Content uploaded by Kai Sonder
Author content
All content in this area was uploaded by Kai Sonder on Apr 06, 2016
Content may be subject to copyright.
CIMMYT Institutional Multimedia Publications Repository
http://repository.cimmyt.org/
CIMMYT Wheat
Impacts of international wheat
improvement research, 1994-2014
Lantican, M.A.
2016
Downloaded from the CIMMYT Institutional Multimedia Publications Repository
Impacts of
International Wheat
Improvement Research
1994-2014
Maximina A. Lantican | Hans-Joachim Braun
Thomas S. Payne | Ravi Singh | Kai Sonder
Michael Baum | Maarten van Ginkel | Olaf Erenstein
Impacts of
International Wheat
Improvement Research
1994-2014
Maximina A. Lantican | Hans-Joachim Braun
Thomas S. Payne | Ravi Singh | Kai Sonder
Michael Baum | Maarten van Ginkel | Olaf Erenstein
CIMMYT – the International Maize and Wheat Improvement Center (www.cimmyt.org) – is the global leader
on publicly-funded maize and wheat research and related farming systems. Headquartered near Mexico
City, CIMMYT works with hundreds of partners throughout the developing world to sustainably increase the
productivity of maize and wheat cropping systems, thus improving global food security and reducing poverty.
CIMMYT is a member of the CGIAR Consortium and leads the CGIAR Research Programs on MAIZE and WHEAT.
The Center receives support from national governments, foundations, development banks and other public and
private agencies.
International Maize and Wheat Improvement Center (CIMMYT) 2016. All rights reserved. The designations
employed in the presentation of materials in this publication do not imply the expression of any opinion
whatsoever on the part of CIMMYT or its contributory organizations concerning the legal status of any country,
territory, city, or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries. The
opinions expressed are those of the author(s), and are not necessarily those of CIMMYT or our partners. CIMMYT
encourages fair use of this material. Proper citation is requested.
Correct citation: Lantican, M.A., H.J. Braun, T.S. Payne, R.P. Singh, K. Sonder, M. Baum, M. van Ginkel, and O.
Erenstein. 2016. Impacts of International Wheat Improvement Research, 1994-2014. Mexico, D.F.: CIMMYT.
ISBN: 978-607-8263-55-4
AGROVOC descriptors:
Wheats; Varieties; Plant breeding; Seed production; Production economics;
Impact assessment; Yields; International cooperation; Regional development.
AGRIS category codes:
A50 Agricultural Research
F30 Plant Genetics and Breeding
E16 Production Economics
Additional Keywords: CIMMYT
Dewey Decimal Classification: 633.11 LAN
Printed in Mexico.
iii
Impacts of International
Wheat Improvement Research 1994- 2014
Tables
Figures
Acronyms
Acknowledgements
Executive summary
CHAPTER 1. INTRODUCTION
CHAPTER 2. DATA SOURCES AND METHODS
Sources and types of data
Analytical methods
CHAPTER 3. EVOLUTION IN BREAD WHEAT IMPROVEMENT AND
INVESTMENTS IN WHEAT IMPROVEMENT RESEARCH
Evolution in bread wheat improvement
Investments in wheat improvement research
CHAPTER 4. GLOBAL WHEAT VARIETAL RELEASES, 1994-2014
Rates of varietal releases
Wheat growth habit and production environments of varietal releases
CGIAR contribution to wheat varietal releases
Private and public sector roles in varietal releases
Breeding objectives and attributes of wheat varietal releases
CHAPTER 5. GLOBAL WHEAT VARIETAL ADOPTION
Wheat varietal adoption
CGIAR contribution to the adoption of modern varieties
Alternative measures of CGIAR contribution to wheat varieties grown
Characteristics of wheat varietal adoption
CHAPTER 6. BENEFITS OF WHEAT IMPROVEMENT RESEARCH
Wheat yields
Benefits of wheat improvement research
Discussion
CONCLUSIONS
APPENDICES
REFERENCES AND RECOMMENDED READING
iv
v
vi
vii
viii
1
2
3
5
8
9
17
18
19
20
22
24
27
28
29
30
33
34
38
39
40
42
44
48
54
CONTENTS
iv
TABLES
2.1 Countries that participated in CIMMYT/CGIAR global wheat impacts studies. 4
3.1 Priority traits in breeding spring bread wheat germplasm at CIMMYT. 11
3.2 Spring bread wheat international yield trials and screening nurseries distributed
yearly by CIMMYT. 16
3.3 Estimated research intensity by region, area, production, and value of wheat
production, 2014. 17
4.1 Average and cumulative number of wheat varieties released by region and period,
1994-2014. 19
4.2 Wheat varietal releases by moisture regime, region, and wheat type, 1994-2014. 21
4.3 Distribution of wheat area by mega-environment, 2014, with percentage
comparisons to values from studies conducted in 1997 and 1990. 22
4.4 Wheat varietal releases by mega-environment and region, 1994-2014. 23
4.5 CGIAR contribution to wheat varieties released worldwide, 1994-2014. 23
4.6 Wheat varietal releases by sector and region, 1994-2014. 25
4.7 Percentage of wheat varietal releases by target trait and region, 1994-2014. 26
4.8 Reported breeding objectives by rank of importance. 27
5.1 Adoption of improved, modern varieties in 2002 and 2014 in a subset of countries
covered in both surveys. 29
5.2 Area sown to different wheat types and variety classes in survey countries, 2014. 30
5.3 CGIAR contribution to modern wheat varieties adopted worldwide, 2014. 30
5.4 Area sown to different wheat types, classified by origin of germplasm, 2014. 32
5.5 Weighted average age of varieties grown by farmers, 1997 and 2014. 35
5.6 Attributes of 499 wheat varieties by region, 2014. 37
6.1 Global wheat yields and underlying growth rates. 39
6.2 Additional annual wheat production due to wheat improvement research based on two
attribution scenarios, 2014. 41
6.3 Benefits from global wheat improvement research (high- and low-end
estimates in parentheses). 41
6.4 Counterfactual scenarios: a world without CGIAR wheat improvement research. 43
A.1 Rates of genetic gain in bread wheat yield, developing countries. 48
A.2 Rates of genetic gain in wheat yield, developed countries. 49
v
Impacts of International
Wheat Improvement Research 1994- 2014
2.1 Distribution of global wheat production. 3
3.1 CGIAR wheat research expenditures, 2002-14. 16
3.2 Number of CGIAR wheat improvement researchers, 2002-14. 16
4.1 Rates of release of wheat varieties, normalized by wheat area, 1994-2014. 20
4.2 Key wheat production sites in the study countries. 21
4.3 Spring bread wheat releases by region and origin, 1994-2014. 23
4.4 Spring durum wheat releases by region and origin, 1994-2014. 24
4.5 Winter/facultative bread wheat releases by region and origin, 1994-2014. 24
4.6 Spring bread wheat releases by region and breeding program, 1994-2014. 25
4.7 Spring durum wheat releases by region and breeding program, 1994-2014. 26
4.8 Winter/facultative bread wheat releases by region and breeding program, 1994-2014. 26
5.1 Spring bread wheat area shares by origin of germplasm and region, 2014. 31
5.2 Spring durum wheat area shares by origin of germplasm and region, 2014. 31
5.3 Winter/facultative bread wheat area shares by origin of germplasm and region, 2014. 32
5.4 CGIAR contribution to spring bread wheat varieties grown worldwide, 2014. 33
5.5 CGIAR contribution to spring durum wheat grown worldwide, 2014. 33
5.6 CGIAR contribution to winter/facultative bread wheat varieties grown worldwide, 2014. 34
5.7 Genetic diversity in the Elite Spring Wheat Yield Trial (ESWYT), with genetic
distance measured as average Rogers distances. 37
A.1 Global wheat sites with stem rust as a production constraint. 50
A.2 Global wheat sites with leaf rust as a production constraint. 50
A.3 Global wheat sites with yellow rust as a production constraint. 51
A.4 Global wheat sites with powdery mildew as a production constraint. 51
A.5 Global wheat sites with drought and heat stresses as production constraints. 52
A.6 Global wheat sites with drought stress as a production constraint. 52
A.7 Global wheat sites with heat stress as a production constraint. 53
FIGURES
vi
Benefit-cost ratio
Borlaug Global Rust Initiative
Consultative Group on International Agricultural Research
International Maize and Wheat Improvement Center
Coefficient of parentage
CGIAR Project on Diffusion and Impacts of Improved Crop Varieties in Sub-Saharan Africa
deoxynivalenol
Durable Rust Resistance in Wheat Project
End-point royalty
CIMMYT Elite Spring Wheat Yield Trial
Food and Agriculture Organization of the United Nations
Fusarium head blight
Full-time equivalent
Genealogy management system
Global Trade Analysis Project Agro-Ecological Zone
Hectare
International Center for Agricultural Research in the Dry Areas
International Crop Information System
International Model for Policy Analysis of Agricultural Commodities and Trade
Turkey-CIMMYT-ICARDA International Winter Wheat Improvement Program
Kenyan Agriculture and Livestock Research Organization
Meters above sea level
Mega-environment
Improved, modern variety
Plant breeders’ rights
Sub-Saharan Africa
Triennium ending
International Union for the Protection of Plants
U.S. Department of Agriculture–National Agricultural Statistics Service
Weighted average age
West Asia and North Africa
CGIAR Research Program on Wheat
BCR
BGRI
CGIAR
CIMMYT
COP
DIIVA
DON
DRRW
EPR
ESWYT
FAO
FHB
FTE
GMS
GTAP-AEZ
ha
ICARDA
ICIS
IMPACT
IWWIP
KALRO
masl
ME
MV
PBR
SSA
TE
UPOV
USDA-NASS
WA
WANA
WHEAT
ACRONYMS
vii
Impacts of International
Wheat Improvement Research 1994- 2014
We owe a tremendous debt of gratitude to the
many people who have contributed to this report.
Wheat researchers from around the world provided
the basic data for this study. We are grateful for
their time and effort to provide the best information
available, especially Negusse Abraha, Suresh
Acharya, Adnan Adary, Michel Afram, Khoshgeldi
Danatarovich Agakishiev, Husnu Aktas, Ana Sofia
Almeida, Azzildeen M. Al-Shamma, Fanny Alvaro,
Hussien Al-zubi,Bakit Alpysbaevna Aynebekova,
Mathewos Ashamo, Moustafa Azab, Carlos Tomás
Bainotti, Naresh Chandra Deb Barma, Savas Belen,
Roi Ben-David, Sahar Bennani, Cresencio Calle
Cruz, Miguel Alfonso Camacho Casas, Ravish
Chatrath, Fran Clarke, M. Nazim Dincer, Chen
Dongsheng, Jose Coutinho, Fran Clark, Francisco
de Assis Franco, Srbislav Dencic, Pasquale de Vita,
Mahmud Duwayri, Mira Dzhunusova, Rola El Amil,
Fetah Elezi, Dario Fossati, Dnyandeo Gadekar,
Nihorimbere Gaspard, Nutan Raj Gautam, Seyed
Mahmoud Tabib Ghaffary, Tanja Gerjets, Lisardo
Jorge González, Bob Graybosch, Ana Gulbani, Edgar
Guzmán Arnéz, Innocent Habarurema, Sherif Thabet
Issa Hafez, Victor Manuel Hernández Muela, Julio
Huerta-Espino, Bahromiddin Husenov, Makhdoom
Hussain, Abdelraheem H.A. Hussein, Muhammad
Ibrahim, Talai Javanshir, Saffour Kaddour, Ishwar K.
Kalappanavar, Roland Kazarian, Arthur Klatt, Man
Mohan Kohli, Umran Kucukozdemir, Uri Kushnir,
Liu Jianjun, Sergey Nikolaevich Kulinkovich,
Masatomo Kurushima, László Láng, Benvindo
Macas, Nestor Machado, Peter Mavindidze, Ibrahim
Mamuya, Herbert Masole, Andrea Massi, Vladimir
Meglic, Ruben Miranda, Rose Mongi, Carlos Monar
Benavides, Moussa Mosaad, Goodarz Najafian,
Nsarellah Nasserlehaq, Zdenek Nesvadba, Peter
Njau, Oluwasina Gbenga Olabanji, Izzet Ozseven,
Irfan Ozturk, S. Venkata Sai Prasad, Pu Zongjun,
Martin Quincke, Mahbubjon Rahmatov, Teshome
Regassa, Salah Rezgui, Carlos Roberto Riede,
Mozaffar Roostaei, Victor Kuzmich Ryabchun,
Sami Sabry, Gharbi Mohamed Salah, Keremli
Saparmuradov, Nicolae Saulescu, Ruzanna Sayodan,
Ursula Schafer, Cemal Sermet, Yahya Shakhatreh,
Vladimir Petrovich Shamanin, Ali Chafiq Shehadeh,
Shi Xiaowei, Mahboob Ali Sial, Hoshiyar Singh,
N.B. Singh, Julius Siwale, Virinder Singh Sohu,
Ignacio Solis Martel, Ernesto Solís Moya, Valentina
Spanic, Vija Strazdina, Sun Daojie, Muluken Bayable
Tadege, Seyfi Taner, Mekuria Temtme, Dhruba
Bahadur Thapa, Primoz Titan, Vicki Tolmay, Richard
Trethowan, Sangay Tshewang, Riza Unsal, Patrizia
Vaccino, Ondrej Veskrna, Merja Vetelainen, Eben Von
Well, Wagoire William Wamala, Wu Xiaohua, Xiang
Cheng, Yang Fangping, Yang Jinhua, Yang Wuyun,
Yao Jinbao, Selami Yazar, Omar Zeghouane, Zhang
Pingzhi, Zhang Yelun, Zhang Yueqiang, Zhao Hong,
Zhu Zhanwang, and Zafar Ziyaev.
We are also grateful to those who forwarded the
query/questionnaires to their colleagues/other wheat
experts, particularly Carl Bulich, Hadi Bux, Luigi
Cattivelli, L. Cseuz, Ephrame Havazvidi, Marja Jalli,
Francois Koekemoer, Bona Lajos, Zacharia Malley,
Hafiz Muminjanov, Frank Ordon, Ivan Rwomushana,
and Roberto Tuberosa. Thanks also to Eric Huttner
for the useful reference material he gave for this
report.
Many CIMMYT colleagues also afforded support for
this report, especially in following up on the return of
questionnaires, data collection, and data validation.
We thank Bekele Abeyo, Beyhan Akin, Akhter
Ali, Karim Ammar, Jorge Enrique Autrique Ruiz,
Sridhar Bhavani, He Zhonghu, Muhammad Imtiaz,
Arun Kumar Joshi, M.R. Jalal-Kamali, Muratbek
Karabayev, Li Simin, Alexei Morgounov, Rajiv Kumar
Sharma, Wang Xin, Amor Yahyaoui, and Yuriy
Zelenskiy. Other CIMMYT colleagues have provided
additional wheat-related information: we are grateful
to Irene Christiansen, Cesar Mariano Cossani Rial,
Maria Laura Donnet, Susanne Dreisigacker, Maria Luz
George, Moti Jaleta, Dil Rahut, Matthew Reynolds,
Darell Sison, Graham Sim, Chloé Konig Kleinsmidt,
Petr Kosina, Jennifer Marie Nelson, Maria Emilia
Arredondo Fernandez Cano, Newman Edmundo
Montes Samayoa, Martin Rodríguez Velasquez, Maria
Teresa Rodríguez, and Ariel Saffer. We also thank
María Concepción Castro Aragón for the Spanish
translation of the questionnaire for Latin America
and Karlygash Dyussembayeva for the translation
of questionnaires from English to Russian for former
Soviet Union countries, as well as for translating
responses from Russian to English.
Likewise, other former CIMMYT colleagues have
provided help during the initial stage of this study,
in particular, Federico Carrión, Morten Lillemo and
Reynaldo Villareal.
Colleagues from ICARDA provided wheat-related
information and helped validate data; for this we
thank Ahmed Amri, Filippo Bassi, Zewdie Bishaw,
Salah Chouki, Mohamad El Mourid, Mesut Keser,
Erwin Lopez, Hassan Machlab, Ahmed Mazid,
Ram Chandra Sharma, Abdul Rahman Manan, and
Tadesse Wuletaw.
We are deeply grateful to Jesper Norgaard for
his tireless curation of wheat variety pedigrees,
providing an additional parser, and the BROWSE
application. Similarly, we thank Arlett Portugal and
Graham McLaren for their help in the initial phase of
the pedigree analysis, and Erico Panergalin for his
excellent assistance in data processing/graphics.
We are extremely grateful to Victor Kommerell and
the WHEAT CRP for funding and supporting
this study.
Finally, we greatly thank CIMMYT science writer
Mike Listman for editing this report and Bosen Zhou
and Eliot Sánchez Pineda for its design, overall
production, and printing.
– Mina Lantican
ACKNOWLEDGEMENTS
viii
This study documents for 1994-2014
the global use of improved wheat
germplasm and the economic benefits
from international collaboration in
wheat improvement research funded by
CGIAR and involving national agricultural
research systems,1 CGIAR organizations,
and advanced research institutes.
Conducted by the CGIAR Research
Program on Wheat (WHEAT), this is the
fourth in a series of global wheat impact
assessments (Byerlee and Moya 1993;
Heisey et al. 2002; Lantican et al. 2005)
initiated by the International Maize and
Wheat Improvement Center (CIMMYT).
It updates data and earlier analyses from
the most recent, previous study, covering
1988-2002 (Lantican et al. 2005).
Data were collected through
questionnaires sent to public and private
wheat breeding programs in 94 countries
that produce at least 5,000 tons of wheat
per year. Responses were received from
66 countries (a response rate of 70%)
representing about 80% of world wheat
production and from 44 developing
countries that account for 99% of the
wheat grown in the developing world.
Survey data were complemented with
information from other sources, including
inter alia online resources, published
varietal guides, figures on wheat varietal
area insured or grown, scientific journals,
technical bulletins, the US Department of
Agriculture National Agricultural Statistics
Service (USDA-NASS), Annual Wheat
Newsletter, and wheat area, production
and yield statistics from the Food and
Agriculture Organization of the United
Nations (FAO). Pedigree analysis using
the BROWSE2 application served to
determine the CGIAR contribution to
wheat improvement research. A simple
economic surplus model was used
to estimate the benefits attributable
to international wheat improvement
research.
Adoption of high-yielding improved
varieties of wheat had increased since
the previous study. A paired comparison
of 32 countries revealed an increase in
adoption from 93% in 2002 to 97% in
2014. Globally, CGIAR-related varieties
covered about 106 million (64%) of the
study countries’ 165.7 million hectares
sown in 2014. This area coverage
represented three-quarters of the world’s
wheat area (222 million hectares3) in
2014. The rest of the area not covered
is mainly in developed countries such
as France, the United Kingdom, other
EU-28 member countries, other areas
of the Russian Federation (represented
in this study by the Omsk region only)
and Australia’s wheat areas aside from
Western Australia (covered in
this study).
Output, as measured by the rate of
releases of improved wheat varieties, has
been particularly high in recent years:
2010-14 accounted for nearly a quarter
of the 4,604 improved varieties released
by public national research organizations
and private seed companies since 1994,
which may be due to the introduction of
rust-resistant varieties in recent years.
Public breeding programs were the
main source of varietal releases (63%),
followed by the private-sector (37%). In
Latin America, especially Argentina and
Brazil, private companies had a greater
role, accounting for 53% of wheat varietal
releases.
CGIAR-related varieties accounted for
63% of all releases. In South Asia – home
EXECUTIVE
SUMMARY
Impacts of International Wheat
Improvement Research, 1994-2014
DATA AND METHODS
RESULTS
1 This includes publicly-funded breeding and extension programs,
private companies, universities in developing countries, and non-
governmental and community-based organizations.
2 BROWSE is a part of the International Crop Information System
(ICIS) program that extracts the required pedigree information,
counting selng generations and identifying common ancestors of
sister lines (McLaren et al. 2007).
3 Derived from FAOSTAT January 2016.
ix
Impacts of International
Wheat Improvement Research 1994- 2014
to more than 300 million undernourished
people and whose inhabitants consume
over 100 million tons of wheat each year–
92% of the varieties released contained
CGIAR breeding contributions and half
of the spring bread wheat varieties were
direct releases of CGIAR breeding lines.
In Latin America, 70% of the spring
durum (pasta wheat) varietal releases
were CGIAR breeding lines used directly.
In Sub-Saharan Africa, direct releases of
CGIAR lines comprised 63% of the spring
durum wheat varieties and in West Asia
and North Africa, 52%. CGIAR breeding
contributions were present in 71% of
released winter/facultative bread wheat in
West Asia and North Africa.
The CGIAR share of improved wheat
area in 2014 was highest in the main
target regions of the developing world
(South Asia and Africa). The share with
contributions from CGIAR centers was
quite large in high-income countries. In
China, 28% of all wheat area was sown
to CGIAR-related germplasm in 2014.
