Random Daodejing

Miscellaneous Things

Animal-Vehicle Collision Prevention w/ GIS  /  Virtual Reality Presentation Slides  /  Metropolis Ghost Town Documentary  /  Laboratory Information Management System  /  Buckminster Fuller Web Site  /  Record Database

Animal-Vehicle Collision Prevention

7 locations in Elko County where animal-vehicle collisions are more likely to occur.

full map

Deer crossing sign in Elko County.

I made a map to target areas where animal-vehicle collisions were more likely to occur in Elko County, Nevada.  The results could be used to help reduce incidents, for example the target areas may serve as locations for constructing wildlife crossing bridges.  

Data used includes defined areas from BLM, road information and traffic data from NDOT, and eco-region data from a Columbia University study.

Research notes

-  According to "Animal Crash Trends 2008-2010," deer, cattle, and horses form the three predominant animals involved in animal-vehicle collisions within Elko County.  This understanding leads me to conclude that roads can be at risk from ranches as well as from wilderness areas.

- According to a 2010 WSDOT research report, animal habitat areas are significantly associated.  Also, proximity of wildlife areas to roads in rural areas should be expected to have higher frequency rates of animal-vehicle collisions.

- According to a 2008 Federal Highway Administration report, wildlife-vehicle collisions have a higher frequency on low-volume roads.  Also, deer collision locations in particular are associated with 'edge habitat,' or transition areas.

Assumptions, criteria and project constraints

With this data in mind I am making several assumptions and proposing the following criteria to find areas:

- Within Elko County
- On Public land
- Close to low-volume traffic roads, which have a higher priority as areas of interest.
- Close to wilderness areas and particularly 'edge habitat, ' or buffer zone.
- Close to ranches. (Definition query of "ranches" within grazing allotment.  This is a big assumption.  FFRs and "Ranches" as such are clearly defined, but other possible ranches exists without "ranch" in the title.  Also, as FFRs wouldn't fall into the category of "problem areas" they were not included.  This was eventually decided in light of providing a cursory example of how this analysis could be used in a more extensive study). 


The layers and datasets were filtered using definitional queries to isolate specific attributes.  To calculate 'LOWVOL,' a created field within the Traffic layer, the following expression was used ("STANUM" LIKE '07-%' AND "LOWVOL" < 200) which was derived from taking a 5 year average sum based on annual average daily traffic (AADT) data (from 2005 to 2009) and limiting results to traffic counters within Elko.  Initially a higher threshold was used to reduce I-80 (potentially skewed due to so many counters), but even at 200 it seemed conservative.  'Horses' was filtered down to simply just horses, as burro generally occupy a significantly smaller rate of animal-vehicle collision based on research.

The big assumption mentioned above concerning ranches had to do with not being able to properly tease out all actual ranches from the Grazing Allotment Boundaries data (instead of "ALLOT_NAME" LIKE '%Ranch%' OR "ALLOT_NAME" LIKE '%ranch%'), as this denotes a Federalized jurisdiction which covers areas not strictly limited to "ranches" as such.   With further data it would be possible to perform the same analysis for property concerns (intersection between wilderness areas and ranches, for example, and further road analysis).

A 0.5 mile buffer zone around wilderness and herd (horse) areas combined with a 2 mile buffer zone for the road seemed to provide the right kind of 'edge' zone I was looking for.  The constraints can be loosened to show problem areas of secondary and tertiary importance.

Results and Conclusions

Seven locations were discovered which fit the criteria.  From this the analysis concludes that these areas seem to be more likely than in other areas to be subject to animal-vehicle collisions.  The map, though incomplete as it is given other factors and sources of data which may make for a more complete survey, nonetheless serves as a good starting point for developing a more thorough framework around analyzing not only animal-vehicle collisions, but other areas involving land use, property, and wildlife.


https://drive.google.com/file/d/1JaarYiwRr4JxqHu9sVEGdzVyLY_4UBEZ/view?usp=share_link (91.8 MB, ArcGIS)