Raster Analysis for Suitability Model

Introduction:

For this assignment, the Geography 337 class at UW- Eau Claire was asked to create a suitability report for potential sand mining locations in Trempealeau Country, Wisconsin. By looking at county data and rasters, both localized suitability and potential risks or hazards also associated with the area could be determined. From this analysis, a map that looks at plausible future mining sites could be established.

Methods:

To begin this assignment, first some organizational steps needed to be undertaken. This included making sure there was a home geodatabase to use as a work space, and updating the environments within ArcMap. Some of the elements updated within environments were:

• Work space
• Output geodatabase
• mapping extent
• Mask used for raster analysis
• Raster cell size.

Once these organizational tasks were completed, raster analysis could begin. As previously mentioned, there were two models that were analyzed for the assessment of possible mine locations, suitability and risk. Figure 1.1 is a data flow model used to create the suitable options needed for potential site criteria. This suitability criteria included:

• Geology (Jordan and Wonewoc Bedrock)
• Land Use (herbaceous or cultivated)
• Distance to railroads
• Slope 
• Water table criteria

To accomplish the analysis of the above criteria, a series of reclassifying was done in efforts to rank suitability within the specific data fields. Ranking was done mostly on a of 1-3 with one being the least suitable and 3 equating out to being the most suitable. Once reclassifying was complete, all the suitable rasters were overlayed together using raster calculator to create one final image of suitability.

Figure 1.1: Data flow model for potential suitability of sand mining site location.

This process was essentially repeated for the establishment of a risk assessment model. For this proceedure, figure 1.2 shows the work flow model established to get the final risk output. To assess risk, the potential impact of different local features were analyzed on a  1-3 scale where 1 was lowest risk and 3 was highest risk. These features included:

• Streams
• Prime Farmland
• Residential or populated areas
• Schools
• Outdoor recreational facilities (i.e. golf courses and state parks)

Figure 1.2: Data flow model for potential risks associated with possible sand mine locations

Results and Discussion:

In association with the data flow model connected to mine site suitability, six maps were created. Figure 1.3 is a data table that explains the criteria established for the development of the individual suitability maps.

Figure 1.3: Variables, ranking, and reasoning behind each suitability map

 From this criteria, six maps were created and can be seen in figure 1.4. As noted, the scale remains constant for each map. Again figure 1.3 explains the variables in which ranking was done, and the reasoning behind it. However, a few personalized choices should be noted for figure 1.3. Such as, land cover types were reclassified based on land that would be easy to clear, and suitable land cover then is then the bi-product of this reclassification. Also, for distance to terminals, the decision to use terminals over railways was made because, mine trucks carrying product would be traveling directly to terminal location, not to the rails.

Figure 1.4: Series of suitability maps. Top row, starting from the left: Suitable geology, land cover types, suitable land cover (in figure 1.3, listed as non-suitable land cover). Bottom row, starting from the left: distance from terminal, slope, water table elevation

Figure 1.5 is then a data table that depicts the variables, ranking and reasoning behind the layers associated with risk analysis. From this variable criteria, an additional 5 maps were made (figure 1.6). Just like the last series of maps, a few topics should be noted about their construction. The first is that choice of river systems used for the risk analysis. Primary and perennial river systems are systems that are consistently flowing through the county and thus were deemed most associated with potential risk. Also, on the school risk map, the schools found were based on a point layer source from ESRI that contained school locations within the entire United States. This data was then projected and clipped to fit the extent and coordinate system of the work space. Lastly, outdoor recreation was the personal risk factor chosen because of potential noise and aesthetics issues with mines in close proximity.   

Figure 1.5: Data table depicting associated variables, ranking and reasoning for risk analysis layers.

Figure 1.6: Map series for risk analysis. Top row, starting from left: potential impact to recreational areas, and distance of impact to streams. Bottom row, from left: impact to prime farmland, impact to regional schools, impact to residential areas

From these eleven maps, one final maps was created to locate the final sand mine suitability locations. Figure 1.7 is this map. As for the numbers associated with this data, 1 is least suitable for potential locations and 3 is the most suitable.

Figure 1.7: Final suitability map for potential sand mining sites within Trempealeau county.

Conclusion:

In conclusion, by assessing raster models for potential suitability and risk variables a final suitability map can be produced using simple reclassification, and raster calculator methods. This data can then be used as a reference source for consultation for many different stakeholders associated with future mine development.

Sources:

Land Records. Trempealeau County Land Records. Geodatabase. Retrieved May 11, 2016, from http://www.tremplocounty.com/landrecords/ 

United States Department of Transportation. Retrieved from http:www.rita.dot.gov/bts/sites/bts/sites/rita.dot.gov.bts/files/publications/nationa_transportation_atlas_database/index.html

United States Geological Survey, National Map Viewer. National Land Cover Database (NLCD) raster and DEM. Retrieved May 11, 2016, from http://nationalmap.gov/viewer.html

Wisconsin Geological and Natural History Survey (GNHS). Water table contours. GIS data. Retrieved December 6, 2015, from http://wgnhs.uwex.edu/map-data/gis-data/