GIS: Geographical Informa3on Systems l Each point has a X, Y loca:on à X/Y loca:on triangle, hexagonal shapes à most efficient Vector Raster
Pros 1. compact data structure 1. good for complex analysis
Computer-based tool that collects, analyze, store, is on a sphere but pretend the Earth is flat 2. efficient for encoding topology efficient for overlays
manipulate and visualize geographical l Feature class: collec:on of many features Data structure: 3. true representation of shape 2. efficient for overlays
informa:on on a map à problem-solving l Layer: contains feature of same type of 1. Grid (assuming square grids) à Cell = pixel 3. data structure common for
imagery
technology geometry à gets a separate entry in the - Spa:al resolu:on depend on cell size Cons 1. complex structure 1. large datasets
- Components: computer hardware pla=orm, Contents Panel - Features are generalized to a constant value 2. overlay operations difficult 2. topology hard to represent
GIS soAware, data storage, data input l Aaribute table: rows VS column à associate - Lines & polygons appear jaggy 3. might imply false sense of 3. maps less realistic
accuracy
hardware, informa:on output hardware, GIS with the vector data - Direct computa:on possible (for lengths,
data and GIS personal Row feature/ record perimeters and areas)
Column attribute/ field
- SoAware func:ons: data entry, data - Diff spa:al resolu:on: X mean higher
l Records have an internal ID number column (OBJID, Data format
management, thema:c mapping, data resolu:on is beaer
OIC, FID etc.): unique number may have gaps File extension à 3-4 leaer codes designate types
analysis, may layout l Geometry type in the shape column (only appear in 2. Run-length encoding
of a file (.aprx/ .gdb/ .lyrx/ .tbx)
- Two relevant geospa:al technique: remote attribute table) - Compact/ compressed version to save data
sensing (RS), global posi:oning system (GPS) l Visualize attribute in attribute table storage
Shapefiles (green): contains only 1 feature class
- GPS: satellite-based naviga:on system made Data structure: - Ordered row-by-row/ column/ con:nuous
with many features of same geometry + 1
up of network of 24 satellites (min) placed 1. Spagheb: line following method line
aaribute table
into orbit by US Dept of Defense 1973 à - Point of intersec:on X coded explicitly - Useful in image scans, digital satellite image
- in Catalog: look like a single file (.shp) à
- Data stored as strings of coordinates data & raster output device
provide precise loca:on (100m to sub-cm) on copy everything
- Shared/ common borders duplicated i.e. - SVF, IDRIS, ERDAS
the earth’s surface à receiver measure - in Windows: many diff files for each single
unnecessary storage 3. Quadtree representa:on
distance using travel :me or radio signals shapefile
- Lines X connected/ unlinked i.e. X suitable - Comprise square cells of varing sizes
from at least 3 satellites - always use Catalog to copy/ move delete
for network analysis - Recursive encoding method by successively
n Rural: receive more satellite info à - Raw data from graphics viewing only dividing squares into smaller squares
shapefile
higher accuracy - aaribute table: stored in .dbf file à Excel can
2. Arc-Node or topological - Rapid search and data manipula:on
- RS: iden:fy, observe, measure object read
- Suitable for network analysis & spa:al - Original data must be rela:vely
without coming into direct contact à from queries homogenous: small storage
Geo data bases (grey): .gdb/ mdb.
aircraA/ satellite à key input of GIS raster - X duplicated lines i.e. storage efficiency - Compact & efficient storage for maps with
- Icons show geometry
data - Resource intensive coding exercise large homogenous area
- Feature class (points, lines, polygon) à no
1. Smart, interac:ve func:onal map - Computa:onal :me required for loca:ng
extension
2. Set of tools and procedures (language to features (e.g. nodes, polygons, links)& Raster edi:ng & enhancement
- Feature dataset “folder” for beaer grouping
perform tasks) calcula:ons (lengths, perimeter etc.) - Reduce storage requirement
- Must use Catalog to look inside
3. Well-managed system of informa:on - Rapid screen presenta:on of a magnified - Enable vectoriza:on of raster
- GeoDB store many data types (table,
- Geographically analysis what microscope, por:on - Image enhancement: techniques applied to
topology, networks etc.)
