Project 6 — Site Suitability Analysis¶

Goal¶

"Find the best 5 sites for a new community park in [your city] using multiple criteria: proximity to population, slope, land use, and floodplain."

This is the classic GIS portfolio project — multi-criteria, raster-based, decision-support analysis.

Data needed¶

Tools used¶

• Spatial Analyst: Slope, Reclassify, Euclidean Distance, Weighted Overlay
• Polygon to Raster, Extract by Mask, Raster Calculator
• Cartography: discrete and continuous symbology

Workflow¶

Step 1 — Set up¶

1. Project: site_suitability_park.
2. Project all data to State Plane (US Survey Feet works fine for this).
3. Set Environments → Cell Size = 10 m, Snap Raster = DEM.

Step 2 — Define criteria & scoring¶

Each criterion gets a 1–10 score (10 = best for park).

Step 3 — Build each criterion raster¶

Slope¶

1. Slope tool on the DEM, output in percent.
2. Reclassify: 0–5 → 10, 5–10 → 8, 10–15 → 6, 15–25 → 3, > 25 → 1.

Distance to existing parks¶

1. Euclidean Distance from parks layer.
2. Reclassify: < 0.25 mi (1320 ft) → 1, 0.25–0.5 mi → 4, 0.5–1 mi → 8, > 1 mi → 10.

Land cover¶

1. Use NLCD raster directly.
2. Reclassify by class code:

21  Developed Open Space    → 10
71  Grassland               → 9
81  Pasture/Hay             → 8
41  Deciduous Forest        → 7
22  Developed Low           → 5
23  Developed Medium        → 3
24  Developed High          → 1
11  Open Water              → 1

Distance to road¶

1. Euclidean Distance from roads.
2. Reclassify favoring close-to-road.

Floodplain¶

1. Polygon to Raster on the FEMA polygon.
2. Reclassify: in flood = 1, out = 10.

Distance to population¶

1. Convert blocks to a point feature class with POP field.
2. IDW or Kernel Density with POP weight → population density raster.
3. Reclassify favoring high density.

Step 4 — Weighted Overlay¶

Open Weighted Overlay:

Adjust weights to fit your stakeholder priorities. Document them.

Step 5 — Identify top sites¶

1. Find cells with score ≥ 8 (high suitability).
2. Raster to Polygon on those cells.
3. Filter polygons by area (e.g., > 1 acre — must be big enough for a park).
4. Pick the top 5 by area or by mean suitability score.

Step 6 — Validate against parcels (vector check)¶

For each candidate:

• Is the parcel vacant?
• Is it publicly owned (or at least available)?

Filter your top sites against parcels. Often half drop out at this step.

Step 7 — Map and report¶

A multi-frame layout:

1. Headline map: city + final 5 recommended sites.
2. Methodology insets: small thumbnails of each criterion raster.
3. Table: the 5 sites with parcel ID, area, score, parcel status.

Skills learned¶

• Raster reclassification
• Euclidean distance / IDW
• Multi-criteria weighted overlay
• Combining vector and raster analysis
• Decision-support narrative
• Stakeholder-aware weighting

Portfolio value¶

This is the gold-standard portfolio project. Demonstrates raster fluency, decision-modeling thinking, and the ability to deliver actionable recommendations.

Stretch goals¶

• Run the same model with 3 different weight sets ("equity-weighted", "cost-weighted", "ecology-weighted") and compare results.
• Build a ModelBuilder workflow so the entire analysis can re-run with one click when criteria change.
• Wrap it in a Python script tool for stakeholders to run themselves.
• Publish as a public Story Map.

→ Next: Environmental Risk Analysis.