Sampling is a simplified method used to examine a whole population. Data is acquired for a small portion of the total population to inform on what the total population generally consists of (Royal Geographical Society).
Various Sampling Techniques:
- Random- samples selected through a completely random process
- Systematic- samples are selected in a methodical, consistent systematic way
- Stratified- samples are selected after the entire population has been divided into subgroups, and samples are selected from the subgroups
The objective of this lab was to provide students the opportunity to create and execute a plan to survey and map the surface of the earth. A wooden sandbox measuring 114 cm by 114 cm was given filled with sand. In groups of three students created a terrain within the sandbox consisting of a ridge, valley, depression, plain, and hill. This group completed this assignment on Friday, January 27th, 2017 between 11:30 am and 1:45pm.
Methods
The following materials were used to complete this lab:
- Sand
- Sandbox
- Meter stick
- Thumbtacks
- String
- Measuring tape
- Spoon
- Pencil
- Paper
- Computer
This group selected to use the systematic point sampling method of sampling. This method was selected because it provides a model with evenly distributed data that covers the sampling area consistently. Also given the materials, string and thumbtacks, a grid could be made that at intersections provide a precise location to take samples. Other methods were considered, but were not selected because clear concise sampling locations were not utilized. This seemed to allow for too much variability and selection made by the sampler. Also other methods in order to obtain complete datasets would require extensive numbers of samples. Systematic point sampling designates specific locations for samples do there is no discrepancy on the number of samples that need to be taken.
The sample plot, a 114 cm by 114 cm wooden sandbox, was located on the University of Wisconsin-Eau Claire campus. On campus it was located on the east side of the Philips Science Hall, on the opposite side of Roosevelt Avenue from the building.
First, a terrain to be mapped had to be created out of the sand of the sandbox. In a pre planning meeting a design (Figure 1) was created to incorporate all five required elements: hill, ride, depression, plain, and valley. Hands and a small spoon were used to create the design. Figure 2 shows group members creating the design in the sandbox. Figure 3 shows the finished creation of the terrain inside of the sandbox.
Figure 1. A model of the design to create in the sandbox to be mapped. Each required terrain element: depression, plain, valley, hill and ridge are labeled.
Figure 2. Group members Maxfield and Shannon work to create the terrain design in the sandbox.
Figure 3. The finished product of the terrain in the sandbox.
The sandbox was next measured to obtain its dimensions: 114 cm by 114 cm. The group then decided to insert the thumbtacks into the edge of the sandbox at 6 cm intervals around the perimeter of the sandbox. String was then threaded between thumbtacks across the sandbox. This creates a 20 box by 20 box grid out of the sandbox. Figure 4 shows a model of what the plan created in a pre planning meeting. Figure 5 shows the completed grid created outside on the real sandbox.
Figure 4. A model used to represent the dimensions of the sandbox and at what intervals the grid will fall. The 114 cm by 144 cm sandbox will be divided into a 20 by 20 box grid by threading string at 6 cm distances across the sandbox.
Figure 5. The finished product of the sandbox grid.
Each sample was taken at an intersection of two strings using a meter stick. The meter stick was placed flat against the sand at the intersection held in a vertical position by one of the group members to obtain the measurement (Figure 6). Another group member was seated a few paces away from the sandbox and read out loud the measurement from where the string crossed the meter stick. The third group member using a pencil and paper recorded each measurement. To orient the data the left bottom corner was considered the origin (0, 0). X values read to the right counting by 1 and Y values read up counting by one. Z values were considered the meter stick measurement. Pencil and paper were used to record the data because the temperate at the time of sampling data points was 25°F and if an electronic device would have been used to record data the battery would have been depleted at a rapid rate. The data was recorded using a three-column chart with three fields, X, Y, and Z. Sea level was considered where the flat plane of string fell. This means that most, in fact all, of the data points fell below sea level.
Figure 6. A meter stick properly held in the vertical position flat against the terrain at an intersection of two strings measuring a Z value.
After returning indoors a group member manually entered the data into an excel spreadsheet.
A total of 400 sample points were recorded. The minimum value is -11.0 cm and the maximum value is -0.8 cm. The mean is -5.185 cm and the median is -5.2 cm. The standard deviation of the data is 2.016 cm. The sampling directly related to the method chosen. The sampling method met the objective as desired. The group did not make changes to the sampling method over the survey, the group stuck to the original plan. This resulted in a dataset that was expected from the beginning.
A few problems were encountered while sampling. One problem was that the string began to sag as the sampling took place. The wood supports of the sandbox were wet due to moisture from the snow on the ground and this caused the wood to become softer than expected. Some of the pins started to loose their placement and caused the strings to no longer be taunt but have sag to them. To over come this half way through the survey some of the problem pins were readjusted to tighten the string. Another problem that was encountered was that the spools of string were not long enough to cover a whole side of the sandbox. To make up for this multiple spools of string were used per sandbox side.
Conclusion
The sampling used in this lab relates to the definition of sampling and is classified as a type of sampling because the whole terrain was examined in a simplified fashion while still considering the whole population. Sampling is spatial situation because it is impossible to collect a data sample from every point on the earth. Sampling methods have to be utilized to form an overall consensus on the terrain. This activity related to sampling spatial data over large areas because the sandbox represented a realistic earth terrain with common features: valley, plain, hill, ridge, and depression. The whole sandbox could not be realistically sampled for every single point, but rather a sampling method had to be utilized to simplify the sampling process. The sampling process used is a common one that is used to sample real earth terrain. The survey did perform an adequate job of sampling the area tasked to sample. One data point was sampled every six centimeters for an area of 12,996 centimeters. That means each data point represented an area of 32.49 centimeters. Those are reasonable measurements. To refine the survey more data points could be collected, for example perhaps one centimeter could be sampled every three centimeters to obtain a more accurate dataset.
Sources
Royal Geographical Society. Retrieved February 5th, 2017 from http://www.rgs.org/OurWork/Schools/Fieldwork+and+local+learning/Fieldwork+techniques/Sampling+techniques.html.
