university of groningen the use of agricultural resources ...maría josé ibarrola rivas . colophon...

University of Groningen

The use of agricultural resources for global food supplyIbarrola Rivas, Maria José

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Ibarrola Rivas, M. J. (2015). The use of agricultural resources for global food supply: Understanding itsdynamics and regional diversity. University of Groningen.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 25-08-2021

https://research.rug.nl/en/publications/the-use-of-agricultural-resources-for-global-food-supply(31bb5a69-b6e7-48f5-9f94-d2c8469954ff).html

The use of agricultural resources for

global food supply

Understanding its dynamics and regional diversity

María José Ibarrola Rivas

Colophon

This PhD project was carried out at the Center for Energy and Environmental Sciences

(IVEM), which is part of the Energy and Sustainability Research Institute (ESRIG) of the

University of Groningen in the Netherlands. IVEM funded the participation of courses

and conferences during the project and an extension of the PhD scholarship.

The PhD scholarship was financed by the CONACYT scholarship Becarios en el

Extranjero No. 209 358 and the Erasmus Mundus ECW Lot 20 programme during four

years (2010 - 2014).

ISBN: 978-90-367-7787-2

ISBN: 978-90-367-7786-5 (electronic version)

© 2015, María José Ibarrola Rivas

Cover design and drawings by Julio Pastor---www.juliopastor.com

Printed by Grafimedia, Facilitair Bedrijf Rijksuniversiteit Groningen

The use of agricultural resources for

global food supply

Understanding its dynamics and regional diversity

PhD thesis

to obtain the degree of PhD at the

University of Groningen

on the authority of the

Rector Magnificus Prof. E. Sterken

and in accordance with

the decision by the College of Deans.

This thesis will be defended in public on

Friday 1 May 2015 at 12:45 hours

by

María José Ibarrola Rivas

born on 7 March 1983

in Mexico City, Mexico

Supervisor

Prof. H.C. Moll

Co-supervisor

Dr. S. Nonhebel

Assessment Committee

Prof. J.W. Erisman

Prof. D. Strijker

Prof. M.J. Wassen

para mi gezin

7

Content

Chapter 1. General Introduction……………………………………………………………….... 9

Chapter 2. Increasing inequality between poor and rich countries as to

availability of land and water by 2050……………………………………………………... 17

Chapter 3. Estimating future global needs for nitrogen based on regional

changes of food demands……………………………………………………………………….... 33

Chapter 4. Nitrogen fertilizer use per person and its trade-off with land

use: An international comparison of agricultural production systems and

diets………………………………………………………………………………………………………... 57

Chapter 5. Farm labour footprint of food: an international comparison

of the impact of diets and mechanization………………………………………………….. 79

Chapter 6. Future global use of resources for food: the huge impact of

regional diets…………………………………………………………………………………………. 101

Chapter 7. Identifying challenges for a sustainable future of the global food

system…………………………………………………………………………………………………... 127

References…………………………………………………………………………………………….. 147

Summary…………………………………………………………………………………………........ 157

Resumen...................................................................................................................................... 161

Samenvatting............................................................................................................................ 165

Acknowledgments.................................................................................................................. 169

About the Author..................................................................................................................... 171

Chapter 1. General Introduction

9

Chapter 1.

General Introduction

1.1 Food and agricultural resources

Every person on earth needs a daily supply of food. The production of all this

food requires resources. A piece of land to cultivate the crops, nutrients and

water to let it grow, and work to plant it and harvest it.

Living on a finite planet means that these resources are limited. Human

consumption is reaching the limits of the planet, and food production is a major

player (Rockström et al., 2009). The global use of resources for food is

enormous. Agricultural production uses 40% of the global land (FAO, 2013c)

and 70% of the anthropogenic global use of water (FAO, 2013a). The increase

of land use in the last decades has led to serious environmental impacts such as

loss of both biodiversity and many ecosystem services (Foley et al., 2005). The

increase of irrigation with underground water has led to an overuse of water

reserves and declining water tables (Foley et al., 2005).

Nitrogen is the most important limited nutrient for crop production (Engels &

Marschner, 1995). In the late 19th century, the discovery of artificial nitrogen

fertilizer production by the Haber-Bosch synthesis of ammonia increased

enormously the potential of crop yields (Smil, 2001). During the so-called

“green revolution” during the 20th century, synthetic nitrogen fertilizer was

widely spread around the globe resulting in a doubling of global food

production in just few decades (Alexandratos & Bruinsma, 2012; Bumb, 1995)

leading to large health benefits: decreasing malnutrition and improving food

security in low income countries. Though, its use has caused strong

environmental impacts (Sutton et al., 2013) such as local pollution (Eickhout et

al., 2006; Shindo et al., 2006), affecting the global nitrogen cycle (Galloway et

al., 2008; Smil, 1999) and represents an important indirect energy use in

agriculture (Woods et al., 2010). Half of the global energy use in agriculture is

related with the production of synthetic nitrogen fertilizer and the other half is

related with fuel for running machinery (Woods et al., 2010).

10

In the coming decades, global food demand will increase due to population

growth and change in dietary patterns (Godfray et al., 2010b). The strong

environmental impacts already caused by the global food production could

increase further, which may compromise the development of a more

sustainable food system (Foley et al., 2011; Fresco, 2009; Godfray et al., 2010a;

Tilman & Clark, 2014; Tilman et al., 2011). A global sustainable food system

should supply enough nutritional food for everyone with the lowest

environmental problems as possible.

1.2 Regional difference

The IPAT approach (Ehrlich & Holdren, 1971) is a methodology commonly used

in environmental sciences (Chertow, 2000). It is useful to understand the

individual impact of the main factors driving the use of agricultural resources.

This approach indicates that the environmental impact of a country (I) depends

on the size of the population (P), the affluence (A) and the technology (T). As an

analogy, the use of agricultural resources needed for food (I) depends on the

size of the population (P), on the type of diet (A) and on the type of agricultural

production system (T). These three parameters and their dynamics show

enormous variations throughout the world which are described below.

Population numbers and population growth can be strongly different among

regions. Some countries have large population in relation to their land resulting

in densely populated areas such as Bangladesh and the Netherlands in

comparison with other countries with low population density such as Botswana

and Australia. Population density is not related with socioeconomic

development. In contrast, the growth rate of a population is closely related

with socioeconomic development. The demographic transition theory

(Chesnais, 1992) states that the increase of per capita income results in a

decrease of the fertility rate, so population growth decreases with the increase

of economic development. Low income countries such as Angola, Uganda and

Zimbabwe had an annual fertility rate of 3% in 2013 while high income

countries such as Denmark, Germany and the Netherlands had an annual

fertility rate of less than 0.5% in the same year (World Bank, 2014). But also,

other demographic factors affect the fertility rate. For example, in general, the

fertility rate in rural areas is higher than in urban areas (Li & Wang, 1994).


11

Diets vary around the globe. The nutritional requirements for a healthy diet can

be obtained from a variety of food items resulting in different composition of

menus around the globe (Menzel & D'Aluisio, 2005). These differences can be

related with culture, food availability, production of local products, income

level, religion or others. For example, all people consume cereals but not the

same type. Rice is mainly consumed in Bangladesh, maize in Mexico and wheat

in Turkey (FAO, 2013d). In rural areas of Africa, people mainly consume

traditional grains such as fonio (Garí, 2001), in contrast with people in urban

areas of industrial countries who mainly consume modern grains such as

hybrid cereals. Similar differences exist with other food items.

The dynamics of the diets are different among regions. The changes in food

consumption patterns of a population are related with socioeconomic

development as well as with social factors such as culture and urbanization.

The Nutrition Transition Theory (Popkin, 1993) has been widely used in the

literature to describe dietary changes in relation to income level. This theory

states that diets change with increase of income levels as follows: starting with

an under nutrition state, first the amount of calories increases including mainly

staple food such as cereals, roots and pulses. Then, diet diversifies increasing

the consumption of fats (e.g. animal products), sugars and processed food,

leading to health problems like obesity. Lastly, diets reach a final state by

increasing the consumption of fruits, vegetables and carbohydrates, and

reducing fats, leading to health improvements. Some studies have calculated

the relation between income levels and caloric consumption of different food

categories for different countries (Gerbens-Leenes et al., 2010; Poleman &

Thomas, 1995), showing the clear increase of affluent food products

consumption (e.g. animal products) with income levels. However, some

countries have deviated from this trend due to other factors. Since the last

decades, these dietary changes in developing countries have happened in lower

levels of income due to fast urbanization, lower food prices, globalization of the

food system, fast food and retail services (Kearney, 2010). Also, cultural

differences result in different dietary patterns. For example, for a similar

socioeconomic development stage, the per capita meat consumption in Brazil

and China has been relatively higher than the world’s average; in contrast, the

meat consumption in India has been relatively lower (Alexandratos &

Bruinsma, 2012).

The type of agricultural systems of a certain population is related with the

socioeconomic development, the physical conditions of the soil, the climate, the

12

type of management practice and the population density. The type of

agricultural system can be classified in intensive and extensive systems.

Extensive systems use low amount of yield-related inputs (e.g. fertilizers,

irrigation) resulting in low crop yields, and intensive systems use large amount

of these inputs resulting in high crop yields. The intensity of the agricultural

system is closely related with economic development. Pimentel (Pimentel,

2009; Pimentel & Pimentel, 2008) have shown that production systems in low

income countries use almost no yield related inputs nor machinery resulting in

low crop yields in comparison with production systems in high income

countries which use large amounts of these inputs resulting in high crop yields.

But, the intensity of the system and the economic development do not always

show a linear relation because other factors affect the choice of the

management practice. For example, increase of population density can increase

labour productivity and intensification of agriculture (Boserup, 1965). Some

studies have shown a clear trend between the increase of population density

and the increase use of fertilizers (Arizpe et al., 2011; Smil, 2001, fig 8.8). In

addition, local physical conditions such as climate and the type of soil, influence

the productivity of the system. This leads to different crop yield potentials

throughout the world (Licker et al., 2010; Lobell et al., 2009; Neumann et al.,

2010).

This overview shows that regional differences of population numbers, diets and

agricultural systems are large, and their dynamics are driven by several factors

from different disciplines: socioeconomic development, culture, nutrition,

agronomy, climate, geography and others. It is essential to consider these

regional differences when studying global food security because food supply is

mainly regionally organized. Globally, only 10% of the countries’ food supply is

originated from food imports (FAO, 2013d). This means that food is mainly

produced and consumed in the same country.

Some studies have analysed the future sustainability of the global food system

by using global values (Godfray et al., 2010a; Tilman et al., 2011). However, the

global average of food security indicators do not show the strong regional

variations such as population dynamics, resource availability per capita,

agricultural productivity (crop yields), food demand per capita, and others. This

thesis focuses on the analysis of these regional differences relevant for the

global use of agricultural resources.


13

As mentioned before, a sustainable future of the global food system includes

producing the food with the lowest environmental impact as possible. So, it is

necessary to consider the use of all major resources at the same time because

the use of each resource results in a different environmental impact. The use of

these resources have trade-offs between them. Many studies have analysed, in

more detail than the global studies, the use of one of these resources: land

(Fader et al., 2013; Kastner et al., 2012; Ramankutty et al., 2008; White, 2007),

water (Hoekstra & Mekonnen, 2012), nitrogen (Bouwman et al., 2011; Leach et

al., 2012; Liu et al., 2010; Pierer et al., 2014; Shindo et al., 2006; Xiong et al.,

2008) and energy (Berners-Lee et al., 2012; González et al., 2011). But, by

focusing on only one resource, the trade-off with other resources is not

considered leading to biased sub-optimal conclusions. For instance, the studies

about nitrogen fertilizer use suggest a need to decrease its use. But, the strong

trade-off between nitrogen and land could, in some cases, result in stronger

environmental implications (such as deforestation, biodiversity and erosion)

than a high nitrogen use. So, in addition to consider the regional differences,

this thesis analyses the main trade-offs among these resources.

1.3 Aim and scope of the thesis

The main aim of this thesis is to assess the sustainability of the global food

supply system. The sustainability of the food system is defined as supply

enough food for everyone with the lowest environmental impact as possible.

The environmental impact is discussed with the use of the main agricultural

resources (land, water, nutrients and work) and their trade-offs. In order to

have an integrative understanding of the global food system, the thesis takes a

demand perspective: food demand is the starting point. The impact of food

demand on the use of agricultural resources is assessed by considering the

dynamics and regional diversity of population numbers, diets and agricultural

systems. An interdisciplinary analysis is needed to integrate the drivers of the

different disciplines related with food demand and food production (population

dynamics, dietary changes, agricultural systems, socioeconomic development,

culture, demography, etc). The results are used to discuss future implications

on the use of agricultural resources.

The main research questions that are addressed throughout the thesis are:

How do the main drivers of food demand impact the use of resources? What are

the relevant global differences in relation to the use of resources? What have

14

been the main trends (changes in the last decades)? What are the trade-offs

and/or feedback loops in the use of agricultural resources? What can we learn

from these findings to have an integrative assessment for the future use of

resources?

The goal of integrating regional differences, interdisciplinary indicators and

trade-offs among resources results in a methodological challenge to achieve an

integrated overview of the global use of resources. Therefore, a methodology

was developed throughout the thesis to simplify the global food system in a

way that the regional differences relevant for the use of resources can be

analysed as well as the trade-offs among resources. At the end of the thesis

(chapter 7), this methodology is described in detail as well as the consequences

for the future use of resources based on the integrative understanding obtained

throughout the thesis.

1.4 Structure and approach in each chapter

The thesis is divided into 7 chapters. Chapter 1 is the general introduction in

which a literature review shows the challenges to analyse the sustainability of

the global food system, and the general aim of the thesis is described. Chapters

2 to 6 show the results of the different sub-projects of the thesis. Throughout

these chapters, the use of the main agricultural resources is analysed: land,

water, nitrogen fertilizer and labour. In each chapter, one or two resources are

explored in detail. Each chapter has a specific approach and focuses on

different indicators and drivers to discuss the use of resources.

The regional differences in resource availability are analysed in chapter 2. This

chapter focuses on the impact of population growth on the availability of land

and water in relation to the socioeconomic development of the population. The

indicators to discuss the use of resources are the availability of land and water

per capita in different socioeconomic groups.

The main trade-offs among resources are between nitrogen and land, and

between human labour and machinery. These trade-offs are analysed in

chapters 3-5. Chapter 3 focuses on the trade-off between nitrogen and land by

studying the impact of population density and type of diets on the intensity of

the agricultural production system. The indicator to measure the intensity of

the system is the nitrogen fertilizer application rate. A model is developed and


15

validated with country level data to show the relationships among population

density, diets and production system. Chapter 4 also studies the trade-off

between nitrogen and land but with a different approach. It focuses on the

impact of diets and production systems on the per capita use of both nitrogen

fertilizer and land. A footprint approach is followed, so the indicators to discuss

the trade-offs between nitrogen and land are the amount of nitrogen fertilizer

and land use per capita. Chapter 5 focuses on the trade-off between human

labour and machinery. This chapter studies the impact of diets and production

systems on the amount of farm labour required for food production. Similar to

chapter 4, a footprint approach is followed, so the indicator to discuss global

differences in labour is hours of farm labour needed per person.

The regional differences in diets are studied in detail in chapter 6. A type of

footprint approach is followed, but in contrast with chapter 4 and 5, the

production system is kept constant to only explore the impact of regional

dietary differences on the use of resources. The indicators to discuss the impact

of diets on resources are the per capita use of land, water, energy and GHG

emissions.

Finally, chapter 7 is the general discussion of the thesis in which the general

methodology is described in detail and the results of the sub-projects are

integrated. The future challenges for the use of resources are discussed using

the new insights obtained with the integrative analysis done in this thesis.

Chapter 2. Increasing inequality as to availability of land and water

17

Chapter 2.

Increasing inequality between poor and rich countries as

to availability of land and water by 2050*

Abstract

Future global food demand will increase resulting on a strong demand of

agricultural resources. The amount depends on population numbers, diets and

agricultural practices. Furthermore, the availability of land and water is

essential for achieving food self-sufficiency.

In this paper, we assess the availability of land and water for food in relation to

the economic development of the population, using the Gross Domestic Product

(GDP) as indicator. We group the global population into six “GDP groups”. We

study how population growth and diets affect the availability of land and water

per capita for food production.

We show large differences between rich and poor countries, which can strongly

increase in the coming decades because of the large population growth of the

poor countries. For instance, by 2050, the richest quarter of world population

will have three times more arable land per person than the rest. In addition, the

people changing diets to more affluent consumption will be the ones with less

available resources per person. More than two thirds of global population will

not have enough land and water to produce the food for an affluent food

consumption by 2050. Thus, the large land and water constrains of the poor

will result in stronger challenges for food security then predicted in other

studies.

* Under Review in Agricultural Systems as: M.J. Ibarrola Rivas & S. Nonhebel, Increasing

inequality between poor and rich countries as to availability of land and water by 2050

18

2.1 Introduction

Global food demand is expected to increase in the coming decades due to

increasing global population and changing food consumption patterns

(Alexandratos & Bruinsma, 2012; Foley et al., 2011; Godfray et al., 2010a;

Godfray et al., 2010b). The production of this food will require large amount of

resources (Tilman, 1999) i.e. land as cropland and pastures (Alexandratos &

Bruinsma 2012; Ramankutty et al., 2008), water (Rockström, 2003) and energy

inputs mainly fertilizers and machinery (Smil, 1999; Woods et al., 2010).

The amount of resources needed to feed a population depends firstly on the

size of the population: more people need more food, secondly on the

consumption pattern of this population: diets rich in animal products require

more than vegetarian diets mainly based on staple foods, and thirdly on the

agricultural production system (Kastner et al., 2012; Leach et al., 2012;

Mekonnen & Hoekstra, 2011a; Xiong et al., 2008). Agricultural production

systems with high crop yields require less land than systems with low crop

yields, but to obtain these high yields a trade-off exists with a larger use of

energy and water resources (Evans, 1980; Licker et al., 2010).

Most countries in the world are food self-sufficient to a certain extend. Globally,

only 12% of domestic food supply is originated from imports (data of 2009:

FAO, 2013d). The availability of agricultural resources is thus key to achieve

food self-sufficiency. Land and water inputs are local resources in contrast with

energy inputs which can be imported. So, the availability of land and water in a

country is crucial to achieve national food self-sufficiency (Fader et al., 2013).

Fader et al. have studied the availability of land and water for food production

throughout the world. They show that some countries have strong constrains of

land and water which currently results on 16% of global population dependent

on food imports. And future population growth can strongly increase this

number up to 50% which can strongly affect food security for these countries.

Thus, food security is strongly dependent on the availability of land and water

for food production.

The aim of this paper is to assess global food supply in relation to the economic

development of the population and the availability of agricultural resources:

the difference of resource availability between poor and rich countries. In

contrast to Fader et al. 2013, we study the availability of land and water per

capita not from a geographical perspective but from a demand perspective


19

because the drivers of food demand are related with economic development.

The increase in socioeconomic development result, first on a decline of

population growth (demographic transition (Chesnais, 1992; Lutz and Samir

2010)); and second on dietary changes from diets based on staple foods to

more luxurious diets including sugar, animal products, etc. (nutrition transition

(Kearney, 2010; Poleman & Thomas, 1995; Popkin, 1993)). We use the

differences in welfare throughout the world as the starting point of our study.

With this approach, our analysis deviates considerably from existing studies on

global food supply (Ausubel et al., 2013; Fader et al., 2013; Foley et al., 2011;

Licker et al., 2010; Lobell et al., 2009; Ray et al., 2013) which use the

geographical situation as the starting point. They determine production

possibilities in various climates and soil conditions which determine yield

potentials to analyse whether agricultural production is enough to feed the

population. In this study, first we analyse what is needed; and then, we assess

the availability of agricultural resources to supply this food. We use land and

water available per capita as indicators to assess future food supply for several

economic development groups.

2.2 Methodology

We collected data for 187 countries from 1960 and 2010 on food consumption

(FAO, 2013d), land availability (FAO, 2013c), water availability (FAO, 2013a),

Gross Domestic Product (GDP) per person (World Bank, 2014), population and

expected population for 2050 with medium fertility rate (United Nations,

2011). These countries accounted for 99% of global population in 2010. For the

average diet, we aggregated the food consumption data into five categories:

staple food (cereals, roots and pulses), affluent vegetal (sugars, vegetable oils,

vegetables and fruits), meat (bovine, pig, poultry and fish), other animal (milk,

eggs and animal fats) and other (alcoholic beverages, tree nuts, stimulants and

spices). For water availability, we used the “Total renewable water resources

(actual)” given by Aquastat (FAO, 2013a) which is the annual amount of surface

and ground water available in a country. For the data of water and arable land

in 2050, we used the same numbers of 2010. Water availability is relatively

stable throughout the years (FAO, 2013a). Agricultural land changes in time

due to different land use changes. But, even though arable land is projected to

increase for 2050, the increase rate is expected to be lower than in the last

decades, globally “only” 4% (Alexandratos & Bruinsma, 2012). It will expand

20

differently among regions (Alexandratos & Bruinsma, 2012, table 4.8), and the

projection depends on the assumption for agricultural expansion. Therefore, in

our analysis we assume no agricultural expansion to avoid accumulated

assumptions.

We ordered the countries according to their GDP per capita in 2010. Then, we

clustered the countries into six “GDP” groups. The population in each group in

2010 is between 0.7 billion and 1.4 billion people to be relevant for a global

analysis (see table 2.1). The countries in each group are the same for 1960,

2010 and 2050. Countries and groups are in Appendix 2.

The groups are numbered 1 to 6 in the order of increasing GDP per capita i.e.

from poor to rich. These groups may contain countries in different regions.

Group 1 includes almost all countries in Sub-Sahara Africa and some from Asia

like Pakistan, Vietnam and Bangladesh. Group 3 includes countries from all

over the world like Philippines, Egypt, Indonesia, Guatemala and Ukraine.

Group 5 is a mixture again with Iran, Brazil, Russia, Mexico, Turkey, Thailand,

South Africa and others. Group 6 includes Western Europe, the USA, Korea,

Japan and others. Group 2 and group 4 are only one country each, India and

China respectively, because of their large population. So, the countries in each

group do not necessarily come from the same continent or climatic region; but

the population in each group follows similar trends on the drivers of food

demand because they have similar economic development.

2.3 Changes in the drivers of food demand

Figure 2.1b shows the characteristics of the diets in 2010. In the first two

groups (GDP less than $600 per capita per year), people consume around 2200

kcal per person per day and 80 % of their calories originate from staple food.

This is in strong contrast with the rich groups on the right part of the graph

where only 30% of the calories is from staple food, 30% is from animal origin

and 30% is affluent vegetal.

The groups in the middle show a gradual change from the staple menus to the

luxurious diets, by declining the shares of staple foods and increasing the

shares of animal products and affluent vegetables. Figure 2.1a shows the data

from 1960. Again, the groups with GDP values below $1000 per person per year

have diets with mainly staple food; and the two richest groups show more


21

luxurious consumption patterns. Up to now, the nutrition transition has been

studied on a country level. Here, we show that the nutrition transition can also

be recognized in these “GDP” groups.

Figure 2.1 The composition of the diets in the six groups considered in 1960 and 2005.

Data based on FAO (2013d) and United Nations (2011).

Table 2.1 Changes in population.

Data source from United Nations (2011).

Population [Billion people]

Avg. annual population growth rate [% per year]

GROUPS 1960 2010 2050 1960-2010 2010-2050

1 0.4 1.4 2.7 4.8% 2.5%

2 0.5 1.2 1.6 3.1% 0.9%

3 0.2 0.7 1.0 4.1% 1.3%

4 0.7 1.4 1.4 2.0% 0.0%

5 0.7 1.2 1.5 1.8% 0.5%

6 0.7 1.0 1.1 0.9% 0.3% Notes: a) We use data of population projections for 2050 with medium fertility rate. b) We excluded all ex-USSR countries to calculate the average annual population growth rate.

22

It is important to point out that the groups in the two graphs are not in the

same order. In 1960, group 4 was the poorest group, while group 1 was on the

third position. The shift in position is caused by different economic growth

rates.

The demographic transition theory is reflected on table 2.1: the groups with

low GDP have the highest population growth rates. With exception of group 4

which shows lower growth rates than groups 5 and 6, because group 4 is China

with its one child policy.

Figure 2.2 Distribution of global population, total land and fresh water (ground and

surface water) among the 6 groups. The numbers in the pie charts indicate the group

number. The global water data is from the year 2010; and population projections for

2050 are with medium fertility rate. Data based on FAO (2013a), FAO (2013c) and

United Nations (2011).


23

The population of each group changes over time due to the differences in

population growth rates. This is clearly shown in table 2.1 and figure 2.2. In

1960, two thirds of the global population lived in groups 4, 5 and 6. In 2050,

these groups will account for only one third of the global population. Note that

in 1960 global population was 3.2 billion people, in 2010 6.9 billion and in 2050

we expect 9.3 billion people. The largest share of these 2.5 billion extra people

will be born in groups 1,2 and 3.

2.4 Differences in land and water availability on a global scale

We calculated the distribution of total available land and water among the

groups, and compared it with the distribution of global population in 1960,

2010 and 2050 (figure 2.2). The unequal distribution is striking: 64% of the

land and 58% of the water are available for groups 5 and 6. Since population is

more equally distributed among the groups, land and water availability per

capita shows enormous variations. For instance, in group 5, 17 times more land

and 23 times more water is available per capita in comparison with group 2.

Table 2.2 Share of agricultural area (arable land and pastures) from the total land for

each group. Data source from FAO (2013c)

Groups 1960 2010

Arable land Pastures Arable land Pastures 1 8 % 26 % 12 % 28 % 2 54 % 5 % 57% 3 % 3 7 % 31 % 13 % 30 % 4 11 % 26 % 14 % 42 % 5 8 % 20 % 9 % 24 % 6 12 % 26 % 11 % 22 %

Only a share of the total land is used for agricultural practices. This share

changes over time: as nature areas are converted into agricultural production

or the other way around: when agricultural land is abandoned and nature takes

over again or when agricultural land is needed for infrastructural purposes

(housing areas, roads). From 1960 to 2010, the change in agricultural land

surface was relatively small: globally it only increased 10% (FAO, 2013c).

24

The distribution of agricultural land also shows variations among the groups.

Group 5 only uses 33% of their total land for agriculture; while group 2 uses as

much as 60% (table 2.2). In general, groups with small amount of total land use

large share of it for agriculture, and countries with a lot of land use a smaller

share. In groups 1-5, agricultural land increased in the last 40 years, while in

group 6 it declined. Groups differ in whether land is used for pastures or for

arable land. Group 2 has hardly any land as pastures, while in the other groups

pastures are the largest share of agricultural land use.

Figure 2.3 Arable land and water availability per person and changes over time. The

green bars indicate the arable land per capita and the blue bars indicate the availability

of water. Data based on: arable land from FAO (2013c), water from FAO (2013a), and

population numbers from United Nations (2011).

To discuss the availability of land per capita for food production, we focus on

arable land since most of the food is produced on it (FAO, 2013c). Figure 2.3

shows the arable land and water availability per person in the 6 groups over

time, which is decreasing for all groups. This decline is due to population

growth, since water availability is relatively constant throughout the years and

only small changes in total agricultural area took and will take place (table 2.2


25

in this paper; FAO, 2013c; table 4.8 in Alexandratos & Bruinsma, 2012). The

decline rates differ among the groups because population growth is different

(table 2.1). As a result, the poor groups have a stronger decrease in the

availability of land and water per capita.

Not only the changes in land and water availability are different but also the

absolute values. In 2010, group 5 has 3800 m2/cap for arable land; in contrast,

group 2 has 1400 m2/cap and group 4 only 900 m2/cap. Differences in water

availability are stronger. Group 3 and group 5 have as much as 21,000 m3/cap

and 37,000 m3/cap respectively, while group 2 and 4 only have 1,600 and 2,000

m3/cap respectively.

Changes on water availability per capita only depend on population growth

since it is relatively stable throughout the years. Changes on arable land

availability, on the other hand, depend on both population growth and arable

land expansion (see table 2.2). For this reason, changes on arable land and

water show different patterns. This is clearly illustrated in group 1 for which in

1960 arable land per person is relatively lower than water per person, and in

2010 arable land is relatively higher (see figure 2.3).

2.5 Future implications and discussion

Presently, two thirds of the total agricultural land and fresh water are available

for one third of the global population (groups 5 and 6). These people also have

the highest welfare levels. They passed the nutrition transition: their diets

contain large amount of luxurious products and no major changes in their

consumption pattern are expected. Next to this, these countries also went

through the demographic transition phase: the expected population growth is

small. Therefore, it can be expected that these countries will not need more

resources to feed their population in the future.

In group 4, the situation is less positive. Even though population growth is low

(table 2.1), which makes the land availability per person relatively constant

(figure 2.3), this group is in the middle of the nutrition transition phase. Diets

are changing rapidly: meat consumption doubled over the last 20 years as well

as the consumption of other affluent products (FAO, 2013d). Their

consumption patterns are still not as luxurious as the ones consumed in group

5 and 6 (figure 2.1); and large increases in consumption of luxurious food items

26

are expected in the coming decades. Kastner et al. (2012) showed that for this

group the changes in diets were the major cause for the increase on land

demand for food. In this group, more land and water will be demanded for food

due to dietary change but not due to population growth.

The present situation in group 1 and 3 shows some resemblance (figure 2.3). In

2010, both groups had similar availability of agricultural land and water per

capita. However, the expected population growth rate in group 3 is lower than

in group 1. This will have large impacts on the future availability of resources

e.g. availability of arable land in group 3 will drop from 2000 m2/cap to 1400

m2/cap; while in group 1 will drop from 2000 m2/cap to only 1000 m2/cap; and

water availability in group 3 will drop from 20,000 m3/cap to 14,000 m3/cap

and in group 1 from 15,000 to 8,000. Note that in 1960 group 1 had twice the

amount of water than group 3, and in 2050 group 1 will only have half the

amount of group 3. This shows the strong impact of population growth in the

availability of resources. In both groups, diets are mainly based on staple foods.

With the increase of welfare a change to more luxurious food patterns can be

expected and more resources per person will be needed to produce these

luxurious foods. Future demand of resources will result from the combination

of high population growth and a change to more luxurious diets which will

result in a challenging situation for future food supply.

The projections for group 2 show the most challenging situation. The land and

water availability per capita are the lowest among the groups and will decrease

even more in the coming decades. Agricultural expansion might be limited since

60% of the land is already in use for food production (table 2.2). And, (almost)

all the agricultural land is arable land which is the high productivity land. This

group still shows high population growth rates (table 2.1) and the nutrition

transition phase still has to start (figure 2.1: 80% of the calories are from staple

food). On the other hand, the population of this group is India where the

majority is Hindu with vegetarian diets. This may result on relative smaller

demand for resources per capita than the resources required for diets with

large amount of animal products in other parts of the world.

Thus, on the short term the situation in group 4 is the most urgent but in the

coming decades the countries in groups 1 and 2 will face comparable problems.

The data we used for the projections of 2050 are the arable land and water

availability in 2010. This assumption might be an underestimation or


27

overestimation considering that water availability might change with climate

change, and arable land will expand (Alexandratos & Bruinsma, 2012).

Therefore, it is obvious that our results cannot be considered as face values, but

they should be considered as an order of magnitude that helps us understand

and recognise the various underlying patterns. We don’t intend to predict the

future with our analysis; we only show the impacts of the demographic and the

nutrition transition on future per capita availability of arable land and water.

In this paper, we studied food security from the demand side: number of people

and their diets, and we used the economic situation of the individual countries

as the starting point. We disconnected the people from their geographical

location. Appendix 2 indicates the countries in each group. It shows that within

one group, countries from different continents and from different climatic

zones are together. Within one group, climates can differ from mainly tundra to

the wet tropics. As a consequence, the agricultural production potential will

differ a lot. In wet tropics, up to three harvests per year are possible while in

the northern latitudes the cold winter only allows one harvest per year. So, the

choice for starting from the demand side makes it difficult to assess whether

there is enough land and water to feed the people (production side) since we

don’t know what can be produced.

However, Kastner et al. (2012) calculated that for producing a luxurious diet

(including meat) using a high yielding agricultural production system (group 6)

2300 m2 of arable land per person was required. Only in group 5 and 6, such

area of arable land is presently available per person and due to the limited

population growth it will also be available in 2050. In group 2, even the total

land surface in 2050 will be smaller (1,800 m2/cap), so there is no option that

2300 m2 of arable land per person will become available in future to feed this

population. In group 1, 3 and 4, arable land expansion will be needed to achieve

this amount of arable land per person.

In relation to water, Mekonnen & Hoekstra (2011a) calculated the water

required to produce the food for several countries, what they call the water

footprint. They show large variations driven by the type of diets and

agricultural systems with a global average diet requiring 1390 m3/cap/yr of

water (Mekonnen & Hoekstra, 2011a, fig 11). Groups 2 and 4 will have 1500

and 2000 m3/cap/yr respectively by 2050, which means that in order to use the

global average requirement of water they need to use almost all their water

available, which represents large infrastructure and management challenges.

28

The most striking result of this analysis is the upcoming huge inequality in

available arable land and water per person between the poor and the rich. In

2050, the richest population (groups 5-6) will have three times more arable

land per person than the rest of the world (groups 1-4). These people, 72 % of

the global population (groups 1-4), will live in countries where arable land will

be less than half the amount needed to produce the food for a luxurious diet

(2,300 m2/cap: based on Kastner et al. (2012)). . Also, the strong land and

water constrains will have effects on global food markets. This has already been

studied by Fader et al. (2013). Here we show that these people will be the

poorest on Earth. It is very likely that the poor countries will increase their food

imports, and the rich countries with larger availability of agricultural resources

will turn into large food exporters.

2.6 Conclusions

In this paper, we only study the availability of resources for food production to

assess future global food supply. However, the amount of food that can be

produced depends not only on these resources but also on the agricultural

practices (production side), which are outside the scope of this paper.

These agricultural practices are usually related with economic development.

High income countries generally have high productivity systems with high crop

yields (Licker et al., 2010; Hengsdijk & Langeveld, 2009; Lobell et al., 2009).

These efficient systems are possible due to optimal conditions on

infrastructure, access to technical and management knowledge and access to

investments for fertilizers, irrigation, crop protection, soil conservation, etc.

