Your SlideShare is downloading. ×
0
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE�
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE

106

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
106
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. USING INSTITUTIONAL ARRANGEMENTS AND HYBRID AUTOMATA FOR REGIONAL SCALE AGENT-BASED MODELING OF LAND CHANGE Sérgio Souza Costa Orientadores: Dr. Gilberto Câmara Dra. Ana Paula Dutra de Aguiar
  • 2. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 2Land Change Introduction
  • 3. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 3Land Change: Amazonia Introduction
  • 4. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 4Level of Analysis Micro (Social Science) Macro (Natural Science) Introduction
  • 5. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 5Level of Analysis Micro (Social Science) Macro (Natural Science) Introduction
  • 6. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 6Pattern-based models ! How changes in the regional (or national) scale affect the smaller units of the model, usually pixels or cells. Introduction Relation between land changes and spatial variables
  • 7. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 7Pattern-based models ! Introduction Relation between land changes and spatial variables Analysing a deforestation hotspot
  • 8. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 8Level of Analysis Micro (Social Science) Macro (Natural Science) Introduction
  • 9. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 9Agent-based models Introduction 1.  Agents, environmental and interactions. 2.  Emergence concepts (Local actions lead to global patterns) 3.  No central authority (autonomos) 4.  Heterogeneity Macro Pattern Interactions between agents and the environment.
  • 10. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 10 Agent-Based Modeling of Land change Introduction TheoreticalModels Empiricalmodels Agent-based models (ABM) range from theoretical to empirical. Theoretical models use simple generalizable ideas, whereas empirical models require more complexity and case-specific data..
  • 11. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 11 Agent-Based Modeling of Land change Introduction TheoreticalModels Empiricalmodels Agent-based models (ABM) range from theoretical to empirical. Theoretical models use simple generalizable ideas, whereas empirical models require more complexity and case-specific data..
  • 12. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 12Problem definition Introduction The modellers rarely have access to individual data to represent and locate households in land change agent-based model.
  • 13. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 13Problem definition Introduction The lack of data is an even greater problem in large frontier area, like Brazilian Amazonia. The modellers rarely have access to individual data to represent and locate households in land change agent-based model.
  • 14. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 14Problem definition Introduction Large extension of municipalities The lack of data is an even greater problem in large frontier area, like Brazilian Amazonia. The modellers rarely have access to individual data to represent and locate households in land change agent-based model.
  • 15. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 15Problem definition Introduction Intense land changes and land market. Large extension of municipalities The lack of data is an even greater problem in large frontier area, like Brazilian Amazonia. The modellers rarely have access to individual data to represent and locate households in land change agent-based model.
  • 16. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 16Intense changes 1985 1995 2006 2010 Area of farms (km²) 2394 9601 17917 * Number of farms 1375 5893 7148 * Deforestation (km²) 426 4070 16762 19271 Head of cattle 24494 230875 1912033 2290538 Population 14016 69117 85751 125030 Evolution of deforestation, population and agrarian structure in the study area. Sources: (IBGE, 2006; INPE, 2010b) Problem definition
  • 17. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 17Intense changes Problem definition Evolution of deforestation patterns in the study area.
  • 18. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 18Intense changes Introduction Concentrationoflandownership inTucumã. In 1996, farms with individual areas greater than 1000 ha accounted for 8% of the total area of farms. In 2006, they accounted for greater than 60% of the total. In the same period, the number of farms decreased from 2518 to 1039
  • 19. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 19Problem definition Introduction Different actors and strategies. Intense land changes and land market. Large extension of municipalities The lack of data is an even greater problem in large frontier area, like Brazilian Amazonia. The modellers rarely have access to individual data to represent and locate households in land change agent-based model.
  • 20. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 20Different actors and strategies Problem definition
  • 21. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 21 Problem definition 61,000 ha 30,000 ha 50-200 ha 60,000 ha 30,000 ha 50 ha 20,000 ha 20,000 ha 200 ha Different actors and strategies
  • 22. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 22Problem definition Introduction The modellers rarely have access to individual data to represent and locate households in land change agent-based model. The lack of data is an even greater problem in large frontier area, like Brazilian Amazonia. The behaviour of actors changes in response to internal and external conditions. Intense land changes and land market. Large extension of municipalities Different actor and strategies.
