Bottomup Methodology To Assess Bioenergy Potentials

Food For Freedom

Ideas for Surviving Food Shortages

Get Instant Access

The methodology used to make bottom-up estimates of the technical production potential for bioenergy includes various databases, the most important are:

2 This research is part of the FairBiotrade project which is funded by the Dutch electricity company Essent N.V. and NOVEM (Netherlands Organisation for Energy and the Environment).

population;

per capita food consumption;

per capita use of biomaterials;

land use patterns;

food crop yields;

natural forest growth;

animal feeding patterns;

bioenergy crop yields; and feed conversion efficiency in animal production.

Historic changes will be derived from existing databases and studies. Future trends are analysed by means of scenario analysis, which allows examination of the impact that various parameters have. The various parameters are grouped and correlations are included in a spreadsheet tool summarised in Figure 9.1.

Demand for food

Consumption per capita

• crop products

• animal products

Crop yields

Yields and areas suitable for crop production determined by level of advancement of agricultural technology

Population growth

Population growth

Production efficiency in the animal production system Animal production system (landess to pastoral)

Feed conversion efficiency

Surplus or shortage cropland land

Yields bioenergy crops determined by level of technology

Demand for wood

Demand fuelwood

Fuelwood from non-forest sources

Fuelwood from forests

Supply of bioenergy

Bioenergy from surplus cropland Surplus wood Residues

Residues & scavenging

Grass & fodder

Pasture

Supply of wood

Forests available for wood production (no deforestation, excl. protected areas)

Demand fuelwood

Fuelwood from forests

Demand

Industrial

industrial

roundwood

roundwood

from forests

Bioenergy from surplus cropland Surplus wood Residues

Forests available for wood production (no deforestation, excl. protected areas)

Figure 9.1 Overview of the key elements in the assessment of the bioenergy potential from specialised bioenergy crops

The model portrayed in Figure 9.1 can be divided into five sections relative to important determinants of bioenergy potential:

Demand for food. The demand for food is modelled as a function of population growth and income growth.

• Crop yields and land use. The available resources, the level of advancement of agricultural technology and the spatial distribution (optimisation) of production determine the area cropland required for the production of food crops and feed crops. Forest areas are excluded from this analysis.

• Feed use efficiency in the animal production system. The production efficiency is determined by the type of animal in question, the production system (pastoral/grazing vs. industrialised stall-fed/landless production) and the feed composition.

• Demand for biomaterials. The demand for wood is the sum of the demand for industrial roundwood and fuelwood.

• Supply of wood. The supply of wood is determined by the forest and plantation area and the growth rate.

In addition the potential for bioenergy can be aggregated into three categories:

• Bioenergy from surplus agricultural land. This potential is determined by the yield of bioenergy crops and the surplus areas of cropland (food production is given priority above bioenergy production).

• Bioenergy from agricultural and wood processing industry residues. The supply of residues is based on the production and processing volumes multiplied by conversion efficiencies.

• Bioenergy from surplus forest growth. The supply of bioenergy from natural forest growth is limited to the surplus forest growth (the use of industrial roundwood or wood used as traditional fuel is given priority above the use as a source for bioenergy).

For each of these factors scenarios are included that capture the uncertainty related to land use patterns and food production. The results are aggregated into

11 world regions, but results can also be generated on a national or sub-national level when sufficient data are available. This methodology is also used as a basis for the assessment of the sub-national economic potential for bioenergy. The technical potential is translated into economic potential by estimating the production costs based on the level of (advancement of) technology applied to produce the bioenergy. Figure 9.2 shows an overview of the procedure used to calculate the production costs of bioenergy.

This methodology requires detailed data on the costs of various production factors such as pesticides, fertilizers, labour, fuels, land, and insurance. Sufficient data have to be available to implement this approach in a region. Preliminary results indicate that there can be a large variation in production costs on a subnational level for energy crop production, which emphasises the need for accurate and detailed data.

