3. The Nexus: Water Land Biofuels

The UN Sustainable Development Goals (SDGs) state access to affordable, clean, and reliable energy as a major challenge of our century.^[1] Achieving this goal entails the transformation of the current systems of energy production and distribution which may ultimately also increase the competition for natural resources (i.e. water and land) with the food system. Land, water, and energy are interconnected, both in terms of the natural resources directly used for energy production, and also as a result of land use change and alterations of the carbon and hydrological cycles. In addition, the availability of natural resources, such as water, might be a limiting factor in the implementation of some of the new energy technologies aimed at decarbonizing the power sector. Technologies such as biofuel production, concentrated solar power, and carbon capture and storage require large amounts of water and land.

To curb the increasing atmospheric GHG concentrations in recent years, energy policies have mandated a certain degree of reliance on renewable energy sources as alternatives to fossil fuels.^[2] The synthesis of biofuels from plant biomass (mostly crops) provides the opportunity to rely on energy from geologically recent carbon as an alternative to fossil fuel, especially in the transport sector.^[3] The ability to produce and consume renewable energy locally can help achieve energy independence, and thus energy security,^[4] particularly in countries that lack direct access to fossil fuel deposits.

However, the production of biofuel crops, especially crops for the production of first generation bioethanol and biodiesel, can also have negative impacts on the environment, particularly through land use change and deforestation.^[5] Although bioethanol consumption is for the most part domestic, at least one third of the global biodiesel use is available through international trade, mostly associated with palm oil from Indonesia and Malaysia for the European market. Moreover, biofuels require water and land resources that may otherwise be used for food production.^[6] Therefore, the competing needs for land and water resources by food and biofuel sectors is at the forefront of the energy-food debate.^[7] As a result, a number of outstanding questions on the water-energy-food nexus have arisen, including those related to the effects on food security, environment, and the displacement of land use due to the reliance on the trade of biofuel feedstock.^[8]

3.1 Energy Security and Food Security: The Role of Biofuels

During the second half of the 20th century, rapid population growth and changes in diets generated increasing concerns about the ability of the limited renewable resources of the planet to meet the food and energy needs of humanity.^[9]

For over 1.3 billion people with no access to electricity, bioenergy can help improve energy security and can be available in rural areas to decrease poverty.^[10] Access to reliable and affordable energy is essential for economically and environmentally sustainable development, as emphasized by the SDGs (SDG 7 specifically).

Although not the focus of this report, it is important to note that today 2.8 billion people in the world burn wood and agricultural waste (solid fuels) for cooking and heating.^[11] This is an inefficient 'traditional' bioenergy source, which causes respiratory illness and approximately 1.6 million deaths per year, mainly of women and children.

In India, solid fuels contribute to about 63% of the total household energy consumption, having great impacts significantly to both CO₂ emissions and hazardous indoor air quality. Cambodia, with an estimated 1,304 deaths per million people and India, with some 954 deaths per million, occupy the top two positions in deaths attributed to indoor pollution, one of the leading causes of mortality in the world.^[12]

In addition to improving indoor air quality, modern bioenergy can potentially help improve food security by maximizing land productivity and agricultural management, building cooperation throughout the biomass and food supply chain. Modern bioenergy production, however, requires post-farm processing and related infrastructure (e.g., biorefineries) that is often lacking in rural areas across the developing world, suggesting the existence of a gap between biofuel crop production and its use to meet local energy needs. Around 70- 80% of food insecurity occurs in rural areas, where energy insecurity or energy poverty are also concentrated.

Bioenergy might also play a part in sustainable energy supplies, even with increasing food demands from rising urbanization. The bioenergy sector can also create a new market for producers while at the same time offering new kinds of employment which will spur economic growth, increasing rural incomes and lowering poverty rates. Opportunities could arise in the areas of biofuel production, processing, transportation, trade and distribution. Employment can grow both geographically and in related sectors. IRENA reports liquid biofuels as one of the major employers in the renewable sector, with jobs concentrated in feedstock supply. Brazil, China, the United States and India are key bioenergy job markets.^[13] In addition, the provision of power generated from biomass sources can contribute to rural development by improving the energy access for rural communities who lack grid connectivity, though biofuel crops are often hardly usable for local energy needs. Energy access can enhance agricultural productivity, food preservation, and access to markets, all of which have direct consequences for food security. Nevertheless, bioenergy can add an element of competition for certain food stocks which then changes incomes and food prices. Income influences both the quantity and quality of food consumed by households. In general, higher food prices hurt net food consumers but farmers who are net food producers are likely to benefit from higher prices and increase their incomes.

Hence, questions have arisen about the sustainability of biofuels and their potential role in energy security because of the tradeoffs that biofuel production generates in terms of food, water, and land consumption.^[14] Biofuels can affect food security through two principal pathways. First, they compete for the same natural resources used to support food production. Second, they may compete with traditional agricultural commodities and affect food security outcomes (Figure 3.1).^[15]

Additionally, biofuel production and food security are linked through food prices, employment and incomes, rural development, and poverty reduction.^[16] However, some studies in the existing energy literature (e.g., Da Silva, 2005; Billen et al., 2015; Nelson et al., 2009; West et al., 2014) show the possibility to increase resilience in the food-energy-water nexus, minimizing the exploitation of water and land resources through the use of green technologies and "sustainable intensification" of food production, as discussed later.^[17]

3.1.1 Food and Energy Trade-Offs: Biofuel Competition with Food

This section addresses food and energy trade-offs considering the Food and Agriculture Organization's (FAO) four pillars of food security (i.e. availability, access, utilization, stability) and the interconnections among them.^[18] It considers the impact of competition amongst finite natural resources, first on food availability and utilization, then on price variations (affecting access), and finally on food stability through analyzing market trends.

Availability

Biofuel production can have direct and indirect effects on food availability.^[19] On the one hand, first generation biofuel crops can be used directly as food; on the other, feedstock production implies the use of water and land resources that could be available for additional food production.^[20] Therefore, first generation biofuels are at the core of the food-energy-land-water nexus because of the destination of edible crops as fuel for energy production rather than as food for reducing malnutrition.^[21] In this regard, analysis made by Brown shows that the fuel that fills a regular car tank could feed one person for one year.^[22] The interlinkage between food security and first generation biofuels has been analysed by Rulli et al. in terms of food unavailability.^[23] In particular, Table 1. shows that about 200 million people could be fed with crops (in terms of calories) used for bioethanol and 70-80 million people with the caloric content of total biodiesel production. Although second and third generation biofuels do not have a direct impact on food availability, they can potentially have indirect impacts since they could be used as animal feed, e.g. wheat, barley, and oat straw.^[24] Greater availability of food, however, could also result in the aftermath of investments in bioenergy as a result of positive spillovers that could increase food supplies through intensification (i.e., increasing crop yields), such as mechanization, hydraulic infrastructures and technology in general.^[25] This issue is explained in detail later.

