Introduction to spatial analysis (part 1)

Rachel Franklin

Newcastle University

Goals for this session

• Spatial analysis overview
• Common tools/methods in a GIS environment

What is “spatial analysis”?

Spatial analysis for urban and regional analysis

Often we use spatial analysis tools to extract information

• Length of paved road in a region
• Number of universities in a place
• Number of inhabitants within a distance of a service
• Area within a distance of a road or feature (like a bus stop or metro station)

A variety of types of spatial analysis

According to Longley

• Queries and reasoning – no changes are made to the database and no new information is produced
  • e.g., how many cities within 300km of L’Aquila?
• Measurements – Describing aspects of geographic data, like length, area, or shape
  • For example, calculating the size (or area) of a parcel
• Transformations – Changing or combining data to create new data, using logical, mathematical, or geometrical rules
• Descriptive summaries – summary statistics for spatial data
• Optimization – Finding the best locations for a set of objects
  • For example, best locations for public services
• Hypothesis testing – Making generalizations about a population from a sample
  • Could this spatial pattern have occurred by chance?

Types of spatial analysis

continued

• We’ve covered queries and reasoning
• Measurement is either embedded in feature class information (e.g., length of a line segment) or is straightforwardly calculated in a GIS—this is GIS bread and butter functionality!
• This lecture focuses on transformations

Transformations

• Buffers – Create an area of a specific and constant width around a point, line, or polygon This can be used to identify all objects falling within a certain distance of the original feature
• Point in polygon – Associates points with polygons
  • Counts the number of points within a polygon or…Attaches polygon characteristics to points
  • This is usually what people mean when they refer to a spatial join
• Polygon overlay – Determining whether two polygons overlap, the extent of their overlap, and what new polygons are created by the overlap
  • Can be thought of as an intersection or a union of units and information
• Spatial interpolation – Estimating the value of a variable for locations where no measurement has occurred. For example, rainfall, temperature, or elevation
  • e.g., Inverse distance weighting or Kriging
• Density estimation and potential – generates a surface from a set of discrete points

Where to find QGIS spatial analysis tools

QGIS pulldown menus

Check out all the options!

QGIS toolbox

• Located on the main QGIS toolbar
• Can be overwhelming—take the time to poke around
• Most tools wook with a common interface/intuition
  • Practice makes…comfortable (if not perfect)
• Search is your friend
  • The key is to know what tools you are searching for!

The most common tools

Buffers

Creates a new layer of polygons around a set of input features

Appealing way to visualise and calculate areas surrounding features

Appealing way to visualise and calculate areas surrounding features

Why would we want buffers? What purpose do they serve?

Calculating how many features fall into an area

Point in polygon (aka spatial join)

Calculating how many features fall into an area

Point in polygon (aka spatial join)

What if we want to know areas where two (or more) layers overlap?

Intersect

What if we want to know areas where two (or more) layers overlap?

Intersect

What if we want to know how many people fall into an area?

Buffers + Intersect (with census area data) for an approximation

How is this useful?

What if you only want to rent an AirBnb that’s close to water and cultural locations?

How is this useful?

What if you only want to rent an AirBnb that’s close to water and cultural locations?

Grouping features together, based on a common attribute

Dissolve

Combining polygons

Dissolve example

Combining polygons

Dissolve example 2

Summary

• So many more tools—this is just scratching the surface
• Most of these tools generate a new layer
  • Pay attention to naming AND to where you’re saving!
• If you mess up, you can simply run the tool again
• Using Help and the internet is key
• It’s common to need to string together tools to get to a final answer
• Layer names can matter:
  • No spaces or strange symbols.
  • Use _
  • No starting a layer name with a number (e.g., Pop_2011, not 2011_pop)

Next up: Tutorial 3!