Data Collection for a DTM

Introduction

When my team was charged with the task of creating a complex terrain out of snow to be represented as a Digital Terrain Model (DTM) we first brainstormed as to how we were going to accomplish this task. The underlying goal of the assignment was to force us to be able to conceptually capture topographic and locational data by creating a grid system and applying it in the field. The features created include: cliffs, valleys, mountains and ridges.

Methodology

For this project we first sculpted the terrain out of snow.  Thankfully for us we already had an outline for the terrain thanks to the planter box which was full of snow that we used. The first step was to clear the top of the box of snow so that we had a level surface of snow which would serve as the maximum height for the terrain. Next we continued to dig out snow to create the rest of the terrain to represent the features mentioned above. After we sculpted the landscape to our satisfaction we began to lay out a grid of string over the planter box to serve as a Cartesian coordinate system (figure 1.) that we would later use to collect the elevation and location data. In figure 1. below you can see the grid layout over the model. The grid formed 420 intersections which we used for data collection.

Figure 1. 3x3 inch grid over the snow terrain model.

We used a 3inch by 3inch grid to gather an accurate representation of the terrain totaling in 420 total points recorded (30rows by 14 columns). To record the points we used a meter stick and stuck it in the box until it hit snow then we measured the depth of the snow using the intersections of the grid as the location of the measurement. Figure 2 below shows the data collection method. As the depth was measured it was entered directly into an Excel spreadsheet and normalized with an X, Y, and Z, field so that it could be displayed within ArcMap. Figure 3 below shows a screenshot of the data once entered into Excel.

Figure 2. Data collection using a meter stick measureing the depth to the nearest tenth of a centimeter.

Figure 3. A screenshot of the data collected showing the X, Y, and Z (elevation) fields.

Discussion

Communication and cooperation was key in completing the creation and collection of data for the terrain model. Using a grid with more intersections would greatly improve the quantity of the data and most likely the quality of the DTM. None-the-less our data collection was efficient with no errors in the collection and the grid was measured accurately to give us accurate X and Y fields. The only things that could have made our survey more accurate would be to increase the number of points collected as well as using a more accurate form of measuring depth. With a meter stick we measured to tenths of a centimeter but if we could get to hundredths that would increase the portrayal of the depths.

Working in the cold was almost unbearble but in figure 2 above you can see that the person (me) entering data into his laptop is wearing a face mask and wearing gloves while using a pencil to push buttons in order to enter the data without completely freezing. Working in warmer conditions would have made this collection of data a lot more bearable and probably more accurate.

Conclusions

The team worked well together to collect data but I would have liked to have met more with them after the data was collected to ensure that everyone understood the data and why it was collected how it was. We communicated well online and during the collection of data so I believe this will not be an issue. Even though we do not have a "sea-level" elevation what we have is sufficient to be worked with in any kind of geographic information systems (GIS) platform and we can always set a "sea-level" once we begin to symbolize the DTMs. Learning to create an improvised surveying technique is a very useful and interesting skill to have. I would love to perform this exercise more in the future for my own purposes but on a much larger scale.