The goal of this assignment was to find a suitable location for a frac sand mine in Trempealeau County Wisconsin. To do this many raster tools and data sets were used and manipulated. Two models were created, one finding suitable locations and the other finding areas with the lowest impact. These two models were combined to find the best location for a frac sand mine.
Methods:
Suitability Model
In order to find the most suitable location for a new land mine in Trempealeau County, soil geology, landcover, distance to rail, slope suitability, and water table elevation were taken into account. These layers were created into rasters by specific criteria and ranked accordingly. Areas that were more suitable were given the rank of 3, and the least suitable areas were given the rank of 1. This was done for all the layers then the layers were added up using a raster calculator to find the most suitable location.
 |
| Figure 1 |
The first step to finding a suitable location for a sand mine was to find a location with the right geologic layer. To do this a Trempealeau County geology layer was used from the Trempealeau County Geodatabase downloaded from their website. The geology layer was converted to raster based on the geologic unit field and was then reclassified. Since the Jordan and Wonewoc formations are most suitable for sand mining, those formations were grouped together and there rest were put in a different group. This allowed for a map (Suitable Geology: figure 1) to be made with suitable geology and not suitable.
A sand mine located in the middle of a heavily wooded forest would be more expensive to log out, in order to find land with the proper landcover, landcover data from the USGS was reclassified and ranked to suitable and not suitable. Specific landcover classifications are located in figure 3 below. The suitable landcover map can be viewed in figure 1.
It is important for a sand mine to be located close to a railway. This allows for sand to be transported to the rail at a cheaper cost. The Distance to Rail map in figure 1 shows the areas located closest to a rail terminal. The areas closest to rail terminals were given the highest rank
The Slope map in figure shows the areas that have the best slope for sand mining. A gradual slope or no slope at all is ideal for a sand mining operation. The areas with a gradual slope were ranked the highest
The fifth map shows Water Table elevation in Trempealeau county. It is important for a sand mine to have access to water for the mining process. Higher water table elevation was given a higher rank because it allows for easier access to water.
The final map shows the Calculated Suitability of the best location for a sand mine. This model added the ranks of all the other maps together and reclassified them into 5 categories. Areas that had the highest number were the most suitable and areas with the lowest number were label the least suitable. This is demonstrated by the model below.
 |
| Figure 2 |
The figure below is a chart of how each of the layers were categorized and ranked.
 |
| Figure 3 |
Just because the land found in the model above was the most suitable for a sand mine, does not mean it's the best spot. In order to find the best location, an impact model was created to find areas where a sand mine would have the least impact on the people and environment. This was done by using distance to streams, distance to farmland, distance to schools, distance to residential area, and distance to wildlife areas. These were categorized and given ranks of 1-3 with 1 having the least impact and 3 having the most impact. These were then all added together to create a map of the areas that had the least and most impact.
 |
| Figure 4 |
 |
| Figure 5 |
 |
| Figure 6 |
Each of the maps in figure 6 were created by running the euclidean distance tool on their shape file then reclassifying and ranking them based on the table in figure 5 above. These were then added together using raster calculator to come up with the calculated impact map that can be seen in the lower right hand corner of figure 6. In this model the areas with lower numbers from the raster calculator were shown as low impact and the high numbers were high impact.
In order to find the best possible location for a sand mine, the Impact model and the Suitability model were combined to get the best areas. This was done by using the raster calculator on the results of both of the models. To get the best areas, the impact model was subtracted from the suitability model. This was because the suitability model had high numbers for the best areas and the impact model had low numbers for the best area. When the impact model was subtracted from the suitability model, the higher numbers were the best areas and the lower numbers were the worse areas. Figure 7 below shows the model that was used to combine both models. Figure 8 below that is the final suitability map.
 |
| Figure 7 |
 |
| Figure 8 |
Viewshed Model
Trempealeau County is known for its beauty and nature. One of the many outdoor recreation activities they help provide access to is bike trails. They have a large number of paved and unpaved trails that go all throughout the county. It would be a shame for a sand mine to be located within sight of one of these trails. ArcMap provides a tool that can show all the land that is visible from a from a specific spot. The Viewshed tool was used to find all the land that was visible from the Beauty and Diversity Abound trail on the eastern portion of the county. This allows the user to find areas that the sand mine would be visible from.
 |
| Figure 9 |
Discussion:
Without the use of raster analysis none of this high quality information for the location of a new sand mine in Trempealeau County could have been created. Raster analysis is a powerful tool that when used properly can help in almost any situation. Only five variables were used in both models. If doing this for a real life situation more variables could have been used and they could have been categorized and re-classed into more classes. This would have resulted in a map that was more specific. The viewshed map was not even utilized into the analysis. Viewsheds could have been created for a slew of other variables and those results could have been created into a variable to go into the raster calculation.
I found the process of finding distances from specific inputs then classifying those and ranking them an efficient way to perform raster analysis in this situation. It ended up being a more streamlined process than using shapefiles and data in vector format.
Overall the final risk and suitability map effectively displayed what it was trying to display. It would be a great map to use for people looking to for a new spot to put a sand mine. Raster modeling and the process in this lab could be used for many situations and shows the power of GIS and the importance it has and will have in the world of business.
No comments:
Post a Comment