Raster Modeling and Sand Mine Suitability

Goals and Objectives

This latest and final extension into our investigation of sand mines in Trempealeau County contains raster modeling of both risk and suitability. The extension of spatial analysis had to be activated in order to work with raster models. All of the tools that were used to perform our analysis were located in the spatial analyst toolbox. Model builder was also utilized in order to model our workflow for both presentations sake and organisation. A suitability index, risk index, and viewshed were all created to best model this region in Trempealeau County. The goal of this project was to show areas of risk where the environment or population could be affected, places of great suitability for mining and to show regions where they are overlapping to show where low risk and high suitability match.

Data

Digital Elevation Model (DEM) - USGS data
Geology - Trempealeau County Class Folder
Landcover - National Land Cover Dataset (NLCD)
Streams - Trempealeau database
Railroads - Trempealeau Data from class
Populated Areas - ESRI folder
Prime Farmland - Trempealeau database
Schools - ESRI folder
Water Table Depth - Hydroline data in WGS

Methods

The tools that were used in this study were slope, reclassify, Raster Calculator, Euclidean Distance, Viewshed, Topo to Raster and Filter. Slope is a tool used to calculate a slopes percentage from elevation data in DEM. The reclassify tool was used to change the values field using class breaks. The reclassify tool allows easier performance of the raster calculator tool. The raster calculator tool allows for the junction of several rasters. It uses mathematical operators to perform further analysis. The Euclidean Distance tool that describes each pixel cells relationship to a source pixel. The viewshed tool represents the visible area that can be seen from a polyline/point in an elevation model. The topo to raster tool allows for the creation of a raster through many factors and in our case, contour lines. The final tool used was the filter tool, which either runs a low or high pass filter on a DEM to bring out or subdue features in the image.

Figure 1: Suitability Model

 

Figure 2: Risk Model

 



Figure 3: Viewshed Model and Risk/Suitability Combination Model

 

 

Results

After each model came to a conclusion 4 maps were the ending result. My first map, shown in figure 4, was created as suitability model using several factors that can be found in figure 1 above. The two geologic layers that we used to find the most prime sand came from the Wonewoc and Jordan formations. The second layer that I used to create my suitability model was NLCD data. When I performed the reclassify on my data, I chose barren land as the most suitable, then shrubland, herbaceous, and cultivated as the next most suitable. Finally forested land was the least suitable. In the reclassify function I also discluded land features that were wetlands, water, or developed regions.

The next step that I ran was the euclidean distance function from the railroad terminals. Then I reclassified my function which listed regions that are near a railroad terminal as more suitable than that was further away. This function appeared to have more of an impact than other tools on the output as the rail terminal in the County is in the northern part of the county while all of the land in the southern portion is less suitable.

After the railroad data was finished I then ran the slope tool on the DEM. I made it so the slopes that were lower were more suitable for mining while steeper slopes were not. After I ran the slope tool, I filtered output using the focal feature from the slope tool to make the map appear visibily more appealing. Once I finished, I performed a reclass on the water table feature class. The water table feature was not a raster so I used the topo-to-raster tool on the data. After the data was in raster format I performed a reclass which made areas closer to the water table more suitable than areas further away.

Figure 4: Suitability Map
 

Figure 5: Risk Map 

 

 In the second portion of our lab I was given the task of creating a map of impact/risk in Trempealeau County. This map can viewed in Figure 5 above. There were main pieces of data used in this investigation. They are as follows: proximity to streams, proximity to schools, proximity to wildlife areas, impact on residential areas, and impact on prime farmland. In order to get all of this data into raster format I used the Euclidean distance tool. This tool allowed me to get viable distances from these features that would consist of impact or lack of impact. I then had to reclassify all of the data into a form where they could be utilized to create a model. The final step to the risk evaluation came with the raster calculator. The final product can be seen in figure 5.

Figure 6: Viewshed of Weaver Hill in Trempealeau County

Figure 6 above introduces the next section of this lab where I was tasked with using an example of a prime recreational feature in our area of interest to run a viewshed on. As see above, from atop Weaver hill one can see a vast majority of the county. Through analysis of whether or not this region would be affected by suitable land to mine, it appeared that it would indeed. Some of the most suitable land for mining was actually located just around Weaver hill. With this said, if Weaver hill really is a prime recreational area then that means that some of the most suitable land for frac sand mining in the county could potentially be unavailable to private companies.

Figure 7: Trempealeau County Risk/Impact vs. Suitability Map

Figure 7 above represents both suitable areas of mining with proportional impact in the map as well. The final output shows areas in green where it is both suitable and has a relatively low risk. On the other hand, the regions in red show very unsuitable areas with high impact on the environment and people.

Conclusion

 This lab was the final installment of a series of investigation looking at frac sand mining in western Wisconsin. My study focused on Trempealeau County and helped me to further understand both the region and tools that help me to be a better geospatial analyst. This course endowed me with great skills in network analysis, raster/spatial analysis, etc. This project was very enjoyable to work on. Sand mining continues to be a rising, and often heated, topic in Wisconsin so learning more about it has given me a better view of both sides. In conclusion, Trempealeau County is a very diverse environment that contains a lot of great features, so hopefully the population stays informed as this topic emerges further.

Sources

http://dnr.wi.gov/maps/gis/datahydro.html