For our 5th lab exercise, we worked with editing and analysing vector data. To do this, we designed an emergency response map book used for public notification after a hypothetical oil pipeline rupture effecting parts of Salt Lake County in Utah. To repeat, this is was only an exercise, and not an actual oil spill.
The hypothetical crisis involved a Chevron oil pipeline that ruptured and spilled 30,000 barrels of crude oil into Emigration Creek, east of Salt Lake City, Utah. The contamination and fumes were reported to be causing sickness in residents living along the 7.5 km downstream stretch of creek, but not all residents were yet aware of the situation. Our task was to design a map overlaying the area of concern on the residential property layer so that residents could be quickly notified by a response team.
The only "fake" information we were given was the location of the pipeline rupture, in Emigration Canyon east of Salt Lake City, UT. All other spatial (vector) data was downloaded from the Utah Automated Geographic Reference Center (AGRC). These data included a cadastre for property lots, and a rivers and streams layer. Some of the other information, such as the the pipeline location was given in lab. We were also given a spreadsheet of residential parcel owners and phone numbers (this may have been "fake" data as well), and some other guidelines for how to best organize the map.
After adding these various spatial data to my ArcMap project, I needed to isolate the affected portion of emigration creek, and show the potentially contaminated area due to the fumes. This required use of the selection tools to "pluck out" Emigration Creek, and then create a new Emigration Creek layer. This also involved some editing to accurately include all parts of the creek downstream from the rupture point, since the rupture point was along the middle of one polyline. I then used the buffer tool to create a contamination polygon that extended 250 meters from the creek in all directions.
The affected area of the spill contained a lot of dense information (roads, land parcels, labels), so the suggested approach was to divide the contamination area into smaller sections and split those sections into different pages of a map book. The extents of these pages were provided in lab. This necessitated a reference page, to show the how the subpages were spatially related to each other. The reference page was also a good spot to put some overview information about the incident -- I included analysis such as size of buffer area, length of affected creek, and contact information in a prominent text box:
To download a PDF of this map, click here.
I left out a the typical coordinate system data often found on the map collar to reduce "clutter", and because it wouldn't be necessary at that scale given the other contextual information. Arcmap kept overriding my data driven page order, so the sub pages were not ordered spatially - although depending on how residential tabular data is organised, avoiding overlapping, consecutive pages may be a better choice.
The smaller extent pages of the book had different symbology needs than the reference page. I used the select by location function to extract all of the parcels that overlapped the contamination buffer and created a new layer of affected parcels. I then deleted all those parcels that were federally or state owned (as instructed by the lab, and assuming government facilities would be aware of the incident). One oddly shaped parcel labeled as residential was verified to be a road easement using aerial imagery, and was also deleted.
The entire emergency response book can be seen here. Below is one example page in jpeg format.
Included in public awareness response should be a publicly available online map showing the extent of the incident. This type of map should be kept simple, in order that it is easily read by lay members of the public, at the expense of more detailed information that would be provided through direct contact with the response team:
View Emigration Creek Spill in a larger map
Included with the mapbook would be a list of residents, with contact information, who could be affected by the incident. To make this list, I used the selected cadastral layer of parcels falling within the contamination buffer. I used the Arcmap join function to cross reference that selection of parcels with a list of names and phone numbers. The common identifier between these lists was the property tax ID number. The resulting list, which was then sorted by name, can be found here. I removed some of the fields, such as lot size or length, which did not seem useful for the task. However, I left in certain information, such as parcel ID, just in case a response team member would need to contact the county recorder, or other similar "detective" work in order to contact a resident. As indicated on the reference map page, there is a lot of overlap between owners and properties, and one phone number would often cover several properties. The list does contain some addresses with no telephone number or name, so the response team would need to use the address and physically visit the property.
This lab was good exercise in the stated objectives of creating and manipulating vector data in a realistic situation. It was also a good introduction to using the Join feature, which is essential when working with large, sometimes disjointed datasets.
Had this been an actual environmental response project, I would have created more than just 4 map pages. The extents of those 4 pages did not adequately cover the detail needed if the response team was unable to contact residents by phone and needed to quickly notify residents directly, especially given the dangerous nature of the incident. These pages could also use an index coordinate system, to help reference a name to a location (since street addresses may be more difficult to locate). Having a dynamic spreadsheet file instead of only the static pdf for the list of names might also help a response team quickly cross reference addresses, names, and phone numbers. Also, it was apparent that the cadastral layer and phone list were not necessarily complete, and or possibly inaccurate. This highlights the importance of accurate record keeping, if hazards such as these are to be quickly mitigated.
ESRI, 2013. World Imagery. (http://goto.arcgisonline.com/maps/World_Imagery). Accessed 2/2014.
Utah Automated Geographic Reference Center, (http://gis.utah.gov/data/). Accessed 2/2014.
Joe Wheaton, course: WATS 6920, Utah State University. Accessed 2/2014.
Lab Pages >