Access

Food can be available but not accessible to the poorest groups of the population.^[26] Food access is strictly related to the volatility of commodity prices and to population income.^[27] Hence, access depends on the equilibrium of food supply and demand over time. In case of increases in food crop demand for alternative uses than food, crop prices can rise if supply does not keep pace with the demand.^[28] In fact, Naylor et al. have observed that combining energy and food sectors led to food price spikes after a long time of declining prices.^[29] For example, in 2006, 20% of corn was diverted from food use to biofuels production, resulting in the increase in food prices registered between 2003 and 2008.^[30] Moreover, Hochman et al. suggest that biofuels have contributed to a 25% increase in corn prices in 2011 with respect to 2001.^[31] Considering FAO estimates, the food price indexes of cereals, oils and sugars more than doubled in 2011 compared to the 2002-2004 average.^[32] In general, biofuels production could favour crop producers (e.g. rural farmers) and damage food consumers, since producers may enjoy higher profits from biofuels production rather than food production and the consequent lower crop supply to the food system would result in higher food prices for consumers.^[33] However, several studies assert the opposite suggesting that over time food prices could drop^[34] because investments in mechanization and other management practices aimed at increasing biofuel production to maximize profits could boost crop production, allowing the crop demand to be met, thus balancing the market.^[35] Furthermore, investments in the bioenergy sector (e.g. production, processing, transportation) may lead to the creation of new job opportunities and rural development.^[36] Higher incomes may improve food security when they benefit rural livelihoods, especially small-scale bioenergy producers therefore allowing for rural development, and thus greater food access from the consumer's side.^[37] In conclusion there is no consensus in the scientific arena on the role played by biofuel production on food access, calling for more in-depth cost-benefit analysis at different scales accounting also for environmental externalities.

Stability

Food stability is closely linked both to market trends of crop prices and to the availability of crop supply over time. In fact, the FAO's fourth pillar of food security states that people should have enough and nutritious food at all times.^[38] On the one hand, biofuel markets can improve the security of farmer incomes and energy self- sufficiency; on the other hand fluctuations in food supply can exacerbate food insecurity conditions.^[39] Biofuel energy markets can divert food crops from the food sector to the energy sector making the food system less resilient to shocks. Moreover, potential shortfalls in food supply can lead countries to import commodities, hence reducing their self-sufficiency making them more dependent on international trade and less stable in the face of market fluctuations.^[40]

Considering trade-offs among finite resources, the expansion of first generation biofuels endangers the capacity to ensure adequate and sufficient food, especially in the case of external shocks (e.g. floods, droughts, pandemics) to the food system. Deprived of its surplus, the food system is less resilient to external catastrophes and shortages, and food stability over time is threatened.

Stability can be achieved with direct state intervention in the food economy through tariffs, subsidies, and governmental policies in order to restore balance between supply and demand.^[41] For this reason, the sustainability of biofuel production and energy infrastructure and technologies must be addressed together with food security.^[42]

TABLE 3.1 People that could be fed with the caloric content of the food crops used as biofuels feedstock, in absolute terms and for each TJ of produced biofuel.^[43]

3.2 The Pressure of Biofuels on Land

The environmental effects of first-generation biofuel production expansion have led to criticism and debate. The increasing demand for ethanol implies a consequent increase in demand for crops such as corn in the United States and sugarcane in Brazil, causing concerns over land use change.^[44] The effects can be direct or indirect in countries such as Brazil, with biofuel crop plantations replacing pastures, and new pastures replacing forested areas.^[45] Similarly, the use of corn-based ethanol to replace gasoline in the United States could cause an increase in CO₂ emissions as a result of land cover change, within and outside of the country borders (e.g., expanding cropland area particularly on more marginal lands, including grasslands and wetlands).^[46] Likewise, the boom of oil palm plantations in Southeast Asia, in response to biofuel and oil crop markets, is having important impacts on the high biodiversity of old-growth forests, including substantial emissions of GHGs from deforestation and particularly drainage of carbon-dense tropical peatlands.^[47] Depending on the previous land cover and its carbon storage, the effects of negative net GHG emissions, which are considered to be the potential advantage of biofuels over conventional fossil fuels, can be nullified for decades (and even centuries). This positive carbon balance (i.e., positive greenhouse gas emissions) will persist until the "carbon debt" from the increased GHG emissions caused by deforestation has been paid off.^[48]

3.2.1 Current Land Use for Biofuels Production

Cropland area covers more than 1.56 billion hectares worldwide. These include -- among others -- areas cultivated for the production of grain, oilseeds, protein, sugar, fibres, fruits, and vegetables. The FAO estimates that 34% percent of the total global land surface is "to some extent" prime and good land for rainfed agriculture (4.5 Bha). Of this area, 1.56 Bha is already in crop production and 1.8 Bha is classified as forest, protected areas, or urban.

Thus, there are about 1.2 Bha of additional land that could be used for crop production, likely at the expense of savannas, grasslands, pastures, and ranges, which provide unique habitat to a variety of plant and animal species. About 26% of this land is in Latin America, 32% in Sub-Saharan Africa and most of the remainder in Europe, Oceania, Canada, and the USA.^[49] Union zur Förderung von Oel und Proteinpflanzen (UFOP) (2020) estimates most cropland is used for food production, while only 5% of cropland is dedicated to biofuels production.^[50] Excluding the commercial co-products from the gross biofuel land area, only 2.4% of arable land is used for biofuel production.^[51] On a global scale, corn and soy are the most cultivated crops for the production of biofuels, contributing to 67% of the total cropland area for biofuel. Figure 3.2 shows the amount of cultivated land for each type of crop for biofuel production. Between 2000 and 2010, the net 'increased area' (net of co-products) associated with biofuels was 13.5 Mha (24.9 Mha in total, where 11.4 was associated with co-products). This area is evenly used for the production of bioethanol (6.8 Mha) and biodiesel (6.7 Mha). The additional area necessary for co-products is roughly 6 Mha for bioethanol (almost all dried distillers grains with solubles (DDGS) in the USA) and 5.4 Mha with biodiesel (mostly EU rapeseed and then US soy).^[52]

FIGURE 3.2. Total Cultivated area used to produce Crops for Global Biofuel Production (OECD, USDA, Oil World (2018)).

Biofuel yield varies among crops and planting regions. Ethanol has generally higher yields than biodiesel and sugary feedstocks are more productive when compared with starchy ones (Table 3.2).

TABLE 3.2 Biofuel Past and Future Yields 2005-2030 (IEA)

3.2.2 Land Use Change and Greenhouse Gasses Emissions

Given the different pressures on finite natural resources, biofuels production can contribute directly or indirectly to land use change (respectively dLUC and iLUC).^[53] dLUC includes changing cropping systems on existing agricultural land or through extensification of biofuel feedstock production on available land for cultivation.^[54] At the same time, a controversial aftermath of biofuel production is iLUC which includes the indirect expansion of food crops on high value ecosystems in order to keep pace with food and feed demand.^[55] This is exemplified by the rise in food prices after the shift in end use (e.g. crops produced for fuel instead of food) in 2008 and the consequent expansion of crop production in pastureland or forests.^[56] These land use changes may lead to the release of CO₂, potentially enhancing climate change. An example of the exact impact of two crop expansions in Brazil for biofuel production is demonstrated in table 3.3.

TABLE 3.3 Land Use Change in Brazil^[57]

A widely used argument in favour of biofuel expansion is the need to reduce GHG emissions by replacing emission intensive fossil fuels.^[58] The direct CO₂ emissions are lower for biofuels than for traditional fossil fuels, mainly because the crop cultivation phase of biofuel production contributes to carbon sequestration through the photosynthetic process, which compensates the emissions of the processing and utilization stages.^[59]

However, carbon sequestration can be overwhelmed by indirect emissions of biofuel production due to land use change (e.g. in response to the biofuel expansion plans in Brazil or through the European oil palm demand in Malaysia and Indonesia).^[60] The CO₂ emissions associated with deforestation to make room for biofuel crop cultivation, are the results of burning or decomposition of forest biomass, and the oxidation of organic soil are high and. The magnitide and lifetime of such emissions to may nullify the benefits of biofuel production, especially in the medium and long term.^[61] Fajardy & Mac Dowell report land conversion factors by Fargione et al. in terms of tons of CO₂ produced for each hectare of native land that is converted to biofuel production.^[62]

The values, reported in Table 3.3, vary strongly depending on the type of native ecosystem and on the biomass produced: oilseed plants for biodiesel production produce roughly 600 tons of CO₂ per hectare of converted tropical rainforest; this can more than double when peatland is converted, and the latter value can increase up to 3452tCO₂/ha.^[63] Corn and sugarcane for ethanol production range around 150tCO₂/ha depending on the crop, the location, and the converted native ecosystem. This effect can be mitigated by converting abandoned cropland or marginal land, lowering the conversion factors to 6tCO₂/ha and even 0-70kgCO₂/ha respectively.^[64] This is more viable for second generation biofuels because energy crops require less agricultural inputs than food crops to attain acceptable yields.