The study confirmed that international
wheat improvement research continued
to generate very high returns. Annually
some US $30 million [2010] was being
invested by the CGIAR in international
wheat improvement research. In recent
years funding had come primarily
through bilaterally-funded research
conducted with partners worldwide by
CIMMYT and the International Center for
Agricultural Research in the Dry Areas
(ICARDA) and including since 2012
approximately US $6 million per year for
WHEAT. CGIAR organizations develop
and freely share global public goods
and depend on national partnerships to
achieve meaningful farm-level impacts,
but national co-investments are not
estimated here.
Using two attribution scenarios, the
historic average increase over base yield
and marginal yield increase by longevity,
annual benefits4 generated from global
wheat improvement efforts ranged from
US $6.7 billion to $9.4 billion [2010].
These benefits are attributable to
global wheat research that includes
the contributions of CGIAR, national
agricultural research systems, and
advanced research partners.
The benefits attributable specifically to
wheat improvement research by CGIAR
organizations ranged from US $2.2 billion
to $3.1 billion [2010] per year – levels
that confirm and exceed estimates
from earlier studies and largely reflect
expanded area under improved varieties
and a higher reference price for wheat.
The benefit-cost ratio for CGIAR wheat
improvement efforts ranged from 73:1 to
103:1 and appears dramatically to justify
the investments made. Note that these
estimates do not encompass benefits
from non-yield traits such as improved
grain quality or fodder quality, straw
strength, or shortened growth cycles, all
of which would further boost the returns.
Consistent and sustainable future
funding is critical to maintain an efficient
and effective global wheat germplasm
improvement pipeline, able to respond to
emerging threats and opportunities and
allowing farmers to satisfy the demand
for wheat for the 9 billion-plus world
population expected by 2050.
THE RETURNS ON INVESTMENTS
IN INTERNATIONAL
AGRICULTURAL RESEARCH FOR
DEVELOPMENT FOR WHEAT
4 Annual benets were estimated by applying a simple economic
surplus model, crediting wheat improvement research with the
value of the additional grain production. The physical quantities of
the additional grain production were translated into value terms
by multiplying them with a reference price of wheat. The benets
were expressed in real terms (2010 US $) to remove ination
effects.
01
INTRODUCTION
1
Impacts of International
Wheat Improvement Research 1994- 2014
Wheat is a major source of calories and
protein for consumers in developing
countries. The “Green Revolution”
improved the national food security
and welfare of the poor in developing
countries in the second half of the
20th century. However, investments in
crop breeding research have slowed
down subsequently, putting pressure
on both national and international
wheat improvement programs, and
wheat productivity increases now lag
behind population growth. Continued
investments in agricultural innovation
and productivity growth are as essential
today as in the early years of the Green
Revolution (Pingali 2012), particularly as
global cereal production must increase
by an estimated 56% between 1997
and 2050, with developing countries
accounting for 93% of cereal demand
growth by 2050 (Rosegrant and
Cline 2003).
Since 1990, CIMMYT – the principal
center for wheat research of the
Consultative Group for International
Agricultural Research (CGIAR) – has led
three global studies (Byerlee and Moya
1993; Heisey et al. 2002; Lantican et al.
2005) on the impacts of international
wheat breeding research in the
developing world. These studies
showed that:
• The adoption and diffusion of modern
wheat varieties continued in the post-
Green Revolution era.
• Improved wheat germplasm developed
by CIMMYT’s wheat breeding program
continued to be used widely by breeding
programs in developing countries.
• Public investment in international wheat
breeding research continued to produce
high rates of return.
The present study on the global impacts
of improved wheat germplasm updates
and expands the data and analyses of
the 2002 study and was commissioned
and funded by the CIMMYT-led CGIAR
Research Program on Wheat
(WHEAT; http://wheat.org).
In line with the previous efforts, this study:
• Examined the use of improved wheat
germplasm in the world.
• Documented the contribution of
national agricultural research systems,
the private sector, and the CGIAR
to international wheat improvement
research.
• Estimated the benefits generated
by international wheat improvement
research and CGIAR investments.
• Was designed to increase awareness
about the value of international wheat
improvement research.
Following this introduction, Chapter 2
describes analytical methods and the
sources and types of data used. Chapter
3 discusses the evolution in bread
wheat improvement and investments in
wheat improvement research. Chapter
4 analyzes wheat varietal releases in
the world from 1994 to 2014 by origin,
wheat type, growing environment, and
region. Chapter 5 examines the use
of improved wheat germplasm in the
world using similar categories, as well as
selected adoption characteristics such
as varietal turnover and attributes of
adopted varieties. Chapter 6 presents and
discusses the estimated research benefits
that can be attributed to international
wheat improvement efforts and
specifically, to CGIAR wheat improvement
research. Chapter 7 presents concluding
thoughts and discussion.
2
02
DATA AND METHODS
3
Impacts of International
Wheat Improvement Research 1994- 2014
From some countries where respondents provided information on varietal releases but
no data on varietal use, we used the following sources:
• CANADA. As a proxy for area sown to specific varieties, we used online data for area
insured.
• USA. We used lists of varieties and corresponding area coverage from wheat surveys
and the United States Department of Agriculture National Agricultural Statistics
Service (USDA-NASS) listing of 2014 wheat varieties grown in major wheat-producing
states (Colorado, Kansas, Montana, North Dakota, Oklahoma, South Dakota, Texas,
and Washington), as well as the following state surveys: “Idaho Wheat Commission’s
2013 Wheat Variety Survey” and the “Wheat Commission’s 2014 California Wheat
Variety Survey.”
• AUSTRALIA. We included data only from Western Australia, derived from the “2014
Wheat Variety Guide for Western Australia,” which lists varieties and percentage of
area sown for each.
Information on pedigree, year of release for several previously unknown varieties, and
attributes (in some cases) were obtained from the Journal of Plant Registrations, Crop
Science, Technical Bulletin, Annual Wheat Newsletter, and other scientific papers.
Information captured through the survey
was complemented with data and
information provided by or compiled from
these sources:
• Public agricultural research programs,
including ministries of agriculture,
research and extension institutes, and
universities.
• CIMMYT and ICARDA offices
worldwide.
• Private sector scientists
and managers.
• Diverse sources of information about
wheat varieties, including online lists,
published variety guides, and lists of
wheat varietal areas insured or grown.
• Scientific papers in journals.
• The World Wheat Book (Bonjean and
Angus 2000; Bonjean et al. 2011).
SOURCES AND
TYPES OF DATA
We conducted a global survey of wheat experts, primarily in public wheat breeding
programs, sending questionnaires to the 94 countries that produce at least 5,000 tons
of wheat each year (Figure 2.1). Sixty-six countries that together produce about 80%
of the world’s wheat responded. This is a 70% response rate and represents a greater
number of wheat-growing countries than those covered in previous such studies (Table
2.1). Of the countries from which responses came, 44 are developing countries that
collectively account for 99% of developing world wheat production, 11 belong to the
EU-28, and the remaining 11 are other industrialized countries. The study covers wheat
sown on about 166 million hectares, which represent three-quarters of the world’s
wheat area (about 222 million hectares5 in 2014).6 Production constraints cited in survey
responses for key wheat-growing locations are shown in Figures A.1–A.7.
Figure 2.1. Distribution of global wheat production.
Data (2005) and aggregation based on You et al. 2014.
5 Derived from FAOSTAT, January 2016.
6 The remaining wheat areas not covered in the study were from countries such as France, United Kingdom, Germany (only list of released
varieties received, no adoption data), remaining wheat areas of the Russian Federation (represented in the study by the Omsk region),
Australia’s other wheat areas (represented in the study by Western Australia), Spain’s remaining areas (only Andalusia covered), and other
relatively small wheat-producing countries wherein we did not receive data nor have online data available.
4
a The year is the year of the survey; the gures in brackets are the total number of study countries.
b Only Western Australia’s wheat area is covered in this study.
c Only Omsk region’s wheat area is covered in this study.
d Only Andalusia’s wheat area is covered in this study.
Algeria
Argentina
Bangladesh
Bolivia
Brazil
Burundi
Chile
South China
Colombia
Ecuador
Egypt
Ethiopia
Guatemala
India
Iran
Jordan
Kenya
Lebanon
Libya
Mexico
Morocco
Myanmar
Nepal
Nigeria
Pakistan
Paraguay
Peru
Saudi Arabia
Sudan
Syria
Tanzania
Tunisia
Turkey
Uruguay
Yemen
Zambia
Zimbabwe
Afghanistan
Algeria
Argentina
Bangladesh
Bolivia
Brazil
Chile
China
Colombia
Ecuador
Egypt
Ethiopia
Guatemala
India
Iran
Jordan
Kenya
Lebanon
Mexico
Morocco
Nepal
Nigeria
Pakistan
Paraguay
Peru
South Africa
Sudan
Syria
Tanzania
Tunisia
Turkey
Uruguay
Yemen
Zambia
Zimbabwe
Afghanistan
Argentina
Armenia
Azerbaijan
Bangladesh
Bolivia
Brazil
Chile
China
Colombia
Czech Republic
Ecuador
Egypt
Estonia
Ethiopia
Georgia
Hungary
India
Iran
Kazakhstan
Kenya
Korea DPR
Kyrgyzstan
Latvia
Lithuania
Macedonia
Mexico
Morocco
Nepal
Pakistan
Paraguay
Peru
Poland
Romania
Russia
Slovakia
South Africa
Tajikistan
Turkey
Ukraine
Zambia
Zimbabwe
Afghanistan
Albania
Algeria
Argentina
Armenia
Australiab
Azerbaijan
Bangladesh
Belarus
Bhutan
Bolivia
Brazil
Burundi
Canada
China
Croatia
Czech Republic
Ecuador
Egypt
Eritrea
Ethiopia
Finland
Georgia
Germany
Hungary
India
Iran
Iraq
Israel
Italy
Japan
Jordan
Kazakhstan
Kenya
Kyrgyzstan
Latvia
Lebanon
Mexico
Morocco
Nepal
Nigeria
Pakistan
Paraguay
Portugal
Romania
Russian Federationc
Rwanda
Serbia
Slovenia
South Africa
Spaind
Sudan
Switzerland
Syrian Republic
Tajikistan
Tanzania
Tunisia
Turkey
Turkmenistan
Uganda
Ukraine
United States
Uruguay
Uzbekistan
Zambia
Zimbabwe
Table 2.1. Countries that participated in CIMMYT/CGIAR global wheat impacts studies (the number of countries appears in brackets).a
1990 <37> 1997 <35> 2002 <42> 2014 <66>
Byerlee and Moya 1993 Heisey et al. 2002 Lantican et al. 2005 Lantican et al. 2016
5
Impacts of International
Wheat Improvement Research 1994- 2014
We examined pedigree information for
wheat varieties released since 1994 and
for cultivars grown during 2013-14 to
determine CGIAR contributions, if any.
We also performed pedigree analysis
using BROWSE, an application of the
International Crop Information System
(ICIS; McLaren et al. 2005) that extracts
the required pedigree information,
counting selfing generations and
detecting common ancestors of sister
lines (McLaren et al. 2007). BROWSE can
easily analyze the pedigrees of more than
a thousand varieties for 12 generations or
more. The database used includes ICIS
GMS v. 5.5.013 (central database) and
a local genealogy management system
(GMS) that incorporates the
varieties analyzed.
All pedigrees were curated to ensure
accuracy and correct syntax before
applying BROWSE and, where the output
was not in line with prior knowledge of
genetic contributions, we rechecked and
corrected the pedigree and re-applied
BROWSE. The output for each variety
comes in the form of a Mendelgram
showing a table of progenitors with their
type, contribution, count, and origin. This
essentially represents the coefficient of
parentage (COP) between a line and an
ancestor; that is, the probability that a
randomly chosen, unselected allele in the
target genotype comes from
the progenitor.
ANALYTICAL
METHODS
PEDIGREE ANALYSIS
VARIETAL ORIGIN CATEGORIES
VARIETAL CONTRIBUTION
Wheat varietal releases and adoption
were categorized into five sub-sets based
on the pedigree analysis:
Category 1: CIMMYT/ICARDA (CGIAR)
line. This means that the cross and
selection were made by CIMMYT or
ICARDA or involved a direct cross from
CGIAR collaboration.7 Lines in this
category may have been re-selected by
a national breeding program. In most
cases, these varieties were selected from
international yield trials and observation
nurseries distributed annually by both
centers to the global wheat breeding
community.
Category 2: CIMMYT/ICARDA (CGIAR)
parent. A national program or private
sector cross using one or more CGIAR
parents, these are usually selected from
international yield trials or observation
nurseries or received directly from
CIMMYT or ICARDA on special request.
Category 3: CIMMYT/ICARDA (CGIAR)
ancestry. A national program or private
sector cross that has CGIAR germplasm
as a grandparent or more distant
ancestor, regardless of how far back in
the pedigree tree the center germplasm
has been used.
Category 4: Non-CGIAR variety.
A variety whose pedigree contains
no known contribution from CGIAR
germplasm.
Category 5: Unknown variety. A variety
for which we had no pedigree information
or whose origin was not known.
Categories 2 and 3 include crosses made
by national programs or companies in
their home country and released there,
or varieties introduced and released in
a country other than where the original
cross was made.
Based on the preceding categories, we applied a set of measures (rules) to assign credit
for varietal contributions from specific improvement programs or crosses. The present
study applied three of the same attribution measures as two previous global wheat
impact studies (Heisey et al. 2002; Lantican et al. 2005). The measures are listed here in
decreasing order of restrictiveness and indicating to which varietal origin category the
rule relates:
• The “CGIAR cross” rule is the most restrictive; it assigns credit only to Category 1
varieties.
• The “CGIAR cross plus parent” rule assigns credit to both Category 1 and Category
2 varieties.
• The “any CGIAR ancestor” rule gives full credit to varieties belonging to any of
Categories 1-3. In the present study, we applied this rule only to provide a point of
comparison with the rules above and to pick up the extent of varieties that contain any
degree of CGIAR contribution, without weighting the contribution.
In all cases, BROWSE was used to assess the extent of the contribution of germplasm
from CIMMYT or ICARDA.
7 Includes varieties developed through Turkey-CIMMYT-ICARDA (TCI)
collaboration.
6
The number of continuous years that a
given variety has been sown is a good
gauge of the rate at which varieties are
being replaced. In the “weighted average
age” approach (WA; Brennan and Byerlee
1991), the “age” of a variety since its
release is weighted by the area sown to it.
For a given year, t, the measure would be
computed as follows:
WA = Σ (pit Nit)
Where pit is the proportion of the area
sown to variety i in year t; Nit is the
number of years (at time t) since release
of variety i.
VARIETAL REPLACEMENT
We estimated the yield growth rate using
FAO farm-level wheat yield data for the
44 developing (study) countries for 1994-
2013 (2013 being the most recent year for
which yield data were available) and then
for all wheat-producing countries in the
world for the same period, in both cases
applying a log-linear trend regression:
ln(Y) = α + βX
where α is constant; ln(Y) the natural
logarithm of yield Y; β is the growth rate
of Y; X the time (years). This is a semi-
logarithmic regression where gains are
expressed as the average percentage
change per year.
YIELD GROWTH
Crop breeding is a continuous process
wherein costs are incurred and benefits
obtained over time. Benefits in any given
year are accrued returns to investments
made over an extended period, just as
investments in any given year produce
benefits over an extended period.
Returns to investment are hence ideally
estimated in terms of dynamic flows;
SURPLUS MODEL
There are three major problems in
estimating benefits of crop breeding
programs (Morris and Heisey 2003):
1. Measuring adoption of improved
modern varieties (MVs). It is difficult
to get accurate data on area planted to
MVs. Interpretation of what constitutes
an MV can also be problematic. Here
we refer to an MV as an improved wheat
variety resulting from global wheat
improvement research (CGIAR, national
program, private sector) released since
1994.
2. Estimating the benefits from modern
varietal use. Main difficulties include:
(a) estimating farm-yield gains; (b)
identifying yield gains attributable to MV
adoption versus those from improved
crop management; and (c) drawing
counterfactual scenarios; that is, what
would have happened in the absence
of the evaluated wheat improvement
research? Other difficulties, such as
accounting for non-yield benefits,
modeling aggregate price effects, or
accounting for policy distortions, are not
covered in this study.
3. Attribution. Attributing credit to the
many wheat improvement programs
that contribute to developing an MV
presents challenges. These include
dealing with spillovers between different
research programs and disentangling
complementarities between the
performance of the research system
and that of supporting institutions and
structures; for example, the seed supply
system, the extension service, the
marketing system, and transportation and
communications infrastructure. This is an
important issue which could be pursued
in the future.
BENEFIT STREAM
that is, investments in period A lead
with a lag to benefits in period B and
need to be discounted (Byerlee and
Moya 1993). Crop improvement however
is a continuous activity that requires
annually recurring investments to
maintain an associated benefit stream. A
simplification used in this study compares
annual recurring investment to annual
incremental benefits.
7
Impacts of International
Wheat Improvement Research 1994- 2014
The CGIAR investment in wheat
improvement research goes primarily
to CIMMYT and ICARDA, the two
international centers leading CGIAR
wheat improvement research. Both
centers engage not only in plant breeding
but also in research-for-development
activities around wheat agri-food
systems, including crop and resource
management research, social science
research, training and capacity building,
networking, and knowledge management.
Congruent with Heisey et al. (2002) and
Lantican et al. (2005), to single out the
portion of the centers’ overall budget that
is spent on wheat improvement research,
we used two measures:
• Expenditures 1 is based on the
assumption that all wheat research
staff– both breeders and other
scientists– contribute to wheat
improvement research. The CGIAR
investment in wheat improvement
research is thus estimated by
multiplying the pooled center budgets
by the ratio of wheat senior staff to
the total number of senior staff in the
centers.
• In Expenditures 2, we assume that
65% of the centers’ wheat research
budgets is committed to wheat
improvement, plus a 26% associated
overhead. The current levels of CGIAR
and center investments in wheat
improvement research are discussed in
the next chapter.
The gross annual benefits generated
by international wheat improvement
research were estimated using a simple
economic surplus model, crediting
wheat improvement research with the
value of the additional grain production.
The physical quantities of additional
grain were translated into value terms by
multiplying them with a reference price of
wheat:
Bt = At yt Pt
where B = value of additional production
attributable to wheat improvement
research; A = area sown to improved
wheat varieties; y = yield gain
attributable to wheat improvement
research; P = the price of wheat grain.8
The area sown to modern wheat varieties
was estimated using data from the 2014
global wheat survey and totaled 149.1
million hectares.
The yield gain attributable to wheat
improvement research is the main factor
in the annual benefits reported, and
includes both the growth in yield potential
and the averted yield decline due to yield
maintenance. We used two attribution
scenarios:
• Historic average yield increase
over a base yield. We credited
wheat improvement research with the
observed average yield increase over
the base yield for the reference period.
• Marginal yield increase by longevity.
We credited wheat improvement
research with the observed annual
marginal yield increase at the end of
the reference period multiplied by a
“persistence” factor representing an
expert estimate of the longevity of
the marginal yield gain. We estimate
the yield gain benefits (including both
the growth in yield potential and the
COST STREAM
“maintenance” of yield against factors
such as evolving crop disease strains)
to last fully for only 3 years, and
then to decline linearly to 0 over the
subsequent 8 years. After discounting
the future yield gains at 5%9 p.a., the
persistence factor amounts to 5.6 for a
10-year longevity scenario.10
9 Discount or interest rate used in determining the present value of
future gains.
10 The persistence factor for a 10-year longevity scenario and a high-
end annual increment of 46.4 kg/ha (1.46% p.a.) would yield a
cumulative incremental production of 260 kg/ha – approximating
the actually observed average yield increase of 292 kg/ha from
TE1993 to TE2013 over the TE1993 baseline yield (derived
from FAOSTAT).
8 The additional wheat is valued using the international price of
wheat, based on the export price of the North American hard red
winter wheat (U.S. Gulf ports). In 2014, the average price was
equivalent to US $267/t in 2010 US $. However, instead of this, the
average real prices of wheat for the study period (1994-2014) US$
215/t (2010 $) was used to have a more conservative estimate of
annual benets.
8
03
EVOLUTION IN BREAD
WHEAT IMPROVEMENT AND
INVESTMENTS IN WHEAT
IMPROVEMENT RESEARCH
9
Impacts of International
Wheat Improvement Research 1994- 2014
The structure of the international wheat breeding system was outlined by Heisey et al.
(2002). Likewise, Lantican et al. (2005) described the evolution of the CIMMYT wheat
breeding program, drawing heavily on information in global wheat impacts studies. This
chapter summarizes the evolution in bread wheat improvement to enhance genetic
gains for grain yield, disease resistance, abiotic stress tolerance, and end-use and
nutritional quality, as well discussing current CGIAR investments and research intensity
in wheat improvement.
EVOLUTION IN BREAD
WHEAT IMPROVEMENT
The improved bread and durum wheat
germplasm developed at CIMMYT and
ICARDA targets most wheat-production
environments in the developing world.
The CIMMYT spring bread wheat
breeding program, initiated in Mexico in
1944 by Nobel Peace laureate Dr. Norman
E. Borlaug and continuing today, achieves
two generations of selection each year
by shuttling segregating populations and
advanced breeding lines between two
contrasting field environments in Mexico
(Braun et al. 1996). Shuttle breeding was
expanded in 2008 to include the research
station at Njoro, Kenya. Operated by
the Kenyan Agriculture and Livestock
Research Organization (KALRO) and
located in an area that experiences
frequent and intense natural infections of
wheat stem rust caused by the Ug99 race
group of Puccinia graminis, the facility
is used to screen thousands of wheat
lines each year from breeding programs
worldwide for resistance to
that pathogen.
