Engaging In Raster Data Analysis Using UAS Data

In a previous post we worked through some ESRI tutorials regarding raster data analysis. In this post, we will be engaging in raster data analysis using actual UAS data gathered over the course of a mine dredging operation. The data is called "Litchfielddredge" and was provided to us by our professor. The data was collected on three different dates and we were tasked with performing an aspect, slope, hillshade, and elevation analysis for each one and compare them. Before we get into the analysis, lets look at the two tools we are going to be using.

Extract by Mask (A raster clip within Geospatial Analysis):

• The extraction tools allow you to extract a subset of cells from a raster by either the cells attributes or their spatial location. Extracting cells by specific locations required that you identify those locations through cell identified using a mask raster.

Resample

• The resample function changes the raster pixel size. Reprojecting an image to different coordinate systems creates an image pixel grid on an alignment that is different from the original image.

The dataset we are using is from a real world application/consulting project. We want to work with the UAS data and engage in some raster analysis. The data collection began the 4th of July to where the mine operation tells of the dredging to take place, and for us to establish a base dataset to know how much material was added. They are also concerned about slope failure and low lying flat areas where water will collect. Because the area flown is pretty large, we will want to clip out the area to work with.

The data that we received was:

• dsm_20170704  
• ortho_20170704  
• dsm_20170722  
• ortho_20170722  
• dsm_20170827 
• ortho_20170827  
• Dredgepile_clip

The table below provides the cell size, the projection, the elevation, and the linear units

DSM	Cell Size	Projection	Elevation	Linear Unit
Dsm_20170827	0.01937	WGS 1984	204.62-286.742	Meters
Dsm_20170722	0.01996	WGS 1984	219.232-261.071	Meters
Dsm_20170704	0.02002	WGS 1984	225.044-254.628	Meters

To do our analysis, we will want to work with just the dredge pile of the operation. We will engage in an extract by mask. We will start with the dsm_20170704 raster layer. 

• In the Geoprocessing pane, search for "Extract By Mask"
• Use the following parameters
• Input raster: dsm_20170704
• Input raster or feature mask data: Dredgepile_clip
• Output raster: dsm_20170704_clipped

The lowest elevation is still the same as before but the highest elevation has changed from 254.628 to 244.559 and that's because we cut out the areas we didn’t need so we can focus on the certain area we want. Looking at the table above, our cell size is pretty small and we really don’t need a cell size this small to engage in what we are doing, so we are going to resample. When it comes to resampling, there are four different types:

• Nearest: Performs a neighborhood assignment and is the fastest of the interpolation methods. Primary used for discrete data such as land-use classification
• Majority: Performs majority algorithm and determine the new value of the cell based on the most popular value in the filter window. Majority tends to be smoother than nearest
• Bilinear: Performs a bilinear interpolation and determine the new value of the cells based on a weighted distance average of the four nearest input cell centers
• Cubic: Performs a cubic convolution and determines the new value of cells based on fitting a smooth curve through the 16 nearest input cell centers. The output from cubic convolution is geometrically less distorted than the raster achieved by running the nearest neighbor resampling algorithm

Out of these techniques, we are going to use Bilinear because it creates a smoother looking surface that we can get from using the nearest neighbor. Now were going to run the resample tool with the following parameters:

• Input raster: dsm_2017_clipped
• X and Y cell size 10cm
• Resample technique: bilinear
• Output name: dsm_20170704_clipped_10cm

Now do the exact same operation for the other two DSM datasets. Perform a raster extract and resample for: 

• Dsm_20170722
• Dsm_20170812

Now we can generate a hillshade layer for each of those datasets. I talk more about hillshade's in a previous post that you can find here. As a reminder, to generate a hillshade:

• From the Analysis tab, in the Raster group, click Raster Functions 
• From the Raster Functions pane, expand the Surface group, and then select Hillshade 
• Under Hillshade Properties, for Raster, choose your raster
• Set the Altitude to 40 degrees above the horizon
• If necessary, for Z Factor, type 1
• Click Create New Layer

The next type of operation we will want to perform is a slope analysis. I talk more about slope analysis in a previous post that you can find here. Slope is one of the easiest operations in ArcGIS to perform.

• From the Geoprocessing pane, search for "Slope"
• Click Slope (Spatial Analyst Tools) to open the tool, and then set the following parameters:
• Input Raster: whatever layer you're working with first
• Output Raster: ... _slope

    The Output measurement parameter has two options. The Degree option produces slope values in degrees, while the Percent Rise option produces slope values as percentages. We are going to use percent rise. The horizontal units are feet, and the vertical units are meters.

• Click Run

Next, lets engage in an Aspect analysis. Aspect is a measure of down-slope direction, I talk more about Aspect in a previous post that you can find here. To run an aspect analysis:

• Search for "Aspect"
• Open Aspect (Spatial Analyst Tools) and set the following parameters:
• Input Raster: whatever layer you're working with first
• Output Raster: ... _aspect
• Click Run

After working with all the different types of analysis, we can run them for each different date that the data was collected at the dredge mining operation to see the difference and change over time. The results are as followed:

Figure 112: Litchfield Dredge 7/4/17

Figure 113: Litchfield Dredge 7/22/17

 Figure 114: Litchfield Dredge 8/27/17