  • 23. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 23 Lagoa do Triunfo Farm: 19962 HA Problem definition Changing of behaviour
  • 24. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 24 Problem definition 0 200 400 600 800 1000 1200 1400 1600 1800 2000 APA – Triunfo do Xingu - 2007 Lagoa do Triunfo Farm: 19962 HA Evolution of deforestation in the Lagoa do Triunfo Farm: ~653 ha/year. Changing of behaviour
  • 25. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 25Changing of behaviour Problem definition 0 20 40 60 80 100 120 140 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Prot. Integral – 2005 0 50 100 150 200 250 300 350 400 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 APA – Triunfo do Xingu – 2007 Evolution of deforestation in different regions:
  • 26. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 26 1.#Occupa)on# 2.#Expansion# 3.#Restric)on# Ca7le#head#/#deforesta)on#(ha)# Evolution of Cattle head / deforestation in the São Félix do Xingu. Changing of behaviour Problem definition
  • 27. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 27Question Introduction How to develop agent-based models that are expressive to account for the different strategies of land change agents in frontier areas?
  • 28. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 28Hypothesis Introduction Our hypothesis is that the idea of strategies (supported by hybrid automata) provides a computational model that is capable of expressing complex collective behavior of land change agents.
  • 29. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 29Hypothesis Introduction Our hypothesis is that the idea of strategies (supported by hybrid automata) provides a computational model that is capable of expressing complex collective behavior of land change agents. Hybrid automata represent strategies that are loosely coupled to agents and may vary during the simulation in response to local and institutional variability’s
  • 30. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 30Hypothesis Introduction Our hypothesis is that the idea of strategies (supported by hybrid automata) provides a computational model that is capable of expressing complex collective behavior of land change agents. Hybrid automata represent strategies that are loosely coupled to agents and may vary during the simulation in response to local and institutional variability’s To validate them, we put up a model for the area in Amazonia with the highest deforestation rate in the 1990s and 2000s. The model captures large-scale land change during the 2000s and is used to build scenarios until 2020.
  • 31. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 31Relevance for Computer Science Introduction 3. Computational modeling of complex systems: artificial, natural, socio-cultural, and human-nature interactions Grand Challenges in Computer Science Research in Brazil - 2006 – 2016 (SBC) The complexity of this kind of research grows with the increase in data volume and/or variables to be considered. Another complicating factor is the frequent need for combining several knowledge domains.
  • 32. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 32Contents 1.  Introduction 2.  Agent-based models at regional scale 3.  Implementation 4.  Model description 5.  Model simulation and results 6.  Conclusion
  • 33. AGENT-BASED MODEL AT REGIONAL SCALE
  • 34. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 34Overview Agent-based model at regional scale Spatial Units Arrangement #1 Arrangement #2 Agents Strategy #1 Strategy #2 Strategy #3
  • 35. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 35Spatial Units Agent-based model at regional scale Regular   cell   Farm   Regular cells describe the properties of space. Each cell has information on its land cover and land use and its spatial properties, such as distance to roads. Farms are irregular space partitions that belong to agents. A farm is linked to one of more cells.
  • 36. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 36Overview Agent-based model at regional scale Spatial Units Arrangement #1 Arrangement #2 Agents Strategy #1 Strategy #2 Strategy #3
  • 37. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 37Institutional arrangements Agent-based model at regional scale We refer to institutional arrangements as deals set up between interest groups, social movements and state agencies to respond to rules and norms that are relevant to them (DIETZ; OSTROM, 2003). These pacts define how agents manage natural resources (SCOTT; MEYER, 1994). The arrangements evolve on the time, where for each time period different arrangements are valid
  • 38. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 38Institutional arrangements The Institutional Arrangements are represented as tuples of four components: 1.  Name, which distinguishes an arrangement from the other arrangements. 2.  Context, inform a set of context variables to agents. 3.  Condition, the target, which specifies the farmers that are influenced by the arrangement. 4.  Temporal, the time period for which the arrangement applies. Agent-based model at regional scale
  • 39. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 39Overview Agent-based model at regional scale Spatial Units Arrangement #1 Arrangement #2 Agents Strategy #1 Strategy #2 Strategy #3
  • 40. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 40Agents Agent-based model at regional scale Agent Farm Regular cell Agents are farmers that carry out land change. Agents own farms and have attributes such as capital, technology level and expansion aims.