(Lobell et al., 2009; Godfray et al., 2010a). These factors are usually lacking in

low income countries resulting on low productivity systems.

Our results show that the people who will have strong arable land and water

constrains in the coming decades will be the poor. So, for the reason mentioned

above, these countries will encounter strong socioeconomic and political

challenges to achieve high agricultural productivity. The other solution for

resource constrains, apart from agricultural productivity, is to increase food

imports; but depending on food imports can have strong risks for national food

security especially in poor countries. Thus, in the coming decades, the challenge

of reaching food security for the poor will become stronger because of huge


29

land and water constrains; and so the potential to achieve food security will

become more unequal between the rich and the poor.

Appendix 2

Groups and countries

The following tables show the countries included on the 6 groups used in this paper. The order of the countries in each group is based on their GDP per capita in 2010 (poor to rich). The tables show the name of the country with its region given by FAO.

Table A1. Group 1.Total population in 2010 (United Nations, 2011): 1,368 Million. GDP per capita in 2010 (World Bank, 2014): US $150 to US $1,000

COUNTRY REGION COUNTRY REGION COUNTRY REGION

Burundi E Africa Guinea-Bissau W Africa Timor-Leste SE Asia

Congo, D. R. Mid Africa Haiti Caribbean Chad Mid Africa

Eritrea E Africa C. African R. Mid Africa Zambia E Africa

Malawi E Africa Tanzania E Africa Pakistan S Asia

Ethiopia E Africa Burkina Faso W Africa Uzbekistan C Asia

Liberia W Africa Gambia W Africa Mauritania W Africa

Madagascar E Africa Mali W Africa Sudan N Africa

Niger W Africa Bangladesh S Asia Senegal W Africa

Guinea W Africa Benin W Africa Lesotho S Africa

Rwanda E Africa Kyrgyzstan C Asia Viet Nam SE Asia

Sierra Leone W Africa Kenya E Africa Yemen W Asia

Nepal S Asia Myanmar SE Asia Cameroon Mid Africa

Mozambique E Africa Comoros E Africa P. N. Guinea Oceania

Zimbabwe E Africa Cambodia SE Asia Cote d’Ivoire W Africa

Togo W Africa Ghana W Africa R. Moldova E Europe

Uganda E Africa Laos SE Asia Nigeria W Africa

Tajikistan C Asia

Table A2. Group 2. Total population in 2010 (United Nations, 2011): 1,205 Million. GDP per capita in 2010 (World Bank, 2014): US $1,034 COUNTRY REGION India S Asia

30

Table A3. Group 3. Total population in 2010 (United Nations, 2011): 656 Million. GDP per capita in 2010 (World Bank, 2014): US $1,050 to US $2,820


Solomon I. Oceania Indonesia SE Asia Samoa Oceania

Kiribati Oceania Sri Lanka S Asia Morocco N Africa

Guyana S America Syria W Asia Tuvalu Oceania

Bolivia S America Paraguay S America Swaziland S Africa

Nicaragua C America Bhutan S Asia Armenia W Asia

Mongolia E Asia Georgia W Asia Angola Mid Africa

Philippines SE Asia Congo, Rep. Mid Africa Tonga Oceania

Iraq W Asia Ukraine E Europe Turkmenistan C Asia

Honduras C America Vanuatu Oceania Marshall I. Oceania

Egypt N Africa Guatemala C America Jordan W Asia Table A4. Group 4. Total population in 2010 (United Nations, 2011): 1,390 Million. GDP per capita in 2010 (World Bank, 2014): US $2,870

COUNTRY REGION China E Asia

Table A5. Group 5. Total population in 2010 (United Nations, 2011): 1,229 Million. GDP per capita in 2010 (World Bank, 2014): US $2,950 to US $ 16,000


El Salvador C America Kazakhstan C Asia Turkey W Asia

Iran S Asia Dom. Rep. Caribbean Mexico C America

Azerbaijan W Asia Cuba Caribbean Lithuania N Europe

Cape Verde W Africa Argentina S America Palau Oceania

Algeria N Africa Romania E Europe Chile S America

Thailand SE Asia Costa Rica C America Libya N Africa

Ecuador S America St Vin. & Gren. Caribbean Poland E Europe Bosnia & Herz. S Europe Brazil S America Estonia N Europe

Macedonia S Europe South Africa S Africa Croatia S Europe

Albania S Europe Venezuela S America St Kitts & Nev. Caribbean

Fiji Oceania Saint Lucia Caribbean Hungary E Europe

Tunisia N Africa Mauritius E Africa Ant. & Barbu. Caribbean

Serbia E Europe Botswana S Africa Seychelles E Africa


31


Peru S America Gabon Mid Africa Eq. Guinea M Africa

Colombia S America Panama C America Czech Rep. E Europe

Namibia S Africa Malaysia SE Asia Slovakia E Europe

Belize C America USSR /Russia E Europe Bahrain W Asia

Suriname S America Grenada Caribbean Trin.& Tobago Caribbean

Jamaica Caribbean Dominica Caribbean Barbados Caribbean

Bulgaria E Europe Uruguay S America Oman W Asia

Montenegro E Europe Lebanon W Asia Malta S Europe

Belarus E Europe Latvia N Europe Saudi Arabia W Asia

Maldives S Asia

Table A6. Group 6. Total population in 2010 (United Nations, 2011): 993 Million. GDP per capita in 2010 (World Bank, 2014): US $18,500 to US $ 80,300


Portugal S Europe Italy S Europe Austria W Europe

Slovenia S Europe Greenland N America Netherlands W Europe

Korea, Rep. E Asia Singapore SE Asia Sweden N Europe

Puerto Rico S Europe France W Europe USA N America

Bahamas Caribbean Canada N America Ireland N Europe

Greece S Europe Andorra S Europe Denmark N Europe

Israel W Asia Germany W Europe San Marino S Europe

Cyprus W Asia Australia Oceania Iceland N Europe

U. A. Emirates W Asia Japan E Asia Qatar W Asia

Brunei D.. SE Asia Belgium (2010) W Europe Switzerland W Europe

Spain S Europe Belg.-Lux(1960) W Europe Norway N Europe

New Zealand Oceania United Kingdom N Europe Bermuda N America

Kuwait W Asia Finland N Europe Luxem. (2010) W Europe

Chapter 3. Estimating future global needs for nitrogen

33

Chapter 3.

Estimating future global needs for nitrogen based on

regional changes of food demands *

Abstract

The use of synthetic nitrogen fertilizer is of great importance for global food

supply, but it is also causing large environmental problems. The amount of

fertilizer needed on a global scale is affected by the size of the population (more

people more food), the type of their diets (luxurious diets require more

resources) and the agricultural production system (intensive systems use more

fertilizer than extensive systems). In the coming decades, global population will

increase, diets will change and a strong intensification of agriculture is

expected. In this paper, we assess how the changes in all these factors will

affect future needs of fertilizers on both a global and a national scale. We

developed a model that describes the various relations among population

density, diets and production systems, and validated the outcomes with

historical country level data. The variations over the globe are large. The model

shows that in the regions with low population densities not much change in

nitrogen applications is to be expected. On the other hand, in densely populated

regions, with high population growth rates and expected changes in diets, the

needs for synthetic nitrogen show an exponential increase. By using existing

projections with respect to population growth and dietary changes, by 2050

over 70% of the global population will live in countries where an exponential

increase of the nitrogen application is needed to feed the people. This may

have large consequences for the global environment.

* Resubmitted version under review in Environmental Development as: M.J. Ibarrola

Rivas & S. Nonhebel. Estimating future global needs for nitrogen based on regional changes of food demands

34

3.1 Introduction

The use of synthetic nitrogen fertilizer has had enormous global environmental

impacts (Sutton, et al. 2013). First, it is highly energy intensive. Globally, half of

the energy use in agriculture is related to the production and application of

synthetic nitrogen fertilizer and it is a major greenhouse gas emitter (Woods et

al., 2010). Secondly, its use has strongly affected the global nitrogen cycle

(Galloway & Cowling 2002; Gruber & Galloway, 2008; Rockström et al., 2009)

by doubling the emissions of reactive nitrogen into the biosphere (Smil, 1999)

which has affected the global climate (Gruber & Galloway, 2008). Third, large

application of nitrogen has local effects on the environment such as

acidification, eutrophication and loss of biodiversity (Smil, 1999; Galloway &

Cowling, 2002; Isbell et al., 2013). However, the food for the present global

population cannot be produced without it, and nowadays half of the world food

consumption is based on its use (Smil, 2001; Erisman et al., 2008).

The world population will reach nine billion people by 2050 (United Nations,

2011), and food consumption per capita will increase as well (Alexandratos &

Bruinsma, 2012; Kearney, 2010). Synthetic nitrogen fertilizer will play an

important role in the increase of global food supply as well as in the

environmental impacts related to the increase of food production (Tilman,

1999; Eickhout et al., 2006; Bouwman et al., 2009; Sutton et al., 2013). Tilman

et al. (2011) estimates that if the future global consumption of nitrogen

fertilizer follows a linear extension of the past 50 years’ trend, it will lead to

serious global environmental consequences. So, achieving global food security

while reducing environmental problems is one of the biggest challenges

humanity will face in the next decades (Godfray et al., 2010a; Godfray et al.,

2010b; Foley et al., 2011).

The use of nitrogen for food production has been widely studied. We have

identified three lines of research addressing different drivers of its use.

First, population density drives the amount of nitrogen applied for a certain

population. Smil (2001, fig 8.8) and Arizpe et al. (2011, fig 7) show that for

several countries the increase of population density, due to population growth,

have led to an increase in the nitrogen application per unit of area.

Second, food consumption patterns drive nitrogen requirements. Different

types of diets require different amounts of nitrogen (Leach et al., 2012;


35

Lassaletta et al., 2013). Food products of animal origin require more nitrogen

than those of vegetal origin because of the inefficiency of producing animal

proteins from vegetal proteins (Smil, 2002a; Smil, 2002b). Also, differences

exist among animal products (Chatzimpiros & Barles, 2013), for instance the

production of one kilogram of beef requires more nitrogen than the production

of one kilogram of chicken (Smil, 2002b).

Third, the type of agricultural production system drives the use of nitrogen

fertilizer depending on the application rate level (kilograms of Nitrogen per

hectare). Large regional variability exists in its use and in the corresponding

local environmental impacts because of different application rates (Bouwman

et al., 2005; Galloway et al., 2008; Liu et al., 2010). Some regions, like industrial

countries and Eastern Asia, have an excess of reactive nitrogen in their

environment due to high application rates (Isermann & Isermann, 1998;

Howarth et al., 2002; Shindo et al., 2006; Xiong et al., 2008). Other regions, like

most African countries, have a lack of nitrogen in their soil because of the low

application rates which limits their agricultural production and affects their

food supply (Galloway et al., 2008; Liu et al., 2010).

So, changes in population, diets and agricultural systems affect the need for

nitrogen fertilizer. The expected changes differ largely over the globe. In

Western Europe, population growth has come to a standstill, menus already

contain large amounts of animal products and agriculture is very intensive. So,

no changes in nitrogen needs are expected in this region. In other parts of the

world, population growth rates are still high (Africa) or large changes in meat

consumption are taken place (China) and nearly everywhere agricultural

production systems are intensifying. Based on these expected changes, we can

expect an increase in global needs for nitrogen fertilizers. In this paper, we do

an assessment for the future needs of nitrogen fertilizer on global scale based

on the existing regional variations of population, diets and agricultural systems,

and the expected changes. First, we study the historical changes in the drivers

for different global regions from 1960-2010 (section 3.2). Then, we develop a

model to integrate the impact of the various drivers on the nitrogen demand

(section 3.3). The results of the model are validated with country level data

(section 3.4), and finally we use the model to assess the global nitrogen

demands in 2050 (section 3.5).

36

3.2 Historical analysis of regional changes in population,

menus and nitrogen use

3.2.1 Methods and data

We obtained country level data of various drivers for 1960-2010 from existing

databases: population (United Nations, 2011), food consumption (FAO, 2013d),

arable land (FAO, 2013c) and nitrogen fertilizer use (FAO, 2013c). Population

density was calculated by dividing the size of the population by the area of the

arable land because we are interested in the availability of arable land per

person which is, in general, where the nitrogen fertilizer is being used. Since the

consumption of animal products affects nitrogen demand, we determined the

total caloric intake of animal products per person per day. Finally, the

application of nitrogen was calculated by dividing the total national

consumption of synthetic nitrogen fertilizer by the amount of arable land of

each country. The countries were clustered in seven regions: A: Africa (with

950 million people in 2010), B: India (1210 million people), C: Southeast Asia

(950 million people), D: China (1370 million people), E: Latin America (570

million people), F: North America (340 million people) and G: Western Europe

(400 million people). These regions accounted for 85% of world population in

2010.

3.2.2 Results

In nearly all regions, population density, consumption of animal products and

nitrogen application has been increasing over the last decades (figure 3.1).

However, the absolute values differ per region. In 2010, Asia was the region

with the largest population density, followed by Western Europe, Africa, Latin

America and North America, see black bars of figure 3.1a which indicate the

values in 2010. So, Asia was the region with the smallest amount of arable land

per person and North America with the largest. The population density in China

was twice that of Western Europe, three times larger than that in both Latin

America and in Africa, and as much as ten times larger than that in North

America. During the last decades, global population density increased because

global population doubled (United Nations, 2011) while arable land “only”

increased 10% (FAO, 2013c). Population density has changed differently

among the regions due to different changes in both population numbers and


37

arable land. Africa had the largest population growth and strongly increased its

population density. But India, with a smaller population growth, ended up with

a larger growth in population density because it did not increase their arable

land as much as Africa. Western Europe increased its population density not

because of a large population growth but because its arable land decreased.

The consumption of animal products differs amongst regions. Western Europe

and North America consumed the largest amount of animal products per capita,

followed by Latin America, China, Southeast Asia, India and Africa; see the

values of 2010 of figure 3.1b. During the last fifty years, the global consumption

of animal products strongly increased (FAO, 2013d), but the increase was not

the same in all regions. North America, Western Europe and Africa were more

or less stable, but Africa consumed much less than the former two. North

America and Western Europe were large consumers and they show a levelling

off in contrast with Latin America and parts of Asia which had the largest

increase in consumption of animal products and they did not show a levelling

off. Some of these regions still consume half the amount of animal products

compared to that of Western Europe and North America.

Figure 3.1 Drivers for the use of nitrogen fertilizer for the seven regions. The regions

are: A: Africa, B: India, C: Southeast Asia, D: China, E: Latin America, F: North America

and G: Western Europe.

Nowadays, the region with the highest application of nitrogen per hectare is

Asia, and Africa has the lowest (figure 3.1c). During the last five decades, mainly

38

Asia but also Latin America strongly increased the nitrogen application rate.

Africa did not increase substantially and by 2010 its average application rate

was still very small compared to the rest of the regions. North America and

Western Europe showed a levelling off from 1960 to 2010. North America

increased its application rate until 1980, and then, it remained stable. Western

Europe had by far the largest application rate in 1960. It increased until 1980,

and then, it began to decrease.

3.2.3 Discussion

When we compare our regional results with relations mentioned in literature

some striking differences can be seen. Figure 3.1 shows that large consumption

of animal products is not always related with large nitrogen application as

suggested by Smil (2002a, 2002b) and Leach et al. (2012). The regions with the

largest consumption of animal products did not have the largest nitrogen

application rate. North America and Western Europe had the largest

consumption of animal products but Asia had the largest application rate. In

addition, Smil (2001) and Arizpe et al. (2011) show that countries with a high

population density have high nitrogen application rates. Figure 3.1 shows that

some regions follow this pattern like Asia (B, C and D). These three regions had

relatively high population density and high nitrogen application rate. But Africa

(A) and North America (F) do not follow this pattern. Africa has a larger

population density than North America, but a smaller application rate. In

addition, in Western Europe, population density increases over time while in

the last decades a decline in fertilizer use is measured. So, the relations

mentioned in the literature are not always found on regional scales.

3.3 Integration of the drivers: model development

We integrate the three drivers into a model. The model aims to understand

patterns rather than calculate exact data. The three parameters are population

density (by means of the inverse: ha of arable land cap-1), consumption of

animal products (kcal cap-1 day-1), and nitrogen application rate (kg N ha-1).

Population density and consumption of animal products are independent

variables, and nitrogen application rate is the dependent variable of the other

two parameters. The model illustrates their interrelationships on a crop field

scale.


39

3.3.1 Model description

We take the food production as starting point and we determine the amount of

nitrogen needed to produce a certain amount of food. Under natural conditions

(no fertilizer use) the natural nutrients available in the soil for uptake by the

crop determine the amount of food that can be produced. If we want to

produce more food two options exist: either more land should be put into

production (land expansion) or the crop yield should be increased (the amount

of food produced in a certain area: intensification). In this paper, we focus on

the increase of food production by increasing crop yields, as intensification of

the production is expected to play the major role in the global increase of food

production (Alexandratos and Bruinsma, 2012). Nutrients have to be added to

the soil in order to increase crop yields. The number of people we can feed from

a certain amount of nutrients depends on their menus, and the amount of

animal products plays an important role. We distinguish three hypothetical

diets: a vegetarian diet Dv requiring N amount of nitrogen for its production, a

more affluent diet Da, with a low amount of animal products, requiring 50%

more nitrogen than Dv, and a diet rich in animal products Da ext requiring

twice the amount of nitrogen than Dv.

Our starting point is that in area A we can produce the food for one person with

a vegetarian diet Dv using the natural nitrogen in the soil which is N. If the

population doubles, we need 2N to produce the food for both people. Since the

natural nitrogen in area A is N, then we need to apply another amount of

nitrogen N. Column Dv of table 3.1 shows the amount of external nitrogen that

we need to apply if the population grows from one to ten people. The following

columns show the amount of nitrogen application required for diets Da and Da

ext.

3.3.2 Results

Figure 3.2 shows the results of the model. The application of external nitrogen

increases exponentially with the decrease of availability of land per capita due

to population growth. So, the increase of population density increases the

application of nitrogen fertilizer, as shown by Arizpe et al. (2011, fig 7) and Smil

(2001, fig 8.8). Also, diets rich in animal products require a larger nitrogen

application rate. Our aim with this model is only to show the relationships

40

between the parameters, therefore we do not assign units of numerical values

to the model, so table 3.1 and figure 3.2 do not have units.

Table 3.1 Application of external nitrogen for different

population numbers and diets on a crop field scale

Population N

atu

ral N

itro

gen

in

th

e cr

op

fiel

d [

N]

Diet Dv (vegetarian)

Diet Da (low animal products)

Diet Da ext (large animal products)

Nu

mb

er o

f p

eop

le

Av

aila

bil

ity

of

lan

d

[A

cap

-1]

Ext

ern

al

nit

roge

n

req

uir

ed [

N]

Ap

pli

cati

on

o

f N

itro

gen

[N

A-1

]

Ext

ern

al

nit

roge

n

req

uir

ed [

N]

Ap

pli

cati

on

o

f N

itro

gen

[N

A-1

]

Ext

ern

al

nit

roge

n

req

uir

ed [

N]

Ap

pli

cati

on

o

f

Nit

roge

n [

N A

-1]

1 1.0 1 1 0 1.5 0.5 2 1

2 0.5 1 2 1 3.0 2.0 4 3

3 0.3 1 3 2 4.5 3.5 6 5

… … … … … … … … …

10 0.1 1 10 9 15.0 14.0 20 19

Figure 3.2 Model integrating the drivers of external nitrogen fertilizer use


41

3.3.3 Discussion

Figure 3.2 shows two things. First, the increase in consumption of animal

products increases the application of external nitrogen much more in densely

populated areas than in lightly populated areas. Second, the reduction of

availability of arable land per capita increases exponentially the application

rate of external nitrogen in densely populated areas. Thus, the nitrogen

application rate tends to be much higher in densely populated areas than in low

populated areas for a similar consumption of animal products per capita.

It is important to point out that in the model we determine the nitrogen needed

by the crop to produce the food. This deviates from what a farmer should apply

to obtain a certain yield. In agricultural practice, only a part of the fertilizer

applied is taken up by the crop, the remainder is lost to the environment. In our

analysis, we exclude nitrogen losses. So, the amount of nitrogen calculated with

our model is far lower than what is needed in practice.

3.4. Validation model outcomes with country data

In section 3.3, a model without values was developed showing the relations

between the drivers and the nitrogen needed for food. By using country level

data on population density, meat consumption and fertilizer use, we can make

the same graphs as the one in figure 3.2, but now with values. We plot all

combinations (i.e. countries) of population, meat consumption and nitrogen

application in a graph. Since drivers change over time (figure 3.1) we do this for

the years 1960, 1980 and 2010. In the model, three different consumption

patterns are recognized. In here, we distinguish four depending on the amount

of animal products in the pattern: 0-300, 300-600, 600-900 and 900-1200 kcal

of animal products per day. We were able to do this for 110 countries. We also

determined the global average.

3.4.1 Results

The global availability of land decreased from 0.4ha cap-1 in 1960 to 0.3ha cap-1

in 1980 and to 0.2 ha cap-1 in 2010. At the same time, the global nitrogen

application rate and the consumption of animal products increased; the former

from 10 kg N ha-1 in 1960 to 44 kg N ha-1 in 1980 and to 75 kg N ha-1 in 2010;

42

and the latter from 350 kcal cap-1day-1 in 1960 to 390 kcal cap-1day-1 and to 500

kcal cap-1day-1 in 2010. This shows a linear increase in the nitrogen application

rate. However, large variations exist between the global average and the

country level data.

The data of 1960 of figure 3.3 fits with our model. The trend lines for the three

diets show a polynomial relation between availability of land and nitrogen

application similar to the trend lines of our model, and the type of diet parallels

this relation. So, countries with high amount of meat in the diets use more

nitrogen per ha than countries with small amounts of meat in their diets and a

population densities of less than 0.2 ha cap-1. Also, for the vegetarian menus

fertilization is needed (green dots).

The results for 1980 do not fit our model results. Especially the countries with

high meat consumption values (red and brown dots) use far more nitrogen

than the expectations based on the model. The data for 2010, however, fit with

the model again.

The nitrogen application rates in 1960 were generally lower for a certain

population density than in 1980 and in 2010. One of the reasons is that during

this period global population more than doubled: from three billion to seven

billion people (United Nations, 2011) and global arable land only increased

10% (FAO, 2013c). In contrast, global use of nitrogen fertilizer increased eight

times (see table A3.1 in Appendix 3), and crop yields strongly increased which

lead to a doubling of the global food supply (FAO, 2013d). The three billion

people in 1960 were already using the natural nitrogen available in the soil, so

for the new four billion people extra nitrogen in the form of synthetic nitrogen

fertilizer had to be applied in the crop fields in order to produce their food by

increasing the crop yields. This enormous population growth increased the

population density of all countries (reducing the availability of arable land per

capita). This pattern can be seen in figure 3.3 where from 1960 to 2010 all

countries moved to the left side of the graphs and also upwards showing that

the increase of population density increased the nitrogen application rates, in

accordance with our model. The population density increased the most for the

countries with the lowest consumption of animal products (green dots in figure

3.3). These countries are mainly in Africa.


43

Figure 3.3 Country level data validating the model. The colours indicate the

consumption of animal products per capita. The black diamond indicates the global

average for each year. The average caloric intake of animal product for each year is

indicated in the legend.

44

During the 1960-2010 period studied, countries increased their animal

products consumption and changed colour from green towards brown. In 1960,

large differences in diets existed among countries showing large inequalities.

More than two thirds of the global population had diets with only up to 300

kcal of animal products (green group), one fifth had luxurious diets (up to 1200

kcal of animal products, brown group), and few countries had transition diets

(orange group 330-600 kcal of animal products). In 1980, many countries of the

green group had moved towards the orange group and a new group emerged:

the red group. By 2010, the inequality of diets decreased because of the large

change towards luxurious diets in the transition countries. More than half of the

world population had transition diets (red and orange diets), only one third had

poor diets (green group) and one tenth had luxurious diets (brown group).

As mentioned earlier, in 1980 the red and brown countries do not fit in our

model results. The nitrogen application rates of these countries were much

higher for their values of availability of land, and so the data points exceed the

trend lines of our model. This indicates that these countries were using more

nitrogen per hectare than what we expected based on our model, which results

in overuse of nitrogen fertilizer. In 2010, the country level data fits with our

model again. The countries that did not fit with our model in 1980 reduced the

nitrogen application rates by 2010. (See the values of the Netherlands, Greece,

UK, Denmark and Bulgaria in Table A3.1 in Appendix 3). In some of these

countries, strict environmental regulations with respect to nitrogen

applications were put in practice to reduce the environmental problems related

to fertilizer use in agriculture (eutrophication) (Kronvang et al., 2008; Rougoor

et al., 2001; Schnoor et al., 1997). This phenomenon is also shown in figure 3.1c

where Western Europe reduced their application of nitrogen after the 1990’s.

So, our results show that during the last fifty years the increase of both

population density and consumption of animal products has increased the

nitrogen application rates. These numbers are average country level data which

should not be considered as strict values, but they help us to identify the

different drivers to the use of synthetic nitrogen fertilizer and their

interrelationships.

The combination of the population density and the menus makes it possible to

explain the unexpected results in figure 3.1. The fact that North America has

high meat consumption levels but low fertilizer application rates can be

explained from the fact that the population density is low. The fact that


45

population density in Africa is much higher than in North America, while

nitrogen application rates in Africa are much lower can be explained from the

very low meat consumption levels in Africa.

3.5 Consequences of our findings for the future needs for

nitrogen.

An important finding of our analysis is that when population densities are

below the 5 persons per ha arable land [p ha-1] (0.2 ha of arable land per

person) the relation population density and nitrogen demand is more or less

linear, and it becomes exponential with higher values of population density.

The present global population density is 5 p ha-1, and at global level we

observed a linear increase of the nitrogen applications over the last decades. By

2050, the global population is expected to reach 9 billion people. If no major

increase in arable land area is expected, 9 billion people will imply a population

density of 7 p ha-1. (0.15 ha cap-1). This value is already in the exponential part

of the curves in figure 3.3. And so, the need for nitrogen can no longer be

derived from the historical linear trend.

Next to this, the increase of the global population will not be evenly distributed

over the globe. If we divide the projected 2050 population numbers for each

country by the present arable land, it turn out that by 2050 6 billion people (or

70 % of the global population) will live in countries with less than 0.1 ha cap-1

of arable land.

To make things even worse: the countries with the expected population

increase are also the countries with the present low meat diets. It is rather

likely that meat consumption will increase in these regions this will increase

the need for nitrogen even more. This double effect can be recognized in figure

3.1 for China. In this country, both population density and meat consumption

increased and the increase in nitrogen application rate was huge.

The existing estimates with respect to future needs for nitrogen fertilizer

involve doubling of the present use. This doubling is considered as a serious

treat for the global environment (Tilman et al., 2011). Our analysis shows that

the existing nitrogen estimates for future are too optimistic, and it is more

likely that future use of nitrogen will have far higher values resulting in very

large impacts on global and regional environments.

46

Appendix 3

Table A3.1

Country level data of population, availability of arable land, nitrogen application rate and consumption of animal products. The countries are ordered in relation to the availability of arable land. These values are plotted in figure 3.3. The countries written with font bold italics are not shown in figure 3.3 because these countries have larger values than the axis values of figure 3. However, they are included in the model simulation of figure 3.3 (the trend lines).

Table A3.1.1 1960 Population

[Million people]

Availability of land

[ha cap-1]

Nitrogen application rate

[kg N ha-1 ]

Animal products consumption

[kcal cap-1 day-1]

Japan 94 0.06 117.5 270

Sri Lanka 10 0.06 60.1 101

Switzerland 6 0.07 50.9 1070

Korea, Rep. 26 0.08 102.0 59

Netherlands 12 0.08 284.3 983

Egypt, Arab Rep. 29 0.08 83.7 143

Lebanon 2 0.09 44.6 345

Malaysia 9 0.10 30.8 258

United Kingdom 53 0.14 74.8 1236

Israel 2 0.14 65.5 498

China 691 0.15 7.5 71

Viet Nam 37 0.15 2.3 133

Bangladesh 53 0.16 3.1 64

Germany 74 0.16 80.5 949

El Salvador 3 0.17 39.6 215

Philippines 28 0.18 9.3 206

Nepal 10 0.18 0.2 149

Indonesia 97 0.19 5.2 52

Peru 11 0.19 33.0 263

Costa Rica 1 0.20 42.2 290

Cuba 7 0.20 33.5 432

Dominican Republic 4 0.21 11.2 220

Colombia 17 0.21 9.4 363

Norway 4 0.23 65.1 1046

Austria 7 0.24 36.8 1036

Ghana 7 0.24 0.3 102

Guatemala 4 0.25 8.9 173

Italy 50 0.25 28.7 485

Albania 2 0.25 5.5 376


47


[Million people]


[ha cap-1]


[kg N ha-1 ]


[kcal cap-1 day-1]

Saudi Arabia 4 0.27 1.7 137

Jordan 1 0.28 2.4 134

Portugal 9 0.28 28.0 347

Mali 5 0.31 0.0 199

Mozambique 8 0.31 1.1 58

Brazil 77 0.31 2.3 290

India 466 0.34 2.0 111

Greece 8 0.34 34.0 400

Pakistan 48 0.35 3.0 302

Paraguay 2 0.35 0.1 441

Venezuela 8 0.36 3.7 333

Madagascar 5 0.36 0.6 294

Ecuador 5 0.36 3.6 319

Thailand 29 0.37 1.3 171

Panama 1 0.37 19.0 380

Bolivia 4 0.37 0.3 299

Malawi 4 0.37 1.7 58

Benin 2 0.38 0.0 71

Kenya 9 0.41 1.3 234

France 47 0.42 35.8 1028

Cote d'Ivoire 4 0.43 1.2 124

Uganda 7 0.44 0.4 143

Sweden 8 0.45 34.7 995

Chile 8 0.45 5.8 423

Zimbabwe 4 0.48 8.7 204

Congo, Rep. 1 0.49 0.6 110

U. Rep. Tanzania 11 0.51 0.2 138

Uruguay 3 0.51 4.6 1129

Hungary 10 0.52 23.4 942

Bulgaria 8 0.52 18.1 431

Spain 31 0.52 20.6 377

Romania 19 0.53 13.6 421

Poland 30 0.53 20.4 918

Morocco 12 0.54 2.0 157

Mexico 41 0.55 7.5 282

Nicaragua 2 0.55 5.3 278

Nigeria 48 0.55 0.0 50

Ireland 3 0.55 20.6 1221

Algeria 11 0.56 2.2 168

48


[Million people]


[ha cap-1]


[kg N ha-1 ]


[kcal cap-1 day-1]

Finland 4 0.60 24.8 1305

Denmark 5 0.60 51.5 1106

Honduras 2 0.61 4.7 207

Iran, Islamic Rep. 23 0.65 0.6 209

South Africa 18 0.66 5.8 406

Tunisia 4 0.71 1.4 161

Turkey 30 0.79 2.0 459

Zambia 3 0.79 1.9 127

Cameroon 6 0.89 0.3 90

Sudan 12 0.90 2.0 299

Argentina 21 0.90 0.6 973

Senegal 3 0.92 0.9 185

USA 192 0.93 19.7 1006

Guinea 4 0.96 0.1 47

Libya 1 1.18 1.2 166

New Zealand 2 1.21 1.3 1252

Syrian Arab Republic 5 1.28 1.5 267

Canada 19 2.21 3.0 1078

Australia 11 2.97 1.6 1254


49


[Million people]


[ha cap-1]


[kg N ha-1 ]


[kcal cap-1 day-1]

U. Arab Emirates 1 0.02 142.9 992

Japan 116 0.04 139.2 539

Egypt, Arab Rep. 45 0.05 237.6 206

Korea, Rep. 37 0.05 214.1 229

Netherlands 14 0.06 612.0 1170

Sri Lanka 15 0.06 91.6 111

Switzerland 6 0.06 166.7 1292

Malaysia 14 0.07 134.2 405

Lebanon 3 0.08 73.3 461

Israel 4 0.09 113.9 620

China 1006 0.10 117.5 170

Viet Nam 54 0.11 19.8 114

Philippines 47 0.11 42.3 246

Bangladesh 81 0.11 28.5 63

El Salvador 5 0.12 100.8 265

Indonesia 151 0.12 45.7 72

Costa Rica 2 0.12 147.3 428

United Kingdom 56 0.12 190.8 1141

Jordan 2 0.13 17.1 298

Colombia 27 0.14 40.0 348

Rwanda 5 0.15 0.1 56

Nepal 15 0.15 7.4 152

Germany 78 0.15 182.9 1129

Italy 56 0.17 109.1 827

Yemen 8 0.17 8.7 218

Ghana 11 0.17 4.6 100

Guatemala 7 0.18 41.4 160

Dominican Rep. 6 0.18 29.5 294

Peru 17 0.19 28.4 267

Saudi Arabia 10 0.19 14.6 438

Venezuela 15 0.19 32.1 473

Ecuador 8 0.20 25.2 355

Norway 4 0.20 133.7 1211

Austria 8 0.20 104.0 1186

Albania 3 0.22 120.8 399

Panama 2 0.22 28.4 438

Burundi 4 0.22 0.6 66

Cote d'Ivoire 9 0.23 8.1 185

50


[Million people]


[ha cap-1]


[kg N ha-1 ]


[kcal cap-1 day-1]

India 700 0.23 23.0 118

Kenya 16 0.23 7.4 226


Portugal 10 0.25 60.1 505

Pakistan 81 0.25 40.9 287

Gambia, The 1 0.26 4.9 120

Congo, Rep. 2 0.27 0.3 122

Mali 7 0.28 2.9 229

Madagascar 9 0.30 1.4 261

Greece 10 0.30 121.4 684

Cuba 10 0.30 96.3 721

Malawi 6 0.32 10.6 93

Uganda 13 0.32 0.0 145

France 54 0.32 123.6 1324

Ireland 3 0.32 240.0 1396

Mexico 69 0.33 41.2 517

Nicaragua 3 0.34 26.4 259

Chile 11 0.34 13.6 430

Zimbabwe 7 0.34 33.3 201

Thailand 47 0.35 8.5 198

Sweden 8 0.36 84.0 1065


Bolivia 5 0.36 1.0 437

Algeria 19 0.36 9.8 281

Nigeria 76 0.37 3.0 105

Brazil 122 0.37 17.4 388

Burkina Faso 7 0.38 0.8 87

Morocco 20 0.39 11.7 192

Zambia 6 0.40 21.5 126

Honduras 4 0.41 9.1 246

Poland 36 0.41 88.4 1154

Benin 4 0.42 0.4 97

Spain 37 0.42 56.3 714

South Africa 29 0.43 37.4 409

U Rep. Tanzania 19 0.43 2.6 136

Bulgaria 9 0.43 120.2 738

Romania 22 0.44 78.5 752

Hungary 11 0.47 110.6 1231

Uruguay 3 0.48 15.1 1050


51


[Million people]


