GIS II: Vector Data and Frameworks

• Study notes for GIS II fundamentals are outlined below

Vector Data & GIS

• Data represents observations about social, environmental, or physical phenomena
• Geographic data subset explicitly combines coordinate and attribute data
• Position info (location, shape, size) and characteristics (attributes, fields) can be stored and managed in many ways
• Focus is on vector data model, filtering vector features using attribute/spatial filters, and creating new vector datasets
• Filtering methods are ways to indicate co-location of features from existing vector layers

Data Organization & Frameworks

• Vector data spatially organizes related attribute information
• Statistics Canada organizes data collection/analysis around geographic enumeration units
• Data is logically structured by geographic boundaries for joining selected information
• Geographic boundaries are referred to as "framework data"
• Boundary layers and tabular data released by Statistics Canada contain identifiers and specific information
• Identifiers see little change over time
• Managed approach requires planned integration of coordinate and attribute data
• Organizations may lack resources for managed geographic information
• Government agencies have responsibility to provide data

Spatial Data Elements

• Spatial data is a conceptual combination of coordinate (where) and attribute (what) data
• Vector data model is a collection of discrete entities with coordinate data describing location and attributes describing characteristics

Data Accuracy & Metadata

• Overall analysis is limited by least accurate data
• Metadata clarifies the meaning of data and varies in quality
• Statistics Canada and City of Toronto metadata vary in clarity
• Both datasets are good for their purposes, but one is easier to work with due to clearer metadata
• Meanings are interpreted as best as possible in the absence of provided metadata

Attribute Interface Selection

• Attribute query interface found under Map Ribbon > Selection Section > Select by Attribute, with examples
• Set-based result descriptions use Venn diagrams (A, B, A AND B, A OR B)
• Compound logical queries compared to sequenced operations using the Method option in query builder interface
• The Select by Attribute methods and Venn diagrams provide equivalent Boolean logic
• Software only selects all features in a layer if none are selected or if "create a new selection" is selected
• If features are selected, those are used as the initial selection for the chosen method ("add to", "remove from", "select from")
• "Switch the current selection" and "Clear the current selection" are convenience methods for bulk selection changes, regardless of how the current selection was created

Spatial Queries

• Accuracy will be limited by the least accurate data
• Sequence-able interface for queries can combine attribute and spatial queries
• An equivalent set of method options are provided for Select by Location, similar to those for Select by Attributes
• Complex query sequences create selection sets based on spatial predicates (e.g., "within a distance of")
• Topological relations for "relationship" offer spatial predicate descriptions based on input dimensions
• Relationship types and supported geometry types change

Vector Overlays

• Vector overlay creates a result layer from two or more input layers using geometric combination rules based on spatial overlap

Workshop Exercises

• Combining layers using spatial operations involves vector overlay (creating a result layer from two or more input layers using predefined geometric combination rules)
• Spatial joins are a more direct analogy for attribute join because one target layer has attributes joined from another layer according to specified spatial relations
• Review methods and results of attribute selection and expand that to introduce spatial selections ("selection by location")

Data for Examples

• Example queries can be tried using the dataset for Part 2 of Assignment 1
• Union overlay operations replicate output shown in-assignment - examine the Overlay toolset and look for the tool help for each of the overlay tools you will be using in the first part of the assignment (intersect, union, erase, etc.)

Spatial Logic

• Vector overlay creates output feature geometries based on input layer geometries overlap and the spatial logic of selected overlay tool
• At least some tools (e.g., Intersect) have different possibilities for creating output feature geometries of variable dimensionality (point, line, or polygon output)
• Vector overlay produces geometries as input layer with lowest dimensionality geometry
• Note that with Intersect, it is impossible to increase dimensionality of the output above the dimension of the lowest input

Combining Layers with Spatial Operations

• Applying Vector overlay and spatial join tools requires consideration of active selection sets (whether resulting from attribute- or location-based selections) because these tools will only apply the selection using all features in a layer if none are selected
• If some features are selected, then only those features will be used in any tool operation
• ArcGIS maintains a selection list for each layer independently
• Creating a “new selection” retains both selection sets

Buffering

• Buffering is not an overlay, but often used to create polygon features from points or lines to be used as mask layer for overlay analysis

Attributes & Joins

• Transferring attribute data from one layer to another may be combining attributes from one or more layers
• Data written permanently (output feature class created) or temporarily (attribute joins in GIS or databases) that can be made permanent (exporting data from layer)
• Joins attributes from one (join) layer into a target layer based on topological relationships
• New output layer created
• Inner (target features only kept with at-least one join-layer satisfaction) joins versus outer

LiDAR Data Collection & Use

• Light Detection and Ranging (LiDAR) systems collect topography and terrain
• Laser pulses are aimed at the ground, and the intensity and timing of the reflections are recorded
• Partial returns arrive back at sensor before final return
• Manage LiDAR data to reduce computing requirements

Raster Data Models

• Models raster data by describing a continuous space in terms of spatial zones of attributes
• Cells usually square or rectangular with fixed dimensions if location and orientation of a single cell is known
• Easier to summarize a column based on possible values to give clarification on the information
• Metadata can tell the story on how the information is to be summarized

Combining Raster Layers

• Consist of three major parts: combining all the development restrictions into a single layer, cover information to a "restricted areas" raster layer
• Raw analysis from the input layer is converted into a suitability rating layer where values are 1-5, where ones indicate "not very suitable" and five means "most advantageous.“
• Suitability ratings are compiled together into a suitability rank using a weighter sum

Weighted Sum & Rasters

• Output Cell Value is the result of adding weighted values from the same location in a set of input rasters
• Allows a GIS analyst to define an expression combining input rasters, constants, and the returned values from functions computed using input raster values to produce a new output raster
• Zones raster may be specified that influence final determination of values at pixels

Iterators in Model Builder

• Iterators make it easy to run sets of operations across model elements like data, files, fields

Setting Up Vector Topology

• Specific data structure that inherently maintains topological relations between features within a single feature layer to detect and prevent/fix topographical errors between layers with different dimensionality

Map Algebra & Raster Analysis Workflows

• During this week's and last week's exercises we are discussing raster functions for basic suitability analysis.
• These functions can be classified according to their scope, defined as the set of raster cell locations relative to a particular raster cell location that will be used to determine the value computed for that cell location in the output raster layer created by each function (Figure 1).
• The scope categories are local, neighbourhood, zonal, and global. Each is described briefly below along with example geoprocessing tools available in ArcGIS that use each scope class in computing output raster layers.

Introduction to Time Series Data

• This lecture deals mainly with data that explicitly references time, recording information for events that happened at specific times or observations of phenomena taken during specific periods of time.
• The use of time in both vector and raster models is introduced. Vector data can hold time variables as attributes but raster data needs to have a level of structure added to enable the explicit storage of time.
• Mosaic data structures are introduced to manage time