Unit 4: Data Types and Data Analysis in GIS

1. Core Concepts in GIS Data

1.1 Data Types in GIS

Geographic Information Systems (GIS) manage and analyze spatial data. Understanding the different ways spatial data is represented is crucial. The two primary data models in GIS are Raster and Vector.

Definition: Spatial Data refers to information that has a geographical or spatial component, describing the location, shape, and relationship of features on the Earth.

1.1.1 Raster Data

Raster data represents geographic features as a grid of cells or pixels. Each cell holds a value representing a specific attribute, such as elevation, temperature, or land cover. These cells are typically of uniform size and arranged in rows and columns.

• Characteristics: Each pixel represents a geographic area, and the resolution is determined by the pixel size. Raster data is excellent for continuous data like elevation, temperature, or satellite imagery.
• Examples: Satellite images, aerial photographs, Digital Elevation Models (DEMs), land cover maps, temperature maps.
• Common Mistake: Confusing pixel size with file size. A smaller pixel size (higher resolution) results in a larger file size.

1.1.2 Vector Data

Vector data represents geographic features using discrete geometric objects: points, lines, and polygons. Each object has a defined shape and location and is associated with attributes stored in a table.

• Points: Represent features too small to be depicted as lines or polygons, such as individual trees, wells, or specific addresses.
• Lines (or Arcs): Represent linear features like roads, rivers, pipelines, or boundaries. They are defined by an ordered sequence of points (vertices).
• Polygons: Represent area features such as lakes, countries, land parcels, or buildings. They are defined by a closed loop of lines.
• Characteristics: Vector data provides precise location and shape information. It's ideal for representing discrete features and clearly defined boundaries.
• Examples: Roads, property boundaries, city limits, rivers, utility lines.

1.1.3 Raster vs. Vector Data

Choosing between raster and vector depends on the nature of the data and the analysis required.

1.2 Data Analysis in GIS

GIS data analysis involves applying various operations and tools to spatial data to extract meaningful information, discover patterns, and support decision-making. It transforms raw spatial data into actionable insights.

2. Spatial Reference Systems

2.1 Projection Systems

The Earth is a three-dimensional, irregular shape (a geoid or spheroid). To represent it on a two-dimensional map, a projection system is used. This process inevitably introduces some form of distortion.

Definition: A Map Projection is a systematic transformation of the latitudes and longitudes of locations on the surface of a spheroid or ellipsoid into locations on a planar surface (a map).

No projection can preserve all spatial properties simultaneously. A map projection will distort one or more of the following:

• Shape: The geometric forms of features.
• Area: The relative sizes of features.
• Distance: The scale along certain lines.
• Direction: The angles from a central point.

Different projection types minimize certain distortions while maximizing others, making them suitable for specific applications.

2.1.1 Common Types of Projections (based on projection surface)

• Cylindrical Projections: Project the Earth's surface onto a cylinder. They are good for showing true direction and are often used for world maps (e.g., Mercator). Distortion increases towards the poles.
• Conic Projections: Project the Earth's surface onto a cone. They are good for showing regions with an east-west extent, like the USA. Distortion is minimal along the parallel where the cone touches the globe.
• Azimuthal (Planar) Projections: Project the Earth's surface onto a flat plane. They are often used for polar regions or for showing great circle routes, preserving true direction from a central point.

2.2 Georeferencing

Georeferencing is the process of assigning real-world coordinates to an image or a dataset that lacks spatial reference. This makes the data usable within a GIS alongside other spatial datasets.

Definition: Georeferencing is the process of spatially aligning a dataset (e.g., a scanned map, an aerial photo) to a known coordinate system, allowing it to be integrated with other spatially referenced data.

• Purpose: To accurately place an unreferenced image or map onto its correct location on the Earth's surface, enabling it to be analyzed with other spatial data.
• How it works: It typically involves identifying a set of Ground Control Points (GCPs) on the unreferenced image and matching them to their corresponding known coordinates (from a referenced map or GPS data).
• Transformation: Once GCPs are established, a mathematical transformation (e.g., affine, polynomial) is applied to warp the unreferenced image into the correct spatial location and orientation.
• Observation: The accuracy of georeferencing heavily depends on the number and spatial distribution of accurate GCPs.

3. GIS Data Operations

3.1 Overlaying of Data

Data overlay is a fundamental GIS operation that combines the spatial and attribute data from two or more input layers to create a new output layer. This process allows for the analysis of relationships between different geographic features.

Definition: Data Overlay in GIS is the process of combining two or more spatially referenced layers (e.g., maps, datasets) to produce a new layer that integrates information from all inputs, revealing spatial relationships.

The type of overlay operation depends on whether you are working with vector or raster data.

3.1.1 Vector Overlay Operations

Vector overlay operations are typically performed on polygon layers, but can also involve points and lines.

• Intersect: Computes the geometric intersection of input features. Only features or portions of features that overlap in all input layers are preserved in the output. The output contains attributes from both input layers.
• Union: Computes a geometric union of all input features. All features from all input layers are preserved in the output, along with their attributes. This operation creates a new polygon for every unique combination of overlapping areas.
• Erase: Removes areas of the input layer that overlap with the erase layer. It's like cutting a hole in the input layer using the erase layer as a cookie-cutter.
• Identity: Computes the geometric intersection of input features and identity features. All features of the input layer are maintained, but those features that overlap with the identity layer have the attributes of the identity layer appended.

3.1.2 Raster Overlay Operations

Raster overlay often involves combining cell values from multiple raster layers through mathematical operations.

• Local Operations: Performed on individual cell values, e.g., adding two raster layers cell-by-cell.
• Focal Operations: Calculate an output value for each cell based on the values of neighboring cells within a defined window (e.g., calculating a mean or maximum within a 3x3 neighborhood).
• Zonal Operations: Calculate an output value for each cell based on the values of cells within a defined zone (e.g., calculating the average elevation within a specific watershed polygon).
• Weighted Overlay: Assigns weights to different input rasters based on their importance to a specific analysis, then combines them. This is common in site suitability analysis.

3.2 Querying of Data

Querying data in GIS involves selecting features or subsets of data based on specific criteria. It's a way to extract information from the dataset that meets certain conditions.

Definition: Data Querying in GIS is the process of retrieving a subset of features or records from a dataset based on specified conditions, which can be either attribute-based (non-spatial) or spatial.

There are two main types of queries:

3.2.1 Attribute Queries (Non-Spatial Queries)

Attribute queries select features based on their non-spatial characteristics stored in the attribute table. These queries often use SQL-like expressions.

Example: