BioEconomic Simulation Modelling: Approaches and Applications

1
Bio-Economic Simulation Modelling Approaches and
Applications

• Ken Belcher
• Visiting Researcher
• Institute of Rural Sciences
• University of Wales Aberystwyth

2
Outline

• Background and history of bioeconomic models
• Bioeconomic model example
• Introduction to underlying theory
• Complex systems
• Scale
• System evolution
• Bioeconomic model applications

3
Background

• As originally defined "bioeconomics" referred to
  the study of how organisms of all kinds earn
  their living in "nature's economy," with emphasis
  on co-operative interactions and the division of
  labor (Reinheimer, 1913).
• Modern bio-economics began with work by
  economists (eg. Scott Gordon, Gary Becker) and
  biologists (eg. Milner Schaefer, Michael
  Ghiselin).
• A general characterization has bioeconomics
  applied to the fundamental problem shared by all
  living species, namely survival and reproduction
  through adaptation - the strategies and tactics
  that an organism (or various functionally-organize
  d groupings of organisms) utilize to meet
  its/their basic "needs."

4
Background

• Today the bioeconomics is a rapidly growing area
  of research that uses expanded microeconomic
  tools and models to examine animal behaviour,
  human behaviour and human social institutions by
  explicitly linking the biological and economic
  relationships of the system.
• While the bioeconomic literature is developing a
  wide scope there has been a strong focus on
  management problems associated with renewable
  resources.
• One of the largest areas of bioeconomic model
  development has focused on the optimal
  exploitation of fisheries resources.

5
Static Bioeconomic Model (Schaefer-Gordon model)

• 1) Logistic growth function of the biological
  stock.
• X stock size
• r intrinsic growth rate
• K carrying capacity
• 2) Harvest function
• h harvest
• q efficiency or catchability coefficient
• E harvesting effort
• 3) Profit expression
• ? - profits
• p output price
• c effort price

6
Static Bioeconomic Model

• The simple static model can take into account
  resource renewal and incorporate stock growth
  while maintaining a profit maximization stance
  and an equilibrium or steady state sustained
  yield harvest rate (hF(X)).
• In the absence of entry limitations the model
  will equilibrate total revenue and total costs
  and all resource rents will be dissipated
  providing a classic open access solution where
  Xc/pq.

7
Dynamic Bioeconomic Models

• When dealing with natural processes and
  biological populations it is generally more
  satisfying to use dynamic specifications - stock
  adjustment.
• These dynamic models have used either
• A framework that optimizes (optimal control
  theory) profitability or some social welfare
  function.
• A simulation framework that may assume away an
  optimal equilibrium or steady state solution.
• Dynamic models provide predictions of parameters
  (eg. stock, harvest, price etc.) that are time
  dependent and that may lead to an internal
  equilibrium or collapse.

8
Bioeconomic Model Theory

• Since bioeconomics focuses on addressing
  questions that explicitly and dynamically
  integrate the development and adaptation of
  biophysical and economic systems the discipline
  has depended heavily on the development of
  appropriate models.
• Bioeconomics has advanced in the last decade
  (Journal of Bioeconomics 1999) partly due to
  the range of modeling techniques that have become
  available through advances in computer speed
  enabling broad interdisciplinary systems views.

9
Bioeconomic Model Theory

• Bioeconomic models can be developed to meet a
  number of objectives which, in turn, determine
  the nature of the model.
• Holling (1966) described tradeoffs in three
  fundamental model criteria generality,
  precision and realism.

10
Bioeconomic Model Theory

• 1) High generality conceptual models (low realism
  and/or precision) simple linear and nonlinear
  economic and ecological models.
• 2) High-precision analytical models - model
  accurately reflects data often with poor
  generality and realism. Often see models with
  high resolution but with simplified relationships
  (few properties characterizing the system) and
  short time frames. Examples include economic and
  biological input-output models, large statistical
  and econometric models (predict short-run
  behaviour with precision).
• 3) High-realism impact-analysis models
  concerned with accurately representing the
  underlying processes in a specific system. Often
  include dynamic, nonlinear evolutionary systems
  models at moderate to high resolution (can be
  very site-specific).
• 4) Moderate-generality, moderate-precision
  indicator models help to determine the overall
  magnitude and direction of system change (trading
  off realism for some amount of generality and
  precision) - include aggregate measures of system
  performance GDP, GNP, ecosystem health indicators

11
Theory Complex Systems

• Bioeconomics deals primarily with complex systems
  (biophysical systems linked to economic systems).
  
• Complex systems are characterized by
• Strong (non-linear) interactions between the
  parts
• Complex feedback loops making it difficult to
  distinguish cause from effect.
• Significant time and space lags, discontinuities,
  thresholds and limits (Costanza et al. 1993).

12
Theory Complex Systems

• These characteristics make it inappropriate to
  simply add up or aggregate small-scale
  (time/space) behaviour to arrive at large-scale
  results.
• Example partial equilibrium analysis (single
  market) compared to general equilibrium analysis
  (all markets).
• In bioeconomic systems it cannot be assumed that
  there are no important linkages and feedback
  loops between the biological system and the
  economic system emergence, co-evolution.