telescope and computer have been to other - Intelligent data digital images to change their appearance &
- can import/ export shapefiles from/ to
sciences - DLG, DIME, ESRI Arc, shapefile, make them easier to interpret à generally
GeoDB
- Integrate spa:al and infos within a system done with specialized soAware called image
- manipulate and display geographical Spa:al topology types processing soAware
Project file (aprx): opens ArcGIS project directly
knowledge in new and exci:ng way - Adjacency: next to, border on - image enhanced by 1. Varing levels of
- X contain any GIS data
- Access to administra:ve records via - Connec:vity: from/ to brightness/ contrast according to diff
- Contains only links to GIS data files
geographical posi:on - Containment/ coincidence: located in/ on, mathema:cal formulas or shapes & size or
- ! = X find file
- Convergence of technological fields and comprise, belong to frequency of histograms for brightness &
- Delete project file X delete data files it links
tradi:onal discipline contrast value 2. Applying spa:al filters (low-
to
Data Models (conceptual): describe only in terms Raster data (square grid): collec:on of points pass or high-pass)
- Store info about appearance only: symbol
of models à Vector/ Raster (pixels, cells) as 2D matrix
type, color of features etc. per layer/ layout,
Data structure: concerns the physical - Each cell correspond to a loca:on on Earth
prin:ng
arrangement of data in the computer à - Each cell contains one or more number
- Digital eleva:on model DEM: cell encodes Display and symboliza3on
resembles some form of data compression
eleva:on but shown as colour Repair bad links (!) à name or Path to data file
techniques
- Image: cell contain number as colour loca:on is wrong à need to find and set
- Uniform, regular cells of rectangular, square, correct path
Vector Data: single en:ty à point, line, polygon
, *Layer à Proper:es à Source à set data of the data Standard devia3on: how far away is the number coordinate grid if coordinate of loca:on must
source - hollow polygons, outline only, internal colour from the mean à useful for ra:o data be looked up
Map scale: ra:o/ frac:on à communicate detail set to invisible - lower than mean: Red
level of the map toggle labels on [right-click: label] - higher than mean: blue Map: Aim à target audience
- 1:10000: 1 length unit on map = 10000 Unique values: categorical à represent dis:nct - further away from mean: darker colour - target audience: info they need
length unit in reality (10000 à scale no.) categories/ groups - data layer needed for the objec:ve
- 1:10000 (0.0001) > 1:50000 (0.00002) - draws geometry & visualize categorical Unclassed colors: color picked directly from - color and symbols for layers à visualize data
- Large scale maps: zoomed in à small scale aaribute (column) smoothly varying linear color ramp à many - context helper layer e.g. major roads, ci:es
no. = large scale map à more detail - categorical values: X numerical à X do math slightly different color à can use color scheme etc. à perform GIS analysis to create
- Easy: metric units - data values “words”, ID, codes, type, that changes through mul:ple color (hues) addi:onal
- Tricky: imperial units category - users cannot easily convert a color to a value
- Required: physical size & page size (paper/ - X implied ordering Graduated symbol: make classes à symbol size Layout: add more map elements depending on
monitor) - Many features share same data value (min à max) the audience’s needs
1. Physical size (metric unit): map distance à - Typically: 3-20 categories à each expressed - size: in points, same as for fonts - N arrow, legend, scale bar, scale text, other
solve equa:on (measurement tools) with unique symbol - if symbol too big: make symbol transparent/ text, overview maps, images
2. Cal Scale: convert units (e.g. 1 yd= 36 inches) Graduated colour: numerical à define class hollow - balance elements
à set up the ra:on (Rule of Three) boundaries for easier differen:a:on Propor3onal symbol: no discrete classes with - readability: adjust size
1. Rank/ order (ordinal): do math in limited class breaks à scaled propor:onally - print map: pdf/ jpg
Mapping GIS data: cartography sense e.g. rank of state by pop, grades - need to choose how many entries the legend
1. Show loca:on of features 2. Decimal numbers/ integers e.g. T°C, $ has Map elements
2. Express aaribute values visually - Ra:o: true 0 pt e.g. height, weight Charts: similar to excel chart à select a field from - scale number: round to 2-3 digits
- Avoid info cluaer: choose suitable info only - Interval: lack true 0 pt e.g. T°C, calendar year the layer aaribute table à export as SVG to use in - scale bar: round number à only a few
Classifica3on à histogram à define break values other applica:on (illustrator etc.) subdivisions
Data types (width of each class, no of class) - can set the number of bins - text: :tle, coordinate system info, author,
Nominal: no repeated values e.g. state name Depends on data distribu:on and objec:ve of - can apply a transform date
Categorical: dis:nct categories/ groups e.g. road map - format axis, color, :tle etc. - legend: use nice value ranges à round up/
types, land cover Equal interval: same width for each class à bad Layer à create chart à chart type down
Ordinal: categorical data with meaningful order/ interval when there is crummy no. Field: [variable à number] - clean layer names (aaribute) in legend
ranking e.g. edu level - Need number of intervals - even distribu:on of all map elements
Numerical or con3nuous data Defined interval: good for comparing maps Map layout and design (balance)
- Ra:o: can be arranged along a scale but the - Calculate number of number classes Normaliza3on (graduated color): adjust data to
scale begins at non-arbitrary 0 point e.g. - Need width of each class allow fair comparison and analysis across diff Map alignment and balance
popula:on, eleva:on Natural breaks (Jenk’s method): automa:c dataset/ scale Process map from top à boaom, leA à right
- Interval: can take any value within a relevant defini:on of class boundaries à op:mize variety - density: fairer approach than raw no. - align edges of boxes and frames exactly using
range of scale and arbitrary 0 e.g. T°C of colours shown on map à good-looking maps - transform skewed distribu:on into normal guides and snapping
- Not good for comparison: class boundaries distribu:on à equal intervals make more - use margins (best: same distance
Aaribute table diff sense everywhere) à even for legends
- Quan::es (no.): visualized by symbol size, Automa3c calcula3on: sort bin and finds - center the most important element à fill
<none>: no normaliza:on
thickness, colour “best” n clusters à find the biggest remaining space evenly
<percentage>: % instead of no.
