Network Analysis

INTRODUCTION

The network analysis performed builds off the previous projects on silica mines in West Central Wisconsin. Although the booming silica industry generates revenue for the company and the county, there are also damages incurred. For this project, network analysis was performed to track the mileage of transporting silica from mine to rail for all mines that don't possess an on-site rail terminal. The nearest facility function was performed for each county and calculations were made to account for total mileage and cost of damage to roads by county.

METHODS

As posted in the python script post, a series of queries were made from DNR gathered data to select only active mines with no on-site rail transportation. From there, network analysis began (As shown by Figure 1) using streets from the ESRI Geospatial database, the silica mines created from script as the facilities, and the rail terminals as the incident. Next, the results were projected into a shapefile and intersected by county. Finally, a summary was made to calculate total mileage by silica transportation per county.

Figure 1: Workflow Model Used to Create Network Analysis on Silica Transportation

Lastly, a total cost to repair road damage was calculated based off of an imagined cost of 2.2 cents/mile. Figure 2 shows these results per county and are clearly arbitrary rather than based off collected research based on the clearly understated cost.

RESULTS

The network analysis performed found the nearest rail terminal in relation to each silica mine with no on-site rail terminal and calculated the distance and time required along the most efficient route as shown by Figure 2.

As shown by Figure 2 and more accurately in Figure 3, there are many mines that require significant travel distances to get the silica to the railroads for additional transporting.

Counties such as Trempealueau, Jackson, and Clark who have multiple silica mines and limited rail terminals are looking at additional costs. The counties should also be aware because it causes the wearing down of roads and traffic congestion. This is not ideal for both the company and the county and this pattern is significant enough for consideration when planning responsible silica mines in West Central Wisconsin.

Network analysis can be used in many capacities to maximize efficiency and plan responsibly. This hard data can be utilized quickly and gives insight into costs and efficiency that otherwise would not be available. This information can be used to make more informed decisions, improve efficiency, and be used as reference in the field.

SOURCES

Data from mines and rail terminals obtained by the Wisconsin DNR
Data from streets obtained from ESRI Geodatabase

Trempealeau County Land Data

INTRODUCTION

Following the first blog post on sand mining in West-Central Wisconsin, land data was obtained from multiple sources for Trempealeau County, Wisconsin. This exercise shows the ability to collect, manage, and edit data to compile into useful maps for further analysis--specifically land suitable both economically and environmentally for sand mining in Trempealeau County. The goals of this exercise are outlined below:

1. Download and collect data from multiple sources
2. Import and manipulate data for viewing
3. Write python script to project, clip, and aggregate data into a single geodatabase
4. Compile cartographically pleasing maps
5. Write a report on the process, accuracy, and data sources

The next section will describe the process of collection, manipulation, and editing base data into a single geodatabase, followed by the resulting maps and data shown.

METHODOLOGY

Data Collection and Manipulation

Base data used and where it was derived from is listed below:

• Bureau of Transportation statistics from the NTAD
• NLCD from the USGS
• DEM from the USGS
• NRCS soil survey from the USDA
• Land records from the Trempealeau County website

All data was freely obtained by accessing each deptartment website and directly downloading from the websites or through email request. The rasters and geodatabase were then reorganized by unzipping directly to a personal working folder. Data from the NTAD and NRCS was then manipulated before python functions were performed. 

NTAD data for updated rail lines were clipped to Trempealeau county and moved into the Trempealeau geodatabase obtained from the county land records.

The NRCS soil survey is SSURGO data which required conversion from text files to a shapefile using Microsoft Access. The new shapefile along with an existing table with the original fields were then imported into the Trempealeau geodatabase. This shapefile was then put into a 1:1 relationship class with the table and joined based on this relationship. 

Python Script

Python script was then made to reproject the NLCD, DEM, and NASS rasters to a single projection and crop around Trempealeau County. For a full view of script refer to the Python Script post under Exercise 5.

