HOW SCALE MATTERS: SOME CONCEPTS AND FINDINGS

1
HOW SCALE MATTERS SOME CONCEPTS AND FINDINGS
Thomas J. Wilbanks Oak Ridge National
Laboratory USA
Prepared for the Millennium Ecosystem Assessment
Workshop on Bridging Scales and Epistemologies in
Multiscale Assessments Alexandria, Egypt March
2004
2
This Presentation Will Briefly

• Step out from an overarching intellectual
  challenge across the sciences to understand
  relationships between macrocale and microscale
  phenomena and processes
• Summarize some (more or less) theoretical
  concepts related to how scale matters in
  conducting large, integrative nature-society
  assessments
• Consider issues related to both
• How phenomena and processes differ between scales
• How phenomena and processes at different scales
  affect each other

3
Consider Some Basic Concepts

• Arrayed along a scale continuum, most processes
  establish certain dominant frequencies a kind
  of lumpiness, organizing themselves more
  characteristically at some scales than others
• Recognizing this lumpiness, we can concentrate
  on scales related to particular levels of system
  activity (e.g., family, neighborhood, city,
  region, and country) and, at any particular
  level, subdivide space into a mosaic of regions

4
Consider Some Basic Concepts (contd.)

• In many (most?) cases, smaller scale mosaics are
  nested within larger scale mosaics therefore we
  can often think in terms of spatial hierarchies
• In some (many?) cases, there are relationships
  between spatial and temporal scales
• As we look across scales, the salience of
  external linkages increases as the scale shrinks
• Place is more than an intellectual and social
  construct it is a real context for
  communication, exchange, and decision-making

5
Consider Some Basic Concepts (contd.)

• Integrative research on complex sustainability
  issues is best carried out in a place-based
  context (e.g., small-regional)
• Sustainability science needs to be sensitive to
  multiple scales rather than to be focused on a
  single scale
• Selection of a single scale can frame an
  investigation too narrowly
• Phenomena, processes, structures, technologies,
  and stresses operate differently at different
  scales
• A particular scale may be more or less important
  at different points in a single cause-consequence
  continuum
• Institutions important for decision-making
  operate at different scales
• No single scale is ideal for broad-based
  investigation

6
(No Transcript)
7
Consider Some Basic Concepts (contd.)

• Integrative research on complex sustainability
  issues is best carried out in a place-based
  context (e.g., small-regional)
• Sustainability science needs to be sensitive to
  multiple scales rather than to be focused on a
  single scale
• Selection of a single scale can frame an
  investigation too narrowly
• Phenomena, processes, structures, technologies,
  and stresses operate differently at different
  scales
• A particular scale may be more or less important
  at different points in a single cause-consequence
  continuum
• Institutions important for decision-making
  operate at different scales
• No single scale is ideal for broad-based
  investigation

8
(No Transcript)
9
Consider Some Basic Concepts (contd.)

• Integrative research on complex sustainability
  issues is best carried out in a place-based
  context (e.g., small-regional)
• Sustainability science needs to be sensitive to
  multiple scales rather than to be focused on a
  single scale
• Selection of a single scale can frame an
  investigation too narrowly
• Phenomena, processes, structures, technologies,
  and stresses operate differently at different
  scales
• A particular scale may be more or less important
  at different points in a single cause-consequence
  continuum
• Institutions important for decision-making
  operate at different scales
• No single scale is ideal for broad-based
  investigation

10
What Are We Learning About Scale Differences?

• Observations of many variables at a more
  localized scale show greater variance and
  volatility (larger scales lose valuable
  information)
• Analyses and assessments at different scales tend
  to be associated with different research
  paradigms and styles e.g., regarding climate
  change responses
• Global or national scale quantitative analysis,
  using net present value metrics
• Small-regional or local scale integrated
  assessments including significant stakeholder
  involvement

11
(No Transcript)
12
What Are We Learning About Scale Differences?

• Observations of many variables at a more
  localized scale show greater variance and
  volatility (larger scales lose valuable
  information)
• Analyses and assessments at different scales tend
  to be associated with different research
  paradigms and styles e.g., regarding climate
  change responses
• Global or national scale quantitative analysis,
  using net present value metrics
• Small-regional or local scale integrated
  assessments including significant stakeholder
  involvement

13
What Are We Learning About Scale Differences
(contd.)?

• Downscaling and upscaling are likely to
  contribute different insights, and bottom-up
  investigations often provide different
  understandings compared with top-down
  investigations see AAG/GCLP re prospects for
  GHG emission reduction
• Different scales are related to different
  institutional roles, and the scale of decisions
  is often poorly matched with the scale of
  processes being decided upon
• The choice of a scale and a set of boundaries is
  not politically neutral, even if the choice is
  not based on political considerations

14
Toward Hypotheses About How Scale Matters
Ability to Capture Complexity Integrated Analysis
Variance
L Scale G
L Scale
G
Structure
Influence On Actions
Others?
Agency
L Scale
G
L Scale
G
15
What Are We Learning About Scale Relationships ?

• In many cases, cross-scale interactions are more
  significant than aggregate differences between
  scales
• Cross scale interactions can be considered in
  terms of certain basic dimensions
• Strength
• Constancy
• Directionality
• Resolution e.g., focused or broadcast
• Context e.g., additive or contradictory
• Effect e.g., stabilizing or destabilizing,
  controlling or enabling
• Intent

16
What Are We Learning About Scale Relationships
(contd.)?

• Cross-scale interactions are often associated
  with distinctive bridging-type institutional
  roles
• In many cases, important kinds of data about the
  interactions are elusive e.g., relationships
  between local phenomena and national or
  international corporate decision-making
• In many cases, relationships are too complicated
  to be incorporated in hierarchy theory
• In some (many?) cases, increasing understanding
  calls for laying out narrative story lines and
  then exploring the connections from multiple
  base points (e.g., Root-Schneiders strategic
  cyclical scaling)

17
(No Transcript)
18
In the Long Run, We Need to Be Able to Integrate
Both Scale Differences and Scale Relationships in
Multiscale Analysis
A
Macroscale analysis
B
B
Cross-scale dynamics
Cross-scale dynamics
Metascale synthesis
A
A
C
B
Cross-scale driving forces
Mesoscale analysis
Multiscale transient relationships
B
Cross-scale dynamics
A
Microscale analysis