Direct land use change may not take place for first generation biofuels when already cultivated land is diverted to biofuel production; instead it triggers indirect land use change. Converting food crops to biofuel production reduces the availability of land for food production and increases the price of food crops.^[65] Thus, rangeland and cropland are expanded through conversion of native ecosystems, in response to the market alteration.^[66] One study found that, "By using a worldwide agricultural model to estimate emissions from land-use change...corn- based ethanol, instead of producing a 20% savings, nearly doubles greenhouse emissions over 30 years and increases greenhouse gases for 167 years." The magnitude of the effect is such that the replacement of fossil fuels with ethanol from corn results in a 20% decrease in GHG emissions not including indirect land use change but a 97% increase in GHG emissions when accounting for indirect land use change (Table 3.8).^[67] The use of trade and of second generation biofuels from residues can be a step in the right direction, but it may not be enough.

Hertel et al. find that resorting to market and by-product use to produce corn ethanol reduces GHG emissions per unit energy by 75% with respect to previous estimates, but this reduction is still not enough to reach a non- positive net GHG balance.^[68]

The areas where land and water are available for biomass production are not necessarily those where biofuel demand originates or where fossil fuel production occurs. Thereby, an expansion of biofuel production able to meet future energy demand scenarios, be it first or second generation, would require the setup of an international supply chain for biofuels comparable to the one currently available for fossil fuels, with the consequent CO₂ emissions.^[69]

Another issue is that CO₂ absorption effects are longer term than the almost immediate GHG emissions derived from land conversion. This generates a 'carbon debt' whose payback times often exceeds the life cycle of cars and power plants.^[70] Payback times have been estimated to range between 2 and 9 decades for bioethanol and 1 and 4 centuries for biodiesel.^[71] Table 3.4 reports carbon debt per unit area and payback time for direct and indirect land use change of different ecosystems in Brazil, Indonesia and Malaysia, and the U.S.

In summary, first generation biofuels may produce more GHG emissions than they could mitigate if agricultural expansion is not well planned.^[72] The fact that the consequences of indirect land use change are delayed in space and time generates a variety of spillover effects and socio-economic externalities. The land claimed to expand rangeland and cropland in response to food crop diversion from food to biofuel production displaces the effect on people and ecosystems to other locations from those that could benefit from biofuel production. Moreover, the marginal land that can be used for low-emission land conversion may be vital for subsistence farmers.^[73] Further comments on the possibility to sustainably expand biofuel production using marginal land can be found later in this chapter. Use of crops residues seems again to be the most sustainable option for biofuel expansion, but whether residual biomass alone will be able to meet future bioenergy demand remains an open question.

TABLE 3.4 Carbon debt per unit area and payback time for direct and indirect land use change of different ecosystems in Brazil, Indonesia and Malaysia, and the U.S.

3.3 The Pressure of Biofuels on Water

Biofuels have received increasing attention and support by policy makers as an instrument for sustainable development to ensure economic growth while reducing fossil fuel dependency and increasing the renewable share of energy consumption.^[74] However, the impact of biofuel production on freshwater resources has only been evaluated recently, despite the much greater rates of water consumption for biofuels with respect to traditional fossil fuels.^[75] An often overlooked, yet interesting explanation for the difference in water consumption between biofuels and fossil fuels is that fossil fuel water consumption only accounts for extraction and processing water inputs. Indeed, the biomass that generated fossil fuels was produced in past geological eras by the same transpiration process that is needed to sustain biomass today, and thus fossil fuel production likely required the consumption of similar amounts of water.^[76] This water is not accounted for because it was consumed in previous geological eras. Conversely, biofuel production adds stress to currently available water sources. Considering the water crisis the world is facing, with two thirds of the global population living in conditions of water scarcity for at least a part of the year, additional pressure on water resources for the biofuel industry and its competition for water with the food industry is a problem that needs to be addressed thoroughly.^[77]

3.3.1 Current Water Use for Biofuels Production

Globally, irrigation water used for biofuel production is estimated by the World Water Assessment Programme at 44 km3, or 2% of all irrigation water in 2014.^[78] With the existing production technologies it takes an average of roughly 2,500 litres of water (about 820 litres of irrigation water) to produce 1 litre of liquid biofuel.^[79]

The share of irrigation water used for biofuel production is negligible in Brazil and the European Union, where crops are mostly rainfed, while it is estimated to be 2% in China and 3% in the United States. Analysis by Jeeam in 2014 showed that implementing all current national biofuel policies and plans would take 30 million hectares of cropland and 180 km3 of additional irrigation water, almost four times the current water demand.

Water is required in most stages of biofuel production. Most water use for biofuel production -- roughly 99% -- is for the cultivation of crops, but it is also important, especially in a policy context, to consider both water withdrawal and consumption in the processing stage, which might have more intense local effects.^[80]

Water in the cultivation stage is essentially needed to support the plant evapotranspiration process. Evapotranspiration depends on climate and weather conditions, principally air humidity, radiation, wind speed and temperature; the potential value rendered by the atmospheric demand of water is then modulated by the needs of the specific crop in each of its growing phases.^[81] Plants use photosynthesis to chemically convert water and carbon dioxide into primary and secondary metabolites: primary metabolites are simple organic molecules such as glucose, cellulose, and starch, while secondary metabolites are less abundant, more complex, and more valuable.^[82] These secondary metabolites are the feedstock that is refined to produce biofuels. Photosynthesis relates to transpiration through stomatal regulation in that, as plants open the stomata to sequester atmospheric carbon, they lose water vapor (transpiration). Plants capture water primarily from the ground through the root apparatus. When the water required by the plant's vital processes is more than what the plant can currently abstract (i.e., relying on soil moisture naturally replenished by precipitation), additional water can be supplied by means of irrigation to avoid crop water stress and subsequent yield reduction.^[83] Moreover, especially for food crops (thus for first generation biofuels), N and P fertilizers are used to increase yield, which requires substantial amounts of water (grey water) to dilute pollutants, in addition to having energy- and carbon-intensive supply chains.^[84]

The principal feedstock sources for bioethanol are corn and sugarcane, followed by wheat, sugarbeet, and sorghum, but other starch-rich crops such as potato and cassava or other cereals such as barley, rye, and rice are also used.^[85] Sugarcane is the highest ethanol biomass contributor, whereas corn is the most yielding, most water-intensive, and the most used bioethanol feedstock, accounting for two thirds of world bioethanol production.^[86] Oilseed plants such as soy, oil palm, and rapeseed are the main sources for biodiesel.^[87] Oil palm is the most water- demanding feedstock when compared to soy, however soy yields are smaller per unit of land.^[88] Any local deficit in biomass production is compensated by trade of biomass (and thus essentially a virtual water trade), by importing the biomass from exporters. This traded biofuel accounts for 3% of world bioethanol production and up to 20% of the OECD+EU27 countries' biodiesel production - thus, bioethanol is produced and used mostly domestically; biodiesel is also produced and used mostly domestically but is traded internationally at higher levels.^[89]

The main energy crops for second generation biofuels production are herbaceous plants such as miscanthus or ligneous plants such as pine or eucalyptus.^[90] Energy crops have generally lower water requirements than food crops, and provide satisfying yields with low inputs, but they present usually low biomass-to-biofuel yield.^[91]

Residues for second generation biofuels are typically agricultural and forestry byproducts such as leaves and straw or solid organic wastes.^[92] The water input for residues are low, as they share the water requirement of the primary product they derive from.^[93] Leaving crop residues on the field to increase the soil organic content and to reduce soil evaporation and erosion is a common and advisable farming practice, but using some of them for farming and bioenergy is also a viable option.^[94] However, removing straw and agricultural byproducts from food crop fields creates the need for additional fertilizer, generating additional grey water outputs.^[95] Some competition may arise as residues are also used for livestock feeding, but given the increasing global meat demand, livestock production systems are getting more and more industrialized, thus moving away from this market.^[96] This demonstrates the complexities of even advanced biofuel production and its impact on direct and indirect water usage.

