Search
Create
Log in
Sign up
Log in
Sign up
Get ahead with a $300 test prep scholarship
| Enter to win by Tuesday 9/24
Learn more
GIS Final Exam
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (183)
GIS meaning
Geographic Information Systems
3 S's in GIS
-systems
-science
-studies
geographic information systems
application and software
geographic information science
using gis to address research
geographic information studies
societal context of gis
components of a GIS
-hardware
-software
-data
-procedures
-people
-network
spatial vs attribute data
spatial: location
attribute: characteristics
latitude
longitude
discrete features
-points, lines, areas
-individually distinguishable
continuous features
-features that exist spatially between observations
(temp/precipitation/elevation)
number of dimensions in a point
0
number of dimensions in a line
1
number of dimensions in a polygon
2
(area + perimeter)
what type of spatial model is a shapefile
vector
3 files NEEDED for shapefiles
-.shx
-.shp
-.dbf
(optionally .prj)
types of raster data
-aerial photos
-digital elevation models
-digital raster graphics (scanned topographic maps)
raster file formats (name a few!)
.img
.tif
.jpg.
.bmp
.sid
.BIL
geodesy
science of measuring and monitoring the size/shape of the earth
geoid
global mean sea level affected only by gravity (not tides and currents)
why do we use different ellipsoids in different places
there are locally best fit ellipsoids
major ellipsoid used in the US
GRS 80
major ellipsoid used globally
WGS 84
height above the geoid (to earth's surface)
orthometric height
height above ellipsoid (to earth's surface)
spheroidal height
-used in GPS
how many satellites are used in a GPS system
31
How many orbits are there
34(?)
1st Law of Geography
everything is related to everything else, but near things are more related than distant things
Spatial Interpolation
process of using points with known values to estimate values at other points
Elements of Spatial Interpolation
1. known points
2. interpolation points
what influences accuracy of spatial interpolation
number and distribution of control points
types of spatial interpolation
-global and local
-exact and inexact
-deterministic and stochastic
global interpolation
uses every known available value to estimate GENERAL trend
local interpolation
uses a sample of known points to estimate local/short-range vairation
exact interpolation
predicts a value at point location the same as the known value
-surface passes through the control point
inexact interpolation
aka approximate interpolation
predicts a value at point location that differs from known value
deterministic interpolation
provides no assessment of errors
stochastic interpolation
offers assessment of prediction errors with variance estimates
trend surface characteristics
-global
-deterministic
-inexact
regression characteristics
-global
-stochastic
-inexact
thiessen characteristics
-local
-deterministic
-exact
density estimation characteristics
-local
-deterministic
-inexact
inverse distance weighted characteristics
-local
-deterministic
-exact
splines characteristics
-local
-deterministic
-exact
kriging characteristics
-local
-stochastic
-exact
local methods
use a sample of known points
Thiessen polygon method
assumes that any point within a polygon is closer to the polygon's known point than any other points
Thiessen polygon aka
Voronoi polygons
Thiessen polygons created by
connecting lines drawn perpendicular to each side of a triangle
line density
calculates a magnitude per unit area from POLYLINE features that fall within a RADIUS around each cell
point density
calculates a magnitude per unit area from POINT features that fall within a NEIGHBORHOOD around each cell
2 methods of density estimation
simple and kernel
simple density estimation
calculates density of point features around each output raster cell
-number of points in neighborhood totaled, divided by neighborhood area
kernel density estimation
calculates density of features in a neighborhood around each output raster cell
-smooth surface fit over each point
-value highest at location of the point and decreases
-reaches 0 at radius distance/bandwidth from point
kernel density compared to simple density
produces smoother surface
IDW interpolation
inverse distance weighted
in IDW all estimated values are
between max and min values of known points
idw values
k=1, constant rate of change
k=2+, value higher near known point
spline
exact interpolation model
spline method
bends rubber to pass thru points while minimizing curvature
-gentle slopes
2 types of spline
-thin plate
-thin plate with tension
spline vs IDW
spline is not limited within range of known min/max points
Kriging
assesses quality of prediction with prediction errors
Kriging namesake
D.G. Krige
-mining engineer from south africa
3 spatial components in kriging
-spatial trend
-spatial autocorrelation
-random
spatial trend
increase/decrease in a variable that depends on direction
spatial autocorrelation
tendency for points near each other to have similar values
random
statistically defined by probability function
spatial autocorrelation
correlation of a variable with itself through space
-any systematic pattern in distribution
positive spatial autocorrelation
nearby areas are more alike
negative spatial autocorrelation
neighboring areas are unalike
-random patterns
positive autocorrelation
random autocorrelation
(random grid)
negative autocorrelation
high autocorrelation means
points near each other are alike
kriging concept of lag distance
distance between sample points
semi-variance
measure of the degree of spatial dependence between observations along a specific support
semivariogram
plots the average semi-variance against average distance
nugget
initial semivariance when autocorrelation is highest
sill
the point where variogram levels off and there is little autocorrelation
range
lag distance at which the sill is reached
semivariogram
anisotropy
the existence of directional differences in spatial dependence
two kriging methods
ordinary and universal
ordinary kriging
assumes no drift/trend
universal kriging
assumes drift/trend in addition to spatial correlation between sample points
cross-validating techniques
-remove known point
-estimate known point
-compare and calculate error
trend tool
uses global polynomial interpolation
-fits smooth surface defined by function to input sample points
flat plane with no bends
first-order polynomial (linear)
plane with one bend
second-order polynomial (quadratic)
plane with two bends
third-order polynomial (cubic)
maximum number of bends in trend tool
12 (twelfth order)
regression method/models
-include variables in model
-must use spatial complement
benefits of python
-easy to read
-large library
-supports object-oriented programming
uses of python
-analysis
-cartography
-conversion
-data management
-editing
-geocoding
python can
manipulate layers in a map
ways to use python in ArcGIS
-python window
-Python add-in
-Python script tool
-Python toolbox
-Field calculator
-Python script
arcpy
The Python add-in that allows for ArcGIS tools and settings to be used in Python scripts.
