Bioversity International scientist Carlo Fadda presents to the World Bank on the results we have had so far working with partners in Ethiopia to tap into the genetic diversity of the country and the knowledge of farmers, to help them adapt better to climate change.
Find out more about Seeds for Needs: www.bioversityinternational.org/research-portfolio/adaptation-to-climate-change/seeds-for-needs/
Scaling up Ethiopia’s ‘Seeds for Needs’ approach of using agricultural biodiversity to adapt to climate change
1. Scaling up Ethiopia’s ‘Seeds for Needs’ approach of using agro-biodiversity to
adapt to climate change
Carlo Fadda, Senior Scientist, Bioversity International
World Bank, Addis Ababa, 15/01/2015
6. •Climate change, floods, droughts,
unpredictable temperatures and
rainfall
•Changing pest and pathogen
populations and levels of pollination
efficiency
•Increased soil and land degradation
Major Environmental Threats to Sustainable Production
7. Adaptation to climate change: recommended actions
by the IPCC
Improving crop tolerance to high temperature is a frequently
identified adaptation for almost all crops and environments
worldwide as high temperatures are known to reduce both
yield and quality
8. Adaptation to climate change: recommended actions
by the IPCC
Improving gene conservation and access to extensive gene
banks could facilitate the development of variety with
appropriate thermal time and thermal tolerance characteristics
9. Adaptation to climate change: recommended actions
by the IPCC
Indigenous Knowledge (IK) has developed to cope with
climate hazards contributing to food security in many parts of
the world
10. Ethiopian Green Economy Strategy
Target set for agriculture:
• Boost agricultural productivity (+40% increased production for
major crops such as teff, maize, wheat);
• Intensify agriculture through usage of improved inputs and better
residue management resulting in a decreased requirement for
additional agricultural land that would primarily be taken from
forests;
• Create new agricultural land in degraded areas through small-,
medium-, and large-scale irrigation to reduce the pressure on
forests if expansion of the cultivated area becomes necessary
• Introduce into cultivated areas lower-emission agricultural
techniques, ranging from the use of carbon- and nitrogen-efficient
crop cultivars to the promotion of organic fertilizers
11. Under conditions of change
(reducing the probability of loss of agricultural
productivity in the future, while enhance productivity
today)
The fundamental question:
Productivity and reduced vulnerability
How can we ensure that agricultural productivity increases are
accomplished in ways that create and enhance ecosystem
resilience and services for the poor?
12. Minimum Goals for 2050
Environmental GoalsDevelopment Goals
Total Agricultural Production
Nutritionally Complete Production
Biodiversity Conserved
Carbon Sequestered
Food Security Goals
Increased Farmer Livelihoods
And Resilience
Increase Farm Self Reliance
Adapted from Foley et al 2011
The Case of Durum Wheat
Improve Human Health
Food Distribution and Access
Conserve agrobiodiversity
Water Conserved
Improved Water Quality
Soil Formed
13. Unexploited Potentials in Landraces
• disease resistance (Leppik,
1970; Negassa, 1986; Klindworth
et al., 2007; Jemanesh et al.,
2013);
– e.g. Ethiopian landraces are the
source for Sr13 gene, which is
responsible for stem rust resistance
• Drought tolerance/resistance
(Tesfaye, 2001; Mondini et al.,
2010, our study)
• Very diverse for qualitative and
quantitative traits
15. Phenotyping trials
The genotypes, 373 landraces and 27 improved wheat varieties, were phenotyped at
two locations (Hagreselam and Geregera) in 2012 and 2013 main cropping seasons for 10
important traits:
A. Phenological traits
• days to 50% booting (DB);
• days to 50% flowering (DF) and;
• days to maturity (DM)
B. Morpho – agronomic traits
• plant height (PH);
• number of effective tillers per plant (NET);
• spike length (SPL);
• number of seeds per spike (SPS);
• above ground dry biomass (BY);
• grain yield (GY).