Population growth

Population growth is an important cause of increased demand for food. In this study, population growth and changes in the capita consumption are analysed separately. Generally, population projections have been found to be fairly accurate for 5 to 10 years (Heilig, 1996), but long-term population projections are less reliable. The population projections of the United Nations Population Division (UNPD) reflect this uncertainty by encompassing six different scenarios based on mortality, fertility and migration rates. These projections are available at a national level and contain projections for rural and urban segments. The medium population growth scenario is the most likely scenarios and regional cases of it are shown in Figure 9.3.

Examples Bio Energy
Figure 9.2 Data requirement for cost analysis for production system A (in this example) on medium quality land

-1-Japan

-Western Europe

—*—C.I.S. and Baltic States —■— sub-Saharan Africa —A—Carribean & Latin America —X— Near East & North Africa ■ ■■%■■■ East Asia South Asia

Figure 9.3 Population growth, 1960-2050 (1,000 heads) Source: UNPD (2003).

The speed of growth generally decreases in the future compared to recent decades, but the absolute number of people continues to grow rising from 6.0

billion in 2000 to 8.9 billion in 2050 under the medium scenario. In the low and high scenarios the population increases to 7.3 and 10.5 billion people in 2050, respectively. According to the UNPD projections the difference between the high and low population scenarios are largest in developing countries, reflecting the present high fertility rates and future uncertainty concerning their decrease. The strongest population growth is projected for the developing regions, e.g. sub-Saharan Africa +138%, South Asia +65%, Caribbean & Latin America +48% and East Asia +21%. The population in Oceania and North America is also projected to increase significantly (+30% and +42% respectively). Regions with a decreasing population are Western Europe, Japan, Eastern Europe, C.I.S. and the Baltic States (-2%, -13%, -17%, -17% respectively).

Uncertainty related to population projections has increased during recent years, with population projections developed in the last decade being frequently downscaled. E.g. the medium projection of the global population is 1.1 billion lower than projected in 1990, mainly stemming from projected levels of HIV infections.

Food consumption

The main driver behind increasing or changing per capita food consumption is an increase in income (expressed in purchasing power). The methodology for estimating future consumption patterns is based on supply-demand equilibrium, considering the impact of the various underlying variables (e.g. agricultural policies and payments, Gross Domestic Product (GDP), population growth, technological development, cultural preferences etc.), their evolution over time and the correlations between these factors. Such an exercise is problematic, due to methodological problems related to the calculation and use of elasticities that describe correlation between for example GDP and consumption or food supply and prices and other parameters. In addition, data on the capacity of the natural resource base of the food production system to support an increasing food production level are often uncertain and insufficient and a detailed understanding of many of the underlying biological and physiological processes is not available. As a result, a considerable amount of expert judgement is involved in estimating future consumption patterns, particularly in the long term.

This uncertainty results in a considerable range of outcomes. They range from forecasts of a global food crisis to more mainstream (and optimistic) projections such as those of the Food and Agriculture Organisation (FAO) of the United Nations. The consensus is that consumption is likely to increase, although at a slower pace than in the past and that under-nourishment is likely to decrease.

FAO projections are used in this study, since these have the longest time horizon (to 2030) and may be regarded as the most widely used and reliable source of projections. The FAO projections are based on a combination of supply demand modelling based on a model developed by the International Food Policy Research Institute (IFPRI) in combination with iterative rounds of adjustments involving expert judgment of FAO consultants. The FAO projections are trend extrapolated to 2050 based on data from the IFPRI, the National Institute of

Public Health and the Environment (RIVM) and our own assumptions. Figure 9.4 and Figure 9.5 show the projected increase in consumption in various regions.

—A—Carribean & Latin America —*— Near East & North Africa ■■■%■■■ East Asia ...%... South Asia -World

1960

1980

2000

2020

2040

1960

- North America

-Oceania

-Japan

-Western Europe

1980

2000

2020

2040

2060

—A—Carribean & Latin America —*— Near East & North Africa ■■■%■■■ East Asia ...%... South Asia -World

Figure 9.4 Historic and projected per capita total food intake, 1961-2050 (kcal/capita/day)

Source: IFPRI (2001a); IMAGE-team (2001); FAO (2003a, b); own calculations.