- Categories (text): visualized by symbol shape, discon:nui:es (“jumps”) à good to isolate - neatlines: boxes à structure space
line type, paaern, font, colour and visualize clusters of classes - landscape
Assembling a map: selec:ve of amount of info
- Beware: some:mes text looks like number Quan3le: each class contain same number of - spread layers over different data frames &
states Color choice
provide common context for all layouts (base
Symbology type - Overemphasis on large states compare to - mimic nature tones: blue for water, green for
map, roads etc.)
Single symbol: nominal à draw feature geometry Natural breaks à not good for skewed forest etc.
- important elements stand out (color/ size)
but use same appearance for all features distribu:on - pastel or subdued color for most of the map
- tell story with diff frames: current situa:on,
- no aaribute needed as data à show value Geometrical interval: op:mize number of à bold color only for emphasis
problem, solu:on à add chart, images, text
via label elements à compromise between equal - sequen:al color ramps: increase intensity for
to support story
- typical but not enforced: X repeated values intervals, natural breaks and quan:les à works just 1 color à express increase in numeric
- provide spa:al (geographic) context: show
- good for gebng an overview of the loca:on well for skewed distribu:on aaribute
roads, zoom-in from big picture à add
Pros 1. compact data structure 1. good for complex analysis
Computer-based tool that collects, analyze, store, is on a sphere but pretend the Earth is flat 2. efficient for encoding topology efficient for overlays
manipulate and visualize geographical l Feature class: collec:on of many features Data structure: 3. true representation of shape 2. efficient for overlays
informa:on on a map à problem-solving l Layer: contains feature of same type of 1. Grid (assuming square grids) à Cell = pixel 3. data structure common for
imagery
technology geometry à gets a separate entry in the - Spa:al resolu:on depend on cell size Cons 1. complex structure 1. large datasets
- Components: computer hardware pla=orm, Contents Panel - Features are generalized to a constant value 2. overlay operations difficult 2. topology hard to represent
GIS soAware, data storage, data input l Aaribute table: rows VS column à associate - Lines & polygons appear jaggy 3. might imply false sense of 3. maps less realistic
accuracy
hardware, informa:on output hardware, GIS with the vector data - Direct computa:on possible (for lengths,
data and GIS personal Row feature/ record perimeters and areas)
Column attribute/ field
- SoAware func:ons: data entry, data - Diff spa:al resolu:on: X mean higher
l Records have an internal ID number column (OBJID, Data format
management, thema:c mapping, data resolu:on is beaer
OIC, FID etc.): unique number may have gaps File extension à 3-4 leaer codes designate types
analysis, may layout l Geometry type in the shape column (only appear in 2. Run-length encoding
of a file (.aprx/ .gdb/ .lyrx/ .tbx)
- Two relevant geospa:al technique: remote attribute table) - Compact/ compressed version to save data
sensing (RS), global posi:oning system (GPS) l Visualize attribute in attribute table storage
Shapefiles (green): contains only 1 feature class
- GPS: satellite-based naviga:on system made Data structure: - Ordered row-by-row/ column/ con:nuous
with many features of same geometry + 1
up of network of 24 satellites (min) placed 1. Spagheb: line following method line
aaribute table
into orbit by US Dept of Defense 1973 à - Point of intersec:on X coded explicitly - Useful in image scans, digital satellite image
- in Catalog: look like a single file (.shp) à
- Data stored as strings of coordinates data & raster output device
provide precise loca:on (100m to sub-cm) on copy everything
- Shared/ common borders duplicated i.e. - SVF, IDRIS, ERDAS
the earth’s surface à receiver measure - in Windows: many diff files for each single
unnecessary storage 3. Quadtree representa:on
distance using travel :me or radio signals shapefile
- Lines X connected/ unlinked i.e. X suitable - Comprise square cells of varing sizes
from at least 3 satellites - always use Catalog to copy/ move delete
for network analysis - Recursive encoding method by successively
n Rural: receive more satellite info à - Raw data from graphics viewing only dividing squares into smaller squares
shapefile
higher accuracy - aaribute table: stored in .dbf file à Excel can
2. Arc-Node or topological - Rapid search and data manipula:on
- RS: iden:fy, observe, measure object read
- Suitable for network analysis & spa:al - Original data must be rela:vely
without coming into direct contact à from queries homogenous: small storage
Geo data bases (grey): .gdb/ mdb.