In the last decade phenotyping in
CIMMYT’s global wheat program has
expanded significantly to address the
performance of wheat breeding lines
under heat and drought, resistance to a
range of diseases, end-use processing
traits, and nutritional quality. The lines
developed through this accelerated
breeding and testing process are
distributed and tested worldwide in yield
trials and screening nurseries, and the
fraction of materials thus selected are
used to make new crosses.
The concept of wheat mega-
environments (MEs; Rajaram et al.
1994) was introduced to better target
the crossing program in Mexico and the
deployment of appropriate germplasm
to diverse production environments
worldwide. Mega-environments
are geographical areas, though not
necessarily contiguous, where wheat
adaptation can be expected to be similar,
due either to similar climatic, disease or
crop-management constraints. In recent
years there appears to be more frequent
overlap between mega-environments
that were previously clearly delineated,
a phenomenon possibly due to climate
change effects and expected to become
more pronounced. As one result, new
wheat varieties will need to feature not
only superior yield potential but also
increased tolerance to drought and
heat stress, better disease and pest
resistance, more stable processing
traits, and better nutritional qualities.
The CIMMYT wheat breeding program is
evolving continuously to develop superior
and diverse improved germplasm that
can continue enhancing productivity and
nutrition in target areas of adaptation.
10
Increasing grain yield, yield stability, resistance/tolerance to biotic and abiotic stresses,
end-use and nutritional quality characteristics are among the most important breeding
objectives at present and will remain so in the future, considering that most of the wheat
in developing countries will be consumed by humans. In developing countries, where
population pressure continues to increase while land and water resources decline due
to urbanization and unsustainable use, the options to raise productivity are genetic
enhancement or improved crop-management.
Led by CIMMYT under WHEAT and in collaboration with other international centers
and numerous national and advanced research institutions, the International Wheat
Improvement Network (IWIN) continuously adjusts breeding objectives and schemes for
effectiveness and efficiency and to tailor required germplasm products. In one example,
as water resources decline farmers will need new wheat varieties that are both high-
yielding and that use water more efficiently for irrigated areas, or that feature improved
drought tolerance for rain-fed growing conditions. Improved varieties that tolerate heat
stress are required for all MEs. Improved germplasm distributed through international
trials and nurseries must feature a range of maturity types, as part of adaptation
in diverse environments.11 Finally, new varieties also need desirable end-use and
nutritional qualities for local and global markets.
Researchers are thus attempting to breed new varieties that combine core traits listed in
Table 3.1 and to add resistance to diseases and pests.
BREEDING OBJECTIVES
BREEDING FOR GENETIC YIELD GAINS
Various studies have shown that increases in wheat yield potential are associated
mainly with increased biomass, kernel number, kernel weight, and harvest index.
Recent CIMMYT studies show that yield potential continues to increase and that large
kernel size, an important trait in local markets of various developing countries, could
be contributing (Lopes et al. 2012, Sharma et al. 2012). Wheat germplasm recently
developed at CIMMYT has shown both increased yield potential and a kernel weight of
over 50 milligrams (mg) in northwestern Mexico, compared with about 40 mg for most
wheat germplasm developed during the 1980s-90s. New high-yielding varieties also
tend to be more tolerant to heat and drought stress (Mondal et al. 2015).
Early gains in the yield potential of semi-dwarf wheat varieties came through
the incorporation of dwarfing genes; subsequent progress can be attributed to
quantitatively-inherited additive genes. Intense breeding efforts over the last four
decades have already selected additive genes that have greatly contributed to
enhanced yield potential. Further progress is expected from selecting genes that have
much smaller effects, thus making it necessary to modify traditional breeding schemes.
Alternatively, introgression of new genetic diversity from unrelated wheat germplasm,
including wide hybridization, can bring in genes not present in commonly used
wheat germplasm. At the same time, it has become crucial to increase the number of
advanced lines in yield trials, to find new lines with superior yield potential.
11 The ver y signicant impact of IWIN is based on co-operation between national program and CGIAR wheat partners and in particular on the
principle of free germplasm exchange. Current efforts to increase annual genetic gains in farmers’ elds would be very difcult to realize if the
free exchange of wheat germplasm among IWIN partners were restricted.
11
Impacts of International
Wheat Improvement Research 1994- 2014
Table 3.1. Priority traits in breeding spring bread wheat germplasm at CIMMYT.
Wheat improvement often utilizes simple crosses, three-way (top) crosses, four-way
(double) crosses, or repeated backcrossing approaches. Various wheat breeders also
practice pedigree or bulk methods of selection. In the 1960s-70s, CIMMYT relied on
simple, top, and double crosses, followed by pedigree selection. During the early
1980s, CIMMYT breeders applied simple and three-way crosses and occasionally
single backcrosses, followed by modified bulk selection where individual plants were
harvested in the F2 generation to grow the F3 generation, with bulk selection in the F3-F5
generations. Individual plants or spikes were once again harvested in the F5 or
F6 generation.
Singh et al. (1998) showed that the choice of parents, rather than the selection scheme,
determined the performance of progeny lines. Following that study, as of the mid-1990s
a selected bulk-breeding scheme was introduced for bread wheat improvement. Under
this approach, one spike from each of the selected plants is harvested as bulk and a
sample of seed is used in growing the next generation, in all segregating generations
until F4 or F5. Individual plants or spikes are harvested in the F4 to F6 generations as
needed. This scheme allows breeders to retain a larger sample of selected plants at the
same cost and is operationally efficient. Moreover, retaining a large sample of plants in
segregating populations increases the probability of identifying rare progenies that carry
most desired genes.
Initially to incorporate multiple, additive minor genes for resistance to wheat rust,
CIMMYT breeders instituted single-backcross crossing (Singh and Huerta-Espino 2004).
It soon became apparent that this also favored selection of genotypes with higher
yield potential.
CIMMYT breeders normally make about 2,000 targeted, simple crosses each year.
Some 700 F1 progeny are then used to make top (three-way) crosses and another 700
to make single backcrosses with the higher-yielding parents. About 400 hybrid seeds
are produced per top and backcross. About 2,000 F2 populations are then grown from
F1-simple, F1-top, and BC1 crosses. These crosses are meant to combine multiple traits
from different parents and to increase the probability of finding superior progenies at the
end of the selection cycle.
In collaboration with Kansas State and Cornell universities, the CIMMYT spring
bread wheat breeding program is attempting to accelerate genetic gains for yield by
developing and testing genomic prediction models and high-throughput phenotyping.
Durable resistance to key diseases/pests for
specic mega-environments
• High and stable yield potential • Septoria leaf blight (ME2)
• Durable resistance to all three rust diseases • Spot blotch (ME5)
• Water use-efciency / drought tolerance • Tan spot (ME4)
• Heat tolerance • Fusarium head blight and mycotoxins (ME2, 4, 5)
• End-use quality • Karnal bunt (ME1)
• Enhanced Zn and Fe content for nutrition • Root rots and nematodes (ME4)
Core traits
12
Support from various projects over the last decade has allowed CIMMYT to expand
yield testing significantly and under diverse environments. As indicated above, about
9,000 new advanced lines are now yield tested during the first year in trials with two
replicates under optimally irrigated conditions in Ciudad Obregón. In the 1980s only
about 1,000 new lines were tested in replicated yield trials and a decade back this
number had increased to about 4,000 and then dropped again to about 2,500 entries for
a few years, due to significant reductions in funding.
About 1,500 lines are retained from first-year yield trials using phenotypic data for
grain yield, heading, maturity, height and resistance to rusts, including Ug99. These
lines undergo rigorous phenotyping for end-use quality at El Batán; for yellow rust
and septoria tritici blight at Toluca; for leaf rust, fusarium head blight, and tan spot
EXPANDED PHENOTYPING FOR GRAIN YIELD AND
OTHER TRAITS UNDER DIVERSE ENVIRONMENTS
SELECTION FOR DROUGHT TOLERANCE IN SEGREGATING GENERATIONS
Selection for drought tolerance is begun by growing the same set of F3 and F4
populations as unreplicated yield trial plots, together with checks, under artificially
managed drought stress in Ciudad Obregón, a desert location that receives little
rainfall during the crop season. The field plots are screened visually and for canopy
temperature depression and normalized difference vegetative index; finally, grain yield
is determined. The 40% of the populations that score highest for those measures
are grown as space-sown F4 and F5 in Toluca. Selection is carried out and spikes are
individually harvested, grain selection conducted, and about 30,000 retained for sowing
as small plots in Ciudad Obregón. About 3,000 advanced lines are finally retained
after selecting them at Ciudad Obregón, Toluca, and El Batán for agronomic and grain
characteristics and disease resistance. These lines are then included in first-year
replicated yield trials alongside sister lines from Mexico-Kenya shuttle breeding.
MEXICO–KENYA SHUTTLE BREEDING FOR UG99 RESISTANCE
Following the launch of Borlaug Global Rust Initiative (BGRI) in 2005, breeding for
resistance to the Ug99 stem rust race group began in 2006. The F3 and F4 populations
derived from the first set of targeted crosses were first grown for selection at the Njoro,
Kenya (latitude -0.341368, longitude 35.947650, 2,165 masl), research station of the
Kenya Agriculture & Livestock Research Organization (KALRO) in 2008. In addition
to allowing researchers to screen wheat lines for stem rust resistance under Njoro’s
intense infections of the Ug99 race, the location provides another selection environment
with respect to day-length and temperatures, broadening the adaptation of CIMMYT
wheat germplasm. The Durable Rust Resistance in Wheat (DRRW) Project funded
construction of an irrigation system at Njoro to facilitate growing and selecting wheat
for two generations per year. Since 2008 CIMMYT has moved F3 and F4 generation
populations from its research station at Toluca, Mexico (latitude 19.25, longitude -99.58,
2,607 masl) to Njoro, selecting them for two consecutive generations under high stem
rust pressures, and then bringing back the F5 and F6 populations to Ciudad Obregón,
Mexico (latitude 27.33, longitude -109.93, 32 masl). A selected-bulk selection scheme,
as described earlier, is being used. Individual plants are then selected and harvested
in the F5 and F6 populations that are grown at Ciudad Obregón, where selection is
conducted for plants with large, plump grains. At present researchers retain over 40,000
plants after grain selection to grow them in small plots and select for agronomic traits
and disease resistance in the F6 and F7 generations in Mexico. About 6,000 advanced
lines are finally retained after selection at Ciudad Obregón, Toluca, and El Batán (latitude
19.53, longitude -98.844481, 2,250 masl) as well as for grain characteristics to conduct
first-year replicated yield trials at Ciudad Obregón and to phenotype for stem and
yellow rust at Njoro.
13
Impacts of International
Wheat Improvement Research 1994- 2014
at El Batán; and for stem rust and yellow rust at Njoro. Simultaneous, initial seed
multiplication takes place at El Batán. All data are used to retain about 1,200 lines for
further yield testing in trials with three replicates under six environments12 at Ciudad
Obregón. These lines are phenotyped for leaf rust and Karnal bunt at Ciudad Obregón,
spot blotch at Agua Fría, Mexico (latitude 20.455, longitude -97.64111, 109 masl), and
stem rust and yellow rust at Njoro. DNA is extracted for genotyping with genotyping-by-
sequencing markers at Kansas State University to develop genomic prediction models,
supplemented with high-throughput aerial phenotyping of all trials. Seed multiplication
for international distribution is done at Mexicali, Mexico (latitude 32.29, longitude
-115.25, 8 masl), under quarantine conditions.
All data are utilized to select the 500-600 best lines for distribution through international
screening nurseries; phenotyping of the lines continues for disease resistance (to leaf
rust, yellow rust, stem rust, tan spot, and septoria nodorum blotch in the greenhouse)
and end-use quality. Molecular markers for genes of interest are also applied. Resulting
data are used to select some 280 white grained lines for another year of yield testing
at Ciudad Obregón under optimum irrigation, severe drought stress, and late-sown
heat stress, as well as for large-scale seed multiplication at Mexicali to supply three
international yield trials of white grained entries. The 135 entries selected for these
international trials have thus undergone rigorous testing for grain yield and other traits.
ENHANCING THE FREQUENCY OF LINES WITH
DURABLE RESISTANCE TO WHEAT RUSTS
The three rusts – stem (or black), leaf (or brown), and stripe (or yellow), caused
respectively by Puccinia graminis f. sp. tritici, P. triticina, and P. striiformis f. sp. tritici –
continue to reduce wheat harvests worldwide and constitute a key focus and constantly
“moving target” for wheat breeding. This is because rust fungi are highly-specialized
pathogens and display significant variation for avirulence/virulence to specific resistance
genes. They also evolve quickly through migration, mutation, and recombination,
followed by selection, whereby evolved strains able to overcome resistance genes
rapidly dominate pathogen populations. To reduce/curtail this evolution of virulence,
breeding programs now seek to identify and use combinations of plant resistance genes
that individually have small-to-intermediate effects – for example, merely slowing rather
than fully blocking rust development – but which together produce additive effects
that confer resistance levels approaching immunity. The combined effects compare
in disease-stopping value to that of a single, major, race-specific resistance gene but,
given their genetic complexity, are more difficult for the pathogen to overcome than a
single gene and are therefore more durable. Three recent review papers (Rosewarne
et al. 2013; Li et al. 2014; Yu et al. 2014) summarize current knowledge on genes
and genetic diversity for slow rusting, “adult plant” resistance (that is, expressed at
advanced plant development, rather than in the seedling stage).
Slow rusting resistance genes now form the backbone of leaf rust resistance breeding
for CIMMYT, as over 60% of wheat lines distributed by the Center possess near-
immune resistance and the importance of leaf rust has gone down in areas where
varieties derived from the lines are grown. The proportion of wheat lines with complex,
adult plant resistance to stem rust has also increased since 2012, with Mexico-
Kenya shuttle breeding. Nonetheless, another four-to-five years are needed to attain
a high frequency of wheat lines that combine the highest yield potential with near-
immunity to stem rust, as required for the eastern African highlands where wheat is
grown year round and stem rust is prevalent and virulent. Diverse sources of race-
specific resistance genes have also been incorporated and distributed in high-yielding
backgrounds through CIMMYT international nurseries and trials.
12 (1) Flat-sown with optimum irrigation; (2) sown on raised beds with optimum irrigation; (3) sown one month earlier on raised beds for heat
stress at juvenile growth stages; (4) sown on raised beds with moderate drought stress; (5) at sown with severe drought stress; and (6) sown
three months late for continuous heat stress.
14
IMPROVING RESISTANCE TO DISEASES OTHER THAN RUSTS
Table 3.1 lists the foliar diseases of importance in targeted MEs. CIMMYT began
breeding for resistance to septoria tritici blotch, caused by Mycosphaerella graminicola
(anamorph Septoria tritici), in semi-dwarf wheat in early 1970, with steady progress and
the development of several high-yielding, semi-dwarf wheats with good resistance.
Resistance is derived from diverse sources, including synthetic wheats. The high-
rainfall site of Toluca, Mexico, is used for selection. Inter-crossing parents with
different resistance sources has produced lines with high levels of resistance based
on genes with additive effects, where disease development is restricted to the lowest
two or three leaves and with low severity. Lines that show good resistance in Toluca
maintain their resistance levels in target areas such as the eastern African highlands.
Some lines derived from synthetics also show excellent resistance that appears to be
leading towards immunity to the disease, and offer new genetic diversity of resistance
originating from durum wheat and Triticum tauschii.
First crosses to incorporate spot blotch (caused by Bipolaris sorokiniana) resistance
into CIMMYT wheats were made about 25 years ago. Testing for resistance is currently
conducted at Agua Fría, a hot-spot for the disease. Use of diverse sources of mostly
intermediate levels of resistance has enabled the development of early-maturing lines
targeted for the Eastern Gangetic Plains of South Asia and which feature high-to-
adequate resistance levels. Sb1, the first designated gene for resistance, turns out to be
the pleiotropic, multi-pathogen, partial rust and mildew resistance gene Lr34/Yr18/Sr57/
Pm38 (Lillemo et al. 2013). The gene confers moderate resistance that is sufficient to
prevent losses in areas where disease pressure is normally not high.
Tan spot (caused by Drechslera tritici-repentis) is on the increase in areas where wheat
stubble is retained through successive crop cycles, as part of conservation agriculture
practices common in rainfed areas of South America and Central Asia, and where
rotation options are limited and a single crop is grown each year. Tan spot phenotyping
routinely takes place in the greenhouse and field at El Batán. Good-to-moderate
resistance is common in newer wheats.
Wheat lines are also screened for septoria nodorum blotch (caused by
Parastagonospora nodorum, previously Phaeosphaeria nodorum, synonyms
Stagonospora nodorum; Septoria nodorum) in seedlings in the greenhouse.
Several species of Fusarium cause fusarium head blight (FHB) or scab, a chief
production constraint where humid and semi-humid conditions coincide with wheat
flowering, such as in the Yangtze River basin of China. Disease outbreaks leading to
epidemics are now more frequent in countries where residues are kept on the soil for
First detected in Mexico in 2002, a new, aggressive, and heat-tolerant yellow rust race
group has been causing serious problems and become the predominant race in various
countries (Ali et al. 2014). As in other countries, in Mexico also the original race has
evolved several times and overcome resistance conferred by at least four race-specific
resistance genes. In some cooler areas where wheat remains longer in its vegetative
phase, this race group is able to establish early, multiply sufficiently, and damage the
foliage before the stem elongates and when the slow rusting, minor-gene-based adult
plant resistance becomes functional. Despite the challenges to breeding that this
presents, the pathogen’s presence in Mexico has facilitated selection for resistance and
its phenotyping at other selected field sites, such as Ludhiana, India, and Njoro, Kenya.
The genetic basis of resistance in germplasm with high levels of resistance at all sites
is complex. Mapping studies so far indicate that it often involves combinations of slow
rusting minor genes with moderately effective, race-specific genes that are often difficult
to phenotype in seedlings in the greenhouse (Basnet et al. 2014; Lan et al. 2014),
but field selection has been effective in building such resistance gene combinations,
which are likely to be a better solution than using only large-effect, race-specific
resistance genes. CIMMYT breeders are using field response and seedling reaction
data, combined with molecular markers where available, to select resistant lines for
international distribution.
15
Impacts of International
Wheat Improvement Research 1994- 2014
conservation agriculture (Argentina, Brazil, and Uruguay, for example), and in areas
where maize, which FHB also infects, is on the rise in cropping systems (China and the
eastern African highlands). The fungus not only cuts crop productivity but also produces
mycotoxins, such as deoxynivalenol (DON), that accumulate in the grain and render it
unsafe for humans or livestock to eat.
Resistance to FHB is under quantitative genetic control but a moderate-effect gene
from the Chinese cultivar ‘Sumai 3’ on the short arm of chromosome 3B, known as
Fhb1, has shown the largest and most consistent effects in reducing disease severity
and mycotoxin accumulation (Anderson et al. 2001). The Chinese varieties and their
derivatives remain the best resistance sources available and are being combined with
others. Progress in FHB resistance breeding at CIMMYT has been hindered by the
widespread use of the stem rust resistance gene Sr2, also located on the short arm
of 3B but in repulsion to Fhb1. However, new Sr2 + Fhb1 recombinants obtained from
CSIRO, Australia, are being incorporated in high-yielding wheat lines already possessing
moderate FHB resistance.
BREEDING FOR INDUSTRIAL AND NUTRITIONAL
QUALITY IN HIGH-YIELDING WHEAT
Wheat figures in a broad range of foods and provides essential nutrients. Bread wheat
is generally milled into flour (both refined and whole meal) and made into leavened
breads, flat breads, biscuits, and noodles. Industrial millers require wheat grain of very
specific characteristics. The genetics of wheat industrial quality is well understood and
our understanding continually increases as more alleles and their effects are discovered.
Some attributes, such as protein content and alpha-amylase activity, are influenced by
environmental factors. Protein content tends to be higher when the plant is under stress
and lower under well-watered or N-limiting conditions. Protein content also affects
other aspects of quality, such as dough strength, dough mixing time, and loaf volume.
The largest fraction of total protein is gluten, made up, in turn, of glutenins and gliadins.
Gluten influences the viscoelastic properties of wheat flour and largely determines how
a particular variety is used. While a relatively small portion of total variation in protein
content across years and locations is genetic, the quality of protein is controlled by
known high and low molecular weight glutenins and gliadins. Genome loci that control
both high and low molecular weight glutenins can be determined using ID SDS-PAGE
and the information used to breed higher-quality cultivars. The CIMMYT wheat quality
lab applies both modern and traditional methods to determine end-use quality in wheat
lines. The frequency of lines with poor quality un-extensible gluten has been reduced
to below 20% in international trials and nurseries – a very significant change from a
decade back.
