Modeling Future Forest Fragmentation - UConn CLEAR

Modeling Future Forest Fragmentation - UConn CLEAR

Simulating Future Suburban Development in Connecticut Jason Parent, Daniel Civco, and James Hurd [email protected] Center for Land Use Education and Research Dept. of Natural Resources Management and Engineering University of Connecticut Study Objectives Simulate suburban development over the next 30 years Analyze impact of simulated development on forest fragmentation 2

The Study Area decreased amount decreased contiguity decreased amount and contiguity no significant change Salmon River Watershed 3 Study Area Properties 150 sq. miles in area Heavily forested (approx. 72% in 2002) Population is growing 18.6% increase between 1990 and 2000 Urban and associated land cover is increasing Urban land increase of 16.6% between 1985 and 2002

Turf / grass increase of 18.6% between 1985 and 2002 4 General Approach Multi-Criteria Evaluation create suitability maps Buildout Analysis identify potential building sites TimeScope Analysis assign build dates to potential building sites 5 General Approach Buffer potential buildings account for land cover change associated with building Update land cover map Forest Fragmentation Model (Riitters et al. 2000)

identify types of forest present with respect to fragmentation 6 Multi-Criteria Evaluation Based on grids depicting relevant criteria Two kinds of criteria: constraints and factors Factors Constraints Two categories (suitable, unsuitable) Grid cells have values of 0 or 1 0 = feature present (unsuitable)

1 = feature absent (suitable) i.e. water More than two categories Grid cells have a range of values (criterion scores): 0 to 100 100 = most suitable categories 0 = least suitable categories i.e. soil types 7 Multi-Criteria Evaluation Constraint Grids Floodzones

Hydrography Wetlands Floodzones (50 buffer) Hydrography (50 buffer) Wetlands (50 buffer) 8 Multi-Criteria Evaluation Constraint Grids DEP land

Municipal Open space Slope > 20% 9 Multi-Criteria Evaluation: Factors: Soil Types Criterion scores systematically assigned to each soil type Initially assigned maximum scores (100) to each soil type Reduced score for negative properties of soil types hydric septic potential min. slope rockiness

no high <8 not rocky 0 no medium 8 to 15 stony or rocky 10

no low or very low NA very stony or very rocky 20 no extremely low > 15 extremely stony 30

yes NA rock outcrop 100 NA score reduction 10 Multi-Criteria Evaluation: Soil Factor Shapefile depicting criterion scores for soil types

11 Multi-Criteria Evaluation: Roads Factor 970 910 850 790 730 670 610 550

490 430 370 310 250 190 130 70 7% 6% 5% 4%

3% 2% 1% 0% 10 Satisfy home owner preference Reduce development costs Mandatory setback percent developed Structures tend to be located within a certain range of distances from a road Building Distances to Roads (Bolton) distance to road

Maximum development approximately 100 feet from the nearest road 12 Multi-Criteria Evaluation: Roads Factor low high major roads Map distances from roads calculate % land developed

in each class Group into classes with 20 intervals scale from 0 to 100 Grid depicting criterion scores 13 Multi-Criteria Evaluation: Calculation of Suitability Values binary raster format (unsuitable = 0) Constraint 1 multiply

Constraint 2 all constraints (unsuitable = 0) Constraint 3 multiply suitability raster Factor 1 Factor 2 raster format; values range from 0 to 100 x

x weight1 add weighted factor combination weight2 Sum of weights = 1 14 Constraints (water, wetland) Roads factor Soils factor Suitability Map

15 Multi-Criteria Evaluation: Suitability Map (Bolton) unsuitable highly suitable 16 Buildout Analysis Community Vizs Scenario 360 Buildout tool Places points at all potential future building locations uses zoning information to determine lot size, building separation distance, etc. excludes constraint areas from analysis works with parcel data

17 Buildout Analysis Constraints Unsuitable land suitability value = 0 Fully developed parcels Parcels containing a structure and less than 240,000 sq ft (6 builders acres) 50 ft buffer around existing structures 18 Buildout Analysis Zoning Regulations Required zoning information: Zone name

Minimum lot size (acres) Building efficiency The percentage of available land that can be built upon Efficiency less than 100% because of roads, open space requirements, etc. Building separation distance Minimum distance between the center points of two buildings ZONE C DD I PO / R R-1 R-2 R-3 R-4 RL VC LOT SIZE

1 5 1 0.5 1.5 1.5 1.6 2.1 0 0.5 EFFICIENCY (%) 80 80 80 80 80 80 80 80 0

80 BLDG_SEPARATION 208 467 208 148 256 256 264 302 0 148 Zoning regulations for East Hampton 19 Buildout Analysis Results

20 TimeScope Analysis Community Vizs Scenario 360 TimeScope tool Assigns a build date to each buildout building Simulates the order in which building construction will occur over a specified period of time Based on parcel suitability Building growth rate is specified by user For a given time step (i.e. year), the value of the current time step is assigned to: A number of building locations, equal to the annual building growth rate Building locations for which a build date has not already been assigned Locations with the highest remaining suitability value 21

TimeScope Analysis Building Growth Rates Building growth was assumed to parallel population growth Census data indicate that population growth has been linear over the past 40 years Population extrapolated out to 2036 by linear regression of past census data Hebron: Past Populations 10000 Population 8000 6000 4000 2000 0

1960 1970 1980 Year 1990 2000 22 TimeScope Analysis Building Growth Rates Gt = Pt2036 - Ht2004

