Terrain Modeling

Introduction

The second of our activities was to create 3D terrains using interpolation methods. The data for this activity was the X, Y, and Z coordinates from activity 1. The five interpolation methods were IDW, Natural Neighbors, Kriging, Spline, and TIN. The idea was to determine which of these five would yield the best view for the group’s results.

Figure 1: Here are the points in ArcScene to make a 3D image

Methods

The data was entered into ArcMap and ArcScene. To create the 3D terrain the Interpolation tool had to be utilized. This is found in the ArcToolbox under Spatial Analyst then Interpolation. The terrains started as 2D models but by using base heights in ArcScene they were changed to 3D scenes. The points were placed in ArcScene to make a 3D image. These points are the points that were taken and then placed into XYZ format.

IDW

Figure 2: This is the IDW terrain. This looks to be one that should be used less than the others. IDW uses a weighted average to predict cell size.

Natural Neighbor

Figure 3: Natural Neighbor Interpolation finds the closest input samples and applies weight to them based on other cells that are proportionate.

Kriging

Figure 4: Kriging uses natural phenomena to predict the value of cells. It takes into account uncertainty. 

Spline

Figure 45 Spline uses a method that estimates values mathematically. The Spline method uses a curvature of the surface to estimate values.

TIN

Figure 6: TIN looks to be the most reliable and has a higher accuracy than most. Triangles are used to show elevation and the triangles can be placed in irregular areas.

Discussion

3D modeling of imagery is one of my favorite aspects of geography. I rather enjoyed looking at the images and analyzing them. That’s what got me through the data collection part. All in all our group worked well together. Everyone involved brought something to the table and we all agreed on almost everything. This part of the class did not really involve much group cooperation rather than talking with data.

Conclusion

This activity was rather an enjoyment and it helps with future endeavors. To have this knowledge and apply really works in the real world. The ability to run these tools is fairly easy if you enter the data correctly. For visualization purposes Kriging looks to be a better measure of accuracy. Kriging also takes into account natural phenomena in its algorithm to predict values. Using IDW and weighted average could skew areas because terrains that differ and are relatively close could skew the results. Natural Neighbor and looking at cells that have similar values could make a better view. Spline uses curvature to estimate values and by using that curvature could throw off results. A TIN is a very common DTM method because of its ability to place triangles in irregular areas that would be harder to measure.