Given wheat’s widespread use as food by low-income consumers, since around 2005
breeders have been working to biofortify the crop, identifying and selecting for higher
grain concentrations of key micronutrients, particularly iron (Fe) and zinc (Zn). There is
significant variation for those traits in certain un-adapted landraces and wheat relatives,
such as spelt wheat, diploid Aegilops tauschii, and some wild tetraploids (Monasterio
and Graham 2000; Cakmak et al. 2002). The work has moved forward under the CGIAR
program HarvestPlus. New rapid, cost-effective, non-destructive methods to determine
Zn and Fe grain levels, such as the XRF machine, allow phenotyping of large numbers
of lines. To facilitate selection for grain Zn, ZnSO4 was applied to research plots to
reduce soil variation for this element. As a result of targeted crossing, maintaining large
population sizes, and phenotyping advanced lines, breeders have been able to develop
and share high-yielding lines with significantly enhanced grain levels of Zn and Fe.
The target region for this work is South Asia, where partners have being growing the
HarvestPlus Yield Trial and HarvestPlus Screening Nursery for five years. Several high-
Zn and -Fe lines identified in Mexico produced grain with good concentrations of these
elements at multiple sites in India and Pakistan, indicating high heritability for the trait
(Velu and Singh 2012) and increased grain Zn correlates with increased Fe. A high-Zn
line from this work has been released as the variety ‘Zinc Shakti’ in India and ‘Zincol
2015’ in Pakistan; both feature about 40% higher grain Zn than other commercial
varieties while providing comparable yields.
16
SPRING BREAD WHEAT
INTERNATIONAL YIELD TRIALS
AND NURSERIES UNDER
ANNUAL DISTRIBUTION
Table 3.2 lists the spring bread wheat
international yield trials and screening
nurseries distributed by CIMMYT each
year at no cost to those who request
them. They can be used by national
partners as a source of direct releases or
for crossing programs. The targeted yield
trials are designed for partners rapidly
to identify new, high-yielding lines for
promotion to variety registration trials.
About 150 sets of international yield
trials and screening nurseries are being
distributed each year, an increase of 40 to
50% during the last decade. Partners are
asked to return data on yield, agronomic
performance, and disease resistance; the
response rate for this is about 60%. The
data are collated and made publically
available by CIMMYT via its web page;
they also help CIMMYT scientists to
identify the best-adapted parents for
crossing programs.
Costs incurred in the aforementioned
wheat improvement research activities
(that is, breeding for genetic yield
gains, selection for drought tolerance
in segregating generations, etc.) are
included in the investments discussed in
the following section.
0
10
20
30
40
Expenditure (2010 US$ millions)
2002
2004
2006
2008
2010
2012
2014
Expenditure 1
Expenditure 2
Figure 3.1. CGIAR wheat research expenditures, 2002-14.
Figure 3.2. Number of CGIAR wheat improvement researchers, 2002-14.
Number of reseachers
2002
2004
2006
2008
2010
2012
2014
Total staff
Senior staff
Post-doctoral staff
0
10
20
30
40
50
60
70
Trial/nursery Abbreviation Number of entries Target environment Grain color
Yield trials (replicated)
Elite Spring Wheat Yield Trial ESWYT 50 ME1, ME2, ME5 White
Semi-Arid Wheat Yield Trial SAWYT 50 ME4 White
High Rainfall Wheat Yield Trial HR WYT 50 ME2, ME4 Red
Heat Tolerance Wheat Yield Trial HTWYT 50 ME1, ME4, ME5 White
HarvestPlus Yield Trial HPYT 50 ME1, ME5 White
Screening nurseries
International Bread Wheat Screening Nursery IBWSN 250-300 ME1, ME2, ME5 White
Semi-Arid Wheat Screening Nursery SAWSN 200-250 ME4 White
High Rainfall Wheat Screening Nursery HRWSN 100-150 ME2, ME4 Red
HarvestPlus Advanced Nursery HPAN 100 ME1, ME5 White
Disease-based nurseries
Stem Rust Resistance Screening Nursery SRRSN 100-150 All MEs White/Red
International Septoria Observation Nursery ISEPTON 100-150 ME2, ME4 White/Red
Leaf Blight Resistance Screening Nursery LBRSN 100-150 ME4, ME5 White/Red
Fusarium Head Blight Screening Nursery FHBSN 50-100 ME2, ME4 White/Red
Karnal Bunt Resistance Screening Nursery KBRSN 50-100 ME1 White/Red
Table 3.2. Spring bread wheat international yield trials and screening nurseries distributed yearly by CIMMYT.
17
Impacts of International
Wheat Improvement Research 1994- 2014
Area Production Value of production
FTE scientists per FTE scientists per FTE scientists per US $100
Country/region million hectares million tons of wheat million of wheat
China 33.0 7.0 2.5
South Asia
1
40.5 15.6 5.8
Sub-Saharan Africa
2
134.0 52.9 19.8
West Asia and North Africa 99.0 56.3 21.0
Latin America 29.2 13.8 5.2
Former Soviet Union countries 55.9 23.8 8.9
EU and other high-income countries
6
84.2 25.5 10.0
INVESTMENTS
IN WHEAT
IMPROVEMENT
RESEARCH
CGIAR wheat research investments (Figure 3.1) are considered for the period 2002-
14, rather than the entire study period (1994-2014), given that the last CIMMYT global
wheat impacts study (reported in Lantican et al. 2005) was conducted in 2002 and
covered the earlier years of CIMMYT investments in wheat genetic improvement.
ICARDA’s wheat program was launched in 2004; it is assumed that, prior to 2004,
ICARDA incurred about US $1.2 million [2010] in operational costs for wheat research
collaborations with CIMMYT in West Asia and North Africa (WANA).
Expenditures 1 was computed based on the number of the CGIAR wheat program
staff relative to total staff, so any change in CGIAR staff numbers will raise or lower
Expenditures 1. Given that Expenditures 2 is based on the assumed percentage
of the centers’ budgets allocated to wheat improvement (see “Analytical Methods”
section, Chapter 2), it is considered a more accurate measure of CGIAR investments
in wheat improvement research. CGIAR invests an average of about US $30 million
[2010] per year in wheat improvement research – a small portion of the total investment
in international wheat improvement research.13 Included in the CGIAR investment are
about US $6 million [2010] annually for WHEAT during 2012-14. In spite of these added
funds, by 2014 the total CGIAR investment on wheat improvement had slightly declined
for both measures.
Scientific staff account for a major share of CGIAR investments during 2002-14. Total
staff includes senior staff and post-doctoral fellows (Figure 3.2). The number of CGIAR
scientists involved in wheat improvement research has fluctuated between 35 and 65
p.a., with a low in 2007.
National investments in wheat improvement research are ideally estimated by examining
research expenditure data, but complete and accurate data of this type are not
available. The number of full-time equivalent (FTE) scientists provides a proxy, but
can result in over- or underestimations of research investments, given the difficulties
of adequately accounting for all personnel involved in wheat improvement research
or their activities. To facilitate comparisons across countries and regions, we present
the “research intensity” of wheat improvement research in terms of the ratios of FTE
scientists to wheat area, production, and value of production (Table 3.3). As expected,
regions or countries characterized by smaller wheat areas and values of production
have higher estimated research intensities than those with larger areas, production, and
values of production. Some small wheat-producing countries were excluded from the
estimation to avoid inflating the averages. For Sub-Saharan Africa, we excluded Eritrea,
Burundi, Uganda, and Zimbabwe; for Latin America, Ecuador; for West Asia/North
Africa, Jordan; for South Asia, Bhutan; and for EU and other high-income
countries, Slovenia.
Table 3.3. Estimated research intensity by region, area, production,
and value of wheat production, 2014.
13 Heisey (2002) estimated that in the 1990s wheat breeding research expenditures across developing countries ranged from US $110 to US $170
million (1996 US $) per year.
1 Includes Bangladesh, India, Nepal and Pakistan.
2 Includes Ethiopia, Kenya, Rwanda, South Africa, Sudan, and Zambia.
6 Includes Albania, Croatia, Czech Republic, Hungary, Japan, Italy, Latvia, Portugal, Romania, Serbia, Spain (Andalusia), and Switzerland.
18
04
GLOBAL WHEAT VARIETAL
RELEASES, 1994-2014
19
Impacts of International
Wheat Improvement Research 1994- 2014
This chapter describes global wheat varietal releases for 1994-2014 and presents
several related indicators: rates of varietal releases, associated wheat growth habits and
production environments, the CGIAR contribution, public and private sectors’ roles, and
varietal attributes.
RATES OF WHEAT
VARIETAL RELEASES
A total of 4,604 improved wheat varieties
were released by public national
research organizations and private seed
companies between 1994 and 2014. The
number of releases per year averaged
219, ranging from 208 (1994-99) to 231
(2005-09) (Table 4.1) and with high-
income countries accounting for nearly
half, likely reflecting a greater resource
allocation to wheat genetic research and
the increased participation of the
private sector.14
Because wheat area varies greatly by
country and region, varietal releases
per million hectares of wheat serves
as a useful indicator for comparison.
More varieties per unit area of wheat
were released in Latin America and
Sub-Saharan Africa than in the rest of
the developing world (Figure 4.1) and
there was higher variability in the rates of
release for these two regions, possibly
associated with their smaller wheat areas,
greater diversity in mega-environments
(MEs), faster evolution in wheat disease
complexes, and greater involvement of
the private sector15 in wheat improvement
(Heisey et al. 2002; Lantican et al. 2005).
In contrast, varietal release rates for large
wheat producers like China and India are
lower, reflecting their larger wheat areas
and the economies of scale at relatively
modest national investment levels.
Annual average Cumulative
1994-99 2000-04 2005-09 2010-14 1994-2014 1994-2014
China 13 14 10 6 11 226
EU and high-income countries 96 105 118 102 105 2,205
Australia 4 10 10 5 7 152
Germany 12 15 12 12 13 269
Canada 8 8 9 15 10 207
United States of America 19 19 31 17 21 447
Former Soviet Union countries 11 17 17 17 15 318
Latin America 25 27 32 36 30 630
South Asia 16 12 14 19 15 320
Sub-Saharan Africa 13 15 17 10 14 291
West Asia and North Africa 34 28 22 32 29 614
World 208 218 231 222 219 4,604
Source: 2014 survey.
14 Private-sector varieties dominate Australia and Germany, while
for Canada and the USA, varieties released are largely from the
public-sector.
15 In South Africa and Zimbabwe for Sub-Saharan Africa.
Table 4.1. Average and cumulative number of wheat varieties released by region and period, 1994-2014.
20
WHEAT GROWTH HABIT AND PRODUCTION
ENVIRONMENTS OF VARIETAL RELEASES
Figure 4.2 shows key wheat production
sites listed by respondents to the global
wheat survey. Wheat releases (Table 4.1)
are normally targeted to specific moisture
regimes (Table 4.2) and MEs (Tables 4.3-
4.4) defined by type of wheat, biotic and
abiotic stresses, predominant cropping
systems, and consumer preferences.16
Some varieties are recommended for
more than one moisture regime, with
56% of global releases being suitable for
irrigated cropping, 48% for high-rainfall
settings, and 44% for dry rain-fed areas.
The most commonly targeted water
regimes included dry rain-fed (33%),
a combination of irrigated and high-
rainfall (31%), and irrigated (16%), with
significant regional variations. Fifty-five
percent of varietal releases in South
Asia are recommended only for irrigated
areas, 19% specifically for dry rain-
fed areas and 11% for both these two
moisture regimes. Sixty percent of varietal
releases in former Soviet Union countries
are recommended only for dry rain-fed
zones of high-latitude areas. Similarly,
30% of varietal releases in the EU and
high-income countries target dry, rain-
fed, high-latitude areas, whereas 51%
are recommended for both irrigated and
high-rainfall areas.
Wheat releases include both bread
and durum and their spring and winter/
facultative variants, with variations by
moisture regime (Table 4.2). Nearly half
of spring wheat releases (both bread
and durum) are for use in dry rain-fed
cropping zones, with the rest targeted
for irrigated and high-rainfall production;
25% of spring bread wheat releases and
19% of spring durum wheat releases are
recommended for irrigated areas alone.
In contrast, nearly two-thirds (64%) of
winter/facultative bread wheat releases
are meant for use in either irrigated or
high-rainfall areas.
Figure 4.1. Rates of release of wheat varieties, normalized by wheat area, 1994-2014.
China
EU and other high income countries
Former Soviet Union countries
Latin America
South Asia
Sub-Saharan Africa
West Asia and North Africa
World
0
10
20
30
40
50
60
Number of varieties per million hectares per year,
ve-year moving average
2002
2004
2006
2008
2010
2012
2000
1998
1996
16 Rajaram et al. 1994; Braun et al. 1996; Lantican et al. 2005.
21
Impacts of International
Wheat Improvement Research 1994- 2014
Figure 4.2. Key wheat production sites in the study countries.
Table 4.3 shows the distribution of wheat
area by ME in 2014, with percentage
comparisons to 1997 and 1990. Irrigated
spring bread wheat (ME1) has dominated
world wheat area in all periods, whereas
winter, facultative, and high-latitude
spring wheat areas pertain mostly to the
EU and high-income countries and former
Soviet Union nations. Durum wheat
cropping in residual moisture drylands
(ME4C) has significantly declined, while
irrigated spring durum wheat area has
increased four-fold, possibly driven by the
higher world price for this wheat type.
The greatest number of varietal releases
was targeted to ME11 (30%), followed
by ME4 (18%) and ME1 (15%), with
less than 10% of releases for any of
the other MEs and significant regional
variation. Two-thirds of South Asia’s
varietal releases and nearly a third of
Sub-Saharan Africa’s releases were
targeted to irrigated ME1 (Table 4.4).
Latin America’s varietal releases were
suited primarily for the low rainfall ME4
(41%), as were more than a quarter of
varietal releases in both South Asia (28%)
and West Asia and North Africa (WANA)
(26%). Eighteen percent of WANA’s
varietal releases were also meant for
dry winter areas (ME12). More than half
(58%) of varietal releases in the EU and
high-income countries targeted ME11,
followed by the dry, high latitude areas
(ME6, 14%) and ME4 (11%). Nearly
a quarter (24%) of varietal releases in
former Soviet Union countries targeted
dry winter areas (ME12), while 21% were
suited for high-latitude ME6 areas.
Table 4.2. Wheat varietal releases (%) by moisture regime, region, and wheat type, 1994-2014.
High- Irrigated Irrigated High-rainfall All three
Region/ rainfall (well- Dry and high- and dry and dry moisture
wheat type Irrigated watered) rain-fed rainfall rain-fed rain-fed regimes
China 36 16 13 30 3 2 0
EU and high-income countries 3 4 30 51 1 0 10
Former Soviet Union countries 24 1 60 7 9 0 0
Latin America 13 25 37 15 2 7 1
South Asia 55 0 19 8 11 0 7
Sub-Saharan Africa 35 32 27 0 1 4 0
West Asia and North Africa 24 11 45 7 9 4 0
World 16 9 33 31 3 2 6
Spring bread wheat 25 16 47 3 4 3 1
Spring durum wheat 19 22 48 1 3 6 1
Winter/facultative bread wheat 7 0 16 64 2 0 11
Winter/facultative durum wheat 18 0 56 11 4 0 12
All wheat 16 9 33 31 3 2 6
Source: 2014 survey.
22
CGIAR CONTRIBUTION TO
WHEAT VARIETAL RELEASES
Table 4.5 summarizes CGIAR
contributions to global varietal releases.17
Overall, 63% were CGIAR-related, with
the highest contribution in South Asia
(92%). Sub-Saharan Africa ranked
second (73%) and Latin America (72%)
third. In China, half of the varieties
released were CGIAR-related, as was the
case in the EU and high-income countries
whereas, in the former Soviet Union,
CGIAR contributions figured in nearly half
(48%). Wheat research by CGIAR targets
the developing world, but significant
spill-overs from the work benefit wheat
farmers and consumers elsewhere.
CGIAR contribution by wheat type.
Direct releases of CGIAR lines dominated
the spring bread wheat varietal releases
in South Asia (50%), Sub-Saharan Africa
(54%), and WANA (47%) (Figure 4.3). In
Latin America more than 70% of wheat
varietal releases were CGIAR-related,
although the direct use of CGIAR lines as
spring bread wheat releases had declined
from levels documented in previous
impact studies, particularly in Argentina
and Brazil, due to the increasing
participation of private companies in
wheat seed markets and the presence of
strong national research programs that
incorporate CGIAR germplasm in varietal
development research.
Table 4.3. Distribution of wheat area by mega-environment (ME), 2014, with
percentage comparisons to values from studies conducted in 1997 and 1990.
ME Bread Durum Bread Durum Bread Durum Bread Durum
Spring
1 47.2 3.9 29.6 2.5 36.3 0.6 32.3 0.4
2 5.9 1.4 3.7 0.9 6.7 2.0 7.6 2.4
3 1.3 0.0 0.8 0.0 1.4 0.0 1.7 0.0
4A 9.0 4.7 5.6 2.9 5.6 3.8 5.5 4.8
4B 1.6 0.1 1.0 0.1 3.0 0.1 3.2 0.0
4C 2.9 0.0 1.8 0.0 6.4 9.1 4.4 1.5
5 2.1 0.1 1.3 0.0 3.6 0.0 7.1 0.0
6 21.0 2.1 13.1 1.3 4.6 0.0 4.9 0.0
Subtotals
b
90.8 12.3 56.9 7.7 67.7 6.5 66.8 9.1
Facultative
7 17.6 0.0 11.0 0.0 9.4 0.0 5.6 0.0
8 4.1 0.0 2.6 0.0 0.2 0.0
9 2.2 0.2 1.4 0.1 3.2 0.0 4.5 1.2
Subtotals 23.9 0.2 15.0 0.1 12.8 0.0 10.1 1.4
Winter
10 4.4 0.0 2.8 0.0 2.9 0.0 6.6 0.2
11 19.2 0.0 12.0 0.0 3.4 0.1
12 8.0 0.9 5.0 0.6 5.4 1.0 6.0 1.2
Subtotals 31.6 0.9 19.8 0.6 11.7 1.1 12.6 1.4
Totals 146.2 13.4 91.6 8.4 92.3 7.7 89.5 10.5
Area, 2014a Percentage, 2014 Percentage, 1997 Percentage, 1990
(million ha) (Heisey et al.2002) (Byerlee and Moya 1993)
17 Following the CGIAR’s Project on Diffusion and Impacts of
Improved Crop Varieties in Sub-Saharan Africa (DIIVA) which
assessed the contributions of CGIAR centers to varieties of various
crops in Sub-Saharan Africa (see Walker et al. 2014)
a Excludes area grown to wheat varieties with unknown wheat type and ME.
b Figures may not add up exactly to subtotal and total amounts shown, due to rounding.
Source: 2014 survey.
23
Impacts of International
Wheat Improvement Research 1994- 2014
Number of CGIAR-related Share (%) of CGIAR-related varieties
Region wheat varieties released to all wheat varietal releases
China 121 54
EU and high-income countries 1,225 56
Former Soviet Union countries 154 48
Latin America 455 72
South Asia 293 92
Sub-Saharan Africa 211 73
West Asia and North Africa 434 71
World 2,893 63
Source: 2014 survey.
Figure 4.3. Spring bread wheat releases by region and origin, 1994-2014.
ME1 ME2 ME3 ME4 ME5 ME6 ME7 ME8 ME9 ME10 ME11 ME12
China 13 17 1 11 2 4 23 13 0 4 12 1
EU and high-income countries 4 4 0 11 0 14 0 4 0 0 58 5
Former Soviet Union countries 3 1 0 13 0 21 14 0 3 14 7 24
Latin America 14 21 7 41 1 0 0 14 2 0 0 0
South Asia 67 0 0 28 5 0 0 0 0 0 0 0
Sub-Saharan Africa 31 33 0 17 6 0 0 0 8 0 0 5
West Asia and North Africa 27 15 0 26 2 0 3 3 2 2 1 18
World 15 9 1 18 1 9 3 5 1 2 30 7
Table 4.4. Wheat varietal releases (%) by mega-environment (ME) and region, 1994-2014.
Table 4.5. CGIAR contribution to wheat varieties released worldwide, 1994-2014. For durum wheat, the share of direct
releases of CGIAR lines remained around
half in South Asia and WANA, but was
substantially greater in Latin America
(70%) and Sub-Saharan Africa (63%)
(Figure 4.4). It is remarkable that 77%
of the durum wheat varieties released
in the EU and high-income countries
were CGIAR-related, with a particularly
substantial contribution of CGIAR
ancestry. Overall, more than 70% of
the world’s spring durum wheat varietal
releases between 1994 and 2014 are
CGIAR-related – on aggregate similar to
bread wheat, but with a higher share of
direct releases.
For winter/facultative bread wheat, WANA
claimed the highest share of CGIAR-
related varietal releases (71%), 36% of
which were direct releases (Figure 4.5)
and largely a result of three decades
of strong Turkey-CIMMYT-ICARDA
collaboration and an earlier CIMMYT
partnership with Oregon State University;
both of significant benefit for Afghanistan,
Iran, and Turkey. Direct releases were
second highest in former Soviet Union
countries, probably as a result of both
CIMMYT and ICARDA having strong local
presences there in the past 20 years.
