Introduction:
Volumetric analysis allows you to find the volume of a given sample in arcmap along with pix4d. There area many different tools that can be used to help you define the volume of your sample. Volumetric analysis can be used for a variety of reasons, the reason that we are working on right now is to find the aggregate volume of dirt/ rock piles at the Litchfield mine. There are different methods when trying to calculate your volume of your sample, one method is the cut/ fill method. A cut and fill operation is a procedure in which the elevation of a landform surface is modified by the removal or addition of surface material. This tool summarizes the areas and volumes of change. To complete this tool you have to take images of the surfaces at two different times allowing you to see the change. This tool then identifies regions of surface material removal, addition, and areas that haven't changed. This tool would be very helpful when trying to calculate the volume of rock that needs to be removed from an area for a mine.
When trying to determine the volumetric analysis a few different types of data is needed. First you need images of the area. You can capture these images easily using a UAS or have the images provided for you. Along with the images it is ideal to have the GCP's of the area. Like I talked about in prior labs, GCP's allow you to tie down your points on a map giving you a more accurate reading of the surface. The reason why using a UAS is ideal is because many of these allow you to take the pictures along with having the GPS location of the pictures. If your working on a mine, many of these places have permanent GCP's which you can then tie down with the GPS locations you gained from your UAS. Using a UAS allows you to cut down on your field time along with being relatively cheaper.
Methods:
When computing volumetrics it is essential to know the methods you are working with and how they can help. One of the tools used in this lab was the raster clip tool. This tool allows an area to be extracted based on a rectangle, polygon, points, mask, or circle even. When you are trying to extract an area you are cutting through cells all the time. The tool needs to know whether the cell is inside or outside your area you are trying to incorporate. The center of the cell is the judge for this. If the center is inside the area then it is calculated in, while if it is outside it is left out.
Another important method used is the raster to tin tool. The purpose of this tool is to create a triangulated irregular network (TIN) whose surface doesn't deviate from the input raster by more then a specified z tolerance. The TIN allows you to have a 3D surface model by calculating elevations. A TIN also uses cells center points to fully cover the perimeter of the raster surface. The more cells it uses the smoother of a transition in elevation gradient it will have.
figure 1
figure 2
figure 3
Polygon volume and surface volume are essentially the same tool, however polygon volume calculates the volume and surface area between a polygon and terrain or TIN surface. Once the polygons are made, it will only be calculated for the areas of the polygons and TIN that overlap.
Methods II:
The first step to calculating volume for the Litchfield mine was to bring the imagery of the congregate piles of rocks/ dirt and the area into Pix4d. Using the volume tool in Pix4d, I was able to calculate the volume of three different mounds on the Litchfield mine. I choose three different mounds that didn't connect to other mounds but rather stood alone. This would allow for more accurate calculations. When using the calculation tool in Pix4d, all I had to do was click around the outside of the mounds. This would then allow Pix4d to calculate the volume so I could compare it to the volumetric analysis that I would later run in Arcmap. The first mound was the smallest of the three that came back with a volume reading of 21 meters cubed. Figure 4 shows my first mound I calculated.
The second mound that I calculated the volume for in pix4d was much larger, and the elevation was well more defined in the picture. It was easier tracing the outside of the mound with the volume calculation tool which gave me a more accurate reading. Figure 5 is the second mound that I calculated. The second mound had a volume of 3952 meters cubed. This is the largest of my three mounds that I calculated. In the picture you can see in comparison just how big this mound is judging by the large dump truck that is parked right next to it. When trying to make comparisons in pictures it really helps to have something you know the size of as a reference.
figure 5
The third and final mound I calculated in pix4d was roughly half the size of the mound in figure 5. Although there is no visual aid for a comparison, such as the dump truck, it is still easy to see that it falls in-between the first and second mound. Likewise, the volume also falls in-between the first and second mound. Figure 6 shows the mound that the volume was calculated for along with the volume.The volume in figure 6 is 1977 meters cubed. Along with the volume, the cut and fill are also calculated, and not just only in 3D but 2D as well.
figure 6The next tool that was used was the surface volume tool. The reference was set to below, because I took the highest point in each of the raster clips and wanted to find the volume below that. before calculating the volume it was key to use the information tool to allow me to find the elevation for each of the rasters. This information was put in as my z value when calculating the surface volume tool. The volume for the first mound raster was 179. I computed the surface volume for the second mound and got a volume of 6957. I finished computing the surface volume for the third raster mound and got a volume of 3918.
Now all this volumetrics was calculated with just one dataset for surface volume and raster clip. The tool cut fill would need more then one dataset. The cut fill tool lets one see the difference over time. For example if one was looking at the Litchfield mine and wanted to see how the humps either grew or disappeared, one would need atleast two datasets showing the difference between the two. With the cut fill tool, it could then be shown where areas of the mounds is gained or lost area. This type of tool would need temporal data, or data that shows a change over time.
The final process of this lab was to take our three raster clips and convert them into a tin. The tool raster to tin allowed me to convert my raster of the given mounds into a defined TIN. Figure 7 is the output from the tool raster to tin for mound 1.
figure 7
In figure 7 it is easy to see the different elevations for mound 1 from the TIN. It allows for a 3D representation of elevation change while being a 2D picture. After getting the TIN for mound 1 I had to run the tool 'add surface elevation'. This tool allowed me to add the z mean or average elevation for mound 1. The final step was to find the volume for figure 7. By using the polygon volume tool I was able to calculate the volume for figure 7. It allowed me to input my TIN from figure 7 and gave me an output figure 8.
figure 8
The output volume from mound 1 was 6.33 meters cubed. This is significantly different from both our raster, assuming that is in feet, and pix4d which is also in meters. This would pose a problem.
For the second mound the raster to TIN tool was also performed and gave an output of figure 9. Again it is easy to see the elevation change in a TIN from the color gradient.
figure 9
figure 10
This output gave the volume of 785 meters cubed. Once again this is significantly different compared to my raster volume of 6957 feet, and my pix4d volume of 3952 meters cubed. I have ran multiple volumetric analysis and still have not figure out why the volumes are drastically different compared to one another.
On the third and final mound, I again used the raster to TIN tool allowing for a 3D representation of that area. This is the middle mound in volume and area compared to the two others. Figure 11 is the output TIN from running the raster to tin tool.
figure 11
Figure 11 I personally think is the best to show elevation change. It is constantly on an up slope judging by the color scheme used. Upon admiring this TIN I was able to add the surface information of the z mean score allowing me to run the polygon volume tool. After tracing the surface area around the outside of the mound with the polygon tool I got an output reading figure 12.
figure 12
The final volume that was captured from the polygon volume tool for my third TIN was 515 meters cubed. Again this was different compared to my other volumes in the raster and pix4d.
After completing this lab it is easy to see that there is error in calculating volumetrics for this lab. With all the readings coming back differently, it is hard to say which one is right and which one is wrong. Using volumetrics can be an essential piece when trying to find the volumes for important surface data such as for a mine. When looking at mines in the future and ways to computing the volumes of aggregate mounds for those mines, I believe that UAS will be viewed as a step forward. However, one problem that the use of UAS could cause is for the mine workers whose jobs are on the line while competing against these aerial systems. It could cause for people to lose their jobs even. With more people getting into UAS mine companies would be smart to look at the background of the workers they higher especially if the field is taking a turn to computing volumetrics, and at a lower cost rate.