—■— sub-Saharan Africa —A— Carribean & Latin America —X— Near East & North Africa East Asia South Asia -World

1960

1980

2000

2020

2040

■ North America

- Western Europe

1960

1980

2000

2020

2040

2060

—■— sub-Saharan Africa —A— Carribean & Latin America —X— Near East & North Africa East Asia South Asia -World

Figure 9.5 Consumption of animal products 1961-2050 (% of total daily caloric intake)

Source: IFPRI (2001a); IMAGE-team (2001); FAO (2003a, b); own calculations.

Figures 9.4 and 9.5 indicate that daily kcal intake in the industrialised regions is approaching an equilibrium level. The consumption share accounted for by animal products has decreased during recent decades and this decrease is

3 RIVM data are derived from the Integrated Model to Assess the Global Environment (IMAGE). The IMAGE model is a dynamic integrated assessment modeling framework for global change.

projected to continue. In the transition economies, the collapse of communism and following economic restructuring caused a strong decrease in GDP, agricultural payments and consumption. It will take several years to decades before consumption and production have reached the level of the communistic era. In developing countries, food consumption is projected to increase, particularly in the economic booming region of East Asia, although consumption levels remain below saturation levels in 2050 in all developing regions. The regions with the highest shares of under nourishment are presently sub-Saharan Africa and South Asia. In the latter, relatively high economic growth is projected to reduce poverty and under nourishment, while in sub-Saharan Africa, strong population growth and poor economic performance limits the increase in consumption, particularly the consumption of animal products. Consequently, the relative incidence of under nourishment in the developing countries is likely to decline from 17% in 1997/99 to 6% in 2030, with 776 and 443 million malnourished, the bulk being in sub-Saharan Africa and South Asia. We acknowledge that food production should be prevented above bioenergy crop production, but in reality food shortages are often the result of armed conflicts, rather than a lack of suitable cropland.

The FAO and IFPRI projections could be the best available, but forecast errors for food consumption and production at the regional level in the range of +/-10 to 40% are common, with errors as large as 90% occurring in the past (IFPRI, 2001b). Globally aggregated data show much smaller projection errors. Since consumption levels in the developing regions are below saturation levels, consumption in those regions is very responsive to further increases in income or decreases in food prices compared to the scenario underlying the projections included in this study. A small change in GDP or prices may significantly increase consumption in these regions. Consumption in regions with consumption levels near the saturation level is likely less sensitive to changes in prices and GDP. This means that the projected increase in consumption is more uncertain for countries with low levels of consumption.

Wood consumption and production

During the 1990s several outlook studies and reviews investigating fuelwood and industrial roundwood consumption have been published. Despite public attention for concerns about deforestation, particularly with respect to rainforests, the data on such activities is relatively weak. Particularly data on illegal cutting and the use and production of fuelwood are largely based on estimates that may provide too little information for the production of reliable forecasts (EFI, 1996). In addition, the methodological problems encountered are similar to those of encountered when projecting future food consumption and land use patterns.

Because of a high degree of uncertainty related to wood consumption and production, only a few projections go beyond 2010. Most that do, only give data on total consumption of roundwood with a regional subdivision limited to industrialised and developing countries. The projections are difficult to compare due to a lack of information on key assumptions and methodologies applied. Also not all studies are intended to produce equivalent results, but also to mimic the effects of various factors (EFI, 1996). The range of projections for the demand for industrial roundwood, fuelwood, plantation production and natural forest growth are translated into a set of demand and supply scenarios (Table 9.1). Since no supply-demand matching is included, the combination of low consumption and high supply results in the highest potential for bioenergy, but also ignores many interactive issues.

Table 9.1 Demand and supply scenarios for wood in 1998 and 2050 (million m3)

2050 2050 2050 1998 low medium high

Demand

Industrial roundwood 1,672

Fuelwood 1,807

Supply

Industrial plantations 330

Non-industrial plantations 85

Total natural forest growth 9,402

Source: various.