RESULTS

After the python script created reprojected the rasters, a comprehensive view of Trempealeau County was then made using ArcMap shown by Figure 1. The NLCD and NASS unique classification maps were overlain by a somewhat transparent DEM to give an idea of elevation along with land cover for environmental impacts of sand mining. The separate DEM is also shown in the bottom-right corner to give a full understanding. Both NLCD and NASS include rail and road systems for the economic conditions of sand mine site location.

Figure 1: NLCD and NASS data with elevation and separate DEM and reference maps

Although data can be collected from various sources for free, it should be done so with caution. Metadata should be examined closely for individual satellite imagery and especially when working with multiple rasters and shapefiles. Credibility of work produced will be based on accuracy and precision of base imagery collected; this includes both accuracy (how close represented features match their respective features in real life or true values) and precision (to what degree or decimal place at which data was collected).
Accuracy can be measured by positional and categorical subgroups. Scale, effective resolution, the minimum mapping unit, and planimetric coordinate accuracy all contribute to positional accuracy whereas temporal accuracy measures when the imagery was taken. Attribute accuracy contributes to categorical accuracy and is how closely the description or labels match the represented feature in the real world.

Figure 2: Geospatial accuracy for all data gathered in Trempealeau County

As Figure 2 shows, much of the metadata used to assess data quality and credibility is missing for the web soil and transportation statistics so using this data would not be wise for reference material or large projects.
Attribute accuracy would preferably need to be obtained by fieldwork. However,  ancillary data with higher accuracy can be used as reference to edit and correct inaccuracies. However, this is time consuming and often times the ancillary data should just be used instead of the image with inaccuracies.
Finally, the temporal inaccuracies present a major problem in data especially land use data where rotating crops, land use, and even elevation change with the natural process of the Earth and human-environmental interactions.

RESOURCES

"BTS | National Transportation Atlas Database." BTS | National Transportation Atlas Database. Accessed March 14, 2017. https://apps.bts.gov/publications/national_transportation_atlas_database/.

"Land Records." Land Records. Accessed March 14, 2017. http://www.tremplocounty.com/tchome/landrecords/.

NRCS. "Web Soil Survey - Home." Web Soil Survey - Home. Accessed March 14, 2017. http://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm.

USGS TNM 2.0 Viewer. Accessed March 14, 2017. https://viewer.nationalmap.gov/viewer/.

"USDA:NRCS:Geospatial Data Gateway:Home." USDA:NRCS:Geospatial Data Gateway:Home. Accessed March 14, 2017. https://datagateway.nrcs.usda.gov/."USGS TNM 2.0 Viewer." 

Python Script

The following exercises demonstrate the ability to perform functions using python script and a familiarity of coding and python language fundamental to GIS applications.

EXERCISE 7

Exercise 7 prepares data for network analysis to be performed. In preparation, a series of queries and select by location was performed to select only the data needed to be used in network analysis

#-------------------------------------------------------------------------------

# Name:        Exercise 7

# Purpose:

#

# Author:      kraegebj

#

# Created:    04/23/2017

# Copyright:   (c) kraegebj 2017

# Licence:     <your licence>

#-------------------------------------------------------------------------------

import arcpy

from arcpy import env

arcpy.env.workspace ="Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_7\ex7.gdb"

arcpy.env.overwriteOutput=True

#Set up the variables

Fc1 = "all_mines"

Fc2 = "rail_terminals"

Fc3 = "rails_wtm"

Fc4 = "wi"

Fc5 = "active_mines"

Fc6 = "status_mines"

Fc7 = "mines_norail"

Fc8 = "mines_norail_final"

#Set up the field delimiters for the SQL statments

field1 = arcpy.AddFieldDelimiters (Fc1, "Site_Statu")

field2 = arcpy.AddFieldDelimiters (Fc1, "Facility_T")

#SQL statement to select active mines

activeSQLExp = field1 + "=" + "'Active'"

mineSQLExp = field2 + " LIKE " + "'%Mine%'"

norailSQLExp = " NOT " + field2 + "LIKE" + "'%Rail%'"

