• Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off
Reading...
Front

Card Range To Study

through

image

Play button

image

Play button

image

Progress

1/162

Click to flip

Use LEFT and RIGHT arrow keys to navigate between flashcards;

Use UP and DOWN arrow keys to flip the card;

H to show hint;

A reads text to speech;

162 Cards in this Set

  • Front
  • Back
GIS
Geographic: spatial component
Information: tabular data or relational database
System: integration of computer power/ dynamic hosting for information
OR
Science: methodology for answering a spatial query

GIST: Geographic Information Science and Technology
One Definition of GIS
a collection of practices, software and hardware with the ability to collect, store, display and analyze spatial/geographical content
Views of GIS
Views of GIS
Views of GIS contd
Views of GIS contd
What "makes GIS so special" ?
analytical power in layering information; insight in spatial display of quantitative information
Goals of GIS
to utilize spatial component of data to maximum extent/maximum insight; relatable/interchangeable/combinable data
Analytical GIS
used for robust spatial analysis, cartography, spatial modeling, spatial statistics
Simple GIS applications
demographic change, policy analysis, regression maps, distance/proximity analysis, exploring accessibility (public transit setups), interpolation (where to buy cheap gas?), density/heat maps, raster analysis (combining population data with risk; questions like "how many people would die in a tsunami .. here?", temporal analysis w/ animation
Raster and 3D analysis
Technical GIS
software packages, scripting, data management and compatibility,
Levels of Representation in GIS
1. reality, 2. conceptual model, 3. logical model, 4. physical model

traditional maps are mute and still! "can't be queried"
GIS' obvious 'problem'
some abstraction still required b/c of practical hardware and data limitations (how fine to render the shoreline?)
Two discrete Conceptual models to pursue in GIS rendering and data treatment
discrete objects (only by data provided) (VECTOR) vs continuous fields (must account for areas where data doesn't exist) (RASTER)
Geographic data components
space, time, attributes (categorical/qualitative?)
Data attributes in GIS
identification, associated data, spatial relationships; classified by scale of measurement (nominal, ordinal, ratio, interval, cyclic)
Some GIS history
hearkened by computer growth, cold war, but John Snow's 1854 cholera map is classic 'GIS concept' work i.e. 'statistical map'

others: Charles Booth poverty map, DuBois' 'Philadelpha Negro', Hull-house maps

Harvard's ODYSSEY (1970) is computational turning point with topological 'arc-node' (vector) data structure
cartography
science of construction, use and principles of maps

important because data needs to be on the same projection!
Definition of 'map'
representation of geographic phenomenon as a set of symbols "and at a scale whose representative fraction is less than one to one"
Coordinate system
location reference system for spatial features on Earth's surface

also known as geocode, needs to be unique assignment

X-direction = ' EASTING'
Y-direction = 'NORTHING'
Earth models/ 'datums'
models to capture Earth's dimensions

Datum: adjusted ellipsoid; can be locally adjusted by changing the location of the spheroid to the geoid to preserve accurate measurements of a particular area
Projections
Mathematically represent the surface of the earth for different purposes in different mediums
Three models of earth projection
sphere, ellipsoid, geoid (geoid most accurate because lumpy)

The earth's ellipsoid is only 1/297 off from a sphere!
Commonly used ellipsoids
Clarke 1866, GRS80, WGS84
To compare or edge-match maps in a GIS...
both maps must be in the same coordinate system
graticule
a grid of parallels and meridians shown as lines on a map
Commonly used coordinate systems in the U.S.
Geographic coordinate system (lat-long)

OR grid-based coordinate systems:

UTM (universal transverse mercator, is projection system AND coordinate system)
State Plane coordinate system
PLSS (public land survey system)
Universal Transverse Mercator
UTM Zones
State plane coordinate system
plane rectangular coordinates for each of 50 states; scale accuracy up to 4x of UTM, in feet, each state has false origin for various zones
public land survey system
not used for eastern original colonies and Texas, six-mile squares (farm flyovers); all land titles W of ohio river use PLSS for parcel descriptions
Measuring with coordinates
Measuring with coordinates: Geographic vs projected
Map projections
secant projection
Map projections contd
Map projections contd
Cylindrical projections
conformal
shape-preserving; no flat map can be both equivalent and conformal
equal area
preserves area
equidistant
preserves distance ;)
Plate carree
default ARCGIS 10.1, evenly placed graticules
Mercator
used for navigation because straight line on map represents constant bearing and direction

lines called 'rhumb line' or 'loxodrome'

lines are NOT the shortest distance to travel
lambert conformal conic
preserves shape; often used for regional maps; used for most state-plane coordinate zones which spread east-west
Standard parallels
Albers equal area conic
relative area- preserving conic; like lambert conformal, not suitable for global projections
Azimuthal equal area
preserves area BUT center of the map is only point with no distortion
Compromise projections
Review of projection classes and associated distortions
NO FLAT MAP...
can be equivalent AND conformal
Units on GIS?
None! Scaleless
types of map scale
two main GIS data models
raster and vector ; need to make choice because analysis, representation and spatial data relationships can be limited
Raster data
grid of cells; cell contains representative values

cell has resolution (in ground units), stores info about location and characteristics; every cell has a value even if 'missing'

points and lines have to move to a cell center, each cell can be owned by only one feature, all cells have to be able to hold the max value

WELL SUITED FOR values CONTINUOUSLY CHANGING over AREAS (elevation, temperature, precipitation, population density)

can also represent discrete values (landcover, vegetation type)

non-displayed (not graphically displayed) information about each raster cell in VAT (value attribution table)

data not necessarily georeferenced (most software accepts straight .gif etc images)
vector data
vector data represents features using three types of geometry: points, lines, polygons (areas)

points are given in order, lines are ordered; i.e. vectors can store relationship information (topology)
DEM
Digital Elevation Model aka Digital Terrain Model
Raster Advantages
simple data structure, easy to edit, easy to render, good at representing surfaces / continuous data, good for satellite and aerial photography, easy to incorporate math because all spatial entities have a relatively simple shape
mixed pixel problem
Vector polygons...
must have the same starting and ending vertex, can have areas of exclusion (hollow centers), may include multiple parts (Hawai'i)
What is topology?
a mathematical procedure for explicitly defining spatial relationships; enables GIS software 'adjacency, connectivity, proximity, coincidence'
Major topological concepts
connectivity: arcs connect to each other at nodes
defn area: arcs that inscribe an area define a polygon
contiguity/adjacency: arcs have direction and Left/Right sides
In topology, position vs 'spatial relationship'
Advantages of topology in GIS
Complete topology makes map overlay feasible! Allows many GIS operations to be done w/o accessing point files.
Arc/node map data structure w/files
Bounding rectangle
Slivers
Unsnapped node
Vector advantages
v. accurate point, line and area features, compact data structure (way smaller files than raster), works well w/ manual data creation methods, topology can be represented (--> network analysis)
Vector DISadvantages
more complex data structure than raster so BECOMING SIMPLIFIED --> object-oriented model

more difficult to render than raster (HQ and complicated drawing)
Supported Raster Formats
Supported Vector Formats
Data conversion
vector --> raster is easy,
raster (back) to vector is hard
Discernable difference in representation between raster and vector data
Graphical example of vector/raster conversion error
3D/Voluminous data structures
can be in either raster or vector format

