GEG 410 GIS II Lecture 3: Vector & Raster Model Fall 2024 PDF

GEG 410 GIS II Lecture 3: Vector & Raster Model Fall 2024 Instructor: Dr. Guimin Zhu Email: [email protected] Representing Real-world Objects with GIS There is usually more than one way to represent a single object or space with GIS. For example, ta...

GEG 410 GIS II Lecture 3: Vector & Raster Model Fall 2024 Instructor: Dr. Guimin Zhu Email: [email protected] Representing Real-world Objects with GIS There is usually more than one way to represent a single object or space with GIS. For example, take Campo Sano Building GIS Data Models A GIS data model is the representation of real-world objects in GIS GIS data can be either vector or raster: Vector data model corresponds to object-based conceptual representation; discrete objects such as points, lines, and polygons Raster data model corresponds to field-based conceptual representation; cell matrixes (arrayed in rows and columns) that store numeric values GIS Data Models Contents Vector Model Raster Model Rasterization & Vectorization TIN Vector Data Model In the vector data model, features are real-world objects such as roads, property boundaries, schools, administrative boundaries, etc., that have two elements: Geometry and Attributes FEATURE GEOMETRY ATTRIBUTES Point Line Polygon Table Vector Data Model – Point, Line, & Polygon Human Computer Point: an (x, y) coordinate pair Point Line: a sequence of ordered coordinate pairs Polygon: enclosed shape with a sequence of interconnected lines, Line and the first and last (x, y) coordinates are the same Polygon Vector Topology Explicit information describing adjacency, connectivity, and relative spatial relationships of features Why Topology is important Enables relationships to be established between features Allows error detection to be done in a GIS Are all polygons completely closed? Do the ends of arcs match up (“snapped together”)? Vector Topology - Arc/Node Model Started in the 1960s Based on the fact that each feature type is composed of features with one fewer dimensions The problem is that it is necessary to keep track of the links between features in an arbitrary and sometimes cumbersome way  Nodes File 1 x y 2 x y 3 x y 4 x y 5 x y 6 x y 7 x y 8 x y 9 x y 10 x y 11 x y Arcs File File of Arcs by Polygon 12 x y 1 1, 2, 3, 4, 5, 6, 7 A 1, 2, Area, Attributes 13 x y 2 1, 8, 9, 10, 11, 12, 13, 7 Vector Data Model – Spaghetti Model The spaghetti model stores individual objects separately as geometric entities Vector Data Model – Spaghetti Model Advantages Easy to create, store, update, and delete an object Used in computer mapping systems Disadvantages Does not explicitly record connections of line segments when they cross or meet Duplication: objects are stored twice and therefore may not be the same (gaps or overlap) Difficult to do analysis: which objects are adjacent? How objects are spatially related? Attributes of Vector Data Every vector data has an attribute table Select the attribute table Manipulate the columns/fields: Add, delete, visible, etc. Advantages of Vector Data Looks the same at any scale Advantages of Vector Data Easy to re-project (change coordinate systems) Contents Vector Model Raster Model Rasterization & Vectorization TIN Raster Data Model Represent continuous fields Elevation, temperature, precipitation, air pollution, etc. Source: Haywood, Cornelius, Carver. An Introduction to Geographical Information Systems. Person. Raster Data Model ArcGIS Pro tool: Raster to ASCII https://pro.arcgis.com/en/pro-app/latest/tool- reference/conversion/raster-to-ascii.htm Raster Data Model – Cell Size & Resolution https://www.usgs.gov/landsat-missions/landsat-9 https://modis.gsfc.nasa.gov/data/dataprod/mod11.php Types of Raster Data Continuous: representing a surface or data that is constant Floating point or integer Examples Topology Satellite imagery Proximity Types of Raster Data Discrete Could be binary (0 & 1) Could be a limited number of unique values Examples Land cover classification Suitability Presence/absence  You can always Reclassify continuous to discrete, but never discrete to continuous Introduction to Remote Sensing Remote Sensing is the process of detecting and monitoring the physical characteristics of an area by measuring its reflected and emitted radiation at a distance (typically from satellite or aircraft). Introduction to Remote Sensing Ground objects have different reflectance rates to different wavelengths. Satellites take pictures where the pixels store the reflectance rates for different bands. Examples – Landsat Landsat 1 was launched by NASA in 1972, becoming the first earth-observing satellite explicitly designed to study planet Earth. Landsat 9: launched in 2021 Resolution: 30-meters Orbits duration: 99 minutes orbits per day: ~14 Revisit: every 8 days (with Landsat 8) Examples – Sentinel European Space Agency Sentinel-1 was launched in 2014 at an approximately spatial resolution of 20 meters Sentinel-2 Sentinel-2A (2015) and Sentinel-2B (2017) Resolution: 10m Revisit: every 10 days Examples - LiDAR LiDAR: an active remote sensing system that can be used to measure vegetation height across wide areas LASER: Light Amplification by Stimulated Emission of Radiation Additional reading: https://www.neonscience.org/resources/lear ning-hub/tutorials/lidar-basics Examples – Digital Elevation Models A digital elevation model (DEM) consists of an array of uniformly spaced elevation data New techniques for DEM generation include optical sensors, interferometric synthetic aperture radar (InSAR), and light detection and ranging (LiDAR) Examples – Google Earth Engine Google Earth Engine Multi-petabyte catalog of satellite imagery and geospatial datasets Planetary-scale analysis capabilities Timelapse: https://earthengine.google.com/timelapse/ Examples – Remote Sensing Data Product Land Use Land Cover map Advantages & Disadvantages – Vector Advantages Good representation of reality More efficient data storage Topology can be described in a network Accurate graphics Disadvantages Complex data structures Simulation may be difficult Some spatial analysis operations are difficult or impossible to perform Advantages & Disadvantages – Raster Advantages Simple data structure Easy overlay Various kinds of spatial analysis Uniform size and shape Cheaper technology Disadvantages Large amount of data Less “pretty” Projection transformation is difficult Different scales between layers can be a nightmare May lose information due to generalization Vector vs Raster Raster Vector Representation Suitable for fields without Good for objects with discrete discrete boundaries boundary and uniform inside Storage requirements Redundant Compact Topology No topology stored (no Efficient topology relationships) Spatial accuracy Lower positional accuracy Excellent positional accuracy (depending on cell size) Compatibility to computer Easy to code in an array Often complex (multiple structure in most computer tables) language Display and output Good for fields, discrete Map-like, with continuous features may show “stair-like” curves borders Analysis Simple for many layer Preferred for network combinations analysis; many other spatial operations are complex Contents Vector Model Raster Model Rasterization & Vectorization TIN Rasterization & Vectorization Vector -> Raster: Rasterization Raster -> Vector: Vectorization Some information is always lost when converting from one data format to the other Digitizing Rasterization Rasterization loses topological information No information about relationships Positional accuracy decreases, particularly at boundaries “winner-take-all” strategy In ArcGIS Pro: Tools -> Conversion -> To Raster -> Point to Raster Polyline to Raster Polygon to Raster Vectorization Feature boundaries become jagged in the vector representation Topology is created In ArcGIS Pro: Tools -> Conversion -> From Raster -> Raster to Point Raster to Polyline Raster to Polygon Contents Vector Model Raster Model Rasterization & Vectorization TIN Triangulated Irregular Network (TIN) A TIN approximates the terrain with a set of nonoverlapping triangles Ideal for irregularly spaced datasets such as LiDAR or drone-based photogrammetry Fewer nodes can be used in areas that don’t change much The data structure of a TIN