Introduction:
For this week's project, the Geography 336 class at the UW of Eau Claire (UWEC) was asked to revisit the topographic landscape that was constructed previously on January 28th, 2016. The purpose of this week's project was to recreate the "snow-scape" from the x, y, and z coordinates collected through the use of ArcMap, and then to resurvey the landscape from last week's project in order to get more points of elevation in designated areas. By doing this, a more accurate depiction of said landscape can be created.
Methods:
The first step in reconstructing the surveyed terrain was to import the previously constructed excel spreadsheet, that contained the recorded x, y, and z values into ArcMap. This was done by clicking on the "file" tab, then add data, and finally clicking on the "add xy data" option. This gave a simple grid of points that contained the x,y,z data collected. Once this was completed, the data points were interpolated. Data interpolation is used to predict unknown geographic values, such as elevation, by giving a specific value to the cells within a raster. This then creates terrain that depicts elevation out of the previously flat point-grid structure. (Essentially, interpolation adds the z value to the x,y data points.)
For this project, the class was asked to interpolate the data using five separate methods:
1. IDW: This method of interpolation creates cell values by by averaging the sample data values of the neighboring raster cells. Thus figure 2.1, is a depiction of this interpolation method for the surveyed data points.
2. Natural Neighbors: In order for this interpolation method to work, natural neighbors locates the closest subset of input samples applies weight to proportionate areas of value. Thus, figure 2.2 shows this method.
3. Kriging: In this form of interpolation, points are estimated to create a surface. This is done by the assumption that distance between sample points reflect a spatial correlation that explains variations in the surface. Figure 2.3 is an example of the Kriging method.
4. Spline: This method works by estimating the points different points in a way that minimizes surface curvature. Thus the resulting surface passes through each point of elevation, creating a smooth appearance without distorting height too much. Figure 2.4 is an example of such.
5. TIN: Finally, this vector-based method functions by modifying the input to reclassify the images edges into a triangular network. The resulting surface becomes very angular in appearence. Figure 1.5 show how this method looks in response to the surveyed data points. While this image is the best in in dealing with the actual hieght differences, the surface remains very angular in appearance, thus slightly distorting the landscape.
Results and Discussion:
The results portion of this report is split up into two parts. Part one is meant as a discussion for the initial maps as they relate to the surveyed landscape. Whereas part two is a discussion of the the landscape after a second survey was conducted on the same landscape.
Part One:
**To better organize this portion of results, discussion for each interpolation method will be numbered as they appear in the methods portion of this report
1. Although figure 2.1 was a good depiction of the relative heights gathered, the image appears choppy, with no sense of flow from one point of elevation to the other. For this reason, this image does not give the best depiction of the original surveyed area.
2. Compared with the originally surveyed terrain, figure 2.2 is a relatively close depiction. In that it appears smoother in general, yet relative depth and height remain in good proportion. However, points that are especially higher or lower than the surrounded points of elevation appear pointed.
3. When figure 2.3 is compared with the original surveyed landscape, the smoothness of the image is more comparable. However, where this image lacks is in elevation and depth, as this method makes any depression look less noticeable. For this reason, figure 2.3 was not a good depiction for elevation related details.
4. Of all the interpolation methods used, figure 2.4 is the best example of the landscape surveyed. As previously stated in the methods section, the image has a smooth appearance, without taking away from the initial elevation heights.
5. While figure 2.5 is the best in in dealing with the actual height differences, the surface remains very angular in appearance, thus slightly distorting the landscape. For this reason, it was not chosen as the best example of the original terrain.
Due to the overall lack of exact representation, it was decided to resurvey the original terrain. The previously used grid method remained the same, however the new grid system was done at half of the original scale (the new grid cells measured 2.5cm²) and only on areas of interest (i.e. land features of variable elevation, such as the hill, valley bottom, the top of the ridge and the bottom of the depression).
After the second survey, the same methods for creating a 3D image was used on the new surveyed points. Figure 2.6 is the final image created using the spline interpolation method on the 2.5cm² grid pattern.
Conclusion:
In conclusion, interpolation becomes a major source in visualizing the differences in point data. Although interpolation is often used to depict elevation (as was done in this project), it can also be used to help show different levels of rainfall of noise variation, from location to location. By conducting this project, a better sense of geographic thinking and project execution was developed and maintained.
For this week's project, the Geography 336 class at the UW of Eau Claire (UWEC) was asked to revisit the topographic landscape that was constructed previously on January 28th, 2016. The purpose of this week's project was to recreate the "snow-scape" from the x, y, and z coordinates collected through the use of ArcMap, and then to resurvey the landscape from last week's project in order to get more points of elevation in designated areas. By doing this, a more accurate depiction of said landscape can be created.
