irena-dbfz_biomass potential in africa

7/29/2019 IRENA-DBFZ_Biomass Potential in Africa

1/44

International Renewable Energy Agency

IRENA

BIOMASS

POTENTIAL

IN AFRICA

K. Stecher

A. Brosowski

D. Thrn


2/44

2 IRENA-DBFZ: BIOMASS POTENTIALS IN AFRICA

Copyright IRENA 2013

DISCLAIMER

The designations employed and the presentation o materials herein do not imply the expression

o any opinion whatsoever on the part o the International Renewable Energy Agency or

German Biomass Research Centre (DBFZ) concerning the legal status o any country, territory,city or area, or concerning their authorities or the delimitation o their rontiers or boundaries.

About IRENA

The International Renewable Energy Agency (IRENA) is an intergovernmentalorganisation that supports countries in their transition to a sustainable energyuture, and serves as the principal platorm or international cooperation, a centre oexcellence, and a repository o policy, technology, resource and inancial knowledgeon renewable energy. IRENA promotes the widespread adoption and sustainable useo all orms o renewable energy, including bioenergy, geothermal, hydropower, ocean,solar and wind energy in the pursuit o sustainable development, energy access,energy security and low-carbon economic growth and prosperity.

www.irena.org

About DBFZ

The mission o the German Biomass Research Centre (DBFZ) is to support the eectiveintegration o biomass as a valuable resource or a sustainable energy supply in thecontext o applied scientiic research - including technical, environmental, economic,social and energy economics issues along the entire chain o exploitation.

www.dbz.de

Unless otherwise indicated, material in this publication may be used reely, sharedor reprinted, but acknowledgement is requested. IRENA and DBFZ would appreciatereceiving a copy o any publication that uses this publication as a source.

No use o this publication may be made or resale or or any other commercial purposewhatsoever without prior permission in writing rom IRENA/DBFZ. Any omissions anderrors are solely the responsibility o the authors.


3/44

IRENA-DBFZ: BIOMASS POTENTIALS IN AFRICA 3

Acknowledgement

This report was prepared by Deutsches Biomasseorschungszentrum gGmbH (DBFZ) or the International RenewableEnergy Agency (IRENA).

Questions or comments about this report can be directed to IRENA or to the authors at DBFZ:

IRENA

E-mail: [email protected]

DBFZ

Kitty Stecher (Dipl.-Ing. agr.)

Tel.: +49-341-2434-580

E-Mail: [email protected]

Andr Brosowski (Dipl. Geogr.)

Tel.: +49-341-2434-718


Daniela Thrn (Pro. Dr.-Ing.)

Tel.: +49-341-2434-578


C67 Oce Building

Khalidiyah (32nd) Street

P.O. Box 236

Abu DhabiUnited Arab Emirates

Internet: www.irena.org

International Renewable Energy Agency

IRENA

Torgauer Strae 116

D-04347 Leipzig

Tel.:+49-341-2434-112

Fax:+49-341-2434-133E-Mail: [email protected]

Internet: www.dbz.de


4/44


Contents

Abbreviations 7

Executive Summary 9

1. Introduction 13

2. Literature Review 152.1 Approach 15

2.2 Overview o ormer reviews and studies analysed in the report 16

2.3 Results 18

2.3.1 Findings rom comparisons o the studies 18

2.3.2 Other important driving actors 20

2.3.3 Results o the studies 24

2.4 Discussion 31

2.4.1 Area potential 31

2.4.2 Energy crops 32

2.4.3 Forestry biomass 33

2.4.4 Residues and waste 34

2.5 Conclusion 38

3. Summary 41

References 42

4


5/44


Figures

Figure 1: Based on International Energy Agency (IEA) data or 2009 10

Figure 2: Total primary energy demand or energy sources in Arica (IEA, 2010) 13

Figure 3: Structure o the analysis ollowed in this report 15

Figure 4: Important driving actors to consider when assessing biomass potential 20

Figure 5: Total technical biomass potential or dierent Arican regions and time periods 26

Figure 6: Area potential as dependent on time-period, Arican region and study 27

Figure 7: Energy crops potential as dependent on time-period, Arican region and study 28

Figure 8: Forestry biomass potential as dependent on time-period, Arican region and study 29

Figure 9: Residues and waste potential as dependent on time-period, Arican region and study 30

Tables

Table 1: An overview o the reviewed studies 17

Table 2: Dierences in timerame, geographical coverage and type o potential 19

Table 3: The biomass resources in each study, and its classication 21

Table 4: Other important driving actors 22

Table 5: Residue and waste biomass potential energy data or Arica (PJ/yr.) 25


6/44


Sugarcane FieldHywit Dimyadi/Shutterstock


7/44


AbbreviationsCEMA Conerence o Energy Ministers o Arica

CGPM Crop and Grass Production Model

DBFZ Deutsches Biomasseorschungszentrum gemeinntzige GmbH

EJ Exajoule: 1018J

EU European Union

FAO Food and Agriculture Organization (o the United Nations)

FRA Global Forest Resources Assessment

GAPP Global Agriculture Production Potential Model

GJ Gigajoule : 109 J

GLC Global Land Cover

GLUE Global Land Use and Energy Model

ha Hectare

IEA International Energy Agency

IIASA International Institute or Applied Systems Analysis

IMAGE Integrated Model to Assess the Greenhouse Eect

IPCC Intergovernmental Panel on Climate Change

IRENA International Renewable Energy Agency

LPJml Lund-Potsdam-Jena managed Land Dynamic Global Vegetation and Water Balance Model

Mtoe Million tonnes o oil equivalent

NGO Non-Governmental Organisation

OECD Organisation or Economic Co-operation and Development

PJ Petajoule: 1015 Joule

PBL,NEAA PBL Netherlands Environmental Assessment Agency

SRC Short Rotation Coppice

SSA Sub-Saharan Arica

TPES Total Primary Energy Supply

UN United Nations

UNDP United Nations Development Programme

UNO United Nations Organization

UNSD United Nations Statistical Division

WBGU Wissenschatlicher Beirat der Bundesregierung Globale Umweltvernderungen

yr Year

Conversion factors: 1 Mtoe = 41.868 PJ = 0.041868 EJ1 EJ = 23.88 Mtoe

1 PJ = 0.02388 Mtoe


8/44


Ngorongoro Crater,TanzaniaGary C. Tognoni/Shutterstock


9/44


Arica is currently experiencing strong

economic growth and showing positive

trends in human development indicators.

Access to modern energy will be critical

to sustain these positive signals. The

continent possesses abundant renewable resources

that could spur continued economic growth, accelerate

social development and help the transition to a

sustainable energy system that can provide universal

energy access.

The Arican Union Assembly o Heads o State and

Government in February 2009 decided to develop

renewable energy resources to provide clean, reliable,

aordable and environmentally riendly energy.

The political will to promote renewable energy

was reairmed in the 2010 Arican Union Maputo

Declaration, which established the Conerence o

Energy Ministers o Arica (CEMA).

Ministers o energy and heads o delegations romArican countries, along with the Arican Union

Commission and CEMA, met at the invitation o the

International Renewable Energy Agency (IRENA) in

Abu Dhabi, United Arab Emirates, in July 2011, hosted

by the UAE Government. The subsequent Communiqu

on Renewable Energy or Accelerating Aricas

Development, endorsed by 46 Arican countries

and 25 Arican energy ministers, called or the

promotion o increased utilisation o the continents

vast renewable energy resources to accelerate

development. In the communiqu, the Arican Energy

Ministers also agreed to urther engage with IRENA as

a key intergovernmental orum on renewable energy.

They urged IRENA to provide strategic support or

renewable energy in its message to the international

community at Rio+20-United Nations Conerence on

Sustainable Development.

Arican countries seek to attract large investments in

renewable energy technologies. A thorough planning

phase is necessary to estimate the energy share that can

be supplied by renewable energy sources; to compare

energy costs with that o conventional solutions; and

to identiy the most eective technologies that are

adaptable to local conditions. This enables countries

to develop adequate policy rameworks and inancing

instruments, to nurture human capacity and to create

the necessary inrastructure to ulil their sustainable

energy aspirations.

For countries to properly analyse and plan the use orenewable energy, they need accurate and documented

estimates o their renewable energy potential, as well

as to identiy the most suitable locations or investment

and deployment in renewable energy technologies. The

accuracy o this inormation correlates directly with

the risk taken in the decision-making process . Accurate

data strengthens each countrys national strategy to

deploy renewable energy technologies.

Perorming an accurate estimate o the potential

requires a high level o technical knowledge, extensiveconsultations and large upront investments in

measurement campaigns. This situation is set

to improve as IRENA increasingly provides ree

inormation and tools to assess the renewable energy

potential within a country. The recent release o the

solar and wind component o the Global Renewable

Energy Atlas, intended to expand to other resources,

has initiated a valuable process o knowledge-sharing

among countries and expert institutions on renewable

energy resource data.