DEM (digital elevation model) is by far the most used raster structure for volume, and TIN (triangulated irregular network) most popular for vectors
TIN
triangulated irregular network, way to handle 3D vector field data; more efficient than a grid, uses an optimal triangulation of a set of irregularly distributed points

no attribute table but can calculate elevation, slope and aspect on the fly

good for hydrological analysis
Which data structure to choose?
Behind the Map ~~
attribute data files or relational database management systems (RDMS) are back-end, simple to complex depending on data

tables can be linked through common 'keys' (common data fields)
Database Managment System
DBMS
DBMS contains/provides...
Normalization
Object- and Object-relational DBMS
Object-oriented data model
Example: Normalized database topology model
Geodatabase Types
Simple geodatabase example
Creating a geodatabase
Three steps:

create a new geodatabase

create a new Feature Dataset (assign coordinate information here which all data within has to comply to)

add a new Feature Class (import shapefiles, raster, images etc)
Geodatabase data management
Feature class
Feature datasets
Mosaic dataset
type of raster organization (w/ raster catalog)
Raster catalog
type of raster organization (w/ mosaic dataset)
Geodatabase topologies
ESRI formats: what to choose?
ESRI formats cont'd
Why geodatabase is better
Why is data quality important?
A ton of people are using the internet now! That's a secondary source/ 'low quality' source of data
Who's responsible for data standards?
US govt standards have been regulated for a long time, but market standards are up to the user
Data quality components
Sources of error
your only hope...
Metadata! Documentation of data, 'data about data'

Federal geographic data committee

Readme.txt one example but ARCGIS uses .xml
GIS Data principles
geospatial data is EXPENSIVE (US federal govt data only source to not charge; check before using!)
Distributed GIS
desktop-free GIS; server, handheld, web-enabled
Advantages of distributed GIS
platform free often (web browsers make the interface)
Other Country's GPS systems
GLONASS - russia, 21 satellites
Galileo - EU, 30 satellites, 2013
CHina

on the ground versions: OnStar, Coast Guard, FAA, public works depts
Handheld GPS capable devices
magellan, garmin, sprint, verizon, att&t
GPS error management
selective degradation -- govt reduces accuracy of signals to deny full access to unauthorized users - was supposed to be turned off May 2000

visible satellites -- ground-level obstructions can block signals / cause position errors

signal multipath -- signals reflect off of objects, increasing travel time of signal and causing error

receiver clock error -- built in clock not as accurate as on-board satellite atomic clocks

orbital error -- satellite misreports its location

atmospheric delays -- GPS receiver corrects for using algorithm w avg values

satellite geometry -- relative position of satellites; best when satellites are at wide angles relative to one another

can find computers (findmymac) or cell phones (the police, opencellID, navizon) with ground systems (wifi, cell phone towers)
Important GPS contact
US coast guard navigation center, natl geodetic survey, FAA, GPS.gov
Civil engineering survey
Uses a theolodite or survey transit; records distance / elevation from known location with a stadia rod; trigonometry/ geometry determines size and shape of landscape

accuracy varies from manual equipment to laser guided automation
Remote sensing
'the science of obtaining information through analysis of data acquired at a distance'

most common is aerial sensing or satellite data

sensors collect portion of electromagnetic spectrum at certain pixel size (resolution)

EM waves out of visible used, e.g. vegetation in IR is red but is green in 'visible'
Remote sensing applications
land use mapping / land use change detection, natl disaster and risk mapping, global warming info, biodiversity info, landscape prediction/measurement (rainfall, temp, humidity), environmental/health problems (water salinity, pollution, oil spill)
Aerial photography
low fliers can produce sub-centimeter resolution; interpretation is 'art' but useful qualities are tone, size (pond v lake michigan), shape (geometry - > humanity usually), texture (cement v. forest), pattern/spatial arrangement (forest v. orchard), shadow (can help determine height and can also obscure things), site (do we expect sinkholes in central Canada?, association (do we usually see nuclear power plants in residential zones?)
Stereo photography
30 to 50 percent overlap, creates depth judgement by parallax in stereoscope
orthophotography
normal photo can't be used for mapping because of camera tilt and topography distortion; photo must be RECTIFIED into an ORTHOPHOTO (a uniform scale img), 

transformed using math algorithm, ground control points, overlapping photos, and precise photogr
normal photo can't be used for mapping because of camera tilt and topography distortion; photo must be RECTIFIED into an ORTHOPHOTO (a uniform scale img),

transformed using math algorithm, ground control points, overlapping photos, and precise photography settings
Satellite imagery
first used by govts for espionage;