Methods:
The first step in reconstructing the surveyed terrain was to import the previously constructed excel spreadsheet, that contained the recorded x, y, and z values into ArcMap. This was done by clicking on the "file" tab, then add data, and finally clicking on the "add xy data" option. This gave a simple grid of points that contained the x,y,z data collected. Once this was completed, the data points were interpolated. Data interpolation is used to predict unknown geographic values, such as elevation, by giving a specific value to the cells within a raster. This then creates terrain that depicts elevation out of the previously flat point-grid structure. (Essentially, interpolation adds the z value to the x,y data points.)
For this project, the class was asked to interpolate the data using five separate methods:
- Inverse Distance Weighted (IDW)
- Natural Neighbors
- Kriging
- Spline
- Triangular Irregular Network (TIN)
1. IDW: This method of interpolation creates cell values by by averaging the sample data values of the neighboring raster cells. Thus figure 2.1, is a depiction of this interpolation method for the surveyed data points.
 |
| Figure 2.1: IDW interpolation of surveyed points. |
2. Natural Neighbors: In order for this interpolation method to work, natural neighbors locates the closest subset of input samples applies weight to proportionate areas of value. Thus, figure 2.2 shows this method.
 |
| Figure 2.2: Natural Neighbors interpolation method of surveyed points. |
3. Kriging: In this form of interpolation, points are estimated to create a surface. This is done by the assumption that distance between sample points reflect a spatial correlation that explains variations in the surface. Figure 2.3 is an example of the Kriging method.
 |
| Figure 2.3: Kriging interpolation of surveyed points. |
4. Spline: This method works by estimating the points different points in a way that minimizes surface curvature. Thus the resulting surface passes through each point of elevation, creating a smooth appearance without distorting height too much. Figure 2.4 is an example of such.
 |
| Figure 2.4: Spline Interpolation for surveyed points |
5. TIN: Finally, this vector-based method functions by modifying the input to reclassify the images edges into a triangular network. The resulting surface becomes very angular in appearence. Figure 1.5 show how this method looks in response to the surveyed data points. While this image is the best in in dealing with the actual hieght differences, the surface remains very angular in appearance, thus slightly distorting the landscape.
 |
| Figure 2.5: TIN interpolation of surveyed data points. |
Results and Discussion:
The results portion of this report is split up into two parts. Part one is meant as a discussion for the initial maps as they relate to the surveyed landscape. Whereas part two is a discussion of the the landscape after a second survey was conducted on the same landscape.
Part One:
**To better organize this portion of results, discussion for each interpolation method will be numbered as they appear in the methods portion of this report
1. Although figure 2.1 was a good depiction of the relative heights gathered, the image appears choppy, with no sense of flow from one point of elevation to the other. For this reason, this image does not give the best depiction of the original surveyed area.
2. Compared with the originally surveyed terrain, figure 2.2 is a relatively close depiction. In that it appears smoother in general, yet relative depth and height remain in good proportion. However, points that are especially higher or lower than the surrounded points of elevation appear pointed.
3. When figure 2.3 is compared with the original surveyed landscape, the smoothness of the image is more comparable. However, where this image lacks is in elevation and depth, as this method makes any depression look less noticeable. For this reason, figure 2.3 was not a good depiction for elevation related details.
4. Of all the interpolation methods used, figure 2.4 is the best example of the landscape surveyed. As previously stated in the methods section, the image has a smooth appearance, without taking away from the initial elevation heights.
5. While figure 2.5 is the best in in dealing with the actual height differences, the surface remains very angular in appearance, thus slightly distorting the landscape. For this reason, it was not chosen as the best example of the original terrain.
Due to the overall lack of exact representation, it was decided to resurvey the original terrain. The previously used grid method remained the same, however the new grid system was done at half of the original scale (the new grid cells measured 2.5cm²) and only on areas of interest (i.e. land features of variable elevation, such as the hill, valley bottom, the top of the ridge and the bottom of the depression).
After the second survey, the same methods for creating a 3D image was used on the new surveyed points. Figure 2.6 is the final image created using the spline interpolation method on the 2.5cm² grid pattern.
 |
| Figure 2.6: final image of surveyed landscape using spline interpolation method.
This image then become the best example of the original land that was surveyed. This is due again to the smooth change in elevation. However because more points of elevation were added, the landscape features show a better depiction of the gradual changes in elevation.
|
Conclusion:
In conclusion, interpolation becomes a major source in visualizing the differences in point data. Although interpolation is often used to depict elevation (as was done in this project), it can also be used to help show different levels of rainfall of noise variation, from location to location. By conducting this project, a better sense of geographic thinking and project execution was developed and maintained.
Comments
Post a Comment