In the meantime, however, no detailed standard

methodology exists to estimate the theoretical,

technical, and economic potential or renewable energy

development. An analysis o the available literature on

each renewable energy resource inds discrepancies in

the deinitions used, even or undamental terms such

as renewable and potential 1.

Each study in the literature investigated uses a

signiicantly dierent approach to estimate renewable

Executive Summary

1 Aaiabe te Gba Ata ebite .iena.g/GbaAta


10/44


Figure 1: Based on International Energy Agency (IEA) data or 2009. The charts show the shares o bioenergy in

the total primary energy supply. (A) Arican continent. TPES: 673 Mtoe (28 177 PJ). (B) Sub-saharan Arica. TPES: 517

Mtoe (21 646 PJ) (77% o Arican TPES). (C) Sub-saharan Arica, excl. South Arica. TPES:372 Mtoe (15 575 PJ) (55%

o Arican TPES. 72% o Sub-saharan TPES)

A) AFRICAN CONTINENT

Bioenergy Shares

Solid biouels 47.6%Other renewables 0.2%

Coal and peat 15.7%

Oil 22.4%

Natural Gas 12.4%

Nuclear 0.5%

Hydro 1.3%

B) SUB-SAHARAN AFRICA

Bioenergy Shares

Solid biouels 61.2%

Other renewables 0.2%

Coal and peat 19.7%

Oil 14.1%

Natural Gas 2.7%

Nuclear 0.6%

Hydro 1.4%

C) SUB-SAHARAN AFRICA

(excl. SOUTH AFRICA)

Bioenergy Shares

Solid biouels 81.2%

Other renewables 0.3%

Coal and peat 1.0%

Oil 13%

Natural Gas 2.7%

Hydro 1.9%


11/44


energy potential. With variations in the input datasets

and the main parameters considered, the results o the

various analyses are hardly comparable. Consequently,

decision-makers are aced with a number o

assessments, all o which use dierent methods and

provide diverse results. There is a need or clarity on

the key actors decision-makers must consider about

renewable energy. IRENA, with its ability to convene

countries and stakeholders in the ield, is well placed

to help determine suitable methods and the best

practices to ollow.

This report ocuses on bioenergy in Arica , as this orm

o renewable energy represents almost 50% o the total

primary energy supply (TPES) or the Arican continent

(International Energy Agency (IEA), 2009a), and more

than 60% o the Sub-Saharan TPES. Bioenergy is astrategic asset or Aricas energy uture and needs to

be assessed in a transparent manner (Figure 1).

At IRENAs behest, the German Biomass Research

Centre (Deutsches Biomasseorschungszentrum

gGmbH - DBFZ) has collected recent studies assessing

bioenergy potential in Arica, compared their various

methodologies, benchmarked the results, and identiied

the key dimensioning elements or those assessments.

The outcomes o the analysis are as ollows:

1. The studies show an enormous range o calculated

biomass potentials;

2. The calculated area potential or energy crops

ranges between 1.5 million hectares (ha) and 150

million ha;

3. The various assessments indicate a potential or

energy crops rom 0 PJ/yr to 13 900 PJ/yr, between

0 PJ/yr and 5 400 PJ/yr or orestry biomass and 10

PJ/yr to 5 254 PJ/yr or residues and waste in Arica

by 2020;

Dried cornluckypic/Shutterstock


12/44


4. Signiicant variations ound between the studies

are primarily due to the selected timerame, the

geographical coverage, the type o potential and

biomass resources analysed. Other important

reasons or deviations are the underlying

assumptions, the so-called driving actors, and the

accuracy o the input databases;

5. All the biomass calculations rom today through to

the year 2100 show a considerable range in potential.

For the year 2050, a potential range between 2

360PJ/yr and 337 000 PJ/yr is revealed or Arica

(energy crops: 0 PJ/yr to 317 000 PJ/yr; orestry

biomass: 14 820 PJ/yr to 18 810 PJ/yr; residues and

waste: 2 190 PJ/yr to 20 000 PJ/yr). The potentials

or the year 2100 determined in two studies, vary

between 74 000 PJ/yr and 181 000 PJ/yr;

6. The calculated area potentials and the assumed

production yields have a signiicant impact on the

potential amount o energy crops, orestry biomass,

and residues and waste that are available or energy

provision;

7. In the studies, both the area potentials and

production yields are analysed at dierent levels o

detail and ound generally to have a high degree o

uncertainty;

8. A precondition or achieving high energy crop

potential is an increase in productivity to a level

similar to that o industrialised countries, or a

stronger ocus on energy crop production, e.g.

through the cultivation o energy crop rotations;

9. In contrast studies showing little or no potential

o energy crops in Arica assume that the enormous

population growth and the associated increase

in per capita consumption especially o animal

products will prevent an energy-related use obiomass;

10. One o the key parameters to determine the

potential o orestry biomass, or residues and waste,

is the competitive use o materials;

11. In general, climate change impacts, biodiversity

and social criteria, e.g. land ownership, remain

insuiciently considered in the studies reviewed;

12. A large problem in determining biomass

potentials or Arica is the poor data availability.

Inaccurate and obsoleted databases oten provide

the initial values or the estimations o the current

and uture potentials.

Due to the large range in results presented by the

reviewed studies, no deinite igures regarding the

availability o biomass in Arica can be provided. Anyanalysis based on these studies needs to account

or their underlying assumptions. The existing data,

methodological approach and results, can however be

used to identiy areas or urther research.

This report is a irst step towards bringing clarity

to decision makers on the inormation available

on bioenergy potentials. By understanding the

discrepancies between the existing approaches and

visualising the variability o results between the

dierent methods, decision makers can evaluate

the impact dierent assessments can have on their

strategic decisions.

This report highlights the need for

d ev e l o p i n g r eco m m en d a t i o n s

and standard methods to provide

accurate estimates of bioenergy

potentials.

Bioenergy resource assessment is a particularly

complex subject. The assessment needs to actor

possible conlicts with other land uses, orecast

increases in agricultural productivity, and project

ood demand over time (Belward, et al., 2011). Since

the assessment is extremely sensitive to the local

context and political choices, assessing the bioenergypotentials is a country-driven exercise.

The Global Renewable Energy Atlas initiative began by

oering ree and open access to data on wind and solar

energy. IRENA will progressively expand this initiative,

with the objective o providing relevant inormation,

access to datasets and tools or decision-makers to

assess all types o renewable energy potential, including

biomass. IRENA will continue to work with the bioenergy

community to identiy the relevant inormation, methods

and tools that can help countries wanting to assess their

national potential or biomass development in detail.


13/44


1. Introduction

Bioenergy is currently the primary energy

resource or about 2.7 billion people

worldwide (Wicke, et al., 2011), playing a

traditional role in Arica. The total primary

energy demand (TPED) or Arica is predominantly

determined by biomass demand (Figure 2), with almost

hal o the energy demand (47.9%) being covered by

biomass and waste. The International Energy Agency

(IEA) projects a decline in the total energy share o

biomass and wastes by 20352, but biomass will stillcontinue to remain an important energy resource or

Arica in the uture (IEA, 2010).

As well as the TPED, the total inal consumption (TFC)

or Arica is also predominately composed o biomass

and wastes. The IEA estimates that the total inal

consumption biomass/waste share will lie between

51% and 57% by 2035 depending on which scenarios

are assumed (IEA, 2010). However, there are enormous

regional dierences within Arica. Particularly those

countries characterised by high poverty, the proportion

o biomass is much higher. In some countries, such as

Burundi, Rwanda and the Central Arican Republic,

the energy provision rom biomass is 90% or greater

(Dasappa, 2011). In Sub-Saharan Arica (SSA) people

are more dependent on traditional biomass energycompared to other regions (Eleri and Eleri, 2009). In

this report traditional biomass reers to the unsuitable

use o uel wood, charcoal, tree leaves, animal dung and

agricultural residues or cooking, lighting and space

heating. In comparison to a modern use o biomass,

which normally includes an eective technical utilisation

2 Te jetin ae bae n te Ne iie enai. Ti enai take ant te ba i itent an an tat ae annne b ntie an

te , t take eite eninenta eneg eit nen, een ee te eae t ieent tee itent ae et t be ientiie annne

(IEA, 2010).

Figure 2: Total primary energy demand or energy sources in Arica (IEA, 2010)

40 000

35 000

30 000

25 000

20 000

15 000

10 000

5 000

0

1990 2008 2015 2020 2025 2030 2035

J

Other renewables

Biomass and waste

Hydro

Nuclear

Natural Gas

Oil

Coal


14/44


to produce energy eiciently, traditional biomass is used

with very low energy eiciencies. Moreover, traditional

biomass use can have serious negative impacts on

health and living conditions, e.g. pneumonia, chronic

obstructive pulmonary diseases or lung cancer (IEA,

2009b; Chum, et al., 2011; Legros, et al., 2009).

Despite an increase in energy use, many poor households

in Arica have no access to modern energy sources.

