Reflections on a “systems approach” for Drylands CRP-Brian Keating

1. Reflections on a “systems
approach” for Drylands CRP
Brian Keating and ISAC colleagues

2. “…in spite of fashionable lip-service to systems ideas,
and in spite of frequent exhortations to use a systems approach,
we are rarely told what it consists of, or exactly how we might use it.
There has been a notable lack of determined persistent efforts, first to
define what ‘a systems approach’ means and then to go out and use it in
tackling problems,
in order to experience that interaction between theory
and practice which is the best recipe for intellectual
progress.”
Peter Checkland (1981) Systems Thinking, Systems Practice.
This topic has a long and deep literature ....
(often outside of agriculture)

3. Models and on-farm participative research
• Agriculture is part of a “human activity system” with
production and management elements
• Agriculture is only part of a systems approach to food
and nutritional security ad poverty reduction
• Our systems approach should be focused on “problem
solving” at the science-practice interface
• Innovation in agri-food systems needs more than
research.
Four observations
Self-evident

4. Systems focus
Farm System
O
U
T
P
U
T
S
CropsSoil
Animals
I
N
P
U
T
S
CLIMATE
PRODUCTION SYSTEM
MARKETS
MANAGEMENT SYSTEM
Monitoring
Deliberative planning & control
Action
Internal Environment
Needs, values, goals, etc
Knowledge, cognitive limits, etc.
Resources
SURFACE AND
GROUND
WATER
Adapted from Sorensen &
Kristensen, 1992
Policy
environment

5. Hierarchy of scales and multiple drivers of change
Herrero et al, Science (2010)

6. Models and on-farm participative research
• Agriculture is part of a “human activity system” with
production and management elements
• Agriculture is only part of a systems approach to food
and nutritional security and poverty reduction
• Our systems approach should be focused on “problem
solving” at the science-practice interface
• Innovation in agri-food systems needs more than
research.
Four observations

7. AGRICULTURE - Sustainable Productivity Improvement
Innovation
research &
extension
Infrastructure
Finance &
Responsible
Investment
Human
Resources
Markets &
Trade
Resource
Tenure
Enabling Policies, Regulations & Institutions
Supporting Private Sector and Civil Society Engagement and Investment
International Leadership
& Coordination
Global Public GoodsDomestic Policy Reform
Understanding Risks and Opportunities – Foresight and Scenario Analysis
Transforming Small Scale
Agriculture / Agribusiness
Food & Nutrition Security
(availability, access, utilization, stability)
Inclusive Growth & Jobs for
rural women, men and youth
Growth, Jobs
and Resilience
Agriculture and Food Sector
Growth and Efficiency
Food
Availability
& Utilisation
Food
Affordability
Food
Availability
Nutrition
Interventions
Food
Utilisation
Human Productivity
Security and stability
Food Affordability
Social
Protection
Food
Affordability

8. Models and on-farm participative research
• Agriculture is part of a “human activity system” with
production and management elements and a wider context
• Agriculture is only part of a systems approach to food and
nutritional security and poverty reduction
• Our systems approach should be focused on “problem
solving” at the science-practice interface (Impact focused)
• Innovation in agri-food systems needs more than research.
Four opening observations

9. Systems thinking - systems practice
– Creation of knowledge relevant to system design
and management
• --- but there has always been the problem of “adoption”.
– Use of scientific knowledge in intervention in
system owners’ design and management
• This is the essence of “systems PRACTICE”.
• Embedded in a strong “problem solving” paradigm
Peter Checkland (1981) .

10. increasing ‘integration’
Jackson (2000) Systems Approaches to Management
The Systems Movement
Systems thinking in
the disciplines Study of systems in
their own right
Systems thinking for
“problem solving”
(Multidisciplinary)
(in practice)
organisational
integration
interactive science -
practice integration
Systems thinking is integrative
(cf. analytical thinking)

11. Models and on-farm participative research
• Agriculture is part of a “human activity system” with
production and management elements
• Agriculture is only part of a systems approach to food
and nutritional security and poverty reduction
• Our systems approach should be focused on “problem
solving” at the science-practice interface
• Innovation in agri-food systems needs more than just
research.
Four opening observations

12. What is an Innovation System?
= The conditions that are needed to enable innovation.
Definition: A network of organizations, enterprises, and individuals focused on bringing
new products, new processes, and new forms of organization into economic and social
use, together with the practices or institutions and policies that affect their behavior and
performance.