Phenotyping (external characteristics)
16. 0
10
20
30
40
50
60
70
80
<µ - 2SD µ - 2SD to µ -
SD
µ - SD to µ +
SD
µ + SD to µ
+2SD
>µ +2SD
DISTRIBUTION OF
GENOTYPES INTO
VARIOUS CLASSES
HS GER Com
Performance of Genotypes Across Locations
17. Landraces Performance Compared With the Best
Improved Variety
The table tells that:
•21%, averaged over traits, of
the landraces are superior to
the best performer IM variety
•Many landraces mature
earlier than the IM varieties
•A yield advantage of 61%
obtained from the best
landrace over the best IM
variety (Robe)
Trait
Superior
(IM)
Superior
(LRs) no‡ %age
No
Geregera
%
Geregera
DB* 59.69 55.54 1 0.3 1 0.3
DF* 70.8 69.88 1 0.3 5 1.6
DM* 116.59 109.34 57 18.4 71 23.0
PH 110.34 115.07 8 2.6 5 1.6
NET 7.14 7.48 90 29.1 48 15.5
SPL 7.94 9.5 125 40.5 19 6.1
SPS 41.67 41.83 1 0.3 2 0.6
BY 7.17 9.99 97 31.4 47 15.2
GY 2.17 3.49 68 23.9 22 7.1
20. Crop Improvement
Grain Yield as
quantitative trait in
Hagereselam 2012
Plot overall
performance in
Hagereselam 2012
21. Modified from Yu et al. 2008
Principal Facts
• 52 RIL families
• 180 – 200 lines
• > 9,000 F6 lines in
Dec. 2014
• wide phenotypic
variation
Development of a Structured Multiparental Population
Nested Association Mapping - NAM-Population
26. (1)
Genetic
diversity
(2)
Selection &
cultivation
(3)
Harvest
(4)
Value addition
(5)
Marketing
(6)
Final
use
Outcomes
Empowerment of communities: more
resilient to eco-socio-economic changes,
more resilient food systems
Outcome
Preservation of options
for resilient systems
Outcome
Self-reliance of value chain
actors on broader set of
options, making them more
resilient to market changes
From Farm to Fork: Biodiversity Contribution along
the Value Chain
IMPACT
Improved
nutrition,
incomes and
other
livelihood
benefits
27. 3. Farmers test
and report back by
mobile phone
3. Environmental data
(GPS, sensors) to assess
adaptation4. Data are used
to detect
demand for new
varieties and
traits
4. Farmers receive tailored variety
recommendations and can order seeds
The process
2. Each farmer gets a different
combination of varieties
1. A broad set of varieties
is evaluated
28. Participatory Evaluation
• 30 farmers per location (15
male + 15 female)
• Individual score on 5 traits for
800 plots
• > 200,000 data points
30. 3. Farmers test
and report back by
mobile phone
3. Environmental data
(GPS, sensors) to assess
adaptation
1. A broad set of varieties
is evaluated
4. Data are used
to detect
demand for new
varieties and
traits
4. Farmers receive tailored variety
recommendations and can order seeds
The process
2. Each farmer gets a different
combination of varieties
32. • 32 genotypes (pre-selected by farmers)
• 24 farmers
• 4 genotype per farmer
• 3 times replication of each genotype
Mother and Baby Trial Approach
32
Monthly report by a group of farmers Method of communication with farmers
I-button in each farmer plot
33. Crowdsourcing
• 4 genotypes per farmer
• All package was sent to each
farmer
• 200 farmers, 12 different villages
• 21 genotypes, 1 common for all 200
farmers
• 30 times replication of each genotype
37. 2. Each farmer gets a different
combination of varieties
1. A broad set of varieties
is evaluated
4. Data are used
to detect
demand for new
varieties and
traits
4. Farmers receive tailored variety
recommendations and can order seeds
The process
3. Farmers test
and report back by
mobile phone
3. Environmental data
(GPS, sensors) to assess
adaptation
40. 3. Farmers test
and report back by
mobile phone
2. Each farmer gets a different
combination of varieties
3. Environmental data
(GPS, sensors) to assess
adaptation
1. A broad set of varieties
is evaluated
4. Data are used
to detect
demand for new
varieties and
traits
4. Farmers receive tailored variety
recommendations and can order seeds
The process
42. Upscaling and Outscaling Seeds for Needs
Reaching more farmers and for more crops
• Capacity development
• Approach institutionally embedded in extension services and agro-
dealer networks
• Methodology improved and expanded using ICT-based solutions
43. Crowdsourcing plan
• Initial investment for a new crop such as teff, sorghum, pulses,
targeting 500 farmers/site for 2 years
• Crop technical characterization
• Participatory evaluation
• Capacity development
• Crowdsourcing
• Subsequent distribution through crowdsourcing: targeting
10,000,000 HH over 3-4 years including seed multiplication
45. To strengthen the seed network one needs more than one
seed bank/landscape (roughly 10 to reach 10,000
households).