Biomass is converted to biofuels or bioenergy by thermochemical or biochemical processes, depending on the source and the final product. The grain product must be ground and mixed with water to enable the cleavage of polysaccharides into glucose by yeast.^[97] The bacteria then anaerobically digest the glucose generating ethanol and carbon dioxide. The mixture is then distilled to separate the ethanol: water is lost in this process (though in amounts much smaller than water losses in the field by transpiration), mainly through the heating and cooling process, incorporation in the final product, and in the DDGS (distiller's dried grains with solubles) by-product.^[98] Beside the direct loss of water due to evaporation or system efficiency, cooling water generates a grey water footprint due to side production of ammonia and sulfuric acid as well as thermal pollution.^[99]

Second generation bioethanol is produced from lignocellulosic biomass with a hydrolysis-fermentation process analogous to first generation bioethanol, but the complex structure of lignocellulosic biomass requires some additional pre-treatment that can be water- and heat-intensive, such as steam explosion to disintegrate biomass.^[100] Water consumption in the processing phase for first generation bioethanol has been estimated as 3 liters of water per liter of bioethanol (L/L), excluding the water consumption associated with the treatment plant setup, whereas for second generation bioethanol the value rises to 9.8 L/L.^[101]

Whole-crop biorefineries are the plants dedicated to the conversion of biomass into a range of products, including biodiesel, through a stepwise process analogous in principle to traditional crude oil refineries.^[102] In biorefineries, vegetable oil is extracted from oil crops and converted into biodiesel through transesterification, a chemical reaction that requires alcohol as a reagent.^[103] Water use and loss in the biorefinery process include washing water to remove residual catalyst and condensation water to recover solvents.^[104]

Third generation biodiesel is obtained from algae, also through a biorefinery process. The extraction process of fatty biomass is easier for algae than for oil crops, and the by-products are still marketable.^[105] Water consumption in biorefineries is about 1 L/L, regardless of the feedstock.^[106]

3.3.2 Water Footprint of Biofuels

The 2016 World Energy Outlook by the International Energy Agency (IEA) states that, while agriculture is and remains the most water-intensive sector, energy production and power generation are projected to have substantially increased impacts on water resources by 2040, with water withdrawal and consumption rising by 2% and 60% respectively.^[107] This stronger relative growth of water consumption to withdrawals is also related to the aforementioned policy-driven boosting of the biofuel sector and is caused by the water demands of the crops cultivated to generate biomass.^[108] The best and most widely used approach to assess water consumption by a process or a product (in this case, biofuels) is the water footprint approach proposed by Hoekstra et al.^[109] The water footprint is made up of 3 components: a green, blue, and grey water footprint. The green water footprint is an indicator of the use of water originating from rain that does not generate runoff, but is instead stored in the soil or in plant biomass and eventually evaporates or transpires. The blue water footprint is water withdrawn from a surface or subsurface water body and consumed in a given process, e.g. irrigation water for non-rainfed crops.

The grey water footprint is an indicator of pollution representing the volume of water necessary to absorb the water pollution load associated with a given process. In the case of biofuels, not only the water footprint itself, but also its repartition among its green, blue, and grey components is strongly dependent on the crops used; the location, in terms of climate and soil; and on the type of biofuel generated - first, second, or third generation biofuels.^[110]

Table 3.5 shows water footprint values for a range of first and second generation biofuel feedstocks. First generation biofuels have an average water footprint ranging between roughly 40 and 150m3/GJ, 80% of which is green water.^[111] Second generation biofuels from crop residuals have a much lower green water footprint, since only a small part of crop evapotranspiration goes into the production of the residual biomass used for producing this type of biofuels.^[112] In fact, water footprints of crop residuals in Table 3.5 are lower both in terms of feedstock mass and energy production, meaning that, even when the processing stage is more complex than for first generation biofuels, the lower water footprint in the cultivation stage propagates along the whole producion chain. For second generation biofuels, energy crops are not usually irrigated nor fertilized, so their biomass water footprint has no blue or grey component.^[113] Although their water footprint per ton of feedstock is comparable, if not lower, than that of first generation biofuels, their green water footprint per unit energy is often higher, as can be seen in Table 3.5. This is because the conversion process is more water intensive and the caloric yield of energy crops is lower. Moreover, the longer life cycle of energy crops may alter the groundwater recharge system for longer periods, and their processing stage is water-intensive.^[114] Nonetheless, the second generation biofuel water footprint does not directly compete with food, as their biomass has no alternative food-related use.^[115]

However, competition may be possible for energy crops in case of future biofuel expansion, as they would use water and land resources that could otherwise be destined to food crops for food production.^[116] Third generation bioenergy presents complementary issues. Since third generation bioenergy is derived from algae, it has only a blue water footprint, ranging from 8 to 35 m3/GJ, but if bioenergy production were to expand through third generation biofuels, the increase in blue water consumption to fulfill future demand as given by the 2040 IEA energy scenario would pose serious challenges to global freshwater reserves.^[117] The effects of water consumption, especially if compared with greenhouse emissions, are typically dependent on where and when water is consumed.^[118] First, volumes of consumed water that have a negligible marginal effect on the global freshwater availability may have significant impacts on a regional or local scale.^[119] Second, differences between distribution in space and time of biofuel demand and water availability would be compensated by virtual water trade, resulting in market alterations as explained by Rulli et al. in the "3.1.1.Competition with food section" of this chapter.^[120] From a water footprint perspective, a more viable strategy to tackle the increasing demand of biofuels, and bioenergy in general, without entering competition with food seems to be the employment of second generation biofuels from crop residuals.

TABLE 3.5 World average water footprint (WF) per unit mass of feedstock and unit energy for different types of biofuel.

3.4 Land and Water Availability without Competing with Food and Causing Unwanted Ecological Impacts

In the face of a steadily increasing world demand for food, the two rooted complementary strategies for increasing agricultural production are intensification and expansion.^[121] Agricultural intensification is the increase in land productivity through increased inputs, such as water and fertilizers, or optimized practices, whereas extensification is the increase in production through expansion into uncultivated areas.^[122] These two approaches should not be practiced without consideration for spillover effects and sustainability factors, since, as Table 3.6 shows, we are already pushing the limits of our planet in terms of water and land resources, and fertilizers can have significant negative environmental impacts.^[123] Several studies, for example Tilman et al., analyzed the extent of these combined effects according to present trends, finding that there could be an increase in 1 billion hectares of land and in nitrogen use of 250 Mt per year by 2050.^[124] To meet future demand within planetary boundaries it is widely recognized that the only viable solution is the sustainable intensification of agriculture.^[125]

TABLE 3.6 Natural Resources use and Planetary Boundaries

Therefore, finding the combination of interventions able to increase agricultural production in a sustainable way is not only a core issue for the biofuel sector, but for the whole food system, and thus requires comprehensive policies.