common script errors
-spelling mistakes
-incorrect case
-indentation error
-missing colons
-missing quotations
-missing double equal sign
free and open source GIS system
QGIS
measures of ocation
-mean
-median
-mode
measures of location
what is the location or center of the data
measures of variability
how do the data vary?
central feature
input lots of dots get one dot in the general middle of them
directional distribution
input lots of dots get one ellipse that encompasses most
linear directional mean
input bunch of arrows, get arrow pointing average direction
mean center
input lots of dots, get dot in average distance from all of them
standard distance
input lots of dots, get big dot over most of them
point pattern analysis
-random
-evenly dispersed
-clustered
1st law of geography (Tobler's law)
everything is related to everything else, but near things are more related than distant things
spatial autocorrelation
correlation of a variable with itself through space
positive spatial autocorrelation
nearby areas are more alike
no spatial autocorrelation
random patterns
negative spatial autocorrelation
nearby areas are unalike
null hypothesis is
complete spatial randomness
MOran's I
Compares the value of the variable at any one location with the value at all other locations
P-value
probability that the observed pattern was created at random
small p-value means
reject the null
small p value and very high/low z-score
it is unlikely that the spatial pattern is random
Moran's 1
-if E(I) is near 1 and p is significant, like values are clustered
-If E(I( is near -1 and p is significant, then high/low values are dispersed
g statistic
measures clustering of high/low values in a dataset
high g values
clustered high values
low g values
clustered low values
LISA
local indication of spatial association
-z score for every point
high z score
indicates feature is near similar values
low z score
feature is near dissimilar values
hotspot analysis
computes a z score for every feature
-cluster of high positive z scores indicates hotspot
-cluster of high negative is coolspot
Moran's I summary
measures autocorrelation
-whether like values are close to each other
-distance and value
LISA
local measure of autocorrelation
-for each point/polygon indicates whether the adjacent or nearest feature is similar or dissimilar
G-statistic summary
global statistic
-shows clustering of like high values
ord-getis summary
hotspot analysis (local)
-tells whether a high value is close to one another
(or low is next to low)
modelbuilder
series of operations linking tools
-visual
-reusable
-editable
-convenient
model elements
-inputs
-geoprocessing operations
-outputs
fist step in building a model
determining what the output should be
spatially explicit
what form should final products take? (point/line/polygon)
spatially aggregated
what attributes are needed in the table?
intermediate data
data that was used in the process that was only used to connect another process
model is clear/white
model is not ready to run
model is colored
model is ready to run
model is red
model is running
model has drop shadow
model has already run
descriptive models
describe existing condition
prescriptive models
predict the condition
deterministic models
doesn't assume randomness
stochastic model
assumes presence of some randomness
dynamic model
changes the spatial data and the interactions between the variables
static model
deals with the state of spatial data at a given time
binary model
logical expressions to select features in the raster
-1 and 0
index model
calculates index value for each unit area
index model uses
-habitat suitability analysis
-vulnerability analysis
-DRASTIC
DRASTIC model
-EPA
-Depth to water
-net Recharge
-Aquifer media
-Soil media
-Typography
-Impact of the zone
-hydraulic Conductivity
four ways to run a tool
-toolbox
-modelbuilder
-python script
-python script standalone
network
system of interconnected linear features
lines...
-meet at intersections
-can't have gaps
-have directions
pure network
only typology and connectivity are considered
flow network
topology and flow characteristics
direct path
most direct path
optimum routing
numerous places at one time
-efficient
-aka pizza route
closest facility
finding nearest location
measure link impedence
distance+speed limit
junction
street intersection
turn
transition from one stree segment to another
turn impedance
time taken to complete a turn
-directional
negative turn impedance
prohibited turn
over/underpass methods
-nonplanar features
(continuout lines no nodes)
or
-2 arcs meet at a node
neagtive minutes in turn impedance
no turn
multimodal network
-modeling a more than one mode of transport
3d network
interior of buildings/caves
True/False: In an IDW interpolation technique, estimated values are always between max and min values of known points
true
Process of using points with known values to estimate values at other points
spatial interpolation
What kinds of technique is IDW
-local
-deterministic
-exact
Which spatial interpolation method assumes that any point within a Voronoi polygon is closer to the polygon's known point than any other known points?
Theissen Polygon
First law of geography
everything is related to everything else, but near things are more related than distant things
When neighboring things are alike, this is called
positive spatial autocorrelation
In which estimation is the surface value highest at location of the point an diminishing with increasing distance from the point
Kernel density
In which interpolation technique are all points in the data used to estimate the values of locations for which the data is unavailable
global
Moran's I close to 1 with a low p value indicates
spatial autocorrelation
simlified representation of phenomenon in the system
model
In a model where logical expressions like & and "or" are used, will the output values have index values?
NO
What would a model that preditcts land use change be (without giving error estimation)
prescriptive and deterministic
Predicting hurricane path and giving error estimation
dynamic and stochastic
A descriptive model predicts the current condition?
true
is a model predicting hurricanes dynamic?
yes!
;