[ha cap-1]


[kg N ha-1 ]


[kcal cap-1 day-1]

Finland 5 0.50 81.2 1309

Tunisia 6 0.50 8.3 243

Paraguay 3 0.51 1.0 482

Denmark 5 0.52 144.5 1172

Libya 3 0.57 12.9 590

Senegal 5 0.58 2.3 172

Turkey 44 0.58 29.9 448

Syrian Arab Rep. 9 0.59 15.4 450

Sudan 20 0.62 5.1 392

Cameroon 9 0.65 2.6 121

Guinea 4 0.70 0.2 70

Togo 3 0.73 0.6 70

USA 230 0.82 55.0 962

New Zealand 3 0.83 8.2 1282

Argentina 28 0.92 2.2 1055

Niger 6 1.70 0.1 219

Canada 25 1.82 20.4 988

Australia 15 2.97 5.5 1046

52


[Million people]


[ha cap-1]

Nitrogen application rate [kg

N ha-1 ]


[kcal cap-1 day-1]

U. Arab Emirates 7 0.01 591.5 717

Jordan 6 0.03 88.0 409

Korea, Rep. 48 0.03 162.9 510

Egypt, Arab Rep. 80 0.03 518.3 299

Japan 127 0.03 114.4 571

Lebanon 4 0.03 86.5 504

Colombia 46 0.04 274.7 518

Israel 7 0.04 123.3 794

Costa Rica 5 0.04 327.2 608

Bangladesh 147 0.05 152.1 94

Switzerland 8 0.05 126.5 1151

Yemen 23 0.05 14.5 172

Sri Lanka 21 0.06 161.5 154

Philippines 92 0.06 90.0 393

Netherlands 17 0.06 211.6 1066

Malaysia 28 0.06 530.4 495

Viet Nam 87 0.07 192.3 508

Chile 17 0.07 303.3 742

China 1366 0.08 320.7 668

Nepal 29 0.08 6.9 162

Dominican Republic 10 0.08 56.0 410

Ecuador 14 0.08 128.6 553

Slovenia 2 0.09 151.0 936

Venezuela 29 0.09 103.9 539

Indonesia 237 0.10 122.8 158

United Kingdom 62 0.10 163.1 1012

New Zealand 4 0.10 551.0 998

Guatemala 14 0.10 79.8 256

Georgia 4 0.11 35.5 475

Portugal 11 0.11 94.3 1072

El Salvador 6 0.11 89.3 396

Tajikistan 7 0.11 65.4 201

Burundi 8 0.11 1.1 48

Italy 60 0.12 77.0 941

Pakistan 171 0.12 154.5 498

Rwanda 10 0.12 0.9 73

Saudi Arabia 27 0.12 47.8 397 Congo, Rep. 4 0.13 0.9 132

Peru 29 0.13 62.9 278


53


[Million people]


[ha cap-1]


N ha-1 ]


[kcal cap-1 day-1]

India 1208 0.13 99.1 206

Eritrea 5 0.13 0.4 107

Kenya 39 0.14 13.2 283

Honduras 7 0.14 41.5 418

Cote d'Ivoire 19 0.14 7.9 102

Germany 82 0.14 137.2 1086

Madagascar 20 0.15 1.9 163

Armenia 3 0.15 23.3 667

Panama 3 0.16 33.1 550

Uzbekistan 27 0.16 148.7 458

Austria 8 0.16 60.6 1127

Ethiopia 81 0.17 9.3 97

Norway 5 0.17 103.4 1109

Ghana 24 0.18 4.2 135

Albania 3 0.19 51.4 858

Croatia 4 0.20 196.0 839

Uganda 32 0.20 1.3 175

Azerbaijan 9 0.21 11.1 417

Greece 11 0.21 62.5 844


Algeria 35 0.21 4.0 338

Thailand 69 0.22 91.5 330

Mexico 112 0.22 42.4 645

U. Rep. Tanzania 44 0.22 5.8 141

Gambia, The 2 0.22 3.5 165


Malawi 14 0.23 24.4 78

Nigeria 155 0.23 2.5 106

Syrian Arab Rep. 20 0.23 42.5 474

Kyrgyzstan 5 0.24 18.9 578

Zambia 13 0.24 20.9 101

Ireland 4 0.25 320.5 1023

Slovakia 5 0.25 61.5 734

Morocco 32 0.25 25.1 267

Tunisia 10 0.27 24.7 356

Bosnia&Herzegovina 4 0.27 17.3 518

Spain 46 0.27 65.5 860

Libya 6 0.28 26.2 392

Sweden 9 0.28 62.9 1058

54


[Million people]


[ha cap-1]


N ha-1 ]


[kcal cap-1 day-1]

Senegal 12 0.29 3.6 198

Guinea 10 0.29 0.8 97

South Africa 50 0.29 28.5 466

Benin 9 0.29 0.8 112

France 62 0.29 109.9 1215

Turkey 72 0.30 60.0 427

Czech Republic 10 0.31 73.5 877

Cameroon 19 0.31 3.3 121

Brazil 193 0.32 42.5 714

Cuba 11 0.32 12.5 424

Poland 38 0.33 90.9 919

Zimbabwe 12 0.33 15.2 189

Nicaragua 6 0.34 21.1 296

Burkina Faso 16 0.36 5.8 141

Bolivia 10 0.37 4.3 408

Togo 6 0.38 0.3 86

Mali 15 0.40 9.8 255

Romania 22 0.40 33.8 905

Bulgaria 8 0.41 93.6 611

Finland 5 0.42 89.8 1230

Denmark 6 0.43 81.8 1272

Sudan 42 0.45 2.4 502

Hungary 10 0.46 61.8 1021

Uruguay 3 0.48 71.5 594

Republic of Moldova 4 0.51 9.3 497

USA 308 0.53 67.8 1030

Belarus 10 0.57 96.9 807

Lithuania 3 0.57 32.0 1075

Paraguay 6 0.59 17.8 490

Ukraine 46 0.71 22.5 672

Argentina 40 0.80 22.4 938

Russian Federation 143 0.85 10.0 770

Niger 15 1.00 0.2 260

Canada 34 1.34 42.9 903

Kazakhstan 16 1.45 1.0 886

Australia 22 2.06 21.6 1037


55

Table A3.2 Global changes in the drivers: Population, nitrogen fertilizer and arable land

over 50 years

Availability of arable land

Year TOTAL

Low 0.0-0.2 ha cap-1

Low medium 0.2-0.4 ha cap-1

Medium high 0.4-0.6 ha cap-1

High 0.6-1.5 ha cap-1

Number of countries

1960 85 19 28 21 17

1980 91 30 35 19 7

2010 109 50 40 13 6

Population [Million people]

1960 2 691 1 247 792 315 337

1980 3 861 1 901 1 403 260 297

2010 6 299 4 289 1 275 441 293

Nitrogen Fertilizer [Mton]

1960 11 4 1 2 4

1980 48 20 11 6 11

2010 99 67 14 13 5

Arable Land [Mha]

1960 887 177 255 160 294

1980 951 197 392 122 241

2010 1 261 429 338 224 269

This table shows the aggregated number of countries, population, nitrogen fertilizer and arable land of the data points of figure 3.3. The data points are grouped into four groups based on the availability of arable land (low, low-medium, medium-high and high). This table shows two main findings. First, in the period 1960-2010 not only more countries moved to densely populated areas (number of data points) but also the total number of people increased in densely populated areas. In 1960, “only” 46% of global population were living in countries with less than 0.2 hectares of arable land per capita and in 2010 it increased to 65% of global population. Second, the use of nitrogen fertilizer became more unequally distributed. In 1960, the share of nitrogen fertilizer use in relation to the share of arable land was more equally distributed than in 2010, see the percentage values of nitrogen fertilizer and arable land in 1960. In 2010, 68% of global use of nitrogen fertilizer is used in only 34% of the global arable land which are the densely populated countries with less than 0.2 hectares per capita. This shows the large intensification of agriculture in densely populated areas.

Chapter 4. Nitrogen fertilizer use per person and its trade-off with land use

57

Chapter 4.

Nitrogen fertilizer use per person and its trade-off with

land use: An international comparison of agricultural

production systems and diets*

Abstract

Nitrogen fertilizer will play a vital role in future global food supply. The amount

needed depends on the production system used, on the number of people to be

fed, and on their type of diets. In this paper, we determine nitrogen fertilizer

use per capita for five combinations of production systems and diets. Varying

from the low input production systems in Africa supplying staple food diets to

the high input production systems in Europe supplying affluent diets with large

consumption of animal products. We also calculated land use per person for

these production systems and diets because of the strong link between the use

of synthetic nitrogen fertilizer and crop yields and, therefore, with land use for

food.

The global differences are enormous: from 3 to 30 kilograms of nitrogen

fertilizer per person, and from 1800 to 4500 m2 per person. The methodology

used in this paper allowed us to study production systems and diets

individually. The analysis showed that diets have a large impact: in all

production systems an affluent diet with large meat consumption uses at least 4

times as much nitrogen fertilizer and 4 times as much land than a staple food

diet. In the coming decades, global population will increase and diets will

change to more affluent diets. Based on our analysis, the future needs of

synthetic nitrogen fertilizer will be far larger than the ones expected up to now.

* Resubmitted version under review in Industrial Ecology as: M.J. Ibarrola Rivas & S.

Nonhebel. Nitrogen fertilizer use per person and its trade-off with land use: An international comparison of agricultural production systems and diets

58

4.1 Introduction

Mineral nitrogen fertilizer plays a vital role in global food supply. At the end of

the 20th century, the food for half of the world’s population was produced with

it (Smil, 2001). However, nitrogen fertilizer gives rise to strong environmental

impacts (Sutton et al., 2013). So, it has a controversial trade-off between

supplying food and associated environmental impacts. In this paper, we refer to

mineral nitrogen fertilizer only as nitrogen fertilizer.

The nitrogen fertilizer use is different among food products. The difference

depends on the way the food is produced (the agricultural system) and on the

type of food itself. For instance, some crops are produced in agricultural

systems using large nitrogen application rates resulting in high crop yields

(Engels & Marschner, 1995). In other agricultural systems, the same crop is

produced with almost no nitrogen fertilizer, but with low crop yields resulting

in a larger use of land per amount of crop (Pimentel, 2009). This close relation

between nitrogen fertilizer application and crop yields indicate a trade-off

between nitrogen fertilizer and land use.

In addition, the type of food makes a difference in the requirements of nitrogen.

In general, animal products require more nitrogen per amount of food

produced than vegetable products; but within these categories differences exist

(Pierer et al., 2014). Legumes require less nitrogen per kilogram of crop than

other crops, and milk requires less nitrogen per kilogram of food than meat. So,

the production system and the type of food have a large impact on the use of

nitrogen fertilizer.

In the coming decades, global food demand is projected to increase due to

population growth and the change to more affluent diets (Godfray et al., 2010a;

Godfray et al., 2010b); and nitrogen fertilizer will be essential to assure global

food supply (Sutton et al., 2013; Erisman et al., 2008). The land available per

person for food production will be reduced (Alexandratos & Bruinsma, 2012,

fig 4.3). As a result, food production systems are expected to change especially

in developing countries mainly to increase crop yield (Alexandratos &

Bruinsma, 2012). In this paper, we investigate how the type of agricultural

system and the type of food consumption drive the per capita use of nitrogen

fertilizer and land in order to assess future global use of nitrogen fertilizer and

its trade-off with land use.


59

There is a wide range of literature on the environmental impacts of nitrogen

related to food consumption. Several global studies have calculated the

nitrogen emissions and nitrogen flows throughout the world and have

discussed the differences among regions (Beusen et al., 2008; Bouwman et al.,

2009; Bouwman et al., 2011; Cui et al., 2013; de Vries et al., 2013; Eickhout et

al., 2006; Galloway & Cowling, 2002; Galloway et al., 2008; Liu et al., 2010;

Shindo et al., 2006; Smil, 1999; Xiong et al., 2008). However, these studies do

not link the emissions with the consumption patterns of the population.

Moreover, some studies have calculated the nitrogen emissions per capita for

several countries (Billen et al., 2012; Bleken & Bakken, 1997; Chatzimpiros &

Barles, 2013; Howarth et al., 2002; Isermann & Isermann, 1998; Lassaletta et

al., 2013; Ma et al., 2010; Pierer et al., 2014), what some call a “Nitrogen

Footprint”. These studies do link the nitrogen emissions with the food

consumption patterns. But, they do not discuss global differences among

production systems and diets since they only study one country or two

countries with similar diets in respect to the global differences (USA and the

Netherlands, see Leach et al., 2012).

Most of the literature mentioned above focus on the total nitrogen emissions

(Nr) into the environment resulting in large environmental problems i.e. local

pollution and affecting the global nitrogen cycle. In contrast, this paper does a

global analysis of only the use of nitrogen fertilizer per capita because of its

vital role for future global food security.

The aim of this study is to understand how global differences in production

systems and diets drive the use of nitrogen fertilizer use per person and its

trade-off with land use. We use examples of diets and agricultural systems of

different countries which illustrate the global variations. We analyse how the

agricultural factors (nitrogen fertilizer application rate and crop yields) and the

consumption factors (amount of food consumed per person) drive the per

capita use of nitrogen fertilizer and land. The results are used to discuss future

use of nitrogen fertilizer and land for a population changing both diets and

agricultural production systems.

60

4.2 Methodology

The footprint studies in the literature (Hoekstra & Mekonnen, 2012; Kastner et

al., 2012; Leach et al., 2012; Pierer et al., 2014) calculate the use of agricultural

resources per capita. Their methodology is useful to discuss the impact of both

production and consumption on the use of resources by linking food

consumption data with crop production data. In this paper, we follow a similar

methodology to calculate the per capita use of nitrogen fertilizer and arable

land. So, we link food consumption data: kilograms of food per person per year,

with agricultural production data: nitrogen fertilizer use per tonne of food &

arable land per tonne of food. We used national level data of five countries

(France, USA, Mexico, the Philippines and Tanzania) which illustrate the global

differences of diets and production systems.

For the consumption data, we used food supply data from the Food Balance

Sheets (FAO, 2013d). The food supply is not the actual food consumption of the

population since it includes food losses. A more accurate source of diets is

household level surveys. But these surveys are only available on a national

level, and do not include global data. However, Gerbens-Leenes et al. (2010, fig

7) show a clear relation between food supply data and household levels

surveys. Throughout the paper, we refer to diets when discussing food supply

data.

For the production data, we calculated the nitrogen fertilizer use per ton of

food by combining nitrogen application rates (kg N/ha) with crop yield data

(ton/ha); and the arable land per kilogram of food with the inverse of the crop

yield. We used crop yield data from the FAOSTAT (FAO, 2013b); however,

FAOSTAT does not provide data of nitrogen application which in general is local

data that is not widely available for a global scale. FertiStat (FAO, 2007) has

national average data for some countries and some crops at a certain year. We

used their data and linked it with the crop yield given by FAOSTAT. To validate

this assumption of combining two different databases, we compare our

calculations with some studies that provide crop field scale data of both

nitrogen application and crop yield, see Appendix 4.

Fertistat has data of fertilizer use per crop for 10-15 crops for 92 countries. For

some crops, only a share of the harvested area of the country is fertilized, and

FertiStat indicates this percentage. Since we link this application rate with the

national crop yield given by FAOSTAT, we multiply the application rate by the


61

share of the area, resulting in a weighted nitrogen application rate. For

instance, the nitrogen application rate of maize production in Tanzania in 1997

was 80 kg N/ha, but only 10% of the harvested area was fertilized. So, the value

of nitrogen rate that we used is 8 kg N/ha (80 kg N/ha*10%). This is a

limitation of our methodology since the weighted application rate is not really

linked with the national crop yield. However, using the original application rate

(80 kgN/ha) with the national crop yield would lead to a strong overestimation

of nitrogen use especially for crops which their share of fertilized area is very

small. This assumption is validated in Appendix 4 in which we show that our

calculations are in the same order of magnitude as crop field studies with

accurate nitrogen application rates data.

We linked agricultural production data with consumption data for the average

diet of each country in 2009 (FAO, 2013d). The diet consists of several food

items and each one is produced in different systems. Simplification of the diet is

needed to identify the factors driving the use of nitrogen fertilizer and land.

Therefore, we used data of the 14 main food categories given by the Food

Balance Sheets (FAO, 2013d) to represent the diet of each country. The food

categories are: cereals, roots, sugars, pulses, vegetable oils, vegetables, fruits,

alcoholic beverages, milk, eggs, beef, pork, mutton meat and poultry meat. The

food items included in each food category have similar agricultural production

characteristics. For instance, cereals (wheat, rice and maize) in general require

similar nitrogen application rates and obtain similar crop yields in comparison

with roots (cassava, sweet potatoes and potatoes).

For the vegetable products, one crop was chosen to represent each food

category. We choose the crop of each food category with the largest harvested

area according to FertiStat. For instance, for Mexico FertiStat gives data of

maize, rice and wheat production, maize has the largest harvested area so we

chose maize to represent cereals. We converted the food supply data into its

crop equivalent using conversion factors from Kastner & Nonhebel (2009,

Table A1). Note that not all vegetable food categories have to be converted. For

this paper, only sugars, alcoholic beverages and vegetable oils were converted.

We divided the nitrogen application rate (kgN/ha) by the crop yield (ton/ha) to

come up with the amount of nitrogen fertilizer use per amount of crop (kg N

/ton crop). We multiplied the nitrogen fertilizer per amount of crop by the crop

equivalent of the food supply. Then, we multiplied the data of production (kg of

nitrogen fertilizer per ton of crop) by the data of food supply (kg food per

person) for every food category in the diet and divided it by 1000 to convert

62

tonnes into kilograms. This resulted in the use of nitrogen fertilizer (kg

Nitrogen fertilizer per person). Finally, we aggregated the use of nitrogen

fertilizer of all food categories to arrive at the total nitrogen fertilizers use per

person.

For animal products, we calculated it separately for bovine meat, pig meat,

poultry meat, mutton meat, milk and eggs; and we excluded fish and animal

fats. The nitrogen and land use for each product is the indirect use for the feed

crop. First, we calculated the nitrogen fertilizer per ton for the feed crop. We

assumed that all animal products in all countries are produced with the same

feed crop: maize, which is the feed crop most widely used globally (FAO,

2013b). The amount of meat, milk or eggs produced with this feed depends on

the feed-food efficiency factor (see table 4.2). The feed-food efficiency factor is

the ratio of kilogram of feed crop needed to produce a kilogram of meat, milk or

eggs. We multiplied the kilogram of nitrogen fertilizer per amount of feed

produced with the feed-food efficiency factor to calculate the nitrogen fertilizer

per amount of animal food (kg N/ton). Then, we multiplied it by the food supply

of the animal food item. We assumed the same feed-food efficiency factors for

all countries (kg dry mass feed/kg output). We used the values for the global

average and Industrial system given by Mekonnen & Hoekstra (2012), see table

4.2. This is a simplified assumption since feed efficiencies vary a lot among

regions and production systems; but in this way, we avoid adding extra

variables to the calculation of the nitrogen fertilizer requirement and we can

identify the role of each animal food item.

To calculate the land needed per person, we divided the food supply data in its

crop equivalent (kg /cap/yr) by the crop yield (ton/ha) for all vegetable

products. The crop yield data was converted from tonnes into kilograms and

hectares into m2. For the animal products, we multiply the food supply

(kg/cap/yr) with the feed-food efficiency factor (table 4.2) and divided it by the

crop yield of the feed. By doing so, we calculated the arable land per person.

Throughout the paper, we refer to arable land when discussing the trade-off

between nitrogen fertilizer and land.

4.3 Results

In this section, we show the results of the nitrogen fertilizer and land use per

person for different production systems and diets. First, we show the data of


63

the production systems. We discuss the differences among countries and food

categories, and how the different factors (nitrogen application, crop yield, feed-

food efficiency factor) drive the nitrogen fertilizer and land use per kilogram of

food for each food category. Second, we discuss the differences in food

consumption patterns among the five countries studied in this paper. Third, we

discuss the nitrogen fertilizer use per person and its trade-off with land. Finally,

we discuss the individual impact of the production systems and diets on the

nitrogen fertilizer use per person and its trade-off with land use.

Production systems:

Nitrogen fertilizer and land use per kilogram of food

We gathered data of nitrogen fertilizer application rate in kg N/ha and crop

yield in ton/ha for each food category of the vegetable products of the diet and

for the animal feed. With this data, we calculated the nitrogen fertilizer and

land use per amount of food produced (see table 4.1). The nitrogen application

rates and the crop yields are different among the countries and also among the

food categories.

In general, USA and France have both the largest values of nitrogen application

rates and crop yields, Tanzania have the lowest, and the Philippines and Mexico

have middle values. Though, the nitrogen fertilizer per amount of food

produced does not follow this pattern (column 5 of table 4.1). The largest

nitrogen application does not necessarily result in largest nitrogen fertilizer use

per kilogram of food because it has to be weighted with the crop yield. For

instance, the maize production in Mexico has smaller nitrogen application and

crop yield than the wheat production in France. But the nitrogen fertilizer per

amount of maize produced in Mexico is more than two times larger than the

wheat production in France. The reason is the relation between nitrogen

application and crop yield. With only 30% higher nitrogen application, the crop

yield obtained in France is more than 3 times higher than in Mexico. Similar

relations have the sunflower seed, barley and maize production among these

two countries.

The production systems of Tanzania result in much lower nitrogen use per

kilogram of food due to the very low nitrogen application rates. In addition, the

land use per kilogram of food is generally much larger due to the low crop

64

yields obtained. So, a clear trade-off is shown between nitrogen and land use

per kilogram of food produced.

Table 4.1 Production systems data for the vegetable products and feed.

Data sources: 1FertiStat (FAO, 2007) 2FAOSTAT (FAO, 2013b) 3calculations from the

authors, see the methodology section for details.

Crop

N application [kg N/ha]1

Crop yield [ton

crop/ha]2

kg N/ton crop3

kg N/ton food3

m2/ton crop3

m2/ton food3

Cereals

France Wheat 80 7 11 11 1 405 1 405

USA Wheat 63 3 22 22 3 444 3 444

Mexico Maize 60 2 26 26 4 268 4 268

Philippines Rice, paddy 43 3 14 18 3 138 4 048

Tanzania Maize 8 1 7 7 8 541 8 541

Roots

France Potatoes 35 40 1 1 253 253

USA Potatoes 209 38 5 5 260 260

Mexico Potatoes 108 20 5 5 488 488

Philippines potato 68 12 5 5 806 806

Tanzania Sweet potato 1 2 0.3 0.3 5 447 5 447

Pulses

France Beans, dry 150 3.0 49 49 3 283 3 283

USA Beans, dry 64 1.8 36 36 5 625 5 625

Mexico Beans, dry 17 0.6 28 28 17 027 17 027

Philippines Beans, dry 8 0.7 10 10 13 376 13 376

Tanzania Pulses 1 0.7 1 1 14 531 14 531

Sugar & Sweeteners

France Sugar beet 145 76 1.9 10 132 702

USA Sugar beet 120 50 2.4 13 199 1 061

Mexico Sugar cane 90 78 1.2 14 129 1 603

Philippines Sugar cane 68 70 1.0 12 143 1 782

Tanzania Sugar cane 1 87 0 0 116 1 436


65

Crop

N application [kg N/ha]1

Crop yield [ton

crop/ha]2

kg N/ton crop3

kg N/ton food3

m2/ton crop3

m2/ton food3

Vegetable Oils

France Sunflower seed

130 2.5 52 148 3 974 11 407

USA Soybeans 21 2.6 8 21 3 821 10 088

Mexico Sunflower seed

60 0.8 79 227 13 154 37 753

Philippines Coconuts 6 4.2 1 7 2 395 11 497

Tanzania Seed cotton 9 0.5 19 67 21 299 74 334

Vegetables

France Vegetables 45 23 2 2 440 440

USA Vegetables 171 26 7 7 390 390

Mexico Vegetables 42 16 3 3 634 634

Philippines Vegetables 0.025 8 0 0 1 179 1 179

Tanzania Vegetables 10 6 2 2 1 540 1 540

Fruits

France Fruits 50 11 5 5 915 915

USA Fruits 93.5 24 4 4 413 413

Mexico Fruits 40 11 4 4 882 882

Philippines Fruits 38 11 3 3 886 886

Tanzania Fruits 1 3 0.3 0.3 3 224 3 224

Alcoholic Beverages

France Barley 120 6 19 3 1 580 237

USA Barley 60 3 19 3 3 095 464

Mexico Barley 32 2 21 3 6 513 977

Philippines Rice, paddy 43 3 14 2 3 138 549

Tanzania Barley 1 2 1 0 6 000 1 080

Feed

France maize 170 9 19 1 102

USA maize 150 8 18

1 185

Mexico maize 60 2 26 4 268

Philippines maize 46 2 25

5 495

Tanzania maize 8 1 7 8 541

66

A pattern can be recognized in the values of nitrogen and land use per kilogram

of crop among the food categories. Vegetables, fruits, potatoes and sugars have

the lowest values. These categories show one order of magnitude lower values

of nitrogen and land use per kilogram of crop produced than pulses and cereals.

Though, note that for sugars the nitrogen and land use increases ten times in

the conversion of the crop equivalent (sugar cane or sugar beet) into its food

category (sugar), which is the food item consumed in the diet. The production

of oil crops for vegetable oils production show large variation among the

countries due to large differences in nitrogen application and crop yield. Note

that the crops chosen to represent this food category are also wide differently:

from soybean which is a legume that generally do not require large nitrogen

application, to coconuts, cotton and sunflower.

For the animal products, the amount of nitrogen fertilizer and land use per

kilogram of food was calculated using the data of the feed (table 4.1) and the

feed-food efficiency factor (table 4.2). The nitrogen fertilizer and land use per

kilogram of food for the animal products, especially for the meat products, is

much higher than the ones for the vegetable products.

The difference in the amount of both nitrogen and land per kilogram of food

among the animal products is related with the feed-food efficiency factor. Beef

is the product with the largest feed-food efficiency factor and it requires the

largest amount of both nitrogen and land use per kilogram of food, followed by

mutton and goat meat, pork, poultry, eggs and milk.

The difference among the countries in the values of the amount of nitrogen and

land use per kilogram of food is related with the feed shown in table 4.1.

Tanzania uses the largest amount of land and the lowest amount of nitrogen

fertilizer per kilogram of food due to the low nitrogen application and low crop

yield of maize. The nitrogen application of France and the USA is much higher

than the ones of the Philippines and Mexico, the former obtaining much higher

crop yields than the latter. As a result, the nitrogen and land use per kilogram of

maize in the USA and France is lower than in the Philippines and Mexico. This

shows, once again, that larger nitrogen application rates do not necessarily

result in larger nitrogen use per kilogram of food depending on the crop yield

obtained.


67

Table 4.2 Production data for the animal food products

Data sources: 1Calculations from the authors 2Mekonnen & Hoekstra (2012,Appendix 1)

kg N/ton food1

m2/ton food1

kg N/ton food1

m2/ton food1

Beef Poultry

France 356 20 932 France 52 3 085

USA 338 22 517 USA 50 3 318

Mexico 487 81 096 Mexico 72 11 951

Philippines 484 104 407 Philippines 71 15 386

Tanzania 130 162 282 Tanzania 19 23 915

Mutton & goat meat Milk

France 249 14 653 France 21 1 212

USA 236 15 762 USA 20 1 304

Mexico 341 56 767 Mexico 28 4 695



Pork Eggs

France 73 4 297 France 43 2 534

USA 69 4 622 USA 41 2 726

Mexico 100 16 646 Mexico 59 9 817



feed-food efficiency factors in kg feed/ kg weight carcass or milk and eggs2

Bovine Meat 19.0

Mutton meat 13.3

Pork 3.9

Poultry Meat 2.8

Milk 1.1

Eggs 2.3

68

Food consumption patterns

The nitrogen fertilizer and land use per person depends not only on the

production system (table 4.1 and 4.2) but also on the amount of food consumed

(table 4.3). We combined the data of production shown in tables 4.1 and 4.2

with the food supply per capita of each country in 2009. The data is shown in

table 4.3. Large variation exists among the countries. Tanzania consumes larger

amount of staple food such as roots and pulses than the rest of the countries.

And USA and France consumes larger amount of affluent food products such as

sugars, vegetable oils, vegetables and animal products followed by Mexico, the

Philippines and Tanzania.

Table 4.3. Food consumption data for each country

Source of data: Food Supply data of 2009 from FBS (FAO, 2013d)

kg/cap/yr France USA Mexico Philippines Tanzania

Cereals 120 108 162 154 105

Roots 53 57 16 31 162

Pulses 2 5 12 2 20

Sugars 41 64 52 23 10

Vegetable oils 22 28 10 5 7

Vegetables 93 123 57 62 34

Fruits 115 111 109 122 77

Alcohol 87 98 61 15 63

Beef 26 40 17 4 7

Mutton meat 3 1 1 1 1

Pork 31 30 15 18 0

Chicken 22 49 30 10 1

Milk 247 256 113 13 38

Eggs 14 14 18 4 1

The consumption of the animal products is relevant for the nitrogen fertilizer

and land use per person due to the larger requirement of resources than the

vegetable products as shown in tables 4.1 and 4.2. The USA has the largest


69

consumption of animal products in kg/cap/yr followed by France, Mexico, the

Philippines and Tanzania. The composition of the food categories in the diet is

different. In general, the consumption of milk is the largest among the animal

products for all countries, except for the Philippines, followed by beef, chicken

and pork, then eggs, and the consumption of mutton meat is the lowest. The

importance of beef, chicken and pork in the diet is different among the

countries (see table 4.3). For instance, in France pork consumption is larger

than chicken and in the USA chicken consumption is larger than pork.

Nitrogen fertilizer use per person and its trade-off with land use

The combination of the production data (table 4.1 and 4.2) and the

consumption data (table 4.3) results in the nitrogen fertilizer and land use per

person for each country (see figure 4.1). Large differences exist among the

countries due to the differences in both production and consumption data

discussed above. The order of the countries, left to right, is in accordance to the

type of diet: from affluent to staple, and in accordance with the production

system: from large to low nitrogen application. Though, for the nitrogen

application, this trend is not followed by all the food categories and, in some

cases, a country in the right side might have larger nitrogen application than

the country in the left.

The USA is the country with the largest nitrogen fertilizer use per person

followed by France and Mexico, then the Philippines, and Tanzania has much

lower nitrogen fertilizer use per person than the rest of the countries. As

mentioned before, due to the relation between nitrogen application and crop

yields (Engels & Marschner 1995), a trade-off between nitrogen fertilizer and

land use is expected in which large nitrogen fertilizer use result in low nitrogen

land use. However, the trade-off between nitrogen fertilizer and land use per

person is not linear. Tanzania is the country with the largest land use per

person followed by Mexico, the USA, the Philippines and France. So, some

countries use both large land and nitrogen fertilizer. The USA has both larger

nitrogen fertilizer and land use per person than France, and Mexico show the

same relation in comparison with the Philippines.

The composition of the nitrogen fertilizer and land use per person is different

among the countries. For the USA and France, the animal food products account

for the largest share of the nitrogen fertilizer and land use per person. The

70

reasons are both the large consumption of animal food products and the low

nitrogen fertilizer use per kilogram of food of the vegetable products. In

contrast, for Tanzania, the staple food account for the largest share of the land

use per person. The reasons are the large consumption and the low crop yields

of these food products.

Figure 4.1 Nitrogen fertilizer and land use per person. Staple food includes the

consumption of cereals, roots and pulses. Affluent vegetal includes sugars, vegetable oils,

fruits and vegetables. Animal products include beef, pork, chicken, mutton meat, eggs

and milk. Source of data: calculations from the authors, see text for details.

Role of production systems and diets in the use of

nitrogen fertilizer and land use

Figure 4.1 shows large differences in nitrogen fertilizer and land use per person

due to production systems and diets. It is difficult to identify to what extent is

the diet or the production system driving the differences. In this section, we

discuss how individually the production system and the diet contribute to the

nitrogen fertilizer and land use per person. To identify the impact of the diets,

we keep the production system constant and compare different types of diets;

and to identify the impact of the production system, we keep the diet constant

and compare different types of production systems. We use three examples of

diets representing the global spectrum of diets: Tanzania representing a staple


71

diet, Mexico representing a transition diet, and the USA representing an affluent

diet (see table 4.3 for the data of the diets).

Figure 4.2 shows the use of nitrogen fertilizer and land per person for the three

diets produced in the production systems of each country. The differences

among figures 4.2a, 4.2b and 4.2c show the impact of the diets (with a constant

production system), and the differences within each graph (left to right) show

the impact of the production systems (with a constant diet).

The differences in diets can result in as much as four times more nitrogen

fertilizer and land use per person, comparing the diet of Tanzania and the USA.

The staple diet of Tanzania, low in animal products, requires less nitrogen

fertilizer than the affluent diet of the USA rich in animal products. Also, the

composition of the use of resources is different. With an affluent diet such as

the one of the USA, animal products account for 75% (produced in Mexico) to

90% (produced in the Philippines) of the nitrogen fertilizer use per person. And

with a staple diet such as the one of Tanzania, animal products account for only

40% (produced in the USA) to 60% (produced in the Philippines).

The differences in agricultural production systems can also result in large

differences in the use of nitrogen fertilizer and land. For example, look at figure

4.2a. The production systems of Mexico uses 4 times more nitrogen fertilizer

per person than the production systems of Tanzania, and Tanzania uses 7 times

more land per person than the production system of France.