3.4.1 Surplus from Boosting Feedstock Productivity (Intensification)

Intensification can help in ensuring increased productivity through irrigation, mechanization, and regionally specific inputs including fertilizers and seeds, but it may also generate several externalities including water scarcity, freshwater resources pollution, the emergence of dead zones, biodiversity losses, and large-scale land acquisition (LSLAs).^[126] For example, intensification through increased inputs is sustainable only if this increase does not generate additional environmental costs. Sustainable intensification allows water to be withdrawn for irrigation only where the withdrawal does not generate water scarcity. Likewise, in sustainable intensification practices the environmental optimum for fertilizer use must be sought, although it may not coincide with the economic optimum.^[127] Alternative approaches to yield boosting can be achieved through improved management practices, and through optimizing choices in crop varieties. Irrigating crops that are currently rainfed can raise yield and boost production: adding about 400 km3 of irrigation water on currently rainfed agricultural land could feed 2.8 billion people.^[128] Additional water inputs, if well calibrated and combined with fertilization, can increase the production of biomass in a way that the GHG emissions per unit of biofuel-generated energy are substantially lowered.^[129]

Fertilizers, in particular nitrogen (N), are popular and established instruments for yield boosting but their efficiency depends on the agricultural practices they are combined with.^[130] Moreover, only a carefully tuned fertilization rate can increase yields without negating the positive environmental effects of biofuels.^[131] Tilman et al. points out the need for a global improvement in management practices of fertilizers for sustainable intensification.^[132] There is a need for farming practices that can sustainably boost yields worldwide through crop replacement, efficient management practices, and improved seeds.^[133]

Sustainable intensification can be also obtained by growing crops in the most suitable place for their cultivation, in order to save water and inputs while boosting crop yields.^[134] For example, it is possible to plant crops that use less water and do not induce water scarcity conditions. Therefore, it is crucial to define the suitability of a certain crop considering different environmental boundary conditions (e.g. temperature, soil slope, texture).^[135]

Improved technologies and mechanization can produce higher yields both directly for biofuels cultivation and indirectly for the other crops grown in the area with the same machinery.^[136] Therefore, yield gap closure can be achieved by bringing technology to biofuel producing countries.^[137] The study by Johnston et al. shows that closing the yield gap at the 50th percentile for 20 crops can increase biofuel production by more than 100 billion liters of ethanol and roughly 9 billion liters of biodiesel.^[138] However, mechanization, besides altering the life cycle of the soil, requires investments in the agricultural sector that small farmers cannot afford that could lead to LSLAs^[139] and environmental and social issues, as discussed later in section 3.8.

Gap closure is efficient especially in countries where there is a big difference between the maximum attainable yield and the present yield (e.g. sub-Saharan Africa).^[140] Changing the crop used for biomass generation can push the attainable yield to higher values where the gap is small. Moderately fertilized high-mixture grasslands can provide overall satisfying biomass yields, on areas that would yield much less with food crops, such as abandoned or degraded lands, as reported in the next section.^[141]

Sustainable intensification of agriculture can have positive effects not only on the biofuel production chain, but on the food system as a whole; however a necessary condition to that is the employment of fine-tuned, mixed- approach strategies that are policy-driven. There could be negative social and environmental impacts because of the shift from subsistence farming to high-input and industrialized agriculture, as previously said.^[142] However, sustainable intensification can also help in reducing the competition among resources, therefore increasing the efficiency of water and land use and contributing to maintaining higher environmental quality.^[143]

3.4.2 Surplus from Activating Under-Utilized Low Carbon Land (Especially Degraded Land) for Feedstock Production

Strategies to increase feedstock production for biofuel expansion include intensification, as explored in the previous section, and extensification, which is the focus of this section.

Using marginal land has often been invoked as a potentially successful strategy to produce biofuels without competing with food systems.^[144] Often touted as a win-win solution to this coupled food-energy problem, this approach hinges on a myopic and anthropocentric perspective that does not recognize the environmental value of "marginal land" because of its perceived lower primary productivity and biodiversity. It also ignores possible uses that local indigenous communities often make of that land (e.g., for fuelwood collection, grazing, hunting) by simply labeling "marginal lands'' as "unused".

Marginal lands include areas that are unsuitable or unproductive for conventional crops because of the soil characteristics, climatic conditions, or due to contamination or degradation.^[145] Marginal lands may also be exposed to erosion, salinization, and nutrient depletion.^[146] Therefore, they may not be suitable for food crops unless substantial irrigation and fertilization are adopted, thereby offsetting the positive environmental effects of producing biofuels on marginal lands. In order to use marginal lands for biofuel production, we need to consider the consequences of the use of these lands from an environmental, economic, and social lens. Second generation biofuels from residues do not have a direct impact on land, as they share it with the main product(s) from which they derive. Agricultural residues play a role in maintaining the carbon balance and the fertility of the soil, and thus removing them for biofuel production can significantly reduce crop yields and soil health.^[147] Sacrificing the land's natural yield to increase productivity for a secondary product would not have a net positive outcome in terms or sustainable land management.

Urban residues and residual oil are not constrained from this point of view, as they have no such alternative value. The main problems in this case remain technologic and logistic, as conversion of such complex biomasses is costly and energy demanding, and the supply chain for such feedstock should be integrated from the household level to the biorefinery.^[148] Thus, for extending biofuel production on marginal lands, the focus is on energy crops. The yield of energy crops varies substantially among species, climates, and soils.^[149] Perennial grasses such as Miscanthus and Switchgrass have low agricultural requirements, are drought resistant and, thanks to their developed root system, help in reclaiming soil and preventing erosion, thus eventually improving the soil conditions of marginal lands.^[150] Miscanthus, in particular, requires half of the land and one third of the water used by corn to produce the same amount of biofuel.^[151] Yet, cellulosic crops work best if irrigated, so policies to incentivize feedstock production from energy crops may induce farmers to irrigate permanent grasses, zeroing one of the positive environmental effects these crops had in the first place.^[152]

Another interesting perspective is offered by succulent crops such as Agave. The advantages here are that the water requirements are notoriously low, and that the leaves present high content of soluble non-structural carbohydrate at the expense of lignin content, therefore being much easier to convert, but the global availability of suitable marginal lands for agave is uncertain.^[153] In order to define a comprehensive framework, numerous local studies should be integrated in a global analysis to map the marginal lands and determine their suitability for food crops or energy crops.^[154]

Extensification of feedstock production on marginal land also has a range of secondary effects, both positive and negative. A careful selection of the cultivated species can improve water quality, prevent erosion, and help restore biodiversity.^[155] Moreover, many herbaceous energy crops are almost completely dried out at their harvesting date, reducing the costs associated with transport and drying.^[156] Marginal lands could attain their productive potential if properly managed; a non-invasive management would avoid the carbon costs of conversion, and a minimal and well calibrated fertilizer input may increase the field productivity more than it would increase the GHG emissions associated with the fertilizer use.^[157] However, not all marginal lands are ready and available to be used for the production of herbaceous feedstocks. As noted earlier, land classified as "marginal" is often grazed or used by subsistence farmers in rural poor areas; in these cases, the positive impact on GHG emissions may be questioned, and the legal nature of the acquisition process of these lands would be critical for its socio-economic effects on the local population.^[158] Moreover, not all marginal lands are close enough to biorefineries or to areas that could potentially host one.^[159] Sacrificing one ecosystem or the socio-economic equilibrium of one community to obtain marginal benefits at a global scale may not be a sustainable pathway. Therefore, the conditio sine qua non for biofuel expansion on marginal land is a leap in agricultural practice research, land management policies, feedstock conversion technology, and logistics. There are several studies underway in the U.S. to better understand how to use marginal lands to sustainably produce biofuels, as showcased in the final chapter of case studies in this report.

