1. © K.Fedra 20004 1 Dynamic Land Use Change Modeling A simple spatial modeling application

2. © K.Fedra 20004 2 Land Use Change modeling Land cover: Natural land surface, vegetation Land use: Economic interpretation and classification of land cover Land cover: Natural land surface, vegetation Land use: Economic interpretation and classification of land cover

3. © K.Fedra 20004 3 Land Use Change modeling Describes the change of land use or land cover over time. Provides inputs and boundary conditions for other estimates such as: Population developmentPopulation development Regional economy, GRPRegional economy, GRP Resource requirements (water, energy)Resource requirements (water, energy) Waste and environmental pollutionWaste and environmental pollution Describes the change of land use or land cover over time. Provides inputs and boundary conditions for other estimates such as: Population developmentPopulation development Regional economy, GRPRegional economy, GRP Resource requirements (water, energy)Resource requirements (water, energy) Waste and environmental pollutionWaste and environmental pollution

4. © K.Fedra 20004 4 Land Use Change modeling How to model land use change ? Divide the area under study into small cells or units (a grid) of homogeneous land use each; Describe the evolution of each cell as a sequence of discrete states (land use) over time. How to model land use change ? Divide the area under study into small cells or units (a grid) of homogeneous land use each; Describe the evolution of each cell as a sequence of discrete states (land use) over time.

5. © K.Fedra 20004 5 Land Use Change modeling Turing machine (general purpose computer) A.M.Turing (1936) John von Neumann (1950): Self-reproducing machines Cellular automata: J.H. Conway: LIFE (game) Turing machine (general purpose computer) A.M.Turing (1936) John von Neumann (1950): Self-reproducing machines Cellular automata: J.H. Conway: LIFE (game)

6. © K.Fedra 20004 6 Land Use Change modeling Cellular automata:J.H. Conway: LIFE (game) Complex behavior with only two very simple rules: Survival: with 2 or 3 neighbors Death: less than2, more than 3 Birth: empty field with 3 neighbors/ Cellular automata:J.H. Conway: LIFE (game) Complex behavior with only two very simple rules: Survival: with 2 or 3 neighbors Death: less than2, more than 3 Birth: empty field with 3 neighbors/

7. © K.Fedra 20004 7 Land Use Change modeling

8. © K.Fedra 20004 8 Land Use Change modeling Continuously moving GLIDER

9. © K.Fedra 20004 9 Land Use Change modeling Simple dynamic modelSimple dynamic model Spatially distributedSpatially distributed Individual cells (objects, parcels of land) show SIMPLE behaviorIndividual cells (objects, parcels of land) show SIMPLE behavior Complexity through interaction in space and timeComplexity through interaction in space and time Simple dynamic modelSimple dynamic model Spatially distributedSpatially distributed Individual cells (objects, parcels of land) show SIMPLE behaviorIndividual cells (objects, parcels of land) show SIMPLE behavior Complexity through interaction in space and timeComplexity through interaction in space and time

10. © K.Fedra 20004 10 Land Use Change modeling Basic modeling principle: State transition (traffic light)State transition (traffic light) Markov chain: a series of random events or states from a given set, each determined only by its predecessor (A.Markov, 1856-1922).Markov chain: a series of random events or states from a given set, each determined only by its predecessor (A.Markov, 1856-1922). Basic modeling principle: State transition (traffic light)State transition (traffic light) Markov chain: a series of random events or states from a given set, each determined only by its predecessor (A.Markov, 1856-1922).Markov chain: a series of random events or states from a given set, each determined only by its predecessor (A.Markov, 1856-1922).

11. © K.Fedra 20004 11 Land Use Change modeling Complexity through interaction in space and time, probabilistic state transitions: Transitions depend on neighborhood and history: if each cell shares information on all other cells intended transitions, it may change its own naïve strategy. EXAMPLE: urban development, building apartment blocks (planning and zoning !) Complexity through interaction in space and time, probabilistic state transitions: Transitions depend on neighborhood and history: if each cell shares information on all other cells intended transitions, it may change its own naïve strategy. EXAMPLE: urban development, building apartment blocks (planning and zoning !)