Worldwide, SSA and India have the greatest proportion

o population dependent on traditional biomass use.

In Arica, a total o 657 million people (80% o the

population) rely on the traditional use o biomass or

cooking (IEA, 2010). It is important given this scale o

biomass use that modern bioenergy systems are able

to provide an important contribution to uture energy

systems and to the development o sustainable energy

supplies (Berndes, Hoogwijk and Broek, 2003).

It is thereore crucial to use the potential renewable

energy resources eectively. In addition, against

the background o ood security, i biomass energy

is developed in a sustainable manner, this creates

opportunities or increased ood production, especially

in rural areas (Eleri and Eleri, 2009).

In general, the availability and quality o biomass energy

data in Arica is scarce. There are some country-speciic

studies, as well as various international research projects

that ocus on the estimation o global biomass potentials,

which provide some inormation about Arica. Although

a ew publications engage with this ield o research,

there is no scientiic literature review or Arica to date.

It is important that in making statements about the

prospects or development o sustainable bioenergy

rom biomass sources in Arica, that a realistic order o

magnitude or biomass potentials is assessed. This project

report aims to provide an overview and comparisono the available studies relating to biomass potentials

in Arica. The considerable discrepancies among the

studies results will be illustrated and explained, and the

requirements concerned with determining parameters

and the other challenges associated with potential

estimates will be discussed.

This report contains an overview o the selected studies.

The indings are presented and discussed; they concern

the methodology and analysis o results or dierent

biomass categories, including the uncertainties in the

database and the broad deviations o the study results.

The report closes with conclusions and remarks.Food on market placeZurijeta/Shutterstock


15/44


2. Literature Review

2.1 ApproAchThe approach in this study (Figure 3) rst

evaluates existing literature reviews on

global biomass potentials. In particular,

attention is given to those biomass studies published

ater 2005 that have included inormation about Arica.

The number o studies ound within this timerame is

small and the search expanded to include those studies

published ater 2000. Additional studies ocus on specic

Arican countries. By including some country-specicstudies it became possible to compare the national

biomass potentials and to make comparisons with larger-

scale continental and global studies. A total o 14 studies

are used or the analysis.

The study analysis is divided in two sections (Figure 3). In

the rst section, the methodology and assumptions used in

the selected studies are analysed. To enable a comparison,

several criteria are established, which include the denition

o the biomass potential, the timerame, the geographic

coverage and the underlying biomass resources considered

in the studies. Another criteria used to compare studies

is the group o other important driving actors, which

combines all parameters that impacted on the amount o

biomass potential. These parameters are both listed and

classied in a database, with special attention paid to their

source and data quality.

In the second section o the analysis, study results anddeclared potentials are evaluated, creating a comparative

basis where the results are grouped by the our categories:

available land or energy production; energy crops; orestry

biomass; and residues and waste.

The discussion identied the main causes or discrepancies

in the results. All results rom the our biomass potential

categories are discussed, and the main parameters

compared. The question o where to ocus urther research

is highlighted in the conclusion.

Figure 3: Structure o the analysis ollowed in this report

Literature research

Existing reviews

Studies on a

global scale

Studies on a

continental scale

Selection

of

relevant

studies

Conclusion

Analysis

Timeframe

Geographical

coverage

Biomass

resources

Other important

driving factors

Methodology

used in the studies

Area potential

Energy crops

Forestry biomass

Residues & waste

Biomass

potentials Discussion


16/44


2.2 ovErvIEw of formEr rEvIEwsANd sTudIEs ANAlysEd IN ThE rEporTThree literature reviews are presented hereunder,

comparing dierent methodologies used to assess

global biomass potentials, the inal results ound and

the input databases used.

In Berndes, Hoogwijk and Broek (2003), 17 studies are

analysed. The conclusions o the reviewed studies are

discussed with regard to the underlying assumptions and

methodology (timerame, geographical aggregation and

bioenergy resources).

Moreover, the review concentrates on the classiication

o studies in demand-driven3 and resource-ocused4

assessments. The possible contribution o biomass touture global energy supply ranges rom below 100

exajoules per year (EJ/yr) to over 400 EJ/yr by 2050.

As a matter o comparison, the global energy demand

in 2010 reached 12 380 Mtoe in 2010, equivalent to 518

EJ (IEA, 2012).

Most o the studies consider biomass plantations as the

most important source o biomass or energy (e.g. biomass

plantation supply is orecast to be between 47 EJ/yr and

238 EJ/yr by the year 2050). Berndes, Hoogwijk andBroek (2003), ound two major parameters explaining

the discrepancies among the literature reviewed: land

availability; and yield levels in energy crop production.

Both parameters although uncertain, are crucial to

estimating uture biomass potentials.

In a comparison o 19 studies by Thrn, et al. (2010a),

there was ound to be a large range in results or

biomass ractions o energy crops, organic residues

and waste. The largest dierence between the results inthis literature review is ound in the potential o energy

crops. While some studies indicate an energy crop

potential o 0 EJ/yr by 2050, optimistic estimations o

values reach 1 272 EJ/yr or this time period. The global

potential o orest residues ranges rom 0 EJ/yr to 150

EJ/yr in 2050. In addition to analysing results, Thrn, et

al. (2010a) highlights the dominant inluencing actors

on the amount o biomass potential, with the most

important being global population growth; ollowed

by uture per capita consumption and development o

speciic yields or ood, including odder and biomass

production (Thrn, et al., 2010a).

The third literature review, by Chum, et al., (2011),

analyses 15 scientiic studies in which global biomass

potentials are calculated. Although the various actors

inluencing the amount o the potential are named, there

is no precise speciication attached to these actors.

The literature review comes to the conclusion that the

global potential o biomass or dierent categories5

in the studies ranges rom less than 50 EJ/yr to more

than 1 000 EJ/yr by 2050 (Chum, et al., 2011).

All the literature reviews highlight the enormous

range in global biomass potential studies. Although

some reviews analyse the regional dierentiation o

global biomass potentials, no igures on Arica are

reported, or example, Berndes, Hoogwijk and Broek

(2003), distinguished between the calculated biomass

potentials o developed and developing countries.

According to the estimates from mostof the reviewed studies, developing

countries will account for the largest share

of global energy supply in the longer term

(Berndes, Hoogwijk and Broek, 2003).

Fourteen studies analysed in the ollowing section

(Table 1), make more precise statements about thebiomass potential in Arica. Those ourteen studies

include nine global, three continental and two country-

speciic studies. All studies indicate numerical estimates

o the biomass potentials or dierent biomass ractions.

The exact geographical coverage or the calculated

potentials can be taken rom Section 2.3.1.

3 dean-ien aeent anae te etitiene bia-bae eetiit an bie, etiate te eqie bia t eet exgen taget a

iate-neta eneg (ean ie ) (Bene, hgijk an Bek, 2003).

4 In ee-e aeent te tta bieneg ee bae an te etitin bet een ieent ee e ( i e) ae anae (Bene, hgijk an Bek, 2003).

5 Te ategie ine eie agite (15 EJ/ t 70 EJ/.), eiate bia tin n agita an (0 EJ/ t 700 EJ/.), eiate bia

tin n agina an (0 EJ/ t 110 EJ/.), et bia (0 EJ/ t 110 EJ/.), ng (5 EJ/ t 50 EJ/.) an gani ate (5 EJ/ t e 50 EJ/.).


17/44


Table 1: An overview o the reviewed studies.

Acronym 6 Reference

BATIDZIRAIBatidzirai, B., A.P.C. Faaij and E.Smeets (2006), Biomass and bioenergy supply rom Mozambique, Energy or

Sustainable Development, Vol. 10, No. 1, pp. 54-81 (doi:10.1016/S0973-0826(08)60507-4).

BMVBS

BMVBS (2010), Globale und regionale rumliche verteilung von biomassepotenzialen - status quo und mglichkeit

der przisierung; Endbericht, BMVBS-Online-Publikation, (www.bbsr.bund.de/cln_032/nn_629248/BBSR/

DE/Veroeentlichungen/BMVBS/Online/2010/DL_ _ON272010,templateId=raw,property=publicationFile.pd/

DL_ON272010.pd).

COOPERCooper, C.J. and C.A. Laing (2007), A macro analysis o crop residue and animal wastes as a potential energy

source in Arica, Journal o Energy in Southern Arica, Vol. 18, No. 1, pp. 10-19.

DASAPPA

Dasappa, S. (2011), Potential o biomass energy or electricity generation in Sub-Saharan Arica, Energy or

Sustainable Development, Vol. 15, No. 3, pp. 203-213 (doi:10.1016/j.esd.2011.07.006).

DUKUDuku, M.H., S. Gu, and E.B. Hagan (2011), A comprehensive review o biomass resources and biouels potential

in Ghana, Renewable and Sustainable Energy Reviews, Vol. 15, No. 1, pp. 404-415.

