Lecture Notes - GIS for the Natural Sciences

Pedagogical Structure, etc

  • The difference between learning GIS and learning GIS software

  • Learning by doing

  • Teaching vs Learning

  • My background

  • History of this class

  • Grading the term

Basic GIS

  • Representing things (example)

    • Points, Lines, Polygons, Raster, Tin, Attributes (like spreadsheets!)

  • Measuring things

    • Categorical -discrete

    • Ordinal - discrete with order

    • Continuous - quantitative

  • GIS Files

  • 2 problems: file formats and coordinate systems

  • Projections and coordinate reference systems (reference ellipsoid) (Dana's overview)

    WGS84 vs NAD83

    • Lat/Long, UTM zones, State Plane (overlay example)

    • Geoid, Relationship between Geoid, Spheroid, Datum

    • NAD 27/NAD83 (North American Datum)

    • Why different coordinate systems?

    • Minimize distortion in area and/or angle

    • Parameters used in defining a projection or coordinate system

    • Converting DMS to DD

    • Reprojecting data: within the same Datum vs different Datum!

  • "On-the-Fly" vs "Physical" reprojection, and the Data Frame

  • Data Frame [CRS/SRS/Projection] defined by first added Object (as long as Object has SRS defined!)

  • Defining the Projection

  • Storing Data

  • Converting Data

  • Resampling Data (Continuous vs Categorical) some details

  • Selecting Data (SQL Select)

  • Symbolizing or classifying data (covered in tutorial, right?)

  • Scale: "large" vs "small", remember that ratio is like dividing

    • 1523 vs 8341 glaciers from 1:100K vs 1:24K

  • Maps as Models vs modeling physical systems

  • Ecological Fallacy


Types of Purposes (coastal erosion example)

  • Organization

    • basic database construction, warehousing large amounts of data, ongoing updates

  • Visualization

    • visual display of data shows patterns not evident in tables of data (duh!), spatial display shows patterns not evident in graphs!

  • Spatial Query

    • show me all places where "X > Y". identify objects, ie "what is this thing?" (Arcview examples)

  • Combination

    • also overlay analysis (often vector)
    • map modelling, eg, intersect (impose in Bonham-Carter), example: soils and geology
    • combine values from different layers with algebraic statements (often raster)
    • subset of geoprocessing, ie union, intersect

  • Analysis

    • wide open, tons of possibilities
    • spatial analysis takes into account the spatial relationships, standard statistical analysis typically does not
    • single map vs multiple map
    • modeling

  • Prediction

    • perhaps the ultimate goal in science!
    • detailed example from Bonham-Carter

Some map types for area data or thematic mapping

Metadata (example)

other tools
  • Lineage
  • Positional Accuracy
  • Attribute Accuracy
  • Logical Consistency
  • Completeness

Anatomy of a GIS Dataset

  • Shapefile
  • Coverage
  • Geodatabase
  • Raster Imagery

Databases

  • Terminology: tables, fields, records, indexes, joins

  • Relational vs Hierarchical vs OO (object oriented)

  • SQL (structured query language)

  • Foreign key, or how do I relate multiple tables

  • levels of "Normality", or "Do I sacrifice usefulness for Elegance?"

  • bad example, good example

 

Useful stuff for scientists

  • Downloading and unzipping files (illustration)

  • Creating basemaps

  • Creating basemaps to put your own data on

  • Finding basemap layers

  • Creating your own data: Digitizing, GPS, etc

  • Scale issues, microscopic to intergalactic

Analysis, the payoff!

  • Too tired!

  • Independence and Randomness

  • Analysis of point, line and polygon patterns

  • Geostatistics

  • Geospatial Analysis 

  • Raster analysis, con() function, other algebraic functions

  • Directional or Orientation analysis

  • What can be done, vs what is useful, ie, computing power makes algorithms do-able, not always meaningful

From de Smith book:

  • Core components of geospatial analysis, including distance and directional analysis, geometrical processing, map algebra, and grid models
  • Exploratory Spatial and Spatio-temporal Data Analysis (ESDA, ESTDA) and spatial statistics, including spatial autocorrelation and spatial regression
  • Surface analysis, including surface form and flow analysis, gridding and interpolation methods, and visibility analysis


Extending the GIS

  • Limitations of commercial GIS

  • GIS as a platform for modelling and analysis (vs programming from scratch...)

  • Where to look for extensions

  • Simple programming structures to "roll your own" algorithm


Sharing using standards

  • OGC standards body

  • Open Standards and Open Source

  • WMS vs WFS, the raster/vector paradox on the web!

  • What about KML?

  • Mashups! Fun for all!


Image data

Environmental Modeling

Error!

Bandon Coastal Images

Krigging















end