12. © K.Fedra 20004 12 Land Use Change modeling What makes land use change ? 1.Demographic development 2.Economic, technological development (incl. pollution, exhaustion) 3.Political development (regional planning, borders, war) 4.Climate change (suitability) What makes land use change ? 1.Demographic development 2.Economic, technological development (incl. pollution, exhaustion) 3.Political development (regional planning, borders, war) 4.Climate change (suitability)

13. © K.Fedra 20004 13 Land Use Change modeling Typical time scale: Decades Regulatory framework: Land use plan, zoning Building laws Ownership, property and inheritance laws (agriculture) Typical time scale: Decades Regulatory framework: Land use plan, zoning Building laws Ownership, property and inheritance laws (agriculture)

14. © K.Fedra 20004 14 Land Use Change modeling Model components: 1.Land use classes, initial conditions (map) 2.Transition matrix (probabilities) 3.RULES modifying the probabilities expressing global or regional constraints Model components: 1.Land use classes, initial conditions (map) 2.Transition matrix (probabilities) 3.RULES modifying the probabilities expressing global or regional constraints

15. © K.Fedra 20004 15 Land Use Change modeling CORINE land use classes: 1.1.Urban fabric 1.2 Industrial, commercial and transport units 1.3 Mine, dump and construction sites 1.4 Artificial non-agricultural vegetated areas CORINE land use classes: 1.1.Urban fabric 1.2 Industrial, commercial and transport units 1.3 Mine, dump and construction sites 1.4 Artificial non-agricultural vegetated areas

16. © K.Fedra 20004 16 Land Use Change modeling CORINE land use classes: 1.1.Urban fabric 1.1.1. Continuous urban fabric 1.1.2. Discontinuous urban fabric CORINE land use classes: 1.1.Urban fabric 1.1.1. Continuous urban fabric 1.1.2. Discontinuous urban fabric

17. © K.Fedra 20004 17 Land Use Change modeling CORINE land use classes: 2.1 Arable land 2.2 Permanent crops 2.3 Pastures 2.4 Heterogeneous agricultural areas 3.1 Forest 3.2 Shrubs and/or herbaceous vegetation associations 3.3 Open spaces with little or no vegetation CORINE land use classes: 2.1 Arable land 2.2 Permanent crops 2.3 Pastures 2.4 Heterogeneous agricultural areas 3.1 Forest 3.2 Shrubs and/or herbaceous vegetation associations 3.3 Open spaces with little or no vegetation

18. © K.Fedra 20004 18 Land Use Change modeling CORINE land use classes: 4.1 Inland wetlands 4.2 Coastal wetlands 5.1 Inland waters 5.2 Marine waters CORINE land use classes: 4.1 Inland wetlands 4.2 Coastal wetlands 5.1 Inland waters 5.2 Marine waters

19. © K.Fedra 20004 19 Land Use Change modeling Starting point: a historical land use map: Batroun/Tripoli, Northern Lebanon Starting point: a historical land use map: Batroun/Tripoli, Northern Lebanon

20. © K.Fedra 20004 20

21. © K.Fedra 20004 21 Land Use Change modeling

22. © K.Fedra 20004 22 Land Use Change modeling Shrubs, herbaceous veg. 17.0 Permanent crops 14.7 Forest, mixed 7.9 Heterogeneous agriculture: 2.9 Urban fabric 2.8 Open space, little veg. 2.7 Mines. dumps, construction 0.8 Industry, commerce, transport 0.4 Coastal wetlands 0.1 Unclassified 50.7 Shrubs, herbaceous veg. 17.0 Permanent crops 14.7 Forest, mixed 7.9 Heterogeneous agriculture: 2.9 Urban fabric 2.8 Open space, little veg. 2.7 Mines. dumps, construction 0.8 Industry, commerce, transport 0.4 Coastal wetlands 0.1 Unclassified 50.7