  • 41. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 41Overview Agent-based model at regional scale Spatial Units Arrangement #1 Arrangement #2 Agents Strategy #1 Strategy #2 Strategy #3
  • 42. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 42Strategies A set of consistent actions and agent decisions. Each agent chooses one strategy at a time, which it can change later. Examples of strategies include land speculation, intensive farming, and subsistence agriculture. Agent-based model at regional scale MIGRATION Search for land SUBSISTENCE Deforest 10%/year CATTLE Extensive cattle raising Deforestation > 50% Land exhaustion Farmer gets land
  • 43. IMPLEMENTATION
  • 44. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 44 Implementation Model source code • model syntax semantic checking • model execution TerraMEINTERPRETER LUA interpreter TerraMEframework TerraLib database TerraView MODEL DATA TerraME (Carneiro, 2006)
  • 45. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 45Spatial Units Implementation TerraMe Cell Farm TerraME suports regular and irregular cells. We support the Farm data type, which includes a set of operations for land market submodels, like farm creation, farm split and farm merge
  • 46. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 46Agents Implementation TerraMe Agent Farmer Terrame support agents and its operations. We support the Farmer data type as an extension of the TerraME agent; this type includes common functions performed by and attributes of farmers.
  • 47. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 47Strategies Implementation •  To implement our proposal, we need software that expresses strategies. In computational terms, strategies make up a discrete state machine. •  When an agent changes his strategy, he moves from one state to another. Inside a state, an agent carries out continuous actions. •  We propose the formalism of hybrid automata to capture the idea of strategies.
  • 48. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 48Strategies A hybrid automaton H has three parts (HENZINGER, 1996): •  A finite set of variables X = {x1, x2, ... xn} which is the automaton internal status. •  A finite directed graph G = (V, E). The set of vertices V are states, and the set of edges E are jumps. Each edge jump connects a source state to a target state, following a condition. If this jump condition is true, the automaton discrete state will change from the source state to the target state. •  A set of flow rules assigned to each state. When a flow rule is evaluated it changes the automaton internal status, defined by the variables {x1, x2, ... xn}. Implementation
  • 49. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 49Strategies A hybrid automaton H has three parts (HENZINGER, 1996): •  A finite set of variables X = {x1, x2, ... xn} which is the automaton internal status. •  A finite directed graph G = (V, E). The set of vertices V are states, and the set of edges E are jumps. Each edge jump connects a source state to a target state, following a condition. If this jump condition is true, the automaton discrete state will change from the source state to the target state. •  A set of flow rules assigned to each state. When a flow rule is evaluated it changes the automaton internal status, defined by the variables {x1, x2, ... xn}. Implementation TerraME supports hybrid automaton.
  • 50. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 50Strategies State Flow rule Jump Condition New state MIGRATION Search for land Farmer gets land SUBSISTENCE SUBSISTENCE Farmer deforests 10% of land/year Deforestation > 50% EXTENSIVE CATTLE EXTENSIVE CATTLE Extensive cattle raising Land exhaustion MIGRATION Implementation Example: A hybrid automaton model for poor migrant farmer
  • 51. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 51Institutional arrangements We support the institutional arrangement as a new Lua data type. This code shows an example of an institutional arrangement. Implementation arrang1 = Arrangement (! {market = HIGH, enforcement = LOW }, ! function (agent) return agent.capital> 1000 end, ! function (event) return event:getTime() > 2008 end! )!