Goal: to reach 200,000 households in 20 landscapes
across the country
Notas do Editor
PARTNER LOGOS SLIDE
If you have 4 or more logos, place them on a separate slide like this.
Projected rainfall change over sub-Saharan Africa in the mid- and late 21st century is uncertain. In regions of high or complex topography such as the Ethiopian Highlands, downscaled projections indicate likely increases in rainfall and extreme rainfall by the end of the 21st century.
Climate model projections under the SRES A2 and B1 scenarios over Ethiopia show warming in all four seasons across the country, which may cause a higher frequency of heat waves as well as higher rates of evaporation (Conway and Schipper, 2011).
Changes in precipitation are difficult to predict: downward trends in some part of the country and upwards in others with increased inter-annual variability.
Despite the fact that durum wheat was broadly grown in Ethiopia until the 60’s it has been now largely replaced by bread wheat and now it only contributes to 20% of Ethiopian wheat production and mainly in marginal areas. Yet, Ethiopia importes most of its durum wheat from abroad
Lack of awareness on the importance (i.e. commercial use) of durum wheat at farm level for bread , drinks and other local products
Heavy dependence of pasta making industries on imported durum wheat than promoting the within country production despite potential alternatives for high value locally produced pasta
Landraces is what farmers developed and varieties are developed by breeders
A) Phenotypic data
The collected phenotypic data of each trait was analyzed for:
Variability (ANOVA)
Heritability (h2)
Genotype by traits relationship (GGE)
B) Genetic data
The SNP data was analyzed for:
Relatedness (phylogeny, PCA, Structure)
Marker – traits association (MTAs)
The figure depicts:
The majority, about 60%, of the genotypes have average performance in both locations for all traits
About 20% of the tested genotypes have superior performance in both locations
On the other hand, nearly 20% of the genotypes showed inferior performance.
This performance distribution assumed normality which is statistically expected.
Mediterranean groups
The improved Ethiopian durum wheat varieties clustered together with the Mediterranean groups
This implies that improved durum wheat varieties cultivated in Ethiopia are genetically very distant from the diverse landraces
Ethiopian durum wheat landraces much diverse from both Ethiopian improved durum wheat varieties and Mediterranean durum wheat groups.
The majority of improved durum wheat varieties cultivated in Ethiopia are exotic introduction.
The landraces are genetically diverse for phenological and agronomic traits. So far, the breeding program of Ethiopia overlooked the use of these diverse landraces but the result urges revising the ignored resources to produce adaptable durum wheat varieties. Re-thinking on the breeding strategies is needed!!
Landraces collected from very distant locations are found more closer genetically than those from the same area of collection. This implies that grouping on the basis of their geography of collection does not necessarily indicate that the landraces from different areas are genetically different.
Molecular markers, SNPs, revealed the diversity of the landraces for the studied traits.
Also focus on crops holding comparative advantages to grow in marginal areas, where poor communities are most challenged- selection based also on nutritional benefits.
Many of the so called underutilized species hold great promises in terms of nutrition security, adaptation and income- many other non material benefits also related to their continued use..
4 predetermined traits and 1 general scoring
Earliness
Tillering capacity
Spike quality
Disease tolerance
This is a much faster process than conventional breeding, can be part of variety release strategy (it has been tested in Nicaragua with some success) and can be used to more decentralized participatory breeding schemes.
Yet, one can think at strengthening all sort of intermediate seed systems, including local cooperatives of seed producers (ATA, ISSD, GIZ)