#Make a layer from the feature class with mine status = active

arcpy.MakeFeatureLayer_management (Fc1, "active_mines", activeSQLExp)

#Make a layer from the feature class with facility type = mine

arcpy.MakeFeatureLayer_management (Fc5, "status_mines", mineSQLExp)

#Make a layer from the feature class without rails

arcpy.MakeFeatureLayer_management (Fc6, "mines_norail", norailSQLExp)

#Select by location

arcpy.SelectLayerByLocation_management (Fc7, "INTERSECT", Fc4)

arcpy.SelectLayerByLocation_management (Fc7, "WITHIN_A_DISTANCE", Fc3, "1.5 KILOMETER", "REMOVE_FROM_SELECTION")

arcpy.CopyFeatures_management (Fc7, Fc8)

print "The script is complete"

EXERCISE 5

Exercise 5 reprojects previously gathered tifs--NLCD, DEM, and NASS from the USDA and USGS (see Exercise 5 blog)--into a single projection to be used in conjunction to give a comprehensive view of Trempealeau County, Wisconsin.

#-------------------------------------------------------------------------------
# Name:        Exercise 5
# Purpose:
#
# Author:      kraegebj
#
# Created:     08/03/2017
# Copyright:   (c) kraegebj 2017
# Licence:     <your licence>
#-------------------------------------------------------------------------------

#Import system modules and Spatial Analyst
import arcpy
from arcpy import env
from arcpy.sa import *
arcpy.CheckOutExtension("spatial")

#Set environment settings
arcpy.env.workspace = "Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_5\Working\Script"
arcpy.env.overwriteOutput = True
print "{}".format(env.workspace)

#Get a list of rasters in the workspace
listOfRasters = arcpy.ListRasters()
print "{}".format(listOfRasters)

#Loop through the rasters in the loop
for raster in listOfRasters:
    #define the outputs
    rasterOut = '{}_Out.tif'.format(raster)
    rasterExtract = '{}_Extract.tif'.format(raster)

    #Project the rasters
    arcpy.ProjectRaster_management(raster, rasterOut, "Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_5\Working\Script\TrempWebDATA.gdb\Boundaries\County_Boundary")

    #Extract the raster and copy the raster into the geodatabase
    outExtractByMask = ExtractByMask(rasterOut,'Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_5\Working\Script\TrempWebDATA.gdb\Boundaries\County_Boundary')
    outExtractByMask.save(rasterExtract)
    arcpy.RasterToGeodatabase_conversion(rasterExtract, 'Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_5\Working\Script\TrempWebDATA.gdb')
    print "Raster to Geodatabase conversion {} sucessful".format(rasterExtract)

>>>
Q:\StudentCoursework\CHupy\GEOG.337.001.2175\KRAEGEBJ\Excersices\Exercise_5\Working\Script
[u'DEM.tif', u'NASS.tif', u'NLCD2011.tif']
Raster to Geodatabase conversion DEM.tif_Extract.tif sucessful
Raster to Geodatabase conversion NASS.tif_Extract.tif sucessful
Raster to Geodatabase conversion NLCD2011.tif_Extract.tif sucessful

>>>

Frac Sand Mining in Western Wisconsin

INTRODUCTION

Frac sand mining is a relevant topic because of the revenue it can generate but is also very controversial and is carefully monitored and regulated because of the environmental and community impacts it causes. The recent boom in West-Central Wisconsin has caused a need for research for county zoning regulations and DNR health and water quality reports. GIS is a useful tool that can be used to easily evaluate frac sand site location and evaluating negative health and water impacts.

