Frac Sands Mines and Related Facilities

Northern American Frac Sand Mines

Pattern, Process, Quality, Quantity, and US Frac Sands
By Ted Auch, OH Program Coordinator, FracTracker Alliance;
Daniel Berghoff, The Ohio State University; Elliott Kurtz, Intern, FracTracker Alliance

Part I, Frac Sands Locations and Silica Geology Map Description


Click on the arrows in the upper right hand corner of the map for a fullscreen view and to access the legend.

This is a map of silica sands/frac sands mines, drying facilities, and value added facilities in North America. The map includes addresses and facility polygons. We present production for only 24 of these facilities all of which are in Wisconsin. The remaining Wisconsin and other state facilities do not have production or acreage data associated with them pursuant to a lack of disclosure requirements at the state level and USGS’s confidentiality agreement with all firms. The sandstone/silica geology polygons presented herein – in certain instances – include a breakdown of each polygon’s land cover distribution across agriculture, urban/suburban, temperate deciduous forest, and conifer forests. At the present time we only have this type of delineation for the primary frac sands producing US state, Wisconsin, along with Ohio, with Minnesota soon to arrive. The identification of each polygon’s land cover gives a sense for the types of ecosystem services present and/or threatened from a macro perspective. During our tour of select West Central Wisconsin frac sand mines it became apparent that the mining industry was essentially picking off forested “bluffs” or drumlins because these are generally the areas where frac sand deposits are deepest and closest to the surface. In return landowners are returned these parcels with less dramatic slopes making them more amenable to grazing or crop production. Consequently understanding the current land cover of each sandstone polygon will give us a sense for how much forest, grasslands, or wetlands acreage could potentially be converted to traditional agricultural usage.

Part I of this series can be found here.

Data Sources

Industry data was provided by or sourced from the following organizations, individuals, or websites:

Methodology

Land Cover Data Methodology:

State Level Primary and Secondary Silica Sand Geology – polygons extracted from USGS Mineral Resources > Online Spatial Data > Geology. Primary and secondary polygons are dissolved by Unit Age.Land cover in km2 and as a % of the entire polygon are presented using the following:

  1. “Select By Attributes” tool in ArcMAP
  2. _geol_poly_dd
  3. “ROCKTYPE1” = Primary; “ROCKTYPE1” = Secondary
  4. Using the following protocol we have begun to code each Silica Sand Geology polygon for land cover in terms of km^2 and % of polygons. The protocol fractionates polygons into forest, crop, pasture, urban, and wetlands:Used zonal statistics, which is in the spatial analyst toolbox in ArcGIS.

Here’s the basic procedure:

  1. Download national land cover dataset which can be found at: http://www.mrlc.gov/nlcd2006.php
  2. Before recoding the raster, it may be easier to manage after clipping it to a smaller extent such as the state you are interested in. Simply use Arc’s Clip tool to do this. I also found that QGIS has a fast, easy, clipping tool called Clipper. Once the raster is a bit more manageable, use the legend for the dataset that is on the above webpage to recode the raster into a set of rasters for each land cover type you’re interested in. Use Arc’s Reclassify to set all the values you want to 1 and all other values to 0. This process can also be done in QGIS which I found to be easier and faster. For QGIS, use Raster Calculator and create an expression that connects all the rasters of interest with “OR.” The syntax should be something along the lines of: ([name of raster @ band1] = first forest value) OR ([name of raster @ band1] = second forest value) and so on for all your values.
  3. Use the zonal statistics tool in Arc (Zonal Statistics as Table) to get the sum (it is important that is the sum) of the new binary raster for each polygon for each shapefile you’re using. The tool used should export a table of values.
  4. Add the table that the zonal statistics tool outputs and then join it to the shapefile you used to generate it.
  5. Repeat steps 3 and 4 for the other raster layers you generated with reclassify.
  6. Export the shapefile with the joined data.
  7. Put the shapefile back in Arc and open the attribute table.
  8. Add a new column.
  9. Use field calculator to calculate this column as 900 times the sum you got from your first zonal statistics run (because the data are in 30mX30m resolution, this will give you a good approximation of the square meters of land cover affected).
  10. Repeat steps 8 and 9 for your other zonal statistics results.
  11. Repeat step 2 for other raster classes you are interested in (developed, cultivated, wetland, etc.).
  12. Repeat steps 3-10 for the other shapefiles you are using.
2 replies

Trackbacks & Pingbacks

  1. […] The Loop reports that farm land is being bought up by fracking companies to convert into Silica Sand mining.  Not only is fracking devastating the environment and public health, but it is leaving a desert […]

Comments are closed.