Geographic Information Systems for Public Policy - Lecture X

Vector Analysis: Buffering

Sergio Rey

Last Time

  • Detecting Geometry Problems
  • Understanding Geometry Problems
  • Fixing Geometry Problems

Today

  • Deterministic Spatial Analysis
  • Buffers and Buffering
  • Select by Location
  • Health Deserts

Deterministic Spatial Analysis

Dimensions of vector geometries

  • Point: 0-dimension
  • Line: 1-dimensional (length)
  • Polygon: 2-dimensional (area)

Spatial Relations

DE-9IM

Dimensionally Extended nine-Intersection Mode (DE-9IM)

DE-9IM

[url]

Food Deserts url

fd

[url]

Today: Clinic Deserts

Buffers and Buffering

Buffers

image

Buffers (Variable)

image

Buffers (Rings)

image

Buffers (Dissolve)

image

Buffering the Clinics

  • food deserts: 0.5 mile distance
  • use for the clinic points
  • will revisit other distances later

.5 mile clinic buffer

image

buffer menu

image

buffer result

image

buffer result: geometry

image

buffer rename

image

Buffering

Using the Buffers for Locational Analysis

  • Find what tracts intersect with the buffers
  • Create layer with these tracts
  • Use: Select by Location

Processing: Select by Location

(Bring up Processing Toolbox with CTRL-ALT-T)
image

Processing: Select by Location Dialog

image

What tracts intersect with the .5 mile buffers?

image

Save Layer

image

Saved Layer

image

Health Deserts

Traditional Approach

  • Tracts not intersecting buffers
  • Add up population in desert polygons

Identifying Clinic Desert Tracts

image

Clinic Desert

image

Clinic Desert Layer

image

What is the population residing in Clinic Desert?

Desert Tracts Tables

image

Basic Statistics for Fields

image

Basic Statistics for Fields: Dialog

image

Basic Statistics for Fields: Results

image

Basic Statistics for Fields: File

image

Basic Statistics for Fields: File

image

Total County Tract Population

image

Population Estimates

  • County Total: 2,189,641
  • In Clinic Desert Tracts: 1,881,643
  • In Buffer Tracts: 307,998
  • Population with access problems: 1,881,643

Issues in measuring access to clinics

  • all individuals in the tracts intersecting the buffers are in the buffer
  • overestimation of people within .5 miles of a clinic
  • underestimation of people in the clinic desert

Why overestimation of buffer population?

image

We can do better

Buffer Tracts

image

Random Points

image

Random Points Generated

image

Random Points Layer

image

Random Points Layer Symbology

image

Random Points Layer

image

Count points in buffers: Layers

image

Count points in buffers

image

Points in polygons

image

Count Field

image

Count Summary

image

Count Table

image

Revised Estimates

  • County Total: 2,189,641
  • In Clinic Desert Tracts: 1,881,643
  • In Buffer Tracts: 307,998
    • 90,630 in buffer, in tract
    • 217,358 out of buffer, in tract
  • Population with access problems: 2,099,011
  • Population within .5 miles of clinic: 90,630

Issue

  • double counting
  • points contained in more than a single buffer are counted more than once

Overlapping Buffers

image

Count Table With Overlapping Buffers

image

Fix

  • dissolve on buffering

Buffer and Dissolve

image

Dissolved Buffer

image

New Buffer

image

New Buffer Feature Count

image

Points in Polygon Count

image

Population within .5 Miles of a Clinic

image

Revised Estimates

  • County Total: 2,189,641
  • In Clinic Desert Tracts: 1,881,643
  • In Buffer Tracts: 307,998
    • 50,190 in buffer, in tract (no double counting)
    • 257,808 out of buffer, in tract
  • Population with access problems: 2,139,451
  • Population within .5 miles of clinic: 50,190

Next Up

  • Vector Analysis: Clipping and Spatial Joins

Creative Commons License
GIS for Public Policy’20 by Sergio Rey is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.