  • 52. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 52Institutional arrangements We support the institutional arrangement as a new Lua data type. This code shows an example of an institutional arrangement. Implementation arrang1 = Arrangement (! {market = HIGH, enforcement = LOW }, ! function (agent) return agent.capital> 1000 end, ! function (event) return event:getTime() > 2008 end! )! Context Condition Temporal Name
  • 53. MODEL DESCRIPTION
  • 54. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 54 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 55. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 55 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 56. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 56Purpose The model attempts to describe landscape dynamics and replicate the deforestation fluctuations in the time period simulated by considering changes in the institutional context and their effects on households’ strategies. Model Description
  • 57. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 57 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 58. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 58 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 59. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 59Scales Entities, attributes and scales Study&Area& Main&rivers& Deforesta3on&4&2006& Main&roads& The model simulates a rural region that covers 60,000 Km2 in the south- east of Pará, Brazil. Spatial resolution 225x225 ha, 25x25 ha and 1x1 ha. Temporal extension: 1985 to 2020. Temporal resolution: 1 year São Félix do Xingu Tucumã
  • 60. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 60 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 61. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 61Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 62. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 62Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 63. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 63Spatial Units: Farm A farm is a spatial unit entity, and corresponds to the basic decision unit referring to Land Market Decisions. The farm attributes include •  area, •  price, •  quantity of animals, •  pasture area, •  area of degraded pasture and •  area of remaining forest. Entities, attributes and scales
  • 64. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 64Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 65. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 65Spatial Units: Regular cells A RegularCell represents the geographic space attributes, including biophysical (e.g. slope, soil quality) and accessibility features (e.g. distance to roads and distance to river). •  Area of each cell. •  Land cover (Forest, Pasture, Secondary Forest, River, Other •  Number of animals per cell •  Minimum Euclidean distance (normalized) to roads, rivers and urban centers. •  Slope Entities, attributes and scales
  • 66. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 66Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 67. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 67Agents Based on fieldwork and literature review, we defined a agent to be a rural entrepreneur with the following dynamic attributes: •  Inclination to obey the law. •  Number of farms. •  Average size of the farm he wants to buy. •  Technological capability (Low, Medium, High). •  Investment capital. Entities, attributes and scales
  • 68. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 68Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 69. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 69Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 70. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 70Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 71. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 71Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 72. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 72Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 73. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 73Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 74. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 74Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 75. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 75Agents and Strategies Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity A farmer performs land use and land market decisions on these spatial units. These decisions are guided by different strategies that an agent can take over time. A farmer chooses between five strategies:
  • 76. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 76Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 77. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 77Context variables Context variable Description Forest  code  enforcement  how  is  the  Forest  Code  being  followed?   Law  enforcement    is  there  control  over  poaching  of  public  lands?   Market  for  ca=le    how  strong  is  the  beef  market  chain?   Credit  for  small  farmers   how  easy  and  cheap  is  the  credit  for  small  farmers?   Credit  for  large  farmers    how  easy  and  cheap  is  the  credit  for  large  farmers?   Credit  for  reforesta?on    how  easy  and  cheap  is  the  credit  for  reforesta?on?   Entities, attributes and scales The agent uses six guidelines to decide on his strategy, which are:
  • 78. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 78Entities, attributes and scales Migrant Extensive Farming Extensive Expanding Intensive Farming Speculate Abandon Rural activity Land use and land market decisions SPATIAL UNITS: Farms AGENTS: Farmers Context variables Rule parameters SPATIAL UNITS: Regular cells Environmental, economic and accessibility information and feedbacks INSTITUTIONALARRANGEMENTS: RULES AND NORMES
  • 79. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 79Institutional Arrangements We defined six institutional arrangements in our model: 1.  Government-induced occupation: prevalent from 1970s to mid 1990s, with the government encouraging people to occupy Amazonia. 2.  Beef market chain organization: From the mid 1990s until today, following initial occupation with easy access to land, the beef market chain grew. 3.  Deforestation control: From 2005 onwards, law enforcement increased and the government created new protected areas Entities, attributes and scales
  • 80. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 80Institutional Arrangements We defined six institutional arrangements in our model: 4.  Green market: In the late 2000s, given pressure from consumers, NGOs and public attorneys, part of the private sector changed. Some farmers and part of the industry agreed to comply with sustainable practices, in exchange for market and credit access. 5.  Sustainable Development: a possible future arrangement to bring about equilibrium between social, environmental and economic goals. This choice combines strong law enforcement with green market practices. 6.  Economic development: a possible future arrangement based on a return to 1970s model, where economic growth prevails over environmental or social concerns. Entities, attributes and scales
  • 81. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 81Arrangements and Variables INSTITUTIONAL ARRANGEMENTS FOREST CODE ENFORCE LAND POACHING ENFORCE CATTLE MARKET CREDIT LARGE FARMS CREDIT LARGE FARMS CREDIT REFOREST Government- induced occupation 2 Beef market chain organization 2 2 1 Deforestation control 1 2 Green market 3 3 3 3 2 Sustainable development 3 3 1 1 3 3 Economic development 2 2 1 Relevance: 3 = strong, 2 = medium, 1 = low, no value = no relevance Entities, attributes and scales
  • 82. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 82 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 83. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 83 Process Oververview and Schedulling Initialization For each year T  Allocate agents in the landscape Compute attributes at regular cell level Update agent quantities at time step t (immigration) Update institutional arrangements acting at time step t Each farmer execute the land use and land market decisions Compute and aggregate attributes at farm level
  • 84. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 84 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Initialisation 5.  Input data 6.  Submodels Model Description
  • 85. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 85Initialization •  In theoretical models, the initialization may be simple random placement (EPSTEIN; AXTELL, 1996). •  Empirical models require much data, such as census and cadastral data. Model description
  • 86. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 86Initialization The initialization should answer the following questions at least: •  How many farmers are in the study area? •  What is the size of each farm? •  Where are these farms located? Model description
  • 87. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 87Initialization The initialization should answer the following questions at least: •  How many farmers are in the study area? •  What is the size of each farm? •  Where are these farms located? Model description
  • 88. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 88Initialization Area range (HA) Quantity Sum of area by range < 50 685 27625 50-100 574 44980 100-200 62 8877 200-500 30 9618 500-1000 9 6467 > 1000 14 141863 Total 1374 239430 Model description Number of farms and total area aggregated for various area ranges (IBGE, 1985).
  • 89. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 89Initialization The initialization should answer the following questions at least: •  How many farmers are in the study area? •  What is the size of each farm? •  Where are these farms located? Model description
  • 90. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 90Initialization Model description The deforestation map is derived from classification of LANDSAT TM images, which use the following classes: forest, deforestation, not-forest, cloud and river. In 1985, the deforestation class accounted for an area of 426 km2. Deforested  Area  /  Area  of  farm 1985: 0.17 2006: 0.93
  • 91. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 91Initialization Model description 120,000 km2 São Félix do Xingú - 1985
  • 92. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 92Initialization Model description SãoFélixdoXingú-Contexto-1985 Indigenous land
  • 93. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 93Initialization Model description SãoFélixdoXingú-Contexto-1985 Indigenous land Settlement (1981)
  • 94. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 94Initialization Model description SãoFélixdoXingú-Contexto-1985 Surge um novo Brasil no Sul do Pará ... Localizado perto de polos importantes como Carajas, Tucurui, servido por vias de transporte e privilegiado pelas condições naturais. Panfleto de divulgação do Projeto Tucumã, 1981. Arquivo pessoal de Rosângela Sampaio