FISCHERFischer, G. and L. Schrattenholzer (2001), Global bioenergy potentials through 2050, Biomass and Bioenergy,

Vol. 20, No. 3, pp. 151-159, (doi:10.1016/S0961-9534(00)00074-X).

HABERLHaberl, H., et al. (2011), Global bioenergy potentials rom agricultural land in 2050: Sensitivity to climate change,

diets and yields, Biomass and Bioenergy, Vol. 35, No. 12, pp. 4753-4769 (doi:10.1016/j.biombioe.2011.04.035).

HAKALAHakala, K., Kontturi, M., and Pahkala, K (2009), Field o biomass as global energy source, Agricultural and Food

Science, Vol. 18, No. 3-4, pp. 347-365.

HOOGWIJKHoogwijk, M. (2004), On the global and regional potential o renewable energy sources, PhD Thesis, Utrecht

University, Utrecht.

KALTSCHMITTKaltschmitt, M.,H. Hartmann, and H. Hobauer (2009), Energie aus biomasse: grundlagen, techniken und

verahren, Springer, pp 1032(ISBN: 9783540850946).

SMEETS

Smeets, E.M.W., et al. (2007), A bottom-up assessment and review o global bio-energy potentials to 2050,

Progress in Energy and Combustion Science, Vol. 33, No. 1, pp. 56-106 (doi:10.1016/j.pecs.2006.08.001).

WECWorld Energy Council (2010), 2010 Survey o energy resources, London, pp. 618, www.worldenergy.org/

publications/3040.asp

WICKE Wicke, B., et al. (2011), The current bioenergy production potential o semi-arid and arid regions in sub-Saharan

Arica, Biomass and Bioenergy, Vol. 35, No. 7, pp. 2773-2786 (doi:10.1016/j.biombioe.2011.03.010).

YAMAMOTO

Yamamoto, H., J. Fujino, and K. Yamaji (2001), Evaluation o bioenergy potential with a multi-regional

global-land-use-and-energy model, Biomass and Bioenergy, Vol. 21, No. 3, pp. 185-203 (doi:10.1016/S0961-

9534(01)00025-3).

6 An ae e in Tabe 1 t itingi te tie te iteate eeene ae tgt ti et.


18/44


2.3 rEsulTs

The substantial dierences ound between the studies are

due to complex reasons. The main dierences between

the various methodologies and assumptions used are:

Timerame

Geographical coverage

Denition o potential

Biomass resources

Other important driving actors

Section 2.3.1 includes all results rom the calculated

potentials in the reviewed studies. The biomass potentials

are classied into the ollowing categories: area potentialand energy crops, orestry biomass, and residues and

waste.

2.3.1 Findings rom comparisons o thestudies

The irst signiicant dierence between the studies is

the selected timerame (Table 2). The majority o the

studies show values or the present-day situation. In

order to project uture energy potentials in Arica,

most studies calculate up to the year 2050. Very little

inormation is available or the short- and mid-term

orecasts (2020-2030), or or the longer-term (year

2100). In the studies, various parameters such as

uture ood and eed demand, yields rom crop species,

environmental issues, etc. (see section 2.3.2) are taken

into account. Their signiicance in assessing the uture

biomass potential depends strongly on the time period

selected.

Secondly, the studies dier in the geographical area

covered. With the exception o country-speciic

studies, all other studies calculate and summarise

biomass potentials or a number o Arican countries

that are then reerred to as overall Arica or SSA

(Table 2). In the SSA studies, the analysis o potentials

in North Arica is combined with West Asia, Middle

East or Near East. This geographical delimitation

means it is not possible to use these studies to

make a statement on the potential or the whole o

the Arican continent. Furthermore, the studies also

varied in the number o countries considered, i this

value was given at all. These variations in area need

to be considered when interpreting the results o the

studies.

Thirdly, the heterogeneous use o the term potential in

the studies diers, but can generally be distinguished as

theoretical, technical or economic potential.

The theoretical potential describes the theoretical

maximum energy supply that is physically available in

a given region within a speciied period o time, (e.g.

energy saved in the entire crop mass). It is determined

by its limits o physical utilisation and thereore is

the upper limit to the provision o energy that can

be theoretically realised. As not all o the theoretical

energy potential can be realised it is o no practical

relevance in assessments o actual biomass availability.

The technical potential describes the theoreticalpotential that can be used ater accounting or energy

losses through the technical conversion process. The

calculation o the technical potential oten includes

the structural, ecological, administrative and social

limitations as well as legal requirements, since they

ultimately can be regarded as insurmountable, in

a way that is similar to the technically determined

limitations.

The economic potential is the share o the technical

potential which is economically proitable in given

conditions (Rettenmaier, et al., 2008).

To clariy urther the available level o potential,

additional terms are used besides the theoretical,

technical and economic potential. For example,

the implementation potential describes the actual

contribution to energy supply. This potential depends

on a variety o other socio-political and practical

constraints (Thrn, et al., 2010b). An estimation o the

implementation potential can only be assessed at aregional level, requiring an enormous eort, because

o the number o restrictions that have to be taken into

account (e.g. considering the willingness o the biomass

producers to sell the eedstock). An assessment o the

implementation potential or the whole o Arica is

currently not practical.

All the studies reviewed evaluate the technical potential

(Table 2) and in three studies, the economic potential

is also determined. Although the same term is used in

the studies, the underlying assessment actors o the

potentials vary considerably. Technical, structural and

environmental constraints, as well as legal requirements


19/44


Table 2: Dierences in timerame, geographical coverage and type o potential

Acronym Time period Geographical coverage

Type of potential

Technical Economical

BATIDZIRAI 2015 Mozambique X X

BMVBS2007; 2010; 2015;

2020; 2050Arica X

COOPER Current Arica X

DASAPPA Current SSA X

DUKU Current Ghana X

FISCHER 2050 SSA X X

HABERL 2050 SSA X

HAKALA 2006; 2050 Arica X

HOOGWIJK 2050; 2100 Arica X

KALTSCHMITT Current Arica X

SMEETS 2050 SSA X

WEC Current Arica X

WICKE Current

Botswana, Burkina Faso,

Kenya, Mali, Senegal, South

Arica, Tanzania, Zambia

X X

YAMAMOTO 2100 SSA X

are accounted or as each author sets out a range

o actors which maniest themselves to dierent

extents, depending on the system boundary and the

biomass raction considered. The biomass potential is

thereore not strictly deined, but is related to numerous

assumptions and boundary conditions (BMVBS, 2010;

Kaltschmitt, Hartmann and Hobauer, 2009). Anoverview o the driving actors considered in the studies

is given in Section 2.3.2.

Finally, there is no uniorm classiication or the

description o types and ractions o biomass resources.

Biomass being deined as the biodegradable raction o

products , waste and residues rom biological origin rom

agriculture (including vegetal and animal substances),

orestry and related industries including isheries and

aquaculture, as well as the biodegradable raction o

industrial and municipal waste (European Union (EU),

2009).

To compare the amount o potentials calculated in thereviewed studies, the dierent biomass resources are

divided into three biomass ractions:

Energy crops

Forestry biomass

Residues and waste


20/44


Table 3 shows the biomass resources considered in the

studies and their allocation within the three biomass

ractions. The resources included in Table 3 identiy

the biomass ractions used exclusively in the Arican

studies. All o the studies provide an estimate o the

potential or the biomass raction rom residues and

waste, whereas the eedstock type is dierent orevery study. In ten studies, the energy crop potential

is determined, while the orestry biomass potential is

only calculated in six o the studies.

The type o biomass ound in the dierent biomass

ractions varies greatly. While some studies calculated

only the potential o a particular crop type (e.g.

eucalyptus), other studies reer to a large number

o crop species. Some studies ail to identiy the

eedstock composition or its raw materials raction,

instead providing a general description o the biomassresources (e.g. energy crops on grassland or crop

residues).

Moreover, a stringent classiication o eedstock is not

always possible, because the biomass ractions overlap

each other (e.g. orest products and wood residues).

Not all studies are clear about the exact eedstock they

consider, which may inluence greatly the biomass

potential.

2.3.2 Other important driving actors

The biomass potential is largely determined by the actors

shown in Figure 4. The demand or ood grows with an

increasing population; and the per-capita consumption is

also dependent on a countrys economic development.

With limited potential area, the increased demand may beoset by increases in crop yield. Otherwise, the agricultural

acreage needs to expand (e.g. through deorestation) or

ood imports need to be increased to meet the populations

demand.

These actors can all operate as driving actors that cover

certain social, environmental or economic dimensions, or

can be seen as restrictions that directly limit the biomass

availability. The driving actors in the selected studies are

determined by parameters that are considered by the

authors o each study, dierently.