The consumption of industrial roundwood is projected to increase from ca. 1.7 billion m3 to 1.9 to 3.0 billion m3 in 2050, although in the most extreme scenario found in the literature the consumption is projected to increase to some 7 billion m3 in 2050.

Despite many uncertainties and conflicting trends there seems to be general agreement that the demand for fuelwood is not going to change rapidly (EFI, 1996; FAO, 2003b). Increasing income and urbanisation encourage a switch from fuelwood to more modern commercial fuels (gas, oil) while rapid population growth in many developing regions and increasing (but still low) income levels on behalf of the majority of the mainly rural, fuelwood consumers counteract this effect. Data from existing studies indicate a consumption of fuelwood between 1.7 in 2050 and 2.5 billion m3 in 2020, compared to the present ca. 1.8 billion m3. The upper range is based on a constant per capita consumption and results in a consumption of 2.6 billion m3 in 2050.

Fuelwood and industrial roundwood stem from very different sources and production systems, ranging from well-managed plantations to full deforestation of virgin forests or gathering of twigs for use as fuelwood. In this study we distinguish plantations and natural forests.

The contribution of plantations to the global supply is significant and increasing. According to a study on future wood production from plantations, the production from plantations may increase to 0.8 to 2.0 billion m3 in 2050. The theoretical production from natural forests is estimated based on forest area data and data on gross annual increment (GAI). Note that the data on GAI are

1,900 1,700

609 173 9,402

2,500 2,200

863 245 9,402

3,100 2,600

1,488 479 9,402

considered very uncertain. We also assume 10% of the forest area is set-aside for biodiversity protection and nature conservation. In turn, annual forest growth is constant at 9.4 billion m3, assuming no deforestation. The total surplus forest growth is estimated at 72 EJ/yr, maximum. However, most of the production is classified as unavailable4 or consists of non-commercial species, for which there is presently no market due to poor quality or characteristics of the species. In addition, roughly half of the global forest area is old-growth undisturbed forest. For reasons of nature protection, these areas may be excluded from supply. More detailed analysis show that if all three limiting factors are included, the wood demand in 2050 cannot be met and the potential for bioenergy is zero.

In reality, any gap between demand and supply is closed, since there is a general agreement that 'the technological global wood production capacity is sufficiently large to fulfil the largest projected increases in demand' (EFI, 1996); in line with the supply and demand situation shown in Table 9.1. Further, standing stocks may serve as a buffer to reduce the effect of regional or temporary market fluctuations. The volume in standing stocks is more than 120 times the current total wood consumption.

It is not known to which extent the three scenarios include the effects of technological improvements. Recent studies indicate that both through increasing conversion efficiencies and the development of new wood products which make more efficient use of resources (e.g. medium density fibre board), the growth of demand for industrial roundwood will slow down (FAO, 2003b). Energy efficiency improvements (e.g. improved stoves or the use of modern bioenergy carriers such as liquid fuels) on the other hand have (in theory) the potential to more than offset increasing demand up to 2030.

Agricultural land use and agricultural management

In this analysis, the production of bioenergy from specialised crops is limited to surplus land or land not suitable for agriculture. The mainstream studies on agricultural land use project an increase in yields and an increase in the area under crop production in the developing regions during the coming decades, partially at the expense of forests. Globally, the arable land area is projected to increase 13% until 2030 (FAO, 2003b). Cropland area in the transition and industrialised regions is expected to increase marginally, if not remain stable or decrease, though the FAO states that a potential decline could be partially offset by emerging trends towards de-intensification and the increasing demand for ecologically produced crops (without or with minimum use of fertilizers and chemicals). Pastures are not included in the FAO calculations, although globally pastureland area is likely to decrease due to increasing mixed farming, improved pastures and stall-fed systems, and demand for animal products in the developing countries.