  • 95. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 95Initialization Model description SãoFélixdoXingú-Contexto-1985
  • 96. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 96Initialization Model description
  • 97. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 97Initialization Model description 0 2000 4000 6000 8000 10000 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 (a) Deforested areas in 1985 (b) Number of farmers from 1985 to 2010 (b) Distribution of farms Deforestation - 1985
  • 98. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 98Initialization Model description 0 2000 4000 6000 8000 10000 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 (a) Deforested areas in 1985 (c) Number of farmers from 1985 to 2010 (b) Distribution of farms Deforestation - 1985
  • 99. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 99Input data 1985 < 50 50-500 500-2000 > 2000 total qt 685 666 12 11 1374 area 27625 63475 10844 137486 239430 1995 < 50 50-500 500-2000 > 2000 total qt 871 1811 93 41 2816 area 42650 274944 103678 320823 742095 2005 < 50 50-500 500-2000 > 2000 total qt 3895 1756 353 105 6109 area 85006 270872 363661 738062 1457601 Model description
  • 100. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 100Input data 1985 < 50 50-500 500-2000 > 2000 total qt 685 666 12 11 1374 area 27625 63475 10844 137486 239430 1995 < 50 50-500 500-2000 > 2000 total qt 871 1811 93 41 2816 area 42650 274944 103678 320823 742095 2005 < 50 50-500 500-2000 > 2000 total qt 3895 1756 353 105 6109 area 85006 270872 363661 738062 1457601 Model description
  • 101. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 101 Overview, Design concepts, and Details The ODD (Overview, Design concepts, and Details) protocol is used to provide a standardised description of our model (GRIMM et al., 2006, 2010). 1.  Purpose 2.  Entities, Attributes and Scales 3.  Process overview and scheduling 4.  Design ideas 5.  Initialisation 6.  Input data 7.  Submodels Model Description
  • 102. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 102Submodels: Pasture degradation Model description For cattle famers, the productivity of a farm is characterised by the number of animals per ha. This variable is complex and depends on many different factors. We simplified this variable by making it a function of only technology level and pasture age: Animal  unit  =  degradation  (technology  level,  pasture  age) Where: technology  level  =  {low,  medium,  high}
  • 103. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 103Submodels: Pasture degradation Model description The degradation is a discrete function that has a single value for each technology level and pasture age. This function may be calibrated for based on the region. In our experiments, we use the discrete function represented in Figure 5‑9. 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1 2 3 4 5 6 7 8 9 10 11 12 high medium low
  • 104. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 104Submodels: Farm locator/creator Model description This submodel creates farms, both at initialisation and during the simulation. For each new farm in the model, it 1.  first produces a sample of 500 cells. 2.  it calculates the potential for each cell (x, y) using the multi-criteria evaluation (MCE). 3.  chooses the cell with the greatest potential. 4.  The chosen cell is marked as the location of the agent’s house. 5.  The Farm Creator operator takes the chosen cell and a specified farm area as parameters. 6.  Finally, the submodel calculates and updates the spatial relations amongst farms using the Neighbourhood operator. MCE
  • 105. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 105Submodels: Farm locator/creator Model description This submodel creates farms, both at initialisation and during the simulation. For each new farm in the model, it 1.  first produces a sample of 500 cells. 2.  it calculates the potential for each cell (x, y) using the multi-criteria evaluation (MCE). 3.  chooses the cell with the greatest potential. 4.  The chosen cell is marked as the location of the agent’s house. 5.  The Farm Creator operator takes the chosen cell and a specified farm area as parameters. 6.  Finally, the submodel calculates and updates the spatial relations amongst farms using the Neighbourhood operator. MCE
  • 106. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 106Submodels: Farm locator/creator Model description 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Create (1) Create operator, which instantiates a new farm given an initial settlement cell and a farm area. 0 0 2 2 0 0 0 0 2 2 0 0 3 3 1 1 0 0 3 0 1 0 0 0 4 0 0 0 5 5 4 4 0 0 5 5 0 0 2 2 0 0 0 0 2 2 0 0 3 3 1 1 0 0 3 0 1 0 0 0 4 0 0 0 5 5 4 4 0 0 5 5 Neigh (1) Neighbourhood operator, which identifies the spatial adjacency relations amongst the farms. Farm operators used for farm locator/creator submodel:
  • 107. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 107Submodels: Farm locator/creator Model description 1.  Receives an initial cell and farm area. The choice of initial cell may depend on factors such as the proximity of rivers or roads. 2.  Adds the available neighbouring cells until a given area is attained.
  • 108. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 108Submodels: Land-market Model description Puts on sale
  • 109. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 109Submodels: Land-market Model description Puts on sale Farms on sale given a set of restriction (price and area)?
