In today’s era of computers and technology, almost everything in this world needs to go through a digital conversion process to work smoothly. The sounds and voices around us act as signals that are analog in nature. These analog signals are then converted to digital form in order to be processed in computers. The process of converting these analog signals to digital ones is known as ‘Sampling’.
GIS uses the sampling method to convert data into digital form. The purpose of GIS is to present high quality results of locations around the world. In order for this to take place, it is important that the data being input to the GIS is sampled accurately. A spatial database represents a certain part of the world with its topological features. GIS enables its user to view that part of the world in real, through that database. Therefore, everything has to be perfect, e.g. the landmarks, landscapes, topological features, weather, culture etc.
Fig. 1: Sampling via GIS
As shown in Fig. 1, the GIS collects data of the spatial features in layers. GIS uses the spatial data entered in each layer to produce a high quality image of the specific part of the world being represented. All the analog data is converted to digital form to make it user friendly.
Another example can be seen in Fig. 2 below:
Fig. 2: Layers representing spatial information
The layers named topography and hydrography, soils and wetlands, land use and roads represent a certain part of the world. The process of sampling converts the spatial information from these layers to a digital map image, enabling the user to view that particular part of the world in front of him, from anywhere in the world.
Sampling and Its Representation
The whole process of sampling requires a spatial database which includes the following:
- Digital versions of real objects: this covers the topological features, e.g. houses, roads, rivers etc.
- Digital versions of unreal objects: this covers borders or political boundaries that are invented and do not exist naturally.
Reality is represented by the constantly changing values, such as terrain changes, or as discrete objects, that remain definite, such as boundaries, roads etc.
Definite or discrete features can be represented in the form of lines, such as the roads. Whereas, the varying features such as climate, temperature or terrain can be represented in the form of pixels. The change can be seen there by varying the value of the pixel.
Layers in a GIS can be attributed using 4 measuring scales:
- Nominal: this includes names and labels
- Ordinal: this includes sequential attributes where the scale of numbers can be differentiated easily (e.g. 2 is greater than 1)
- Interval: this includes numbers whose mathematical variation matters (e.g. temperature)
- Ratio: this is where the measurements start at zero and the variation in numbers is significant
Discrete features of any landscape are often represented nominally and ordinally, e.g. houses, roads). However, features that tend to change constantly, such as temperature or elevation, use interval and ratio scales of measurement.
Sources of Data Collection
There are 2 sources of data collection – primary and secondary. Primary data collection involves data collected via field sampling or remote sensing. It involves direct measurement of the spatial database. The time span and distance at which the data is collected represent the resolution of the image. For example, data collected every hour will miss out shorter term representations.
A sample can be random (i.e. from anywhere), systematic (i.e. following a rule) or stratified (i.e. general knowledge of the researcher determines the sample data).
On the other hand, secondary data collection refers to data obtained from third party sources, such as governments, maps or other databases. However, this type of data collection often tends to be inaccurate and unreliable.
In order for GIS to fulfill its true purpose of producing high quality images based on spatial features, it is important to sample the data collected. With the help of sampling, data collected at various levels or layers, is represented in the form of pixels, dots or lines. These representations are assigned certain values that determine and calculate the changes or stagnancy of the various spatial features. Sampling combines the data and represents it in the form of a high quality image that looks like the actual picture of the part of the world being sampled.
Errors tend to decrease substantially when data is collected in digital mode. However, if the image resolution is unclear, errors tend to occur in locating wells, for example, as unclear data can be misinterpreted. The importance of sampling the world comes with the rapidly changing technology worldwide. As technology changes and grows, the need for high quality data collection in a user-friendly mode will keep increasing.