Guide: Mapping Population Density in ArcGIS Desktop

This guide shows you how to map population density across Japan using ArcGIS Desktop 10.7 with the Spatial Analyst extension. Starting with population data tied to administrative polygons, you’ll convert these to points and then explore density patterns with both simple and kernel methods.

Time Required: Around 35 minutes
Requirements:

• ArcGIS Desktop 10.7 (Basic, Standard, or Advanced)
• Spatial Analyst extension

Step 1: Acquire the Data

1. Download the dataset if it’s not already on your system.
2. Store it in a folder like C:\Training\GISData.

Step 2: Open the Map

1. Start ArcMap.
2. Navigate to C:\Training\GISData and open PopulationMap.mxd.
   • The map shows Japan’s administrative boundaries, colored by total population per unit.

Step 3: Configure Analysis Settings

1. From the Geoprocessing menu, select Environments.
2. In the settings window, set:
   • Output Coordinates:
     • Coordinate system to Same As Display.
   • Processing Extent:
     • Extent to match the AdminUnits layer.
   • Raster Analysis:
     • Cell size to Maximum Of Inputs.
     • Mask to AdminUnits.
3. Click OK to apply.

Step 4: Generate Population Points

1. Convert administrative polygons to points:
   • With an Advanced License:
     • Open the Feature To Point tool via search.
     • Set:
       • Input Features: AdminUnits
       • Output Location: C:\Training\GISData\Output\PopPoints.gdb\Centroids
       • Check Inside to place points within polygons.
     • Click OK.
   • With a Basic or Standard License:
     • Open the Catalog window.
     • Add the pre-made Centroids layer to the map.
2. Each point now represents the total population of its original administrative unit.

Insight:

• Centroids work well here because Japan’s administrative units are small and evenly spread.

Step 5: Build a Simple Density Surface (Large Radius)

1. Open the Point Density tool from the search bar.
2. Configure:
   • Input Points: Centroids
   • Population Field: Population
   • Output Raster: C:\Training\GISData\Output\PopDensity50.gdb\DensityLarge
   • Cell Size: 2500
   • Neighborhood: Circle
   • Radius: 50000 (50 km)
   • Area Units: SQUARE_KILOMETERS
3. Click OK.
4. Hide the AdminUnits and Centroids layers.
5. Adjust the DensityLarge layer:
   • Open its Properties.
   • Under Symbology, switch the color ramp to Yellow to Red.
   • Click OK.

Insight:

• A 50-km radius smooths out local details, emphasizing broad trends with a coarser look due to the larger area divisor.

Step 6: Analyze Density Patterns

1. Reopen the DensityLarge layer Properties.
2. Under Symbology, click Classify:
   • Set Method to Standard Deviation, then click OK.
   • Click OK again to apply.
   • This highlights extreme high and low density areas.
3. Try another method:
   • Set Method to Quantile, Classes to 9.
   • This evenly distributes cells across classes, masking peak densities.
4. Finally:
   • Set Method to Natural Breaks (Jenks).
   • This balances highlighting high-density zones with subtle variations.

Insight:

• Different classifications reveal unique aspects of the data—experiment to find what fits your needs.

Step 7: Build a Simple Density Surface (Small Radius)

1. Run the Point Density tool again with:
   • Same settings as Step 5, but:
     • Output Raster: C:\Training\GISData\Output\PopDensity25.gdb\DensitySmall
     • Radius: 25000 (25 km)
2. Style DensitySmall with Natural Breaks (Jenks), 9 classes, and Yellow to Red ramp.
3. Compare the surfaces:
   • Right-click DensitySmall and select Zoom To Raster Resolution.
   • Open the Effects toolbar (Customize > Toolbars > Effects).
   • Set the layer to DensitySmall.
   • Use the Swipe Layer tool to compare DensitySmall (25 km) with DensityLarge (50 km).

Insight:

• The 25-km radius reveals more localized patterns compared to the broader 50-km version.

Step 8: Create a Kernel Density Surface

1. Open the Kernel Density tool.
2. Set:
   • Input Points: Centroids
   • Population Field: Population
   • Output Raster: C:\Training\GISData\Output\PopDensity25.gdb\KernelDensity
   • Cell Size: 2500
   • Search Radius: 25000
   • Area Units: SQUARE_KILOMETERS
3. Click OK.
4. Style KernelDensity with Natural Breaks (Jenks), 9 classes, and Yellow to Red ramp.
5. Use the Swipe tool to compare KernelDensity with DensitySmall.

Insight:

• Kernel density smooths the surface more than the simple method, blending overlapping neighborhoods for a clearer, generalized pattern.

Wrap-Up

1. Save your map with a new name in C:\Training\GISData\Output.
2. Close ArcMap.

Key Takeaways

• Why convert polygons to points?
  • It simplifies density calculations for evenly distributed areas.
• Simple vs. Kernel Density:
  • Simple is direct but blocky; kernel smooths transitions for easier interpretation.
• How does radius affect results?
  • Larger radii generalize; smaller ones highlight local details.