An explanation of Raster data that is used in GIS and Remote Sensing
What is Raster Data?
Raster data uses millions of tiny squares that are referred to as pixels or grid cells to produce a digitized graphical image such as aerial or satellite photographs, or the image is of an elevated surface. The cells are comprised of colored cells, typically red, green, and blue that gives the features a more detailed natural appearance. Raster data presents a higher resolution of the area being portrayed in a model that is better suited for analytical modelling of data that has discrete characteristics such as an urban or forested area. This makes the use of raster data better suited for some spatial analysis methods since it represents how the real world looks such as data that is discrete like a forest (GISGeography.com). Since raster data uses pixels, the higher the resolution, the more space is required for storage of the data and could cause data analysis to run slower. The structure of raster data being pixels or cells means they can be sized, just like a JPEG image. The larger the pixel size the less clear the image appears and causing structures and features to be blurry, or unrecognizable. Typically a map using raster data can reflect only a single characteristic or area attribute due to the large amount of data that may exist. Modeling of raster data can produce various layer combinations or continuous data such as elevations or movement of contaminated water, and where temperatures increase and decrease (Bolstad 59-60). Geographical Information System (GIS) technology can produce information needed by analysts using the grid cell raster data (support.esri.com). The attributes presented in a raster map can provide the spatial data that would allow an analyst to determine coastline changes as a result of rising ocean levels due to changes in the climate. Therefore the spatial entity is realistically seen by the viewer as it would exist in the natural sense. According to the lecture notes, there are two advantages the raster model, simplicity and flexibility. Simplicity due to the "ease of input, code and data structure and flexibility of use of cell values of specific locations can be updated easily (Lecture 9). Therefore the spatial entity is realistically seen by the viewer as it realistically exists (planet.botany.uwc.ac.za).
Vector data models use lines that typically have a starting point, a direction, and length built from using coordinate information provided in an X,Y coordinate format. As discussed in class, "most vector analysis uses points, lines and polygons" (Lecture 7).They can use polygon features, lines, and points to represent the boundaries on maps, elevation, roads, rivers and streams, and other topographical features (GISGeography.com). Data for vector maps is more compact and since it is used in developing hard copy maps no data conversion is required but the vertex location has to be explicitly stored (planet.botany.uwc.ac.za). Vector maps provide a more accurate representation of the area, features, and structure locations such as what is typically seen in a map that uses a grid coordinate scale using latitudes and longitudes where distance and degree calculations can be performed. However, these advantages come at a high cost since the display of these features requires a scaled detailed map of high color combinations and data. As opposed to the simple image of an area that uses raster data (Bonstad 59-60). A vector data map uses spatial data in the form of these coordinates and topology as a complex data structure to achieve the graphical accuracy needed identify natural and constructed features such as buildings, roads, and water sources. Since vector data uses mathematical calculations in forming the lines and shapes that give the user a drawing view of the area, rather than an image, analysis of spatial data is more flexibly obtained except within polygons where analysis and filtering are impossible. So while vector maps provide a better visual representation of the area being viewed, it does not allow for continuous data such as elevations to be effectively represented (planet.botany.uwc.ac.za). But one benefit it does have is its scalability. We discussed in class how this feature are "independent of map projections and topological data models and therefore maintains the relationships among spatial objects" (Lecture 7). The advantages and disadvantages of each data model really comes down to which is better for the application or intended use of the data by the user. Vector data looks more detailed and accurate but with the technological advances today in Geographical Information Systems pixel resolution can be increased to match that of something that resembles vector resolution (giscommons.org).