Table 4 presents the associated databases or the driving

actors used by authors in their studies. These include

or example, areas o land with special protection status

(e.g. biodiversity, conservation areas) that can reduce

the potential area or bioenergy production, and climate

change with its eect on production yields. Yields are

also infuenced by both economic development and

agricultural productivity o a country. Besides these

broader dierences between the models used in the

Figure 4: Important driving actors to consider when assessing biomass potential

Population

FOOD DEMAND

Yields Area potential

Per-capita

consumption


21/44


Table 3: The biomass resources in each study, and its classifcation.

Acronym

Biomass fraction

Energy crops Forestry biomass Residues and waste

BATIDZIRAI Eucalyptus Logging residues, industrial waste,bagasse

BMVBS

Cereals, maize, rape, sunfower, soya,

sugar beet, sugarcane, oil palm,

grassland

Fuel-wood

Straw, municipal waste (bio-waste

and wood residues), animal residues,

industrial wood waste, logging

residues, coconut, oil palm, bagasse

COOPERCrop residues, animal waste,

bagasse, coconut shells

DASAPPA Fuel-wood Crop residues, logging residues,industrial waste wood

DUKU Crop residues, coconut, oil palm

FISCHERBiomass (e.g. energy crops) on

grasslandForest products Crop residues

HABERLEnergy crops on cropland and on

other land (i.e. grazing land)

Residues on cropland

HAKALA Not claried Crop residues

HOOGWIJK Energy crops

KALTSCHMITT Not claried Fuel-wood Crop residues, animal residues

SMEETS

Woody energy crops (e.g.

eucalyptus, willow), conventionalcrops (e.g. sugarcane, wheat, maize)

and grasses (e.g.miscanthus)

Crop harvest residues, crop process

residues, wood harvest residues,wood process residues, wood waste

WEC Bagasse

WICKEAtrophy, cassava, short rotation

coppice

YAMAMOTO Not claried Fuel-wood

Human aeces, kitchen reuse,

animal dung, harvesting residues,

timber scrap, paper scrap, sawmillresidues, black liquor, uel wood/

industrial elling residues


22/44


analysis, variations occur with regard to the application

and combination o data sources. Sometimes, very

complex phenomena (e.g. climate change and the issue

o sustainability) are excluded or not dealt with to the

correct level o detail in the studies, and yet some o the

reviewed studies use complex data models to simulatecurrent and uture material fows. Variations also extend to

the database chosen and the time period considered, and

the dierences may signicantly infuence the extent and

the range in results.

In the ollowing box, inormation about the individual

databases is compiled. Complex models tend to use

statistical data rom the FAO and other UN entities.

Unortunately there are no real alternatives to these

datasets and the data quality or Arica is poor. For

example, the data used rom remote sensing, e.g. GlobalLand Cover (GLC) oten has no reerence to any reporting

requirements rom the country concerned. The GLC

database was produced in 2000, and is now out o date,

because o the dynamic land use in Arica.

Table 4: Other important driving actors

Driving factor Database used

Social dimension

Food demand

- Population

- Per capita consumption

United Nations Development Programme (UNDP), United Nations (UN), Global Land

Use and Energy Model (GLUE), Integrated Model to Assess the Greenhouse Eect

(IMAGE), International Institute or Applied Systems Analysis (IIASA), Global Agricultural

Production Potential Model (GAPP)

Food and Agriculture Organization (FAO), GLUE, UN, GAPP

Environmental dimension

Biodiversitywww.biodiversityhotspots.org, Wissenschatlicher Beirat der Bundesregierung Globale

Umweltvernderungen (WBGU)

Conservation areas World Database on Protected Areas (WDPA), GLUE

Climate change/

environmental conditions

like deorestation

Lund-Potsdam-Jena managed Land Dynamic Global Vegetation and Water Balance

Model (LPJml), Intergovernmental Panel on Climate Change (IPCC), WBGU, IMAGE, FAO,

GLUE, Literature

Economic dimension

Economic growth FAO, World Bank, Quickscan Model, IMAGE

Costs (production,

transport, conversion)FAO, World Bank, Literature, Expert decision

Eciency o agricultural

production system

Crop and Grass Production Model (CGPM), IMAGE, FAO, IIASA, GLUE, IPCC, GAPP

Literature

General

Land use Global Land Cover(GLC); FAO, GLUE, IMAGE, IIASA, GAPP, Literature

Biomass demand/material

utilisationIMAGE, GLUE, Literature


23/44


FAO: The FAO is a UN specialised agency that manages a comprehensive database o reported national data. In

general, data provision is based on the reporting o governments (ministries), non-governmental organisations(NGOs) or private, as well as international authorities (FAO, 2011a). The FAO Statistics Division is responsible or the

management o the data pool e.g. FAO holds statistics about ood, agriculture, orestry and sheries in FAOSTAT,

a statistical databases collection (FAO, 2011b). The database itsel is connected with national or regional statistical

inormation systems (e.g. CountrySTATS; FAO, 2011c). Additionally, the FAO uses databases rom other international

organisations, e.g. United Nations (UN) in order to carry out secondary statistical research (United Nations Statistics

Division, 2011a).

UN: The United Nations Statistical Division (UNSD) perorms, or example, energy statistics through national surveys

within the member countries (Energy Statistics Database; UNSD, 2011). The UNSD manages the data pool o the UN

(web based inormation systems UNDATA) and is supervised by the United Nations Statistical Commission (UNSC).

UNDP: As a member within the United Nation system, the UNDP is part o the UNO data pool. The organization also uses

surveys rom other international institutions, such as the Organisation or Economic Co-operation and Development

(OECD), World Bank, etc. (UNDP, 2011).

GLC: The Global Land Cover, produced in 2000 provides global land use data. The data on global spatial inormation

with a resolution o 1 km were recorded during a 14 month period (November 1999 to December 2000) using SPOT4

satellites (European Commission, Joint Research Centre 2011).

IMAGE: This database was initiated by the Environmental Assessment Agency o the Netherlands (PBL, NEAA, 2011a).

The model simulates the environmental impacts o human activities and processes demographical, technological,

economic, social, cultural and political data (Bouwman, Kram and Klein Goldewijk, 2006). Secondary statistical data

sources are used, including values rom the World Bank, UN, FAO, etc.

IIASA: This database relates to the development o orestry and above-ground biomass, the IIASA used the results o

the Global Forest Resources Assessment 2005 (FRA 2005) (FAO, 2011d). This biomass data has been extrapolated

with a linear inter-relationship to country level on a reerence area with a 0.5 grade (longitude/latitude) (IIASA, 2011).

The report provides an estimation o the global orestry inventory or 1990, 2000 and 2005. It comprises reports rom

229 countries and results developed in regional workshops (FAO, 2011d).

CGPM: The CGPM ollows the approach developed by the FAO (between 1978 and 1981) using agro-ecological zones

and is based on the world soil map o the FAO (1991). It considers in particular, three energy crops; sugarcane, corn andwood. Input data or this model include temperature, rainall, soil data, carbon dioxide concentrations, plant specic

growth parameters and yields. Furthermore, the model results serve as the data source or other models like IMAGE

or the LCM2000 (PBL, NEAA, 2011b).

GLUE: This Model developed by Yamamoto, Junichi and Kenji (2001), considers land use competition and overall

biomass fow. The model is based on data rom FAO 1992, World Bank 1993, Intergovernmental Panel on Climate

Change (IPCC) 1990/1992 and also data collected rom the literature (Yamamoto, Junichi and Kenji, 2001).

GAPP:The Model developed by the University o Hohenheim, calculates the area potential. Several structural parameters,

e.g.areas under cultivation, animal livestock, yields, population and per-capita consumption, are considered (BMVBS,

2010).

dETAIls of ThE mAJor dATABAsEs usEd By ThE sTudIEs


24/44


2.3.3 Results o the studies

Figure 5 presents the overall results categorised by time

period and region as a logarithmic interpolation. The range

o all study results lies between 242 PJ/yr (WEC) and

337 000 PJ/yr (SMEETS) or the entire period under

review. The results or the Arican continent up to 2020vary between 242 PJ/yr (WEC) and 6 679 PJ/yr (BMVBS).

Looking long-term (2050 to 2100) the technical potential

is expected to lie between 2 360 PJ/yr (HAKALA) and

337 000 PJ/yr (SMEETS). A detailed presentation o

results according to dierent biomass categories area

potential and energy crops, orestry biomass, and residues

and waste can be ound in the ollowing sections.

Area potential and energy crops

The area available or the biomass production is a crucialparameter to calculate the quantity o raw material and total

energy crop potential. Figure 6 summarizes the values o

areas potentials calculated by ve o the reviewed studies.

According to these ve studies, SSA has an area potential

o at least 17.6 million ha, with Arica reaching a maximum

area potential o 150 million ha. In the long term (2050), an

area potential up to 717 million ha is assumed.

Over the short- and mid-term time period (2015-2020)

the area potential calculated by BATIDZIRAI provides

the highest values ound in any paper, albeit only or

Mozambique. Other high values ound include those rom

KALTSCHMITT (Figure 6). For the time period leading

to 2020, BMVBS orecasts the largest range o values

or a number o dierent scenarios (1.56 million ha to

20.6 million ha).