3.5 Impacts of Biofuels Expansion on Ecosystems

The environmental impacts of European palm oil imports from Malaysia and Indonesia have been highlighted by a number of recent studies. Such impacts from oil palm plantations include high deforestation rates and large carbon emissions in Malaysia and Indonesia as well as losses of habitat and threats to biodiversity.^[160] In response, the European Union has taken some action to limit these unwanted effects on the environment.^[161] For instance, biofuels produced from feedstocks grown on land with "high biodiversity value" (e.g., primary forests, peatlands, wetlands, certain woodlands and grassland) are not accepted under EU renewable energy mandates. The direct and indirect effects of biofuel production on these ecosystems, however, remain difficult to verify.^[162]

In addition to the environmental impacts, biofuel production has important societal implications that can be better understood by examining the energy-food-water nexus of biofuels as discussed in section 3.7. A variety of biofuel production schemes are showcased in the last chapters of this report to highlight real world challenges and successes in the global biofuel marketplace.

3.6 Impacts of Climate Change on Bioenergy Crop Cultivation

Monia Santini, Euro-Mediterranean Center on Climate Change Foundation

While there is robust consensus on the need of biomass for bioenergy according to various national and global energy pathways, future estimates of biomass availability are highly uncertain, especially for those concerning dedicated energy crops.^[163] Among various sources of uncertainty, including the competition for lands, production costs, and sustainability factors, one must also consider the potential effects on crop growth due to a changing climate, which modifies temperature and precipitation patterns. In particular, increasing global water stress, due to rising water demands and reduced supplies, can be further exacerbated in some locations by climate change, with evaporative requirements of plants rising with temperature as vapor pressure deficit rises.^[164]

However, climate change impacts depend on several factors: the crops and regions in question, the modelling approach used, and the consideration of land-use constraints and CO₂ fertilization effects.^[165] This is why evaluations based on model ensembles are often adopted in order to consider a range of likely outlooks, by averaging results and/or labeling them based on likelihood. For example, Cronin et al. created a land suitability approach for a range of future climate and land-use conditions under which the suitability for energy crops could change.^[166] They applied five general circulation models (GCMs) driven by two GHG Representative Concentration Pathways - RCP 2.6 and RCP 8.5 - representing a low and high climate change scenario respectively. The models were also driven by two pathways of socio-economic development, one assuming medium population growth, urbanization and land-use for food agriculture (SSP2) and the other assuming high levels of fossil-fuel driven development, high population and GDP growth, and food consumption with a high share of meat and waste generation (SSP5). Results suggest that the area of marginally suitable land has increased globally but the area of optimal land has decreased, with very different impacts between northern and southern latitudes as described below.

Considering the variety of pathways and models considered in this report, climate change would result in North America and Northern Asia (including China) increasing their global share of suitable land area suitable for biofuel production from 17 to 26-35%, while a decrease from 58% to 39-43% is projected for Sub-Saharan Africa, Brazil, Australia, and Southeast Asia. The same authors also project the suitability for different energy sources (wood, grass, oil, sugar/starch crops) under different climate change scenarios, revealing that the largest increase in crop suitability is expected to occur by the end of the century for grass, sugar/starch, and oil crops in the northernmost regions and under the strong climate change scenario (SSP5-RCP8.5), while the highest absolute losses are projected for all crops in Central Africa and Brazil.

FIGURE 3.3. Maps of percent changes in land suitability classes (marginal-top; moderate-centre; high-bottom) across countries for 2040-2069 vs. 1980-2009. The values are averaged among scenarios SSP2-RCP2.6, SSP5- RCP2.6 and SSP5-RCP8.5 considering land availability restrictions (i.e. excluding - from land availability - urban, forest, protected and food agricultural lands). (Source: author's elaboration from Cronin et al. 2020).

Using an ecological niche model driven by temperature and precipitation variables aggregated at the monthly, seasonal and annual level, Hu (2017) found even more concerning results for Jatropha (Jatropha curcas), a biodiesel) feedstock which is commonly grown without irrigation in subtropical regions.^[167] Hu's results project, under the same RCPs assumptions as above but using a only one GCM, an overall reduction in land suitability by more than 35% and 45% under RCP2.6 and RCP8.5, respectively. Under the same RCPs but using five climate models, Jatropha was also included among the nine crops analyzed by Yan et al. (2021) for China with a multi-factor analysis approach.

In each climate change scenario, marginal suitable land increases for seven out of nine crops in the medium-term (2050-2059), - including Jatropha. This confirms the findings of Hu (2017) over southern China. The potential production of these crops is projected to reach just one fourth of the current values due to both climate change and the poor yield that results from using marginally suitable lands to grow energy crops.

Jaime et al (2018) conducted species distribution modelling under five GCMs, driven by both RCP8.5 and RCP4.5 (high and intermediate climate scenarios respectively) to compare two oilseed crops - Brassica napus and Sinapis alba - in their suitability to the Mediterranean basin and other western European countries.^[168] Due to decreased resilience of B. napus under the arid conditions expected for the area, the study confirmed S. alba a good alternative bioenergy crop better preadapted to future climatic conditions.

By using a vegetation model, Gernaat et al. (2021) analyzed likely modifications in the potential for bioenergy for the end of this century (2070-2100) with respect to the reference period (1970-2000). They found contrasting results when CO₂ increase is considered or not in the model to account for CO₂ fertilization effects, in addition to changes in climate variables, confirming that CO₂ fertilization effects are an important source of uncertainty.^[169]

The above-mentioned findings suggest that the possible effect of climate change and variability on energy crops must not be neglected for robust planning and investments in biofuel pathways' development. Although future projects inherently maintain some level of uncertainty, this can be addressed by adopting the likelihood and confidence approach as used by IPCC.^[170]

3.7 Social Impacts and Controversies of Biofuel Expansion

Jampel Dell'Angelo, Vrije Univeristeit Amsterdam

Biofuel expansion through large-scale agricultural land investments

Biofuel production expansion plays a fundamental role in shaping the direction of global agrarian development. Supported by environmental narratives on decarbonization policy and driven by financial incentives and financial returns on investment, the recent expansion of biofuels in Sub-Saharan Africa, Latin America, South-East Asia and Eastern Europe, has been tightly associated with the phenomenon of large-scale land investments and a driving force of the redefinition of the agrarian landscape in these countries.^[171] Over 90 million hectares of arable land, approximately the size of Pakistan, have been acquired by foreign investors in the last 20 years.^[172] In many instances, the land that is being appropriated through these investments is transformed from small-scale, semi- subsistence, traditional farming, to large-scale industrialized commercial agriculture.^[173]

The extent of this phenomenon, in terms of both the dimension of land property reconfigurations, and also in terms of the socio-political impacts produced, leads scholars to describe it as a new 'global land rush'[6], evoking images of neo-colonial dispossession.^[174] A variety of scholars have pointed at the development of the biofuels industry as one of the key drivers of this process.^[175]

In addition to the environmental trade-offs that have been exposed in the previous parts of this chapter, agrarian scholars have highlighted a number of concerning socio-political transformations that can be associated with this agricultural transition, with biofuel expansion playing a driving role.

Commodification of agriculture. Biofuels are produced following the logic of commercial opportunity and profit. Large investments seek financial returns that are competitive on international commodity markets. This commoditization and commodification of agriculture is particularly accentuated by the logic of 'flex crops', a term which describes investment in a product that can satisfy different types of financial and economic demands. The system allows for the same type of crop to be sold for bioenergy purposes, for feed, and for food. The finality of agricultural production in this context is completely driven by commercial and financial aims.^[176] Agricultural values that have traditionally been associated with the land, emic values, and the perspective on land as a transgenerational family asset are lost and substituted by the logic of land as a mere factor of production which needs to be exploited for commercial purposes.^[177] There are a variety of ethical, cultural, and anthropological implications that result from this type of transition.

