Creation of a Digital Elevation Surface Field Activity

Introduction

          To begin this lab, it was fundamental to understand the concept of sampling. The Royal Geographical Survey defines "sampling" for its purposes as "A shortcut method for investigating a whole population" and "Data is gathered on a small part of a whole parent population or sampling frame, and used to inform what the whole picture is like." Sampling allows inferences to be made on the true nature of something based on the information gathered in regards to it. For the purpose of this lab, sampling a land area for elevation measurements, specifies sampling a spatial component. We want to get a picture of the elevation of a land area oriented in space by gathering measurements that represent the area. Sampling is done because it is not feasible to measure every aspect of a population or sampling frame (Royal Geographical Society).
          There are three main sampling techniques: random, systematic, and stratified. Random sampling is the least biased of the three, and randomness can be achieved through random number tables and generators. It involves the selection of random numbers or coordinates to measure for sampling. Subdivisions of random sampling are random point, line, and area sampling. Poor representation of the area can arise from points not being randomly selected. Systematic sampling involves collecting samples in accurate and regular distances or patterns in a sampling frame (in a spatial context). Point, line, and area sampling methods also apply. Systematic sampling is more biased and can lead to over and under representation of certain areas based on the structure. Stratified sampling is used when a sampling frame is made up of different and known proportion sub-sets, and sampling is done according to this proportion. Stratified sampling can be done randomly or systematically (Royal Geographical Society).
          The main objective of this lab was to think critically and apply improvised survey techniques to ultimately create a Digital Elevation Map of terrain we artificially created in a sandbox.

Methods

          We decided to use the systematic point sampling method because we thought that it was the most appropriate method given the materials we had: string, tacks, field notebook, meter stick, and measuring tape. We didn't want to chance having certain areas under represented with a random sampling method, and we weren't sure of the proportions to utilize a stratified method. Sampling sites uniformly with the systematic method was chosen to best represent the elevation of our terrain.
          The sandbox where we would sculpt our terrain was located on the UW - Eau Claire campus - on the east side of L.E. Phillips Hall and Roosevelt Avenue. There are four 115 cm x 115 cm sandboxes next to a small swamp. The sandbox we used was the western-most one.
          We sculpted our terrain in the sandbox, consisting of plains, ridges, hills, valleys, and depressions (Figure 1). Using string and tacks, we established a grid over the sandbox at 10 cm intervals, creating 121 10x10 cm cells (Figure 2). The 4x10 cm cell on the remaining x and y-axis were discarded to keep our samples within uniform-sized cells and x-y coordinates in integers. The plane perpendicular to the wood barrier served as sea level, easily identified in space by the strings of the grid.

                                     Figure 1: Sandbox with our terrain. Photograph facing north.

                                     Figure 2: 10x10 cell grid above the terrain. Photograph facing north.

                    The northwest corner served as our coordinate origin (0, 0). Data was recorded in a field notebook as coordinates (X, Y) and elevation (Z), starting at (0, 0) and measuring Z values along the y-axis (west to east) and measured with a meter stick (Table 1). Written data was transferred to an Excel sheet at a later date. This organization was chosen so the data could easily be imported into ArcGIS.

                                              Table 1: Page 1 of the data entry in a field notebook.

Results

• In all, 144 samples were taken.
• Min. = -14.4 cm
• Max. = -0.4 cm
• Mean = -6.19 cm
• Standard Deviation = -2.58 cm

Discussion
          Our sampling method remained true to the systematic method definition, and an adequate representation of the elevation should be rendered from such evenly distributed sample sites. Our plan didn't change throughout the lab as we thoroughly discussed our next action as we proceeded through the steps. Minor mix-ups were promptly fixed, such as grid set-up and missing a coordinate point.

Conclusion
          Like the definition of sampling, we gathered elevation for only a portion of the sampling frame to understand what the whole terrain is like. Sampling in a spatial situation can give us information about our environment that might otherwise be unknown to us without viewing a model of it. This lab represented sampling over larger area in the environment on a much smaller and simplified scale. I think our data adequately represents our terrain, but a Digital Elevation Map will show how well our data holds up. 5 cm intervals would increase our accuracy and render a more detailed map - a revision I would make if I were to repeat this activity.

Sources

Royal Geographical Society. (n.d.). Sampling Techniques. Retrieved September 18, 2017, from Royal Geographical Society: http://www.rgs.org/OurWork/Schools/Fieldwork+and+local+learning/Fieldwork+techniques/Sampling+techniques.htm