Homologue areas of Quesungual pilot sites

1
Homologue areas of Quesungual pilot sites
2
Dominios de Extrapolación(Extrapolation Domains)

• Jorge Rubiano Victor Soto

3
(No Transcript)
4
Spatial Extrapolation Domains Potential areas to
scale up CPWF research projects
5
Objective

• To identify places with similar characteristics
  (social and biophysical) to those of pilot
  research sites of CPWF Projects in which outputs
  are highly probably to replicate

6
Rationale (1)

• We describe an extrapolation domain as the
  geographical area that is likely to behave in the
  same way as a project site, based on evidence of
  its physical and social attributes. Similarity
  is defined by the probability that a search site
  has the same value as the project site in a
  limited number of attributes that are defined by
  project specialists as being important.
  Attributes commonly used include climate, soil,
  land use, socio-economic characteristics, etc.

7
Rationale (1)

• We describe an extrapolation domain as the
  geographical area that is likely to behave in the
  same way as a project site, based on evidence of
  its physical and social attributes. Similarity
  is defined by the probability that a search site
  has the same value as the project site in a
  limited number of attributes that are defined by
  project specialists as being important.
  Attributes commonly used include climate, soil,
  land use, socio-economic characteristics, etc.

8
Rationale (2)

• The assumption is that adoption is related to the
  degree of similarity in key variables. The key
  variables condition constrain the likelihood that
  a given site is favourable for extrapolation.
• The degree to which a site is favourable for
  extrapolation depends upon the weight of evidence
  that the key variable is present.
• Maps indicate where similar conditions exist.

9
Methods

• Weights of Evidence
• Homologue

10
Bayesian Approach
H1
H2
H3
11
Bayesian Approach
H2
H1
H3
E1
12
Bayesian Approach
H1
H2
H3
E1
E2
13
Why a Bayesian Approach?

• Using these techniques it is possible to obtain a
  probabilistic distribution of the occurrence of
  modelled events, and reduce the uncertainty and
  sample space for further inquiries.
• The Bayesian paradigm offers a natural and
  consistent way of framing a problem and achieving
  data integration and developing methodological
  solutions (Herriges and Kling, 1998).
• It answers questions like 'How should a piece of
  evidence change what we currently believe?'
  (Spiegelhalter et al., 1999).
• An advantage of a Bayesian approach is that it
  allows explicit recognition of multiple
  perspectives (Spiegelhalter et al., 2000).

14
Project 6 Natural resource conservation and
management for increased food availability and
sustainable livelihoods empowering farming
communities with strategic innovations and
productive resources in dryland farming.

• The outputs relating to crops and soils are
• 1)      New varieties of staple food crops
  developed
• 2)      Soil and water management practices
  developed
• 3)      Drought probability map
• 4)      Manuals on crop production, and soil and
  water conservation developed.
• The outputs relating to water are
• 1)      Appropriate domestic water harvesting
  reservoirs designed to meet household water needs
• 2)      Dugouts enhanced to retain water
• 3)      Formal structures for governance of
  community water resources developed
• 4)      Manuals on appropriate water harvesting
  systems produced.

15
(No Transcript)
16
Key Variables

• Existence of fish production
• The status of sanitation facilities
• The poverty line as describe by the below UD 1
  per/day index, and
• Climate and other biophysical conditions

17
Location of pilot sites for project 6 in Africa
18
Location of pilot sites for project 6 in Africa
19
Extent of extrapolation domains
20
Improved Sanitation ()
Improved sanitation includes any of the following
excreta disposal facilities connection to a
public sewer, connection to a septic tank,
pour-flush latrine, simple pit latrine, and
ventilated improved pit latrine. WHO emphasizes
that these data measure access to an improved
excreta disposal system--access to a sanitary
system cannot be adequately measured on a global
scale
Irrigation Number of ha per pixel with
irrigation facilities (pixel size of 2500ha
21
Education Average length of schooling (years)
Children's Health Oral re-hydration therapy
(ORT) use rate ()
Oral rehydration therapy (ORT) use rate refers to
the percentage of children under the age of 5
with diarrhea (in the two weeks preceding data
collection) who received either oral rehydration
therapy (oral rehydration solutions or
recommended homemade fluids) or increased fluids
and continued feeding in order to prevent
dehydration and diarrheal diseases.
22
Population 2000 (x 1000)
23
Probabilities of socio-economic similarities
(ii) Based on agroecological variables
Probabilities of bio-physical similarities
24
Areas with highest SE and BF probabilities in
Africa
25
Probabilities of socio-economic similarities
(ii) Based on agroecological variables
Probabilities of bio-physical similarities
26
Areas with highest SE and BF probabilities in
South America
27
Population and areas with highest SE and BF
probabilities in the tropical world
28
Summary

• Where else in the world current funded projects
  are reliable
• Integrate diverse sources of information for
  targeting new sites
• Present a logical and quantitative method explore
  geographical impact
• Produce basic information for scenario analisis

29
Questions for extrapolation and scenario analyses

• 1. Please provide geographic co-ordinates of
  current pilot sites, alternatively details and
  exact names of sites.
• 2. Caracteristicas y/o problemas claves de los
  sitios piloto que son condicion que justifica o
  valida la implementacion de su proyecto en otro
  lugar por ejemplo
• 3. Project outputs expect to be adapted outside
  project site? (Productos que se esperan seran
  adaptados fuera del sitio de trabajo)
• 4. Areas where project outputs will be
  disseminated (please quantify in terms of numbers
  of expected adopters and changes in yield, etc.,
  if possible)?
• 5. If crop technologies form part of your
  project, what yield and area increases by
  irrigation/rainfed crop do you expect and by
  when year?
• 6. If water productivity increases are part of
  your project, what increase do you project from
  what baseline productivity or efficiency?, and
  for which crop rainfed or irrigated or which
  water use sector irrigation, domestic, industry,
  etc., and by which year?
• 7. Do you expect beneficiaries to engage in other
  activities, like crop diversification into which
  crops?
• 8. Critical factors that affect spread of
  outputs?

30
Bayesian Approach
Area173
Pilot sites10
Prior P0.0578
Pattern Area 22 Area of Pilot sites in Pattern
5 Conditional Prob5/220.227
Relational factor of prior and posterior P
5/100.5
Posterior Probability0.522/1730.063