Sand Box Digital Elevation Surface

Introduction

In this lab we created a terrain in a sand box which included a ridge, a valley, a depression, a hill, a plain, and create it with some creativity.  There are a few different ways to go about this; The first way would be to use random sampling which means to just basically pick random points in the landscape, although this would seem the least effective at creating a model for defining the landscape. Therefore, another way to do it is systematic sampling.  This means to take it in an order from the origin designated.  The way we used was Stratified sampling which means to divide of the landscape into known values.  Sampling is a very useful tool because there is not nearly enough time in the world to sample every single spot to get a value, therefore we need to make interpretations and use our knowledge of the landscape to understand the values needed to make our model of the landscape.  This labs objective is to create the landscape and use a grid system to create points to make a spread sheet to use with a computer program that will make a digital representation of our terrain that we created.  

Figure 1:  A picture showing the topography and the grid system used to take our sample points.  

Methods

The sampling techniques used for this lab was sequential sampling because it makes the most sense with the terrain created, and it seemed like it would be the fastest way to get the project done to create our model.  Another method that is similar is the random sampling method.  it is similar because it is taking fewer points than the systematic sampling, but with stratified sampling there will be more accuracy because we will be taking points that are similar and combining them together. The materials being used in this lab include a sandbox made out of wood along with a large amount of brown sand. Also being used is a meter stick, string for making sample locations into a grid making it easier to identify.  The sampling scheme was set up at 6 cm x 6 cm. This was chosen because it is small enough to get some data collected, but large enough that there is not an overwhelming amount of data being taken.  It was also set up with an origin in one of the corners and then each point correlates with 2 points like a general x-y chart.  There is also a Z value to give it a value of elevation based off of our decision of putting sea level.  The zero elevation was defined by the top of the sandbox.  The data was entered by going 1-1-Z , 2-1- Z, 3-1-Z, etc.  We choose this data entry method because it makes it easy to enter into a spreadsheet, and it is easy to read and go back to create if needed.  

Results/Discussion

The resulting number of sample points was 321.  The minimum number was -14 and the maximum number was +4.  the mean -2, and the standard deviation was +3.  The sampling we took did relate to the method that we choose, and the group stuck to the original plan to use each grid square as one point.  Therefore, we will not have any inaccuracy due to changing methods at all. Some problems that were encountered were that the grids we created were not completely tight making some slack more than others. Another problem was trying to average the terrain over a square.  We took the average of 4 points and used that number, and we tried to make the string as tight as we could.

Figure 2:  This is a grid showing the data we collected for each square 6cmX6cm.  It is actually remarkable that numbers can show valleys, hills and ridges so well. 

Conclusion 

It relates to using the sequential method because we took every grid mark as a point in an order that can easily be used to create a digital model. It also relates to the other methods because  it makes a large area and takes small points of it to make a model which has interpretation in it.  It is important to use sampling in spatial situations because there is no possible way to get every single small point in the sampling area.  It is nearly impossible to do it, but a small sample spread out throughout the area makes it easier to understand the data and has the ability to allow smart interpretation.  This activity relates to sampling larger areas because it teaches strategies to handle larger areas which include setting up a grid, taking an average value of what is being addressed in the grid and creating a model of the area.  The numbers gathered did provide an okay amount of detail and did show the high and lows within the map, but there should be some areas that need more detail.  To refine the survey it would be beneficial to make the areas with higher relief to have a smaller grid size then 6x6cm; it would possibly be better to make it 3x3 just to get mroe data in those regions.