Transformation of systems of agricultural production. Biofuel production is associated with a process of industrialization and intensification that drastically changes the modes of agricultural production. Traditional systems of rural production, in many areas where biofuels are being developed through large-scale land acquisitions, are being fundamentally transformed. Small-scale farming, semi-subsistence agriculture, and pastoralist systems are affected by the imposition of new agricultural models which entail a multidimensional radical transformation of the agrarian landscape. The transformation of these systems entails a change in the way land is cultivated, the types of fertilizers, pesticides, nutrients, tillage, rotation, and crop diversity, very often moving from organic to fossil-based agriculture. Moreover, traditional ecological knowledge and cultural practices are affected as well as drastic reconfiguration of property rights, land tenure, and access to land resources.^[178]

Impact on labor and working relations. The impact of the development of commercial agriculture for biofuels can be understood by looking at the labor market effects of large-scale agricultural investments. A common narrative supporting these types of agricultural investments is that they favor the development of new employment opportunities. Nevertheless the extent to which these types of investments provide employment benefits for local communities is very strongly debated. A recent global empirical study addressing the question of whether large-scale agricultural investments create or destroy employment found that these investments massively crowd out smallholder farmers. This effect is mitigated to a very small extent by the cultivation of labor-intensive crops and contract farming schemes.^[179]

Reconfiguration of property relations and access. Large-scale land acquisitions (LSLAs) have a direct impact on land access and they fundamentally alter property rights and land tenure in the targeted areas. Most contemporary large-scale land investments in Sub-Saharan Africa, Latin America, Southeast Asia and Eastern Europe are implemented through land concessions that attribute exclusive access and use rights to investors.^[180] In many instances, this reconfiguration of land titles is directly developed and enforced by national governments and a large body of literature has pointed at dynamics of dispossession, eviction, and coercive imposition of new institutional arrangements.^[181] A common narrative in the reallocation of land titles for commercial agricultural development is the one of "idle", "empty", or "unused" lands. It has been denounced as a developmentalist narrative that pushes forward a system of legalized dispossession at the expense of traditional communities, such as the pastoralist and other rural communities that directly rely on natural resources, the so-called 'marginal lands'. This process is also happening with clear negative implications for the ecosystems.^[182]

Dispossession of commons. The frontier of expansion of large-scale land acquisitions for industrial agriculture is being pushed in territories with traditional land use and institutional arrangements. In many of these instances it has been reported that land governed through customary common property systems is appropriated and privatized by land investors.^[183] LSLAs are happening at the expense of the commons which represent a fundamental system for the subsistence and maintenance of social norms and traditions of small-scale farmers, indigenous people, pastoralists, and other rural groups. Often governed in a sustainable way, the commons exhibit a variety of positive socio-ecological features. Their dispossession affects rural communities in multiple ways including productive security, food security, and employment.^[184] There is strong evidence that land investors are specifically targeting common systems as often governments step in to favour land investments and grant concessions on lands that de facto are managed through traditional common property systems. The users of the commons are often evicted and a meta-study of the literature demonstrated that this is happening with high levels of coercion and violence.^[185]

Violence, coercion and repression. It has been denounced that the dynamics of commons grabbing are inherently characterized by violence, power imbalance, and coercion [ Dell'Angelo et al,. 2017]. A recent synthesis paper added novel information and characterizations about the ways in which coercion and violence manifest more frequently in these types of land claim confrontations. What emerges is an oppressive dynamic that begins with the violation of communities and collective interests, and leads to collective reactions that eventually are suppressed through coercion and violence. There are several studies that show that, when communities organize to oppose these types of agricultural land acquisitions they face conditions of oppression and violence. Repression, displacement, violent targeting, criminalization, and assassination of activists are more common than other non-violent social outcomes such as legislative and institutional changes.^[186]

3.8 Life Cycle Analysis

Joaquim E. A. Seabra, Universidade Estadual de Campinas

The growing societal concern with sustainability requires appropriate tools to inform decision-making. In this regard, life cycle assessment (LCA) methods have been increasingly used in the private and public sectors to provide a conceptual basis for identifying and understanding the impacts associated with a given process or product, from the extraction of raw materials up to final disposal and recovery.

A traditional LCA study addresses the environmental aspects and their potential impacts throughout a product's life cycle. The comprehensive scope of LCA aims to avoid shifting problems, for example, from one phase of the life cycle to another, from one region to another, or from one environmental problem to another.^[187]

Different from the other renewable technologies, bioenergy is a part of the terrestrial carbon cycle. The CO₂ emitted due to biofuels use was earlier sequestered from the atmosphere and will be sequestered again if the bioenergy system is managed sustainably, although emissions and sequestration are not necessarily in temporal balance with each other (e.g., due to long rotation periods of forest stands).^[188] Therefore, opposed to the typical case of fossil fuels, the net contributions to the biofuels life cycle emissions are not associated with their final use, but with non-CO₂ GHG and fossil CO₂ emissions from auxiliary energy use in the supply chain, as well as carbon from land use change (LUC).

Even though LCA studies usually employ methodologies in line with ISO 14040:2006 and 14044:2006 standards, there is no single method for conducting an LCA. Examples of key issues related to the evaluation of biofuels are product system definition (including spatial and dynamic boundaries) and the method for considering energy and material flows across system boundaries.^[189] Furthermore, many processes create multiple products, which result in significant data and methodological challenges because environmental effects can be distributed over several decades and in different geographical locations.^[190]

Figure 3.4 illustrates the ranges of life cycle GHG emissions for biofuels and their fossil alternatives per unit energy output. Given the wide variation in cultivation conditions as well as methodological differences between studies, estimates of life cycle emissions for the same bioenergy options vary over a wide range, even for the same temporal and spatial considerations. A broader comparison of biofuels options is shown in Figure 3.4 for the specific context of the EU's Renewable Energy Directive.^[191]

FIGURE 3.4. Ranges of life cycle GHG emissions of petroleum fuels, first-generation biofuels and selected next- generation lignocellulosic biofuels without considering land use change (Edenhofer, et al., Renewable Energy Sources and Climate Change Mitigation).

FIGURE 3.5 EU's Renewable Energy Directive estimations of typical life cycle GHG emissions of (a) biofuels and (b) future biofuels that were not on the market or were only on the market in negligible quantities in 2016 (European Union, Directive (EU) 2008/2001).

Of particular relevance for the biofuels life cycle performance are the LUC effects, the nitrous oxide (N₂O) emissions, the methods used for handling the co-products, the process efficiency, and the fuel used in the biomass conversion step. The hydrogen supply is another special point of concern for the cases involving hydrotreatment, such as in several pathways dedicated to the production of aviation biofuels.^[192] Table 3.7 illustrates the GHG emissions breakdown for four mature, commercial biofuels. Usually, feedstock production dominates life cycle emissions, but process fuel can drastically reduce the climate benefit of biofuels. For example, Wang, Wu, and Huo showed that GHG emissions for US corn ethanol can vary significantly -- from a 3% increase if coal is the process fuel to a 52% reduction if wood chips are used.^[193] Brazilian sugarcane ethanol plants, in turn, use bagasse as process fuel to meet their own energy demand, and modern mill configurations can actually export large quantities of surplus electricity to the grid.^[194]

TABLE 3.7 Breakdown of GHG emissions per life cycle stage for four commercial biofuels (gCO₂eq/MJ) (Souza et al., Bioenergy & Sustainability).

As discussed in Chapter 3, LUC has been the most contentious issue in the evaluation of GHG effects of biofuels, which can lead to significant reduction or increase of carbon stocks in biomass and soil. Even though there are large uncertainties about the overall carbon stock changes, the dLUC effects can be measured and observed with time, while iLUC implications result from projections using economics models, which are only able to capture both effects (dLUC and iLUC) together.