Sub-Saharan Africa was represented by
South Africa’s facultative bread wheat
varietal releases, developed mostly by
private companies.
South AsiaEU and other
high income
countries
Former Soviet
Union
countries
Latin
America
Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
Unknown varieties Non-CGIAR CGIAR ancestry CGIAR parent CGIAR line
China
24
Figure 4.5. Winter/facultative bread wheat releases by region and origin, 1994-2014.
China EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
Unknown varieties Non-CGIAR CGIAR ancestry CGIAR parent CGIAR+TCI line
Figure 4.4. Spring durum wheat releases by region and origin, 1994-2014.
South AsiaEU and other
high income
countries
Former Soviet
Union
countries
Latin
America
Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
Unknown varieties Non-CGIAR CGIAR ancestry CGIAR parent CGIAR line
PRIVATE AND PUBLIC SECTOR ROLES
IN WHEAT VARIETAL RELEASES
The role of private companies in wheat varietal development and seed marketing has
grown in many developed countries in recent decades, partly as a result of reduced
public investment in agricultural research and of a more attractive climate for investment
in the private sector. There is a similar trend in emerging economies, with South Africa
leading and greater participation in Argentina and Brazil.
In aggregate over 1994-2004, the public sector still dominated the world’s varietal
releases (63%), with the private-sector accounting for 37% of the releases in high-
income countries and half of those in Latin America (Table 4.6). Argentina was among
25
Impacts of International
Wheat Improvement Research 1994- 2014
Public- Private-
Region sector sector
China 92 8
EU and high-income countries 50 50
Former Soviet Union countries 97 3
Latin America 47 53
South Asia 99 1
Sub-Saharan Africa 68 32
West Asia and North Africa 77 23
World 63 37
Source: 2014 Global Wheat Impacts survey.
Figure 4.6. Spring bread wheat releases by region and breeding program, 1994-2014.
Private-public sector roles by wheat
type. Both for bread and durum wheat,
the public sector provided the bulk
of the spring wheat varietal releases
globally and by region (Figures 4.6 and
4.7). Among developing country regions,
Latin America was the exception for
spring bread wheat, in that the public
and private sectors accounted for equal
shares of varietal releases (Figure 4.6).
In high-income countries, 60% of spring
durum wheat releases came from the
private sector (Figure 4.7). For winter/
facultative bread wheat, varietal releases
worldwide came in nearly equal portions
from public and private sources, but
the aggregate figure masks substantial
regional differences, with the private
sector dominating winter/facultative
bread wheat releases in Latin America,
Sub-Saharan Africa (represented by
South Africa) and high-income countries,
whereas public sources provided most
releases elsewhere (Figure 4.8).
Table 4.6. Wheat varietal releases (%) by
region and breeding program, 1994-2014.
Private Sector Public Sector
China EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
South Asia Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
the first countries in the world to establish some form of Plant Breeders’ Rights (PBR)
(Pray 1992). As discussed in Heisey et al. (2002), Argentina was also among the first
of the developing countries to become a member of the International Union for the
Protection of Plants (UPOV) and varieties developed by the private sector in this
country are also grown in Brazil and Uruguay. Likewise, some Brazilian varieties are
sown in Argentina. Sub-Saharan Africa has the third-largest private sector share in
wheat varietal releases (32%), primarily the products of private seed companies based
in South Africa and Zimbabwe.
The public sector accounted for most wheat varietal releases across developing
country regions (Table 4.6). The high figure for the share (97%) of public varietal
releases in former Soviet Union countries must be interpreted with caution because, for
the Russian Federation, we received data only from the Omsk region.
Public- and private-sector roles were split for high-income countries (Table 4.6).
Australia applies the End Point Royalty (EPR),18 a value capture system used by plant
breeding companies to generate returns on their investment.
18 This is a risk-sharing mechanism, wherein a crop grower pays a
royalty based on production instead of a set fee for a particular
variety (http://varietycentral.com.au/end-point-royalties).
26
Figure 4.8. Winter/facultative bread wheat releases by region and breeding program,
1994-2014.
Private sector Public sector
China EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
Table 4.7. Percentage of wheat varietal releases by target trait and region, 1994-2014.
Figure 4.7. Spring durum wheat releases by region and breeding program, 1994-2014.
Private sector Public sector
EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
South Asia Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of releases
Resistance to Tolerance to Other: Early or
Region High yield Better quality biotic stresses abiotic stresses late maturity
China 89 5 7 10 0
EU and high-income countries 38 60 57 56 1
Former Soviet Union countries 77 9 85 92 1
Latin America 27 25 62 25 6
South Asia 32 21 60 77 0
Sub-Saharan Africa 51 34 47 37 1
West Asia and North Africa 51 23 63 61 0
World 49 36 58 56 1
Source: 2014 survey.
Rows do not add up to 100, as there were combined responses.
27
Impacts of International
Wheat Improvement Research 1994- 2014
Table 4.8. Reported breeding objectives by rank of importance (%).
Breeding objectives 1
st
2
nd
3
rd
Overall
High yield 71 9 8 30
Biotic stress resistance 9 54 27 29
Abiotic stress tolerance 17 12 43 24
Better quality 7 23 29 19
Wide adaptation 1 5 1 2
Early maturing varieties 1 1 3 2
Yield stability 0 1 1 1
Short plant stature 0 1 0 0
Source: 2014 survey.
1 Columns do not add up to 100 as there were some combined responses.
Rank of importance
1
BREEDING OBJECTIVES AND
ATTRIBUTES OF WHEAT
VARIETAL RELEASES
Survey responses provided information
on varietal attributes for about a third
of the wheat releases reported. The
attributes most often targeted included
resistance to biotic stresses (58%),
tolerance to abiotic stresses (56%), high
yield (49%), and better quality (36%)
(Table 4.7), with substantial variation
among regions. High yield (89%) was the
predominant attribute for wheat in China,
whereas in former Soviet Union countries
yield was second to biotic and abiotic
stress tolerance. Attributes reported for
wheat releases in high-income countries
included better quality and biotic and
abiotic stress tolerance in equal measure.
In Latin America, biotic stress resistance
was most often targeted, whereas in
South Asia abiotic stress tolerance
ranked somewhat higher than biotic
stresses. Some breeders/respondents did
not prioritize yield as the most important
attribute because it is considered
obvious. In general however, yield is still
the overarching driver in all countries
where varieties are tested in registration
trials. For instance, in the EU, no variety
will be released that is not competitive
for yield.
Evidently, yield was still important but
other traits were considered of equal
or greater importance during the 15
years leading up to the current study,
due to factors such as the increasing
ability of disease organisms to evolve
and overcome resistance and the rising
emphasis on end-use quality.
Respondents were also asked to name
the three most important breeding
objectives for the next five years. High
yield was ranked first, followed by biotic
stress resistance, abiotic stress tolerance,
and quality, in priority ranking and overall
unweighted average (Table 4.8). These
findings align with the attributes reported
for varietal releases over the last two
decades and also the important breeding
objectives mentioned in Chapter 3.
28
05
GLOBAL WHEAT
VARIETAL ADOPTION
29
Impacts of International
Wheat Improvement Research 1994- 2014
This chapter looks into global wheat varietal adoption and particularly the CGIAR’s
contribution. We also review some of the characteristics of wheat varietal adoption –
including the lags in adoption/varietal replacement, wheat varietal attributes, and the
effects of adoption on genetic diversity.
WHEAT VARIETAL
ADOPTION
Use of improved varieties – also known
as modern varieties (MVs) – in wheat
production has long been widespread,
with universal use in the developed
economies. Still, based on the results of
the current and previous global impacts
survey, adoption of MVs increased from
93% in 2002 to 97% in 2014 (Table 5.1 –
covering only the 32 countries included
in both surveys). The most notable
increases were observed in West Asia
and North Africa (WANA) and particularly
Sub-Saharan Africa (SSA). Aggregate
area sown to improved varieties across
the 32 countries had expanded by 4.6
million hectares in 2014, a phenomenon
seen in the EU and high-income
countries, former Soviet Union countries,
South Asia, Sub-Saharan Africa,
and WANA.
Table 5.1. Adoption of improved, modern varieties (MVs) in 2002 and 2014, in a subset of countries covered in both surveys.
Number of Improved varieties, 2002 Improved varieties, 2014
Region paired countries Area (000 ha) Adoption (%) Area (000 ha) Adoption (%)
China 1 26,033 100 24,213 100
EU and high income countries 4 4,140 100 4,237 100
Former Soviet Union countries
1
7 5,650 100 7,808 100
Latin America 7 9,051 99 7,594 100
South Asia 4 36,324 100 39,337 100
Sub-Saharan Africa 4 1,019 76 1,753 98
West Asia and North Africa 5 16,644 80 18,537 86
Total/weighted average 32 98,861 93 103,479 97
In contrast, the area sown to improved
wheat varieties contracted in China and
Latin America over the same period.
Spring wheat in China was largely
replaced by maize, a crop in high
demand as a source of feed grain.
Across all 2014 study countries, global
wheat area amounted to 165.7 million
hectares, comprising 149.1 million
hectares (90%) of improved varieties, 3.1
million hectares (2%) of landraces, and
13.4 million hectares (8%) of unidentified
or unknown varieties, some of which may
be improved (Table 5.2). Spring bread
wheat comprised more than half (58%)
of the global wheat area, followed by
winter/facultative bread wheat (34%),
spring durum (8%), and winter/facultative
durum (about 1% - Table 5.2).
1 The excess area coverage (13.6 million hectares) in 2014 for some countries in this
group was excluded in the analysis to avoid bias in the actual MV area expansion.
Source: 2002 and 2014 global wheat impacts database.
30
Table 5.2. Area (million ha) sown to different wheat types and variety classes in survey countries, 2014.
CGIAR CONTRIBUTION TO
MODERN VARIETIES ADOPTED
Globally, the area sown to CGIAR-related wheat varieties in 2014 was nearly 106 million
hectares – 71% of the area sown to modern, improved varieties. Area under non-
CGIAR wheat varieties developed by public programs in 2014 was 39.5 million hectares
(26.5%), while 3.8 million hectares (2.5%) of the area was sown to non-CGIAR wheat
varieties from private companies.
Use of CGIAR-related MVs varied significantly among regions (Table 5.3). As expected,
the CGIAR share was highest (>90%) in the main target regions of the developing
world (South Asia and Sub-Saharan Africa) and lower in Latin America (Table 5.3).
Contributions from CGIAR centers figured significantly in high-income countries,
whereas the shares for China and the former Soviet Union were below average, with
the area grown to unknown varieties excluded from the calculation. For comparison
we included the CGIAR share of varietal releases and, overall, the CGIAR contributed
slightly more to adoption (71%) than to varietal releases (63%). In Sub-Saharan Africa
and WANA, the CGIAR share in the use of improved varieties was substantially larger
Table 5.3. CGIAR contribution to modern wheat (MV) varieties adopted worldwide, 2014.
Adoption Release Difference
Estimated between CGIAR adoption
Country / region adoption (%) CGIAR share (%) CGIAR share (%) and release shares (%)
China 100 28 54 (26)
EU and high-income countries 80 82 56 26
Former Soviet Union countries 90 25 49 (24)
Latin America 84 78 73 5
South Asia 99 98 92 6
Sub-Saharan Africa
a
97 97 72 25
West Asia and North Africa 84 98 71 27
Weighted average
b
90 71 63 8
a Excluding South Africa due to unavailability of adoption data.
b Weighted by total area, except the share in adoption estimates that are weighted by total adopted area in each region.
Source: 2014 global wheat impacts survey.
than the share in varietal releases, but
in China and the Former Soviet Union19
it was considerably lower than the other
regions. Our estimate of the CGIAR share
in improved wheat adoption in China
(28%) corroborates the findings of Huang
et al. (2015), whose study reported that
more than 26% of major wheat varieties
in the country since 2000 contained
contributions from CIMMYT breeding and
that these had enhanced the performance
of China’s wheats for traits such as
yield potential, grain processing quality,
disease resistance, and early maturity.
19 Use of CGIAR germplasm however had increased in Central Asia in
the last decade.
Wheat type Improved, modern varieties (MV) Landraces Unknown varieties
a
All
Spring bread wheat 89.8 1.0 5.1 95.9
Spring durum wheat 10.7 0.1 1.9 12.6
Winter/facultative bread wheat 47.8 2.0 6.3 56.1
Winter/facultative durum wheat 0.9 0 0.2 1.1
All wheat types 149.1 3.1 13.4 165.7
Source: 2014 Global wheat impacts survey;
a Some may be modern or improved varieties.
31
Impacts of International
Wheat Improvement Research 1994- 2014
Figure 5.1. Spring bread wheat area shares (%) by origin of germplasm and region, 2014.
Figure 5.2. Spring durum wheat area shares (%) by origin of germplasm and region, 2014.
Unknown varieties Landraces Non-CGIAR CGIAR ancestry CGIAR parent CGIAR line
China EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
South Asia Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of area sown
Unknown varieties Landraces Non-CGIAR CGIAR ancestry CGIAR parent CGIAR line
EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
South Asia Sub-Saharan
Africa
West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of area sown
The aggregate CGIAR contribution can
be grouped into varietal origin categories
based on pedigree analysis and by wheat
type: spring bread wheat (Figure 5.1),
spring durum (Figure 5.2), and winter/
facultative bread wheat (Figure 5.3).
Of all wheat types, spring bread wheat
accounted for the largest area share
sown to CGIAR-related varieties (more
than 70%; Figure 5.1). Sub-Saharan
Africa stood out with a near universal use
of CGIAR-related varieties20 and a heavy
reliance on CGIAR-lines. This is followed
by South Asia, with more than 90% of
the area sown to CGIAR-related cultivars.
Likewise, more than 85% of wheat area
in WANA is sown to CGIAR-related
germplasm. In China, CGIAR-related
varieties occupy nearly half the area sown
to spring bread wheat.
A somewhat smaller area of spring
durum wheat was sown to CGIAR-related
varieties (approaching 70%; Figure 5.2).
In SSA use of CGIAR-related varieties
was very high, with particularly heavy
reliance on CGIAR ancestry. In WANA,
more than 90% of the wheat area was
sown to CGIAR-related varieties, a result
of long collaboration between national
programs and CIMMYT and ICARDA.
There was significant use of CGIAR
durum lines in WANA and Latin America.
CGIAR-related cultivars were grown on
about 70% of the durum wheat area in
the EU and high-income countries, with a
major presence in Italy and Spain.
20 South Africa is not included due to the unavailability of adoption
data.
32
The share of CGIAR-related varieties
sown in winter/facultative bread wheat
areas was slightly more than 40% (Figure
5.3). Still, those varieties accounted
for nearly 70% of the winter/facultative
area in WANA, including Afghanistan,
Iran, Turkey, and other countries – an
outcome in part of nearly three decades
of Turkey-CIMMYT-ICARDA collaboration,
as well as of an earlier joint program of
CIMMYT with Oregon State University. An
impact assessment by Jilani et al. (2013)
showed that CGIAR-related varieties
performed well in Afghanistan, particularly
in irrigated areas. Similarly, a quarter of
China’s winter/facultative bread wheat
area was sown to CGIAR-
related varieties.
In Table 5.4 we further disaggregate the origin of improved wheat varieties by source of
cross and CGIAR contribution. Across all wheat types, 35.7 million ha (22%) were
sown to CGIAR crosses, 99.5 million ha (60%) to public national program crosses,
13.9 million ha (8%) to private sector crosses, 3.1 million ha (2%) to landraces, and
13.4 million ha (8%) to unidentified varieties. Both the national program and private
sector crosses showed a heavy reliance on CGIAR-related breeding material. As
expected, CGIAR germplasm was particularly dominant in spring bread and durum
wheats. Nearly 27 million hectares of spring bread wheat area was sown to direct
releases of CGIAR lines. CGIAR germplasm has figured as parents in another
26.6 million ha of spring bread wheat varieties released by publicly-funded national
programs. An additional 12.9 million ha was sown to varieties whose pedigrees featured
CGIAR lines as grandparents or earlier ancestors utilized by national programs. CGIAR-
related germplasm was in the parentage or earlier ancestry of private, spring bread
wheat releases grown on nearly 7.0 million hectares.
More than one-third (37%, or 4.6 million ha) of spring durum area is sown to CGIAR
lines, with an additional 3.9 million ha (31%) of CGIAR-related public and
private-sector crosses.
Figure 5.3. Winter/facultative bread wheat area shares (%)
by origin of germplasm and region, 2014.
Unknown varieties Landraces Non-CGIAR CGIAR ancestry CGIAR parent CGIAR + TCI line
EU and other
high income
countries
Former Soviet
Union
countries
Latin
America
China West Asia
and North
Africa
World
0
20
40
60
80
100
Percentage of area sown
Table 5.4. Area (million ha) sown to different wheat types, classied by origin of germplasm, 2014.
Public national Private sector
program crosses releases
CGIAR CGIAR CGIAR Non- CGIAR- Non- Land Unknown
Type line parent ancestry CGIAR related CGIAR races varieties All
Spring bread wheat 26.9 26.6 12.9 14.5 6.9 2.0 1.0 5.1 95.9
Spring durum wheat 4.6 1.3 1.5 1.8 1.1 0.3 0.1 1.9 12.6
Winter/facultative
bread wheat 4.3 4.2 13.0 22.8 2.0 1.5 2.0 6.3 56.1
Winter/facultative
durum wheat - - 0.5 0.3 0.1 - - 0.2 1.1
All wheat types 35.7 32.1 27.9 39.5 10.1 3.8 3.1 13.4 165.7
Source: 2014 global wheat impacts survey.
33
Impacts of International
Wheat Improvement Research 1994- 2014
Figure 5.4. CGIAR contribution to spring bread wheat varieties grown worldwide, 2014.
Figure 5.5. CGIAR contribution to spring durum wheat grown worldwide, 2014.
ALTERNATIVE MEASURES OF
CGIAR CONTRIBUTION TO
WHEAT VARIETIES GROWN
As discussed in Chapter 2, the rules for
crediting CGIAR contributions can be
used to generate aggregate indicator
estimates of the CGIAR contribution
to the varieties grown. The “CGIAR
cross” rule underestimates the CGIAR
contribution, since it restricts it to
include only CGIAR crosses or lines. In
contrast, the “any CGIAR ancestor” rule
overestimates the CGIAR contribution,
because it attributes all varieties with a
CGIAR relation to the CGIAR, regardless
of how far back in the pedigree tree
the CGIAR germplasm was used. The
most accurate estimate is provided
by BROWSE, which accounts for the
CGIAR contribution at different stages
of varietal development and gives more
credit for the most recent cross. The
“CGIAR cross” (0.28) and “CGIAR cross
plus parent” (0.35) rules approximate the
BROWSE estimate (0.33).
The CGIAR contribution to wheat
varieties grown can be disaggregated by
wheat type: spring bread wheat (Figure
5.4), spring durum (Figure 5.5) and winter/
facultative bread wheat (Figure 5.6).
It is interesting to contrast the CGIAR
contribution across all study countries
and solely in developing countries: as
would be expected and regardless of the
attribution measure used, developing
countries have benefited more from
CGIAR contributions to spring bread and
spring durum wheat (Figures 5.4 and 5.5),
but spillovers in other countries have also
been significant.
40
Percentage
Study countries
Developing countries
0
15
30
45
60
75
90
CGIAR cross rule CGIAR cross plus
parent rule
Any ancestor rule BROWSE
Study countries
Developing countries
CGIAR cross rule CGIAR cross plus
parent rule
Any ancestor rule BROWSE
0
20
40
60
80
100
Percentage
34
It is remarkable that for winter/facultative
bread wheat, the “any ancestor” rule
generates a higher CGIAR contribution
for all study countries than for developing
countries alone (Figure 5.6). This implies
that CGIAR lines are being used by
several developed countries during early
stages of varietal development.
Overall, the findings highlight the
widespread use of CGIAR wheat
improvement outputs in the developing
world, as well as significant spillovers
in more developed economies. The
continued use of improved wheat
germplasm is expected to have benefited
smallholder farmers in the developing
world and to have enhanced the
availability of food to poor consumers.
Both conclusions are beyond the scope
of this study but are supported by
studies like Shiferaw et al. (2014), which
documented the enhanced availability
and affordability of food to poor
consumers in Ethiopia from the adoption
and use of improved wheat varieties in
that country.
CHARACTERISTICS OF
WHEAT VARIETAL ADOPTION
Figure 5.6. CGIAR contribution to winter/facultative
bread wheat varieties grown worldwide, 2014.
Study countries
Developing countries
CGIAR cross rule CGIAR cross plus
parent rule
Any ancestor rule BROWSE
0
10
20
30
40
50
60
70
Percentage
Our study showed that most farmers in developing countries had adopted improved
wheat varieties, but that a large share of these were older improved varieties and that
the replacement of old varieties was slow. Slow uptake of new improved varieties delays
and reduces the potential benefits from wheat improvement research, as farmers who
grow old varieties forego gains from the improved yield potential and better disease
resistance of the newer varieties. A recent stochastic frontier analysis by Battese et
al. (2014), for example, substantiates the decreased technical efficiency of wheat
production in Punjab, Pakistan, from slow varietal replacement.