4 Unavailable areas are defined as:

• Physically inaccessible areas due to factors such as steepness of terrain.

• Areas far from industrial sites due to transportation distances or lack of infrastructure.

• Areas too low in commercial volume, degraded forest or some other legitimate reason specific to each country.

The FAO projections of agricultural land use are based on the same methodology as used in determining the consumption projections. Again the projected land use changes are uncertain. Many studies indicate that there are large 'exploitable yield gaps', both with respect to crop yields and the efficiency of production of animal products. Yield advances are thus a key determinant of future bioenergy production, since the efficiency of food production determines the area of surplus cropland and pastureland available for bioenergy production. The closing of these yield- and efficiency gaps is a matter of agricultural management,5 which is the prime target of agricultural, economic policies. A newly emerging market for bioenergy production and related policies could further speed up the adoption of more efficient agricultural production systems.

The impact of yield increases is analysed by translating the total demand for food into a demand for cropland. A more sophisticated management system results in higher yields per hectare, larger areas suitable for crop production, higher production per animal, lower demand for feed and an overall lower demand for agricultural land.

For the production of crops six management levels are defined that vary with respect to the level of advancement of agricultural technology (including the level of agricultural inputs) and the use of natural rainfall and/or irrigation:

• Low, rain-fed: using no fertilizers, pesticides or improved seeds, equivalent to subsistence farming.

• Intermediate, rain-fed: average of high and low.

• High, rain-fed: full use of all required inputs and management practices as in advanced commercial farming.

• Very high, rain-fed and/or irrigated: use of high level of technology on very suitable and suitable soils, intermediate level of technology on moderately suitable areas and low level on moderately and marginally suitable areas. The rationale for this methodology is that it is unlikely to make economic sense to cultivate moderately and marginally suitable areas under the high technology level, or to cultivate marginally suitable areas under the intermediate technology level.

• Super high, rain-fed and/or irrigated: the high and very high level of agricultural technology exclude the impact of the development of technology beyond the best available technologies presently used in the industrialised regions. We consider it likely that agricultural technologies will continue to become more efficient and productive, although at a much slower pace than previously. Based on various sources, we assume that the total bio energy potential may be 25% higher than in a very high level of technology without further specifying the origin of this increase. This is referred to as the super high level of technology.

5 The term management usually refers to the use of fertilizers, pesticides, mechanised tools, improved breeds, double cropping, and the application of irrigation. In this chapter the term also includes the level of agricultural technology and the optimalisation of land use patterns to minimize land use or optimize profits.

The impact of the application of these management systems is analysed separately for crop production and the production of animal products as described in the following section.

Cropland and agricultural management

The impact of management systems on crop yields is analysed using data from a crop yield model from the International Institute of Applied Systems Analysis (IIASA) (IIASA/FAO, 2002). The crop yield model uses georeferenced data on climate, soil quality etc. and may be regarded as the state-of-the-art in crop growth modelling considering the global coverage and number of crops included. In total, data for 19 different crops are included. Since the data from the crop growth model are based on georeferenced datasets (employing Geographic Information Systems), this type of data can be made available per region, subregion, and country or per grid cell.

The data are specified for yield and area by country and follow a classification of suitability for crop growth. The classification is based on the maximum constraint free yield (MCFY): very suitable (VS, 80-100% of MCFY), suitable (S, 60-80% of MCFY), moderately suitable (MS, 40-60% of MCFY), marginally suitable (mS, 20-40% of MCFY) and not suitable (NS, <20% of the MCFY).6 No yield levels are included for areas classified as NS. A dataset that indicates the total extent of cropland not under forest cover is also provided. In addition, a set of simple allocation rules was used to determine use of suitable cropland (VS, S, MS or mS) for other purposes than crop production. Data were derived from the FAOSTAT database (FAO, 2003a) and the IIASA data (IIASA/FAO, 2002). The total global land area is 13 Gha, divided into other land (3.6 Gha), permanent pasture (3.5 Gha), built-up land (0.2 Gha), forest (4.2 Gha, divided into plantations and natural forest), permanent crops (0.1 Gha), and arable land (1.4 Gha). In this study, deforestation is not allowed, so increases in e.g. the areas of built-up land occur at the expense of the area of agricultural land in the base year 1998.