23. © K.Fedra 20004 23 Land Use Change modeling The mechanisms: 1.State transition matrix (simple Markov Model) 2.Rules (shared information described by first order production rules) The mechanisms: 1.State transition matrix (simple Markov Model) 2.Rules (shared information described by first order production rules)

24. © K.Fedra 20004 Land Use Change modeling state transition matrix: S1 S2 S3 S4 S1 S2 S3 S4 S1 0.74 0.01 0.20 0.05 ∑=1.0 S1 0.74 0.01 0.20 0.05 ∑=1.0 S2 0.01 0.98 0.00 0.01 S2 0.01 0.98 0.00 0.01 S3 0.00 0.02 0.93 0.05 S3 0.00 0.02 0.93 0.05 S4 0.10 0.01 0.02 0.87 S4 0.10 0.01 0.02 0.87 state transition matrix: S1 S2 S3 S4 S1 S2 S3 S4 S1 0.74 0.01 0.20 0.05 ∑=1.0 S1 0.74 0.01 0.20 0.05 ∑=1.0 S2 0.01 0.98 0.00 0.01 S2 0.01 0.98 0.00 0.01 S3 0.00 0.02 0.93 0.05 S3 0.00 0.02 0.93 0.05 S4 0.10 0.01 0.02 0.87 S4 0.10 0.01 0.02 0.87

25. © K.Fedra 20004 25 Land Use Change modeling 1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 3.1 3.2 3.3 4.1 4.2 1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 3.1 3.2 3.3 4.1 4.2 1.10.86 0.05 0.02 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.00 1.20.01 0.95 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00 1.30.05 0.10 0.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.40.01 0.00 0.00 0.99 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.10.10 0.05 0.00 0.10 0.40 0.10 0.05 0.10 0.05 0.05 0.00 0.00 0.00 2.20.10 0.05 0.05 0.05 0.05 0.56 0.05 0.05 0.01 0.01 0.02 0.00 0.00 2.30.10 0.05 0.01 0.05 0.01 0.10 0.55 0.10 0.01 0.01 0.01 0.00 0.00 2.40.10 0.05 0.01 0.05 0.05 0.10 0.11 0.50 0.01 0.01 0.01 0.00 0.00 3.10.05 0.02 0.01 0.01 0.05 0.01 0.01 0.01 0.82 0.00 0.01 0.00 0.00 3.20.05 0.02 0.01 0.04 0.00 0.01 0.02 0.02 0.03 0.80 0.00 0.00 0.00 3.30.06 0.02 0.01 0.03 0.01 0.01 0.03 0.02 0.00 0.01 0.80 0.00 0.00 4.10.00 0.00 0.01 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.03 0.94 0.00 4.20.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.98 1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 3.1 3.2 3.3 4.1 4.2 1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 3.1 3.2 3.3 4.1 4.2 1.10.86 0.05 0.02 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.00 1.20.01 0.95 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00 1.30.05 0.10 0.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.40.01 0.00 0.00 0.99 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.10.10 0.05 0.00 0.10 0.40 0.10 0.05 0.10 0.05 0.05 0.00 0.00 0.00 2.20.10 0.05 0.05 0.05 0.05 0.56 0.05 0.05 0.01 0.01 0.02 0.00 0.00 2.30.10 0.05 0.01 0.05 0.01 0.10 0.55 0.10 0.01 0.01 0.01 0.00 0.00 2.40.10 0.05 0.01 0.05 0.05 0.10 0.11 0.50 0.01 0.01 0.01 0.00 0.00 3.10.05 0.02 0.01 0.01 0.05 0.01 0.01 0.01 0.82 0.00 0.01 0.00 0.00 3.20.05 0.02 0.01 0.04 0.00 0.01 0.02 0.02 0.03 0.80 0.00 0.00 0.00 3.30.06 0.02 0.01 0.03 0.01 0.01 0.03 0.02 0.00 0.01 0.80 0.00 0.00 4.10.00 0.00 0.01 0.00 0.00 0.00 0.00 0.01 0.00 0.01 0.03 0.94 0.00 4.20.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.98