One major driver of varietal turnover has been the emergence of new, highly-virulent
strains of fungal pathogens – particularly those associated with the rusts – that are able
to overcome the genetic resistance protecting older improved varieties. Khan (1987)
estimated that it takes five to six years for leaf rust resistance to break down in wheat
in northern Pakistan, while Byerlee and Heisey (1990) revealed an average longevity of
three to five years for rust resistance in northern Mexico. The longevity of rust
resistance based on a single race-specific rust resistance gene approaches six years
(Kilpatrick 1975).
LAGS IN ADOPTION/VARIETAL REPLACEMENT
35
Impacts of International
Wheat Improvement Research 1994- 2014
Age (years) 1997a 2014b
<6 Zimbabwe , Afghanistan Argentina, Burundi, Czech Republic, Eritrea, Georgia, Hungary, Kenya,
Lebanon, Spain, Ukraine
6-8 Argentina, Brazil, Chile, China, Guatemala, Brazil, Paraguay, Rwanda, Tajikistan, Uruguay, Zimbabwe
Pakistan
8-10 Bolivia, Colombia, Iran, Nigeria, Afghanistan, W. Australia, Azerbaijan, Bangladesh, Canada,
Uruguay, Zambia China, Ethiopia, Iran, Italy, Japan, Latvia, Mexico, Nepal, Pakistan,
Romania, Tanzania, USA, Uzbekistan, Zambia
10-12 Ecuador, Morocco, Paraguay,South Africa, Armenia, Bolivia, Egypt, Kazakhstan, Nigeria, Turkmenistan,
Tanzania Uganda, Switzerland
12-14 India, Kenya, Lebanon, Mexico, Albania, Belarus, India, Israel, Portugal, Turkey, Serbia,
Syria, Yemen Slovenia, Russian Federation (Omsk Region)
>14 Algeria, Bangladesh, Egypt, Ethiopia Algeria, Bhutan, Ecuador, Jordan, Kyrgyzstan, Morocco,
Jordan, Nepal, Peru, Sudan, Tunisia, Turkey Sudan, Syria
c
, Tunisia
Source: a Heisey et al. (2002);
b 2014 Global Wheat Impacts survey.
c For Syria, adoption data before the country’s troubling situation were used in the analysis.
The breakdown of rust resistance is followed by rust epidemics and, eventually,
replacement of susceptible varieties, a phenomenon known as the “boom-and-bust”
cycle. The lag in the development and release of new resistant varieties seriously harms
the food security and livelihoods of smallholder farm families. To achieve more durable
resistance to rust pathogens, since the mid-1970s CIMMYT has pursued a breeding
strategy that involves endowing varieties with four-to-five partial resistance genes (also
known as “slow rusting” or adult-plant-resistance genes, because they usually become
effective in post-seedling growth stages) that individually have small-to-intermediate
effects but in combination often provide levels of resistance that are comparable to
immunity. This approach is described in greater detail in Chapter 3, in the section
“Enhancing the frequency of lines with durable resistance to wheat rusts.”
Table 5.5 compares the weighted average age (WA) of improved wheat varieties grown
in 1997 and 2014 (though earlier for Syria), with WA divided into 6 categories: less
than 6 years, 6-8 years, 8-10 years, 10-12 years, 12-14 years, and more than 14 years.
Consistent with past studies (Byerlee and Moya 1993; Heisey et al. 2002; Lantican et
al. 2005; Dixon et al. 2006; Krishna et al. 2015), a significant proportion of the current
wheat area is sown to older improved wheat varieties.
The WA had improved in some countries over the two periods, but in others there were
long or lengthening time lags for varietal turnover (Table 5.5). Varietal replacement in
Argentina in 2014 had improved to less than 6 years from the 6-8 year WA for 1997,
possibly due to greater involvement of the private sector. WA in Kenya had improved
tremendously, from 12-14 in 1997 to less than 6 years in 2014, a consequence of the
need to provide farmers with new varieties able to resist the stem rust race Ug99. In
contrast, varietal replacement in Afghanistan and Zimbabwe was occurring in less than
6 years in 1997 but had slowed to 8-10 and 6-8 years, respectively, in 2014, likely due
to farmer preferences for the attributes of certain older improved varieties.
Table 5.5. Weighted average age of varieties grown by farmers, 1997 and 2014.
36
Particular varietal attributes can lead to strong farmer preferences for specific cultivars.
Good examples are CIMMYT’s Attila22 line, released as PBW 343 in India in 1995, and
LOK-1, which has two CIMMYT parents and was released in India in 1981. These two
older improved varieties are now susceptible to stem rust race Ug99, but farmers still
prefer them over other varieties, due to their productivity and other traits. In addition
to high yields, PBW 343 has yield stability – that is, it is dependable under varying
conditions – and it has good heat tolerance, a quality of great value in India, where pre-
monsoon temperatures regularly exceed 40o C. In 2014, this variety was sown on more
than 2 million hectares in India alone. Likewise, LOK-1, which was grown on about 1
million hectares in India in 2014, is broadly adapted and very good for making chapattis.
Table 5.6 presents the foremost reported attributes of 499 wheat varieties grown in
2014, covering resistance to biotic stresses, tolerance to abiotic stresses, high yield,
and superior quality. Some respondents listed several attributes for a particular variety.
High yield was the most important attribute for farmers in China, whereas in Latin
America better quality, high yield, and resistance to biotic stresses were favored equally.
In 1997, wheat varietal replacement in Pakistan was in the 6-8 years category but
slowed to 8-10 years in 2014 (Table 5.5). This seems to differ from results of a recent
study using duration analysis and showing that irrigated areas of Punjab, Pakistan,
have an average varietal turnover rate of four years (Nazli and Smale 2016). The authors
however observed that their results would likely have been different, had they estimated
varietal replacement in rain-fed wheat areas, implying that environment can be a key
factor in varietal turnover. Lantican et al. (2003), on average, also found that varietal
replacement occurred three years sooner in favorable wheat-growing environments than
in marginal ones.
Bangladesh, Ethiopia, and Nepal had markedly improved WAs in 2014, down to 8-10
years from more than 14 years in 1997, while in Mexico WA improved from 12-14 years
to 8-10 years (Table 5.5). Farmers in India were changing varieties in 2014 WA a year
earlier than in 1997 but still remaining in the WA category of 12-14 years; due again to
farmer-preferred attributes in older varieties. Among West Asian countries, only the WA
of Turkey had improved, from more than 14 years in 1997 to 12-14 in 2014. Mazid et al.
(2014) link low adoption of newer improved varieties in Turkey to farmers’ knowledge
and perception of certain varietal attributes, together with the unavailability of adequate
or timely seed.21 Other WANA countries have remained in the slowest WA category. The
same case applies for Sudan.
China’s WA declined from 6-8 years in 1997 to 8-10 years in 2014. This appears to
contradict Huang et al. (2015), who claimed that farmers in China replaced their wheat
varieties in less than four years, but this may simply be a result of the method used
to estimate varietal turnover. They conducted a farm survey in 2011 wherein farmers
were asked to indicate how often they changed varieties, from among options of 0 to 9
years. The results do not indicate that all farmers replaced their older varieties with new,
improved ones. Indeed, from our survey results, some varieties grown in China in 2014
included those released in the late 1990s and even a few released in the 1980s, likely
due (again) to their possessing attributes valued by farmers.
21 The latter has changed in recent years since Turkey has introduced a system that encourages farmers to buy new seed. (Ministry of Food,
Agriculture, and Livestock 2013)
22 Attila’s pedigree is ND/VG9144//KAL/BB/3/YACO/4/VEE#5 and has been released in several countries.
WHEAT VARIETAL ATTRIBUTES
37
Impacts of International
Wheat Improvement Research 1994- 2014
Figure 5.7. Genetic diversity in the Elite Spring Wheat Yield Trial (ESWYT),
with genetic distance measured as average Rogers distances.
Table 5.6. Attributes (%) of 499 wheat varieties by region, 2014.
y = -0.001x + 0.3376
R2 = 0.2405, p = 0.013
0.40
0.35
0.30
0.25
0.20
Average genetic distance
501015
ESWYT
20 25
Several studies (Smale et al. 2001; Dreisigacker et al. 2004; Zhang et al. 2005; Reif et
al. 2005; Warburton et al. 2006) have examined the effects of the widespread use of
CGIAR wheat germplasm on the global genetic diversity of bread wheat. Molecular
marker analysis on the use of synthetic hexaploid wheat has shown that synthetic
backcross-derived lines are ideal for increasing diversity (Dreisigacker et al. 2008), as
they provide a combination of precise elements to incorporate new genes for increased
yield, abiotic stress tolerance, and biotic stress resistance (Trethowan and Mujeeb-Kazi
2008; Zhu et al. 2014). Furthermore, based on genetic distance, Dreisigacker et al. 2012
concluded that a constant level of genetic diversity has been maintained over the years
in CIMMYT’s Elite Spring Wheat Yield Trial (ESWYT) (Figure 5.7).
Genetic distance between individual
ESWYT lines significantly increased
when lines were grouped according
to differences in years of ESWYT
dissemination, suggesting a systematic
change in allele frequencies over
time, most likely due to breeding and
directional selection (Dreisigacker et al.
2012). These studies highlight that
there is no loss of genetic diversity
among CIMMYT/CGIAR-bred bread
wheat materials.
Source: Dreisigacker et al. 2012.
Source: 2014 global wheat impacts survey.
* As part of the survey, participants were asked to list the attributes or traits that they considered valuable in particular adopted varieties. The open-ended responses were grouped into the ve trait
categories listed here, and a percentage given based on the number of times a specic trait was mentioned, compared to the total number of responses received. Sums across rows exceed 100
because many respondents mentioned multiple varietal attributes.
High Better Resistance to Tolerance to Other
Region (n) yield quality biotic stresses abiotic stresses (growth cycle)
China (23) 78 17 4 13 0
EU and high income countries (205) 49 46 63 65 0
Former Soviet Union Countries (44) 45 20 61 79 2
Latin America (20) 50 50 50 35 0
South Asia (44) 30 21 73 69 0
Sub-Saharan Africa (36) 47 15 62 34 6
West Asia and North Africa (127) 47 36 71 61 2
World (499) 48 35 63 60 1
GENETIC DIVERSITY IN THE CGIAR-BREAD WHEAT PIPELINE
38
06
BENEFITS OF WHEAT
IMPROVEMENT
RESEARCH
39
Impacts of International
Wheat Improvement Research 1994- 2014
TE1993 TE2003 TE2013
Wheat yield kg/ha 2,507 2,711 3,176
Yield growth rate (% pa) < 1.14% >< 1.46% >
< 1.18% >
Average increase in yield over base yield TE1993 (kg/ha/yr) < 132 >< 452 >
< 292 >
Source: Derived from FAOSTAT.
TE = triennium ending.
Table 6.1. Global wheat yields and underlying growth rates.
Previous global wheat impact studies have reported large benefits from international
wheat improvement efforts. This chapter discusses whether this positive trend
continues. It reviews wheat yields and estimates and discusses the benefits from wheat
improvement research.
WHEAT YIELDS
Global wheat yields averaged 3.2 tons
per hectare (t/ha) for the triennium ending
2013 (TE2013), up from 2.5 t/ha two
decades earlier and with an underlying
growth rate of 1.2% p.a. Taking the
2.5 t/ha from TE1993 as the base yield,
the cumulative additional production over
the base amounts to 5,842 kilograms by
TE2013, or an annual average of
292 kilograms per hectare (kg/ha;
Table 6.1). The observed yield growth
is understood as the net result of three
factors:
• Growth in genetic yield potential.
• Yield maintenance (for example,
breeding for disease resistance) to avert
yield declines.
• Use by farmers of yield-enhancing crop
management practices.
Yield growth was relatively slow in the
first decade and accelerated during the
second, probably as a result of replacing
rust-susceptible varieties with new,
resistant ones.
The growth in yield potential is the
most easily measured gain from wheat
improvement research. An assessment
of genetic progress for yield using results
from CIMMYT’s Semi-Arid Wheat Yield
Trial23 over a 17 year-period showed
a gain of 1% per year and concluded
that there had been consistent genetic
progress (Manes et al. 2012). In a related
study by Sharma et al. (2012), genetic
yield gains in CIMMYT spring bread
wheat, based on data from the Elite
Spring Wheat Yield Trial (ESWYT) during
1995-2009, ranged from 0.5% to 1.13%
per year. Results of a study conducted by
Kansas State University that quantified
the impact of genetic improvement in
wheat during 1985-2011 showed that
wheat breeding programs had improved
average yields over that period by a
cumulative 917 kg/ha or a cumulative
27% of the base yield (Barkley et al.
2014).
Notwithstanding, the environment directly
influences the expression of many genes,
so it is not always easy to disentangle
the purely genetic component of yield
from environmental effects. Moreover,
unfavorable environments furnish a
lower baseline for yield gain studies. For
example, in ESWYT data for 1979-99,
wheat yields in dry and hot environments
showed higher annual growth rates (3.5%
and 2.1% respectively) than those from
favorable environments (0.8-1.1%).24
23 This replicated yield trial contains spring bread wheat germplasm
adapted to low rainfall, drought-prone environments typically
receiving less than 500 millimeters of water during the
cropping cycle.
24 This is due partly to an increased focus on selection for heat and
drought tolerance (Lantican et al. 2003).
40
Appendix Tables A.1 and A.2 present
updated summaries of rates of yield gains
in various locations and environments in
developing and developed countries.
Maintaining disease resistance in the face
of evolving pathogen biotypes has been
a major thrust in wheat improvement
research and particularly in international
wheat breeding, given resource-poor
farmers’ lack of disease control measures
(Reynolds and Borlaug 2006), but the
yield losses averted through breeding for
disease resistance are hard to quantify.
Singh and Rajaram (2002) claimed that
wheat yield gains are contingent upon
maintaining genetic disease resistance
for the rusts, Septoria diseases, leaf
blight, blotch, and tan spot. Marasas et
al. (2004) estimated a net present value
of US $5.4 billion [1990] from leaf rust
resistance research during 1973-2007.
Likewise, a 2009 study that quantified
the benefits from CGIAR research on
yield stability estimated the annual global
value of genetic resistance to various
diseases at about US $2 billion (CGIAR
2011). Sayre et al. (1998) showed the
maintenance effect to be substantially
larger than the yield potential effect,
under experimental conditions with
increased disease pressure. Byerlee
and Traxler (1995) assumed that, under
farmers’ conditions, the maintenance
effect would equal yield potential gains in
irrigated and high-rainfall areas, and be
somewhat lower in less favorable areas,
due to lower disease pressure.
BENEFITS FROM WHEAT
IMPROVEMENT RESEARCH
The net yield gain attributable to wheat improvement research is the main component
for calculating the annual benefits the research brings. Due to the difficulty of attributing
yield gains in farmers’ fields to the multiple causes involved, we assumed that three
principal underlying factors – yield potential, yield maintenance, and crop management
– contribute equally. For this study, the aggregate effect of crop improvement was
assumed to consist of yield potential + yield maintenance.
We used two attribution scenarios for the annual benefits of wheat improvement
research: (1) historic average increase over the base yield and (2) marginal yield increase
from longevity. For each of the preceding scenarios, we included a base situation
wherein the observed wheat yield gains for 1993-2013 reflected the aggregate effect
of crop improvement (292 kg/ha/year for Scenario 1 and 1.18% p.a. for Scenario 2).
We complemented each base scenario with a range estimate. The “low-end” or more
conservative estimate was calculated as half of the observed gains. An alternative proxy
interpretation for this lower rate would be that it reflects yield potential alone, in the
absence of maintenance breeding. The “high-end” or more liberal estimate is based on
the yield gains during the second decade, 2004-13. Such a higher rate could reflect a
structural upward shift in yield gains associated with crop improvement.
The CGIAR contribution rules were used to estimate the CGIAR share of the gross
benefits of global wheat improvement. The most realistic indicator of the CGIAR
contribution is generated by applying the ICIS program BROWSE – which estimates the
CGIAR share in global wheat germplasm at 33%.
The estimated additional annual wheat production due to international wheat
improvement research, based on the first attribution measure used (historic average
yield increase from base yield) ranged from 21.7 to 67.4 million tons (Table 6.2). Using
the BROWSE-generated CGIAR contribution (0.33), the additional wheat production
attributable to the CGIAR ranged from 7.2 to 22.2 million tons of wheat per year.
The remaining two-thirds of additional annual wheat production are due to wheat
improvement efforts of non-CGIAR partners.
41
Impacts of International
Wheat Improvement Research 1994- 2014
Scenario
Historic average increase Marginal yield increase
over base yield by longevity
Yield increase 292 kg/ha 1.18%
(146-452) (0.6-1.46%)
Annual benets (billion 2010 US$) attributed to:
Global wheat improvement research 9.4 6.7
(4.7-14.5) (3.4-8.4)
CGIAR wheat improvement research 3.1 2.2
(based on BROWSE) (1.5-4.8) (1.1-2.8)
Table 6.3. Benets from global wheat improvement research
(high- and low-end estimates in parentheses).
In the second scenario (marginal yield increase by longevity), the additional annual
wheat production due to international wheat improvement research ranged from 15.7
to 38.8 million tons (Table 6.2). The additional wheat production attributable to CGIAR
wheat improvement research ranged from 5.2 to 12.8 million tons per year.
From the above computed additional annual wheat production and using the average
World Bank’s real price25 of wheat for the study period (1994-2014) – US$ 215 [2010]
per ton – global wheat improvement research generated an annual benefit of US $9.4
($4.7-$14.5) billion [2010] under the first scenario (historic average yield increase over
the base yield) (Table 6.3). Annual benefits are somewhat lower but still substantial
under the second scenario (marginal yield increase by longevity): US $6.7 ($3.4-$8.4)
billion [2010].
Of this, the annual benefits attributable to CGIAR wheat improvement research were
US $3.1 ($1.5-$4.8) billion [2010] for the first scenario and US $2.2 ($1.1-$2.8) billion
[2010] for the second (Table 6.3).
25 Based on price data from Global Economic Monitor (GEM) Commodities (http://databank.worldbank.org/data/reports.aspx?source=global-
economic-monitor-%28gem%29-commodities&savedlg=1&l=en# . )
*Calculated from Table 6.2 using a reference price. For a more conservative estimate of annual benets due to wheat improvement research,
the 1994-2014 average real price (US $215/t) was used instead of the higher 2014 wheat price equivalent to US $267/t [2010].
Table 6.2 Additional annual wheat production due to wheat improvement research based on two attribution
scenarios, 2014 (for each scenario, low-, mid-range, and high-end estimates are given).*
1.Historic average
increase over base yield 0.146 21.7 7.2
0.292 43.5 14.4
0.452 67.4 22.2
2. Marginal yield
increase from longevity 0.0187 15.7 5.2
0.0375 31.4 10.4
0.0464 38.8 12.8
Attribution scenario Assumed yield gain (t/ha)
Additional annual wheat production
due to international wheat
improvement research (million tons)
Additional annual wheat production
attributable to CGIAR wheat
improvement research (million tons)
*Total area grown to improved wheat varieties = 149.1 million hectares; the CGIAR contribution generated through BROWSE = 0.33.
42
The CGIAR invests an average of about US $30 million [2010] per year in wheat
improvement research of late; up from only US $10-15 million in the early 2000s, prior
to the 2008 world food price crisis. Thus, regardless of the scenario or contribution
rule used, annual returns to CGIAR investments in wheat improvement have
been substantial.
The “historic average yield increase” scenario is very sensitive to the assumed base
yield – TE1993 – and the effect of this sensitivity increases over time, being more
pronounced in the second decade of the study period. The “marginal yield increase
by longevity” scenario is more robust and can be used for marginal forward-looking
analysis; that is, to forecast expected returns from an additional year of investment in
wheat improvement research. It takes prior investments as sunk costs – which clearly
lay the foundation for future benefits – and gives cumulative expected benefits from
each year’s investment.
Both scenarios could be variously strengthened by more robust measurements
underlying the various assumptions. The BROWSE application provided a particularly
robust way of estimating the CGIAR contribution to the germplasm, but 8% of the
varieties remained unidentified. Most likely these were improved varieties but with
unknown ancestry, including undisclosed origins. More reliable estimates on the extent
of varietal use across the globe are still much needed; the current study relied heavily
on expert estimates compiled during the survey. New developments such as DNA
fingerprinting offer new prospects to better identify ancestries and the extent of their
use. Stronger attribution of yield gains would be another area that merits improvement.
Aside from these various improvements, it will be advisable to continually monitor
progress and impacts, and possibly take stock and document impacts every five years.
Finally, several aspects – including price and distributional and non-yield effects – were
not accounted for in our study. We discuss them in the next section.
DISCUSSION
Benefits of international wheat
improvement research can be estimated
in various ways. Our study followed a
simple economic surplus approach, as
did two previous global wheat impact
studies (Heisey et al. 2002 and Lantican
et al. 2005). Those studies focused
on developing countries and the latter
study valued the benefits attributable to
global wheat improvement efforts during
1988-2002 at US $2.0-6.1 billion [2002]
per year. Although the studies apply
similar underlying models, their results
are not directly comparable, given the
expanded target areas and wheat areas,
certain of the underlying assumptions,
and differences in the international
reference prices of wheat used for each
study. Still, the outcomes of both reiterate
the continuous and substantial returns
to investments in international wheat
improvement research, even now that
adoption of modern varieties is becoming
near universal and farmers upgrade to
newer and better modern varieties.
