This semester I am undergoing an internship with the Irish Military Archive (IMA) as part of the Practicum module for the MA in Digital Humanities (DH) at National University of Ireland, Maynooth (NUIM). The purpose of the internship is to gain practical experience on an active DH project.
The DH project I am engaged with involves the development of a web-based resource for making available the Aer Corps Aerial Photographs and Vertical Negative collections online. The vertical negatives are aerial photographs of Ireland taken by the Aer Corps from the 1930s to the 1980s. Due to some deterioration of the original negative films, the IMA undertook a large-scale digitisation project of the aerial images and the corresponding flight maps. The digital collection now consists of 230 flight maps and over 45,000 digital aerial images. As part of my approach in this internship, I will endeavour to develop a greater understanding of GIS and GIS data. Thus, the objective of this blog-post is to further my knowledge in GIS terminology for the purpose of understanding the relationships between GIS and the source data for the IMA project.
GIS is an acronym for the term Geographic Information System. A GIS is a computer-based data management system that facilitates the input, capture, storage, retrieval, analysis, management, manipulation, output and display of georeferenced data, otherwise known as data that relates to a location on the surface of the Earth (Scurry; National Geographic). GIS software and tools are employed in a vast array of projects and scenarios such as urban and regional planning, forest and wildlife management, and emergency response analysis. GIS organises geographic data in such a way that “a person reading a map can select data necessary for a specific project or task” (ESRI 1). Therefore, a good GIS program should be capable of processing “geographic data from a variety of sources and integrate it into a map project” (Ibid.). According to Morais, GIS data tends to be separated into two types of categories: spatially referenced data which is represented in vector and raster data forms; and attribute tables which is data in tabular format (Morais “GIS Data Explored”).
There are normally three types of vector data:
- Polygon data: used to represent boundaries on large scale maps, i.e. a city, a forest. They are 2-dimensional and so may be used “to measure the area and perimeter of a geographic feature”, and tend to be distinguished via colour schemes (Morais “GIS Data Explored”).
- Line /Arc data: used to represent linear features, i.e. rivers, roads. They are 1-dimensional with a start and end point and so only measure length, and tend to be distinguished as unbroken, thickened or dashed lines (Ibid.).
- Point data: used to represent distinct data points or nonadjacent features. They are 0-dimensional with no length or area to measure i.e. a school, a library. It might also be used for a place name of a town as an abstract point (Ibid.).
On the other hand, raster data or grid data represents land surfaces as a fourth type of feature. It is cell-based and includes satellite and aerial imagery (Morais “GIS Data Explored”; Briney; ArcGIS “What is raster data?”). When discussing imagery as a data type for GIS applications, it is inclusive of digitised vertical negatives, scanned photographs that were printed from film, or born-digital images. Digitised topographic maps are also included under the domain of raster data imagery (Briney; ArcGIS “What is raster data?”). In raster data, imagery types are often described in terms of hyperspectral, multispectral and panchromatic (Briney). For example, hyperspectral imagery is used for classifying different land types, and it is beneficial for agriculture, forestry management and monitoring the environment (Briney; Morais “Satellite imagery”). A raster is comprised of a grid of cells (or pixels) that are arranged into rows and columns, and each raster cell is assigned a value that represents information such as temperature or elevation (Briney; ArcGIS “What is raster data?”).
Typically, for use in GIS, a photograph/digital image needs to be registered to a specific place on the earth’s surface, a process known as georeferencing (Land Trust GIS). However, for such an image to present itself as an accurate representation of that surface it needs to go through a process of orthorectification (Ibid.). According to Lear, aerial photographs and maps appear to be “a natural combination to record and analyze geographical information: Maps provide geometric information and photographs add realistic, timely detail” (12). However, aerial imagery is recorded via cameras as a “flat plane . . . whereas the earth is curved and its terrain takes on many varied shapes” (Lear 12). Thus, aerial images are distorted and are rendered “invalid for mapping and geographic analysis” (Ibid.). Orthorectification is an attempt to overcome this problem. The process begins with the rectification of a scanned (rasterized) aerial photograph. Rectification “removes distortions arising from the camera lens, the aircraft’s position, and elevation and other topographical features” (Ibid.). The aerial photograph is then transformed into a high-resolution digital image “that correctly represent[s] the geometry of an area and its terrain” (Ibid.).
From this brief exploration of GIS data, it is now possible for me to ascribe the GIS term – raster data – to both the flight maps and aerial images in the IMA Vertical Negatives digital collection. From my recent experiences working with the digital collection, I am aware that the digital aerial images have not been altered or rectified since digitisation, therefore, I am now more equipped to acknowledge the following. Typically, for best practice, in order for the IMA aerial images to be presented as accurate representations of the land surface in Ireland through a GIS application, the images will need to undergo the process of orthorectification beforehand.
- ArcGIS. “What is raster data?” ArcGIS 9.2 Desktop Help, 22 Sept. 2008. Web. 24 Feb. 2015. <http://webhelp.esri.com/arcgisdesktop/9.2/index.cfm?TopicName=What_is_raster_data%3F>
- Briney, Amanada.“Imagery and its use in GIS” GIS Lounge, 05 March 2015. Web. 09 Mar. 2015. <http://www.gislounge.com/imagery-use-gis/1/>
- ESRI. GIS Best Practices: Imagery. ESRI, September 2010. Web. 09 Mar. 2015 <http://www.esri.com/library/bestpractices/imagery.pdf>
- ESRI. “What is GIS?” Web. 03 Mar. 2015 <http://www.esri.com/what-is-gis/howgisworks>
- Land Trust GIS. “Use Aerial or Satellite Imagery.” Land Trust GIS and GreenInfo Network, 2011. Web. 22 Feb. 2015. <http://landtrustgis.org/technology/advanced/imagery>
- Lear, A.C. “Digital Orthophotography: Mapping with Pictures.” IEEE Computer Graphics and Applications 17.5 (1997): 12–14. IEEE Xplore. Web. 8 Jan. 2015.
- Morais, Caitlin Dempsey. “GIS Data Explored – Vector and Raster Data.” GIS Lounge. Web. 9 Mar. 2015. <http://www.gislounge.com/geodatabases-explored-vector-and-raster-data/>.
- Morais, Caitlin Dempsey. “Satellite Imagery.” GIS Lounge 9 Mar. 2011. Web. 9 Mar. 2015. <http://www.gislounge.com/satellite-imagery/>
- National Geographic. “GIS (geographic Information System).” National Geographic – Education Encyclopedia. N.d. Web. 13 Mar. 2015. <http://education.nationalgeographic.com/education/encyclopedia/geographic-information-system-gis/>
- Scurry, J. “What is GIS?” Characterization of the Ashepoo-Combahee-Edisto (ACE) Basin, South Carolina. South Carolina Department of Natural Resources and the National Oceanic and Atmospheric Administration’s Coastal Service Center. Web. 20 Feb. 2015. <http://www.nerrs.noaa.gov/doc/siteprofile/acebasin/html/gis_data/gisint2.htm>
- NCSU Libraries. “Overview of GIS in the Libraries.” New York State University Libraries, N.d. Web. 03 Mar. 2015 <http://www.lib.ncsu.edu/gis/overview.html>