26. © K.Fedra 20004 26 Land Use Change modeling State transition matrix: 1.Educated guesses (hypotheses) 2.Estimated from a series of land use maps (e.g., from satellite imagery) to estimate transition frequencies. State transition matrix: 1.Educated guesses (hypotheses) 2.Estimated from a series of land use maps (e.g., from satellite imagery) to estimate transition frequencies.

27. © K.Fedra 20004 27 Land Use Change modeling Transition probabilities: Source: shrub, meadows Targets: Pastures, grazingPastures, grazing Agriculture (irrigated, rain fed)Agriculture (irrigated, rain fed) Horticulture, orchards, wine, ….Horticulture, orchards, wine, …. Urban fabric (housing)Urban fabric (housing) Industrial/commercial areasIndustrial/commercial areas Transition probabilities: Source: shrub, meadows Targets: Pastures, grazingPastures, grazing Agriculture (irrigated, rain fed)Agriculture (irrigated, rain fed) Horticulture, orchards, wine, ….Horticulture, orchards, wine, …. Urban fabric (housing)Urban fabric (housing) Industrial/commercial areasIndustrial/commercial areas

28. © K.Fedra 20004 Land Use Change modeling A priori transition probabilities are CONTINGENT on 1.Global states (e.g., %S1) 2.Local States: Temporal (history, memory) Spatial (neighborhood) A priori transition probabilities are CONTINGENT on 1.Global states (e.g., %S1) 2.Local States: Temporal (history, memory) Spatial (neighborhood)

29. © K.Fedra 20004 Land Use Change modeling Global/local adjustments of the transition probabilities in relative terms (p= 0.50; 10% increase p = 0.55 in absolute terms (p = 0.50 increase by 10% p = 0.60) absolute (p = 0.50; set to 10% p = 0.10) Global/local adjustments of the transition probabilities in relative terms (p= 0.50; 10% increase p = 0.55 in absolute terms (p = 0.50 increase by 10% p = 0.60) absolute (p = 0.50; set to 10% p = 0.10)

30. © K.Fedra 20004 30 Land Use Change modeling Operators: 1.FRACTION (in space: describes the neighborhood) 2.FREQUENCY (in time: describes the history) 3.LAST (in time: checks for specific events) Operators: 1.FRACTION (in space: describes the neighborhood) 2.FREQUENCY (in time: describes the history) 3.LAST (in time: checks for specific events)

31. © K.Fedra 20004 31 Land Use Change modeling Example Concepts: 1.Cities are more likely to expand than to start in the middle of nowhere. 2.A dense city is likely to retain some last green areas like parks. Example Concepts: 1.Cities are more likely to expand than to start in the middle of nowhere. 2.A dense city is likely to retain some last green areas like parks.

32. © K.Fedra 20004 32 Land Use Change modeling Cities are more likely to expand than to start in the middle of nowhere: IF FRACTION(city,1) < 10% THEN p(*,city) DECREASES 90% If in the immediate surrounding of an area (8 neighbors) there is not at least 1 cell that is already part of a city, the probability of any source class to become city is reduced to 1/10. Cities are more likely to expand than to start in the middle of nowhere: IF FRACTION(city,1) < 10% THEN p(*,city) DECREASES 90% If in the immediate surrounding of an area (8 neighbors) there is not at least 1 cell that is already part of a city, the probability of any source class to become city is reduced to 1/10.