Byerlee and Traxler (1996) likewise
used the economic surplus approach to
estimate the impact of the joint CIMMYT/
national wheat genetic improvement
efforts, focusing on spring bread wheat
and looking at an investment stream
and the associated benefit stream over
time. Their calculations arrived at global
benefits of US $2.5 billion per year due
to wheat breeding research, of which
about US $1.5 billion per year could
be attributed to the CIMMYT/national
wheat program network and varieties of
CIMMYT origin.
43
Impacts of International
Wheat Improvement Research 1994- 2014
Counterfactual scenario
Study Model Wheat output Wheat price
Evenson and Rosegrant (2003) IMPACT -5-6% +19-22%
Stevenson et al. (2013) GTAP-AEZ -43-60% +29-59%
Using the International Model for Policy
Analysis of Agricultural Commodities
and Trade (IMPACT), which is a partial
equilibrium model for the agricultural
sector, and data from 1965-2000,
Evenson and Rosegrant (2003) found
that, in the absence of CGIAR genetic
improvement in wheat, there would have
been a 5-6% decrease in wheat output
and a 19-22% increase in wheat prices
(Table 6.4), with adverse effects on
poverty and nutrition.
Supporting and providing another
perspective on those findings, Stevenson
et al. (2013) used the Global Trade
Analysis Project Agro-Ecological Zone
(GTAP-AEZ) Model, which includes land
rent effects and impacts on land-use
through factor markets, to show that,
without the contributions of CGIAR crop
genetic improvement in the developing
world, wheat production in 2004 would
have been 43-60% lower than observed
output (Table 6.4). This lack would have
been partially offset by increased wheat
imports in developing countries, which
in turn would have driven up the global
weighted average price of wheat by 29-
59% (Stevenson et al. 2013).
Table 6.4. Counterfactual scenarios: a world without CGIAR wheat improvement research.
Irrespective of the model used, all studies
attribute substantial benefits to CGIAR
wheat improvement research. Still, the
simple economic surplus model used
in the current study provides a rather
narrow measure of benefits. Future
impact studies should consider assessing
the benefits of spill-overs, price-effects,
and non-yield benefits such as improved
grain quality, improved fodder and straw
quality, and short growth cycles. In a
recent review of the impacts of CGIAR
research, Renkow and Byerlee (2010)
concluded that direct productivity
impacts and indirect (wage and price)
impacts of modern varieties developed
by international centers and partners
continue to provide huge benefits for the
poor within and outside the
agricultural sector.
Taken with the global importance of
wheat as a food crop and the expected
growth in demand for wheat to 2050, the
current and past impact studies make a
strong case for continued or increased
system-wide investment in genetic
improvement research on wheat. For
CGIAR to continue to generate enormous
benefits from wheat improvement
research, consistent and secure financial
support is crucial.
This message has resonated in studies
that focus on potential wheat production
losses from evolving and virulent crop
pathogens. Byerlee and Dubin (2010)
concluded that sustainable funding was
vital to the success of international wheat
improvement efforts, especially with the
re-emergence of stem rust as a threat to
wheat production and food security in the
developing world. More recently, Beddow
et al. (2015) found that yearly global grain
production lost to wheat stripe rust alone,
estimated at $979 million, warranted a
sustained annual research investment of
at least $32 million.
44
07
CONCLUSIONS
45
Impacts of International
Wheat Improvement Research 1994- 2014
Past global wheat impact studies have
illustrated the significant contribution of
CGIAR-related germplasm to international
wheat improvement efforts. This study
reiterates these findings and provides
further support. It also strongly confirms
the conclusions reached by the earlier
studies: (1) the adoption and spread
of modern wheat varieties has been
sustained in the post-Green Revolution
period, (2) CGIAR germplasm has
continued to be widely used by breeding
programs in the developing countries,
and (3) investment in wheat improvement
research continues to generate
higher returns.
There has been no slowdown in the rates
of release of improved varieties. Between
1994 and 2014, public and private-sector
breeding programs released 4,604 wheat
varieties in the world. Sixty-three percent
of varietal releases were from public-
sector breeding programs, while the
private-sector had accounted for 37% of
wheat varietal releases. Latin America is
the region with the highest percentage
of private-sector releases (53%). EU
and high–income countries had equal
shares (50%) of public and private-sector
releases. In other regions, most wheat
varietal releases came from the
public sector.
More than 60% of wheat varietal releases
since 1994 were CGIAR-related. Direct
use of CGIAR lines was prevalent in
South Asia (50%). In Latin America, the
use of direct CGIAR lines decreased
relative to past impacts studies,
particularly in Argentina and Brazil. There
is an active private seed sector in the
region and, though the pedigrees of
several varietal releases were unknown,
wheat scientists had used CGIAR
germplasm extensively as parents or
grandparents in breeding programs, so
that more than 70% of varietal releases
contained CGIAR contributions.
With the Green Revolution many
decades behind, the use of improved
wheat varieties was widespread but had
continued to expand. A comparison of
32 paired countries showed that the area
under improved varieties had expanded
by 4.6 million hectares during 2002-2014
and adoption had increased from 93% to
97% in some regions, including high-
income countries, former Soviet Union
countries, South Asia, Sub-Saharan
Africa, and WANA.
In contrast to suggestions that yields of
modern varieties are more variable than
those of farmers’ traditional varieties,
Gollin (2006) showed that the relative
variability in wheat grain yields had
actually fallen over 40 years, due to use
of improved varieties. Furthermore, as
discussed earlier, several studies provide
evidence that there has been no loss of
genetic diversity in CIMMYT/CGIAR-
bred varieties.
The significance of CGIAR-related
varieties is evident in farmers’ fields.
The total area sown to CGIAR-related
germplasm in the world is estimated at
about 106 million hectares. Spring bread
wheat occupied more than half (58%)
of the global wheat area in 2014, with
winter/facultative bread wheat coming
second at 34%.
46
As expected, CGIAR made significant contributions to spring bread wheat; 27 million
hectares (28%) were sown to spring bread wheat varieties that constituted direct
releases of CGIAR lines. CGIAR germplasm had contributed indirectly (as parents) in
public national program spring bread wheat releases grown on another 26.6 million
hectares (28% of spring bread wheat area). An additional 12.9 million hectares (14%)
were sown to varieties with CGIAR lines as grandparents or utilized by public national
programs as ancestors in the early development of varieties. Likewise, CGIAR-related
germplasm was grown on nearly 7.0 million hectares of private-sector spring
bread wheat.
By region, use of CGIAR-related spring bread wheat varieties was nearly universal
in South Asia and Sub-Saharan Africa, with the latter region relying heavily on direct
releases of CGIAR lines.
SPRING BREAD WHEAT
Nearly all spring durum wheat grown in
Africa had CGIAR ancestry and – likely
as a result of long collaboration between
ICARDA and CIMMYT – nearly 80% of
the durum wheat area in WANA was
sown to direct releases of CGIAR lines.
Similarly, nearly 70% of the area in
WANA was under CGIAR-related winter/
facultative bread cultivars that can be
credited to almost three decades of
Turkey-CIMMYT-ICARDA collaboration
and an earlier partnership of CIMMYT
with Oregon State University.
DURUM AND WINTER/
FACULTATIVE WHEAT
Developing countries received the greatest benefit from CGIAR contributions,
particularly in spring bread and spring durum wheat areas, an outcome that aligns with
CGIAR’s mandate to help resource-poor farmers and alleviate poverty and malnutrition.
Still, adoption of CGIAR-related cultivars was not limited to developing countries and
our study highlights significant spill-overs.
• In Canada, three-quarters of the wheat area was sown to CGIAR-related cultivars.
• In the USA, nearly 60% of the wheat area was sown to CGIAR-related varieties.
• In Western Australia, CGIAR-related varieties were used on more than 90% of the
wheat area, confirming the findings of Brennan and Quade (2004), who reported high
spillover benefits to Australia from CIMMYT/CGIAR wheat improvement research, as
well as related averted welfare losses.
CHIEF BENEFICIARIES
AND SPILLOVERS
47
Impacts of International
Wheat Improvement Research 1994- 2014
REPLACING
OLD VARIETIES
Most farmers in developing countries have adopted modern wheat varieties, but a large
portion of total wheat area in 2014 was still sown to older improved varieties and few
countries (Argentina, Hungary, Kenya, Lebanon, and Czech Republic) had improved
their varietal replacement rates. Farmers who grow older improved varieties lose out
significantly on gains from improved yield potential or better disease resistance in the
newer varieties, not to mention running the risk of devastating grain losses from disease
outbreaks. A better understanding of the attributes of new varieties, coupled with strong
public and private sector support, are needed to promote adoption of newer improved
wheat varieties.
On average, CGIAR was investing about
US $30 million per year [2010] in wheat
improvement research at the time of the
study, but according to our estimates,
CGIAR wheat improvement efforts
accounted for US $2.2-3.1 billion [2010]
in economic benefits per year, attributable
to increased grain production. The
associated benefit-cost ratio for CGIAR
wheat improvement research ranged from
73:1 to 103:1. So, while it accounted for
a relatively small portion of the global
investment in wheat improvement
research, CGIAR generated a large share
of the total benefits, had substantial
impact, and continued to serve as a
leader and a catalyst in the global wheat
improvement system.
Our analyses could be strengthened
through use of more robust
measurements, but our study provides
valuable information on the trends of
varietal releases, varietal attributes,
adoption of improved varieties, and
the continued importance of CGIAR
wheat improvement research efforts –
a relevance confirmed by published
counterfactual scenarios. Climate
change and evolving disease spectra
call for continued investments in wheat
improvement. Food security concerns
and the limited availability of favorable
agricultural lands also call for revisiting
the potential contribution of less
favorable lands to increase production
while avoiding encroachment into forests.
In general, the CGIAR continues to
generate very high returns from its
relatively modest investment in wheat
improvement research.
Consistent and secure financial support
for international wheat improvement
research is crucial to continue to generate
these benefits, respond to emerging
threats and opportunities, and avoid
recurrences of global food crises.
ECONOMIC BENEFITS AND
RETURNS ON INVESTMENT
48
Genetic gains in yield resulting from the release of new wheat varieties over time.
APPENDICES
Table A.1. Evidence on rates of genetic gain in bread wheat yield, developing countries.
Environment/location Period Rate of gain (%/yr) Data source
Spring habit wheat, irrigated
Sonora, Mexico 1962-75
a
1.1 Fischer and Wall (1976)
1962-83
a
1.1 Waddington et al. (1986)
1962-81
a
0.9 Pat Wall, CIMMYT
b
1962-85
a
0.6 Ortiz-Monasterio et al. (1990)
1962-88
a
0.8 Sayre, Rajaram and Fischer (1997)
1988-96
a
0.8 H.J. Dubin, CIMMYT
b ,c
Nepal 1978-88
a
0.8 Morris, Dubin and Pokhrel (1992)
India 1911-54 0.6 Kulshrestha and Jain (1982)
1967-79 1.2
India 1989-99 1.9 Nagarajan (2002)
Northwest India 1985-95
a
0.9 H.J. Dubin, CIMMYT
b ,c
Pakistan 1965-82
a
0.8 Byerlee (1993)
Zimbabwe 1967-85 1.0 Mashiringwani (1987)
Semi-arid wheat, CIMMYT 1994-2010 1.0 Manes et al. (2012)
CIMMYT spring bread 1977-2008 0.7 Lopes et al. (2012)
Hot, irrigated
Sudan 1967-87 0.9 Byerlee and Moya (1993)
Rainfed
Ethiopia 1967-94 1.2-1.7 Amsal et al (1996)
Uruguay 1966-95
a
1.4 Mohan Kohli, CIMMYT
b
high fertility
1966-95
b
0.9 Mohan Kohli, CIMMYT
b
low fertility
Parana, Brazil (non-acid) 1978-94 0.9 Mohan Kohli, CIMMYT
b
Paraguay 1972-90 1.3 Mohan Kohli, CIMMYT
b
1979-92
a
1.6 Mohan Kohli, CIMMYT
b
Argentina 1912-80 0.4 Slafer and Andrade (1989)
1966-89 1.9 Byerlee and Moya (1993)
1971-89 3.6 Mohan Kohli, CIMMYT
b
unprotected
1971-89
a
2.1 Mohan Kohli, CIMMYT
b
protected
1988-97
a
3.7 Mohan Kohli, CIMMYT
b
Bolivia 1986-96
a
1.0 Mohan Kohli, CIMMYT
b
Central India 1966-91 0.27 Jain and Byerlee (1999)
Acid soils, rainfed
Rio Grande do Sul 1976-89 3.2 Byerlee and Moya (1993)
Rio Grande do Sul, Brazil 1970-90 3.6 Tomasini (2002)
Parana, Brazil 1969-89 2.2 Byerlee and Moya (1993)
1970-96
a
0.2(ns) Mohan Kohli, CIMMYT
b
Facultative/winter habit wheat
South Africa 1930-90 1.4 Van Lill and Purchase (1995)
Northern China 1960-2000 0.48-1.23 Zhou et al. (2007)
Southern China 1949-2000 0.31-0.74 Zhou et al. (2007)
a Semi-dwarfs only. b Unpublished data. c Two-variety comparison only.
Note: This table is an update of Lantican, Dubin and Morris (2005); Heisey, Lantican and Dubin (2002); Rejesus, Smale and Heisey (1999);
and Byerlee and Moya (1993).
49
Impacts of International
Wheat Improvement Research 1994- 2014
Table A.2. Evidence on rates of genetic gain in wheat yield, developed countries.
Environment/location Period Rate of gain (%/yr) Data source
Spring habit wheat, rainfed
Victoria, Australia 1850-1940 0.3 O’Brien (1982)
1940-81 0.8
New South Wales, Australia 1956-84 0.9 Anthony and Brennan (1987)
Western Australia 1884-1982 0.4 Perry and D’Antuono (1989)
(low rainfall)
Hard red spring wheat (CWRS)
Western Canada prior to 1990 0.35 Thomas and Graf (2014)
early 1990s-2013 0.67
Facultative/winter habit wheat, rainfed
Kansas (hard red winter) 1932-69 0.6 Feyerherm and Paulsen (1981)
1971-77 0.8 Feyerherm, Paulsen & Sebaugh (1984)
1874-1970 0.4 Cox et al. (1988)
1976-87 1.2
Oklahoma/Texas (hard red winter) 1932-74 0.8 Feyerherm and Paulsen (1981)
U.S. corn belt winter (soft/hard) 1968-76 1.7 Feyerherm,Paulsen and Sebaugh (1984)
U.S. winter 1958-78 0.7-1.4 Schmidt (1984)
(various regional performance nurseries)
U.S. Great Plains 1996-97 and 1998-99
Older genotypes 0.16 Donmez et al. (2001)
New genotypes 0.63
US Southern region (SRPN) 1959-2008
All entries 1.1 Graybosch and Peterson (2010)
Only most productive entry 1.3
US Northern region (NRPN) 1959-2008
All entries 0.79 Graybosch and Peterson (2010)
Most productive entry 0.79
U.K. 1947-77 1.5 Silvey (1978)
Sweden 1900-1976 0.2 Ledent and Stoy (1988)
Spain (contrasting environments) 1930-2000 0.88 Sanchez-Garcia et al. (2013)
Note: This table is an update of Lantican, Dubin and Morris (2005); Heisey, Lantican and Dubin (2002); Rejesus, Smale and Heisey (1999);
and Byerlee and Moya (1993).
50
Figure A.2. Global wheat sites with leaf rust as a production constraint.
Figure A.1. Global wheat sites with stem rust as a production constraint.
Production constraints per key wheat site as viewed by respondents are summarized,
mapped and presented in Figures A.1 to A.7. Note that no responses on constraints
were received from Australia, Canada and the USA.
51
Impacts of International
Wheat Improvement Research 1994- 2014
Figure A.3. Global wheat sites with yellow rust as a production constraint.
Figure A.4. Global wheat sites with powdery mildew as a production constraint.
52
Figure A.5. Global wheat sites with drought and heat stresses as production constraints.
Figure A.6. Global wheat sites with drought stress as a production constraint.
53
Impacts of International
Wheat Improvement Research 1994- 2014
Figure A.7. Global wheat sites with heat stress as a production constraint.
54
Abdalla, O.S., J. Crossa, E. Autrique, and I.H.
de Lacy. 1996. Relationships among
international testing sites of spring
durum. Crop Sci. 36(1): 33-40.
Ali S., P. Gladieux, M. Leconte, A. Gautier,
A.F. Justesen, M.S. Hovmøller, J.
Enjalbert, C. de Vallavieille Pope. 2014.
Origin, migration routes and worldwide
population genetic structure of the
wheat yellow rust pathogen Puccinia
striiformis f. sp. tritici. PLoS Pathog
10(1): e1003903. doi:10.1371/journal.
ppat.1003903.
Ali, M., and D. Byerlee. 1991. Economic
efficiency of small farmers in a changing
world: A survey of recent evidence.
Journal of International Development
3(1): 1-27.
Alston, J.M. 1991. Research benefits in a
multimarket setting: A review. Review of
Marketing and Agricultural Economics 59
(1): 32-52.
Alston, J.M., and P.G. Pardey. 2001. Attribution
and other problems in assessing the
returns to agricultural R&D. Agricultural
Economics 25(2-3): 141-152.
Amsal, T., G. Getinet, T. Tesfaye, and D.G.
Tanner. 1996. Effects of genetic
improvement on morpho-physiological
characters related to grain yield of bread
wheat in Ethiopia. In Woldeyesus Sinebo,
Zerihun tadele, and Nigussie Alemayehu
(eds.), Proceedings of the Annual
Agronomy and Crop Physiology Society
of Ethiopia. Addis Ababa, Ethiopia:
ACPSE.
Anderson J.A., R.W. Stack, S. Liu, B.L.
Waldron, A.D. Fjeld, C. Coyne, B.
Moreno-Sevilla, F.J. Mitchell, Q.J.
Song, P.B. Cregan, and R.C. Frohberg.
2001. DNA markers for Fusarium
head blight resistance QTLs in two
wheat populations. Theor Appl Genet
102:1164-1168.
Anthony, G., and J.P. Brennan. 1987. Progress
in yield potential and bread-making
characteristics in wheat in New South
Wales, 1925-26 to 1984-85. Agricultural
Economics Bulletin, Division of
Marketing and Economic Services, New
South Wales, Australia: Department of
Agriculture.
Austin, R.B., J. Bingham, R.D. Blackwell, L.T.
Evans, M.A. Ford, C.L. Morgan, and
M. Taylor. 1980. Genetic improvements
in winter wheat yields since 1900 and
associated physiological changes.
Journal of Agricultural Science,
Cambridge 94:675-689.
Barkley, A., J. Tack, L.L. Nalley, J. Bergtold,
R. Bowden, and A. Fritz. 2014. Weather,
disease and wheat breeding effects on
Kansas wheat varietal yields, 1985-2011.
Agronomy Journal 106(1): 227-235.
Basnet B.R., R.P. Singh, A.M.H. Ibrahim,
S.A. Herrera-Foessel, J. Huerta-
Espino, C. Lan, and J.C. Rudd. 2014.
Characterization of Yr54 and other genes
associated with adult plant resistance
to yellow rust and leaf rust in common
wheat Quaiu 3. Mol Breeding 33:385-
399.
Beddow, J.M., P.G. Pardey, Y. Chai, T.M.
Hurley, D.J. Kriticos, H.J. Braun, R.F.
Park, W.S. Cuddy, and T. Yonow. 2015.
Research investment implications of
shifts in the global geography of wheat
stripe rust. Nature Plants DOI: 10.1038/
nplants.2015.132.
Bonjean, A.P., and W.J. Angus (eds.). 2000.
The World Wheat Book. A History
of Wheat Breeding. Paris: Lavoisier
Publishing Inc.
Bonjean, A.P., W.J. Angus, M. van Ginkel
(eds.). 2011. The World Wheat Book
2. A History of Wheat Breeding. Paris:
Lavoisier Publishing.
REFERENCES AND
RECOMMENDED READING
55
Impacts of International
Wheat Improvement Research 1994- 2014
Braun H.J., S. Rajaram, and M. Van Ginkel.
1996. CIMMYT’s approach to breeding
for wide adaptation. Euphytica 92:175-
183.
Brennan, J.P., and D. Byerlee. 1991. The rate
of crop varietal replacement on farms:
measures and empirical results for
wheat. Plant Varieties and Seeds (4):
99-106.
Brennan, J.P., and K.J. Quade. 2004. Analysis
of the Impact of CIMMYT Research
on the Australian Wheat Industry.
Economic Research Report No. 25. NSW
Department of Primary Industries, Wagga
Wagga.
Byerlee, D. 1992. Technical change,
productivity, and sustainability in
irrigated cropping systems of South
Asia: Emerging issues in the post-Green
Revolution era. Journal of International
Development 4(5): 477-496.