These data were integrated into a spreadsheet tool wherein the projected demand for food and feed in 2050 is translated into yield-area combinations. A given demand for crops can be produced for different combinations of yield and area; a small area with very productive land can produce the same amount of crops as a large area of low productive land. The spreadsheet includes:

• Optimisation of production 'geographically'. Geographic optimalisation includes the allocation of crops to areas with high yields first (leaving the least productive areas for bioenergy production) and a cropping intensity of 1 (defined as the ratio harve sted land to arable land).7

6 Because the classifications VS to mS are based on the percentage of maximum constraint free yield (MCFY), not the absolute level of yields, economic optimalization of production is included in this dataset. A VS yield in region 1 can be lower than a VS yield in region 2, but are equally important in the allocation procedure. Production in region 1 on VS areas is however attractive considering the relative high suitability compared with areas in that region.

7 The FAOSTAT database does not include data on total harvested land. Data can be obtained by summing up the harvested areas reported for different crops. Data are available for total arable land in

• Application of a certain level of technology.

• Regional if not global self-sufficiency. The demand for food in a region is allocated within the region. When a region is not self-sufficient (meaning that the projected demand for food and feed in a region can not be produced within that region), the remaining demand is allocated to other regions that are self-sufficient and have surplus areas of cropland.

The area of cropland required to produce the regional demand is calculated and compared with the present agricultural land. When a region exhibits decreasing demand for agricultural land, the surplus land is available for crop production. The potential increases in yields are considerable, globally between 190% and 360%. The calculated (theoretical) potential yield increases for a number of regions for scenario 1 and scenario 4 are shown in Table 9.2 (scenarios are defined below). The 1998 yield levels are set at 1; average increases are weighed averages based on harvested areas.

Table 9.2 Average increase in crop yields (1998=1)

Region

Very high rainfed level of technology (scenario 1)

Super high rainfed/irrigated level of technology (scenario 4)

North America

1.6

3.2

Oceania

2.4

4.6

West Europe

0.9

1.9

C.I.S. and Baltic States

3.2

6.7

Sub-Saharan Africa

5.6

7.7

Caribbean & Latin America

2.8

4.5

South Asia

3.7

5.6

World

2.9

4.6

Source: IIASA/FAO (2002); FAO (2003a), own calculations.

Source: IIASA/FAO (2002); FAO (2003a), own calculations.

Animal production and agricultural management

The management system applied for the production of animal products determines the future demand for various feed categories (pasture biomass, feed from crops, residues & scavenging biomass) based on Equation 9.1.

agricultural use (named 'arable land' and 'land in permanent crops' in the FAO statistics). It is not known to which extent these datasets are consistent, but the cropping intensity can be used as an indicator. Globally the area harvested is 93% of the area arable land, regional aggregated data are between 70 and 130%.

Feed = Demand x Prod x Fce x Fco where:

Demand = demand for animal products based on the consumption scenarios.

Prod = production system. Two extreme production systems are included: pastoral and landless. Combinations of the two are referred to as mixed production systems. The difference between these systems is the source of animal feed (cropland vs. pasture land) and the overallefficiency of production (feed conversion efficiency). Data are derived from the IMAGE projections (IMAGE-team, 2001).

Fce = feed conversion efficiency (total demand of biomass (dry weight = dw) per kg animal product); data are taken from the IMAGE model (IMAGE-team, 2001). The range in feed conversion efficiencies in the year 1995 is used to estimate feed conversion efficiencies in a low, intermediate and high level of technology production system. Table 9.3 gives an overview of feed conversion efficiencies for a selected number of animal products and regions.

Fco = feed composition. Data on feed composition (feed, pasture and fodder biomass, residues, scavenging) are specific for each region, type of animal product and production system. The demand for feed from crops is added up to the demand for food crops and is included in the spreadsheet tool used to calculate land use.