33. © K.Fedra 20004 33 Land Use Change modeling A dense city is likely to retain some last green areas like parks. IF FRACTION(city,2) > 95 THEN p(*,city) ABSOLUTE 0 If there is not at least one free green area in radius of 2 units (a 5by5 neighborhood) around a given plot, no transition to city is possible. A dense city is likely to retain some last green areas like parks. IF FRACTION(city,2) > 95 THEN p(*,city) ABSOLUTE 0 If there is not at least one free green area in radius of 2 units (a 5by5 neighborhood) around a given plot, no transition to city is possible.

34. © K.Fedra 20004 34 Land Use Change modeling Other attributes that can affect potential land use: Elevation, slopeElevation, slope Terrain, reliefTerrain, relief Soil, geologySoil, geology Climate (water)Climate (water) Infrastructure (transport, energy)Infrastructure (transport, energy) Other attributes that can affect potential land use: Elevation, slopeElevation, slope Terrain, reliefTerrain, relief Soil, geologySoil, geology Climate (water)Climate (water) Infrastructure (transport, energy)Infrastructure (transport, energy)

35. © K.Fedra 20004 35 Land Use Change modeling Other driving forces: Population development, pressuresPopulation development, pressures Migration (jobs, income, unemployment, conflicts)Migration (jobs, income, unemployment, conflicts) GRP, property pricesGRP, property prices Other driving forces: Population development, pressuresPopulation development, pressures Migration (jobs, income, unemployment, conflicts)Migration (jobs, income, unemployment, conflicts) GRP, property pricesGRP, property prices

36. © K.Fedra 20004 36 Land Use Change modeling Related analysis: Employment opportunitiesEmployment opportunities GRP, revenues, taxes, incomeGRP, revenues, taxes, income Resource consumption (water, energy)Resource consumption (water, energy) Environmental impacts (waste streams)Environmental impacts (waste streams) Based on land use specific processes or activity specific coefficients Related analysis: Employment opportunitiesEmployment opportunities GRP, revenues, taxes, incomeGRP, revenues, taxes, income Resource consumption (water, energy)Resource consumption (water, energy) Environmental impacts (waste streams)Environmental impacts (waste streams) Based on land use specific processes or activity specific coefficients

37. © K.Fedra 20004 37 Land Use Change modeling Related analysis or models: Traffic systemsTraffic systems Forest fireForest fire DesertificationDesertification Avalanches, mud slidesAvalanches, mud slides Development of a infrastructures:Development of a infrastructures: –Road network, pipelines, district heating Locational analysis (site suitability: schools, hospitals, airports, supermarkets …..)Locational analysis (site suitability: schools, hospitals, airports, supermarkets …..) Related analysis or models: Traffic systemsTraffic systems Forest fireForest fire DesertificationDesertification Avalanches, mud slidesAvalanches, mud slides Development of a infrastructures:Development of a infrastructures: –Road network, pipelines, district heating Locational analysis (site suitability: schools, hospitals, airports, supermarkets …..)Locational analysis (site suitability: schools, hospitals, airports, supermarkets …..)

38. © K.Fedra 20004 38 Land Use Change modeling Assignment: Build a LUC scenario: Use the Lebanon exampleUse the Lebanon exampleOR Use your own real or hypothetical initial conditionsUse your own real or hypothetical initial conditionsAssignment: Build a LUC scenario: Use the Lebanon exampleUse the Lebanon exampleOR Use your own real or hypothetical initial conditionsUse your own real or hypothetical initial conditions

39. © K.Fedra 20004 39 Land Use Change modeling Assignment: Develop a transition matrix Develop a set of RULES, explain the underlying ideas and principles ! Suggest improvements to the model. Assignment: Develop a transition matrix Develop a set of RULES, explain the underlying ideas and principles ! Suggest improvements to the model.

40. © K.Fedra 20004 40 Land Use Change modeling Assignment: LUC Details and material on: http://80.120.147.30/LUC/ http://www.ess.co.at/SMART/luc.html Assignment: LUC Details and material on: http://80.120.147.30/LUC/ http://www.ess.co.at/SMART/luc.html