What is spatial?

SPATIAL |adjective

of or relating to space and the relationship of objects within it

Not that type of space (yet)

Spatial data is centered around a planet (Usually earth - geospatial)

Some other planets are also mapped

• Mars
• Venus

LOCATION LOCATION LOCATION!

Spatial data reference object locations

Different types for different objects

Spatial Types

• Spatial data comes in several types

• These types are related but separate

POINTS

• Fundamental spatial type
• Represents a single location
• Has no size or dimensions

Cholera observations

Linestring

• a curve interpolated between points
• a series of connected points
• includes a direction from start to finish

Transportation Connectivity

Polygons

• a closed series of connected linestrings
• defined by an exterior boundary
• may have interior boundaries

Most Recurrent words on Wikipedia

Raster

• Represents continous data
• A plane composed of regular cells(pixels)
• Satellite images

Terrain of Contiguous US

QGIS

• Free and Open Source Geographic Information System
• Download from www.qgis.org/en/site
• Instructions at www.uagist.net/bootcamp/setup/qgis-install/

SCALE

Measures difference between reality and representation

Map Scale

• Shown as a scale bar
• Depicts map units in proportion to map's scale
• Map distance vs actual distance

Small & Large Scale

• Small scale(1:500,000) - Objects appear smaller

• Large scale(1:20,000) - Objects appear larger

Lexical scale

Map scale represented by words

• 'one centimetre to one hundred to one hundred metres'

Representative Fraction

• Represents the map scale as a fraction or mathematical ratio.
• 1/50,000 or 1:50,000

Point Scale

• Measures distortion caused by projection

Tissot indicatrix

Mercator distortion

Spatial Resolution

4 types:

• Spatial
• Spectral
• Temporal
• Radiometric

Spatial Resolution

Number of indepentdent pixels per unit length

Temporal Resolution

Number of frames or datasets per unit of time

Spectral Resolution

The range of wavelengths that an imaging system can detect

Radiometric Resolution

Refers to bit-depth or the number of levels of brightness a sensor can record

Projections

Projections translate 3-dimensional data onto a 2-d surface

Which is best?

Different projections => different strengths

Distortion

Every projection makes compromises.

• Size
• Shape
• Direction
• Distance
• Area

Cylindrical Projections

• Mercator
• Preserves angles and small shapes
• Distorts size and shape of large objects
• Distortion highest at the poles

Greenland distortion in Mercator

Conic Projections

• Lambert Equal Area Conic
• Useful for representing data in temperate zones.
• Equal Area or Conformal
• Not ideal for global data

Lambert Conformal Conic

Azimuthal Projections

• Useful for representing polar data
• Uses a single reference point
• Similar to squashing a sphere on a surface

Polar equidistant projection

Polyhedral projections

• Minimize distortion by tearing the surface
• Similar to opening a box

Waterman's Butterfly

First Law of Geography

"Everything is related to everything else, but near things are more related than distant things" -Waldo Tobler

Autocorrelation

• Measures the statistical dependence on a group of objects within a region
• Useful in measuring the distribution of phenomena

Autocorrelation

Radil, S. Matthew, Spatializing Social Networks: Making Space for Theory in Spatial Analysis, 2011

Importance

• Measures the relationship between values at different locations
• Quantifies the distribution of a phenomena

Credible Data Sources

Where do I find spatial data for my analysis?

Not all spatial data is created equal

But there's plenty of credible spatial data sources out there

Evaluating data quality

Standard format

Do similar datasets have similar formats?

Accessibility

How can you access the data?

e.g. iPlant Data Store

Accountability

Is the organization considered an authority of this type of data?

Government agencies are an excellent data source with plenty of free data.

Credible data sources

The following slides contain a few examples of credible data sources.

Global Administrative Areas (GADM)

"GADM is a spatial database of the location of the world's administrative areas (or administrative boundaries) for use in GIS and similar software."

United States Geological Survey (USGS)

Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment.

Some notable products include:

• National Elevation Dataset
• Digital Orthophotography
• National Hydrography Dataset

National Aeronautics and Space Administration (NASA)

Some useful data products:

• Modern-era Retrospective Analysis for Research and Applications (MERRA)
• Moderate Resolution Imagagin Spectroradiometer (MODIS)
• Landsat-8 (satellite)
• Global Change Master Directory

National Oceanic and Atmospheric Administration (NOAA)

• Weather and climate data
• Real-time and historical

U.S. Census Bureau

• Social demographic
• Population
• TIGER data products

Local and regional governments

• Local repositories
• Usually free
• Can be more accurate due to scale

Data collection

Exercise can be found at www.uagist.net/data-collection/credible-sources/

Consumable Services

• Variety of methods to access data
• Don't have to download everything

Web Feature Service (WFS)

Useful for modifying features in remote datasets

• Serves vector features
• Can be transactional
• Good for large datasets

Web Map Service (WMS)

• Serves data as images
• Useful for raster data
• Mostly for visual purposes
• Base maps

Web Processing Service (WPS)

• Provides computational resources to users
• Can upload data to be analyzed/modeled
• Leverages HPC resources for users

OPeNDAP

Open-source Project for a Network Data Access Protocol

• Serves NetCDF datasets
• Delivers subsets on demand

Data Wrangling

Methods to make spatial data cooperate

Relational Operations

• Allow us to test for spatial relationships between features
• I.e. do any roads cross a wildlife corridor

Spatial Predicates

• Define specific relationships between objects
• Allow us to extract intersections, and unions of datsets

Relational Operations

Examples of Spatial Predicates

• Equals
• Disjoint
• Intersects
• Touches
• Contains

Methods for Spatial Analysis

• Distance
• Buffer
• Convex Hull
• Intersection
• Union
• Symmetric Difference

Exercise

Exercise can be found at www.uagist.net/bootcamp/data-collection/data-formatting/

Applications

QuantumGIS (QGIS)