  • 110. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 110Submodels: Land-market Model description Puts on sale Farms on sale given a set of restriction (price and area)? Rank-order-select Farm buy
  • 111. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 111Submodels: Land-market Model description 0 0 0 0 0 0 0 2 1 1 0 0 0 2 2 1 1 0 0 2 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Merge (1,2) 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Merge operator, which joins two farms and updates the farm boundaries. Farm operators used for land-market submodel . 0 0 0 0 0 0 0 2 3 3 0 0 0 2 2 3 3 0 0 2 2 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 Split (1) 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 1 1 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Split operator, which divides a given region into several farms of specified areas.
  • 112. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 112Submodels: Land use Model description In our experiments, land-use decisions refer to deforestation and pasture creation, which depend on the technology level. Land-use decisions also include pasture reformation and reforestation. These decisions may yield to different trajectories (A, B and C), These trajectories reflect the different strategies that farmers can employ, each strategy constrains how much, where and when the agents deforest, create pasture and manage the land.
  • 113. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 113Submodels: Land use Model description We propose using a land-allocation submodel (similar to pattern models) to define what cells are changed. In this model, each agent performs a land allocation, using a method such as rank- order-change or a clues-like algorithm, under the constraint that a farm can have at most 60,000 cells. In our simulation, we use a simple allocation strategy in which we rank the cells using the MCE to calculate the potential of each cell. The factors considered include the slope and distance to the house. Then, we order the cells and finally, change the land use.
  • 114. MODEL SIMULATIONS AND RESULTS
  • 115. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 115Simulations Simulation Period Institutional Arrangements Temporal range of arrangement 0 (Calibration) 1985-1997 Government-induced occupation 1970-1996 1 (Validation) 1985-2010 Government-induced occupation 1970-1996 2 (Validation) 1985-2010 Government-induced occupation Beef market chain organization 1970-1996 1997-2010 3 (Validation) 1985-2010 Government-induced occupation Beef market chain organization Deforestation control 1970-1996 1997-2010 2005-2010 Model simulation and results
  • 116. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 116Simulations Simulation Period Institutional Arrangements Temporal range of arrangement 0 (Calibration) 1985-1997 Government-induced occupation 1970-1996 1 (Validation) 1985-2010 Government-induced occupation 1970-1996 2 (Validation) 1985-2010 Government-induced occupation Beef market chain organization 1970-1996 1997-2010 3 (Validation) 1985-2010 Government-induced occupation Beef market chain organization Deforestation control 1970-1996 1997-2010 2005-2010 Model simulation and results
  • 117. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 117Results Model simulation and results 0 100000 200000 300000 400000 500000 600000 700000 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Observed Simulated Deforestation rate 0 10000 20000 30000 40000 50000 60000 70000 80000 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 Observed Simulated Total deforested area
  • 118. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 118Simulations Simulation Period Institutional Arrangements Temporal range of arrangement 0 (Calibration) 1985-1997 Government-induced occupation 1970-1996 1 (Validation) 1985-2010 Government-induced occupation 1970-1996 2 (Validation) 1985-2010 Government-induced occupation Beef market chain organization 1970-1996 1997-2010 3 (Validation) 1985-2010 Government-induced occupation Beef market chain organization Deforestation control 1970-1996 1997-2010 2005-2010 Model simulation and results
  • 119. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 119Results – S1, S2 and S3 Model simulation and results 0 20000 40000 60000 80000 100000 120000 140000 160000 180000 200000 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Observed S1 S2 S3 0 500000 1000000 1500000 2000000 2500000 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Observed S1 S2 S3 Deforestation rateTotal deforested area
  • 120. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 120Results – S1, S2 and S3 Model simulation and results 0 20000 40000 60000 80000 100000 120000 140000 160000 180000 200000 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Observed S1 S2 S3 0 500000 1000000 1500000 2000000 2500000 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Observed S1 S2 S3 Deforestation rateTotal deforested area
  • 121. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 121Results – S1, S2 and S3 Model simulation and results 1200000 1400000 1600000 1800000 2000000 2200000 2400000 2005 2006 2007 2008 2009 2010 Observed S1 S2 S3