13. A dynamic view of “Innovation systems”
Adapted from A. Hall (2012) Partnerships in agricultural innovation - Who puts them together and are they
enough? In OECD Conference on Improving Agricultural Knowledge and Innovation systems
Technology
triggers
Market triggers
Social triggers
Environmental
triggers
Research
Organisations
Enterprises
Support
Organisations
Markets and
Consumers
“Go-between”
Organisations
Protocols
Enabling Policy
Environment
Innovations
of
economic,
environmen
tal or social
significance
New capacity to
innovate

14. Long history of systems research
methods

15. A Research Typology
(Oquist, 1978)
Acta Sociologica l21, 143-163.
• Descriptive research
– e.g. broaden the knowledge of the production and
management system, characterise system resources
• Predictive (nomothetic) research
– e.g., Test understanding by developing predictive and
generalisable models of a system
• Prescriptive (policy / operations) research
– e.g., Use data or models to identify optimal strategies for
desired outcome
• Participatory action research
– e.g., learn via inquiry within the life experience of
participants
Each type assumes and builds on the prior type
How the
world works!
How to
change the
world !
In Drylands we are going to need all four types of
research – in the journey from Discovery through
Diagnosis to Pilot and Scale Out.
This journey will not necessarily be linear

16. Methodological
innovation – FSR
foundation
From Dillon and Anderson (1984), after Collinson (1982)

17. Linking Operations Research to FSR
In the late 80’s and
early 90’s, McCown
and colleagues
combined the
“simulation
modelling of
agricultural systems
with the client-
orientation of FSR”

18. Action Learning
IntentLearning
Action
Knowledge When the main intent is deliberative
“changing of a reality” (action), the learning is
action learning.
Plan
ActReflect
Observe
Action
Learning Cycle

19. An elaborated view of Farming
Systems Research (FSR)
A Framework for
intervention that
substitutes a
production
systems model for
the actual system
in facilitated
action learning
(after McCown,
2008)

20. Models and on-farm participative research
• On-farm
==> Relevance
• Participative
==> Ownership and relevance
• Systems Analysis (incl.
Models)
==> Explanation and generality
• Generality
– Extrapolation in time (over
variable seasons)
– Extrapolation in space (other soils,
climates, livelihood circumstances)
Models and on-farm participative
research

21. A crowded history of research for development
approaches
1980’s
1990’s
2000’s
1960’s
1970’s
On Station
Research
Extension based
technology transfer
NIE
FSR
OFR
RRA
PRA
PAR
PTD
FTR
FFS
PPB
AKIS
PI&D
IRD
BB’s
MBTs
SRLsFARMSCAPE
INRM
IGNRM
IAR4D
IS
IP
ERI
CASE
PLAR
RDs
CCNR
AR
ARD
FS

22. What trends can we observe ?
• Moving from descriptive to
predictive/diagnostic approaches including the
use of systems analysis and modelling tools
• Increasing participation from a broader range
of actors
• Emergence of a value chain focus to
complement an on-farm focus
• Increasing recognition of the significance of
enabling institutions and governance
• Contested paradigms; hard systems vs soft
systems; positivism vs constructivism;
researcher knowledge / farmer knowledge
• Greater recognition of social equity and gender
issues

23. Some propositions for Drylands to
consider in shaping a “systems
approach”

24. 1. A systems approach shaped by
problem solving “in practice”
• A “systems approach” that is best defined in
terms of the outcomes we seek.
• That is, it is a “whatever it takes” approach to improving
food security, reducing poverty and enhancing resilience in
the world’s drylands.
• Our approach does not prejudge the need for a particular
technology, a particular commodity-related intervention or
a particularly disciplinary consideration.
• Approach draws upon diverse sources of scientific and local
knowledge to improve the food security and livelihoods of
the dryland peoples.
Systems research at the scale of impact

25. 2. Agricultural Livelihood System
• The primary focus for our systems approach
(level n) will be the “agricultural livelihood
system”.
• That is the set of farm, farming and human activity
systems that determine the livelihood opportunities for
agricultural households, enterprises or communities.
• Implicit in this focus is consideration of the food and
nutritional security, health and well being, employment
and income generation of dryland peoples.

26. 3. Systems Context
• Our systems context (n+1) is the wider
environmental and institutional setting
• Including government policy, business activity, input
and output markets, value chains, knowledge systems,
social and cultural norms, gender bias etc.
• We consider this wider context to be the “innovation
system” and we recognise scientific research is only one
part of the innovation process, albeit a potentially
catalytic or transformational part

27. 4. Science based diagnosis and
intervention design
• Our explanatory insight (n-1) comes from our
descriptive and predictive capacity around the key
components and the many interactions that shape
agricultural livelihoods.
• Components include but are not limited to crop, livestock and tree
options and technologies within farming systems, agricultural inputs
and output availability and prices, natural resources used in farming
in particular soil fertility and water management, tillage systems,
energy systems, labour and capital, nutrition and health
consequences of diets, education systems and off-farm income
generation …..
• We can’t discard our scientific method/value-add in our efforts to
get more participative and relevant

28. 5. An evolving research methodology
• Diagnosis of constraints and opportunities at the
agricultural livelihood level will be our primary
entry point for “discovery” science in dryland
systems.
• These will be holistically analysed for development
constraints in order to identify the system bottlenecks and
effective remedies.
• For the latter we will draw upon indigenous knowledge as
well as technological discoveries and developments from
other CRPs and the wider agricultural R&D system.

