Study Guide: Geog 621, Exam 1
Intro: Environmental Analysis & GIS
- nature of environmental data, sampling of continuous surfaces, raster GIS
Understanding Raster GIS
- discrete entities : features, named, attributes, many human landscapes,
some physical
- continuous fields : focus on single attribute, measurement, interpretation,
common in nature
- How to deal with continuous fields?
- trend surface : polynomials, advantages, disadvantages. Know at least the first-order formula Z = a + b1X + b2Y
- discretizing : isolines, classifying, tesselation (irregular & regular)
- irregular tesselation : Thiessen/Voronoi polygons and their dual, TIN (Delaunay triangulation)
- regular tesselation : e.g. raster
- examples of continuous & discrete variables
Raster & Surface Display & Analysis
- Purpose: analysis of patterns, visualization of data, communication
- Modes: planimetric (contours, hypsometric tint, shaded relief), projected
perspective surfaces, sections/profiles, longitudinal profiles
- Elevation as a surface; the nature of surface data (2.5D); DEM
- True 3D : XYZ at each point (point feature dataset) or vertex (polyline
or polygon feature dataset) -- allows 3D figures with tops, bottoms, sides.
Useful for underground features like caves, ore bodies, mines, or temperature
plumes in lakes, etc. 3D polygons, points, lines.
- TINs: derived from corner nodes, each triangle has slope, aspect.
Elevation interpolated from nodes.
- Continuous map display schemes: continuous (image-like) vs. classified
(isarithmic)
- classification methods: equal interval, quantile, natural breaks,
standard deviation
- Categorical display of integer grids, using unique values.
- Layers in Arcmap : stores symbology, etc.
- using remotely sensed data in GIS: difficulties, advantages
- Hillshades:
- purpose: visualization, radiation modeling
- illumination variables: azimuth & altitude -- influence of
each on effect
Grid data: structure
- grid structure: rows & columns superimposed on cartesian
- cell size & resolution
- registration & projection (avoid). How to reproject a categorical grid?
- integer vs. floating-point grids. VAT
- Grid storage (folders) : Importance of ArcCatalog to manage grids and other
GIS datasets
- Arcmap & Raster Calculator: temporary vs. permanent grids
- overlaying contrasting grids
Raster Analysis Methods & Conversions
- Geoprocessing tools and environments for raster analysis: toolboxes, ModelBuilder, Map Algebra in Python
- Map Algebra syntax: expressions, functions, etc.
- meaning of expression & assignment statement
- functions & tools
- return integer vs. floating-point grids,…. type conversion
- Analysis Environment: settings for window (extent), cell size, mask.
- analysis extent: union vs. intersection
- vector to raster. cell value assignment – predominant, weighted features
- raster to vector. gridpoly – weed tolerance. gridline – thinning.
gridpoint
- Generalizing boundaries (Douglas-Poiker)
- Reading and mosaicking DEMs
Raster Operations
- Python geoprocessing basics: Python Window, PythonWin, import arcpy, indentation, script tools
- map algebra processing: how is each cell assigned a value. output-oriented
- what happens with null (nodata)
- mask & nodata : using a mask to clip a grid. What is a mask raster and when would you use it? Understand both what it is -- a raster that has data and nodata -- and what it is used for -- e.g. a study area that isn't rectangular.
- map algebra operators: arithmetic (+, -, ...), relational (>,
<, etc.), boolean (and, or, etc.), logical (diff, over, in)
- Understand logical (boolean) expressions!!! Including how to write them in map algebra, using relational and boolean operators, including expressions that combine expressions.
- Extraction tools & nodata values vs. Logical tools and Boolean values: Know the Difference.
- Conditionals: con & setnull
- IsNull: what does it do?
- Know how to write the following in map algebra:
- Calculate the result of an expression with arithmetic and other operators
- Create Boolean rasters using relational operators
- Change nodata to zero, while keeping the other values the same
- Change zero to nodata, while keeping the other values the same
- In general, you should be able to use con, setnull, and isnull for any situation
- Reclassify & Slice, Over and Diff
- resampling: nearest/search/bilinear/cubic
- Combinatorial operations
Statistical Summary Methods and other tools
Local, Focal, Zonal, and Sample Statistics
- Know the difference in the way these work -- consider the nature of data
sources.
- CellStatistics (max, min, mean, med, std, majority, etc.) -- uses multiple
rasters, often temporal, uses python list: e.g.: CellStatistics([ras1, ras2, ras3]), "mean")
- focal stats (focalmax, focalmin, etc.)
- zonal stats (zonalmax, zonalmin, zonalmean, etc.)
- contrast local, focal, zonal, global function classes
- basic syntax for using functions: Spatial Analyst & Raster Calculator
- Local functions. purpose for: reclass, math functions
- Focal functions: meaning of & types of neighborhoods, purpose for various
fucntions, like focalvariety
- Block functions : purpose for, contrast and similarity with focal
- Aggregate
- Zonal functions: zonalstats of various types (zone grid & value grid);
geometric zonal functions
- Sample Statistics: SSUS and proportional sampling
- Generalization tools:
- Resample
Distance & Density Analysis
- Euclidean distance functions: eucdistance, eucdirection, eucallocation. How do they differ from a buffer? Why would we use them instead of a buffer?
- Weighted distance functions: costdistance, costdirection, costallocation,
pathdistance. What is a cost surface and how would you create it (troll analogy)? What really does costdistance create -- what do the values represent?
- costpath, corridor. What is a cost path, and how does it relate to the cost distance result above?
- Density functions. Works with events -- how do these differ from samples? Thiessen polygon method: 1/area