The order of the countries in figure 4.2 (left to right) is in relation to the use of

land: from low to high land use per person. This allows us to identify the trade-

off between nitrogen fertilizer and land use due to the production systems

since the diet is constant in contrast with figure 4.1. Figure 4.2 shows that the

trade-off is not linear: the nitrogen fertilizer use per person does not follow an

opposite trend than the land use (higher to lower values). Mexico and the

Philippines use both larger nitrogen fertilizer and land use per person than

France and the USA. The reason is the values of the nitrogen fertilizer use per

kilogram of food of the different food categories (table 4.1) which were

calculated with the nitrogen fertilizer application and the crop yield. As

mentioned before, even though France and the USA have higher nitrogen

fertilizer application rates, the nitrogen fertilizer per amount of food produced

is lower than the ones for Mexico and the Philippines because of the high crop

yields of France and the USA in comparison with Mexico and the Philippines.

72

Figure 4.2 individual role of diets and production systems in the nitrogen fertilizer and

land use per person. Source of data: calculations from the authors, see text for details.

The most striking result of this analysis is the strong differences that the type of

diet and the type of production system can result in the per capita use of

nitrogen fertilizer and land. An affluent diet produced in the systems of


73

Tanzania uses 24 times more land than a staple diet produced in the systems of

France. And the affluent diet produced in France uses 9 times more nitrogen

fertilizer than the staple diet produced in Tanzania. Changes in diets from a

staple to an affluent diet generally result in a stronger increase in use of both

land and nitrogen (vertical differences in figure 4.2). However, the type of

production system can strongly change the use of resources and the ratio

between nitrogen fertilizer and land use (horizontal differences of figure 4.2).

4.4 Discussion

In this paper, we have identified the individual roles of both the type of

production system and the type of diet on the use of nitrogen fertilizer and

arable land per capita. To do this, we used examples of production and diets

representing the global differences among countries. Our aim was to identify

how agricultural factors (nitrogen application rate, crop yields, feed-food

efficiency) in combination with consumption factors (amount of food

consumed) drives the use of nitrogen fertilizer and land use per person. This

aim represented a methodological challenge due to the complexity of the food

system so simplification was needed which was described in the methodology.

Other studies in the literature (Billen et al., 2012; Bleken & Bakken, 1997;

Chatzimpiros & Barles, 2013; Howarth et al., 2002; Isermann & Isermann,

1998; Lassaletta et al., 2013; Ma et al., 2010; Pierer et al., 2014) have calculated

regional use of resources similar to what we did in figure 4.1, but with more

accurate methodology to calculate the real use of resources per person. We do

not attempt to do this, but we aim to show the strong impact of global

differences of diets and production systems on the use of nitrogen fertilizer and

land per person. In contrast with the studies mentioned above, we can project

the impact of different paths of production systems and food consumption

patterns to assess future use of resources as illustrated in figure 4.2.

In the coming decades, diets and agricultural production systems are expected

to change mainly in developing countries to affluent diets and intensive

production systems (Godfray et al., 2010a, Alexandratos & Bruinsma, 2012).

Due to population growth, the global availability of arable land per person will

decrease to less than 2000 m2 per person (Alexandratos & Bruinsma, 2012, fig

4.3). Tilman et al. (2011) forecast two scenarios for global nitrogen fertilizer

use in 2050, the first following the trends of past decades and the second

74

following a more sustainable intensification of agriculture. Using their

predictions, the global average use of nitrogen fertilizer per capita for both

scenarios will be 26 kgN/cap and 24 kgN/cap respectively (assuming

population growth with medium fertility rate (United Nations, 2011)). Figure

4.2 shows that with this amount of nitrogen and less than 2000 m2, it is not

possible to produce the food for an affluent diet. All systems in figure 4.2a

require 2200 m2 or more. It is possible to produce the food for a transition diet

but only with intensive systems such as the ones of the USA and France because

the other systems require more than 4000 m2/cap.

The countries that will increase their nitrogen fertilizer use because of the

change of diets and /or change of production systems are developing countries.

Some of these countries use relatively small amount of nitrogen fertilizer

similar to the example of Tanzania used in this paper. It means that they have to

strongly change their production systems. It is essential that the increase in

intensification is done with efficient agricultural practices having both high

nitrogen application and high crop yields resulting in low nitrogen fertilizer use

per kilogram of food. Otherwise, it might result in both large use of nitrogen

fertilizer and land the example of Mexico shown in figure 4.2. Also, the change

of food consumption pattern could make a large difference in the per capita

consumption similar to figure 4.2. A sustainable path of food patterns can

reduce by four times the use of nitrogen fertilizer and land.

4.5 Conclusions

We have quantified and discussed the impact of global dietary and production

differences on the nitrogen fertilizer and land use per person. A trade-off exists

between the use of nitrogen fertilizer and land. This trade-off is not linear

between the nitrogen application rates and land use because of the relation

between this application rate and the crop yield obtained. As a result, some

production systems result in large use of both nitrogen fertilizer and land.

We have identified the individual roles of the production systems and the diets

(figure 4.2). We have shown that the nitrogen fertilizer use per person can be 9

times different, and the land use per person can be as large as 24 times

different. These results can be used to assess the impact of the future changes

in dietary patterns and production systems.


75

Appendix 4

Validation of the calculation of nitrogen fertilizer use per tonne of food

We have calculated the use of nitrogen fertilizer per tonne of food for each food

category in the diet (table 4.1 and 4.2) to calculate the nitrogen fertilizer use

per person. We have used country level data to illustrate global differences of

production systems. Several assumptions had to be done due to the availability

of data.

First, we combined crop production data from two databases: nitrogen

application rate from FertiStat (FAO, 2007) and crop yield from FAOSTAT (FAO,

2013b). These databases give country level data averaging the crop production

systems of each country. The production systems within each country can differ

due to local conditions (climate, type of soil, management practices, others).

Second, in some cases, only a share of the harvested area in a country is

fertilized. FertiStat indicates, for each crop, the average application rate of the

fertilized area, and the share of the harvested area that is fertilized. In the cases

that only a share of the harvested area is fertilizer, we calculate a “weighted”

application rate and we used it as the application rate of country for that crop.

The reason to use this “weighted” application rate is because we combine it

with crop yield data of all the harvested area of the country given by FAOSTAT.

It is questionable whether with these assumptions our results illustrate real

situations of production systems. In order to validate our assumptions, we

compare the data we have used and our calculations of kilogram of nitrogen

fertilizer per tonne of crop with some case studies. These case studies have

crop field scale production data of nitrogen application rate and crop yields for

several crops. Similar to table 4.1, we combined the nitrogen application rate

and crop yield to calculate the kilograms of nitrogen fertilizer per tonne of crop

(table A4.1).

Table A4.1 show that for each food category, the nitrogen application rate, the

crop yield and the kilogram of nitrogen fertilizer per tonne of crop is in the

same order of magnitude as the values of table 4.1. The crops for each food

category in table A4.1 are ordered from high to low nitrogen application,

similar to table 4.1. In this way, it is easier to compare the values.

76

Table A4.1 Crop field scale production data from several sources

Crop N

application [kg N/ha]

Crop yield

[ton/ha]

kg N/ton crop

Case Study Area Source

Cereals

Maize 155 9.4 16 USA Pimentel (2009)

Barley 84 4.9 17 Iran Mobtaker et al. (2010)

Wheat 68 2.9 24 USA Pimentel (2009)

Wheat 8 1.5 5 Valley farm, India Triphati & Sah (2001)

Wheat 2 0.9 2 Mid-hill farm, India Triphati & Sah (2001)

Roots

Potato 319 44 7 USA Pimentel (2009)

Cassava 90 19 5 Nigeria Pimentel (2009)

Potato 33 3 12 Valley farm, India Triphati & Sah (2001)

Pulses

Peas 27 4.6 6 High-hill farm, India Triphati & Sah (2001)

Dry beans 16 1.5 11 USA Pimentel & Pimentel (2008)

Pulses 0 0.5 0 Valley farm, India Triphati & Sah (2001)

Oil crops

Soybeans 1561 3.2 48 Iran Mohammedi et al. (2013)

Soybeans 4 2.6 1 USA Pimentel (2009)

Rapeseed 3 0.2 17 Valley farm, India Triphati & Sah (2001)

Vegetables

Tomato 75 80 1 USA Pimentel (2009)

Cabbage 27 11 2 High-hill farm, India Triphati & Sah (2001)

Fruits

Apple 50 54 1 USA Pimentel (2009)

1 The application rate is not only for Nitrogen but also includes Potassium (K) and Phosphorus (P)

For cereals, the case study of maize in the USA in table A4.1 has similar nitrogen

application rate (155 kg N/ha), crop yields (9 ton/ha) and kg nitrogen fertilizer

per ton of crop (16 kg N/ton) than the maize production of USA and France that

we used for our calculations (table 4.1). Similarly, the wheat production in the

USA and in the valley farm in India (table A4.1) has similar values than the


77

maize production in Mexico and in Tanzania, respectively, used in our

calculations (table 4.1). Similar comparisons are shown in the other food

categories.

Also, the differences among food categories in the kilogram of nitrogen

fertilizer per ton of crop resulted from the case studies (table A4.1) is similar to

our calculations (table 4.1). The kilogram of nitrogen fertilizer per ton of

cereals of table A4.1 ranges from 2 kg N/ton to 24 kg N/ton, similar to table 4.1

in which it ranges from 7 kg N/ton to 26 kg N/ton. Roots, vegetables and fruits

show lower values in both tables: lower than around 10 kg N/ton. Vegetable

oils show large deviations among the crops, similar in both tables.

To conclude, table A4.1 validate our assumption in order to use our

methodology to illustrate global differences of production systems.

Chapter 5. Farm labour footprint of food

79

Chapter 5.

Farm labour footprint of food:

an international comparison of the impact of diets and

mechanization*

Abstract

The footprint studies of food have calculated the per capita use of the major

agricultural resources (land, water and nitrogen). We develop a farm-labour-

footprint in which we calculate the amount of farm hours needed to produce

the food for a person. Two examples of production systems and diets illustrate

the global differences. Non-mechanized systems need 200 times more labour

than mechanized systems, and affluent diets need two times more labour than

staple diets. The gain in labour efficiency with mechanization is enormous: only

2-5 hours of labour are needed to produce the food consumed by a person

during a year.

5.1 Introduction

A new line of research has emerged in the last years studying the impact of both

production and consumption of food on the use of agricultural resources such

as land (Kastner et al., 2012), water (Hoekstra & Mekonnen, 2012) and nitrogen

(Billen et al., 2012; Leach et al., 2012; Pierer et al., 2014). What some

researchers call “footprint” studies. These studies take a consumer perspective

and show the impact of diets on the requirement of agricultural resources per

person to produce their annual food consumption. The food in the diet is

traced back to the agricultural production in order to link it with the amount of

resources that were used to produce it.

* Under review in Food Policy as: M.J. Ibarrola Rivas, T. Kastner & S. Nonhebel. Farm labour footprint of food: an international comparison of the impact of diets and mechanization

80

The footprint studies have shown that, in general, affluent diets require larger

amount of agricultural resources than staple diets. However, in some cases, the

production system can overrule the type of diet. For example, the very basic

staple diet of Middle Africa requires similar amount of arable land than the very

affluent consumption of Western Europe (Kastner et al., 2012). The reason is

that the food of Middle Africa is produced in low crop yields systems and the

food of Western Europe is produced in high crop yields systems. With the

footprint studies, it is possible to combine both the differences of production

and consumption throughout the world to assess the impact on the use of

agricultural resources.

In this paper, we develop and discuss a farm-labour-footprint of food which we

define as the hours of farm labour needed to produce the food that a person

consumes in one year. Large global differences exist among food production

systems in relation to labour requirements. The whole spectrum covers from

low input systems where labour is done by hand or by draft animals to high

input systems where the production is mechanized (Pimentel & Pimentel,

2008). Also, diets are different throughout the world in relation to the

consumption of staple and affluent food products (Popkin, 1993). In order to do

an international comparison of the farm labour footprint, we describe four

different scenarios representing the global extremes of diets and production

systems. Our main research question is: How much farm labour is needed for

the production of our food consumption, and does dietary patterns matter?

5.2 Methodology

The footprint studies (Hoekstra & Mekonnen, 2012; Kastner et al.,2012; Leach

et al., 2012; Pierer et al., 2014) combine food consumption data (in kg/cap/yr)

with food production data (resources/kg food) to calculate the resources use

per capita. In this paper, we follow a similar methodology in which the

production data is the hours of farm labour needed to produce a kilogram of

food. So, we define the farm labour footprint as the amount of human working

hours needed in agriculture (crops and livestock) to produce the food that a

person consumes in a year. Similar to the footprint studies, the starting point is

the food consumed per capita, then we trace back to the agricultural production

the labour required to produce that certain amount of food.


81

Table 5.1. Data source of crop production systems

Food category

Production system: 1mechanized & 2non-mechanized

Crop Country and area

Source

Cereals Machinery1 Wheat USA Pimentel & Pimentel (2008) Draft power2 Wheat India, valley Tripathi & Sah (2001)

Roots Machinery1 Potatoes USA Pimentel (2009) Draft power2 Potatoes India, valley Tripathi & Sah (2001)

Pulses Machinery1 Dry Bean USA Pimentel & Pimentel (2008) Draft power2 Lentil India, valley Tripathi & Sah (2001)

Oil crop Machinery1 Soybeans USA Pimentel & Pimentel (2008) Draft power2 Soybeans India, mid-hill Tripathi & Sah (2001)

Vegetables Machinery1 Cabbages USA Pimentel & Pimentel (2008) Draft power2 Cabbages India, high-hill Tripathi & Sah (2001)

Fruits Machinery1 Apples USA Pimentel & Pimentel (2008) Draft power2 Apples India, high-hill Tripathi & Sah (2001)

Cereal for alcoholic beverages

Machinery1 Barley Iran Mobtaker et al. (2010)

Draft power2 Barley India, valley Tripathi & Sah (2001)

Feed Machinery1 Maize USA Pimentel (2009) Draft power2 Maize Mexico Pimentel & Pimentel (2008)

We included labour for production up to the farm gate i.e. planting, harvesting,

weeding, livestock management, others. The data was gathered from different

sources that calculate all the human labour required for each crop and livestock

product in a farm scale (see tables 5.1 and 5.3). We calculated the labour

footprint for four scenarios. Each of the four scenarios is a combination of a

different production system and a diet to illustrate the global differences of

consumption and production of food. Scenario 1 represents an affluent diet

produced in a non-mechanized system, scenario 2 an affluent diet produced in a

mechanized system, scenario 3 a staple diet produced in a non-mechanized

system, and scenario 4 a staple diet produced in a mechanized system.

The average per capita food supply of Western Europe in 2010 was used as the

affluent diet, and the one of Middle Africa in 2010 was used as the staple diet.

Data were taken from FAO (2013d). FAO (2013d) gives food supply data for 92

food items. In order to identify the impact of the production and consumption

data in the labour footprint, simplification of the diet is needed. We grouped

these food items into 12 food categories. The criterion for grouping was based

on the similarity of their crop production systems. So, the food items in each

82

food category have comparable labour requirements e.g. harvesting. The food

categories are: cereals, roots, pulses, vegetable oils, vegetables, fruits, alcoholic

beverages, milk, eggs, beef, pork and chicken. Stimulants i.e. coffee is not

included in the footprint but it is discussed in section 3.4.

The data sources of the production data of the vegetable products are indicated

in table 5.1. These studies quantify the amount of labour (in hours or days) in

the agricultural field to produce a certain amount of crop, and indicate the crop

yield obtained. By combining these data, we calculated the hrs/kg of crop. We

selected two crops for each food category which represent the mechanized and

the non-mechanized production systems of the vegetable products and the feed

for the animal products. In addition to human labour, the mechanized system

uses machinery for mainly all the tasks: ploughing, planting, weeding and

harvesting. The non-mechanized system uses draft animal power mainly for

ploughing and the rest is done only by human labour. Table 5.1 shows the crops

selected for each food category, the production systems, the countries of

production and the data source. The data is shown in table 5.4 in the results

section.

Table 5.2 Feed-food efficiency conversion for animal products.

Values in kilograms of feed (dry mass) per kilogram of output. The output refers to

carcass meat, milk or eggs.

Source: Mekonnen & Hoekstra (2012, Appendix 1)

Livestock food Global average of industrial system

Beef 19 Pork 3.9 Chicken 2.8 Eggs 2.3 Milk 1.1

The labour needed for the production of the animal products includes: the

labour to produce the feed as well as the labour for animal management. To

calculate the labour to produce the feed, we assumed that one type of crop was

used as feed for all the animal products. We selected maize as the feed, and we

used data for both mechanized and non-mechanized maize systems. Similar to

the vegetable food categories, we calculated the hours to produce a kilogram of

maize. Then, we multiplied it by a feed-food efficiency conversion factor (table

5.2) for each animal product to calculate the hours per kilogram of meat, milk


83

or eggs (table 5.3). The feed-food efficiency conversion factors were obtained

from Mekonnen & Hoekstra (2012, Appendix 1). We used the global average

data of an industrial system. For the non-mechanized system, only chicken was

produced with off-farm feed (see details below on the farm livestock labour).

Large scale intensive farms were used as examples of the farm labour for

livestock maintenance in mechanized systems. The data was gathered from

KWIN (2009). The KWIN (2009) indicate the number of animal places that a

farmer can manage in their farm and the life time of the animals before

slaughter (table 5.3). We assumed that each farmer works 2000 hours per year.

We used data of meat production per animal from the FAOSTAT, and milk and

eggs production from KWIN (2009). We combine these data to calculate the

hrs/kg of animal product of maintenance labour in the farm (table 5.4).

Smallholders’ livestock systems were used as examples of the farm labour for

livestock maintenance in non-mechanized systems. The choice of the

production system to illustrate these non-mechanized systems was more

difficult than the choice for the mechanized systems which are well

documented. Large variety exists in small livestock production systems

throughout the world. It is difficult to identify for all animal products an

example of production system with the same degree of labour efficiency. We

have chosen case studies of smallholders producers in different parts of the

world. The number of hours of labour for livestock production in these systems

is not as clear as for the intensive systems because these farmers usually do mix

systems producing other food crops or doing non-agricultural activities to earn

extra income. Also, the labour is shared within the household and sometimes

women’s and children’s labour is not fully quantified. However, we have chosen

some studies that have calculated, based on surveys and interviews, the amount

of labour needed exclusively for the livestock production of the smallholders.

These case studies illustrate a general picture of the labour needed in small

scale non-mechanized systems which show large differences in comparison

with the large scale mechanized systems (table 5.3 and table 5.4). We describe

them briefly below.

For beef, we used data from smallholders in Eastern Amazon, Brazil (Siegmund-

Shultze et al., 2007). The smallholders have pastures for the cattle in

combination with mixed systems of cassava, maize, pepper and others. In

average, one farm has 14.7 hectares of pastures for 8.3 cows of 450 kg of life

weight each. The labour required is in average 12 days of work per hectare

84

which includes planting, weeding, fencing, cutting down secondary vegetation

and animal care. We assumed a bull lives two years, and a dressing factor of the

bull of 0.6 (Elferink & Nonhebel, 2007), so each bull produces 270 kg of carcass

meat.

Table 5.3 Data source of livestock production systems

Livestock product

Production system

Animals places/

farm

Lifespan of animals [months]

Kg food product/

animal Country

Beef Large scale farm 400 1 24 1 338 6 Netherlands Beef Smallholder 8 2 24 1 270 2,8 Brazil Chicken Large scale farm 90 000 1 2 1 1.9 6 Netherlands Chicken Smallholder 17 3 2 1 1.2 7 Zimbabwe Pork Large scale farm 4 000 1 6 1 92 6 Netherlands Pork Smallholder 2 4 6 1 52 4,8 India Milk Large scale farm 50 10 - 8500/yr 1 Netherlands Milk Smallholder 2 5 - 721/yr 5 Bangladesh Eggs Large scale farm 60 000 1 - 22/yr 1 Netherlands Eggs Smallholder 17 3 - 5/yr 9 Zimbabwe

1KWIN (2009) 2Siegmund-Shultze et al. (2007) 3Muchadeyi et al. (2004) 4Kumaresan et al. (2009) 5Uddin et al. (2010) 6kilograms of meat produced per animal in 2010 for North America: FAO (2013b) 7 kilograms of meat produced per animal in 2010 for South Asia: FAO (2013b) 8dressing factors from Elferink & Nonhebel (2007, table 1) cow: 0.6, pig: 0.8, chicken:0.75 9 kilograms of eggs per hen per year in 2010 for Zimbabwe: FAO (2013b) 10 Own calculation based on annual milk production per farm (425 000 kg) and milk production per cow (8 500 kg) both data from KWIN (2009)

For chicken meat and eggs, we used data from smallholders in the semi-arid

region of Zimbabwe (Muchadeyi et al, 2004). These households have a mix

system of maize, cotton, sunflower and a vegetable garden in addition to the

chickens, and some of them also have cattle and small ruminants. The

households have in average 17 chicken, and almost all farms supplemented the

chicken with feed. Labour for chicken demanded the least time of the

household duties and it was carried out all year. Muchadeyi et al. (2004) show

that, in average, one person spends one hour per day for managing the chicken

including penning the chicken at night, releasing them in the morning and

feeding them. To allocate this hour between the chicken meat and eggs

production, we assumed that half of this hour is for the chicken meat, and the

other half is for the eggs. We assumed the chicken live 2 months before

slaughter, so each household produced 102 chickens per year in average, and

each chicken produces 1.2 kg of meat (average data for Zimbabwe in 2010 from


85

FAO (2013b)). For eggs, we assumed that these 17 chicken are hens that each

produce 5 kilograms of eggs per year (yield data for Zimbabwe in 2010 from

FAO (2013b)).

For pork, we used data from smallholders in the mountain area of Northeast

India (Kumaresan et al., 2009). These households also have mixed farming

systems. In average one household has 2 pigs which are kept in the backyard of

the house. Most household practice stall feeding and not scavenging. The

household produced the feed by cooking kitchen wastes, crop residues and

plants collected from the forest or near areas, this feed lasts for 3 days. The pigs

are feed twice a day. The live weight at 12 months is in average 65 kg per pig.

We assume a dressing factor of 0.8 (Elferink & Nonhebel, 2007) and after 12

months the pigs are slaughter, so one pig produces 52 kg of meat per year.

Kumaresan et al. (2009) do not give a number in the hrs of labour needed for

these activities, so we assumed one hours per day is needed all year long.

For milk, we used data from small-scale extensive systems in Bangladesh

(Uddin et al., 2010). In average, the households have two cows in a mixed

farming system including rice production with 0.5 hectares of land for grazing

and also perform off-farm activities. The farmers practice a “cut and carry

feeding system” (Uddin et al., 2010) and, in addition to using their own land,

they use public land for periodic grazing. Family labour is the only labour

involved in managing the cows which involve 2 163 hours per year per

household, and each cow produces 721 kg of milk.

We calculated the amount of farm labour needed to produce a kilogram of meat,

milk and eggs with the data described above. The results are shown in table 5.4.

Finally, we calculated the labour footprint by combining the data of production

(hrs/kg) and consumption (kg/cap /yr) of all food categories to calculate the

agricultural hours of labour needed per person (hrs/cap) for the four scenarios.

The data is shown in table 5.4. The consumption data of vegetable oils and

alcoholic beverages was converted into its crop equivalent following the

methodology of Kastner & Nonhebel (2009). We assumed that the consumption

data of alcoholic beverages is beer and the vegetable oils is soybean oil.

86

5.3 Results

The results of the farm labour footprint are shown in figure 5.2. First, we

calculated the hours needed per kilogram of food product (table 5.4). Then, we

combine these data with the food consumption data in kilograms of food per

capita per year to calculate the farm labour footprint.

Labour hours per kilogram of food

The amount of labour needed per kilogram of crop (hrs/kg) was calculated by

dividing the hours of labour per hectare (hrs/ha) by the crop yield (kg/ha). The

mechanized and non-mechanized systems are widely different in both factors,

so the differences in hrs/kg should be analysed considering both factors. In

general, mechanized systems use less hrs/ha and obtain higher crop yields

which result in much lower hrs/kg than the non-mechanized systems. The

lower labour per hectare is related with the machinery used in comparison

with the non-mechanized systems. The higher crop yields are not related with

the machinery but with the use of other inputs such as fertilizers and

pesticides.

Differences exist among the type of crops. Potatoes result in the largest

difference in hrs/ton: 800 times more hours in non-mechanized system than in

the mechanized system. In this case, the hours per hectare is 50 times higher in

the non-mechanized system, and the crop yield is 15 times higher in the

mechanized system. Cereals and oil crops show the largest difference in labour

in the field [hrs/ha]: 250-350 times more hours in the non-mechanized systems

than in the mechanized systems. But, the difference between crop yields is only

2 times. As a result the hrs/kg is less than potatoes. Fruits and vegetables show

the smallest difference in hrs/kg: “only” 100 times higher in the non-

mechanized system for vegetables and 40 times higher in the non- mechanized

system for fruits. The reason is different for each food categories. For

vegetables, the labour in the field (hrs/ha) is relatively different (30 times) but

the crop yields are very similar. For fruits, both systems show very similar

labour in the field (hrs/ha), but the crop yield is very different: 60 times higher

for intensive system.


87

Similar to the vegetable products, the animal products show a strong difference

in the hrs/kg between the mechanized and the non- mechanized system. The

farm labour requires 600 to 1 500 times more hours in the smallholder farms

for chicken, pork, milk and eggs, and 50 times more hours for beef. The large

scale farms have high labour efficiency because of two main reasons. First, large

numbers of animals are managed by only one farmer in comparison with the

smallholders’ farms. For chicken and pork the difference is more than 2000

times larger in the large scale farms, for beef and milk is “only” 50 and 100

times larger respectively (table 5.3). Second, the amount of meat or milk

produced per animal is larger in comparison with the smallholders (table 5.3).

Table 5.4 Production and consumption data for the 4 scenarios which include:

mechanized system (m), non-mechanized system (n-m), affluent diet (A)

and staple diet (S)

Production data

Consumption data

Vegetable Products

Food category

Crop Yield

[kg/ha] Labour

[hrs/ha]

Labour for kg of food [hrs/kg]

Kg/cap/yr

Cereals Wheat (m) 2 670 8 0.003 115 (A) Cereals Wheat (n-m) 1 500 1 098 0.700 107 (S) Roots Potatoes (m) 40 656 35 0.001 68 (A) Roots Potatoes (n-m) 2 800 1 875 0.700 185 (S) Pulses Dry beans (m) 1 457 10 0.007 1.5 (A) Pulses Lentil (n-m) 300 683 2.300 12 (S) Veg oils Soybeans (m) 2 668 7 0.003 19 (A) Veg oils Soybeans (n-m) 1 300 1 217 0.900 8 (S) Vegetables Cabbages (m) 38 416 60 0.002 100 (A) Vegetables Cabbages (n-m) 11 400 1 834 0.200 64 (S) Fruits Apples (m) 55 000 385 0.007 109 (A) Fruits Apples (n-m) 900 230 0.300 74 (S) Beer Barley (m) 4 800 83 0.017 112 (A) Beer Barley (n-m) 500 960 1.700 54 (S) Feed Maize (m) 8 655 11 0.001 N.A: Feed Maize (n-m) 940 383 0.400 N.A.

88

Table 5.4 (continue)

Production data

Consumption data

Animal Products

Production system External

feed needed

Animal management

labour in farm [hrs/kg]

Labour for feed

[hrs/kg] Kg/cap/yr

Beef Large scale farm (m) Yes 0.0300 0.025 18 (A) Beef Smallholder (n-m) No 1.3000 -- 7 (S) Pork Large scale farm (m) Yes 0.0030 0.005 44 (A) Pork Smallholder (n-m) No 1.7000 -- 4 (S) Chicken Large scale farm (m) Yes 0.0019 0.004 20 (A) Chicken Smallholder (n-m) Yes 1.5000 1.100 8 (S) Milk Large scale farm (m) Yes 0.0050 0.001 260 (A) Milk Smallholder (n-m) No 1.5000 -- 16 (S) Eggs Large scale farm (m) Yes 0.0015 0.003 13 (A) Eggs Smallholder (n-m) Yes 2.2000 0.900 1 (S)

NOTES: The production systems are indicated with (m) = mechanized system and (n-m)= non-mechanized. The type of diets are indicated with (A) = affluent consumption which is the average consumption of Western Europe, and (S)= staple consumption which is the average consumption of Middle Africa. The consumption for beer is for Alcoholic Beverages from FAO (2013d). The data of the vegetable products non-mechanized systems (Tripathi & Sah, 2001) is given in days of work per hectare, we assumed that each working day includes 8 hrs.

The relation between the labour footprint for the feed and for the animal

management in the farm is different among the mechanized and non-

mechanized systems. For the large scale farms, the labour for the feed (in

hrs/kg) is higher than the farm labour and the opposite for the smallholders.

So, the labour for the feed is more relevant in the large scale farms, and the

management labour in the farm is more relevant in the smallholders’ farms.

However, only in the chicken and eggs production system of the smallholders

the labour for the feed is differentiated from the farm labour because is the only

system that buys the feed concentrates from outside the household. The other

smallholders farms include the labour for the feed within the farm.

The labour efficiency is different in the large scale farms among the food

products. Eggs and chicken are the products requiring less labour per kilogram

produced: 0.0015 hrs/kg and 0.0019 hrs/kg respectively. Then pork (0.003

hrs/kg) and milk (0.005 hrs/kg). And beef is the least labour efficient with 0.03

hrs/kg because of two main reasons: first, the lifespan is longer for cattle than


89

for pigs and chicken. Second, the average farm size for cattle is smaller than for

pigs and chicken.

The labour efficiency among the animal products in the smallholders show a

different relation: eggs are the least efficient followed by pork, chicken, milk

and beef. However, the data for these systems was gathered from different

sources, countries, and production systems in comparison with the large scale

farms. As mentioned before, it is difficult to identify the same degree of “labour

efficiency” for these systems. Therefore, the difference in hrs/kg among the

animal products in the non-mechanized systems is not due to the type of

product but by the type of production system. However, this limitation of the

data does not alter our results (see appendix 1). We are interested on

identifying the general impact of mechanized and non-mechanized systems.

The values of farm labour of the non-mechanized systems are within the same

order of magnitude, and differ with the mechanized systems so we only focus

on the general differences.

Food consumption patterns

We used data of food consumption in kilograms per person per year for our

calculations in order to combine this data with production data which is in

hours per kilograms. However, to discuss the dietary difference, we used

different units: daily calories per person, which are commonly used to describe

nutritional differences.

The staple and the affluent diets are different in both amounts of calories

consumed and on the composition of the food products (figure 5.1). The

affluent diet has 500 kcal/cap/day more than the staple diet. The main

difference is in the amount of animal products. For the affluent diet, 30% of the

calories are of animal origin and in the staple diet only 6%. The amount of

calories consumed of vegetable products is relatively similar, around 1900

kcal/dap/day. But the composition differs. In the staple diet, cereals, roots and

pulses account for 80% of the calories; and in the affluent diet, these food items

only account for 50%.

90

Figure 5.1. Differences among the staple and affluent diet. Source: Staple diet: average

food supply of Middle African in 2010 (FAO, 2013d), affluent diet: average food supply of

Western Europe in 2010 (FAO, 2013d)

Farm labour footprint

Figure 5.2 shows the farm labour footprint for the four scenarios. It is striking

the low amount of hours needed to produce the food of a person in a

mechanized system. Less than five hours for an affluent diet and only two hours

for a staple diet are needed to produce the food that a person consumes

throughout the whole year.

The labour footprint varies a lot among the four scenarios. Figure 5.2 shows

these differences, note that the scale is 200 times different in scenarios 1 and 3

than in scenario 2 and 4. The difference in the production systems has the

strongest impact on the labour footprint. A non-mechanized system requires

200 times more hours of labour than a mechanized system. But also, the type of

diet has a relative impact on the labour footprint: an affluent diet requires two

times more hours of labour than a staple diet.

Not only is the amount of hours different among the four scenarios but also the

composition of the hours of labour in relation to the food categories. For staple

diets, most of the labour is for the production of the vegetable products: 60% in

a mechanized system and 80% in a non-mechanized system. The reason is

mainly the low consumption of animal products in comparison with the affluent

diet. In contrast, for the affluent diet, less than 30% of the labour is for

vegetable products.


91

Figure 5.2 Farm labour footprint for the four scenarios.

Looking in detail to the food categories we have identified the following. The

labour for vegetable products produced in non-mechanized systems is more

diversified among the different food categories. In contrast, the labour in

mechanized systems for the vegetable products is mainly for fruits and cereals.

The share of labour for animal products in the total diet is similar for both

systems, but the composition differs. In non-mechanized systems (almost) all

the labour is in the farm, and in the mechanized system only one fourth or fifth

is labour in the farm, and the rest is indirect labour related with the production

of the feed.

92

Impact of food imports

In this paper, the labour required for the production of stimulants such as

coffee was not included in the analysis. Other studies have shown that the use

of resources for stimulants are relevant in relation to the rest of the food

categories in the diet (Kastner et al.,2012; Hoekstra & Mekonnen, 2012). In

general, coffee is an imported affluent food product. The largest per capita

consumers of coffee are people living in industrial countries in North and

Western Europe (FAO, 2013d). These countries import all their coffee

consumption (FAO, 2013d). In contrast, the major coffee producers are

developing countries such as Brazil, Vietnam, Indonesia, Colombia and Ethiopia

(production data in 2012: FAO (2013b)). Therefore, the discussion of labour for

coffee is in relation to the impact of food imports. Other affluent food products

are also commonly originated from food imports. For instance, Northern

Europe imports most of their consumption of fruits (data on 2010 from FAO

(2013d)). In this section, we discuss the impact of mixing food products

produced with different degree of mechanization in the labour footprint. We

use coffee and fruits are examples.

Table 5.5. labour footprint of coffee

hrs/ha kg/ha hrs/kg

Staple diet [hrs/cap]

Affluent diet [hrs/cap]

Coffee production in Sumatra Indonesia

1600 1 540 2 3,0 3 5,3 3 26,4 3

1Budidarsono et al. (2000) table 2.3, we assumed that a working day requires 8 hours of work; 2average crop yield of Indonesia in 2010 (FAO, 2013b); 3calculations from the authors

We calculated the labour footprint for an affluent and a staple diet for which all

food products are produced in a mechanized system except for fruits and coffee

which are imported from a non-mechanized system. For fruits, we used the

production data of Triphany & Sah (2001) which was the example of a non-


93

mechanized system for scenarios 1 and 3 (see table 5.4 for the production

data). For coffee, we used the data of a case study in Indonesia (Budidarsono et

al., 2000).

