Week 8 Flashcards
which data type is more locationally acurate (vector or raster)
Vector because of high res outline of boundary
Main single advantage of raster analysis
- ability to analyse continuous surface variables (e.g slopes, density, proximity)
- suitable for surface analysis (e.g terrains, flow analysis, traffic travel time)
Raster analysis formats
- contour elevation lines
- Red, Green, Blue (RGB) images
- Satellite images
- encoded Integer or real number data
Raster cell value types
- georeferenced (e.i numerical value = land type)
- Cells have ground resolution
- integer (whole #), or real/floating (ei. 4.6)
Zonal raster (with attributes)
- integer valued
- value signifies zone ID
- Zones need not be contiguous
- linked attribute table
Raster (grid cell) overlay
- overlay of grid cells
- each cell has numerical values (can be any scale of measurement)
- two or more layers
- simpler than vector data overlay
- more efficient and no sliver problem
- layers have same coordinate system and projection
- layers comparable in measurement scale
map algebra
- mathematic symbols for manipulating geographic data
- can perform all system analysts tools on arcGIS
scale of analysis
- Local: per cell
- Neighbourhood: surrounding cells of single target cell
- Zonal: groups of cells defined by a zone
- Global: entire raster (e.i. distance, flow)
can map algebra in local cell operation be done on both single layer rasters and multiple layer rasters?
Yes
- single layer: sin, log, sqrt
- Multiple layer: mean min
Local cell operation
- Single cell algebra application (e.i mean of two layers)
- Reclassification (or grouping)
Neighbourhood Operation
- set off cells to use aggregate functions on
Zonal Operation
Two inputs
- Zones: raster with integer value for a group of cells (i.e. administrative area, catchment)
- Value raster: input values of interest to be summarised (i.e population, elevation)
Global operation: type of distances
- Euclidean (cell to cell straight line distance)
- Incremental or path
Arithmetic operators
- produce a new layer through the operations +, -, *, /
- Any arithmetic operation on NODATA results in NODATA
Boolean operations - not, and, or, xor
- Input value of ‘0’ is FALSE and a non-zero value is TRUE
- Output values of TRUE are written as ‘1’
- NODATA=NODATA
Combinatorial operators
-assign new output values to corresponding combination of input values
Logical operators
ei. difference (DIFF), contained in (IN), over (OVER)
Descriptive Cartographic models
- those that describe patterns and pattern associations
Prescriptive cartographic models
- those that prescribe actions, solutions or allocations of space
- focus on solving a problem or proposing alternative options
- show user the benefits of manipulation existing attributes
Raster analysis applications
- Land use analysis
- Terrain and hydrological analysis
- Environmental analysis
Land use analysis
- suitability analysis
- simple models
Suitability analysis def + e.g
- illustrates combination of preference maps
e. g Finding best area of siting a ski resort - combine layers: depth(m), slope (degrees) and sunlight (aspect direction).
- above factors easily obtain as rasters
- reclassify each factor to score
- Combine with map algebra as weighted combination (e.g weighted influence of snow=50%, slope=30%, sun= 20%
Environmental analysis:
- illustrates analysis for data association
e. g broad vegetation groups (BVG) are stratified by elevation. Output is a zonal summary of BVG and elevation to find statistical association (i.e mean and deviation)