1960 MetSats launched
CORONA program 1960 - 1972, film had to be retrieved by planes, resolution 1 to 10 meters

Gov't satellites now achieve 2-4 inch resolution

different satellites capture different EM waves but most common are microwave, visible, UV, IR

ADVANTAGES: efficient (dont have to send a team to remote areas), versatile (can be used by many disciplines)

disadvantages: maybe expensive (if you need high-res img), learning curve -- need good software and good understanding of image analysis
Landsat 7
can collect seven images at once, each image shows specific band of EM spectrum

in order: blue-green, green, red, near IR, mid-IR, thermal IR, High IR
can collect seven images at once, each image shows specific band of EM spectrum

in order: blue-green, green, red, near IR, mid-IR, thermal IR, High IR
Basic satellite image processing
original images are in black and white; images typically 'composed' (stacked) and shown using primary colors 

'False Color Img' combinations below

software can derive data from these combos, and will assign classes to data based on combo labels
original images are in black and white; images typically 'composed' (stacked) and shown using primary colors

'False Color Img' combinations below

software can derive data from these combos, and will assign classes to data based on combo labels
Supervised vs unsupervised classification (in satellite imagery)
Supervised: create training file (identify sample groups/classes), create signature file using training file and image bands, evaluate signature file, classify (apply signature rules to unknown areas), interpret

Supervised classification steps below
u
Supervised: create training file (identify sample groups/classes), create signature file using training file and image bands, evaluate signature file, classify (apply signature rules to unknown areas), interpret

Supervised classification steps below
unsupervised: create signature file by identifying clusters using statistics, evaluate signature file, classify by applying the signature/rules to the whole area, interpret
RADAR
is old, uses radio waves; can detect, range, and map objects like aircraft, rain, ships; adapted for mapping topography of land masses and bathymetry
LIDAR
LIght Detection And Ranging -- uses pulsed light to determine ground surface elevation with high vertical accuracy; comes back as 2D image
Street Address Geocoding
formats different for different countries... GIS doesn't interpet address directly and needs a database "lookup table" to match address against
Geocoding and Address Matching
Geocoding is process of assigning a pair of approximate geographic coordinates to each case (e.g. address, zip code, place name, etc.)

Address matching is most common geocoding method, matches street addresses with reference street data; matching metho
Geocoding is process of assigning a pair of approximate geographic coordinates to each case (e.g. address, zip code, place name, etc.)

Address matching is most common geocoding method, matches street addresses with reference street data; matching methods vary by country

GEOCODING problems: new / irregular street addresses, expensive street reference data (international streets expensive)

Solutions: instead of street, match by zipcode or city centroid; parcel matching, geocode with GPS
Geocoding process in ARCGIS
Build a geocoding service (geocoder, locator) -- define method, reference data & fields, thresholds/criteria, and output data; run service against your table data; manual intervention / correction


lower quality geocoding service => higher geocoding rate (OPTIONAL / use with caution; geocode with second best spatial information, e.g. zipcode instead of street addresses)
Reverse geocoding
1. " I can't geocode a location because there aren't any nearby streets, but I know exactly where it is"

we can assign geographic coordinates by clicking on a map

2. " I know geographic coordinates but I want to know a street address for this location"

By submitting geographic coordinates, some systems can assign closest address
Reference (look-up table) sources
Traditional geocoding services require reference (geocodable streets) data

some options: census TIGER/ lines (Free but data must be processed), TeleAtlas (used by Google maps), NavTec (TA's rival, used by Yahoo Maps)
Zipcode / City matching
Geocoding web services
can use ARCGIS pre-defined online geocoding;

advantages: no reference table necessary;

disadvantages: limited geocodable cases in general (ARCGIS online publicly only 1000 cases at a time), not much control so need to examine results carefully
What is a map?
A graphic depiction of all or part of a geographic realm in which real-world features are replaced with symbols at correct spatial location at reduced scale
Map function in GIS
storage, temporary communication, intermediate data check, final report
Map design
primary goals are to convey information, highlight patterns & processes, and to illustrate results

secondary goal is to create a pleasing and interesting picture

A good map: looks good, is elegant and simple, leads to a map that is fit for the purpose