At 2036 - 2004 Gt = annual number of houses constructed Pt2036 = predicted population in 2036 At = average number of people per house in 2000 Ht2004 = number of houses in 2004 Estimated Population and Building Growth Town Bolton Colchester East Haddam East Hampton Hebron Marlborough Estimated pop. in 2030 6559 21000

11485 18591 13778 9193 Houses existing in 2004 1999 5167 4119 4856 3418 2017 Predicted # of houses in 2030 2572 7778 5548 6136 4974

3307 Persons per house in 2000 2.55 2.7 2.07 3.03 2.77 2.78 Predicted # of houses / year 22 100 55 49 60 50 23

TimeScope Analysis Results: Colchester 24 TimeScope Analysis Results: Colchester 25 Buildout Building Buffers 2005 - 2010 2011 - 2015 2016 - 2020 2021 - 2025 2026 - 2030 2031 - 2036

after 2036 26 Future Land Cover Maps Forecasted land cover for 2010, 2015, 2020, 2025, 2030, and 2036 2002 land cover base map Derived from Landsat satellite imagery Selected buffers, with the appropriate build dates, for each forecast year Land cover grid cells, within the selected buffers, were converted to urban land cover 27 Future Land Cover Maps: Marlborough 28

Forest Fragmentation Analysis Model developed by Riitters et al. (2000) Identifies forest grid cells as one of 5 types based on the percentage of forest grid cells and connectivity of forest grid cells in the surrounding area: Interior: all surrounding grid cells are forest Edge: grid cell is on the exterior edge of a forest tract Perforated: the interior edge of forest tract Transitional: about half of the surrounding grid cells are forest Patch: less than 40% of surrounding grid cells are forest In this study, edge and perforated forest types were within 60 meters of the forest perimeter

29 Forest Fragmentation Maps: East Haddam 30 Changes Predicted to Occur, between 2002 and 2036, in Land Cover 3% of forest cover will be converted to non- forested land cover Urban land and associated turf will increase by approximately 18% Agricultural land will decline by approximately 5.6% 31 Changes Predicted to Occur between 2002 and 2036 in

Forest Fragmentation Interior forest will decline by 28% Perforated, transitional, and patch forest will increase by 67%, 10%, and 8% respectively Edge forest will decline by 15.5% 32 Discussion Analysis does not account for road construction Estimates of forest cover change are conservative Perforated forest area is over-estimated while edge forest area is underestimated Building growth rates used in the TimeScope analysis are probably not applicable to non-residential zones

The effect in this study should be minimal since the towns had little non-residential area The analysis is applicable to regions in which the major forest fragmenting process is suburban development. 33 Future Work Incorporate a model to predict road development Include socioeconomic data into the derivation of the suitability maps Identify land availability Derive building growth rate estimates applicable to non-residential zones Compare of the effects of current zoning scenarios with low impact zoning scenarios

34 Questions? Simulating Future Suburban Development in Connecticut Jason Parent, Daniel Civco, and James Hurd [email protected] Center for Land Use Education and Research Dept. of Natural Resources Management and Engineering University of Connecticut

Recently Viewed Presentations

  • Apresentação do PowerPoint

    Apresentação do PowerPoint

    Tem esperança que a gente não engorde!! O diabetes começa muitos anos antes do açúcar aumentar no sangue. Começa em qualquer idade, quando a gente faz as três piores escolhas: engordar não fazer atividade física se alimentar de forma inadequada...
  • CS186: Introduction to Database Systems

    CS186: Introduction to Database Systems

    CS 405G: Introduction to Database Systems Lecture 2 : Database Design I
  • Introduction - bpums

    Introduction - bpums

    Introduction Fungi Fungus Mycology Mykes = Mushroom Medical mycology 5 kingdom Monera Plant Animal Fungi Protista Eukaryote & Heterotrophic Structure & Ultra structure EM Cytochemical stains Hyphae Yeast Hyphae Tube Solid wall Protoplasm No fixed length Diameter 1-30 μm (...
  • GW CyberCorps Program

    GW CyberCorps Program

    InChI - The Worldwide Chemical Structure Standard Stephen Heller NIST [email protected] Objectives: Develop a computer algorithm to produce unique label - Chemical Identifier (InChI)
  • Projection Methods - Rice University

    Projection Methods - Rice University

    Projection Methods (Symbolic tools we have used to do…) Ron Parr Duke University Joint work with: Carlos Guestrin (Stanford) Daphne Koller (Stanford)
  • Geothermal Energy Projects in Vojvodina

    Geothermal Energy Projects in Vojvodina

    Paris Basin, largest district heating network density in continental Europe. 34 doublets in activity, temperature ~70°C . 22 doublets are operated by Dalkia. Heat production >1100 MWh/y . Supplyheat for the equivalent of 300 000 homes. Over 40 years operating.
  • Classical Tragedy, Tragic Heroes, Catharsis

    Classical Tragedy, Tragic Heroes, Catharsis

    Students will be able to list common Tragic Flaws of a tragic hero. Students will be able to identify what makes a character the . tragic hero. Students will be aware of . Classical Tragedy, catharsis, and . Hubris. Students...
  • Memory and Learning

    Memory and Learning

    TIP: You can think of the three stages of memory processing in the following way: encoding is like listening to songs, consolidation is like recording those songs (or burning a CD), and retrieval is like playing back the songs. Retrieval....