate global climate change. Nature Communications 1, 56 (2010).
Brown, T. R., Wright, M. M. & Brown, R. C. Estimating profitability of two biochar production scenarios: slow pyrolysis vs fast pyrolysis. Biofuels, Bioproducts and Biorefining 5, 54–68 (2011).
Jeffery, S., Verheijen, F. G. A., van der Velde, M. & Bastos, A. C. A quantitative review of the effects of biochar application to soils on crop productivity using meta-analysis. Agriculture, Ecosystems & Environment 144, 175–187 (2011).
Borchard, N. et al. Biochar, soil and land-use interactions that reduce nitrate leaching and N2O emissions: A meta-analysis. Science of The Total Environment 651, 2354–2364 (2019).
Razzaghi, F., Obour, P. B. & Arthur, E. Does biochar improve soil water retention? A systematic review and meta-analysis. Geoderma 361, 114055 (2020).
Woolf, D., Lehmann, J. & Lee, D. R. Optimal bioenergy power generation for climate change mitigation with or without carbon sequestration. Nature Communications 7, 13160 (2016).
Woolf, D., Lehmann, J., Fisher, E. M. & Angenent, L. T. Biofuels from pyrolysis in perspective: trade-offs between energy yields and soil-carbon additions. Environ Sci Technol 48, 6492–6499 (2014).
Woolf, D. et al. Greenhouse Gas Inventory Model for Biochar Additions to Soil. Environ. Sci. Technol. 55, 14795–14805 (2021).
Calvin, K. et al. The SSP4: A world of deepening inequality. Global Environmental Change 42, 284–296 (2017).
Janssens-Maenhout, G. EDGARv4.2 Emission Maps. European Commission, Joint Research Centre (JRC) [Dataset]. https://data.jrc.ec.europa.eu/dataset/jrc-edgar-emissionmapsv42 (2011).
Boden, T., Andres, R. & Marland, G. Global, Regional, and National Fossil-Fuel CO2 Emissions (1751 - 2014) (V. 2017). (2017) doi:10.3334/CDIAC/00001_V2017.
Brassard, P., Godbout, S., Pelletier, F., Raghavan, V. & Palacios, J. H. Pyrolysis of switchgrass in an auger reactor for biochar production: A greenhouse gas and energy impacts assessment. Biomass and Bioenergy 116, 99–105 (2018).
Schmidt, H.-P. et al. Pyrogenic carbon capture and storage. GCB Bioenergy 11, 573–591 (2019).
Kammann, C. et al. Biochar as a tool to reduce the agricultural greenhouse-gas burden – knowns, unknowns and future research needs. Journal of Environmental Engineering and Landscape Management 25, 114–139 (2017).

There is NO Competing Interest.

BergeroetalBiocharSupplementary2022.docx

Download PDF

Journal Publication

published 30 May, 2024

Read the published version in Environmental Research Letters →

Version 1

posted

You are reading this latest preprint version

Biochar as a carbon dioxide removal strategy in integrated long-run climate scenarios

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Biochar Carbon Sequestration

Biochar Demand and the Effects of Varying Assumptions

Syngas Coproduct

Biochar in Cropland

Water Withdrawal Reductions

Discussion

Methods

Overview

Model

Bioenergy in GCAM

Modeling Biochar in GCAM

Biochar Supply

Biochar Demand

Scenario Design

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1