Design influencers: purpose, reality, available data, map scale, audience, conditions of use, technical limits

different media have different cartographic strategies (paper, film, mylar, monitor, digital projection, TV/movie)

its ok to have gaps, but keep them on the upper area NOT the bottom; KISS, balance

cartographers have designed hundreds of map types and choice depends heavily on data to be shown in the map figure; choosing the wrong map type can lead to misinformation and reduces communication effectiveness although symbolization can still be 'perfect'
Cartographic elements
map body, inset/overview map, title, legend, scale, direction/orientation indicator, map metadata, background, border, neatline, inset, page coordinates, ground elements, graticule
map body, inset/overview map, title, legend, scale, direction/orientation indicator, map metadata, background, border, neatline, inset, page coordinates, ground elements, graticule
Map Types: Point Data
Graduated symbols
proportional symbols are similar to graduated symbols but represent quantity more accurately
proportional symbols are similar to graduated symbols but represent quantity more accurately
Dot density maps
Charts (pie, bar, stacked)
Map types: line data
Famous flow map
Map types: Area Data
Map types: volume data
map types: time
Choosing map types
Data scaling
Example: cloropleth mapping
classification questions
Graduated (saturated) colors
Classifications
natural breaks (jenks) shown by jump in values on data scatter
Good for not-evenly-distributed values, but computing good break-points doesn't always work well

quantiles: each class contains an equal number of features; good for evenly distributed values, also good for comparing area roughly the same size; not appropriate for different value sets as greatly different values can end up in the same class

equal interval: each class has an equal range of values
defined interval: each class has a specified interval value

it's easy to understand but isnt appropriate for clustering values

standard deviation: each class defined by distance from mean of values; it's good to see which features are above/ below mean but it's hard to understand w/o knowledge of stats
Map design
visual balance affected by: symbol 'weight', visual hierarchy of symbols and elements, relative location of objects on map

cartographic color conventions: land cover is brown, yellow or green; water is shades of blue; roads are red, white, gray, black,
visual balance affected by: symbol 'weight', visual hierarchy of symbols and elements, relative location of objects on map

cartographic color conventions: land cover is brown, yellow or green; water is shades of blue; roads are red, white, gray, black, or green

in arcgis: RGB (additive primaries), CMYK (subtractive primaries), HIS (hue intensity saturation values) -- saturation/intensity (sometimes 'lightness') maps better onto values than hue; usually for classification

hue, saturation, intensity , simultaneous contrast (same color different background)

human eye limited to about 12 various colors and 7 - 8 gradient colors

color choice based on connotations, conventions, preferences, media
Color impaired- friendly map
Cartographic conventions for scholarly publications
symbol 'weight'
Cartographic conventions for scholarly pubs - text
Cartographic conventions for text placement
above-right is preferred location for city label, path right -> path down | V for words
above-right is preferred location for city label, path right -> path down | V for words
scales and generalization
Map design and GIS
Geovisualization examples
Cartogram example
Dasymmetric map
Graphic editor software
Spatial analysis
examine relationships between geographic features to describe and understand real world being represented

compare and contrast maps

investigate variation over space

predict future or unknown areas / features through modeling

data modification and statistical measures are sometimes required

spatial statistics has its own set of requirements and methods

Tools for modifying data, identifying spatial relationships, and modeling integrated into GIS software
Analytical approaches
testing for the existence of spatial structure, estimating model parameters, prediction of the 'unobserved'
Exploratory spatial data analysis
methods: visualization, numerical summaries, identification of spatial structure, investigation of spatial dependence

in the past 2 to 3 years, GIS analytic toolset has grown statistical modeling tools, spatial analytical tools

real geographical phenomena are dynamic but GIS is static... time-slice and animation helps in this
the THIRD airport!