Bubidarsono et al. (2000, table 2.3) calculate the hours of labour needed for

several coffee production systems in the Lampung province of Sumatra in

Indonesia. Most coffee production systems in this area require around 200

days of work per hectare per year. With this data and the crop yield data (FAO,

2013b), we calculated the hours needed per kilogram of coffee produced (see

table 5.5): 3 hrs/kg. This value is large in comparison with the other vegetable

food categories produced in a non-mechanized system shown in table 5.4. As a

result, even a low consumption of coffee results in relative large need for labour

in comparison with the other food categories.

Figure 5.3. Impact of food imports (coffee and fruits) from a non-mechanized system on

the labour footprint of food produced in a mechanized system.

Figure 5.3 shows the labour footprint for an affluent and a staple diet produced

in a mechanized system with exception of coffee and fruits which are produced

in a non-mechanized system. Coffee and fruits imported from a non-

mechanized systems play a major role in the labour footprint since it overrule

the labour for the other food products produced in a mechanized system. For

example, the affluent diet produced with a mechanized system requires 4.7

hrs/cap, but by adding the labour for coffee and fruits from a non-mechanized

system, then the whole footprint is as much as 50 hrs/cap. This would be the

94

case of an industrial country producing their food with mechanized system but

importing coffee and fruits from a non-mechanized system.

5.4 Discussion

In this paper, we analysed two production systems which are strongly different

among each other. In reality, production systems vary throughout the world, so

by choosing other production systems or other assumptions for the production

data, the values of figure 5.2 could differ. However, not in order of magnitude as

shown appendix 1. Our aim is to illustrate global differences and discuss orders

of magnitude. Therefore, our values should not be considered as facts, but as

starting point to discuss the impacts of production systems and diets on the use

of agricultural labour. In addition, we only considered the labour needed for

food production up to the farm gate. The rest of the food chain also requires

labour for transport, food processing, storage, cooking, etc. This is outside the

boundaries of our research.

The farm-labour footprint (figure 5.2) gives an indication of the number of

people that a farmer can feed in the four scenarios that were studied. Assuming

that a farmer works 2000 hrs/year (8 hours a day during 250 days), then the

farmer with a mechanized system can produce the food for 400 people with

affluent diets (scenario 2) and 1000 people with staple diets (scenario 4), and

with a non-mechanized system the farmer can produce the food for 5 people

with staple diets (scenario 3), and only 2 people with affluent diets (scenario 1).

The differences in production systems and diets are related with the

socioeconomic development of the population. In general, low income countries

have both staple diet (Popkin, 1993) and non-mechanized systems (Pimentel &

Pimentel, 2008). As income increases, diets change to a more affluent

consumption (Popkin, 1993) and mechanization of production systems

increases (Pimentel & Pimentel, 2008). In our results, scenario 3 illustrates a

low income country which in general one fourth of their population is engaged

in agriculture*. The fact that one farmer can produce the food for 5 persons fits

with this figure. Scenario 2 illustrates a high income country which in general

* Population data for Least Developed Countries in 2014 (FAO, 2013b; United Nations, 2011)


95

less than 1% of their population are engaged in agriculture*. This figure fits

with the fact that one farmer can produce the food for 400 people.

One of the main insights obtained in this paper is the strong impact of the type

of production system due to the labour efficiency gained with mechanization.

The labour footprint with a non-mechanized system is 200 times higher than

the one with a mechanized system. The type of diet plays an important role in

the labour footprint, though not as large as the production system. An affluent

diet requires two times more working hours than a staple diet.

The composition of the labour hours in the diet is different among the

production systems. In non-mechanized systems the labour in agriculture is

more diversified among all food categories, and in mechanized systems it is

more specialized (e.g. mainly for fruits). The reason is that each type of crop

increases the labour efficiency differently with mechanization. See the

difference in the values of hrs/kg between the mechanized and non-

mechanized systems in table 5.4. The increase in labour efficiency with

mechanization for cereals, oil crops and pulses is 300 times, in contrast for

vegetables and fruits it is “only” 40-100 times.

The difference in gaining labour efficiency with mechanization is intrinsic to the

crop: whether the management in the field is possible to mechanized. For

example, roots require a lot of labour for harvesting and it is possible to

mechanized. So, the labour reduces strongly with mechanized systems. Fruits

also require a lot of labour for harvesting but, in contrast with potatoes, it is

more difficult to mechanize. So, the gain in labour efficiency with

mechanization for fruits is not as large as for potatoes.

The increase in labour efficiency is not only related with mechanization which

results in smaller hrs/ha, but also with the increase of crop yields (kg/ha).

Some production systems with low mechanization can have relatively high

labour efficiency because of high crop yields due to large use of external inputs

such as fertilizers and pesticides. Though, in general, non-mechanized systems

are performed by smallholders who do not use large amounts of external inputs

resulting in low crop yields, and vice versa.

* Population data for Western Europe in 2014 (FAO, 2013b; United Nations, 2011)

96

To conclude, the farm-labour footprint gives new insights with a consumer

perspective on the impact of diets and production systems on the need for

agricultural labour. The increase of labour efficiency with mechanization is

enormous. With a mechanized system, less than 5 hours of farm labour are

needed to produce the food consumed by a person during a whole year with an

affluent diet. This means that less than a minute is needed to produce the daily

food for this person. This value is much lower than the time the person spends

in buying this food, cooking it or eating it.

Appendix 5.

Sensitivity analysis to validate the calculations of the farm labour

footprint

Our results show that the type of production system has a strong impact on the

farm labour footprint (figure 5.2). These results are based on the selection of

two extremely different production systems regarding mechanization and

livestock farm scale, and on several assumptions. Production systems widely

vary throughout the world, so it is questionable whether our results would

have a different outcome by choosing different assumptions. In this section, we

investigate whether the choice of different mechanized or non-mechanized

systems, and the choice of different assumptions can lead to a different

discussion. Table A5.1 show the results of the labour footprint by changing the

variables of the production systems. The first row shows the results with the

assumption chosen in this paper (figure 5.2), the following rows show the

deviation of these results by choosing a different assumption indicated in the

first column. First, we discuss the assumptions for the crop field systems of the

vegetable food products in the diet; then, we discuss the assumptions for the

feed of the animal food products; finally, we discuss the assumption for the

livestock production systems.


97

Table A5.1. Sensitivity analysis of the assumptions of the production systems used in

this chapter.

mechanized system

Non-mechanized system

Assumptions Staple diet

Affluent diet

Staple diet Affluent

diet

[hrs/cap] [hrs/cap] [hrs/cap] [hrs/cap]

1 This paper 1.9 4.7 355 830

2 Crops more efficient:+100% crop yields & -50% hrs/ha

1.0 3.7 161 666

3 Crops less efficient:-100% crop yields & +50% hrs/ha

4.6 8.0 935 1321

4 Feed: Crops more efficient:+100% crop yields & -50% hrs/ha

1.7 4.0 348 807

5 Feed: Crops less efficient:-100% crop yields & +50% hrs/ha

2.4 7.1 375 898

6 Feed: feed-food efficienty lower -100% 1.8 4.1 349 812

7 Feed: feed-food efficienty higher: +100% 2.2 5.9 365 864

8 Farm: +50% more animals/farmer 1.8 4.1 337 646

9 Farm: -50% more animals/farmer 2.1 5.7 382 1104

10 Farm: +50% lifespan meat animals, milk constant

1.8 4.5 345 787

11 Farm: -50% lifespan meat animals, milk constant

2.0 5.1 369 894

12 Farm: +50% kg of meat/animal & kg milk/cow

1.8 4.1 337 646

13 Farm: -50% kg of meat/animal & kg milk/cow

2.1 5.7 382 1104

14 Farm: mechanized: all constant; non mechanized: chicken & pig -100% hrs/day, beef & milk constant

1.9 4.7 344 762

15 Farm: mechanized: all constant ; non mechanized: chicken & pig +100% hrs/day, beef & milk constant

1.9 4.7 376 965

The vegetable food products in the diet include 75 food items which were

grouped into seven food categories. We have chosen one production system for

each food category assuming that only one crop represents each food category.

For example, the consumption of cereals include wheat, maize, rice and other

cereals, and we assumed that the cereals consumption is only wheat. As a

result, the combination of the crop yield (kg/ha) and the crop field labour

(hrs/ha) of the wheat production system represents all the cereals production

98

systems. In reality, each food item has different values of crop yields and labour

per hectare. For example, according with Pimentel (2008; 2009), mechanized

systems in the USA of maize, wheat and rice have crop yields of 8.7 ton/ha, 2.7

ton/ha and 7.4 ton/ha respectively, and values of agricultural labour of 11.4

hrs/ha, 7.8 hrs/ha and 24 hrs/ha respectively. As a result, the labour per

kilogram of cereal would be 1.3 hrs/kg if we assume the consumption is only

maize, and 3.3 hrs/kg if we assume the consumption is only rice, instead of 2.9

hrs/kg which is the value of wheat that we used in this paper.

Similar variations exist for non-mechanized systems for crops within the same

food category (Triphany & Sah, 2001). In addition, crop yields vary among

years and countries (FAO, 2013b) because of climate, geographical conditions

and management practices. Due to all these differences, we evaluated the

impact of, first, increasing the crop yields two times and reducing the hours of

labour per hectare 1.5 times (row 2), and, second, reducing the crop yields two

times and increasing the hours of labour per hectare 1.5 times (row 3). Higher

crop yields and lower labour per hectare results in a reduction of the total

footprint by 50% for staple diets and 20%-30% for affluent diets (row 2).

Lower crop yields and higher labour per hectare increases the labour footprint

by around two times for both staple and affluent diets (row 3).

The feed for livestock is usually a mixture of crops and is different for each

livestock animal. We have assumed that only one crop is used as feed for all

livestock animals. We chose maize since it is globally the most common crop

used as animal feed (FAO, 2013d). For the same reasons of diversity of crop

systems mentioned above for the vegetable products , rows 4 and 5 of table

A5.1 show the impact of different values of crop yields and crop field labour of

the feed crop. Higher crop yields and lower labour per hectare in mechanized

systems result in a reduction of the total footprint by 10% for staple diets and

by 20% for affluent diets (row 4). Lower crop yields and higher labour per

hectare in this same system increase the labour footprint by 5 times for staple

diets and two times for affluent diets (row 3). The other assumption for the

animal feed is the feed-food efficiency factor (table 5.2). We have chosen the

average global values of industrial production systems for each livestock

product given by Mekonnen & Hoekstra (2012). However, this efficiency factor

is different among livestock animals, type of production system and regions

(Mekonnen & Hoekstra, 2012; Elferink & Nonhebel, 2007). Rows 7 and 8 of

table A5.1 shows that for mechanized systems, the labour footprint reduces by

around 10% by halving the global feed-food efficiency factors, and it increases


99

15% for staple diets and 30% for affluent diets by doubling the factors. The

non-mechanized systems are not relevantly influenced by the feed assumption

because most of the labour for the feed of the livestock is included in the farm

labour (see section 2).

The livestock production systems also widely vary around the world. Especially

for the small scale systems, the labour requirement is not widely documented

and only a limited number of case studies exist evaluating in detail the labour

involved in these systems. Also, it is difficult to label whether a system is a

small scale system since it could depend on the context or reference. For

example, a livestock producer is considered to be a “smallholder” if they have

only one head of cattle or as much as 50 heads (Siegmund-Schultze et al.,2007).

In addition, the lifespan of the animals is different around the world. In general,

in industrial countries it is shorter and in developing countries it is larger

resulting in a higher productivity in the former than in the latter (FAO, 2013b).

We have analysed the impact of changing the main variables of the production

systems: number of animals per farmer by 1.5 times (rows 8 and 9), lifespan of

the livestock animals for meat production by 1.5 times (rows 10 and 11), the

amount of food produced per animal by 1.5 times (rows 12 and 13), and the

hours of labour per day for managing the small scale chicken and pig systems

by 2 times (rows 14 and 15). Table A5.1 shows that the total labour footprint

only changes by 10% to 20% by changing the assumptions and variables

described above for the livestock management labour.

To conclude, this sensitivity analysis shows that even by strongly changing the

variables of the production systems in accordance with global differences in

mechanized and non-mechanized production systems, the order of magnitude

of the labour footprint of the four scenarios is still as strong as shown in figure

5.2. So, the general discussion lead by our results is valid.

Chapter 6. Impact of regional diets on the global use of resources

101

Chapter 6.

Future global use of resources for food:

the huge impact of regional diets*

Abstract

The global use of resources for food is large, it causes strong environmental

problems, and future dietary changes are expected to increase their use. A

global trend exists in which diets change to an affluent consumption with the

increase of socioeconomic development following the Nutrition Transition

Theory. However, it is questionable whether all regions will follow the same

food pattern with the present regional cultural differences. In this paper, we do

a global analysis of the regional dietary changes from 1960 to 2010 to assess

future dietary paths and its impact on the use of agricultural resources. We

show that regional dietary trends are stronger than the global trend. So, we

expect that future diets will follow present regional dietary compositions. The

change to an affluent diet with these regional dietary paths can result in a

different use of resources. We show that for the consumption of animal

products, the use of resources can be doubled depending on the type of meat.

Also, vegetarian diets with large consumption of dairy products can use similar

amount of resources than diets with large consumption of meat, depending on

the type of meat. Our results give new insight to reduce resource use from a

demand perspective, though cultural barriers can be a strong challenge in some

regions.

6.1 Introduction

Global food production uses large amounts of resources mainly land, energy

and water (FAO, 2013a; FAO, 2013c; Woods et al., 2010). The per capita use of

resources is widely different throughout the world, and affluent diets rich in

* Invited paper to Global Food Security, resubmitted version under review as: M.J. Ibarrola Rivas, H.C. Moll & S. Nonhebel. Future global use of resources for food: the huge impact of regional diets

102

animal products use more resources than staple diets (Kastner et al., 2012;

Leach et al., 2012; Mekonnen & Hoekstra, 2011a). Diets have changed due to

different factors e.g. socioeconomic development, income and urbanization

among others. Large changes in dietary patterns are expected in the coming

decades especially in developing countries (Alexandratos & Bruinsma, 2012;

Godfray et al., 2010a). These changes in diets will have major impact on the

increase demand of resources and, therefore, strong environmental impacts

(Godfray et al., 2010b; Smil, 2002b; Tilman et al., 2011).

The most recognized global trend of dietary changes in the literature is

formulated in the Nutrition Transition theory (Popkin, 1993). This theory

states that diets change with socioeconomic development as follows: starting

with an under nutrition state, first the amount of calories increase including

mainly staple food, then it diversifies increasing the consumption of fats (e.g.

animal products), sugars and processed food, and finally reaching a final state

by increasing the consumption of fruits, vegetables and carbohydrates, and

reducing fats. Since the last decades, these changes in diets have happened

faster due to the globalization of the food system and the fast urbanization in

many developing countries (Kearney, 2010). So, it is expected that diets in

developing countries undergoing fast economic development and urbanization

will change rapidly in the future following this pattern (Alexandratos &

Bruinsma, 2012; Kearney, 2010).

Most of the studies calculating future use of resources due to dietary changes

assume a linear relation between economic development (GDP) and caloric

intake in accordance with the nutrition transition (Tilman et al., 2011). In these

cases, the saturation level (last phase of the nutrition transition) is achieved

similarly to the regions currently in this phase (e.g. Western Europe and North

America). However, it is questionable if diets on a global scale will all follow

this same food pattern since food consumption is also influenced by culture and

religion. Alexandratos & Bruinsma (2012) discuss the regional differences in

dietary composition and they show that large diversity exists among countries

in the composition of the diets even though some global trends are clear. For

example, the clearest global trend in the last decades has been in developing

countries towards an increase consumption of livestock products and

vegetables oils, but countries like China and Brazil (which account to 1.5 billion

people) show a much higher meat consumption for their development stage,

and in contrast, India (which account to more than 1 billion people) hardly

changed its diet with a very low consumption of meat for a similar increase in


103

economic development. So, regions can strongly deviate from global dietary

trends.

In this paper, we study the global changes in diets associated with their use of

resources. We study the regional dietary changes in the past to assess future

regional dietary changes. To do this, the paper is divided into two parts. First,

we study the development of diets to assess future dietary changes. In this part

of the paper, our main research questions are: What is the major trend of

dietary changes? Is the global dietary trend or the regional dietary trend

stronger? What can we expect for future changes in diets? To answer these

questions we study the differences in food composition among regions. In the

second part of the paper, we study the use of resources for these future dietary

paths. In this section, our main research question is: What is the impact on the

use of resources of the regional differences in food composition?

6.2 Methodology

We divided the world into 13 regions to discuss the global differences in diets

and the use of resources. We grouped the regions defined by the FAO as

follows: E-M-W Africa (including East, Middle and West Africa), N Africa (North

Africa), S Africa (South Africa), China, India, S-E Asia (South, East and Southeast

Asia), W-C Asia (West and Central Asia), C-S America (Central and South

America and the Caribbean), E Europe (Eastern Europe), N America (North

America), W-N Europe (West and North Europe) and Oceania. China and India

are one region each because of their large population. Appendix 6 shows the

countries included in each region.

General approach to calculate the use of resources per person

The amount of resources needed to produce the food of a person depends on

how the food was produced: the agricultural technology, and on what people

eat: the type of diet. Several studies have calculated the use of resources per

person (Kastner et al., 2012; Leach et al., 2012; Mekonnen & Hoekstra, 2011a),

what some call “footprint” of food. They show large differences among regions

due to both differences in agricultural production systems and type of diets.

They show that, in general, affluent diets rich in animal products require more

resources than staple diets because the production of a kilogram of animal

104

product requires more resources than a kilogram of vegetable products

(Berners-Lee et al., 2012; González et al., 2011; Mekonnen & Hoekstra, 2012;

Mekonnen & Hoekstra, 2011b; Pierer et al., 2014).

However, in some cases the differences in agricultural production systems can

overrule the differences in diets. Kastner et al. (2012) show that regions with

affluent diets including large consumption of animal products (e.g. Western

Europe) require similar amounts of land than regions with very basic staple

diets with low consumption of animal products (e.g. Middle Africa) due to the

differences in agricultural technologies, the former with high land efficient

systems (high crop yields), and the latter with low land efficient systems (low

crop yields). So, with the footprint studies it is not possible to identify the

impact of only the global differences in diets. In contrast with these studies, we

aim to discuss only the differences in diets to identify the impact of different

dietary changes.

In addition, the type of diet is not only related with the amount of food

consumed, but also with the type of food consumed. So, the type of diet depends

on the amount of caloric intake as well as on the food composition of the diet.

Since we only focus on the impact of the differences in diets, we keep the

technology factor constant. By doing so, we can identify the impact of global

differences of both the amount of food consumed and the type of food

consumed (food composition of the diet).

To keep the technology factor constant, we use one production system to

calculate the use of resources per capita representing a global average

production system or a production system of a specific country when the global

data is not available. The data was gathered from several sources and table 6.1

shows the data that we used. All data was gathered from the literature except

for the use of land for the vegetable products for which we used the inverse

value of the crop yield of the global average in the year 2010 given by the

FAOSTAT (FAO, 2013b). The country level data for the animal products was

gathered from de Vries & de Boer (2010). They collected production data from

several countries and production systems in Europe and discuss their

differences. We use their high estimates for each animal food product.


105

Table 6.1. Data for the use of agricultural resource per amount of food.

Sources: 1 Crop yields from FAOSTAT (FAO, 2013b); 2 Mekonnen & Hoekstra (2011b); 3 de Vries & de Boer (2010)

Vegetable Products

Resource Region Unit cereals roots pulses veg oils sugars vegetables fruits

Land1 world avg

m2/kg food

2.8 0.7 11.2 10.2 1.7 0.5 0.9

Water2 world avg

m3/ton food

1 644 387 4 055 4 190 1 666 322 967

Animal Products

Resource Region Unit milk eggs bovine poultry pork

Land3 Europe m2/kg food 2 6 31 9 11

m2/kg protein 59 48 258 52 64

Water2 World

avg

m3/ton food 1020 3265 15415 4325 5988

m3/kg protein 31 29 112 34 57

Energy3 Europe MJ/kg protein 68 95 273 96 129

GHG3 Europe kg CO2-e/kg food 1 5 32 7 10

kg CO2-e/kg protein 38 38 170 36 53

We combined data of resource use with data of food consumption per person.

The food consumption data include around 100 food items depending on the

country (FAO, 2013d). We grouped the food items into food categories to

identify the relevant differences of the use of resources. We used seven food

categories for the vegetable products: cereals, roots, pulses, vegetable oils,

sugars, vegetables and fruits. Within each food category, the use of resources of

the food items is very similar due to the physiology of the crops. For example,

for cereals, the global average of land use to produce a kg of wheat, rice and

maize is 3.3 m2, 2.3 m2 and 2 m2 respectively; in contrast, for root is much

lower, to produce a kg of potatoes and cassava is 0.6 m2 and 0.8 m2 respectively

(we calculated these values from crop yield data of 2010 from FAOSTAT (FAO,

2013b)). So, we grouped the staple food items into three categories which show

relative differences among them: cereals, roots and pulses. For the animal

106

products, a more detailed analysis is needed so we grouped the food items into

7 food categories: milk (excluding butter), eggs, bovine meat, poultry meat,

pork meat, fish and animal fats (though no data of resource use is available for

animal fats and fish). We did not group all the meat products into one category

because of the differences in the use of resources among them (see table 6.1).

For example, bovine meat requires three times more land to produce a

kilogram of meat than poultry and pork meat.

For the use of energy and GHG emissions, no data is available for vegetable food

categories. Some studies have calculated the country level data of the use of

energy and GHG emissions for specific vegetable food items (Berners-Lee et al.,

2012; González et al., 2011). They show large differences among specific food

products produced in Europe, e.g. tomatoes produced in heated greenhouses in

the United Kingdom require 130 MJ/kg and tomatoes produced in the open

field in Spain require only 3 MJ/kg (González et al., 2011). With these large

differences within one food item, it is not possible to group the specific

vegetables into one category. So for energy use and GHG emissions, we only

analysed the animal products.

Table 6.1 provides an overview of the use of resources for food. Large variation

exists in resource use between food categories. For vegetable products, pulses

and vegetable oils are the most resource intensive per kilogram of food, then

cereals, and fruits, roots and vegetables are the lowest. The animal products in

general are more resource intensive, and differences exist among the food

categories. Beef is the most resource intensive per kilogram of food, and milk is

the lowest. However, in some cases some vegetable products require more

resources than animal products. For example, the production of a kilogram of

pulses requires more land than the production of a kilogram of poultry meat or

pork meat. The explanation can be found in the data that we used. The data for

animal products refer to a production system of Europe and the data for the

vegetable products refer to the global average. The European production

system is an intensive system and resource efficient in comparison with other

production systems in the world. As a result, with the data that we used, the

animal products are produced in more efficient systems in comparison with the

vegetable products. The animals in these systems are fed with grains produced

with high yields which use less land per kilogram of feed. Elferink & Nonhebel

(2007, fig 2) show that pork, chicken or beef meat produced with feed from

high yield systems requires half of the land than meat produced with feed from


107

global average crop yield. This means that by using the data of table 6.1, we are

showing an underestimation in land use for animal products in respect to the

average global production systems.

The units of table 6.1 are in amount of resources per kilogram of food and per

kilogram of protein which are the units commonly used in studies of resource

use for food. Studies of food consumption patterns commonly use units of

calories or proteins per person per day to discuss diets. So, proteins are the

common ground unit to combine resource use data and food consumption data.

In section 4.2, we calculate the use of resources for the whole diet: the

vegetable products and the animal products, since no data of proteins is

available for the vegetable products, then we use the data of resources per

kilogram of food and combined it with consumption data in kilogram per

person per year. In section 4.2, we only calculate the use of resources for

animal products. Then, we can use the data of resources per kilogram of protein

and combine it with consumption data in protein per person per day.

Food consumption data

We discuss the differences in diets using country level food supply data of the

Food Balance Sheets of the FAO (FAO, 2013d). We use data of kcal/cap/day

which are the units most commonly used to describe food consumption. We

aggregated the daily caloric supply per capita of all countries in each region to

obtain the average caloric supply of the region. We included 14 food categories,

the 12 food categories included in table 6.1 and fish and animal fats which are

relatively relevant for the diets of some regions. We discuss the changes in the

last five decades for which we used data of 1961 to discuss the year 1960, and

data of 2009 to discuss the year 2010.

The food supply data is not the actual diet of the population since it includes

food losses throughout the food chain. These food losses are different among

regions. For example, for developing countries the food losses are mainly on-

farm and during transport and food processing, and in developed countries the

food losses are mainly in the retail and food service and at home (Godfray et al.,

2010a). A more accurate source of data would be to use household level

surveys. Unfortunately, these surveys are only on a country level, and not for a

global level. However, Gerbens-Leenes et al. (2010, fig7) shows a clear relation

108

between food supply data and household level surveys. For this reason,

throughout the paper, we refer to diets when using food supply data.

Differences in dietary composition

We calculated mathematically the differences in dietary composition among

regions and compared it with the change in time in dietary composition for

each region to show whether the global or regional dietary trend is stronger. To

do this, we used the equation of Euclidean distance between two points:

( ) √∑ ( )

where p and q are two dietary composition of figure 6.2 and k is the number of

dimension of these points which in our case are the food categories, seven for

both the vegetable products and the animal products. The distance gives an

indication of the differences between the dietary composition p and q and we

use it for comparison. First, we calculated the differences in time. So the change

in dietary composition from 1960 to 2010 for every region using:

( ) ( ) √∑ ( )

eq. 1

where p is the region, pt1is the dietary composition in 1960, pt2 is the dietary

composition in 2010, and k are the food categories. Then, we calculated the

difference among regions for each year, 1960 and 2010, using the following

equation:

( ) √∑ ( )

eq. 2

where p and q are the dietary composition of regions i and j in the same year,

and k are the food categories. With equation 1, we obtain a vector ∆ti which

indicate the change from 1960 to 2010, and with equation 2 we obtain a

symmetric matrix dij indicating the differences among regions.


109

6.3 Global and regional dietary paths: what can we expect for

future regional diets?

In this section, we evaluate the regional changes in diets in the period of 1960

to 2010 to assess future developments in diets. The changes have been driven

by two trends: the global dietary trend (following the nutrition transition

theory) and the specific-regional trends (following the regional food patterns).

Both trends have been discussed by Alexandratos & Bruinsma (2012). Most

developed countries have reached a saturated level (last phase of the nutrition

transition) and diets have been relatively stable; in contrast, developing

countries have changed fast in the last decades by a shift towards livestock

products and vegetable oils. This reflects the global trend. But, as mentioned

before, the countries also show country-specific deviations of this trend. For

example, the increase of meat consumption in China and Brazil was faster than

in other developing countries, and in India it was much more slower for a

similar economic development stage (Alexandratos & Bruinsma, 2012). We

study the regional and global trends to find out which has been stronger.

Figure 6.1 shows the daily caloric intake of each region in 1960 and in 2010 for

both vegetable and animal products for the 13 regions and the global average.

This figure shows that regions largely differ from the global average and among

each other. Also, large changes in diets have occurred during this period, and

each region changed differently. The changes have been in both total caloric

consumption and in food composition. To identify the dietary paths we look

only at the food composition of the diets which is illustrated in figure 6.2. Note

that the numbers in figure 6.2 are relative and the difference in total caloric

consumption among regions and among the years 1960 and 2010 can be

strong. For instance, the caloric consumption of animal products in E-M-W

Africa is less than 200 kcal/cap/day and in N America is 1000 kcal/cap/day.

And, the caloric consumption of animal products in China in 1960 is less than

100 kcal/cap/day and in 2010 is more than 600 kcal/cap/day.

In one hand, to identify the regional dietary path we need to compare the

dietary composition of 1960 and 2010 for each region using figure 6.2. If the

dietary composition in both years remains the same, then it means that it

followed its regional path, even if the amount of calories changed. The

consumption of animal products in China is a good example for which the

110

caloric consumption increased 12 times (figure 6.1), but the composition is

relatively similar in both years mainly with pork and eggs (figure 6.2).

In the other hand, to identify the global dietary path, we need to compare the

dietary compositions among the regions. If the dietary compositions are

becoming more similar among each other, it means that regions are changing

towards a global food patterns. This is more difficult to identify just by looking

at figure 6.2, so we use equation 1 and 2 to quantify the differences and identify

which trend is stronger. But before these calculations, we discuss three global

trends and the deviations among the regions.

Figure 6.1 Global differences in diets. The graphs show the daily caloric intake in

kcal/cap/day of the vegetable products and the animal products (next page) in 1960 and

2010. See text for details.


111

Figure 6.1 (continue) Global differences in diets.

First, a global trend during this period was the reduction in the per capita

consumption of staple food (roots, pulses and cereals). However, regions

changed differently and the share of this staple food in their diet is relatively

similar in 1960 and in 2010 (figure 6.2). The large cereals’ consumers, N Africa,

S Africa and S-E Asia, are still relative large consumers compared with the other

regions. E-M-W Africa, E Europe and China are relative large consumer of roots.

They all changed differently during this period, China even decreased, but they

all remained large consumers in relation to the other regions. India and C-S

America are the largest consumers of pulses, and even though they decreased

their consumption, they are still the relative largest consumers.

112

Second, the global per capita consumption of affluent vegetable products

(vegetable oils, vegetables, sugars and fruits) increased. The largest global

increase in the per capita consumption was for vegetable oils and vegetables

which doubled, fruits increased 70% and sugars “only” 20% (FAO, 2013d).

Each region did not increase the consumption of all four products in the same

way; and, again, the share of these food items in the diet is similar in both years

as follows (figure 6.2). Looking at sugars and vegetable oils, N Africa, E Europe,

America and Oceania show a larger consumption of sugars than vegetable oils;

in contrast, E-M-W Africa, China and S Europe show a larger consumption of

vegetable oils than sugars. In the same way for fruits and vegetables, Africa,

America and S-E and W-C Asia have a larger consumption of fruits than

vegetables; in contrast, China and E Europe have a larger consumption of

vegetables than fruits.

Figure 6.2 Regional differences in dietary composition. The graphs show the food

composition of the daily caloric intake of the vegetable products (top) and the animal

products (bottom) in 1960 and in 2010.


113

Third, the consumption of animal products became more diversified (figure

6.2). In general, the consumption of poultry meat increased in all regions,

strongly in S Africa, America, S-E and C-W Asia and Oceania, but still the change

was different following their dietary composition. N Africa, India and W-C Asia

are the relative largest consumers of milk for which milk accounts to more than

half of their animal products calories, even though the per capita consumption

of milk in other regions is larger: N America and W-N Europe. All Africa and

America are the relative largest beef consumers. Even though in some regions

(S Africa and all America) poultry meat replaced part of the beef consumption

during this period. China is by far the largest consumer of pork, but also Europe

and S-E Asia are relative large consumers. The largest consumers of animal fats

are India, all Europe and Oceania. Finally, S-E Asia and E-M-W Africa are the

relative largest fish consumers.

Table 6.2 Differences in dietary composition

These tables show the results of equations 1 and 2 based on the data of figure 6.2.

6.2a.Change in time (∆t) and differences among regions (dij). The first row for vegetable and animal products shows the average values of vector ∆t and matrix dij, in 1960 and in 2010. The following rows show the values of vector ∆t, and the average distance of each region in relation to the other regions (avg dij). The complete matrix dij is shown in table 6.2b where the avg dij is shown in the last column. See text for details.