 

DEFINING FRAC SAND MINING AND SCOPE OF RESEARCH

Hydrofracking pumps a mixture of water, chemicals, and silica sand into a well, fracturing and maintaining fissures in the rock layer to extract petroleum or natural gas. The silica sand required needs to meet specific industry specifications:

• Nearly pure quartz
• Very well rounded
• Tight size gradation standards
• High compressive strength

Described by Wisconsin’s Department of Natural Resources, this type of sand is exclusively found in poorly cemented Cambrian and Ordovician sandstones and from unconsolidated alluvial sands locally derived from these sandstones. The use of hydrofracking has increased rapidly and thus too, has the demand for silica sand and has exponentially increased mining operations in Western Wisconsin where much of this sandstone is located.

In 2012, 60 silica extraction sites and 30 processing facilities were operational with an additional 20 proposed. Figure 1 shows all operational sites in 2011 and the underlying sandstone required for silica extraction. Most all locations are found in West-Central Wisconsin and export over 12 million tons per year.

Figure 2 represents the rapid growth rate silica extraction has seen just in the last five years. There was a total of 128 industrial sand facilities in which 92 were active, 32 inactive, and 4 in reclamation as of May 23, 2016 according to the Wisconsin Department of Natural Resources.

ISSUES WITH FRAC SAND MINING

With any new booming industry comes traffic in and out the facility. Just as the wind turbine industry previously had problems, the frac sand industry can lead to traffic congestion and the wearing down of roads (Hart, etc al. 2013). This traffic has also lead to noise pollution for nearby residents such as Lenny Shier, who lives near a mine in New Auburn who has said the train engine and 24/7 traffic has ruined his family’s quality of life. Additionally, a pollutant called PM 2.5 that can cause lung problems is produced by these mines that weren’t being monitored until recently. (Gloninger 2015).

Environmentally, the predominant issue is with water pollution during storm water runoff which can have a devastating effect on locals and especially farmers. From just 2011 to 2013, there were 20 notices of violations. With few resources for the DNR to closely monitor these mines’ responsibility to environmental, health, and safety concerns, local zoning boards still have the power to decide where mines are located and how close they are located to homes, farms, and schools (Gloninger 2013).

REGULATIONS AND MONITORING

“The department (of Natural Resources) is committed to working with the sand mining industry while protecting natural resources through permits, regulations and compliance. Industrial sand mines must follow the same state requirements to protect public health and the environment. This includes getting necessary air and water permits from DNR. Industrial sand operations are required to provide ambient monitoring data to the DNR on a monthly basis” (WI Dept. of Nat. Resources).

Registration for a mining permit requires a legal description of the property, follow state and local zoning laws, certification by a registered geologist that the deposit is a marketable deposit, and the approval of the county board.

GIS APPLICATIONS

Specifically in GIS II at the University of Wisconsin-Eau Claire, the use of land cover, zoning areas, soil survey, and transportation lines will be combined to produce an overview of where a sand mine might or might not be suitable to Eau Claire County.

RESOURCES

Gloninger, Chris. "Wisconsin sand mining is big business, but health effects questioned." WISN. October 15, 2016. Accessed March 03, 2017. http://www.wisn.com/article/wisconsin-sand-mining-is-big-business-but-health-effects-questioned/6325716.

Hart, V. Maria, Teresa Adams, Andrew Schwartz. “Transportation Impacts of Frac Sand Mining in the MAFC Region: Chippewa County Case Study.” University of Wisconsin-Madison Department of Civil and Environmental Engineering, 2013. Accessed March 03, 2017.

Robertson, M. James. “Frac Sand in Wisconsin.” Wisconsin Geological and Natural History Survey.
2012. Accessed March 03, 2017. http://wcwrpc.org/frac-sand-factsheet.pdf.

“Silica Sand Mining in Wisconsin.” Wisconsin Department of Natural Resources. January, 2012.
Accessed March 03, 2017. http://dnr.wi.gov/topic/Mines/documents/SilicaSandMiningFinal.pdf.

 "Wisconsin Department of Natural Resources." Industrial sand mining - Wisconsin DNR. July 5, 2016. Accessed March 03, 2017. http://dnr.wi.gov/topic/Mines/Sand.html.