Table 9.3 Feed conversion efficiencies in 1995 and in a high level of technology in an animal production system in which all animal feed is derived from residues and feed crops (kg dw feed/kg product)

Region Bovine meat Pig meat Poultry meat and eggs

Table 9.3 Feed conversion efficiencies in 1995 and in a high level of technology in an animal production system in which all animal feed is derived from residues and feed crops (kg dw feed/kg product)

Region Bovine meat Pig meat Poultry meat and eggs

North America

26

6.2

3.1

Oceania

36

6.2

3.1

West Europe

24

6.2

3.1

C.I.S. and Baltic States

21

7.4

3.9

Sub-Saharan Africa

99

6.6

4.1

Caribbean & Latin America

62

6.6

4.2

South Asia

72

6.6

4.1

World

45

6.7

3.6

High level of technology

15

6.2

3.1

Source: IMAGE-team (2001); FAO (2003b); own calculations.

Source: IMAGE-team (2001); FAO (2003b); own calculations.

The potential to increase feed conversion efficiencies in the developing countries is considerable (up to a factor 7), which is mainly the result of the low feed conversion efficiencies in pastoral production systems.

The calculations included in this study allow (in theory) a comparison of the demand for various feed sources with the supply of feed sources based on natural circumstances, prices etc. The demand for feed crops is included in the land allocation procedure. The demand for residues and scavenging biomass is compared to the future production of residues. The use of feed from pastures through grazing is unknown due to a lack of data on and models mimicking the productivity of pastures under various management schemes. Therefore, the relative change in demand for pasture biomass is used as a proxy for the areas of permanent pasture as explained below.

In case of an increase in the demand for grasses and fodder is projected compared to the base year, the increase in demand for grasses and fodder compared to the base year is added up to the demand for feed from crops. The reason for this approach is that an increasing demand for feed from grazing could lead to an expansion of the area permanent pasture by deforestation or higher grazing intensities, which in turn could lead to e.g. soil erosion and other problems related to overgrazing. In case of a decreasing area of pastureland, areas of permanent pasture become available for crop production. The data and methodology described above are also used in the assessment of the economic potential, though sub-national data on livestock production efficiencies and feed sources are generally not available.

Bioenergy yields

Data on bioenergy yields can be determined based on crop growth models or derived from field experiments. In this study, we use yield data for short rotation woody bioenergy crops, because there is extensive experience with woody bioenergy for fibre production for the pulp and paper industry. Also woody biomass can be converted in various types of fuel (e.g. eucalyptus, poplar or willow). We use data from the IMAGE model which are derived from crop modelling (IMAGE-team, 2001). Note that higher bioenergy yields in tropical regions are possible if herbaceous crops (e.g. Miscanthus) are used (Hall et al., 1993).

The calculation of bioenergy production potential is based on the areas of surplus land calculated in the previous section multiplied by the yields of bioenergy crops taking into account the quality of these surplus areas. Figure 9.6 shows the global (modelled) yield-area curve for the production of bioenergy based on a low and high level of technology.

The curves clearly show the impact of both the suitability of the land and the impact of the production system: the area suitable for bioenergy production is higher and yields are also higher in a production system based on a high level of advancement of agricultural technology compared to a low level of technology. The surface under the graph is the total global (technical) production potential for bioenergy. For a low and high level of advancement of agricultural technology this potential is estimated at 1,807 and 4,435 EJ/yr respectively (based on a higher heating value of 19 GJ/ton dw).

800 700 600 500

300 200 100 0

Ns low level of technology high level of technology

Vs ms

Vs s

Figure 9.6 Simulated bioenergy yields (GJ/ha) based on a low and high level of advancement of agricultural technology (VS = Very Suitable Areas, S = Suitable Areas, MS = Moderately Suitable Areas, mS = Marginally Suitable Areas) Source: IMAGE-team (2001); own calculations.

Was this article helpful?

0 0

Post a comment