Vegetable products Animal products

1960 2010 1960 2010

Average values:

│∆t│= 0.13

│dij│= 0.23

│dij│= 0.21

Average values: │∆t│= 0.18

│dij│= 0.30

│dij│= 0.29

∆t Avg.dij Avg. . dij

∆t Avg. . dij Avg. . dij

E-M-W Africa 0,04 0,24 0,24 E-M-W Africa 0,13 0,29 0,26 S Africa 0,10 0,20 0,18 S Africa 0,34 0,30 0,33 N Africa 0,05 0,20 0,20 N Africa 0,17 0,25 0,33 N America 0,12 0,31 0,25 N America 0,16 0,20 0,20 C-S America 0,10 0,20 0,18 C-S America 0,22 0,28 0,23 China 0,20 0,26 0,23 China 0,16 0,52 0,54 India 0,09 0,21 0,18 India 0,15 0,34 0,38 S-E Asia 0,11 0,23 0,20 S-E Asia 0,15 0,34 0,27 W-C Asia 0,12 0,21 0,16 W-C Asia 0,24 0,37 0,29 E Europe 0,18 0,18 0,17 E Europe 0,13 0,22 0,21 S Europe 0,22 0,18 0,22 S Europe 0,17 0,22 0,22 W-N Europe 0,11 0,25 0,21 W-N Europe 0,12 0,29 0,24 Oceania 0,25 0,28 0,23 Oceania 0,22 0,26 0,23

114

Table 6.2 (continue)

6.2b. Detail differences among regions: matrix dij

Vegetable products

Animal products

t=1960

│dij│=0.23 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g t=2010

│dij│=0.21 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g

E-M-W Africa 0 0,28 0,26 0,38 0,21 0,10 0,26 0,26 0,23 0,19 0,18 0,25 0,34 0,24 E-M-W Africa 0 0,23 0,23 0,36 0,21 0,21 0,22 0,23 0,20 0,17 0,26 0,27 0,29 0,24

S Africa 0,28 0 0,08 0,37 0,21 0,28 0,12 0,13 0,14 0,10 0,15 0,29 0,27 0,20 S Africa 0,23 0 0,08 0,36 0,19 0,14 0,06 0,06 0,07 0,16 0,27 0,27 0,30 0,18

N Africa 0,26 0,08 0 0,42 0,24 0,25 0,10 0,08 0,08 0,11 0,14 0,32 0,33 0,20 N Africa 0,23 0,08 0 0,40 0,22 0,11 0,06 0,04 0,09 0,19 0,30 0,31 0,34 0,20

N America 0,38 0,37 0,42 0 0,22 0,44 0,40 0,47 0,41 0,36 0,31 0,13 0,16 0,31 N America 0,36 0,36 0,40 0 0,19 0,42 0,37 0,41 0,31 0,23 0,15 0,10 0,07 0,25

C-S America 0,21 0,21 0,24 0,22 0 0,25 0,21 0,28 0,23 0,17 0,15 0,13 0,15 0,20 C-S America 0,21 0,19 0,22 0,19 0 0,26 0,19 0,23 0,15 0,08 0,15 0,11 0,14 0,18

China 0,10 0,28 0,25 0,44 0,25 0 0,24 0,23 0,24 0,19 0,21 0,31 0,37 0,26 China 0,21 0,14 0,11 0,42 0,26 0 0,12 0,10 0,13 0,21 0,31 0,33 0,36 0,23

India 0,26 0,12 0,10 0,40 0,21 0,24 0 0,14 0,13 0,15 0,16 0,32 0,32 0,21 India 0,22 0,06 0,06 0,37 0,19 0,12 0 0,06 0,07 0,17 0,28 0,29 0,31 0,18

S-E Asia 0,26 0,13 0,08 0,47 0,28 0,23 0,14 0 0,12 0,12 0,19 0,37 0,37 0,23 S-E Asia 0,23 0,06 0,04 0,41 0,23 0,10 0,06 0 0,10 0,20 0,30 0,32 0,35 0,20

W-C Asia 0,23 0,14 0,08 0,41 0,23 0,24 0,13 0,12 0 0,14 0,12 0,31 0,34 0,21 W-C Asia 0,20 0,07 0,09 0,31 0,15 0,13 0,07 0,10 0 0,12 0,21 0,23 0,25 0,16

E Europe 0,19 0,10 0,11 0,36 0,17 0,19 0,15 0,12 0,14 0 0,12 0,25 0,26 0,18 E Europe 0,17 0,16 0,19 0,23 0,08 0,21 0,17 0,20 0,12 0 0,16 0,13 0,17 0,17

S Europe 0,18 0,15 0,14 0,31 0,15 0,21 0,16 0,19 0,12 0,12 0 0,20 0,26 0,18 S Europe 0,26 0,27 0,30 0,15 0,15 0,31 0,28 0,30 0,21 0,16 0 0,10 0,10 0,22

W-N Europe 0,25 0,29 0,32 0,13 0,13 0,31 0,32 0,37 0,31 0,25 0,20 0 0,15 0,25 W-N Europe 0,27 0,27 0,31 0,10 0,11 0,33 0,29 0,32 0,23 0,13 0,10 0 0,03 0,21

Oceania 0,34 0,27 0,33 0,16 0,15 0,37 0,32 0,37 0,34 0,26 0,26 0,15 0 0,28 Oceania 0,29 0,30 0,34 0,07 0,14 0,36 0,31 0,35 0,25 0,17 0,10 0,03 0 0,23

t=1960

│dij│=0.23 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g t=2010

│dij│=0.21 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g


S Africa 0,28 0 0,08 0,37 0,21 0,28 0,12 0,13 0,14 0,10 0,15 0,29 0,27 0,20 S Africa 0,23 0 0,08 0,36 0,19 0,14 0,06 0,06 0,07 0,16 0,27 0,27 0,30 0,18

N Africa 0,26 0,08 0 0,42 0,24 0,25 0,10 0,08 0,08 0,11 0,14 0,32 0,33 0,20 N Africa 0,23 0,08 0 0,40 0,22 0,11 0,06 0,04 0,09 0,19 0,30 0,31 0,34 0,20



China 0,10 0,28 0,25 0,44 0,25 0 0,24 0,23 0,24 0,19 0,21 0,31 0,37 0,26 China 0,21 0,14 0,11 0,42 0,26 0 0,12 0,10 0,13 0,21 0,31 0,33 0,36 0,23

India 0,26 0,12 0,10 0,40 0,21 0,24 0 0,14 0,13 0,15 0,16 0,32 0,32 0,21 India 0,22 0,06 0,06 0,37 0,19 0,12 0 0,06 0,07 0,17 0,28 0,29 0,31 0,18

S-E Asia 0,26 0,13 0,08 0,47 0,28 0,23 0,14 0 0,12 0,12 0,19 0,37 0,37 0,23 S-E Asia 0,23 0,06 0,04 0,41 0,23 0,10 0,06 0 0,10 0,20 0,30 0,32 0,35 0,20

W-C Asia 0,23 0,14 0,08 0,41 0,23 0,24 0,13 0,12 0 0,14 0,12 0,31 0,34 0,21 W-C Asia 0,20 0,07 0,09 0,31 0,15 0,13 0,07 0,10 0 0,12 0,21 0,23 0,25 0,16

E Europe 0,19 0,10 0,11 0,36 0,17 0,19 0,15 0,12 0,14 0 0,12 0,25 0,26 0,18 E Europe 0,17 0,16 0,19 0,23 0,08 0,21 0,17 0,20 0,12 0 0,16 0,13 0,17 0,17

S Europe 0,18 0,15 0,14 0,31 0,15 0,21 0,16 0,19 0,12 0,12 0 0,20 0,26 0,18 S Europe 0,26 0,27 0,30 0,15 0,15 0,31 0,28 0,30 0,21 0,16 0 0,10 0,10 0,22


Oceania 0,34 0,27 0,33 0,16 0,15 0,37 0,32 0,37 0,34 0,26 0,26 0,15 0 0,28 Oceania 0,29 0,30 0,34 0,07 0,14 0,36 0,31 0,35 0,25 0,17 0,10 0,03 0 0,23

t=1960

│dij│=0.30 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g t=2010

│dij│=0.29 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g


S Africa 0,11 0 0,24 0,25 0,06 0,54 0,43 0,36 0,46 0,27 0,26 0,34 0,28 0,30 S Africa 0,29 0 0,42 0,22 0,15 0,49 0,54 0,29 0,35 0,30 0,29 0,35 0,24 0,33

N Africa 0,21 0,24 0 0,15 0,23 0,59 0,22 0,35 0,25 0,16 0,13 0,28 0,17 0,25 N Africa 0,28 0,42 0 0,25 0,30 0,73 0,24 0,40 0,08 0,26 0,32 0,34 0,30 0,33



China 0,54 0,54 0,59 0,45 0,49 0 0,69 0,37 0,71 0,46 0,47 0,40 0,53 0,52 China 0,54 0,49 0,73 0,52 0,51 0 0,74 0,38 0,68 0,49 0,42 0,44 0,52 0,54

India 0,38 0,43 0,22 0,29 0,41 0,69 0 0,45 0,06 0,27 0,24 0,38 0,31 0,34 India 0,37 0,54 0,24 0,33 0,40 0,74 0 0,43 0,26 0,28 0,35 0,31 0,33 0,38

S-E Asia 0,31 0,36 0,35 0,28 0,34 0,37 0,45 0 0,48 0,28 0,25 0,31 0,34 0,34 S-E Asia 0,21 0,29 0,40 0,22 0,23 0,38 0,43 0 0,36 0,18 0,13 0,18 0,21 0,27

W-C Asia 0,40 0,46 0,25 0,32 0,44 0,71 0,06 0,48 0 0,31 0,27 0,42 0,35 0,37 W-C Asia 0,25 0,35 0,08 0,18 0,23 0,68 0,26 0,36 0 0,21 0,28 0,30 0,24 0,29

E Europe 0,26 0,27 0,16 0,07 0,23 0,46 0,27 0,28 0,31 0 0,08 0,13 0,12 0,22 E Europe 0,19 0,30 0,26 0,11 0,17 0,49 0,28 0,18 0,21 0 0,08 0,10 0,13 0,21

S Europe 0,22 0,26 0,13 0,08 0,23 0,47 0,24 0,25 0,27 0,08 0 0,20 0,18 0,22 S Europe 0,18 0,29 0,32 0,15 0,18 0,42 0,35 0,13 0,28 0,08 0 0,09 0,15 0,22


Oceania 0,30 0,28 0,17 0,15 0,27 0,53 0,31 0,34 0,35 0,12 0,18 0,16 0 0,26 Oceania 0,20 0,24 0,30 0,12 0,13 0,52 0,33 0,21 0,24 0,13 0,15 0,16 0 0,23

t=1960

│dij│=0.30 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g t=2010

│dij│=0.29 E-M

-W A

fric

a

S A

fric

a

N A

fric

a

N A

me

rica

C-S

Am

eri

ca

Ch

ina

Ind

ia

S-E

Asi

a

W-C

Asi

a

E E

uro

pe

S E

uro

pe

W-N

Eu

rop

e

Oce

an

ia

Av

g


S Africa 0,11 0 0,24 0,25 0,06 0,54 0,43 0,36 0,46 0,27 0,26 0,34 0,28 0,30 S Africa 0,29 0 0,42 0,22 0,15 0,49 0,54 0,29 0,35 0,30 0,29 0,35 0,24 0,33

N Africa 0,21 0,24 0 0,15 0,23 0,59 0,22 0,35 0,25 0,16 0,13 0,28 0,17 0,25 N Africa 0,28 0,42 0 0,25 0,30 0,73 0,24 0,40 0,08 0,26 0,32 0,34 0,30 0,33



China 0,54 0,54 0,59 0,45 0,49 0 0,69 0,37 0,71 0,46 0,47 0,40 0,53 0,52 China 0,54 0,49 0,73 0,52 0,51 0 0,74 0,38 0,68 0,49 0,42 0,44 0,52 0,54

India 0,38 0,43 0,22 0,29 0,41 0,69 0 0,45 0,06 0,27 0,24 0,38 0,31 0,34 India 0,37 0,54 0,24 0,33 0,40 0,74 0 0,43 0,26 0,28 0,35 0,31 0,33 0,38

S-E Asia 0,31 0,36 0,35 0,28 0,34 0,37 0,45 0 0,48 0,28 0,25 0,31 0,34 0,34 S-E Asia 0,21 0,29 0,40 0,22 0,23 0,38 0,43 0 0,36 0,18 0,13 0,18 0,21 0,27

W-C Asia 0,40 0,46 0,25 0,32 0,44 0,71 0,06 0,48 0 0,31 0,27 0,42 0,35 0,37 W-C Asia 0,25 0,35 0,08 0,18 0,23 0,68 0,26 0,36 0 0,21 0,28 0,30 0,24 0,29

E Europe 0,26 0,27 0,16 0,07 0,23 0,46 0,27 0,28 0,31 0 0,08 0,13 0,12 0,22 E Europe 0,19 0,30 0,26 0,11 0,17 0,49 0,28 0,18 0,21 0 0,08 0,10 0,13 0,21

S Europe 0,22 0,26 0,13 0,08 0,23 0,47 0,24 0,25 0,27 0,08 0 0,20 0,18 0,22 S Europe 0,18 0,29 0,32 0,15 0,18 0,42 0,35 0,13 0,28 0,08 0 0,09 0,15 0,22


Oceania 0,30 0,28 0,17 0,15 0,27 0,53 0,31 0,34 0,35 0,12 0,18 0,16 0 0,26 Oceania 0,20 0,24 0,30 0,12 0,13 0,52 0,33 0,21 0,24 0,13 0,15 0,16 0 0,23


115

Now, we use equation 1 and 2 to quantify whether the global or regional

dietary trend was stronger during this period. The results are shown in tables

6.2 where dij shows the differences in dietary composition among regions and

∆t shows the change in time in dietary composition for each region. The

numbers of these tables do not measure a real value, only show an indication of

the differences of dietary composition and we use them for comparison. The

largest the number, the largest the difference of dietary composition between

two region or throughout the period.

The first row of table 6.2a shows the average of ∆t and dij. For both vegetable

and animal products, the average ∆t is smaller than dij. This means that the

difference among regions (dij)is larger than the change in time within the

regions (∆t). So, in general, the regional trend of dietary pattern is stronger

than the global trend, and regions have followed their own dietary composition.

The average ∆t and dij show that │dij│ in 1960 was larger than in 2010. This

means that in 2010 the dietary composition of the regions are more similar

among each other than in 1960 indicating that regions changed towards a more

similar food consumption (following the global trend). However, this trend of

conversion to a global diet is slow. In the 50 years of study (1960 to 2010) the

differences among regions only changed from 0.23 to 0.21 for vegetable

products and from 0.30 to 0.29 for animal products. The differences of dietary

composition of the animal products is stronger than of the vegetable products

suggesting that regional patterns of animal products are stronger or more

evident than of vegetable products.

Some regions changed more than others which is indicated in the values of ∆t.

E-M-W Africa and S Africa changed very little their vegetable products

composition: their values (0.04 and 0.5 respectively) are much lower than

│∆t│. In contrast, S Europe and Oceania changed a lot. For animal products, S

Africa was the region with the stronger changes, and E-M-W Africa, E Europe

and W-N Europe were the regions with the lowest change.

The columns “avg dij” of table 6.2a indicate the average distance between each

region and the rest. This is the average of each row of dij (see the last columns

of table 6.2b). For most regions, the avg dij is larger than ∆t indicating that the

difference of that region in comparison with the rest is larger than the change

in time. Showing again that for those regions the regional dietary trend was

stronger than the global trend. However, for some regions, the change in time

in food composition was stronger (see red numbers): the change of vegetable

116

products in E Europe, S Europe and Oceania was stronger than the differences

of these region in respect with the rest, and the change in animal products in S

Africa was stronger than the difference of this region with the rest.

Thus, from this analysis, we can expect that future changes in diets will follow

regional dietary paths if we assume that past trends remain in future. Based on

this finding, in the next section we assess the impact of these regional dietary

trends on the use of agricultural resources.

6.4 Future use of agricultural resources:

will regional differences in dietary composition have an

impact?

Large awareness in the literature is focused on the large increase in use of

agricultural resources due to future dietary changes. As we have shown,

regions are following different dietary paths. In this section, we discuss the

impact of the different regional dietary paths on the use of resources. To have a

reference with the present situation, first, we discuss the impact of the present

diets on the use of resources and then we discuss the impact of future dietary

changes.

Impact of present diets

Here, we evaluate the impact of the differences in diets shown in figure 6.1

using the production data of table 6.1. Figure 6.3 shows the use of land, water,

energy and GHG emissions for the diets of each region in 2010. The use of land

and water shows that the largest difference in resource use is caused by the

consumption of animal products, and the use of resources for vegetable

products is relatively similar among the regions (see the green bars in figure

6.3). For energy and GHG emissions no data is available for the specific

vegetable food categories.


117

Figure 6.3 Impact of the regional dietary differences on the use of agricultural

resources.

The differences in total use of resource per person are enormous. The average

diet of N America requires more than three times of land and water than the

average diet of E-M-W Africa and India. The animal products consumption per

capita of N America requires ten to twelve times more land, water, energy, and

emits ten times more GHG than the per capita animal products consumption of

E-M-W Africa and India. This huge difference is mainly due to the very low

consumption of animal products in E-M-W Africa and India. But, the differences

are also large by comparing with a diet with medium consumption of animal

118

products like China and E Europe. A person in N America consumes three times

more land, water, energy respectively and emits three times more GHG than a

person in China, and two times more than a person in E Europe.

Thus, the largest differences in the use of resources are caused by the strong

differences in the consumption of animal products among the regions. These

differences are due to both the caloric intake and food composition (the type of

animal products), and, as figure 6.1 shows, these differences are enormous. The

consumption of animal products is expected to strongly increase specially in

developing countries (Alexandratos & Bruinsma. 2012) which currently

consume relative small amount of animal products. So, to discuss the impact of

resources in future diets, we focus on the consumption of animal products.

Impact of future diets following regional dietary paths

As mentioned before, diets are expected to change in the coming decades

especially in developing countries due to the increase of socioeconomic

development, globalization of the food system and urbanization. A clear

relation has been shown between the increase of income (GDP per capita) and

the consumption of animal products (Gerbens-Leenes et al., 2010; Poleman &

Thomas, 1995). For income values below US$10,000 per capita, a linear

relation has been shown for all countries between the increase of income and

the increase consumption of animal products. This relation starts with very low

values of income and very low consumption of animal products, and reaches

around 1000 kcal/cap/day of animal products with income values of

US$10,000 per capita. Then, with higher values of income, the consumption of

animal products does not change and the diets are relatively stable. This means,

that a saturation level is reached at around 1000 kcal/cap/day of animal

products consumption. Most developed countries already show this saturation

level (see W-N Europe and N America in figure 6.1) and their diets have been

relatively stable.

In the coming decades, developing countries will increase their income and

they are expected to follow this trend between income and consumption of

animal food products. However, as we have shown in section 3, the type of

animal products will be different among the regions in accordance to their

current food composition. To evaluate the future impact in resource use due to

the different choice of type of animal products, we assume that all regions


119

consume the same amount of animal products: 60 gr of protein/cap/day (which

is the saturation level of around 1000 kcal/cap/day) but with the food

composition of each region of 2010 (see figure 6.4). We do this to analyse the

impact on the use of resources if all region change to an affluent consumption

reaching a saturation level similar to the developed countries. We use 60 gr of

protein which is the per capita consumption of animal products of the

European Union in 2010 (FAO, 2013d), and we assume that this is the

saturation level that all region can reach with an increase of income. We use

protein consumption instead of caloric consumption because the values of

resource use per type of food (table 1) are in amount of resource per kilogram

of protein.

Figure 6.4 Daily consumption of 60 gr/cap/day of animal products with different

regional dietary composition. The food composition is based on the values of 2010 of the

Food Balance Sheets (FAO, 2013d).

Figure 6.5 shows the use of resources for the consumption of 60 gr of

protein/cap/day of animal products. The differences among regions are smaller

than the differences shown in figure 6.3, but still are relevant showing the large

impact of the choice of the type of animal product in each region. The regions

with the largest use of resources: E-M-W Africa, C-S America and Oceania, use

60% to 80% more resources per person than the regions with the lowest use of

resources: India and China for the same protein consumption. This is mainly

120

related with the consumption of bovine meat. In general, the share of resources

for bovine meat production is large: a lot of red is shown in figure 6.5, even

though the share of bovine meat in the protein consumption is not as large: a

few red is shown in figure 6.4. This is due to the high requirement of resources

per kilogram of protein of bovine meat. As shown in table 6.1, bovine meat

requires 3-5 times more resources per kilogram of protein than the other

animal products.

Figure 6.5 Use of resources for the consumption of 60gr protein/cap/day with the

differences of regional food composition of figure 6.4.


121

It is widely known that diets with large consumption of meat require a lot of

resources. However, one interesting finding of figure 6.5 is that the animal

products food pattern of E-M-W Africa and C-S America uses more resources

than “western” regions like N America and W-N Europe. Also, the use of

resources (specially of land use and GHG emissions) between India and China is

almost the same, even though the consumption of meat in India is very low.

These two findings can be explained with the efficiency of the animal products

(table 6.1). E-M-W Africa and C-S America have the largest share of beef in their

food composition, which is the animal product requiring the largest amount of

resources per amount of food produced. China and India show very different

composition of their animal products consumption: 60% of the protein

consumption in China is meat and in India only 10% and the rest are dairy

products. The largest share of the meat consumption in China is pork and

poultry, also eggs consumption account to 15% of the protein consumption.

The production of one protein of pork, poultry and eggs requires similar

amount or in some cases less amount of resources than the production of one

protein of milk (table 6.1). As a result, the animal protein consumption of India

with almost no meat and only milk, does not result in lower use of resources in

comparison with a consumption of large amount of meat like in China. Thus, the

type of meat makes a big difference.

6.5 Discussion

In this study, we have used data from several sources and we have made

several assumptions which should be discussed to understand our results in

perspective. We discuss four general points:

1. The data to calculate the use of resources was gathered from one production

system (table 6.1). We assumed that this production system was used in all

regions. By doing this, we are not calculating the real use of resources of the

region, what some call “footprint”. But, we are able to compare only the

differences in diets, and not the differences of production systems, which is the

goal of this paper.

2. The production system that we chose is high resource intensive for bovine

meat which highly influences the use of resources for diets with high beef

consumption. However, in general, all studies in the literature show that beef is

the type of meat requiring most resource. This is due to the physiology of the

122

animals: ruminants animals (cattle) are much lower efficient in producing a

kilogram of food protein than non-ruminants animals (chicken and pig).

Therefore, the absolute values of table 6.1 might be different for other

production systems, but the relation among the type of animal food products is

in general the same. For example, the values of other study by Elferink (2009,

pag 68) show similar relation among the type of animal products. So, the

general differences on the use of resources among the regions that we have

shown can also be expected for other production systems.

3. The aim of this study is to do a global analysis. We used global average data

of production systems when it was available. For the production system of

animal products, a global average was available only for the use of water, for

the other resources we used a European production system. The use of water is

highly dependent on the climate and it largely differs from country to country

(Mekonnen & Hoekstra, 2011a). The European production systems are relative

intensive systems with high yields and high energy inputs in comparison with

other systems of the world. As a result, these systems are land efficient. So our

results of land use are an underestimation in comparison with the global

average, and for the use of energy and GHG emissions are an overestimation in

comparison with the global average.

4. We used food supply data (FAO, 2013d) for the diets of each region. These

values are not the actual food consumption of the population, since they include

food losses. But, they are an indication of the consumption pattern and they are

useful for comparison among countries (Gerbens-Leenes et al., 2010). Also,

since we discuss the use of resources for the production of food, then the

resources needed to produce the food losses should also be included.

Thus, considering these four assumptions, our results should be interpreted

qualitatively and not quantitatively. The differences among regions and the

general trends in time are the aim of this study and not the exact values of our

calculations.

Most studies of future food security and use of resources use the nutrition

transition theory for assessing future dietary changes: change from staple food

to sugars, vegetable oils and animal products. However, this trend is too rough

to assess the impact of dietary changes on the use of resources because of the

large differences among the food categories and food items discussed in this

paper (table 6.1). For example, the differences in resource use among staple


123

crops are large as well as the differences among the type of animal products. As

a result, the composition of the diet has a strong impact on the use of resources.

So, it is essential to differentiate among these food items to assess future use of

resources. Kastner et al. (2012) calculated the land requirement for food for all

regions. They did not differentiate among the type of animal products but they

assumed that the production of a calorie of any animal product requires the

same amount of land. We have shown that the present regional differences in

the type of animal products consumption can result in almost twice use of

resources for the same protein consumption (E-M-W Africa VS India). So, the

results of Kastner et al. (2012) could be an underestimation depending on the

type of animal products consumption of the region.

The present discussion in the literature on the future sustainability of the

global food system focuses on the strong impact of dietary changes in transition

countries towards western consumption patterns. First, we have shown that

there is not one “Western food pattern”. The regions commonly called as

“western” show different food patterns, e.g. N America and W-N Europe show

different food composition. Second, we have shown that developing regions

have changed to affluent consumption (e.g. increased livestock products) but

they have not followed the exact western pattern but their own food pattern.

Assuming that this trend remains, the impact on the use of resources of some

developing regions could be larger than the use of resources of the western

countries. For example, the food pattern of animal products of E-M-W Africa

and S-C America requires more resources than the food pattern of N America

and W-C Europe. Also, vegetarian diets which are commonly considered as low

resource intensive could use similar or more resources than diets with large

amount of meat. For example, the animal products food pattern of China and

India requires similar amount of land even though in China 60% of the animal

product’s protein intake is meat and in India only 10% and the rest are dairy

products.

These differences in resource use for regional differences in animal products

consumption are due to the differences in resource efficiencies of the food

items. Beef is the meat requiring more resources per protein produced, as most

of the literature has shown. So, E-M-W Africa and S-C America, which have the

highest relative consumption of beef, result with the highest use of resources.

Also, the production of one protein of pork or chicken is relatively the same as

the production of one protein of milk. So, vegetarian diets with large

124

consumption of dairy products do not necessarily require less resources than

diets with large consumption of meat like chicken or pork.

The analysis done in this study shows that large reduction of resource use for

food can be obtained through changing from beef to pork or poultry

consumption. In some cases, resource use can be halved. But in several regions

in the world a change from beef to pork will not be a socially acceptable option

due to cultural and religious traditions.

Appendix 6

The 13 Regions used in this paper and their countries

1.- E-M-W Africa: Eastern, Middle and Western Africa. Countries: Angola, Benin, Burkina Faso, Burundi, Cabo Verde, Cameroon, C. African Rep., Chad, Comoros, Congo, Côte d'Ivoire, Djibouti, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Niger, Nigeria, Rwanda, Senegal, Seychelles , Sierra Leone, Togo, Uganda, U. Rep. of Tanzania, Zambia and Zimbabwe. 2.- S Africa: South Africa. Countries: Botswana, Lesotho, Namibia, South Africa and Swaziland. 3.- N Africa: North Africa. Countries: Algeria, Egypt, Libya, Morocco, Sudan and Tunisia 4.- N America: North America Countries: Bermuda, Canada and United States of America 5.- C-S America: Central and South America, and the Caribbean Countries: Antigua & Barbuda, Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, Dominica, Dom. Rep., Ecuador, El Salvador, Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, St Kitts & Nevis, St Lucia, St Vincent & the Grenadines, Suriname, Trinidad and Tobago, Uruguay and Venezuela. 6.- China 7.- India 8.- S-E Asia: Southern, Eastern and Southeast Asia


125

Countries: Bangladesh, Brunei Darussalam, Cambodia, Dem. Rep. Korea, Indonesia, Iran, Japan, Lao, Malaysia, Maldives, Mongolia, Myanmar, Nepal, Pakistan, Philippines, Rep. Korea, Sri Lanka, Thailand, Timor-Leste and Vietnam. 9.- W-C Asia: Western and Central Asia Countries: Armenia, Azerbaijan, Cyprus, Georgia, Israel, Jordan, Kazakhstan, Kuwait, Kyrgyzstan, Lebanon, Palestina, Saudi Arabia, Syrian A. Rep, Tajikistan, Turkey, Turkmenistan, United Arab Emirates, Uzbekistan and Yemen. 10.-E Europe: Eastern Europe Countries: Belarus, Bulgaria, Czechoslovakia (1960), Czech Rep. (2010), Hungary, Poland, Rep. of Moldova, Romania, USSR (1960), Russian Federation (2010), Slovakia and Ukraine. 11.- S Europe: Southern Europe Countries: Albania, Bosnia & Herzegovina, Croatia, Greece, Italy, Malta, Portugal, Slovenia, Spain, Yugoslav SFR (1960) and Rep. of Macedonia (2010). 12.- W-N Europe: Western and Northern Europe Countries: Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Iceland, Ireland, Latvia, Lithuania, Luxembourg, Netherlands, Norway, Sweden, Switzerland and United Kingdom. 13.- Oceania Countries: Australia, Fiji, French Polynesia, Kiribati, New Caledonia, New Zealand, Samoa, Solomon Islands and Vanuatu.

Chapter 7. Identifying challenges for a sustainable future

127

Chapter 7.

Identifying challenges for a sustainable future of the

global food system

7.1 More food, more resources

In the past decades, the global food system has undergone fast changes. These

changes have been caused by population growth, urbanization, globalization of

the food system, increase of income levels, dietary changes, change in

production systems (green revolution), among others (Alexandratos &

Bruinisma, 2012). All of this has resulted in an increase of food demand which

has led to an increase demand for resources: mainly land, water and energy.

The increase of food demand and resources has been different among regions

due to different demographic, geographic and cultural situations. It is expected

that more resources will be required in the coming decades due to increase in

food demand which, again, will be different among regions. Many studies

address the need for a sustainable future of the global food system (Foley et al.,

2011; Godfray et al., 2010a, 2010b; Tilman et al., 2011).

The goal of this thesis was to analyse the sustainability of the global food

system. A sustainable food system should provide enough food for all people

with the lowest environmental impact. The assessment focuses on the impact of

food demand on the use of agricultural resources considering the dynamics and

regional diversity of population numbers, diets and agricultural systems.

Agricultural production requires a mixture of resources which are interrelated

(land, water, nutrients and labour). For instance, a strong trade-off exists

between nitrogen fertilizer and land, and between labour and machinery. The

lowest environmental impact of the food system has been assessed by

considering an efficient use and the major trade-offs of the agricultural

resources. Throughout chapters 2 to 6, the main trends of the last decades and

regional differences in the use of resources have been studied in detail. In this

last chapter, all the insights obtained throughout the thesis are integrated to

discuss the future sustainability of the global food system.

128

7.2 General methodology: integrative assessment as a

response to the existing literature

A rapidly growing line of research has emerged in the last decades studying the

urge for a sustainable future of the global food system. These studies can be

grouped into two main lines of research which have some limitations that

motivated the development of a new approach used in this thesis. The two lines

of research are described below.

The first line of research are global studies that assess the future use of

resources identifying challenges and sustainable solutions (Fresco, 2009;

Godfray et al., 2010a; Tilman et al., 2011). They discuss the future use of

resources based on average global trends, account for all main resources, and

discuss the impact of dietary changes (based on the nutrition transition theory)

and population growth (based on the demographic transition theory).

However, the global studies do not analyse the regional differences which could

strongly deviate from the global average. In some cases, considering the

regional differences may lead to different future projections for the use of

resources in comparison with the global projections, as this thesis has shown.

The second line of research are studies analysing in detail only one resource

accounting for regional differences: land (Fader et al., 2013; Kastner et al.,

2012; Ramankutty, 2008; White, 2007), nitrogen fertilizer (Bouwman et al.,

2009; Leach et al., 2012; Liu et al., 2010; Pierer et al., 2014; Shindo et al., 2006;

Xiong et al., 2008), water (Hoekstra & Mekonnen, 2012) or energy (Berners-Lee

et al., 2012; González et al., 2011). These studies quantify in more detail the use

of resources than the global studies, but do not consider the trade-offs with

other resources. By focusing in detail in only one resource, it is possible to miss

important consequences in the use of other resources. For instance, focusing in

reducing land use can result in large nitrogen fertilizer use to increase crop

yields. The trade-offs among resources have not been analysed in detail in the

existing literature.

The mentioned limitations of these two lines of research suggest the need for

an integrative assessment which considers both the regional differences

relevant for a global analysis and the trade-offs among the resources.

Throughout the thesis, the global differences in the use of resources were

studied in relation to the dynamics of population numbers, type of diets and


129

type of agricultural systems. These three factors are the main drivers for the

use of resources (see figure 7.1) and they differ throughout the world.

First, population numbers change differently among countries in relation to the

socioeconomic development of the population, education, urbanization and

others (demographic transitions: Chesnais (1992)).

Second, the daily diet of a person consists of several food products. The menu of

a person is different throughout the world in relation to the amount of calories

consumed as well as on the type of food consumed due to different income

level, food availability, food preferences, local traditions, urbanization, etc.

(Menzel & D’Aluisio, 2005). Taking cereals as example, a person in Turkey

might mainly consume wheat in contrast with a person that lives in Mexico

who mainly consume maize or a person in Bangladesh who mainly consumes

rice (FAO, 2013d). Also, some regions consume mainly traditional grains (fonio,

quinoa, sorghum) and others modern agricultural grains (hybrid maize) (Garí,

2001).

Third, every food product can be produced in a different production system

which differ in relation to the use of external inputs (organic VS inorganic

farming), the amount of these inputs (resulting in high or low crop yields), the

climate (in tropical regions three harvests a year are possible and in temperate

regions only one harvest), type of machinery (animal draft VS machinery), etc.

All these differences in production systems result in different productivities

and different use of resources per amount of food produced.

The aim to perform a global analysis accounting for all these global differences

and the trade-offs among resources results in a methodological challenge.

Therefore, simplification is necessary to come up with a global overview of the

main relations of these factors relevant for the global use of agricultural

resources.

In order to do this, a methodology was developed to analyse the major trends

and the global differences of the main drivers of agricultural resources and

their trade-offs. National data were used as examples to illustrate the global

spectrum of these differences. Because of this, the food security situation of

these countries is not calculated. Instead, the countries’ data were used to

understand the implications of changes in the drivers, to identify the regions

with the strongest challenges for achieving food security in the coming decades

130

and the origin of the challenge (increase of population, changes of diets,

availability of resources, other). The main two strategies of the methodology to

simplify the global food system are, first, identify the regional differences

relevant for a global analysis and for the use of resources and, second, analyse

the trade-offs among the use of resources.

The level of scale throughout the thesis is a combination of global and regional

scale. In order to compare global differences, it is necessary to always consider

the global scale and at the same time “zoom-in” in a regional scale to identify

the relevant differences for the global scale. The relevant regional differences

are identified with the following approaches:

Grouping the global population taking into account the size of the

group relevant for a global discussion. The grouping is based on the

relevant driver in relation to the use of resources (GDP, population

density or culture). For example, GDP per capita was used to discuss

the differences and changes in diets and population growth because

these two factors are driven by the differences in socioeconomic

development (chapter 2). Population density was used to discuss

differences and changes in production systems because crop yields are

related with the population density of the region (chapter 3). Finally,

culture (as geographical location) was used to discuss differences and

changes in diets because food preferences are related with culture

(chapter 6).

The production systems need to be simplified to identify relations

among the drivers and global differences. As a consequence, several

assumptions need to be done which allows to have a global overview of

the differences in production systems relevant for the use of

agricultural resources even though some details of the system are not

included. For instance, only one crop was used to represent one food

category. Wheat was used to represent cereals in order to identify the

impact of the different factors of crop production (crop yield, nitrogen

fertilizer used, productivity). Also, feed for livestock generally consist

of a mixture of crops but in this thesis it was assumed that one crop

was used as feed to identify the impact of the different production

factors of the feed (crop yield, inputs used, productivity, others) in the

use of resources.


131

The diets need to be simplified to identify regional differences and

their impact on the use of resources. In order to do this, the food

products were grouped into food categories relevant for the use of

resources. For example, all cereals (maize, wheat, rice, sorghum, millet)

were grouped as cereals because their production productivities

(resource use per amount of food) are similar among them. But, for

animal products a distinction needs to be done between the meat types

(beef, pork, chicken) because of the differences in resource use per

amount of meat produced.

Thus, an integrative analysis of the whole system has to follow the strategies

mentioned above. With this approach, less detail is considered but the results

show more relations and trade-offs which lead to discussions of the whole

global food system in relation to the use of resources. It is possible to identify

the regions or groups which will have the strongest challenges for food security

and the source of the challenge (population growth, resource availability,

dietary changes, etc).

The use of resources was studied with a demand perspective in order to

integrate all the different drivers of food demand in one analysis. In this way,

the starting point is what people demand to eat; and, then, trace back to the

agricultural production to assess the amount of resources that were needed for

the production of that demand of food.