According to Figure 7, the studies that show a high area

potential over a specic time also have the highest amounts

o energy crop potential (see KALTSCHMITT, BATIDZIRAI and

SMEETS). However, in general, the energy crops potential,or the period up to 2020, ranges rom 0 PJ/yr (HAKALA)

to 13 900 PJ/yr (KALTSCHMITT). The high energy crop

potential value or Arica o 13 900 PJ/yr determined by

KALTSCHMITT, is in sharp contrast to the average value or

Arica o 992 PJ/yr, rom studies or the same time period.

More values are ound rom studies or the year 2050,

with a low energy crops potential o 0 PJ/yr calculated by

HAKALA and BMVBS, which is in contrast to the value o

317 000 PJ/yr produced by SMEETS. Finally, two o the

evaluated studies provide gures or the year 2100. While in

YAMAMOTO no energy crop potential is expected in 2100,

HOOGWIJK assumes a potential between 74 000 PJ/yr and

279 000 PJ/yr.

Forestry biomass

Five studies present results or the category orestry

biomass (Table 3). The authors use the term uel wood

potential with the exception o FISCHER who reers to

the data with the expression orestry products; a term,

which is not clearly deined. An overview on the resultso the orestry potential is shown in Figure 8.

A direct comparison o results is not possible, because o

variations between the studies. The studies o DASAPPA,

FISCHER and YAMAMOTO all reer to SSA, but relate

to dierent time periods. They present a potential or

SSA increasing rom 0 PJ/yr or the present-day time to

75 000 PJ/yr or the year 2100.

Two o the studies (FISCHER and BMVBS) represent

several scenario values or dierent time periods. TheBMVBS report speciies three scenarios or 2020 and

quantiies uel wood potential between 949 PJ/yr and

2 459 PJ/yr, although these values reer only to our

Arican countries (South Arica, Nigeria, Democratic

Republic o Congo and Ethiopia).

The uel wood potential o Nigeria by 2020 is predicted

to be 0 PJ/yr, whereas South Arica (242 PJ/yr to 1 200

PJ/yr) and the Democratic Republic Congo (558 PJ/yr to 1

123 PJ/yr) provide the highest percentage o the overall

values. The results or dierent scenarios presented by

FISCHER range rom 14 820 PJ/yr to 18 810 PJ/yr. Whereas,

KALTSCHMITT indicates a current orest biomass potential

o 5 400 PJ/yr or the whole Arican continent.

Residues and waste

Nearly all studies used in this report present data or

waste and residual potentials; an overview is shown in

Figure 9. In comparison to the wide range o results in

the previous ields o energy crops and orestry biomass,

the values or waste and residues show only moderatevariation lying between 2 100 PJ/yr and 5 200 PJ/yr. The

mean value rom the studies o KALTSCHMITT, HAKALA,

BMVBS, COOPER, HABERL and FISCHER is calculated

as 2 900 PJ/yr. Even accounting or the dierent time

periods, the deviation rom the calculated average value

is small.

Exceptional low values or waste and residual potentials

can be ound in the studies covering larger areas (WEC,

242 PJ/yr and DASAPPA, 585 PJ/yr), as well as in the

country-specic studies or Ghana (DUKU, 0.1 PJ/yr, data

not presented in Figure 9) and Mozambique (BATIDZIRAI,

10 PJ/yr). Comparatively high gures or the year 2050 are


25/44


ound by SMEETS (12 000 PJ/yr to 20 000 PJ/yr) and or

2100 by YAMAMOTO (25 000 PJ/yr).

The reviewed studies consider dierent biomass categories

in their analysis. In Table 5, the results or each biomass

category are presented. The crop residues raction has been

analysed in the majority o studies in this report; in total 10studies present results including this agricultural by-product.

Comparing these studies in Figure 9, it can be noted that

HAKALA (2 380 PJ/yr to 2 600 PJ/yr), BMVBS (1 161 PJ/yr)

and COOPER (1 089 PJ/yr to 3 588 PJ/yr) report similar

results or the Arican continent or the current time

period. Yet in contrast to these ndings, DASAPPA states

a comparatively low yield o only 135 PJ/yr or the SSA

countries. In the longer-term (2050) HAKALA expects a

potential similar, i slightly lower than that ound today.

Signicant dierences occur in a detailed comparison o

crop residue data rom the SSA. The results range rom

between 2 190 PJ/yr (HABERL) and 20 000 PJ/yr (SMEETS).

A crop residue potential o approximately 10 500 PJ/yr,

which is comparatively high or the year 2100, is calculated

by YAMAMOTO.

A large range in results also occurs or animal residues.

COOPER (1 450 PJ/yr) and KALTSCHMITT (1 200 PJ/yr)

present similar ndings; with the BMVBS study indicating

a low yield o 28 PJ/yr or the same region and time

period. Signiicant dierences are also shown or

industrial wood waste (0 PJ/yr to 356 PJ/yr) and or

logging residues (0 PJ/yr to 822 PJ/yr). The estimated

potential or bagasse (201 PJ/yr to 242 PJ/yr) and

coconut shells (11 PJ/yr to 15 PJ/yr) are almost identical

in the two studies that included them.

Table 5: Residue and waste biomass potential energy data or Arica (PJ/yr.)

StudyCrop

residues

Animal

residues

Municipal

waste

Industrial

wood waste

Logging

residuesBagasse

Coconut

shellsOil palm

KALT-

SCHMITT900 1 200 - - - - - -

HAKALA

2 380-2 600

(current)

2 360-2 370(2050)

- - - - - - -

BMVBS 1 161 28 353 4 822 208 11 88

COOPER 1 089-3 588 1 450 - - - 86-201 15 -

WEC - - - - 242 - -

DASAPPA 135 - - 356 94 - 0.001 0.007

DUKU 0.07 - - - - - - -

BATIDZIRAI 2 1 8 - -

HABERL 2 190 - - - - - - -

FISCHER 4 884 - - - - - - -

SMEETS12 000-

20 000- - 0 0 - - -

YAMAMOTO42 %

(10 500)not specied - not specied not specied - - -


26/44

26 IrENA-dBfZ: BIomAss poTENTIAls IN AfrIcA

Figure 5: Total technical biomass potential or dierent Arican regions and time periods

]ry/JP[laitnetopssamoiB

21400

2380

6

096

2626

242

5223

585

6147

6307

6680

7686

134000

2360

23440

46000

279000

100000

2600

6679

4845

5239

4588

4195

4

162

51000 2

370

122000

125000

5728

4

450

4020

3671

3624

91000

280000

181000

5169

15000

337000

74000

110

100

1000

1

0000

10

0000

100

0000

KALTSCHMITT

HAKALA

BMVBS

2003-2007

BMVBS

2010

COOPER

WEC

WICKE

DASAPPA

BMVBS

2015

BMVBS

2020

BATIDZIRAI

BMVBS

HOOGWIJK

HAKALA

HABERL

SMEETS

HOOGWIJK

YAMAMOTO

SSA

Mozambique

SSA

SSA

2015-2020

2100

2050

currently

Study

Region

Timeframe


27/44

IrENA-dBfZ: BIomAss poTENTIAls IN AfrIcA 27

Figure 6: Area potential as dependent on time-period, Arican region and study


28/44


Figure 7: Energy crops potential as dependent on time-period, Arican region and study



29/44

IrENA-dBfZ: BIomAss poTENTIAls IN AfrIcA 29

Figure 8: Forestry biomass potential as dependent on time-period, Arican region and study


30/44


Figure 9: Residues and waste potential as dependent on time-period, Arican region and study



31/44


2.4 dIscussIoN

The potentials calculated by the reviewed studies reer

to dierent time periods, geographical regions and

biomass ractions. The studies thereore require careul

consideration beore comparing their results. The

comparisons are hampered by a dierent denition orthe technical potentials (see Section 2.3.1). Ultimately only

a ew parameters are incorporated into the calculations.

This report nds major dierences between the reviewed

studies, applied methodology, underlying databases and

the assumptions made.

In the ollowing, the results or the area potential, energy

crops, orestry biomass and residues and waste are

discussed. Signicant dierences are expounded.

2.4.1 Area potential

The area potential is ound to be a critical variable

in determining the quantity o raw material. Most o

the studies evaluated ail to mention the calculated

area potential. The results o the remaining studies

demonstrate that large area potentials can lead to high

potentials o energy crops. The availability o land and

the production yield o the energy crop are inherently

uncertain parameters and both tend to be estimated

dierently (Berndes, Hoogwijk and Broek, 2003). For

assessing the uture area potential, the studies estimated

the available land, which depends on population growth,

eating habits, increased urbanisation, and so on. In

theory an increase in yield could lead to a release o land

or energy crop cultivation.

