Field Activity #1: Creating a Digital Elevation Surface Model

Introduction

The first adaption of a multipart course, this activity was designed to endow students with some basic skills of thinking in a geospatial reference. The task was relatively open in its instructions other than being tasked to create a landscape in a sand box. Some requirements for this landscape included having a ridge, hill, depression, valley and plain to create a more advanced model. Having only suggestions from previous students from the course via their blogs each group had to come up with their own method of measuring their landscape.

Methods

Several methods really could have been used to measure the landscapes that were created. Some groups chose to use intricate grids of yarn, however Group 3 (my group) chose a loftier goal. Each group was given a sand box with an area of 1.10 x 2.35 meters. Our group used the top of the sand box as our base layer of 0 cm. The landscape parameters were then met by piling more snow onto or removing from the initial base layer. Contrast was a goal that all of the groups attempted to meet in order to get a better digital elevation model. The requirements for the landscape again were:

-Ridge
-Hill
-Depression
-Valley
-Plain

Image 1: Bottom left: Hill. At the top of the volcano there is a Depression. A valley can be seen dipping down on the back of the volcano. A ridge is found at the back of the box rising up out of the valley. The plain spans the area at the bottom right of the screen.

As can be seen in Image 1 above the group met all of the expections for the assignment with the addition of a volcano where our depression sits at the top of as a caldera. Because of the massive spectrum that was dealt with it was difficult to record the data. In hind site the group should have decided to dig down to the dirt and use the grid method of measurement. In the next lab though the data may prove otherwise.

As a group we were very ambitious in our measuring methods as well. We used a 5 cm x 5  cm grid to get our x, y and z data. As the day went on and the cold really set in we all started to regret many decisions, but the data was fairly precise and many points were obtained. The biggest hope was that the model would be very exagerated so as to give us the best results possible for elevation.

Image 2: Measuring the Volcano. This process included the use of yard sticks to get flat edge measurements in each cell of our grid at each elevation. The yard stick in my hand (red coat) was placed at the base level while Jeremy's (green jacket) was measured up along the base stick then moved out to each grid for elevation measurements.

The next big step was to create some sort of grid to measure the elevation. We used a piece of yarn to make slight impressions every 5 cm in the snow to create a grid across the snow. As can be seen in image 2 above we used yard sticks to get the elevation (z) values from the base level (0 cm). The major issue with measuring the data was when trying to measure elevation because we would sometimes have to stick the yard stick into the snow to an estimated 0 cm point. This was rather inaccurate at times which once again pointed to the fact that we were most likely too ambitious as a group.



Image 3: Above is an image of the landscape from above once again showing how ambitious the group was.  The large mountain in the middle was created initially, then the depression (caldera) was dug out at the top. The mountain then became a volcano. Our group chose to do this so as to set ourselves apart.

We used two different methods that led to the same result in order to record the data. At first a field notebook was used to write down the x, y and z values (Image 4). These values then were transposed into Microsoft's Excel. The second method that was used included putting the data straight into an Excel spreadsheet with the use of a laptop. The laptop made recording our data much easier and efficient.

Image 4: Field Notebook containing grid data. The x and y points were initially written down as we created our grid system. Our z points (elevation) were then added as we measured each cell. We later utilized a computer to input our data, but were forced to manually add those that were already in the notebook.

Once the data collection portion of the project was finished the group compiled all of the data into one Excel spreadsheet that could be used in ArcGIS.

Looking back on how we collected our data, I think that the group would agree that the landscape should have been dug down below the 0 cm base level to get easier and better measurements for the elevation data.

Discussion

I had a great time doing this project, specifically creating the landscape, however we, as a group, had our flaws as well. I am very excited to see the models that we will create with of our data because of the great elevation differences that were created.

One of our biggest issues was being able to measure the elevation, or z values, because of the incredible height and depth of our model. I do feel though that we got some of our best measurements on the most difficult parts of the project. The grid system using yarn seemed to work best among the other groups, which would have decreased the amount of time that we spent outside as well. In the end ambition was our biggest enemy, but when all is said and done I think that that's the best kind of enemy to have.

In brighter news, I think that our use of a 5 cm grid format will give us the best data possible. While it was tedious, it should give us the best results when it comes to creating an elevation model. We all really worked well as a team too and were able to have fun during the project.

Conclusion

At the end of the day, we had a few take aways from this assignment. If this project were to replicated I would suggest using the yarn grid system, but also stay as ambitious to get the most exageration possible. Another note to be taken would be to start off taking the data with a laptop so that the group wouldn't have to sit and take over an hour to transpose the notes onto a computer. The data should create a great model since we were rather tedious with our use of measurements in small spaces.