The main drivers for the use of agricultural resources are population numbers,

type of diet and type of agricultural systems (see figure 7.1). The first two

determine the demand for food. The type of production system, in addition with

the food demand, determines the demand for resources. Other factors also

drive the use of resources indirectly such as socioeconomic development (GDP

per capita is used throughout the thesis as the indicator), population density

and culture. These three drivers determine the dynamics in population

numbers, diets and agricultural systems. Figure 7.1 illustrates the relations of

all these drivers to the demand of food and resources. These drivers are from

different disciplines so an interdisciplinary approach is needed.

132

Figure 7.1 Relations among the drivers of the demand for food and agricultural

resources. See text for details.

7.3 New insights from an integrative analysis

The insights obtained in each chapter of this thesis have led to identify the main

global relationships of the drivers shown in figure 7.1, their dynamics, the

regional differences regarding these drivers and the trade-offs among the use of

resources. Some of these findings deviate from existing literature and result in

different discussions for the future challenges of the global food system.

The potential for food production based on the availability of land, water and

nitrogen fertilizer was studied in chapters 2 to 4. These chapters analyse the

amount of food that is needed based on the number of people and/or the type

of diets, and the amount of resources that are needed and/or available to

produce it.

The availability of land and water is analysed in chapter 2. The inequality in

availability of land between the poor and the rich countries will strongly

increase by 2050 due to the large population growth of the poor countries. This

leads to discuss whether the available area will be enough for the food demand

of each region. The low availability of land in the poor regions indicates a need

for intensification of the production systems: increase the food production per

area. This is further studied in chapter 3.


133

The intensification of the production systems in relation to the availability of

land and the type of diet was studied in chapter 3. The intensification is

analysed based on the nitrogen application because of the strong relation

between nitrogen application and crop yields (Engels & Marschner, 1995). The

results show that the nitrogen application increases exponentially with the

reduction of land availability below an area of 0.1 ha of arable land per capita.

The type of diet has an effect on the intensification and parallels this relation.

Affluent diets with low population density show low nitrogen application, but

the same diet in highly populated areas shows a large nitrogen application.

These results deviate from existing projections of nitrogen fertilizer use. Tilman

et al. (2011) projects a global nitrogen fertilizer use in 2050 with a linear

increase following the trend of fertilizer use in the last decades. However,

chapter 2 has shown that two thirds of global population will live in countries

with less than 0.1 hectares of arable land per person. So, following the results of

chapter 3, the future use of nitrogen might be larger than the one predicted by

Tilman et al. (2011) because it might increase exponentially instead of

following the linear trend of the last decades.

A clear trade-off between land use and nitrogen fertilizer use is shown in

chapter 3 which is studied in detail in chapter 4. This trade-off is analysed with

a demand perspective by calculating the amount of both land and nitrogen

fertilizer needed per person. The results show the impact of different

production systems and diets on the use of both resources. In general, a

production system with large use of nitrogen per capita results in low land use

and vice versa. However, this trade-off is not linear, and some systems use large

amount of both land and nitrogen fertilizer. Also, a staple diet uses less land

and nitrogen fertilizer than an affluent diet. The nitrogen fertilizer use per

capita can increase a factor three from a staple to an affluent diet with the same

production system. For affluent diets with relative low use of land, only large

nitrogen fertilizer use is possible. The results in chapter 4 show the importance

to consider the trade-off between nitrogen fertilizer and land. This gives new

insights to the discussion of the land and nitrogen footprint studies in the

existing literature. The nitrogen footprint studies (Leach et al., 2012; Pierer et

al., 2014) suggest that a solution to reduce environmental impact caused by

nitrogen fertilizer is the reduction of its use. These studies do not account for

the strong trade-off with land. Chapter 4 have shown that in some cases, where

land is not largely available, the reduction in the use of nitrogen is not possible

even for very basic staple food diets. Similarly, the land studies (Kastner et al.,

134

2012; White, 2007) do not account for the use of nitrogen fertilizer. These

studies suggest that a desirable scenario is the reduction of land use. But, it is

necessary to consider the consequences of increase nitrogen fertilizer use (local

pollution, increase indirect energy use, affecting the global nitrogen cycle and

others) which these studies do not take into consideration.

In addition to production potential, the use of agricultural labour in relation to

diets and production systems is studied in chapter 5. The insights in this

chapter are useful to discuss the trade-off between labour and energy use

related with the use of machinery. Labour efficiency is 200 times higher in a

mechanized system compared with a non-mechanized system. This gain in

labour efficiency is possible with mechanization which replaces human labour.

In general, the degree of labour efficiency of the production system is related

with the socioeconomic development of the population. The labour efficiency is

reflected in the share of agricultural population in a country (Structural

transformation: Timmer (2009)). Low income countries have low labour

efficiency (large amount of hours of farm labour needed per kilogram of food

produced) by using non-mechanized systems (Pimentel & Pimentel, 2008).

With this system, one fulltime farmer produces the food for 5 people (chapter 5

of this thesis), which fits with the share of agricultural population in these

countries: around 30% of the population is engaged in agriculture (FAOc,

2013). In contrast, the share of agricultural population in high income countries

is less than 1 %. These countries have mechanized systems with high labour

efficiency in which one fulltime farmer produces the food for more than 100

people (chapter 5 of this thesis). The use of machinery indicates higher use of

fossil energy for fuel. The results of this chapter contribute to the studies of

energy use for food (González et al., 2011; Berners-Lee et al., 2012). These

studies indicate a need to decrease energy use per product to reduce

greenhouse gas emissions. However, due to the trade-off between machinery

and human labour, the reduction of machinery use might not be an option in

countries with a low share of agricultural population because not enough

labour force is available.

Thus, it is essential to consider the trade-offs among land, energy (fuel and

nitrogen fertilizer) and labour, to evaluate the sustainability of the global food

system. Sustainability is achieved with the lowest environmental impact of food

production. Since food production includes the use of all these agricultural

resources, the environmental impact should be evaluated considering the use of

all the resources at the same time.


135

The type of diet has an important impact on the use of resources as shown in

chapters 2-5. This general statement is not a novel finding and most of the

existing literature addressing the sustainability of the food system concludes

that dietary changes are the key for a sustainable future. For this reason, a

detailed analysis of the global differences in diets and their impact on the use of

resources is performed in chapter 6. The results show that regions change diets

following their own dietary composition and not a global or “western” pattern

as assumed by other studies (Pingali, 2007). This new insight has major

consequences for the use of resources in comparison with existing literature.

The regional dietary composition, especially of the animal food products,

results in different use of resources (with the same production system).

Regional changes to affluent diets will result in different use of resources. For

the same protein consumption, an affluent diet with the current food

preferences of a region in Central Africa needs 30% more land for the

production of animal products than the food preferences of North America.

Similarly, the food composition of animal products in China needs 30% less

land than the food composition in North America (figure 6.5). The assumption

of changing to a “western diet” implies following the dietary pattern of North

America or Western Europe. Chapter 6 shows that the diet in North America or

in Western Europe is not the food consumption pattern requiring larger

amount of resources. Food patterns with large consumption of beef such as the

average diet in Central Africa require larger amount of resources for an affluent

consumption. Thus, future changes to affluent diets could result in higher

amount of resources per capita than the ones in “western countries”.

All the new insights mentioned above can be combined to assess the future

challenges of the food system by 2050. By doing this, the discussion includes

relevant regional differences, the trade-offs and relationships among the

resources and drivers. So, results in an integrative assessment which is

described in the following section.

7.4 Feeding more than 9 billion people in 2050

Future demand for resources will differ among regions because of the

differences in socioeconomic, geographical and cultural situations. By taking

into account all the main drivers for the use of resources (income level,

population density, culture, population numbers, diets and agricultural

136

systems) and considering the trade-offs among the resources (land, water,

nitrogen, labour), an accumulation of challenges is identified for a certain group

of the global population.

Figure 7.2 Global population grouped based on their socioeconomic development to

discuss their expected changes in diets and population numbers, and then sub-grouped

based on their availability of land to discuss the potential of food production. See text for

details.


137

By 2050, 70% of the global population will live in countries with very low land

availability for food production (chapter 2 of this thesis). In this section, the

discussion of chapter 2 is integrated with food production possibilities in

relation to the type of diet and the need for a production system with high or

low crop yields (which was analysed in detail in chapters 3 and 4).

The global population is first grouped based on their GDP per capita in 2010

(World Bank, 2014) similar to chapter 2. Then, these groups are subdivided

based on their availability of arable land per capita. The values of arable land

are the numbers in 2010, see chapter 2 for the justification of this assumption.

The GDP per capita is the starting point because it is the indicator that impacts

all three drivers discussed through the thesis: population growth, dietary

changes, and agricultural production systems. By doing the grouping based on

income level, it is possible to identify the type of changes of food demand for

each group in the coming decades.

In 2010, the global population was around 7 billion people (figure 7.2). 2.6

billion people lived in countries where the average GDP per capita was lower

than US$1,000. This group is referred as the “low income group”. It includes the

countries of sub-Sahara Africa and also countries in Asia such as Bangladesh,

India, Pakistan and Vietnam (groups 1 and 2 of chapter 2). The second group

had an average income level between US$1,000 to US$10,000. This group is

referred as the “transition group”. It includes the countries in North Africa,

some countries in Asia such as China, Indonesia and the Philippines, most

countries in Eastern Europe and Latin America (groups 3,4 and 5 of chapter 2).

The third group had an average income level higher than US$10,000. This

group is referred as the “high income group”. It includes the countries in

Western Europe, some in Asia such as Japan and Korea, Australia, United States

and Canada, among others (group 6 of chapter 2). See Appendix 2 in chapter 2

for the complete list of counties in each group.

The major changes in population numbers are expected in the low income

group, the major changes in diets are expected in the transition group, and no

major changes in both population growth and diets are expected in the high

income group. So, the drivers of food demand are different among the groups.

The low income group will demand more food because of more people, the

transition group will demand affluent food because of dietary changes, and the

high income group will not demand more food. The share of each group in the

global population will be different in 2050 as a result to the different

138

population growth rates. The low income group will almost double and will

account to half of the global population (which only accounted to one third in

2010). In contrast, the high income group will not grow substantially. These

strong differences in population growth rates will increase the inequality in

land availability per capita as discussed in chapter 2.

The transition group is expected to have the largest changes to affluent

consumption due to the changes in socioeconomic development (Kearney,

2010). These countries will demand food for affluent diets. In contrast, not

large changes in diets are expected in the low income group, so these countries

will demand food for staple diets.

Following the discussion of chapter 2, the availability of land per capita is

analysed in more detail. The analysis includes both the need of intensification

of the production systems and the type of diet that will drive the demand for

food. Each group of figure 7.2 was subdivided based on the availability of arable

land per capita in 2050 (figure 7.2c). The requirement of land and nitrogen

fertilizer for a staple and affluent diets were used as the criterion for

production possibilities based on the insights obtained in chapter 4 (figure 4.2).

A staple diet needs 0.4 ha/cap with a low crop yield system with very low

nitrogen fertilizer use and needs only 0.08 ha/cap with a high crop yield system

with large nitrogen fertilizer use. In contrast, an affluent diet needs as much as

1.5 ha/cap in the low crop yield system and 0.3 ha/cap in the high crop yield

system.

The colours of figure 7.2c indicate the production possibilities in relation to the

per capita arable land availability for a certain diet (staple or affluent) , and a

certain production system (with high or low crop yields). The graph shows in

green the population that have enough land to produce an affluent diet with a

low crop yield system (more than 1.5 ha/cap are needed). In yellow, the

population that have enough land to produce a staple diet with a low crop yield

system (more than 0.4 ha/cap are needed). In orange, the population with

enough land for an affluent diet with a high crop yield system (more than 0.3

ha/cap are needed). In red, the population with enough land for a staple diet

with a high crop yield system (more than 0.08 ha/cap are needed). And in

black, the population with not enough land to produce the food with these

systems even for a very basic staple diet.


139

Chapter 3 showed that the global average availability of land in 2050 will be

0.15 ha/cap. The countries show strong variations which deviates from this

global average. With 0.15 ha/cap, more than enough land is available for a

person to have a staple diet produced with high crop yields (more than 0.08

ha/cap are needed). Though not an affluent diet since 0.3 ha/cap are needed.

However, by analysing production possibilities with the global average, the

strong land limitations of a large share of the population are not shown. Figure

7.2c shows that one fifth of the global population will not have enough land for

a basic staple diet with a high crop yield system (population in black). This

shows the need to analyse the land availability in more detail as it is illustrated

in figure 7.2c.

The low income group will demand food for a staple diet. More than one billion

people of this group will live in countries where less than 0.08 ha/cap of arable

land are available (share of population in black). This means that they will not

have enough arable land per person to produce a staple diet with the present

production systems. The rest of the group will have 0.08 ha/cap to 0.2 ha/cap,

which is enough to produce the food for a staple diet but with high crop yield

systems. In general, these countries currently have low crop yield systems. But,

they will have less than half the amount of land needed for a staple diet with

low crop yield systems. The low land availability indicates the urge for these

countries to increase crop yields. It is interesting to point out that none of these

countries will have enough land to produce an affluent diet even with high crop

yield systems.

The transition group will demand food for an affluent diet. In general, these

countries have a higher crop yield systems in comparison with the low income

group. Differently to the low income group, this group will have larger demand

for resources per person due to the changes to affluent diets. Figure 7.2c shows

that the majority of this group will have strong land limitations: more than 80%

of this group will live in countries where not enough land is available to

produce the food for an affluent diet with the present production systems (less

than 0.3 ha/cap available). 600 million people will not have enough land even

for a staple diet in high crop yield systems (less than 0.08 ha/cap available,

population in black), and 2500 million people will have enough land only for

staple diets with high crop yield systems (population in red). In the other hand,

600 million people will live in countries with more than 0.3 ha/cap (population

in orange and yellow), so enough land for producing the food for affluent diets.

These countries include Brazil, Russia, Thailand, Ukraine, among others.

140

The high income group will also have low availability of land but “only” for half

of their population (population in red and black). The rest will have large

availability of land: more than 0.3 ha/cap (population in orange and yellow).

With this amount of land, it is possible to produce the food for more than one

person with intensive systems. This means that they can feed more than their

own population. The type of diet can strongly influence the number of “extra”

people that can be feed with their available land. For example, the countries in

yellow, with a high crop yield system, can produce the food for 4-15 “extra

people” per inhabitant with staple diets or for 0.3-3 “extra people” per

inhabitant with affluent diets.

It is important to point out that in 2050 no country will have the land available

to produce an affluent diet with a low crop yield system and (almost) no

nitrogen fertilizer use: at least 1.5 ha/cap are needed. This value is 10 times

higher than the global average available land in 2050. This means that in 2050,

global population will be ten times higher than what the world could produce

with no nitrogen fertilizer for affluent diets. In other worlds, ten world would

be needed to produce the food for all people with affluent diets and no nitrogen

fertilizer.

To conclude, land availability will be unequally distributed between poor and

rich countries in the coming decades. The largest share of countries with large

land availability is in the rich group. And due to the low land availability of the

poor and transition countries, strong intensification is required for these

groups. This will result in an increase use of inputs to achieve high crop yields,

mainly nitrogen fertilizer. This can result in local pollution if management

practices are not efficient, and in an increase of indirect energy use to produce

the fertilizers.

In addition to the production possibilities (crop yields), the production systems

will change in relation to mechanization. The increase of socioeconomic

development of the transition group will result in a Structural Transformation

of their population (Timmer, 2009) in which the share of agricultural

population decreases. In these cases, mechanization needs to replace human

labour. So, large increase of energy use is expected for this group in relation to

fuel use for machinery in addition to the indirect energy use related with

nitrogen fertilizer use. The high income group already have highly mechanized

production systems, so changes in energy use are not expected for this group.


141

The production possibilities of figure 7.2 for the staple and affluent diet are

based on two examples of diets (figure 4.2). Chapter 6 showed that present

regional differences in dietary composition have a strong impact on the use of

resources. The land needed for an affluent diet in figure 7.2 is based on the

dietary composition of North America in 2010. But, the changes to affluent diets

of the transition group will follow different paths which will result in different

needs for land. Figure 6.5 shows that the land needed for the production of

animal products for an affluent diet is different in relation to the food

preferences of the region, even for the same protein consumption. For example,

an affluent diet with the meat and dairy preferences of a region in Africa can

use 30% more land than the meat and dairy preferences of North America. In

contrast, the meat and dairy preferences of China use 30% less land than the

preferences of North America. Therefore, figure 7.2 might look more optimistic

or pessimistic depending on the choice of food in the diet.

7.5 Looking for integrated sustainable solutions

The solutions for the strong challenges of the future food system mentioned

above and illustrated on figure 7.2 should fulfil food demand with the less

environmental impact as possible. So, a balance should be made between food

needed, resource use per capita and the trade-offs among resources.

The increase of crop yields for 80% of the population will be necessary to fulfil

food demand based on the low availability of land per person (black and red

shares of the population of low income and transition group in figure 7.2). In

general, the low income group have low crop yield systems due to the low use

of inputs resulting in depletion of their soil (Liu et al., 2010). These countries

should overcome the economic barriers to increase the use of inputs such as

fertilizers with efficient agricultural practices. Otherwise, they will not have

enough land to fulfil their food demand with low crop yield systems. Some of

the transition countries already use large amount of nitrogen fertilizer, for

instance countries in East Asia (Shindo et al., 2006; Xiong et al., 2008), though

the use is inefficient and causes large local pollution. Changes to efficient

practices can reduce environmental impact.

As mentioned before, the intensification of the production system for the

transition group will not only include the increase of crop yields (by increasing

the use of nitrogen fertilizer) but also the increase of machinery resulting in

142

higher energy use. However, the increase of crop yields results in lower energy

use for machinery. As shown in chapter 5, the labour efficiency of a certain

amount of crop is related with the amount of hours of labour needed per

hectare and the crop yield obtained. By increasing the crop yield, the labour

productivity increases as well, so the amount of labour needed per kilogram of

crop produced is lower than with low crop yields. In the same way, the use of

machinery is more efficient and less fuel is needed per kilogram of food

produced. So, the increase in indirect energy use related with nitrogen fertilizer

(which increases crop yields) reduces the energy use for fuel.

Throughout the thesis, it has been shown the strong role of diets in the use of

resources. Therefore, the change in diets is a crucial sustainable solution. For

example, affluent diets result in large nitrogen application which can cause

local pollution (chapter 3), affluent diets require both more land and nitrogen

fertilizer use per person than staple diets (chapter 4), and affluent diets require

more labour per person than staple diets (chapter 5) which indirectly require

more fossil energy for machinery use per kilogram of food produced. So, the

change from an affluent to a staple diet can result in: a reduction of both direct

and indirect energy use, reduction of land use (which can increase biodiversity,

afforestation, etc), and a reduction in local pollution due to reduction of

nitrogen fertilizer use. However, the discussion should not finish but start in

this statement. It is essential not to generalize between a staple and an affluent

diet. Strong cultural differences in diets exist throughout the world which have

a relevant impact on the use of resources. It is necessary to do a distinction

among the animal food products, for instance the difference between a diet

with beef or pork consumption results in twice use of resources per capita for

the production of animal food products (chapter 6). Also, affluent vegetarian

diets with large dairy products consumption can result in larger use of

resources than an affluent diet with large consumption of pork or chicken

consumption (figure 6.5).

Thus, in order to come up with an integrative solution for the future of the

global food system, it is necessary to identify the source of food demand (more

people and/or change in diets) by analysing the drivers, and also consider the

availability of resources, the trade-offs among resources and the regional and

cultural differences.


143

7.6 Recommendations for further research

The main achievement of this thesis was to develop an integrative assessment

by identifying the dynamics and regional differences of the main drivers of the

agricultural resources use and its trade-offs relevant for a global assessment.

This allowed to have a global overview of the food production challenges of

future food demand. This was possible by using available parameters of the

countries such as GDP per capita, population density, food supply, crop yields,

etc. The results pinpointed the specific region with the strongest challenges for

future food supply including the relevant drivers for the demand of resources.

The insights of this thesis should be used as starting point for further research

to analyse in detail the food challenges of each specific region and find local

solutions.

Figure 7.2c shows in black the countries with the strongest challenges for

future food security. Detail analysis for these countries should be done to

analyse future food production possibilities including local data which was not

included in this study: climate, soil conditions, specific management practices,

specific food preferences, etc. Then, the local food security situation of the

region can be discussed, and local solutions can be recommended.

Also, the methodology of this thesis was based on country level averages. In

some cases, strong differences within countries exist in socioeconomic

development, diets, and production systems, among others. Further research is

needed considering these differences in some countries and their impact on the

use of resources. A previous study (Ibarrola Rivas, 2010) has shown that the

use of land within the Mexican population is strongly different due to different

production systems among the states and diets among the poor and rich sector

of the population.

7.7 Final conclusion

To assess the future of the global food system, it is necessary to have a global

perspective and at the same time take into account the relevant regional

differences of socioeconomic development, population density, diets, culture

and availability of resources. This perspective allows having an integrative

understanding of the major factors driving the use of resources and results in

new insights for finding solutions. With this approach, this thesis has identified

144

the following main points which summarize the strong challenges for the

future.

The increase of food demand in the coming decades will be in the low

and middle income countries. The present availability of land and

water is unequally distributed between the high and low income

countries. This inequality will increase due to the high population

growth in the low income countries. Because of this, the low and

middle income countries with high population density will have the

strongest challenges for achieving food supply with local food

production. Land and water are non-tradable resources, so their

availability limits the food production possibilities of the region. In

contrast, energy inputs (nitrogen and fuel) are tradable which can

increase the production possibilities by intensifying the production

system.

The low land availability in addition to the increase of economic

development will result in huge increase of energy use in agriculture

(nitrogen fertilizer and fuel). It is necessary to consider the trade-offs

between nitrogen fertilizer and land use, as well as human labour and

machinery use to discuss the implications of the changes in production

systems.

The type of diets will play an important role in the use of resources.

The regional dietary differences should be considered and not only

differences in socioeconomic development. The regional dietary paths

with low resource demand can be used as examples of potential

sustainable solutions for the future of the global food system.

In order to evaluate the sustainability of the food system, it is essential

to consider the use of all major agricultural resources at the same time.

Using energy use as a sustainable indicator for the food system, which

is commonly used in other systems, could have strong side effects due

to the trade-off between energy and land use. The reduction of energy

use for food production can increases the need for land. If land is not

available, food production could not be enough in relation to the

demand of the population. In this case, the food system is not

sustainable since it is not supplying enough food for all people.


145

The different transitions of the drivers for food demand among regions

(population growth, dietary changes and changes in agricultural

systems) will end up in different demand for resources. The

sustainability of the food system in relation to the use of agricultural

resources will depend on the transitions of these drivers.

147

References

Alexandratos, N., & Bruinsma, J. (2012). World Agriculture Towards 2030/2050: The 2012 Revision,

Arizpe, N., Giampietro, M., & Ramos-Martin, J. (2011). Food security and fossil energy dependence: An international comparison of the use of fossil energy in agriculture (1991-2003). Crit. Rev. Plant Sci., 30(1-2), PII 937010094.

Ausubel, J. H., WernIcK, I. K., & Waggoner, P. E. (2013). Peak farmland and the prospect for land sparing. PoPulation and develoPment Review, 38(s1), 221-242.

Berners-Lee, M., Hoolohan, C., Cammack, H., & Hewitt, C. (2012). The relative greenhouse gas impacts of realistic dietary choices. Energ. Policy, 43, 184-190.

Beusen, A. H. W., Bouwman, A. F., Heuberger, P. S. C., Drecht, G. v., & Hoek, K. W. v. d. (2008). Bottom-up uncertainty estimates of global ammonia emissions from global agricultural production systems. Atmos. Environ., 42, 6067-6077.

Billen, G., Garnier, J., Thieu, V., Silvestre, M., Barles, S., & Chatzimpiros, P. (2012). Localising the nitrogen imprint of the Paris food supply: The potential of organic farming and changes in human diet. Biogeosciences, 9(1) , 607-616.

Bleken, M. A., & Bakken, L. R. (1997). The nitrogen cost of food production: Norwegian society. Ambio, 26(3), 134-142.

Boserup, E. (1965). The conditions of agricultural growth: The economics of agrarian change under population pressure. London: George Allen & Unwin.

Bouwman, L., Goldewijk, K. K., Van Der Hoek, K. W., Beusen, A. H. W., Van Vuuren, D. P., Willems, J., . . . Stehfest, E. (2011). Exploring global changes in nitrogen and phosphorus cycles in agriculture induced by livestock production over the 1900–2050 period. PNAS USA, 110 (52), 20882-20887.

Bouwman, A. F., Beusen, A. H. W., & Billen, G. (2009). Human alteration of the global nitrogen and phosphorus soil balances for the period 1970-2050. Global Biogeochem Cy, 23, GB0A04.

148

Bouwman, A. F., Drecht, G. v., & Hoek, K. W. v. d. (2005). Global and regional surface nitrogen balance in intensive agricultural production systems for the period 1970-2030. Pedosphere, 15(2), 137-155.

Budidarsono, S., Kuncoro, S. A., & Tomich, T. P. (2000). A profitability assessment of Robusta coffee systems in Sumberjaya watershed, Lampung, Sumatra Indonesia. (Southeast Asia Policy Research Working Paper No. 16). Bogor, Indonesia: International Centre for Research in Agroforestry.

Bumb, B. L. (1995). World nitrogen supply and demand: An overview. In P. E. Bacon (Ed.), Nitrogen fertilization in the environment (pp. 1-39) Marcel Dekker,Inc.

Chatzimpiros, P., & Barles, S. (2013). Nitrogen food-print: N use related to meat and dairy consumption in France. Biogeosciences, 10, 471-481.

Chertow, M. R. (2000). The IPAT equation and its variants. J. Ind. Ecol., 4(4), 13-29.

Chesnais, J. C. (1992). The demographic transition. stages, patterns, and economic implications (1st ed. ed.). New York: Clarendon Press Oxford.

Cui, S., Shi, Y., Groffman, P. M., Schlesinger, W. H., & Zhu, Y. (2013). Centennial-scale analysis of the creation and fate of reactive nitrogen in china (1910–2010). PNAS USA, 110(6), 2052-2057.

de Vries, M., & de Boer, I. J. M. (2010). Comparing environmental impacts for livestock products: A review of life cycle assessments. Livestock Sci., 128(1–3), 1-11.

de Vries, W., Kros, J., Kroeze, C., & Seitzinger, S. P. (2013). Assessing planetary and regional nitrogen boundaries related to food security and adverse environmental impacts. Curr. Opin. Env. Sust., 5(3–4), 392-402.

Ehrlich, P. R., & Holdren, J. P. (1971). Impact of population growth. Science, 171, 1212-1217.

Eickhout, B., Bouwman, A. F., & van Zeijts, H. (2006). The role of nitrogen in world food production and environmental sustainability. Agr. Ecosyst Environ., 116(1-2), 4-14.

Elferink, E. V. (2009). Meat, milk and eggs: Analysis of animal food environment relations. PhD thesis. University of Groningen.

149

Elferink, E. V., & Nonhebel, S. (2007). Variations in land requirements for meat production. J. Clean Prod., 15(18), 1778-1786.

Engels, C., & Marschner, H. (1995). Plant uptake and utilization of nitrogen. In P. E. Bacon (Ed.), Nitrogen fertilization in the environment (pp. 41-81) Marcel Dekker,Inc.

Erisman, J. W., Sutton, M. A., Galloway, J., Klimont, Z., & Winiwarter, W. (2008). How a century of ammonia synthesis changed the world. Nature Geosci., 1, 636-639.

Evans, L. T. (1980). The natural history of crop yield. Am. Sci., 68(3), 388-397.

Fader, M., Gerten, D., Krause, M., Lucht, W., & Cramer, W. (2013). Spatial decoupling of agricultural production and consumption: Quantifying dependences of countries on food imports due to domestic land and water constraints. Environ Res. Lett., 8(1), 014046.

FAO (2007). FertiStat database. Retrieved from: www.fao.org/ag/agp/fertistat/

FAO (2013a). Aquastat. Retrieved from: www.fao.org/nr/water/aquastat/main/index.stm

FAO. (2013b). FAOSTAT statistical database. Retrieved from: faostat.fao.org/

FAO. (2013c). FAOSTAT statistical database. inputs. Retrieved from: faostat.fao.org/site/348/default.aspx

FAO. (2013d). Food Balance Sheets (FBS). Retrieved from: faostat.fao.org/site/368/default.aspx#ancor

Foley, J. A., Defries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., . . . Snyder, P. K. (2005). Global consequences of land use. Science, 309(5734), 570-574.

Foley, J. A., Ramankutty, N., Brauman, K. A., Cassidy, E. S., Gerber, J. S., Johnston, M., . . . Zaks, D. P. M. (2011). Solutions for a cultivated planet. Nature, 478(7369), 337-342.

Fresco, L. O. (2009). Challenges for food system adaptation today and tomorrow. Environ Sci Policy, 12(4), 378-385.

Galloway, J. N., & Cowling, E. B. (2002). Reactive nitrogen and the world: 200 years of change. Ambio, 31(2), 64-71.

http://www.fao.org/ag/agp/fertistat/

http://www.fao.org/nr/water/aquastat/main/index.stm

http://faostat.fao.org/

http://faostat.fao.org/site/348/default.aspx

http://faostat.fao.org/site/368/default.aspx#ancor

150

Galloway, J. N., Townsend, A. R., Erisman, J. W., Bekunda, M., Cai, Z., Freney, J. R., . . . Sutton, M. A. (2008). Transformation of the nitrogen cycle: Recent trends, questions, and potential solutions. Science, 320, 889-892.

Garí, J. A. (2001). Review of the African millet diversity. Paper for the International Workshop on Fonio, Food Security and Livelihood among the Rural Poor in West Africa,

Gerbens-Leenes, P. W., Nonhebel, S., & Krol, M. S. (2010). Food consumption patterns and economic growth. Increasing affluence and the use of natural resources. Appetite, 55(3), 597-608.

Godfray, H. C. J., Beddington, J. R., Crute, I. R., Haddad, L., Lawrence, D., Muir, J. F., . . . Toulmin, C. (2010a). Food security: The challenge of feeding 9 billion people. Science, 327, 812-818.

Godfray, H. C. J., Crute, I. R., Haddad, L., Lawrence, D., Muir, J. F., Nisbett, N., . . . Whiteley, R. (2010b). The future of the global food system. Philos T Roy Soc B, 365(1554), 2769-2777.

González, A. D., Frostell, B., & Carlsson-Kanyama, A. (2011). Protein efficiency per unit energy and per unit greenhouse gas emissions: Potential contribution of diet choices to climate change mitigation. Food Policy, 36(5), 562-570.

Gruber, N., & Galloway, J. N. (2008). An earth-system perspective of the global nitrogen cycle. Nature, 45(17), 293.

Hengsdijk, H., & Langeveld, J. (2009). Yield trends and yield gap analysis of major crops in the world Wettelijke Onderzoekstaken Natuur & Milieu.

Hoekstra, A. Y., & Mekonnen, M. M. (2012). The water footprint of humanity. PNAS USA, 109(9), 3232-3237.

Howarth, R. W., Boyer, E. W., Pabich, W. J., & Galloway, J. N. (2002). Nitrogen use in the united states from 1961-2000 and potential future trends. Ambio, 31(2), 88.

Ibarrola Rivas, M. J. (2010). Transitions of land requirements for food: A matter of scale (MSc thesis), University of Groningen, the Netherlands.

Isbell, F., Reich, P. B., Tilman, D., Hobbie, S. E., Polasky, S., & Binder, S. (2013). Nutrient enrichment, biodiversity loss, and consequent declines in ecosystem productivity. PNAS USA, 110(29), 11911-11916.

151

Isermann, K., & Isermann, R. (1998). Food production and consumption in Germany: N flows and N emissions. Nutr Cycl Agroecosys, 52(1-2), 289-301.

Kastner, T., & Nonhebel, S. (2009). Changes in land requirements for food in the Philippines: A historical analysis. Land use Policy, 27(3), 853-863.

Kastner, T., Ibarrola Rivas, M. J., Koch, W., & Nonhebel, S. (2012). Global changes in diets and the consequences for land requirements for food. PNAS USA, 109(18), 6868-6872.

Kearney, J. (2010). Food consumption trends and drivers. Philos T Roy Soc B, 365(1554), 2793-2807.

Kronvang, B., Andersen, H. E., Borgesen, C., Dalgaard, T., Larsen, S. E., Bogestrand, J., & Blicher-Mathiasen, G. (2008). Effects of policy measures implemented in Denmark on nitrogen pollution of the aquatic environment. Environ Sci Policy, 11(2), 144-152.

Kumaresan, A., Bujarbaruah, K. M., Pathak, K. A., Anubrata Das, & Bardoloi, R. K. (2009). Integrated resource-driven pig production systems in a mountainous area of northeast India: Production practices and pig performance. Trop Anim Health Pro, 41, 1187-1196.

KWIN. (2009). Kwantitatieve informatie veehouderij 2009-2010. Lelystad, Netherlands: Wageningen UR Liverstock Research.

Lassaletta, L., Billen, G., Romero, E., Garnier, J., & Aguilera, E. (2013). How changes in diet and trade patterns have shaped the N cycle at the national scale: Spain (1961-2009). Reg Environ Change, 14(2), 785-797.

Leach, A. M., Galloway, J. N., Bleeker, A., Erisman, J. W., Kohn, R., & Kitzes, J. (2012). A nitrogen footprint model to help consumers understand their role in nitrogen losses to the environment. Environ Development, 1, 40-66.

Li, S., & Wang, W. (1994). The difference in fertility between urban and rural areas and its impact on the process of urbanization. Chinese J Pop Science, 6(2), 201-210.

Licker, R., Johnston, M., Foley, J. A., Barford, C., Kucharik, C. J., Monfreda, C., & Ramankutty, N. (2010). Mind the gap: How do climate and agricultural management explain the 'yield gap' of crops around the world? Global Ecol Biogeogr, Research Paper, 1.

152

Liu, J., You, L., Amini, M., Obersteiner, M., Herrero, M., Zehnder, A. J. B., & Yang, H. (2010). A high-resolution assessment on global nitrogen flows in cropland. PNAS USA, 107(17), 8035-8040.

