Lab 1: Sand Box Lab

Lab 1: Sand Box Lab

Introduction

The Sand Box Lab is an exercise in improvised surveying. Each group is given a sand box, approximately 1 square meter in size, into which we sculpted the terrain. With limited equipment we then surveyed the terrain, collecting data that will later be used to create a digital elevation model. On Wednesday September 14, 2017 my group, group 1, conducted the exercise in a sand box located approximately 400ft west of Phillips Hall and 300ft south of Schneider Hall on Roosevelt Avenue, Eau Claire Wisconsin.

Methods

Study Area

The study area was a sandbox approximately 1 square meter in size located on the UW Eau Claire campus approximately 400ft west of Phillips Hall and 300ft south of Schneider Hall on Roosevelt Avenue, Eau Claire Wisconsin.

Sculpting the Terrain

The first step was to sculpt a terrain that included at least some hills, valleys, ridges, and plains. We decided to honor our professor, Joe Hupy, by sculpting his name, as seen in Figure 1.

Figure 1: Terrain sculpture of "JOE"

Sampling

Sampling is a method of gathering data from a few locations that can fairly accurately represent the larger area, since measuring elevation by had across the whole are is time consuming and impracticable. There are three sampling techniques, random sampling, systematic sampling, and stratified sampling. Our group employed a systematic sampling technique, where we created a system to decide what points to measure.

Elevation varies greatly across the middle portion of the sand box but is relatively flat along the top and bottom. Therefore we wanted to a denser sampling of the middle portion than of the top or bottom. Using string held across the sandbox by pushpins, we created a grid system, as shown in Figure 2.

Figure 2: Using a ruler and pushpins to set up a grid system

With the origin, (0,0) placed in the bottom right corner of the frame, we began to constructed an X,Y coordinate system with 5cm intervals. Due to limited amounts of string, and knowing that the sample size was going to be different for different parts of the sandbox, it is not a perfect grid. Along the X-axis, the length of the sandbox is sectioned of by 5cm intervals. Along the Y-axis, 25cm to 90cm is sectioned off by 5cm intervals. In this section we took 1 measurement within each 5cm^2 box. In the lower section, everything below Y=25cm, the grid was larger, 25cm x 25cm, with the exception of the leftmost square, which was 15cm x 25cm. One measurement was collected in each section. In the upper section, everything above, Y=90cm, grid sizes were 25cm x 20cm with the leftmost square 15cm x 20cm. See Figure 3 to see completed grid.

Figure 3: Completed Grid

Measurement Technique

The height of the string grid was our arbitrary zero. Since the strings trending North-South were placed on top of those trending East-West and exhibited significantly less sagging, their height was our standard. All measurements were collected as centimeters below zero with a ruler. In the middle section of the grid, measurements were collected in the north-eastern corner of each box. In the upper and lower sections, measurements were collected in the middle of each box. 

Figure 4: Measurement Technique. All measurements in the middle section were collected at the north-eastern corner of the boxes, as exhibited in this picture. The camera is facing west. 

Discussion

The data collected is presented in Figure 5: Data Table. In a later lab this data will be used to create a Digital Elevation Model of the terrain. At that point the data will be analyzed for its accuracy in modeling the terrain. 

Figure 5: Data Table

Error Sources

There were several potentials for human error in conducting this study. The string grid lines exhibited more sagging in the center of the sandbox than at the edges, so the elevation values may be smaller at the center data points. If pressed too hard, the ruler sometimes caused a depression in the sand, artificially increasing elevation values.

Conclusion

For now the data and collection methods seem appropriate. The data will be analyzed further in a later lab, and then I'll be able to better determine if the outcome is valid and our methods hold up.