29. 6. “Fit for purpose” participative approaches
• Efforts to simulate desired change supported by appropriate
engagement/innovation brokering at appropriate scales with
appropriate actors (eg. farmers, community groups, value
chain and market participants, private sector investors,
government policy etc.)
- Not a one size fits all …
• Research contribution will be always informed by a solid
scientific base, including efforts to interpret system
functionality and generalize interventions to other times and
places
• National and regional institutions and development partners
will be drawn in at the outset and the “scale out” objective
adaptively planned as a “research in development” activity

30. 7. Cross-cutting research methods and
capabilities are needed
• Spatial information systems
• Data acquisition and management (includes household survey methods and
human research ethics)
• Farming systems modelling (development, validation, deployment in diagnosis and
participative design )
• Bio-economic modeling / agent based socio-ecological modeling
(Households/communities)
• Value chain and business systems analysis
• Building gender considerations into research for/in development
• Global and regional change scenarios (links to CCAFS ?)
• Capacity building in participative process/ knowledge brokering – Innovation
Platforms, Hubs, Facilities, PPPs etc
Get serious with SRT 1 and 4

31. Thankyou

32. Modelling and systems analysis tools

33. Integrated Modelling Methodology
Systems
characterisation
soil
ANIMAL
crops
TPS
herd
livestock
Biological simulation
system
Databases
limits
constraints
resources
mgmt practices
Multiple criteria
LP models
DYNAFEED
Sustainable
resource
management
strategies
Socio-economic
I/O
Herrero et al. 1996, 1997

34. NUANCES-FARMSIM
Van Wijk et al., 2009, Ag. Systems

35. Integrated analysis tool (IAT)
Crop, forage yield
Feasible/most profitable strategy
Livestock
Model
Economic
Model
Inputs
Climate
Soil
Management
Prices
Costs
Labour
Machinery
Outputs
Crops
Forage
Cattle
Labour
Income
APSIM
Crop-Forage
Model
Herd structure
& Management
Livestock
yield
Forage
yield
Household Modelling Workshop Dockside 19th-20th November 2013 Slide 13

36. Modelling alternative household
resource allocations
Baseline
household survey
N>500
APSFarm-LivSimRelevant
interventions
Better informed discussions,
investments & decisions
Profits Risks SustainabilityFood security
SIMLESA program
http://simlesa.cimmyt.org/
•Stubble management
•N fertilisation
Courtesy of D. Rodriguez

37. • Longitudinal data
• Participatory methods
• Key informants
• Systems’ classification
• Selection of farms
• Household modeling
• Sensitivity analyses
• Participatory appraisals
• Recommendation domains
• Toolboxes of interventions
• Farmers / NARS
• Stakeholder workshops
• Participatory appraisals
Participatory modelling
Ecoregion
Farms
CBA
Case studies
Range of interventions to
test for each system
(filtering)
Scenario formulation
(Farm and policy level)
Selection of a fewer
range of options
Site targeting
Dissemination &
implementation
Policy-making
Testing
options in the
field
(Herrero, 1999)

38. Baseline data
Data collection protocol :
– Climate
– Family structure
– Land management
– Livestock
management
– Labour allocation
– Family’s dietary
pattern
– Farm’s sales and
expenses

39. Systems modelling approaches
Class of model An example of model application Model Examples
Gene to
Phenotype
- more efficient crop improvement
programs
QTL based models (e.g.Yin
et al., 2004)
Crop-Soil - identification of optimal agronomic
practices and/or varietal adaptation
Simulation Models (CERES,
GRO models)
Farming System - farming systems design within soil,
climate and management constraints
Systems Simulation models
(e.g., APSIM, DSSAT,
CROPSYS)
Farm Household
/ Enterprise
- optimal resource allocation (inputs,
labour, enterprise mix) to raise farm
productivity
Bio-economic optimisation
models (e.g. MIDAS,
MUDAS);
NUANCES
Regional
Dynamics
- cross-sectoral responses to change
drivers and intervention strategies
CGE Models, Regional
Stocks and Flows, Agent-
based models
National
Economy
- impacts of interventions in the
agricultural sector on economic growth,
employment, balance of trade etc
National CGE models (e.g.
MONASH) or stocks and
flows models (e.g. ASFF)
Earth Systems/
Global Economy
and Trade
- climate change impacts on global food
supply and agricultural trade
Global Trade models (e.g.
GTAP, IMPACT, IFPRI
models)

40. Global Scenarios
Regional Scenarios
Farmer/village
perspectives
Action research
Participatory
scenario building
Global visioning
activities
Global impacts
modelling
Regional impacts
modelling
Household &
community
impacts modelling
Linking research at different levels
Thornton et al 2012

41. A more integrative approach in the
CGIAR ?
From Stripe Review – NRM Research in CGIAR
http://www.sciencecouncil.cgiar.org/fileadmin/templates/ispc/HOME/NRM_StripeReport_Proof4_WEB.pdf
Natural Resource
Management
(incl. agronomy)
Genetic and
other
technologies
Enabling Institutions
Development
Impact