O all the studies reviewed, only the BMVBS study shows

the development o the area potential or dierent time

periods. In two scenarios, a reduction in the area potential

is predicted. The area potential is assumed to rangerom between 1.5 million ha and 1.7 million ha by 2020,

and 0 million ha in 2050. These calculations are greatly

aected by the increased demand or ood. An increased

potential may only be realised (25.6 million ha in 2050)

i biomass energy use is to be supported, or example,

by government policies that may incentivise armers to

cultivate the most ecient crops (e.g. Short Rotation

Coppice (SRC), corn silage). Policies are not explored

urther in this report. The assessment o the current area

potential by BMVBS assumes that land set aside rom

ood production can be used and that additional land

will be available i agricultural overproduction is used or

bioenergy production.

All other studies indicate the area potential or a given

year, without depicting the development o this potential.

From the literature examined, by ar the highest area

potential was determined by KALTSCHMITT (150 million

ha). Generally the authors assumed that 10% o agricultural

area could be available or energy crop cultivation inthe world. In contrast, a more explicit determination o

area potentials was conducted by WICKE with a GIS-

approach. The available land (34.4 million ha) or energy

crop production resulted rom the dierence between the

total area and the non-usable area. The latter included

unsuitable areas (e.g. cities, sandy desert and dunes), high

biodiversity areas and agricultural land.

The uture area potentials assessment is carried out in only

three o the studies reviewed. In BATIDZIRAI, the calculated

area potential or Mozambique (45 million ha) is highmainly due to the increased eciency in ood production.

Despite the increasing demand or ood, more area is

made available through ecient crop production. It must

be noted that this potential can only be achieved i in 2015

the average crop yields increases by more than our times

(rom 4.9 to 8.9) and the average eed conversion ratio

more than doubles (rom 1.3 to 4.5). The resulting yields

and the eed conversion eciency in Mozambique would

then be equivalent to an industrialised countrys highly

ecient production system.

Similar to the assumptions used in BATIDZIRAI, SMEETS

calculates an area potential o 717 million ha or SSA in 2050

that is dependent on the agricultural production system.

The precondition is also very high levels o technology

or crop production and eed conversion eciencies in

Arica; this includes an increased use o ertilisers and

agrochemicals. To date, globally, SSA has one o the lowest

ertiliser usages in the world (Hakala and Pahkala, 2009).

Another point raised is that ood cultivation is projected

onto areas that achieved the highest yields per hectare ora particular crop. Only through keeping ood cultivation

areas to a minimum and optimising production will there

be more area available or energy crop production.

Due to the small number o studies available, it is dicult to

make any general statements about the major causes o the

signicant dierence ound between the area potentials.

Although it is possible to determine the area potential,

using a general calculation (x% o total agricultural area),

which is the simplest approach compared to those ound

in the studies; however, the assessment o uture area

potentials depends largely on the underlying assumptions.

According to the results analysed in this report, the high


32/44


area potentials required or Arica can only be realised i

ecient production systems are assumed, or i a ocus is

made on biomass use or energy production (e.g. increasing

support). In summary, only by transorming the current

inecient and low-intensive agricultural management

systems into state o the art production systems using theappropriate technologies can the area potential illustrated

in most o these studies be achieved.

2.4.2 Energy crops

The most dominant parameters used to determine the

potential o energy crops are land and yield. All results are

considered with regard to, timerame; geographical coverage;

the type o potential; and the biomass eedstock (see Section

2.3.1). There are undamental dierences between thecalculated potential amounts in the reviewed studies.

For the period up to 2020, the amount o energy

crop potential shows range between 13 900 PJ/yr

(KALTSCHMITT) and 0 PJ/yr (HAKALA). HAKALA

justied this value by reasoning that Arica will continue to

have ood production problems or a signicantly growing

population in the uture. According to the authors, a decit

in ood production also indicates there is no land available

or energy crop production.

When interpreting the results, or example, KALTSCHMITT

used a general estimation that the potential area or

energy crop cultivation was equivalent to 10% o the total

agricultural area. Compared to other studies, this simple

assumption leads to the largest potential o energy crops

ound in any study. Whereas, the calculated value or

Mozambique in BATIDZIRAI (6 670 PJ/yr) can only be

achieved i there is to be a large increase in ood production

eciency that corresponds to a level ound in industrialised

countries (see Section 2.4.1).

The values presented in WICKE indicate area potentials

that exclude other competitive land uses, such as land

or hunting or livestock, due to the chosen methodology.

Excluding these areas leads to a reduction in available land

and a reduction o the calculated energy crop potential. A

dierent picture emerged in the BMVBS study, in which the

calculated values vary according to the chosen scenario. It

should be noted that the highest values are only realised

in the bioenergy scenario, which assumes an increasing

ocus on biomass use, e.g. through a cultivation o SRC

or sugarcane. I previous trends regarding agricultural

development, land use change and population growth

continue, or environmental and conservation restrictions

increase, the calculations show that there can be a decrease

in the energy crop potential (33 PJ/yr to 47 PJ/yr in 2020

and 0 PJ/yr in 2050).

The highest value o 317 000 PJ/yr or the energy

crop potential is determined by SMEETS with the key

parameter in this study being the eciency o ood

production. However, only with very high crop production

eciency and through improved agricultural technology

or geographical optimisation (see Section 2.3.3) can it be

possible to realise this potential by 2025.

Six studies estimate the potential o energy crops in 2050;

the average value ound or Arica is 76 508 PJ/yr. Two

o these studies (HAKALA, BMVBS) indicate that thereis no potential or energy crops by 2050, mostly due to

the high uture population demand or ood. In BMVBS an

energy crop potential o 2 552 PJ/yr is only achieved with a

scenario that has a ocus on energy crop cultivation.

The values or 2050 determined by HABERL (21 250 PJ/yr)

and FISCHER (54 510 PJ/yr to 68 310 PJ/yr) lie in the lower

to middle range o the calculated values. The increase

in biomass potential ound by HABERL, are primarily

explained by the assumption o a global increase o 54%

in yields and also by a 9% expansion in cropland. Although

improvements to the current productivity levels are

implied in this study, there is no massive intensication o

land management assumed.

A wide range o results or the year 2050 is shown in

HOOGWIJK (15 000 PJ/yr to 134 000 PJ/yr). It is assumed

that abandoned agricultural land, low productive land and

rest land (mainly savannah, scrub and grassland/steppe)

are available or energy crop production. The area potential

is determined with the model IMAGE (see Section 2.3.2).An area reduction is achieved by the so-called land-claim

exclusion actor. This actor indicates the percentage o

land that is not available or biomass production, e.g. claims

or nature development, urbanisation, or cattle grazing on

extensive grassland. The authors note that this actor or the

competing land-use claims is associated with a high degree o

inaccuracy. The selection o the land-claim exclusion actors

or rest land area is arbitrary, but has an important infuence

on the total estimated potential. It should also be stated that

on a global scale the potential are assumed to be available or

energy crops corresponds to 30%-40% o the total land area.


33/44


Only two studies assess the potential or 2100. HOOGWIJK,

calculates a potential or energy crops ranging rom

74 000 PJ/yr to 279 000 PJ/yr, depending on the scenario7,

as opposed to YAMAMOTO who assumes that there is

absolutely no potential rom energy crops in 2100. In the

latter study, this result is explained by a high demand orland due to the rising per capita consumption o meat and

despite an increasing availability o land (through the use

o allow and degraded land) and increasing productivity in

developing countries. This demand or meat counteracted

the growth in area, so that nally no areas are available

or energy crop cultivation. In YAMAMOTO, these material

fows are simulated using the model GLUE (see Section

2.3.2).

2.4.3 Forestry biomass

Only ve o the reviewed studies provide values or orestry

biomass potentials. The results present a very wide

range rom 0 PJ/yr to 75 000 PJ/yr (see Section 2.3.3). A

direct comparison o the published results is impractical,

because the ndings are based on dierent regions and

time periods. Yet it is possible to discuss the applied

methodology and the assumed ramework conditions.

For the Arican continent today, KALTSCHMITT derives

a potential o 5 400 PJ/yr using a calculation based on

FAO numbers or raw wood harvesting. The theoretical

increments o timber account or approximately hal o the

calculated potential.

DASAPPA, FISCHER and YAMAMOTO provide inormation

on the SSA region. Although these studies partly use the

same data set, the results are very divergent (Figure 8).

DASAPPA calculates a potential o 0 PJ/yr or the current

time, using a gure o 600 million people living in the

region without access to energy; a per capita consumptiono 0.89 m3/yr o uelwood is calculated or cooking and

heating that eaves no wood resources or other energy

requirements . FISCHER calculates that orestry production

potentials range rom 14 820 PJ/yr to 18 810 PJ/yr or the

year 2050. In this study the evaluation data or yield and

availability are collected rom the literature up to the year

1992 and then combined with FAO land use data rom 1999.