aircraA/ satellite à key input of GIS raster - X duplicated lines i.e. storage efficiency - Compact & efficient storage for maps with
- Icons show geometry
data - Resource intensive coding exercise large homogenous area
- Feature class (points, lines, polygon) à no
1. Smart, interac:ve func:onal map - Computa:onal :me required for loca:ng
extension
2. Set of tools and procedures (language to features (e.g. nodes, polygons, links)& Raster edi:ng & enhancement
- Feature dataset “folder” for beaer grouping
perform tasks) calcula:ons (lengths, perimeter etc.) - Reduce storage requirement
- Must use Catalog to look inside
3. Well-managed system of informa:on - Rapid screen presenta:on of a magnified - Enable vectoriza:on of raster
- GeoDB store many data types (table,
- Geographically analysis what microscope, por:on - Image enhancement: techniques applied to
topology, networks etc.)
telescope and computer have been to other - Intelligent data digital images to change their appearance &
- can import/ export shapefiles from/ to
sciences - DLG, DIME, ESRI Arc, shapefile, make them easier to interpret à generally
GeoDB
- Integrate spa:al and infos within a system done with specialized soAware called image
- manipulate and display geographical Spa:al topology types processing soAware
Project file (aprx): opens ArcGIS project directly
knowledge in new and exci:ng way - Adjacency: next to, border on - image enhanced by 1. Varing levels of
- X contain any GIS data
- Access to administra:ve records via - Connec:vity: from/ to brightness/ contrast according to diff
- Contains only links to GIS data files
geographical posi:on - Containment/ coincidence: located in/ on, mathema:cal formulas or shapes & size or
- ! = X find file
- Convergence of technological fields and comprise, belong to frequency of histograms for brightness &
- Delete project file X delete data files it links
tradi:onal discipline contrast value 2. Applying spa:al filters (low-
to
Data Models (conceptual): describe only in terms Raster data (square grid): collec:on of points pass or high-pass)
- Store info about appearance only: symbol
of models à Vector/ Raster (pixels, cells) as 2D matrix
type, color of features etc. per layer/ layout,
Data structure: concerns the physical - Each cell correspond to a loca:on on Earth
prin:ng
arrangement of data in the computer à - Each cell contains one or more number
- Digital eleva:on model DEM: cell encodes Display and symboliza3on
resembles some form of data compression
eleva:on but shown as colour Repair bad links (!) à name or Path to data file
techniques
- Image: cell contain number as colour loca:on is wrong à need to find and set
- Uniform, regular cells of rectangular, square, correct path
Vector Data: single en:ty à point, line, polygon
, *Layer à Proper:es à Source à set data of the data Standard devia3on: how far away is the number coordinate grid if coordinate of loca:on must
source - hollow polygons, outline only, internal colour from the mean à useful for ra:o data be looked up
Map scale: ra:o/ frac:on à communicate detail set to invisible - lower than mean: Red
level of the map toggle labels on [right-click: label] - higher than mean: blue Map: Aim à target audience
- 1:10000: 1 length unit on map = 10000 Unique values: categorical à represent dis:nct - further away from mean: darker colour - target audience: info they need
length unit in reality (10000 à scale no.) categories/ groups - data layer needed for the objec:ve
- 1:10000 (0.0001) > 1:50000 (0.00002) - draws geometry & visualize categorical Unclassed colors: color picked directly from - color and symbols for layers à visualize data
- Large scale maps: zoomed in à small scale aaribute (column) smoothly varying linear color ramp à many - context helper layer e.g. major roads, ci:es
no. = large scale map à more detail - categorical values: X numerical à X do math slightly different color à can use color scheme etc. à perform GIS analysis to create
- Easy: metric units - data values “words”, ID, codes, type, that changes through mul:ple color (hues) addi:onal
- Tricky: imperial units category - users cannot easily convert a color to a value
- Required: physical size & page size (paper/ - X implied ordering Graduated symbol: make classes à symbol size Layout: add more map elements depending on
monitor) - Many features share same data value (min à max) the audience’s needs
1. Physical size (metric unit): map distance à - Typically: 3-20 categories à each expressed - size: in points, same as for fonts - N arrow, legend, scale bar, scale text, other