13
Theory Scale

• The development of bioeconomic models must be
  done with careful consideration of scale
  including
• Resolution the spatial grain, time step, degree
  of complication.
• Extent time, space and number of components
  modeled.
• Data is generally gathered at relatively small
  scales (small field plots, individual firms or
  farms) - this information is used to build models
  at different scales (regional, national) leading
  to aggregation error.

14
Theory Scale
Natural log of predictability
Data
Model
Lower
Higher
Natural log of resolution
(Source Costanza et al. 1993)
15
Theory Scale

• Model development must include mechanisms to
  incorporate nonlinear fine-scale variability into
  course-scale equations.
• Statistical expectations deriving course-scale
  equations that incorporate fine-scale
  variability.
• Recalibration recalibrate fine-scale equations
  to coarse-scale data.
• Partitioning subdividing the system into many
  relatively more homogeneous levels or zones and
  applying fine scale equations to each partition.

16
Theory Scale

• Partitioning requires an adjustment of the
  parameters for each partition, a choice of the
  number of partitions (resolution) and an
  understanding of the effects of these choices.
• Partitioning may be facilitated by hierarchy
  theory.
• Hierarchy theory - systems can be partitioned
  into naturally occurring levels which share
  similar time and space scales and which interact
  with higher and lower levels in systematic ways.
  Each level experiences the higher levels as
  constraints and the lower levels as noise.

17
Theory Evolution

• Bioeconomic systems often operate away from an
  equilibrium in a state of constant adaptation.
• System evolution is an important characteristic
  that is often required of bioeconomic models
  (genetic evolution, cultural evolution,
  technological evolution, evolutionary game
  theory) .
• The evolutionary paradigm is different from the
  optimization paradigm common in economics
• Evolution is path dependent history is
  important
• Survival of the first rather than of the fittest
  is possible under conditions of increasing
  returns (positive feedback).
• Evolution can achieve multiple equilibria
• No guarantee of optimality

18
Bioeconomic Models and Agriculture

• To develop appropriate bioeconomic models for
  agricultural systems there needs to be
  consideration of particular system
  characteristics
• Property rights for the natural resource may be
  well-defined (eg. private land ownership).
• Poorly defined property rights may also exist
  such that the costs and benefits of resource use
  may not be captured by the decision maker (eg.
  biodiversity loss, surface water quality,
  landscape amenities).
• Production is a function of environmental quality
  which is at least partly an endogenous parameter
  (eg. soil quality, range condition).
• Agricultural systems are spatially heterogeneous
  (eg. soil quality, topography, habitat quality,
  opportunity cost of land).
• Evolutionary development and adaptation of the
  biophysical and economic components of the
  system.

19
Bioeconomic Models and Agriculture

• A wide range of bioeconomic models, both
  optimization and non-optimization simulation,
  have been developed to address management
  questions in agricultural systems.
• Management, exclusion or eradication of invasive
  species (Eisworth and Johnson, 2002 Jones and
  Cachco, 2000).
• Management of wetlands on agric. Landscapes
  (Fernandez and Karp, 1998 Whitten and Bennet,
  2005).
• Habitat and biodiversity conservation (Alexander
  and Shields, 2002 Bulte and Horan, 2003 Foudi,
  2005).
• Climate change mitigation and adaptation
  (Nordhaus, 1991)

20
Sustainable Agroecosystem Model
21
Spatially Explicit Bioeconomic Models

• Historically bioeconomic models have assumed away
  spatial heterogeneity to address the issues of
  scale and scaling.
• However, most if not all environmental issues
  have a strong spatial component.
• Bioeconomic models have recently begun to include
  spatial heterogeneity explicitly .
• Sanchirico and Wilen published a series of papers
  (beginning in 1999) that focused on the optimal
  exploitation of an ocean fishery in a patchy
  environment and the optimal distribution of
  harvest and ocean reserves
• A number of spatially explicit simulation models
  have been developed in the last decade (eg.
  SELSMs process-based, medium to high temporal
  and spatial resolution, complex, dynamic)

22
Spatially Explicit Bioeconomic Models

• A spatially explicit dynamic modeling framework
  is proving to be particularly appropriate for
  addressing management and policy questions in
  agricultural systems given their patchy nature
  (eg. soils, distance to markets, topography,
  policy).
• Dynamic incorporate adaptation, stock recovery
  (depletion), learning.
• Spatial interaction across cells (nutrients,
  pollutants), dispersal and migration.
• A body of research is developing that uses GIS
  databases to explicitly integrate the economic
  and biophysical characteristics of the
  agricultural landscape in a simulation model
  framework.

23
Spatially Explicit Bioeconomic Models

• Some research questions that may be addressed
• What is the optimal spatial exploitation (land
  use, harvest) strategy for resources subject to
  spatial bioeconomic heterogeneity.
• How does this strategy differ from the spatially
  homogeneous case.
• How does the optimal policy depend on
  environmental variability and uncertainty.
• How should policies be designed spatially.
• How should policies be designed temporally.
• How do policy changes influence management
  decisions made outside the target zone.