This data is computed with the IIASA-model. YAMAMOTO

uses their sel-developed model GLUE to predict biomass

potentials up to the year 2100. The calculated orestry

production potential or SSA reaches 75 000 PJ/yr (Figure

8). This calculation is based on the average development

rom 1961 to 1990 and then uses a linear projection annually

to 2010. Their basic assumptions include aorestation

and uture biomass demand. For developing countries

YAMAMOTO assumes a perect reorestation until the year

2025 and a constant biomass demand based on the 1990slevel.

The BMVBS calculations o 949 PJ/yr to 2 459 PJ/yr

biomass potential, use the countrys average annual

change to orest and plantation areas since 1990, which in

combination with a countrys specic annual net increase,

can be used to derive the sustainable extractable raw wood

potential. The calculations link a time series o production

and consumption gures, as well as the gross domestic

product and population data. The main data sources are

the FAO databases. The potentials are calculated or threescenarios with varying ramework conditions.

The highest potential can only be

achieved through the ambitious

implementation of forest plantations.

The lowest potentials can be obtained

in a scenario where 50% of the

primary forest in the tropics and 10%

of commercial forest in the temperate

zones are put under protection.

In this global approach it is impossible to account or all

relevant actors infuencing the raw wood potential orevery country and as such the ndings o the BMVBS

study rely on partial rough estimates and assumptions.

Considering the inconsistent orest types and tree stands

worldwide, net increases can also only be estimated.

Possible developments o inrastructure and technology

are not taken into account.

The studies analysed have a number o points concerning

the data quality, e.g. production and consumption

parameters, as well as the lack o data or the estimation o

the orest biomass potential. Despite the act that almost all

the studies are based on FAO data, the results dier greatly.

7 Te enai e in te t en t te enai bie b te Ipcc.


34/44


This is explained through the use o dierent timelines, as

well as urther data processing using other datasets or

models. In addition, the calculated potentials are infuenced

by the assumptions concerning several determinants, e.g.

the actors or reorestation and biomass demand set by

YAMAMOTO are simplied and do not capture the wholepicture. Another critical actor is the data itsel, which

being generated in the 1990s, is considered rom todays

perspective fawed, i.e. the productivity yield calculation.

Additionally the lack o ocial statistics and the material

use o timber are not considered suciently in the models

used. The dynamic demographic development, especially

in the very poor regions o Arica, is assumed to cause an

increasing demand or uelwood or cooking and heating.

Furthermore a rising share o material use and export o

timber can be assumed.

In light o this discussion, the orest biomass potentials can

only be considered as rough estimates or the assessment

o the respective regions. Regional analyses based on

current data are a precondition to allow sustainable use o

orest resources or energy production.

2.4.4 Residues and waste

In the analysed studies, dierent residual ractions (Table

3) are considered. Although the overall results are mostly

similar, the individual results are compared (Table 5).

Most studies have ndings or crop residues. For the current

investigated period HAKALA (2 380 PJ/yr to 2 600 PJ/yr),

BMVBS (1 161 PJ/yr) and COOPER (1 089 PJ/yr to 3 588 PJ/yr)

provide broadly similar values or the entire continent. In all

three studies the potentials are determined by using crop

yield actors, such as soil quality, crop heating value, crop

straw ratios and the corresponding crop residue ractions.For this purpose HAKALA uses literature data and their

own calculations. The BMVBS calculations are based on

a region-and crop-specic grain-to-straw ratio. Whereas,

COOPER uses actors that are based on literature data or

Asia.

To protect the soil quality only a small part o the total

potential should be used or energy generation. This part

depends on diverse actors (e.g. crop rotation, ertiliser

use) and is estimated in the studies by a (sustainable)

recovery rate. HAKALA estimates this rate to be 50% to

70% while the BMVBS study uses conservative 20% and

COOPER uses 35%.

With the help o specic heating values, the remaining

residue potential is converted into units o energy.

HAKALA and COOPER use a uniorm number o 18 GJ/t,

or all crops. BMVBS calculations utilise a slightly lower

value o 17 GJ/t. KALTSCHMITT evaluates a crop residue

potential o 900 PJ/yr, but gives no explanation on howthis is reached. Nevertheless, this value is still o the same

order o magnitude as those described in the other studies.

Despite a similar methodological approach, the respective

residue and waste energy potential results calculated or

the Sub-Saharan region dier considerably. DASAPPA

calculates a very low 135 PJ/yr or the present-day. For the

year 2050, the results range rom 2 190 PJ/yr (HABERL)

to 20 000 PJ/yr (SMEETS). The amount o crop residue

can be estimated rom the assumptions made or area,

energy crops potential and yield development (Section2.4.2). High energy crop potentials are linked to high crop

residue potentials. In particular, the high yield expectations

ound by SMEETS leads to high values or the estimated

crop residues. This relationship is also evident in FISCHER.

YAMAMOTO indicates no specic value or crop residues

or the year 2100, but estimated a proportion o 42%. As

this value is not mentioned in a global context it is taken to

only be conditionally related to Arica.

The quantity o crop residue is determined by the

parameters such as area, and crop productivity, including

the amount o raw material and the dierent recovery

rates or the crops produced. The results or Arica at

present vary only slightly. The dierences arise rom the

actors used to determine the raction o residue, the

(sustainable) recovery rates, and the varying sample site

conditions. Taking into consideration the methodology and

the relatively small deviations in the results, the potential

o crop residues would range between 1 000 PJ/yr and

3 000 PJ/yr. Further analyses with actual and updated

regional databases are necessary to speciy andcorroborate these igures, given the importance

attached to yield and its role in uture potential energy

evaluations.

In three studies the potentials o animal residues

are presented or the whole continent. The values o

KALTSCHMITT (1 200 PJ/yr) and COOPER (1 450 PJ/yr)

are very similar. In contrast, the potentials calculated

by BMVBS are very low (28 PJ/yr). The calculations are

based on FAO animal data and subsequently multiplied

by speciic energy content. COOPERs calculations

assume 223 million cattle, and each animal is assumed

to produce 50 gallons (gal) o animal residue, with an


35/44


Wooden pelletsJiri Hera/Shutterstock


36/44


energy content o 130 MJ/gal. However, the authors

stress that the results reers to the traditional use o

dried excrement and draws attention to one the biggest

challenges acing Arica, which is the ability to create

an appropriate inrastructure or large-scale excrement

use. KALTSCHMITT also assumes an energy-related use

o a dried solid uel.

The calculations include an additional actor o 50% or

the development o excrement amounts o cattle and

pigs. In the BMVBS calculations, the FAO average values

used are rom 2003-2007 (237 million cattle, 21.6 million

pigs, 1.2 billion chickens). All countries investigated by

the studies are characterised by a poor inrastructure.

Thereore, additional country-specic actors o the uture

development o excreta use are applied. These actorsare assumed to be at a very low level or animal residue

potential: or pigs between 15 to 30% and or chickens 30

to 70%. For bee a development actor o 0% is expected

and could explain the signicant dierence in the results o

COOPER and KALTSCHMITT.

The demand or milk and meat is expected to increase in

the uture with an increasing population. The production o

animal residues is expected to increase. The FAO livestock

data are not suciently detailed to make an estimate,

since the individual stocks (e.g. bee cattle, dairy cows and

calves) are not separately listed. Animal species dierence

will also have a signicant impact on the amount o the

excrements and the possible associated energy yields.

Results using the FAO database must be interpreted with

caution. Moving orward, the improved capture o animal

residue potential in Arica lies in the development o an

appropriate inrastructure.

Only the BMVBS study calculates the municipal waste

potential; 353 PJ/yr. To determine this potential, the FAO

population data and IPCC inormation on the per capita

biomass resources in year 2000 are used. The technical

uel potential is then calculated with a development actor

o 75%. I the population rises, the volume o municipal

waste will increase, especially in major cities. Inaccuracies

in the data arise rom the inormation aabout theoretical

potentials per capita and other dierences in the regional

development actors.

Data or industrial wood waste potential ranges rom 0 PJ/yr

(SMEETS) to 356 PJ/yr (DASAPPA). All authors describe

the FAO database as limited, with ragmented data.

BATIDZIRAI (2 PJ/yr) provide calculations or Mozambique

based on orest area and identied industrial wood waste

potential or sustainable logging. The analysis derives a

potential with regard to the wood processing industry that

takes into account an additional eciency actor o 55%.

BMVBS (4 PJ/yr) and SMEETS (0 PJ/yr) also use this actor,

but assume 75%. DASAPPA assumes a relatively small

actor o 30% or energy production (356 PJ/yr). However,

in general the calculations reveal a high potential and


37/44


such numbers possibly result rom the assumptions being

made, the theoretical potential and the underlying wood

yields being used. In the aorementioned studies, only the

most essential actors are considered. In many cases, they

are only partially taken into account. For example, dierent

wood types cannot be distinguished. Also the material use

can be estimated only very roughly. The results thereore

provide very approximate values

Equally, the data concerning the logging residues is o poor

quality. The BMVBS study calculates a potential o 822

PJ/yr. It is assumed that the logging

irena-dbfz_biomass potential in africa

Documents