This article speaks about the importance of geographic data. Various components forming the geographic data are also described. In the third section of this article various characteristics of geographic data are listed and discussed in detail to understand how geographic data is different and what is its nature.
Data is the key component and heart of every GIS system. It plays a key role in understanding about a particular problem and arriving at a solution/decision based on the analysis of the data.
GIS analysis typically comprises of analyzing or investigating the properties of various geographic features for understanding the variation, relationship among the features. The features are what typically forms the geographic data.
In GIS these geographic features which forms the data can be represented in multiple forms. Depending on how we represent a particular data set, the type of information that we can extract and interpret will vary. Further, The kind of spatial analysis or data analysis which we intend to perform also often depends on how we represent the data in GIS.
Example: The data representing elevation can be represented as contour lines or as a continuous raster. The analysis that which we can perform on contour lines and raster will be different in GIS depending on how we work with them.
Components of Geographic Data:
One can typically define data as a set of observations in unsorted or raw form. The processed raw data (the processing is done based on a question asked) out put is called information. However, all this is well applicable in the general context.
In the context of GIS, data most often means geographic data, and it links 3 important components of data which are listed below.
- Attributes (characteristics)
Of the 3 elements listed above i.e., location is most essential and mandatory element which is required, while time and attributes are optional. Attributes helps in understanding characteristics of a data set and helps in many analysis often.
For Example, if one is investigating accident hot spots in a city, it is important to have the locations of accidents, then time of accidents and other characteristic details such as no of deaths, type of vehicles etc. In this problem of accident hot spots, though it is possible to identify some red zones only with the help of locations, having the details regarding time and other information definitely helps in arriving at a better judgement of accident hot spots in the city.
Nature of Geographic Data:
Geographic data has a typical nature when compared with conventional data sets. In conventional data sets location doesn’t play a much role, here location plays a key role. Other than that, how geographic data varies is governed by several characteristics of it. Following are some of the important characteristics of geographic data which explains its nature.
1. Geographic data varies spatially:
Geographic data contains locations of features. Based on the location, same object may have different characteristics. Some times, it is possible that a complete different data set may present at the new location. This characteristic of variation of data based on the location is called spatial variation. As nature is highly dynamic and unpredictable, spatial variation is seen very commonly in the geographic features.
For example, buildings at different locations have different characteristics such as area, height etc. Forest area at different locations have different vegetation types and density. Also at different locations different such as water bodies, buildings etc may present.
2. Geographic data represents attributes of features:
Though it is not always mandatory to have attributes in the data sets, some times along with the location geographic data presents the details of the data such as width of road, amount of traffic, no of lanes in case of data related to a road.
Temporal variation is another characteristic of geographic data. Temporal variation defines or represents variation of data or feature characteristics with time.
For example, in 1990 Hyderabad map doesn’t have outer ring road as it was not constructed. However, if we look at the map of same Hyderabad city in 2018, it shows outer ring road and several other new structures in the map. One can also notice absence of some water bodies and vegetation area in the map.
This type of variation of data set with time is called temporal variation.
4. Data sets can be discrete or continuous:
Geographic space is continuous however, the geographic data which is represented in GIS can be discrete or continuous. A data set is called discrete, if no observations are possible to make between two observations at given point of time.
For example: Objects such as lakes, buildings etc are discrete. We can clearly define the boundary of these objects for making observations.
A data set is called continuous if it is possible to make additional observations between 2 observations (at 2 different locations).
For example: Phenomenons such as temperature, humidity etc are considered as continuous. Because if we consider any 2 observations of temperature on earth surface, it is still possible to make an additional observation between those 2 locations.
5. Projected data of large areas may have distortions:
Geographic data is acquired from the earth surface which is not uniform in shape. However, for illustration and map making purpose we consider earth as sphere/ellipsoid often. All objects on earth are projected on to reference spheroid for mapping the given study area. This leads to distortion of map at different locations due to improper shape/enveloping of reference spheroid/ellipsoid. These errors include incorrect distance measurements, locations and area measurements.
Hence, in GIS when we are working on a application depending on the project nature and study area we choose the relevant coordinate system accordingly.
6. Spatial Auto correlation:
Spatial data correlation is another property of geographic data. This is the property where near by objects tend to have similar properties. For example, if we look at the contamination of water from source to a distant location. From the source, near by water area is more contaminated compared to the water at distant location. This correlation is called spatial data auto correlation. One can observe similar type of correlation in other data sets also such as elevation data over terrain, temperature data, rainfall data etc.
By making use of statistical measures available in various GIS software it is possible to numerically measure the spatial auto correlation present in a given data set.
For more information about the nature of geographic data and auto correlation concepts etc., one can refer to chapter 4 of textbook: Geographic Information Systems and Science by Paul Langley et al, by Wiley Publishers.