  • 122. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 122Results – S3 Model simulation and results
  • 123. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 123Results – S3 Model simulation and results
  • 124. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 124Scenarios Simulation Period Alternative Institutional Arrangements Temporal range of arrangement 4 (Scenario A) 1985-2020 Government-induced occupation Beef market chain organization Deforestation control Green market 1970-1996 1997-2008 2005-2020 2009-2020 5 (Scenario B) 1985-2020 Government-induced occupation Beef market chain organization Deforestation control Sustainable development Green market 1970-1996 1997-2008 2005-2012 2013-2020 2009-2020 6 (Scenario C) 1985-2020 Government-induced occupation Beef market chain organization Deforestation control Economic development Green market 1970-1996 1997-2008 2005-2012 2013-2020 2009-2020 Model simulation and results
  • 125. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 125Scenarios – A, B and C Model simulation and results Pasture area (ha) Forest area inside farms (%) 0 0,1 0,2 0,3 0,4 0,5 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 A B C 1480700 1580700 1680700 1780700 1880700 1980700 2080700 2180700 2280700 2380700 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 A B C
  • 126. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 126Scenarios – A, B and C Model simulation and results 0 1000 2000 3000 4000 5000 6000 A B C 0 50 100 150 200 250 300 350 400 201020122014201620182020 A B C 0 200 400 600 800 1000 A B C 0 500 1000 1500 2000 2500 3000 A B C (a) Extensive Capitalizing (c) Intensive (d) Abandoning Rural Areas (b) Extensive Expansive
  • 127. CONCLUSION
  • 128. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 128 Using empirical ABM for socio-ecological system is a great challenge: “Agent-based modelling meets an intuitive desire to explicitly represent human decision making. (...) The question is whether the benefits of that approach to spatial modelling exceed the considerable costs of the added dimensions of complexity introduced into the modelling effort.” Helen Couclelis, 2002. We believe that the ideas of institutional arrangements and hybrid automata can help reduce the complexity of agent-based modelling of land change. Conclusion
  • 129. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 129Conclusion •  The thesis shows how to express the ideas of strategy and institutional arrangements in computer simulations. •  Strategies are loosely coupled to agents and may vary during the simulation in response to changes in local context and institutional arrangements. •  The hybrid automata formalism (available in TerraME) turned out to be a good way to express decision-making when agents change their strategies over time. Conclusion
  • 130. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 130Conclusion •  This thesis presents a method for building agent-based models of land change. Most agent-based models in the literature deal with individual decisions in small areas, based on field surveys. •  Our approach extends agent-based models to larger areas, based on collective behaviour. •  We assume that farmers have a limited set of strategies to manage land. Over time, a farmer may change his strategy depending on external conditions. Conclusion
  • 131. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 131Future work •  This model was very important to validate our concepts. however, this work is an is a first step of a long journey where we may suggest the following future work •  Calibration of pasture degradation submodel. •  Land use submodel should consider other land use types, allowing build different scenarious. •  Include actors, like miners and loggers. •  Road submodel, creation of roads should be modeled as endogenous process in future version. •  Connection to market. •  Normative agents Conclusion
  • 132. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 132 Obrigado
  • 133. Using institutional arrangements and hybrid automata for regional scale agent-based modeling of land change 133 StrategiesandJump Conditions State Jump Conditions Migrate 1. When agent gets a farm and has little capital, jump to Extensive Farming. 2. When agent gets a farm and has much capital, jump to Extensive Expanding. 3. If agent is risk prone, jump to Speculate. Extensive Farming 1. If agent gets capital, jump to Extensive Expanding. 2. If law enforcement is high, pasture is degraded and there is credit for small farmers, jump to Intensive Farming. 3. If law enforcement is high, pasture is degraded and there is no credit for small farmers, jump to Abandon Rural Activity. Extensive Expanding 1. If law enforcement is high, pasture is degraded and there is credit for large farmers, jump to Intensive Farming. 2. If law enforcement is high, pasture is degraded and there is no credit for large farmers, jump to Abandon Rural Activity. Speculate • If law enforcement is high, jump to Abandon Rural Activity. Intensive Farming • If there is no credit for large farmers and the beef market chain is weak for more than 5 years, jump to Abandon Rural Activity. Abandoning Rural Activity (final state, no jump conditions)

×