Improved Sandbox Activity

Introduction:

For this revision of our first lab we expanded on our findings from exercise one. The point of this exercise was to try to see what we did well and what we did wrong, then fix any errors that we hay have made in the first exercise. As stated in the first lab write up, fieldwork is where we can collect this data to then observe and analyze later. This lab was focused specifically on the collection and surveying process that is part of any land analysis, and the addition of this data into a GIS for further analysis.

Methodology:

For the first part of the project we were tasked with creating a landscape within our sandbox (for a detailed outline of this process see Lab write-up #1). Then we collected the data and recorded it to put into the GIS when returning (again, detailed outline available in Lab write-up #1).

Since we recorded our data in a graph format of X,Y data it was necessary to convert this by hand into a table format X,Y data so that we could import it as X,Y data in ArcMap. See Figures 1 and 2.

Figure 1:  Raw data collected from first sandbox modeling attempt. Data is in 'graph' format, unusable when importing into ArcMap

Figure 2: Fixed data, now in X,Y format and will be useable for importing into ArcMap. Notice first measurements are listed under Z category, and second measurements are listed under the Nameless category (later renamed Z2).

Next we exported the data as a shapefile and then put that shapefile into our map project. We used this shapefile to create a raster image thourgh the Spatial Analyst tools of Interpolation. Through trial and error I determined that the Kriging method would be the best method for creating the raster file. See Figures 3, 4 and 5.

Figure 3: Data points overlaid on a created raster image.  Dots represent each individual point of data collection. Raster created through trial and error to determine Kriging method as best interpolation

Figure 4: Original measurements Raster file. Created through Kriging method. Notice high values in yellow/red and lower values in green and dark green. 

Figure 5: Second modified measurements Raster file. Created through Kriging method. Notice high values in yellow/red and lower values in green and dark green.

After creating the raster file for the first and second measurements we exported these rasters into ArcScene so that we could create a three dimensional image of the surface of our sandbox. By floating the layers on the Z data surface we were able to create a three dimensional model of the surfaces. See Figures 6, 7 and 8.

Figure 6: Original ArcScene three dimensional model.  Notice high values in yellow/red and lower values in green and dark green. 

Figure 7: Second modified ArcScene three dimensional model.  Notice high values in yellow/red and lower values in green and dark green. 

Figure 8: Side by Side comparison of topographical differences between orignal model and second model. Notice the differences in elevations and between the two. 

With these three dimensional models we have accurately mapped the surface we were working with, and now that this surface is in the GIS we can further analyze it for whatever purposes we deem necessary.

Discussion:

It was beneficial to go back and improve our original data as you can see in Figure 8 it clearly made a difference in our models. Working with an incomplete or inaccurate data could lead to huge mistakes, depending on what type of analysis you are taking part in. Our revised model was far more topographically diverse and I feel it best reflected the surface in real life.

As stated in the previous lab write-up, I think it was important that we used only the necessary materials to prevent any excess baggage in the field, and that we had a backup ‘dumb’ method in mind if needed.

Also similar to the first time, I think we would have benefitted from measuring at a higher accuracy, however the time and weather did not permit us to do so.

Conclusion:

This revision was valuable because it taught us a couple things. First it taught us the process of further developing our surface within a GIS. Knowing how to create a raster, and an ArcScene model will be very helpful in the future in this class, and is a vital part of learning field methods.

This exercise also taught us the importance of careful measuring and how this might change our end results. You can see in our final models (Figure 8) how they vary quite a bit.

Overall this exercise was a good expansion upon the first exercise and should prove to be a valuable learning base for future exercises.