Agent Based Modeling and GIS for Community Resource Management: - - PowerPoint PPT Presentation

agent based modeling and gis for community resource
SMART_READER_LITE
LIVE PREVIEW

Agent Based Modeling and GIS for Community Resource Management: - - PowerPoint PPT Presentation

Agent Based Modeling and GIS for Community Resource Management: Acequia-based Agriculture Sarah Wise & Andrew Crooks Department of Computational Social Science swise5@gmu.edu & acrooks2@gmu.edu What are Acequias? Traditional


slide-1
SLIDE 1

Agent Based Modeling and GIS for Community Resource Management: Acequia-based Agriculture

Sarah Wise & Andrew Crooks

Department of Computational Social Science swise5@gmu.edu & acrooks2@gmu.edu

slide-2
SLIDE 2

What are Acequias?

  • Traditional system of water management

in northern New Mexico

  • Coupled socio-physical system of ditches

and ditch maintenance

  • Committee of water right holders with

precise legal conventions for passing on water versus land rights

slide-3
SLIDE 3

Purpose of the Model

  • Interim goal: to study the viability of acequia-

based agriculture with regard to variations in:

– population pressure – short-term climate – cultural valuation

  • Ultimate goal: to support policy work which

requires an understanding of the importance and impact of acequias

slide-4
SLIDE 4

Study Area: Taos, New Mexico

  • Population: 5,062
  • Acreage - 1,444,480 acres in county

– 50.0% Federally controlled – 6.7% State-owned – 4.3% Native American land – 38.2% Privately owned

  • Acequias built by 1696, town of Taos formally

settled in 1725

All information from Taos County Chamber of Commerce

slide-5
SLIDE 5

Source: Google Maps

slide-6
SLIDE 6

Source: Google Maps

slide-7
SLIDE 7

Source: Google Maps

slide-8
SLIDE 8

Source: Google Maps

slide-9
SLIDE 9

Data

  • 1. Acequia location and
  • wnership
  • 2. Land use
  • 3. Location of urban centers
  • 4. Roads
  • 6. Slope/elevation
  • 7. Hydrological system
  • 8. Population demographics
  • Spatial: a raster space measured at 30m sq
  • Temporal: one tick per year, for 100 ticks

Resolution

slide-10
SLIDE 10
slide-11
SLIDE 11
slide-12
SLIDE 12
slide-13
SLIDE 13
slide-14
SLIDE 14
slide-15
SLIDE 15
slide-16
SLIDE 16
slide-17
SLIDE 17
slide-18
SLIDE 18

Real-world spatial, temporal, and behavioral processes represented Corresponding modeling methodologies implemented Cultivation and sale of crops Linear programming Real estate market Agent-based model Hydrological system Network Acequia participation (maintenance of ditches, use of water, etc.) Agent-based model

slide-19
SLIDE 19

Crop Model

If cropland is adjacent to an acequia which experiences at least some minimum waterflow, crops will grow. Income = Cropprice * Parcelsize

slide-20
SLIDE 20

Hydrological Model

  • Rainfall

– Fed by a uniform application of rainfall across the environment once a tick

  • Propagation

– Water flows downhill until it reaches a hydrological feature, and then along the given (elevation-derived) network. Water flows through the acequia network depending on the permeability of the links

  • Sedimentation

– Every year, acequias experience sedimentation and their permeability decreases slightly. This process is reversed by the Parciantes’ annual cleaning efforts

slide-21
SLIDE 21

Real Estate Model

  • Real estate agents pick random Parciantes and

formulate their bids according to the following properties of the chosen Parciante’s land:

– Constraint = Budget – Transport Cost – Utility = Amenityα * Distanceβ – Offer = Constraint * Utility2/(weight2 + Utility2)

  • Approach taken from Filatova et al’s (2009)

model of land markets

slide-22
SLIDE 22

Acequia Participation Model

  • Every tick, each Parciante agent must pay to maintain

its acequia rights. The cost is proportional to how many rights the agent has, the length of acequia to maintain, and the number of other Parciantes helping them.

Cost = Weight * Rightsnum * Acequialength/ Acequiamembers

  • Parciante agents have different cultural valuations of

their participation in the acequia system.

slide-23
SLIDE 23

Interactions between different kinds of agents

slide-24
SLIDE 24

Verification

  • The model was built

as part of an iterative process, so that each component was tested as it was developed

  • Example:

– The water network

slide-25
SLIDE 25

Verification

  • The model was built

as part of an iterative process, so that each component was tested as it was developed

  • Example:

– The water network

slide-26
SLIDE 26

Sensitivity Test

1.Measured

– Conversion of Agricultural land to Urban – Change in number of Parciantes – Change in number of functioning Acequias

2.Found

– All parameters had the expected impact on the measured variables

slide-27
SLIDE 27

Results

  • Comparison of three scenarios:

1.Traditionalists with low costs 2.Prodigal Sons with high costs 3.Followers with medium costs

  • Constant rainfall, demand profile, and

physical profile

slide-28
SLIDE 28

Traditionalists

slide-29
SLIDE 29

Traditionalists

slide-30
SLIDE 30

Traditionalists

slide-31
SLIDE 31

Traditionalists

slide-32
SLIDE 32

Prodigal Sons

slide-33
SLIDE 33

Prodigal Sons

slide-34
SLIDE 34

Prodigal Sons

slide-35
SLIDE 35

Prodigal Sons

slide-36
SLIDE 36

Followers

slide-37
SLIDE 37

Followers

slide-38
SLIDE 38

Followers

slide-39
SLIDE 39

Conclusions

  • The acequia system is sensitive to variations

in cultural valuation among its members

  • Changing the cost of acequia maintenance

has a significant impact on how tenable farming is as a lifestyle

  • Weather, population pressure in the form of

demand for land, and crop prices influence the system overall

slide-40
SLIDE 40

Thank you! Comments, questions, suggestions?

Emails: swise5@gmu.edu Web: css.gmu.edu acrooks2@gmu.edu socialcomplexity.gmu.edu gisagents.blogspot.com

slide-41
SLIDE 41

Acknowledgements

  • The authors would like to acknowledge of the Department
  • f Computational Social Science and the Center for Social

Complexity at George Mason for providing support for this research.

  • We are deeply grateful to Michael Cox for sharing his data
  • n the Taos, New Mexico acequia system with us.
  • This work is inspired by the research of John Paul

Gonzales of the Santa Fe Institute, who is himself an acequia parciante.