The more recent studies of iLUC report a lower effect than the earlier studies (Table 3.8). For example, estimates for new land brought into cultivation by the expansion of corn ethanol have been reduced by an order of magnitude and by threefold for sugarcane ethanol. These exercises therefore indicate that the land use sectors are able to accommodate a significant part of the projected bioenergy expansion without claiming new land.^[195]

TABLE 3.8 Summary of iLUC factors (Souza et al., Bioenergy Sustainability).

Agricultural intensification, and particularly double cropping, has been suggested as a practical strategy to reconcile biofuel feedstock production with other land-use priorities. Moreira et al. assessed the case of corn ethanol production under representative conditions of the current practice in the west central region of Brazil: corn grown as a second crop with soybean on land that formerly grew a single soybean crop, and energy processed from a combined heat and power plant using plantation-grown eucalyptus chips.^[196] They found that although indirect conversion of natural vegetation is identified, this effect is more than counterbalanced (in terms of GHG emissions) by the expansion of planted forests and a smaller expansion of soybean area on pastures.

Negative LUC emissions have also been anticipated for cellulosic biofuels; Field et al., for instance, showed that on land transitioning out of crops or pasture, switchgrass cultivation for cellulosic ethanol production has per- hectare mitigation potential comparable to reforestation and several fold greater than grassland restoration.^[197] Relevant impacts may as well be expected from alterations in crop cultivation management. For sugarcane in Brazil, studies have indicated the trend for carbon sequestration under unburned cane management, though it is conditioned by several factors.^[198]

Furthermore, bioenergy with carbon capture and sequestration (BECCS) technologies can be an important option for improving life cycle emissions of biofuels. When applied to ethanol plants, the process would be converted into a net carbon absorber, since CO₂ emissions in the biorefinery largely exceed the amount of GHG emissions of ethanol life cycle. For 1G ethanol, Chagas et al. estimated that capturing CO₂ from fermentation would reduce life cycle emissions to -8.8 g CO₂eq/MJ, while for 1G+2G (i.e. sugarcane + bagasse/straw) integrated plants, emissions would drop to -17,2 g CO₂eq/MJ.^[199] For US corn ethanol, a large potential exists. Sanchez et al. found that 216 existing US biorefineries emit 45 Mt CO₂ annually from fermentation, of which 60% could be captured and compressed for pipeline transport for under $25/t CO₂.^[200]

Although the capacity of mitigating GHG emissions is a critical element for biofuels, other environmental aspects can also play important roles in a biofuel-fossil fuel trade-off analysis. Usually, biofuels perform better in terms of global impacts, but biomass cultivation may lead to some higher regional impact emissions, whereas the advantages of waste-based biofuels are naturally more clear.^[201] In a direct comparison between gasoline, ethanol and blends, Luo, van der Voet, and Huppes concluded that in terms of abiotic depletion, GHG emissions, ozone layer depletion, and photochemical oxidation, ethanol fuels are better options than gasoline, whereas gasoline performs better when it comes to human toxicity, ecotoxicity, acidification, and eutrophication.^[202] However, it is important to remark that technological progress in biomass cultivation and processing is expected to bring significant improvements for first generation biofuels.^[203]

3.9 Biofuels Sustainability Certification

Despite the large appeal of bioenergy as a strategy to mitigate emissions, reduce dependency on fossil fuels, and spur economic development in rural areas, various concerns regarding bioenergy sustainability have been raised. As bioenergy policies emerged in the mid-2000s, environmental groups pressured governments to ensure that mandates produced environmental and social gains over the business-as-usual baseline.^[204] As a consequence, policy makers decided to implement sustainability initiatives that set conditions for commercializing liquid biofuels in the most important consumer markets.

These sustainability initiatives can be classified as: (1) technical regulations, technical standards or conformity assessment procedures, according to their scope; (2) public, private or mixed, considering their nature; (3) voluntary or mandatory, according to the flexibility; and (4) aiming at guidance, verification or certification, considering the purpose.^[205]

Examples of technical regulations are the Renewable Energy Directive of European Union (EU-RED), the Renewable Fuel Standard (RFS-2) in the United States, and the Californian Low Carbon Fuel Standard (LCFS). The Global Bioenergy Partnership (GBEP) and the standard ISO 13065 -- Sustainability Criteria for Bioenergy -- are examples of technical standards that aim at informing governments, producers, and consumers about how the production has occurred, its impacts, and what is important in order to verify a product's sustainability. In both cases the adoption of these technical standards is voluntary.^[206]

Conformity assessment procedures aim at verification or certification, and these schemes are mostly private and always voluntary. Examples are the certification schemes recognized by the European Commission, in the context of the EU-RED, such as the RSB standard (Roundtable on Sustainable Biomaterials), the Bonsucro Production Standard, and the ISCC (International Sustainability & Carbon Certification) standard.^[207]

The understanding is that the sustainability initiatives cover, somehow, the general concerns regarding bioenergy production on a large scale.^[208] Compared with more general agricultural certification systems, alternative fuel specific standards are particularly required to address GHG emissions because of the regulatory requirements for life cycle emissions mitigation in comparison to their petroleum counterparts. Additionally, other principles frequently shared among the certification schemes for biomass, alternative fuels and bioenergy include:^[209]

• Sustainable production: Raw materials for biofuels may not come from land that has been converted (e.g. primary forest, protected area, highly biodiverse grassland, areas with high stocks of carbon, or peatlands) and must come from legal sources.

• Other environmental impacts: The production, conversion, and logistics may not lead to negative impacts on soil, water, and air quality.

• Efficient energy conversion: Bioenergy chains should strive for maximum energy efficiency in feedstock production, conversion, and logistics.

• Protection of biodiversity: The production of biomass should not negatively affect biodiversity.

• Contribute to local prosperity and welfare: Bioenergy chains should contribute towards social well-being for employees and local population.

As an example, Table 3.9 summarizes the environmental aspects addressed by EU-RED, GBEP and ISO 13065. In the case of EU-RED and ISO 13065, all aspects should be accomplished by the economic operator (e.g., a biofuel producer), while in the case of GBEP the aspects/indicators are guidance for assessing the impacts of bioenergy production at regional or national level.

TABLE 3.9. Environmental aspects addressed by three selected sustainability initiatives^[210]

As for the social aspects, it must be noted that social principles and criteria required in the sustainability initiatives cannot go beyond what is established by the United Nations Declaration on Human Rights and by the ILO Conventions (International Labor Organization), provided the country has ratified them.^[211] The cases of CBEP and ISO 13065 are presented in Table 3.10. These aspects do not perfectly match, as they are applicable to different contexts (regional or national, in the case of GBEP, and at the operator level in the case of ISO 13065).

TABLE 3.10. Social aspects addressed by GBEP and ISO 13065^[212]

Economic sustainability aspects are not mentioned in EU-RED, while ISO 13065 has only two economic criteria that address the economic and financial feasibility of bioenergy production and trade, besides financial risk management. On the other hand, due to the motivation of assessing impacts at the regional or even national level, GBEP has an extensive list of economic indicators, although some of the indicators are not strictly economic.

Despite the overwhelming proliferation of different standards and certification approaches, there is still no global definition of how sustainability as a concept should be translated into practice, i.e. how to measure sustainability and which criteria and indicators should be included.^[213] Yet, the sustainability initiatives can still be an important tool for the promotion of more sustainable biofuels, even though the compliance with certification schemes does not necessarily translate into a sustainable production. Risks of greenwashing exist, while further investigation is needed to gauge the implications on trade, new producers, and ultimately the effective promotion of sustainable development. To that end, research and development, good governance (helped by appropriate certification schemes), and innovative business models will be essential to address knowledge gaps and foster innovation across the value chain.^[214]