Lobell, D. B., Cassman, K. G., & Field, C. B. (2009). Crop yield gaps: Their importance, magnitudes, and causes. Annu Rev Env Resour, 34(1), 179-204.

Lutz, W., & Samir, K. C. (2010). Dimensions of global population projections: What do we know about future population trends and structures? Philos T Roy Soc B, 365(1554), 2779-2791.

Ma, L., Ma, W. Q., Velthof, G. L., Wang, F. H., Qin, W., Zhang, F. S., & Oenema, O. (2010). Modeling nutrient flows in the food chain of China. J Environ Qual, 39(4), 1279-1289.

Mekonnen, M. M., & Hoekstra, A. Y. (2012). A global assessment of the water footprint of farm animal products. Ecosystems, 15(3), 401-415.

Mekonnen, M., & Hoekstra, A. Y. (2011a). National water footprint accounts: The green, blue and grey water footprint of production and consumption. ( No. Volume 1: Main Report). Enschede, the Netherlands: UNESCO-IHE Institute for Water Education.

Mekonnen, M., & Hoekstra, A.Y. (2011b). The green, blue and grey water footprint of crops and derived crop products. Hydrol Earth Syst Sc Discussions, 8(1), 763-809.

Menzel, P., & D'Aluisio, F. (2005). Hungry planet. what the world eats. Napa, California: Material World Books.

Mobtaker, H. G., Keyhani, A., Mohammadi, A., Rafiee, S., & Akram, A. (2010). Sensitivity analysis of energy inputs for barley production in Hamedan province of Iran. Agr Ecosyst Environ, 137(3-4), 367-372.

Mohammadi, A., Rafiee, S., Jafari, A., Dalgaard, T., Trydeman Knudsen, M., Keyhani, A., Mousavi-Avval, S.H., Hermansen, J.E. (2013). Potential greenhouse gas emission reductions in soybean farming: a combined use of Life Cycle Assessment and Data Envelopment Analysis. J Clean Prod, 54, 89-100

Muchadeyi, F. C., Sibanda, S., Kusina, N. T., Kusina, J., & Makuza, S. (2004). The village chicken production system in Rushinga district of Zimbabwe. Livestock Research for Rural Development, vol. 16 Art. #40. Retrieved from: http://www.lrrd.org/lrrd16/6/much16040.htm

153

Neumann, K., Verburg, P. H., Stehfest, E., & Müller, C. (2010). The yield gap of global grain production: A spatial analysis. Agr Syst, 103(5), 316-326.

Pierer, M., Winiwarter, W., Leach, A. M., & Galloway, J. N. (2014). The nitrogen footprint of food products and general consumption patterns in Austria. Food Policy, 49, 128-136.

Pimentel, D. (2009). Energy inputs in food crop production in developing and developed nations. Energies, 2(1), 1-24.

Pimentel, D., & Pimentel, M. H. (2008). Food, energy and society. Boca Raton, Florida: CRC Press.

Pingali, P. (2007). Westernization of Asian diets and the transformation of food systems: Implications for research and policy. Food Policy, 32(3), 281-298.

Poleman, T., & Thomas, L. (1995). Income and dietary change - international comparisons using purchasing-power-parity conversions. Food Policy, 20(2), 149-159.

Popkin, B. M. (1993). Nutritional patterns and transitions. Popul Dev Rev, 19(1), 138-157.

Ramankutty, N., Evan, A. T., Monfreda, C., & Foley, J. A. (2008). Farming the planet: 1. geographic distribution of global agricultural lands in the year 2000. Global Biogeochem Cy, 22(1), - GB1003.

Ray, D. K., Mueller, N. D., West, P. C., & Foley, J. A. (2013). Yield trends are insufficient to double global crop production by 2050. PLoS One, 8(6), e66428.

Rockström, J. (2003). Water for food and nature in drought-prone tropics: Vapour shift in rain-fed agriculture. Philos T Roy Soc B, 358(1440), 1997-2009.

Rockström, J., Steffen, W., Noone, K., Persson, A., Chapin, F. S., Lambin, S., . . . Foley, J. A. (2009). Planetary boundaries: Exploring the safe operating space for humanity. Ecol Soc, 14(2), 32. [online] http://www.ecologyandsociety.org/vol14/iss2/art32/

Rougoor, C. W., Van Zeijts, H., Hofreither, M. F., & Bäckman, S. (2001). Experiences with fertilizer taxes in Europe. J Environ Plann Man, 44(6), 877-887.

154

Schnoor, J. L., Galloway, J. N., & Moldan, B. (1997). East central Europe: An environment in transition. Environ Sci Technol, 31(9), 412A-416A.

Schroeder, G., & Titgemeyer, E. (2008). Interaction between protein and energy supply on protein utilization in growing cattle: A review. Livestock Sci, 114(1), 1-10.

Shindo, J., Okamoto, K., & Kawashima, H. (2006). Prediction of the environmental effects of excess nitrogen caused by increasing food demand with rapid economic growth in eastern Asian countries, 1961–2020. Ecol Model, 193(3-4), 703-720.

Siegmund-Schultze, M., Rischkowsky, B., da Veiga, J. B., & King, J. M. (2007). Cattle are cash generating assets for mixed smallholder farms in the eastern amazon. Agr Syst, 94(3), 738-749.

Smil, V. (2002a). Worldwide transformation of diets, burdens of meat production and opportunities for novel food proteins. Enzyme Microb Tech, 30, 305-311.

Smil, V. (1999). Nitrogen in crop production: An account of global flows. Global Biogeochem Cy, 13(2), 647-662.

Smil, V. (2001). Enriching the Earth: Fritz Haber, Carl Bosch and the transformation of world food production (First Edition ed.). Cambridge, MA: The MIT Press.

Smil, V. (2002b). Nitrogen and food production: Proteins for human diets. Ambio, 31(2), 126-131.

Sutton, M. A., Bleeker, A., Howard, C. M., Bekunda, M., Grizzetti, B., de Vries, W., . . . Zhang, F. S. (2013). Our nutrient world: The challenge to produce more food and energy with less pollution. (). Edingurgh UK: Global Overview of Nutrient Management.

Tilman, D., & Clark, M. (2014). Global diets link environmental sustainability and human health. Nature, 515(7528), 518-22.

Tilman, D. (1999). Global environmental impacts of agricultural expansion: The need for sustainable and efficient practices. National Academy of Sciences Colloquium "Plants and Population: Is there Time?", Irvine, C.A. , 96 5995-6000.

155

Tilman, D., Balzer, C., Hill, J., & Befort, B. L. (2011). Global food demand and the sustainable intensification of agriculture. PNAS USA, 108(50), 20260-20264.

Timmer, C. P. (2009). A world without agriculture. the structural transformation in historical perspective. Washington, D.C.: The AEI Press.

Tripathi, R., & Sah, V. (2001). Material and energy flows in high-hill, mid-hill and valley farming systems of Garhwal Himalaya. Agr Ecosyst Environ, 86(1), 75-91.

Uddin, M. M., Sultana, M. N., Ndambi, O. A., Hemme, T., & Peters, K. J. (2010). A farm economic analysis in different dairy production systems in Bangladesh. Livestock Research Rural Develop, Vol. 22 Art.122. Retrieved from http://www.lrrd.org/lrrd22/7/uddi22122.htm

United Nations. (2011). World population prospects. the 2010 revision. vol. 1, comprenhensive tables. ( No. ESA/P/WP.220). New York: United Nations.

White, T. J. (2007). Sharing resources: The global distribution of the ecological footprint. Ecol Econ, 64(2), 402-410.

Woods, J., Williams, A., Hughes, J. K., Black, M., & Murphy, R. (2010). Energy and the food system. Philos T Roy Soc B, 365, 2991-3006.

World Bank. (2014). World bank database, indicators. Retrieved from: data.worldbank.org/indicator

World Bank, Poverty Reduction and Equity group. (2012). Food price watch.

Xiong, Z. Q., Freney, J. R., Mosier, A. R., Zhu, Z. L., Lee, Y., & Yagi, K. (2008). Impacts of population growth, changing food preferences and agricultural practices on the nitrogen cycle in East Asia. Nutr Cycl Agroecosys, 80, 189-198.

http://data.worldbank.org/indicator

157

Summary

By María José Ibarrola Rivas

The use of resources for human consumption is reaching the limits of the

planet, and global food production plays an important part. Food is essential to

maintain life and all persons should have a daily supply of food. The production

of this food requires large amount of resources. Future global food demand will

increase so a sustainable food system is needed: enough food for all people with

the lowest environmental impact.

The amount of resources depends on the number of people (population), on

their type of food pattern (diets) and on how the food was produced

(agricultural production system). The dynamics of these three factors depend

on drivers of different disciplines such as socioeconomic development,

urbanization, demography, culture, geography, climate, agricultural practices

and others.

This thesis takes a food demand perspective to analyse the use of agricultural

resources for food production: land, water, nutrients and labour. The aim is to

assess the impact of food demand on the use of agricultural resources taking

into account the dynamics and regional diversity of population numbers, diets

and agricultural systems. A methodology was developed to integrate in one

analysis regional differences, interdisciplinary indicators and trade-offs among

resources. Chapters 2 to 6 study in detail the main trends of the last decades

and/or the relevant regional differences. The main findings of these chapters

were used to assess the future sustainability of the global food system (chapter

7).

The availability of land and water for food production is analysed in chapter 2.

The study shows that the per capita availability of land and water is unequally

distributed between poor and rich countries. Population growth in the poor

countries will increase this inequality. By 2050, one third of global population

(the richest) will have 3 times more land per capita than the rest. The other two

thirds of the population (the poorest) will not have enough land to produce

food for an affluent diet. An affluent diet is commonly composed of luxurious

products such as animal food products, fruits, vegetables and processed food

158

items. The people expected to change to affluent diets are the ones with the

lowest availability of land and water.

The low availability of land indicates a need for intensification: high amount of

food production per area. This is further studied in chapter 3. Intensification is

analysed in relation to the nitrogen application rate. A farm scale model

integrates: population density, diets and nitrogen fertilizer application rate. The

study shows that the nitrogen application highly increases with the reduction

of land availability and the type of diet parallels this relation. The food

production for an affluent diet in countries with large land availability requires

a much lower nitrogen application rate in comparison with a similar diet in

countries with low land availability.

The strong trade-off between nitrogen fertilizer and land is studied in detail in

chapter 4. Nitrogen fertilizer use per person and its trade-off with land was

calculated for five combinations of production systems and diets illustrating the

global differences. The global differences in use of resources are enormous:

from 3 to 30 kilograms of nitrogen fertilizer per person, and from 1800 to 4500

m2 per person. The methodology used in this chapter allows to identify the

impact of production systems and diets individually. Affluent diets in all

systems require 4 times more nitrogen fertilizer and land than staple food

diets.

In addition to production potentials, the use of agricultural labour in relation to

diets and production systems is studied in chapter 5. The hours of farm labour

needed for an affluent and a staple diet were calculated using the extreme

examples of a mechanized and a non-mechanized production system. Only 5

hours of farm labour are needed to produce an affluent diet in a mechanized

system. The increase of labour efficiency with mechanization is enormous: 200

times less farm labour is needed for a person’s diet with a mechanized system.

The type of diet also plays an important role: affluent diets need two times

more farm labour than staple diets. These insights are useful to discuss the

trade-off between labour and energy use related with machinery use.

The global differences in diets were studied in detail in chapter 6 due to the

strong impact of the dietary patterns on the use of resources. The study shows

that regional dietary paths have been stronger than global food pattern trends.

So, future changes in diets will follow current regional dietary composition.

These regional differences have a strong impact on the use of resources. For an

159

affluent diet, the type of meat consumed can change the use of resources by a

factor of two. Some regions have traditional vegetarian diets with dairy

products which do not necessarily use fewer resources than diets with large

consumption of meat.

Future demand for resources will be different among regions because of

different socioeconomic, geographical and cultural situations. By combining the

insights of this thesis, an accumulation of challenges to achieve food supply is

identified for a certain group of the global population (chapter 7). By 2050,

70% of global population (the poorest) will live in countries with very low land

availability. These people are the ones expected to increase food demand due to

population growth and dietary changes. So, strong intensification is required to

increase food production per area. Based on our analysis, future needs for

nitrogen fertilizer will be higher than the ones projected in other studies.

Energy use for agriculture will strongly increase in some of these countries not

only because of the increase of nitrogen fertilizer use but also because of the

increase in machinery use, which will result in huge increase of farm labour

efficiency. Dietary choices could play an important role in the use of resources.

This thesis shows that in order to assess the future of the global food system, it

is necessary to have a global perspective and, at the same time, to take into

account the relevant regional differences of socioeconomic development,

population density, diets, culture and availability of resources. This perspective

allows having an integrative understanding of the major factors driving the use

of resources and results in new insights for finding sustainable solutions.

161

Resumen

Por María José Ibarrola Rivas

El uso de recursos naturales para el consumo humano está alcanzando los

límites del planeta, y la producción de alimento juega un papel importante en el

uso de éstos. Comer diariamente es esencial para la vida y todas las personas

del planeta deben tener un suministro diario de alimento. Producir nuestra

comida requiere una gran cantidad de recursos. La demanda de alimento

aumentará en las siguientes décadas, lo que significa que es esencial que el

sistema global de producción de alimento sea sustentable: suficiente para todas

las personas y con el menor impacto ambiental posible.

La cantidad de recursos para producir el alimento depende de la cantidad de

personas (población), del tipo de alimentación (dietas) y de cómo se produce el

alimento (sistemas de producción agrícola). La dinámica de estos tres factores

depende de indicadores de diferentes disciplinas como por ejemplo del

desarrollo socioeconómico de la población, del grado de urbanización, de la

cultura, de la geografía, del clima y de las prácticas agrícolas, entre otros.

Esta tesis toma la perspectiva de la demanda de alimento para estudiar el uso

de los recursos agrícolas, en concreto el uso de tierra, agua, nutrientes y

trabajo. El objetivo principal es evaluar el impacto de la demanda de alimento

en el uso de los recursos agrícolas tomando en cuenta los cambios desde los

años 60’s y las diferencias regionales en cuanto al tamaño de la población, al

tipo de dietas y al tipo de los sistemas agrícolas. Se desarrolló una metodología

para integrar en un mismo análisis las diferencias regionales, los indicadores

interdisciplinarios y la compensación entre el uso de un recurso agrícola y otro.

En los capítulos 2 al 6 se estudian en detalle las principales tendencias de las

últimas décadas y las diferencias regionales del uso de los recursos. Las

conclusiones de estos capítulos se utilizan en el capítulo 7 para evaluar la

sustentabilidad del suministro mundial de alimento para el año 2050.

La disponibilidad de tierra y agua para la producción de alimento se analiza en

el capítulo 2. Este estudio demuestra que la disponibilidad per cápita de tierra y

agua está distribuida desigualmente entre los países pobres y ricos. Esta

desigualdad aumentará por el gran crecimiento demográfico de los países

pobres. En el año 2050, una tercera parte de la población mundial (la más rica)

162

tendrá tres veces más tierra per cápita que el resto de la población. Las otras

dos terceras partes (los más pobres) no tendrán suficiente tierra para producir

los alimentos para una dieta afluente. Una dieta afluente es considerada como

aquella compuesta por alimentos de origen animal, frutas, verduras y alimentos

procesados. Las personas que cambiarán a dietas afluentes vivirán en los países

con la menor disponibilidad de tierra y agua.

La poca disponibilidad de tierra indica una necesidad de intensificar la

producción de alimento: aumentar la producción por unidad de área. Esto es

estudiado con más detalle en el capítulo 3. La intensificación se analiza en

relación a la aplicación de fertilizante de nitrógeno. Para esto, se utiliza un

modelo que integra tres factores que afectan el uso de fertilizantes directa o

indirectamente: la densidad de población, el tipo de dieta y la tasa de aplicación

de nitrógeno. El estudio demuestra que la aplicación de nitrógeno aumenta

exponencialmente con la reducción en la disponibilidad de tierra;

adicionalmente el tipo de dieta aumenta esta relación. Los países en los que la

población tiene dietas afluentes y poca disponibilidad de tierra aplican mucho

más fertilizantes de nitrógeno en comparación a los países con dietas similares

y alta disponibilidad de tierra.

La clara relación entre el uso de fertilizantes de nitrógeno y el uso de tierra es

estudiada en detalle en el capítulo 4. Para esto se calcula el uso per cápita de

tierra agrícola y de fertilizante de nitrógeno para cinco diferentes tipos de dieta

y cinco tipos de sistemas agrícolas que ilustran las diferencias globales. Los

resultados demuestran diferencias enormes en el uso de fertilizantes y tierra:

de 3 a 30 kilogramos de nitrógeno por persona y de 1 800 a 4 500 m2 de tierra

agrícola por persona. La metodología utilizada en este capítulo permite

identificar por separado el impacto de los sistemas agrícolas y del tipo de

dietas. Por ejemplo, las dietas afluentes requieren cuatro veces más

fertilizantes y tierra que las dietas más básicas.

El uso de trabajo agrícola en relación al tipo de dietas y al sistema agrícola es

estudiado en el capítulo 5. En este capítulo se calculan las horas de trabajo

agrícola necesarias para producir el alimento que una persona consume en un

año. Se analizan cuatro escenarios diferentes ilustrando los extremos de tipos

de dietas (básica y afluente) y de tipos de producción en relación a la mano de

obra (mecanizados y no mecanizados). Los resultados demuestran que sólo se

necesitan cinco horas de mano de obra agrícola en un sistema mecanizado para

producir el alimento anual que una persona consume con dieta afluente. El

163

aumento de la eficiencia en la mano de obra con la mecanización es enorme. Se

necesitan 200 veces más horas de trabajo con un sistema no mecanizado a

diferencia de uno mecanizado. El tipo de dieta también juega un papel

importante. Las dietas afluentes requieren dos veces más horas de trabajo a

diferencia de una dieta básica. Estos resultados pueden ser utilizados para

discutir el uso de energía para la mecanización.

Las diferencias en el tipo de alimentación a nivel mundial son estudiadas en el

capítulo 6. Los resultados demuestran que los cambios en las dietas, siguiendo

patrones regionales, son más recurrentes que siguiendo patrones globales. Por

lo tanto, se puede esperar que en el futuro la población siga las tendencias

regionales y no las globales. Estas diferencias regionales tienen un gran

impacto en el uso de recursos. Por ejemplo, el consumo de carne difiere entre

una región y otra. Para una misma dieta afluente, el uso de recursos agrícolas

puede duplicarse por el tipo de carne dependiendo del consumo regional.

El uso de los recursos agrícolas en las siguientes décadas será diferente en cada

región del mundo. Estas diferencias dependerán de la situación

socioeconómica, geográfica y cultural de cada región. Con los resultados de esta

tesis es posible identificar una acumulación de retos para el suministro de

alimento para un grupo específico de la población mundial (capítulo 7). En el

año 2050, el 70% de la población mundial (los más pobres) vivirán en países

con muy baja disponibilidad de tierra. Son estos países en los que se espera que

aumente la demanda de alimento debido al crecimiento poblacional y a los

cambios en los patrones alimenticios. Por lo tanto, en estos países se necesita

intensificar los sistemas de producción agrícola: más alimento producido por

unidad de área. Con los resultados de esta tesis se estima que el uso global de

fertilizantes de nitrógeno será superior al proyectado en otros estudios. Así

mismo, el uso de energía para la agricultura aumentará enormemente en

algunos países no sólo por el aumento indirecto en relación al uso de

fertilizantes sino también por el aumento en mecanización. Este último

aumentará enormemente la eficiencia en la mano de obra agrícola.

En conclusión, esta tesis demuestra que para evaluar el futuro del suministro

mundial de alimento es necesario tener una perspectiva global y al mismo

tiempo tomar en cuenta las diferencias regionales relevantes en cuanto al

desarrollo socioeconómico, la densidad de población, los patrones alimenticios,

la cultura y la disponibilidad de recursos agrícolas. Este enfoque permite

obtener un entendimiento integral de los principales factores que afectan al uso

164

de los recursos agrícolas obteniendo nuevos resultados que proponen nuevas

ideas para la sustentabilidad del suministro mundial de alimento.

165

Samenvatting

By Sanderine Nonhebel

In de komende jaren zal de wereldbevolking toenemen naar 9 miljard mensen.

Om deze mensen te voeden zal er meer voedsel geproduceerd moeten worden.

De hoeveelheid voor landbouw geschikte grond in de wereld is beperkt en er

zijn grote regionale verschillen. Hetzelfde geldt voor de beschikbaarheid van

zoet water. Hoeveel voedsel je kunt produceren op een stukje grond hangt af

van het gekozen productiesysteem. Je kunt in principe een gewas verbouwen

zonder externe inputs, door gebruik te maken van de natuurlijke

bodemvruchtbaarheid en handelingen op de akker met de hand uit te voeren

(extensieve productie). In dat geval is de opbrengst laag en is er veel arbeid in

het veld nodig. Door kunstmest te gebruiken kan de productie per oppervlakte

worden verhoogd en door het gebruik van machines kan de arbeidsinzet

worden verlaagd. Voor de productie van kunstmest is veel energie nodig en

voor het gebruik van machines ook. Het gebruik van kunstmest heeft daarnaast

grote gevolgen voor de wereldwijde nutriënten kringloop.

Hoeveel voedsel we nodig hebben hangt niet alleen af van het aantal mensen,

maar ook wat deze mensen eten. Een dieet met veel dierlijke producten vraagt

meer grondstoffen dan een vegetarisch dieet. In China is de laatste 20 jaar de

vleesconsumptie verdubbeld en naar verwachting zullen meer landen dit

voorbeeld volgen. In de komende decennia verwachten we dus meer mensen,

die meer vlees gaan eten, die allemaal van een beperkt landbouwareaal gevoed

moeten gaan worden. De productie per hectare van dit areaal kan verhoogd

worden door het gebruik van kunstmest, alleen heeft dat gevolgen voor het

milieu.

De te verwachten veranderingen in bevolkingsaantallen, consumptiepatronen

etc. verschillen sterk voor de verschillende gebieden op aarde; de verwachte

bevolkingsgroei bijvoorbeeld vindt vooral plaats in de arme landen. Daarom

wordt in dit proefschrift onderscheid gemaakt tussen de verschillende regio’s

op aarde. Voor al deze regio’s wordt onderzocht hoe de veranderende vraag

naar voedsel invloed heeft op het gebruik van de beschikbare hoeveelheid

landbouwgrond en water en hoeveel energie (met name kunstmest) en nodig is

om voldoende te kunnen produceren.

166

Eerst wordt er nagegaan hoeveel land en water er in de verschillende regio’s

beschikbaar is voor de productie van voedsel en hoe dit in de komende

decennia zal veranderen door de te verwachten bevolkingsgroei (hoofdstuk 2).

Op het moment zijn er grote verschillen: in sommige gebieden is er meer dan

5000 m2 landbouwgrond per persoon beschikbaar, maar in andere minder dan

1000 m2. Aangezien de bevolkingstoename vooral in de arme regio’s plaats

vindt, zien we dat in die regio’s de beschikbaarheid van landbouwgrond per

persoon sterk afneemt. In 2050 zal een groot deel van de wereldbevolking

(70%) in gebieden wonen met minder dan 1000 m2 landbouwgrond per

persoon.

Deze beperkte beschikbaarheid aan landbouwgrond in veel regio’s betekent dat

er landbouwsystemen met hoge opbrengsten per oppervlakte nodig zijn om

genoeg voedsel te produceren voor de bevolking. De relatie tussen landgebruik

en kunstmestgebruik en het voedingspatroon onderzocht in hoofdstuk 3. Hier

wordt duidelijk dat met name de combinatie van hoge bevolkingsdichtheid en

een consumptiepatroon met veel dierlijke producten een grote vraag naar

kunstmest met zich meebrengt.

In hoofdstuk 4 wordt dit nader onderzocht: voor verschillende soorten

voedingspatronen wordt de kunstmest behoefte uitgerekend. Er wordt

gerekend aan een karig dieet, voornamelijk bestaande uit granen, peulvruchten

en wortels (het gemiddelde dieet in ontwikkelingslanden) een transitiedieet

(met beperkte consumptie van dierlijke producten, kenmerkend voor de

opkomende economieën) en een Westers dieet met heel veel luxeproducten als

vlees, zuivel, dranken, etc. Er blijkt 4 maal zoveel land en kunstmest nodig te

zijn voor de productie van een luxueus Westers dieet dan voor een karig dieet.

We weten inmiddels dat luxueuze diëten meer grondstoffen vragen dan de

karige en dat diëten over de hele wereld aan het veranderen zijn. Dan is het

interessant om na te gaan op wat voor manier de diëten aan het veranderen

zijn, verschuiven alle diëten richting de Westerse hamburger of zijn er

regionale verschillen. Dit wordt onderzocht in hoofdstuk 6. Voor 13 regio’s in

de wereld worden de veranderingen in diëten over de laatste 50 jaar

geanalyseerd. We zien algemene patronen (toename van de dierlijke producten

), maar ook grote regionale verschillen. De regionale verschillen zitten vooral in

het soort dierlijk product. In China neemt de varkensvlees consumptie toe, in

India de melkconsumptie en in Latijns Amerika de rundvlees consumptie, in

West Europa en Noord Amerika de consumptie van kip. Aangezien het gebruik

167

van hulpbronnen als land, water en energie verschilt voor de verschillende

dierlijke producten hebben deze regionale voorkeuren voor vlees en zuivel

gevolgen voor het gebruik van hulpbronnen voor de verschillende diëten.

Regio’s met grote voorkeur voor rundvlees hebben grotere hoeveelheden land,

water en energie nodig voor de productie van voedsel en produceren meer

broeikasgasemissies.

Er is ook energie nodig op de akker voor de mechanisatie (diesel voor de

tractor). De hoeveelheid die nodig is hangt af van het gewas (hoeveel

bewerkingen zijn er nodig), maar nog veel meer van de mechanisatiegraad. In

principe kan alles ook met de hand gedaan worden (alleen duurt het dan wat

langer). In hoofdstuk 5 wordt de farm- labour-footprint uitgerekend: hoeveel

uur een boer bezig is op zijn land voor de productie van voedsel voor 1 jaar

voor een persoon. Ook hier wordt er weer naar de verschillende systemen in de

wereld gekeken en naar de verschillende diëten. De verschillen zijn groot. In

Nederland is er maar 5 uur landbouwarbeid nodig om voor 1 persoon voedsel

te produceren, in Afrika 400 uur. Ook hier heeft het dieet invloed: er is 2 keer

zo veel arbeid nodig voor een dieet met vlees als voor een vegetarisch dieet.

In dit proefschrift zijn op verschillende manieren de verbanden tussen het

gebruik van land, water en energie voor voedselproductie op wereldschaal

geanalyseerd. De bevolkingsgroei in combinatie met veranderende

voedingsgewoonten gaan een grote claim leggen op deze hulpbronnen. Deze

claim is erg ongelijk verdeeld in de wereld. In Europa en Noord Amerika vindt

nauwelijks verandering plaats, er is geen bevolkingsgroei en

consumptiepatronen zijn al luxueus, hier zijn geen veranderingen in

milieueffecten van voedselproductie te verwachten. In andere delen van de

wereld vindt zowel een toename van de bevolking als een verandering van de

voedselconsumptie plaats. Dit heeft tot gevolg dat in 2050 70 % van de

wereldbevolking in gebieden woont met zeer beperkte beschikbaarheid van

landbouwgrond. Alleen intensieve landbouwsystemen (met veel kunstmest)

kunnen dan genoeg voedsel produceren om aan de vraag te voldoen. Dit zal

grote gevolgen hebben voor het milieu in die regio’s.

169

Acknowledgments

Living in Groningen has been one of the best experiences in my life: living in a

foreign country, enjoying and overcoming the struggles of doing a PhD project,

starting a life as a couple and growing to a family of four. I’d like to thank the

people who inspired me and who helped me make things smoother during

these wonderful and challenging years of my PhD.

Sanderine, my daily supervisor, thanks for your support on research and

personal issues. I took literally the term “daily supervisor” since I was knocking

in your door almost every day during my PhD. You taught me to be strong on

my ideas, to defend and to promote my findings. Henk, my promoter, I very

much enjoyed our discussions during the last year of my PhD. Ton, thanks for

your positive input in my papers. Leandro, my external supervisor, you inspired

me to add a social science approach in my research.

My office-mates. Thomas, you have been not only a great colleague but also a

great friend. I am very happy that it has been possible to keep on collaborating

after you left IVEM. Reino and Ron, talking with you during the difficulties of

the last year of my PhD helped me to overcome the problems and to keep on

working.

Gideon, my classmate, colleague, friend and paranimf, you are one of the

sweetest person I’ve met. Thanks for the love you have given me and my gezin.

People from IVEM and SSG. I enjoyed the coffee and lunch breaks to clear my

mind and to get energy to go back to work. It was nice to talk to you or just to

sit around with you all.

The international summer schools that I joined in retreated places (Hokkaido,

Peyresq and Noordwijk) had the best atmosphere for meeting future colleagues

and for gaining inspiration from other PhD students and researchers from

other countries. I met wonderful people, especially Francis, Leah and Irene.

In order to achieve a PhD project, persistence, inspiration and motivation is

constantly needed. I managed to have it thanks to the great friends that I met in

Groningen.

170

Thanks to my Dutch friends. Mieke, Tsjalling, Eduard and Colien. you have

opened your home to us. Gideon, Jorien, Eva and Jeroen you showed us the

coolest part of the Netherlands. Jorien, thanks for your advices about

motherhood, you made it all easier. Charlotte, we have gone through so much in

common these past 7 years; it has been very relaxing to talk with you during all

these moments to have someone who understands and is going through the

same.

Thanks to my international friends with whom we have overcome the

challenges of cultural differences and we have slowly adopted some Dutch

habits. Markella, we have grown together these last 7 years. You are an amazing

friend. María Jesús, not only because of your strong friendship but also for the

love you show to Federico.

Becoming a mother has been the most rewarding and challenging thing during

these past years. Sharing this process as a foreigner with Katrina and Virginie

has made it all easier, you are wonderful friends.

Thanks to my family who has been closer than ever eventhough an ocean has

been separating us. Thanks to my parents and my sisters for their

unconditional support and love, for their frequent Skype talks and their visits to

Groningen. Thank you Guillermo and Tere for your love.

Finally, and most important, I thank my gezin which has started and has grown

here in Groningen. The most important thing I’ve learned in these years has

been the value of having a gezin. This Dutch word should exist in all languages.

Julio, your motivation and example of persistence has given me the strength for

finishing this project. Julio, Federico and Emilio, you are the inspiration of my

life.

171

About the author

María José was born in Mexico City on March 7th 1983. She did a bachelor study

in Physics Engineering in the Universidad Iberoamericana in Mexico City and

she has graduated with honours in May 2007. From September 2008 to August

2010, she completed the master in Energy and Environmental Sciences in the

University of Groningen. On September 2010, she started her PhD at the Center

for Energy and Environmental Sciences of the University of Groningen, and she

has obtained the degree on May 2015.

During her PhD project, she attended several international summer schools for

PhD students In 2011, she participated in the GCOE-INeT summer school

Understanding coupled natural and social systems in Hokkaido University in

Japan. In 2012, she participated in the ALTERNET summer school on

Biodiversity and Ecosystem Services in Peyresq, France. In 2013, she

participated in the summer school Hunger Defeated? organized by Wageningen

University in the Netherlands. In 2014, she did a research visit at the Institute

of Social Ecology of Alpen-Adria Universitaet in Vienna collaborating with Dr.

Thomas Kastner. She obtained the certificate by the SENSE research

partnership of the Netherlands (sense.nl).

Publication during the PhD project

Ibarrola Rivas, M.J., Moll, H.C. & Nonhebel, S (under review, invited paper) Future global

use of resources for food: the huge impact of regional diets Global Food Security Journal

Ibarrola Rivas, M.J. & Nonhebel, S (under review) Increasing inequality between poor

and rich countries as to availability of land and water by 2050. Agricultural Systems

Journal

Ibarrola Rivas, M.J. & Nonhebel, S (under review) Nitrogen fertilizer use per person and

its trade-off with land use: An international comparison of agricultural production

systems and diets. Industrial Ecology Journal

Ibarrola Rivas, M.J., Kastner, T. & Nonhebel, S (under review) Farm labour footprint of

food: an international comparison of the impact of diets and mechanization. Food Policy

Journal

172

Ibarrola Rivas, M.J. & Nonhebel, S (under review) Estimating future global needs for

nitrogen based on regional changes of food demands. Environmental Development

Journal.

Kastner, T, Ibarrola Rivas, M.J., Koch, W. & Nonhebel, S. (2012) Global changes in diets

and the consequences for land requirements for food, PNAS, 109 (18), 6868-6872

Ibarrola Rivas, M.J. & Nonhebel, S. (2011) Methodology to evaluate the impact of the

Nitrogen Footprints in dietary transitions, Conference Paper, IGS-SENSE Conference-

Resilient Societies, University of Twente

Presentations at international conferences (selection)

Ibarrola Rivas, M.J. & Nonhebel, S (2014) Increasing inequality on the availability of land

and water between poor and rich by 2050 Oral presentation at the ISEE Conference:

Wellbeing & Equity within Planetary Boundaries (August 13-15) Reykjavik University,

Iceland

Ibarrola Rivas, M.J. & Nonhebel, S (2014) Integrating the drivers for the global use of

synth. Nitrogen fertilizer Oral presentation at the 2nd GLP Open Science Meeting (March

19-21) Humboldt University, Berlin, Germany

Ibarrola Rivas, M.J. & Nonhebel, S (2013) Global differences in diets and their relevance

for the use of agricultural resources Oral presentation at the First International

Conference on Global Food Security (Sept 29- Oct 2)Noordwijkwehout, Netherlands

Ibarrola Rivas, M.J. & Nonhebel, S (2012) Ditribution of fertilizers, land and food Poster

presentation at Planet Under Pressure (March 26-29), London, UK

Ibarrola Rivas, M.J. & Nonhebel, S (2010) Development of land requirements for food: a

matter of scale Oral presentation at the Ester Boserup Conference (Nov 15-17), Institute

of Social Ecology (SEC) Alpen-Adria Universitaet, Vienna, Austria

university of groningen the use of agricultural resources ...maría josé ibarrola rivas . colophon...

Documents