solve equa:on (measurement tools) with unique symbol - if symbol too big: make symbol transparent/ text, overview maps, images
2. Cal Scale: convert units (e.g. 1 yd= 36 inches) Graduated colour: numerical à define class hollow - balance elements
à set up the ra:on (Rule of Three) boundaries for easier differen:a:on Propor3onal symbol: no discrete classes with - readability: adjust size
1. Rank/ order (ordinal): do math in limited class breaks à scaled propor:onally - print map: pdf/ jpg
Mapping GIS data: cartography sense e.g. rank of state by pop, grades - need to choose how many entries the legend
1. Show loca:on of features 2. Decimal numbers/ integers e.g. T°C, $ has Map elements
2. Express aaribute values visually - Ra:o: true 0 pt e.g. height, weight Charts: similar to excel chart à select a field from - scale number: round to 2-3 digits
- Avoid info cluaer: choose suitable info only - Interval: lack true 0 pt e.g. T°C, calendar year the layer aaribute table à export as SVG to use in - scale bar: round number à only a few
Classifica3on à histogram à define break values other applica:on (illustrator etc.) subdivisions
Data types (width of each class, no of class) - can set the number of bins - text: :tle, coordinate system info, author,
Nominal: no repeated values e.g. state name Depends on data distribu:on and objec:ve of - can apply a transform date
Categorical: dis:nct categories/ groups e.g. road map - format axis, color, :tle etc. - legend: use nice value ranges à round up/
types, land cover Equal interval: same width for each class à bad Layer à create chart à chart type down
Ordinal: categorical data with meaningful order/ interval when there is crummy no. Field: [variable à number] - clean layer names (aaribute) in legend
ranking e.g. edu level - Need number of intervals - even distribu:on of all map elements
Numerical or con3nuous data Defined interval: good for comparing maps Map layout and design (balance)
- Ra:o: can be arranged along a scale but the - Calculate number of number classes Normaliza3on (graduated color): adjust data to
scale begins at non-arbitrary 0 point e.g. - Need width of each class allow fair comparison and analysis across diff Map alignment and balance
popula:on, eleva:on Natural breaks (Jenk’s method): automa:c dataset/ scale Process map from top à boaom, leA à right
- Interval: can take any value within a relevant defini:on of class boundaries à op:mize variety - density: fairer approach than raw no. - align edges of boxes and frames exactly using
range of scale and arbitrary 0 e.g. T°C of colours shown on map à good-looking maps - transform skewed distribu:on into normal guides and snapping
- Not good for comparison: class boundaries distribu:on à equal intervals make more - use margins (best: same distance
Aaribute table diff sense everywhere) à even for legends
- Quan::es (no.): visualized by symbol size, Automa3c calcula3on: sort bin and finds - center the most important element à fill
<none>: no normaliza:on
thickness, colour “best” n clusters à find the biggest remaining space evenly
<percentage>: % instead of no.
- Categories (text): visualized by symbol shape, discon:nui:es (“jumps”) à good to isolate - neatlines: boxes à structure space
line type, paaern, font, colour and visualize clusters of classes - landscape
Assembling a map: selec:ve of amount of info
- Beware: some:mes text looks like number Quan3le: each class contain same number of - spread layers over different data frames &
states Color choice
provide common context for all layouts (base
Symbology type - Overemphasis on large states compare to - mimic nature tones: blue for water, green for
map, roads etc.)
Single symbol: nominal à draw feature geometry Natural breaks à not good for skewed forest etc.
- important elements stand out (color/ size)
but use same appearance for all features distribu:on - pastel or subdued color for most of the map
- tell story with diff frames: current situa:on,
- no aaribute needed as data à show value Geometrical interval: op:mize number of à bold color only for emphasis
problem, solu:on à add chart, images, text
via label elements à compromise between equal - sequen:al color ramps: increase intensity for
to support story
- typical but not enforced: X repeated values intervals, natural breaks and quan:les à works just 1 color à express increase in numeric
- provide spa:al (geographic) context: show
- good for gebng an overview of the loca:on well for skewed distribu:on aaribute
roads, zoom-in from big picture à add
