Using GIS to Assess Parcel Suitability for Potential Park Locations Matthew Mullenax, GISP Berkeley County Planning Commission
Using GIS to Assess Parcel Suitability for
Potential Park Locations
Matthew Mullenax, GISPBerkeley County Planning Commission
Study Background: 2006 Census estimates list Berkeley County as
one of 100 fastest growing counties in U.S. Growth has had impact on variety of public
services, notably parks and recreation. Berkeley County looking to develop long-range
strategy for park land acquisition. GIS can be used to help locate potential park
sites via suitability modeling. Pilot project performed on Martinsburg, WV.
Suitability Modeling
A measure of the relative usefulness of a land unit for some given purpose.
Typically used to locate something. If designed properly, the model results in
potential locations being identified and assigned a relative suitability score for the activity.
Breaks a big problem down into smaller, easily-solved goals which are combined for a measure of suitability.
Suitability Modeling and GIS
Suitability Modeling and GIS
Weighted combination is one method of suitability modeling.
Suitability Modeling and GIS
Weighted combination is one method of suitability modeling.
Weighted Sum Tool available with ESRI ArcGIS 9.2 Spatial Analyst extension license.
Suitability Modeling and GIS
Weighted combination is one method of suitability modeling.
Weighted Sum Tool available with ESRI ArcGIS 9.2 Spatial Analyst extension license.
Three major steps must be completed before running the tool:
1. Select evaluation or utility scale.2. Set scale or suitability values for
inputs.3. Assign weights to inputs.
Building the model
Appropriate evaluation criteria:
1. Population2. Existing parks3. Roads
Determine suitability of each input
1. Population – high population is preferred2. Existing parks – areas distant from existing
parks preferred3. Roads – close proximity to major roads is
preferred
1. Select evaluation or utility scale
Evaluation or utility scale works to normalize all input raster datasets.
A linear scale of 1 to 9 by 1 has proven to be workable.
>9 values difficult to visually comprehend; <9 values can decrease sensitivity
1 = lowest suitability2 = very low suitability3 = low suitability4 = moderately low suitability5 = moderate suitability6 = moderately high suitability7 = high suitability8 = very high suitability9 = highest suitability
Range of Suitability Values on Utility Scale:
2. Set scale or suitability values for inputs. The cells of each input raster in the analysis are
assigned suitability values from the utility scale. The assignment of suitability values for a single
layer is called Single Utility Assignment (SUA).
Population input
Central Dispatch address locations Perform Density calculation Use Reclassify tool to assign suitability values
on utility scale (1 to 9 by 1)
Set scale values for Population input
Reclassify
9 – 11 – 158 – 7.9 – 107 – 6.2 – 7.8 6 – 4.8 – 6.15 – 3.6 – 4.74 – 2.5 – 3.53 – 1.6 – 2.42 – 0.52 – 1.51 – 0 – 0.51
Existing Parks input
Existing Martinsburg parks (9) Perform Distance calculation Use Reclassify tool to assign suitability values on
utility scale (1 to 9 by 1)
Set scale values for Parks input
9 – > 5280 feet8 – 4621 - 52807 – 3961 - 46206 – 3301 - 39605 – 2641 - 33004 – 1981 - 26403 – 1321 - 19802 – 661 - 13201 – 0 – 660
Reclassify
Roads input
Major roads (Routes 9, 11, and 45, King and Queen Streets, Edwin Miller Boulevard, Tavern Road)
Perform Distance calculation Use Reclassify tool to assign suitability values on
utility scale (1 to 9 by 1)
Set scale values for Roads input
9 – 0 – 660 feet8 – 661 – 13207 – 1321 – 19806 – 1981 – 26405 – 2641 – 33004 – 3301 – 39603 – 3961 – 46202 – 4621 – 52801 – > 5280
Reclassify
3. Assign weights to inputs Each input is weighted,
or assigned a percentage influence, based on its importance.
The total influence for all inputs must equal 100 percent.
Total Suitability = (criteria1 * weight1) + (criteria2 * weight2) + ... + (criterian * weightn)
3. Assign weights to inputs Each input is weighted,
or assigned a percentage influence, based on its importance.
The total influence for all inputs must equal 100 percent.
Total Suitability = (criteria1 * weight1) + (criteria2 * weight2) + ... + (criterian * weightn)
(10%)
(45%)
SUAs combined to create Multiple Utility Assignment (MUA)
(45%)
Results in Suitability Map
Implementing the Suitability Map by Parcel Assessment
Implementing the Suitability Map by Parcel Assessment Parcel assessment
conducted using Zonal Statistics as Table Tool.
Implementing the Suitability Map by Parcel Assessment Parcel assessment
conducted using Zonal Statistics as Table Tool.
Tool calculates a variety of statistics on values of a raster within the “zones” of another dataset.
Using Zonal Statistics as Table Tool
Zone Layer is Parcel data.
Value Raster is MUA. Creates table of
statistics joined to Parcel data by unique PIN#.
Parcel Assessment
Parcels mapped by MAX suitability value.
Use of MEAN causes smaller parcels to have higher values overall than larger parcels.
Parcel Assessment (cont’d)
Select all parcels with score > 6.75. Five areas identified. First area apartment complex adjacent to school
playground & church ball field (Martin’s Landing). Second area apartment complex in commercial
area (Courthouse Apartments). Third area in large, phased subdivision in recently
annexed West Side area (The Gallery). Fourth area apartment complex with playground
(Fountainhead Apartments). Fifth area Downtown.
Parcel Assessment (cont’d)
Inspect selected parcels ownership information, current use, etc. and determine feasibility.
If parcels unfeasible, run new query using next highest suitability values and re-perform inspection on newly selected parcels.
Study Limitations/Future Considerations Spatial accuracy of Roads and Address inputs in
few small areas. Lack of separation between residential vs. non-
residential addresses. Parks layer does not include school, church, and/or
subdivision improved open spaces. County inholdings, right-of-ways, municipal
boundary shape have potential to influence results. Method of weight determination not overly rigorous. Consider creating distance grid from all roads vs.
only major roads.
Acknowledgements:
Steve CatlettMartinsburg-Berkeley County Parks & Recreation
Questions?
Matthew Mullenax, GISP GIS Coordinator Berkeley County Planning Commission 400 West Stephen Street, Suite 203 Martinsburg, WV 25401 PH - (304) 267-5107 FX - (304) 262-3127 [email protected]