Byerlee, D. 1993. Technical change and returns
to wheat breeding research in Pakistan’s
Punjab in the post-Green Revolution
period. The Pakistan Development
Review 32(1): 69-86.
Byerlee, D. and P.W. Heisey. 1990. Wheat
varietal diversification over time and
space as factors in yield gains and rust
resistance in the Punjab. 1990. In P.W.
Heisey (ed.) Accelerating the Transfer
of Wheat Breeding Gains to Farmers:
A Strategy of the Dynamics of Varietal
Replacement in Pakistan. CIMMYT
Research Report No. 1 Mexico, D.F.:
CIMMYT.
Byerlee, D., and G. Traxler. 1995. National
and international wheat improvement
research in the post Green Revolution
period: Evolution and impacts. American
Journal of Agricultural Economics 77:
268-278.
Byerlee, D. and P. Moya. 1993. Impacts of
International Wheat Breeding Research in
the Developing World, 1966-90. Mexico,
D.F.: CIMMYT.
Byerlee, D., and H.J. Dubin. 2010. Crop
improvement in the CGIAR as a global
success story of open access and
international collaboration. International
Journal of Commons Vol 4, No 1. DOI:
http://doi.org/10.18352/ijc.147.
Cakmak I., R.D. Graham, and R.M. Welch.
2002. Agricultural and molecular genetic
approaches to improving nutrition and
preventing micronutrient malnutrition
globally. In Encyclopedia of Life Support
Systems, Vol. 2, 1075-1099. UNESCO
publishing.
CGIAR. 2011. Forty Findings on the Impacts
of CGIAR Research 1971-2011. World
Bank, Washington, D.C.: CGIAR Fund
Office.
Cox, T.S., J.P. Shroyer, B.-H. Liu, R.G.
Sears, and T.J. Martin. 1988. Genetic
improvement in agronomic traits of hard
red winter wheat cultivars from 1919 to
1987. Crop Science 28:756-760.
Dixon, J., L. Nalley, P. Kosina, R. La Rovere,
J. Hellin, and P. Aquino. 2006. Adoption
and economic impact of improved wheat
varieties in the developing world. J Agr
Sci [Centenary Review] 144:489-502.
Donmez, E., R.G. Sears, J.P. Shroyer, and G.M.
Paulsen. 2001. Genetic gain in yield
attributes of winter wheat in the Great
Plains. Crop Science 41:1412-1419.
Dreisigacker, S., P. Zhang, M. Warburton,
M. Van Ginkel, M. Bohn, and A.E.
Melchinger. 2004. SSR and pedigree
analyses of genetic diversity among
CIMMYT wheat lines targeted to different
mega-environments. Crop Science
44:381-388.
Dreisigacker, S., M. Kishii, J. Lage, and M.
Warburton. 2008. Use of synthetic
hexaploid wheat to increase diversity
for CIMMYT bread wheat improvement.
Australian Journal of Agricultural
Research 59:413-420.
Dreisigacker, S., H. Shewayrga, J. Crossa,
V.N. Arief, I.H. DeLacy, R.P. Singh, M.J.
Dieters and H.J. Braun. 2012. Genetic
structures of the CIMMYT international
yield trial targeted to irrigated
environments. Molecular Breeding 29(2):
529-541.
Evenson, R.E. 2000. Crop germplasm
improvement: A general perspective.
Paper presented at the annual meeting
of the American Association of the
Advancement of Science. Washington,
D.C. 21 February 2000.
Evenson, R.E., and D. Gollin. 2003. Assessing
the impact of Green Revolution 1960-
2000. Science 300:750-762.
Evenson, R.E., and M. Rosegrant. 2003.
The economic consequences of crop
genetic improvement programmes. In
R.E. Evenson and D. Gollin (eds.), Crop
Variety Improvement and its Effect on
Productivity: The Impact of International
Agricultural Research. Food and
Agriculture Organization. Wallingford,
U.K.: CABI Publishing.
56
Feyerherm, A.M., and G.M. Paulsen. 1981.
An analysis of temporal and regional
variation in wheat yields. Agronomy
Journal 73:863-867.
Feyerherm, A.M., G.M. Paulsen, and J.L.
Sebaugh. 1984. Contribution of genetic
improvement to recent wheat yield
increases in the USA. Agronomy Journal
76:985-990.
Fischer, R.A., and P.C. Wall. 1976. Wheat
breeding in Mexico and yield increases.
Journal of the Australian Institute of
Agricultural Science 42:138-148.
Galushko, V., and R. Gray. 2014. Twenty-five
years of private wheat breeding in the
UK: lessons for other countries. Science
and Public Policy:1-15.
Gollin, D. 2006. Impacts of International
Research on Intertemporal Yield Stability
in Wheat and Maize: An Economic
Assessment. Mexico, D.F.: CIMMYT
Graybosch, R.A., and C.J. Peterson. 2010.
Genetic Improvement in Winter Wheat
Yields in the Great Plains of North
America, 1959-2008. USDA-Agricultural
Research Service/ University of
Nebraska-Lincoln Faculty Paper 915.
Heisey, P.W. 1990 (ed.). 1990. Accelerating
the Transfer of Wheat Breeding Gains
to Farmers: A Study of the Dynamics
of Varietal Replacement in Pakistan.
CIMMYT Research Report No. 1.
Mexico, D.F.: CIMMYT.
Heisey, P.W., and M.A. Lantican. 1999.
International wheat breeding research
in Eastern and Southern Africa. In Tenth
Regional Wheat Workshop for Eastern,
Central, and Southern Africa. Addis
Ababa, Ethiopia: CIMMYT.
Heisey, P.W., C.S. Srinivasan, and C. Thirtle.
2001. Public Sector Plant Breeding in a
Privatizing World. Agriculture Information
Bulletin No. 772. U.S. Department of
Agriculture.
Heisey, P.W. 2002. International Breeding
and Future Productivity in Developing
Countries. Wheat Yearbook/WHS-2002.
United States Department of Agriculture
– Economic Research Service
(USDA-ERS).
Heisey, P.W., M.A. Lantican, and H.J. Dubin.
2002. Impacts of International Wheat
Breeding Research in the Developing
World, 1966-97. Mexico, D.F.: CIMMYT.
Huang, J., C. Xiang, and Y. Wang. 2015. The
Impact of CIMMYT Wheat Germplasm
on Wheat Productivity in China. Mexico,
D.F.: CGIAR Research Program on
Wheat.
Jain, K.B.L., and D. Byerlee. 1999. Investment
efficiency at the national level: Wheat
improvement research in India. In M.K.
Maredia and D. Byerlee (eds.), The
Global Wheat Improvement System:
Prospects for Enhancing Efficiency in
the Presence of Spillovers. CIMMYT
Research Report No.5. Mexico, D.F.:
CIMMYT.
Jilani, A., D. Pearce, and F. Bailo. 2013. ACIAR
wheat and maize projects in Afghanistan.
ACIAR Impact Assessment Series Report
No. 85. Canberra: Australian Centre for
International Agricultural Research.
Khan, M.A. 1987. Wheat Variety Development
and Longevity of Rust Resistance.
Lahore, Pakistan: Department of
Agriculture, Government of the Punjab.
Krishna, V.V., D.J. Spielman, and P.C. Veettil.
2015. Exploring the supply and demand
factors of varietal turnover in Indian
wheat. Journal of Agricultural Science.
DOI:10.1017/S0021859615000155.
Kulshrestha, V.P., and H.K. Jain. 1982. Eighty
years of wheat breeding in India: Past
selection pressures and future prospects.
Zeitschrift Pflanzenzuchtung 89:19-30.
Lan, C., G.M. Rosewarne, R.P. Singh, S.A.
Herrera-Foessel, J.Huerta-Espino, B.R.
Basnet, Y. Zhang, and E. Yang. 2014.
QTL characterization of resistance to leaf
rust and stripe rust in the spring wheat
line Francolin#1. Mol. Breeding 34:789-
803.
Lantican, M.A., H.J. Dubin, and M.L. Morris.
2005. Impacts of International Wheat
Breeding Research in the Developing
World, 1988-2002. Mexico, D.F.:
CIMMYT.
Lantican, M.A., P. L. Pingali, and S. Rajaram.
2003. Is research on marginal lands
catching up? The case of unfavourable
wheat growing environments. Agricultural
Economics 29:353-361.
Ledent, J.F., and V. Stoy. 1988. Yield of winter
wheat, a comparison of genotypes
from 1910 to 1976. Cereal Research
Communications 16: 151-156.
Li, Z., C. Lan, Z. He, R.P. Singh, G.M.
Rosewarne, X. Chen, and X. Xia. 2014.
Overview and application of QTL for
adult plant resistance to leaf rust and
powdery mildew in wheat. Crop Science
54:1907–1925.
Lillemo M., A.K. Joshi, R. Prasad, R. Chand,
and R.P. Singh. 2013. QTL for spot
blotch resistance in bread wheat line
Saar co-locate to the biotrophic disease
resistance loci Lr34 and Lr46. Theor Appl
Genet 126:711-719.
Lopes, M.S., M.P. Reynolds, Y. Manes, R.P.
Singh, J.Crossa, and H.J. Braun. 2012.
Genetic yield gains and changes in
associated traits of CIMMYT spring
bread wheat in a “historic” set
representing 30 years of breeding. Crop
Science 52:1123-1131.
Manes, Y., H.F. Gomez, L. Puhl, M. Reynolds,
H.J. Braun, and R. Trethowan.
2012. Genetic gains of the CIMMYT
International Semi-arid Wheat Yield Trials
from 1994 to 2010. Crop Science 52:
1543-1552.
Marasas, C.N., M. Smale, and R.P. Singh.
2004. The Economic Impact in
Developing Countries of Leaf Rust
Resistance Breeding in CIMMYT-
Related Spring Bread Wheat. CIMMYT
Economics Program Paper 04-01.
Mexico, D.F.: CIMMYT.
Maredia, M.K., and D. Byerlee (eds.). 1999.
The Global Wheat Improvement System:
prospects for Enhancing Efficiency in
the Presence of Spillovers. CIMMYT
Research Report No. 5. Mexico, D.F.:
CIMMYT.
Marshall, G.R., and J.P. Brennan. 2001. Issues
in benefit-cost analysis of agricultural
research projects. The Australian Journal
of Agricultural and Resource Economics
45(2): 195-213.
Mashiringwani, N.A. 1987. Trends in
Production and Consumption of Wheat
and the Role of Variety Improvement in
Zimbabwe. Department of Research and
Specialist Services.
Mazid, A., M. Keser, K.N. Amegbeto, A.
Morgounov, A. Bagci, K. Peker, M.
Akin, M. Kucukcongar, M. Kan, A.
Semerci, S. Karabak, A. Altikat, and S.
Yaktubay. 2014. Measuring the impact
of agricultural research: The case of new
wheat varieties in Turkey. Experimental
Agriculture 51(02): 161-178.
McLaren, C.G., R. Bruskiewich, A.M. Portugal,
and A.B. Cosico. 2005. The International
Rice Information System. A Platform for
Meta-Analysis of Rice Crop Data. Plant
Physiology 139:637-642.
57
Impacts of International
Wheat Improvement Research 1994- 2014
McLaren, C.G., I. DeLacy and J. Crossa, 2007.
Routine Computation and Visualization
of Coefficients of Parentage Using the
International Crop Information System.
Technical Report.
Monasterio I. and R.D. Graham. 2000.
Breeding for trace minerals in wheat.
Food and Nutrition Bulletin 21:393-396.
Mondal S., R.P. Singh, J. Huerta-Espino,
Z. Kehel, and E. Autrique. 2015.
Characterization of heat and drought
stress tolerance in high-yielding spring
wheat. Crop Science 55:1–11.
Morris, M.L. and P.W. Heisey. 2003. Evaluating
the benefits of plant breeding research.
Methodological issues and practical
challenges. Agricultural Economics
29:241-252.
Morris, M.L., H.J. Dubin, and T. Pokhrel. 1992.
Returns to wheat research in Nepal.
CIMMYT Economics Working Paper 92-
04. Mexico, D.F.: CIMMYT.
Nagarajan, S. 2002. The significance of wheat
production for India, in particular under
limited moisture conditions. http://
www.biotech.biol.ethz.ch/India/forms/
Nagarajan.pdf.
Nelson, G.C., M.W. Rosegrant, J. Koo, R.D.
Robertson, T. Sulser, T. Zhu, C. Ringler,
S. Msangi, A. Palazzo, M. Batka, M.
Magalhaes, R. Valmonte-Santos, M.
Ewing, D.R. Lee. 2009. Climate Change:
Impact on Agriculture and Costs
of Adaptation. Food Policy Report.
Washington, D.C.: IFPRI.
O’Brien, L. 1982. Victorian wheat yield trends
1898-1977. Journal of the Australian
Institute of Agricultural Science 48:163-
68.
Ortiz-Monasterio, J.I., K. Sayre, S. Rajaram,
and M. McMahon. 1990. Genetic
progress of CIMMYT germplasm under
different levels of nitrogen. Agronomy
Abstracts 156-60.
Ortiz-Monasterio, J.I., K. Sayre, S. Rajaram,
and M. McMahon. 1997. Genetic
progress in wheat yield and nitrogen use
efficiency under four nitrogen rates. Crop
Science 37:898-904.
Pardey, P.G., J.M. Alston, J.E. Christian, and
S. Fan. 1996. Hidden Harvest: U.S.
Benefits from International Research
Aid. Food Policy Report. Washington,
D.C.: International Food Policy Research
Institute (IFPRI).
Pardey, P.G., J.M. Beddow, D.J. Kriticos,
T.M. Hurley, R.F. Park, E. Duveiller, R.W.
Sutherst, J.J. Burdon, and D. Hodson.
2013. Right-sizing stem-rust research.
Science 340:147-148.
Perry, M., and M. D’Antuono. 1989.
Yield improvement and associated
characteristics of some Australian spring
wheat cultivars introduced between
1860 and 1982. Australian Journal of
Agricultural Research 40:457-472.
Pingali, P.L. 2012. Green Revolution: Impacts,
limits, and the path ahead. Proceedings
of the National Academy of Sciences
of the United States of America (PNAS)
109(31): 12302–12308.
Pray, C.E. 1992. Plant Breeders’ Rights
Legislation, Enforcement and R&D:
Lessons for Developing Countries. In
G.H. Peters, B.F. Stanton, and G.J.
Tyler (eds.), Sustainable Agricultural
Development: The Role of international
Cooperation, Proceedings of the
Twenty-first International Conference of
Agricultural Economists, Tokyo, Japan,
22-29 August 1991,330-342. Brookfield,
Vermont: Dartmouth.
Qin, X., F. Zhang, C. Liu, H. Yu, B. Cao, S.
Tian, Y. Liao, and K.H.M. Siddique. 2015.
Wheat yield environments in China: Past
trends and future directions. Field Crops
Research 177:117-124.
Rajaram, R., M. van Ginkel, and R.A. Fischer.
1994. CIMMYT’s wheat breeding mega-
environments (ME). In Proceedings of
the 8th International Wheat Genetics
Symposium, July 19-24, Beijing, China.
1101-1106.
Reif, J.C., P. Zhang, S. Dreisigacker, M.L.
Warburton, M. van Ginkel, D. Hoisington,
M. Bohn, and A.E. Melchinger. 2005.
Wheat genetic diversity trends during
domestication and breeding. Theoretical
and Applied Genetics 110:859–864.
Rejesus, R.M., P.W. Heisey and M.Smale.
1999. Sources of Productivity Growth in
Wheat: A Review of Recent Performance
and Medium to Long-term Prospects.
CIMMYT Economics Working Paper 99-
05. Mexico, D.F.: CIMMYT.
Renkow, M., and D. Byerlee. 2010. The
impacts of CGIAR research: A review
of recent evidence. Food Policy35(5):
391-402.
Reynolds, M.P. and N.E. Borlaug. 2006.
Impacts of breeding on international
58
collaborative wheat improvement.
Journal of Agricultural Science 144:3-13.
Rosegrant, M.W., and S.A. Cline. 2003. Global
food security: challenges and policies.
Science 302:1917-1919.
Rosegrant, M.W., S. Tokgoz, and P. Bhandary.
2012. The new normal? A tighter global
agricultural supply and demand relation
and its implications for food security.
Amer J Agr Econ 95(2): 303-309.
Rosewarne G.M, S.A. Herrera-Foessel, R.P.
Singh, J.HuertaEspino, X.C. Lan,
and Z.H. He. 2013. Quantitative trait
loci of stripe rust resistance in wheat.
Theoretical and Applied Genetics
126:2427–2449.
Rubenstein, K.D., P. Heisey, R. Shoemaker,
J. Sullivan, and G. Frisvold. 2005. Crop
Genetic Resources: An Economic
Appraisal. Economic Information Bulletin
No. 2: USDA.
Sanchez-Garcia, M., C. Royo, N. Aparicio, J.A.
Martin-Sanchez, F. Alvaro. 2013. Genetic
improvement of bread wheat and
associated traits in Spain during the 20th
century. J Agric Sci 151(1): 105-118.
Sayre, K.D., S. Rajaram, and R.A. Fischer.
1997. Yield potential progress in short
bread wheats in northwest Mexico. Crop
Science 37(1): 36-42.
Sayre, K.D., R.P. Singh, J. Huerta-Espino, and
S. Rajaram. 1998. Genetic progress in
reducing losses to leaf rust in CIMMYT-
derived Mexican spring wheat cultivars.
Crop Science 38:654-659.
Schmidt, J.W. 1984. Genetic contributions to
yield gains in wheat. In W.R. Fehr (ed.),
Proceedings of a Symposium on Genetic
Contributions to Yield Gains of Five Major
Crop Plants. Sponsored by C-1, Crop
Science Society of America, Atlanta,
Georgia, 2 December 1981. Madison,
Wisconsin: Crop Science Society of
America and American Society of
Agronomy.
Sharma, R.C., J. Crossa, G. Velu, J. Huerta-
Espino, M. Vargas, T.S. Payne, and R.P.
Singh. 2012. Genetic gains for grain yield
in CIMMYT spring bread wheat across
international environments. Crop Science
52: 1522-1533.
Shiferaw, B., M. Kassie, M. Jaleta and C.
Yirga. 2014. Adoption of improved wheat
varieties and impacts on household food
security in Ethiopia. Food Policy
44:272-284.
Silvey, V. 1978. The contribution of new
varieties to increasing cereal (wheat
and barley) yields in England and
Wales. Journal of National Institute of
Agricultural Botany 14(3): 367-384.
Singh R.P., and J. Huerta-Espino. 2004. The
use of ‘single-backcross, selected-bulk’
breeding approach for transferring minor
genes based rust resistance into adapted
cultivars. In C.K. Black, J.F. Panozzo
and G.J. Rebetzke (eds.), Proc. of 54th
Australian Cereal Chemistry Conference
and 11th Wheat Breeders Assembly, 21-
24 September 2004, Canberra, Australia,
48-51. Melbourne: Cereal Chemistry
Division, Royal Australian Chemical
Institute.
Singh R.P., S. Rajaram, A. Miranda, J.
Huerta-Espino and E. Autrique. 1998.
Comparison of two crossing and four
selection schemes for yield, yield traits,
and slow rusting resistance to leaf rust in
wheat. Euphytica 100:35-43.
Singh, R.P., and S. Rajaram. 2002. Breeding
for disease resistance in wheat. In
B.C. Curtis, S. Rajaram, and H. Gomez
Macpherson (eds.), Bread Wheat
Improvement and Production, FAO Plant
Production and Protection Series No. 30,
141-156. Rome: FAO.
Slafer, G.A., and F.H. Andrade. 1989. Genetic
improvement in bread wheat (Triticum
aestivum) yields in Argentina. Field Crops
Research 21:289-296.
Smale, M., M.P. Reynolds, M.L. Warburton,
R. Trethowan, R.P. Singh, I. Ortiz-
Monasterio, J.Crossa, M. Khairallah, and
M.I. Almanza-Pinzon. 2001. Dimensions
of Diversity in CIMMYT Bread Wheat
from 1965 to 2000 Mexico, D.F.:
CIMMYT.
Smit, H.A., V.L. Tolmay, A. Barnard, J.P.
Jordaan, F.P. Koekemoer, W.M. Otto,
Z.A. Pretorius, J.L. Purchase, and J.P.C.
Tolmay. 2008. An overview of the context
and scope of wheat (Triticum aestivum)
research in South Africa from 1983 to
2008. S Afr J Plant & Soil 27(1): 1983-
2008.
Stevenson, J.R., N. Villoria, D. Byerlee, T.
Kelley, and M. Maredia. 2013. Green
Revolution research saved an estimated
18 to 27 million hectares from being
brought into agricultural production.
Proceedings of the National Academy of
Sciences of the United States of America
(PNAS). Vol. 110 No. 21.
Thomas, J.B. and R.J. Graf. 2014. Rates of
yield gain of hard red spring wheat in
59
Impacts of International
Wheat Improvement Research 1994- 2014
western Canada. Can J Plant Sci 94:1-
13.
Tomasini, R.G.A. 2002. Impact of Mexican
Germplasm on Brazilian Wheat Cropping:
An Ex-Post Economic Analysis. CIMMYT