Optimal School Location: Raster Multicriteria Analysis (UPDATED)

Geographic Data

Assignment 2: Raster Multicriteria Analysis (UPDATED) Optimal School Location The town is looking to locate a new school to serve its growing population. The mayor has convened three experts to pick the most optimal location: the school board commissioner, the city’s chief planner, and a real estate developer. Each of these stakeholders has different views on the criteria that influence their most optimal site location. In line with the literature on GIS-Multi-Criteria Evaluation/Analysis, you have surveyed the stakeholders to solicit their viewpoints and translated them into the following standardized weights: School Board Commissioner Chief Planner Real Estate Developer Land Use 25% 12% 40% Slope of terrain 15% 13% 30% Proximity to Existing Recreational Centres 30% 50% 15% Proximity to Existing Schools 30% 25% 15% Your task is to identify the most optimal location considering the views of all three stakeholders. Submission Submit your final assignment online via Quercus. Assignments are due by 11:59PM on October 27. Grades , which are all indicated across Part A and B below. Geographic Data You have been presented with several feature classes for this assignment (see zip file on Quercus): • Boundary Polygon for the study area (useful if you need to specify the extent in any geoprocessing tool’s environmental variables, like Euclidean Distance) • Elevation raster (30m resolution, height measured in feet!) • Land Use raster • RecSites (indicating existing recreational sites) • Roads • Schools (indicating existing schools) PART A Use weighted overlay analysis to produce a suitability model. Tools you will use in PART A include: • Raster Functions: Conversion: Unit Conversion • Export Raster • Euclidean Distance (Spatial Analyst) • Reclassify (Spatial Analyst) • Weighted Overlay (Spatial Analyst) • Con (short for conditional, Spatial Analyst) • Majority Filter (Spatial Analyst) • Raster to Polygon (Conversion Tools) • Intersect • Near • Raster Functions: Surface: Hillshade Because we will do the same general workflow 3 times (once for each of the stakeholders), make use of ModelBuilder for this assignment. To do the suitability model, you need to standardize your inputs so that everything is weighted on a scale of 1 – 10, with 10 indicating the most optimal weighting for each of your input criteria, and 1 or RESTRICTED indicating the least optimal weight. First, different types of land use are more suitable for development, while land uses such as Water or Wetlands are not suitable. When producing your suitability model, assume that the suitability for development of different land uses is as follows: Land Use Type Weight Agriculture 9 Barren Land 10 Brush/transitional 5 Built up 3 Forest 4 Water Restricted Wetlands Restricted Second, some slopes are not optimal for building. Assume that any steep slopes with a slope angle of 33° or greater are RESTRICTED for building. Third, proximity to existing infrastructure is important, for different reasons. Locating the new school close to an existing Recreational Centre is weighted positively, as it will contribute to community building. However, proximity to existing schools is weighted negatively, as the new school should serve an area that is not covered by the existing school locations. (with your inputs reclassified, you can now answer Q2 below) Now create your weighted overlay for one of the stakeholders from the reclassified inputs using the weighting criteria noted earlier. Once you have created your combined weighted suitability raster, we need perform some operations: • First, select only sites with a weighted suitability >= 9 (hint: Con tool). • Second, we need to then narrow down the number of sites to those that are large enough to host a new school. Because each cell in your raster is not that large, we need to select contiguous groupings of suitable cells. Filter out any suitable sites that do not have 8 nearest neighbours (hint: Majority Filter tool) • Third, convert your filtered suitable sites to polygons (hint: Raster to Polygon tool) (you can now answer Q3 below) Assuming you built a ModelBuilder model to accomplish all this, you can now easily repeat the same analysis for the other two stakeholders by changing the values of existing Weighted Overlay model, producing ideal site polygons for each stakeholder. (you can now answer Q4 below) Questions 1. Answer the following: a. What is the cell size of the elevation raster? (1) b. What proportion of land use is agricultural? (1) c. What are the statistics for the elevation raster (in metres)? (3) 2. Create a ModelBuilder model that covers all of the steps required to perform the Weighted Overlay suitability analysis up to the end of the workflow required to answer Question 3 (i.e. start from Reclassify and finish before the intersect). (10) 3. Produce a map layout showing the three standardized/reclassified continuous inputs into your suitability model (slope, proximity to recreational centres, and proximity to schools) using the same continuous symbology (colour scheme) to show the reclassified values (one legend for all three maps) (hint: to do this, you can add multiple map frames to a single map layout) (15) 4. Identify the most suitable compromise location(s) that make all three stakeholders happy (hint: Intersect tool on all three stakeholder polygons). a. How many remaining sites are there after the intersect? (5) b. What is the area the site(s)? (2) c. How near are the site(s) to the roads (hint: Near tool)? (2) d. Assuming the stakeholders value site size, which is the most optimal location? (1) e. Provide a map layout with your map at a scale of 1:45,000 that shows the most suitable location(s) relative to roads, symbolized existing recreation centres, and symbolized existing schools in the central part of the town. Also include a transparency of the elevation raster on top of a hillshade as the background of the map like we did in Tutorial 1 for Hong Kong. (15) PART B Oh no! The chief planner has forgotten an important consideration! Because the town lies in a mountainous region, there is risk of flooding! Tools you will use: • Fill (Spatial Analyst) • Flow Direction (Spatial Analyst) • Flow Accumulation (Spatial Analyst) o You can change the output raster’s symbology to standard deviation to better see the results • Zonal Statistics as Table (Spatial Analyst) Questions 5. Your task is to: a. According to the output from the Zonal Statistics as Table tool, what is the average accumulation for the site(s)? (5) b. Based on this, which site do you recommend as best for development, and why? (5)