Using GIS to Track the Dispersal and Spread of the Emerald Ash Borer

Invasive species infestations are a growing environmental economical problem within many regions of the United States. Either introduced intentionally or accidentally, invasive species are capable of firmly establishing populations in ecosystems where they have little or no predators or resource competition, quickly pushing out native species and permanently altering fragile habitats. The emerald ash borer is native to eastern Russia and northern China, Korea, and Japan and can be identified by their small (about 1/2 inch) bullet-shaped bodies and metallic green exoskeletons. As their name implies, these insects attack several species of ash trees, especially, green, white, and black ash, by laying eggs in between the layers of bark. The eggs hatch in about a week and the larvae bore farther into the tree, feeding on the tree's essential inner tissues. Hundreds of emerald ash borer larvae can live in and feed on one ash tree eventually killing in 2-4 years it by cutting off the supply of water and nutrients transported through the tissues. The emerald ash borer was first discovered in North America in 2002 in southeast Michigan. While the exact method of introduction is unknown, experts believe it was introduced through infested ash wood that was used for packing large products or stabilizing cargo ships. Since it's initial discovery, the emerald ash borer has spread to thirteen other states mostly like due to the transportation of infested firewood, ash products, and nursery stocks and has become a major problem in urban areas.

Emerald Ash Borer in Chicago and Cook County, Illinois

Cook County, Illinois is one region that has seen an influx of emerald ash borer infestations since it was first confirmed in the county in July 2006 in Wilmette. Cook County encompasses almost all of Chicago and so far there have been approximately 65 confirmed emerald ash borer sites in the county. Chicago itself is estimated to have over 600,000 ash trees (14.2 percent of all trees) and suburban Cook County (defined as all of the county outside of Chicago) has approximately 3,180,000 ash trees (10 percent of all trees). In both Chicago and suburban Cook County, ash trees are the second most common tree behind Cottonwood and Buckthorn respectively (Nowak). Additionally, Chicago has an extremely high ash tree density with ash trees comprising about 20 percent of all tree canopy cover in the city (Kovacs et. al, 2009). Although several ash trees species are native to Cook County, local governments in both the city and suburbs, as well private landowners, originally chose to plant ash trees on available property. Ash trees are relatively fast growing and hardy with sturdy and dense wood. They are ideal urban trees (especially in Chicago) because they offer large amounts of shade, can easily withstand hot summers and cold, windy winters, and are considered very picturesque in spring, summer, and fall (Empire National Nursery, LLC, 2010).

Trees in general in urban and suburban areas have numerous economical, social, and environmental benefits. They reduce air pollution and soil erosion, save energy by helping to shade buildings and homes, and can help reduce noise pollution. Additionally, trees help boost and maintain local economies; businesses on tree-lined streets attract more customers and both residential and commercial properties with trees have a higher value and sell more quickly. Because ash trees make up such a large percentage of the total number of trees in both suburban Cook County and the city of Chicago, they are undoubtedly an essential part of the county's forestry. In terms of suburban and urban forestry, the Forest Preserve District of Cook County established and currently maintains large amounts of land designated as forest preserves. While very little of forest preserve land falls within Chicago city limits, almost all of it is extensively scattered throughout the suburban areas. Over 50 percent of forest preserve land cover is trees (Nowak) and most are categorized as either flatwoods or floodplain forests, both of which are marked with prominent ash populations. Because of this, many Cook County forest preserves are at high risk for emerald ash borer infestations that could easily spread from private residential and street trees to ash populations in forest preserves. Once emerald ash borers infest one or a few trees in a forest preserve it can quickly spread to other ash trees due to high tree densities.

Previous GIS Applications for the Emerald Ash Borer Problem

GIS has been used before to map the establishment and spread of the emerald ash borer in Ohio, Pennsylvania, West Virginia, Maryland, Delaware, and New Jersey. Ayersman et. al (2009) used ArcGIS's Kernel density calculations feature to look at the density of campgrounds, nurseries, and sawmills (all major sources of emerald ash borer spread) in all of the states. Kernel density looks at points in proximity from one another. Points that are closer are represented by darker colors (higher density) and as distance increases the color gets lighter (lower density). First the study looked at each density as a separate layer and then used an additive model to combine all three of the layers with then created a collective kernel density of all data points. The study concluded that the areas of highest density are the areas with the highest risk of emerald ash borer spread and that the U.S. Forest Service and other agencies could use similar mapping techniques to determine which areas are higher priority for resources and monitoring to reduce spread.

Muirhead et. al (2006) also used GIS to model local and long-distance dispersal of the emerald ash borer. The study used three different models with one modeling local dispersal and two modeling long-distance dispersal. The local dispersal model mapped the epicenter of an infestation in a township/municipality and then mapped centroids of infestation and non-infestation sites for both 2002 and 2003. By comparing the advances in distance, they were able to see the spread over a year in terms of distance from the initial epicenter. The second model looked at human population density as a representation of human activity and thus the transferring of ash products from infested to non-infested areas. It mapped human population density at the subcounty level over several states and then looked at how infestations transferred from one subcounty to another with higher density areas seeing more infestations than lower density areas. The third model was a gravity model that looked at the quantity human-dispersed firewood. Gravity models analyzes the population at point A and the population between point B and the distance between them. In all three examples they were able to show how these techniques could show local and long-distance spread and concluded by suggesting government agencies could use these models to reduce the risk of spread on two different levels.

Using the two studies mentioned above that apply GIS to the emerald ash borer problem, I wanted to conduct a similar analysis using GIS to study high risk areas on a county level. Since ash trees are such important tree species in Cook County and the area has seen a huge influx in positive emerald ash borer identifications, GIS mapping can be applied to aid in the collaboration between local municipalities and the Forest Preserve District of Cook County's Department of Resource Management to assess and determine the most high-risk forest preserves.

Constructing a GIS Model

I originally wanted to look at a model similar to the second model in Muirhead et. al (2006), specifically population density and how it overlapped with the number of positive sites. I spent around an hour looking for a file containing detailed population density for the county but found nothing. I decided to omit this idea and focus on the proximity of positive sites to forest preserves. In order to create a GIS model that showed high-risk areas, I first needed to create a basemap that included the county lines. I obtained the shapefiles for the county from the U.S. Census website and added it to a new mapset in Quantum GIS. This vector layer shows the basic shape of Cook County, Illinois and also includes a rectangular portion of Lake Michigan off the eastern border. Now that I had the basic county outline, I wanted to add more features to show further detail. Next, I added a file that contained outlines of the suburban townships to add local government borders since this is how Cook County divides these areas. Then, I added a shapefile that included the outline of the city of Chicago, which is highlighted in purple. I believed this was a useful vector layer to have because Chicago is a prominent feature of Cook County and can make the map easier to read for those not familiar with the project or the area and it is helpful when understanding the suburban Cook County versus Chicago distinction made in Nowak's study of Cook County's forestry structure and composition. Next, I wanted to add additional vector layers that showed the landscape features of the county. First, I added a shapefile containing hydrographical features, including rivers and lakes, in blue. I then thought it would be useful to add a vector layer that contained all of the roads and streets within the county. The U.S. Census has an "All Lines" shapefile that contains all roadways plus prominent features. Unfortunately, this vector layer proved to be hard to work with since it was so detailed and could only be made out when zoomed in to a half-mile square area. I kept the All Lines vector layer until I was almost done with the map. The next file I needed to add was the one that contained all the points where emerald ash borer was confirmed. This was ultimately the most difficult part of creating my GIS model. I was completely unable to find a shapefile for this data even after searching at government and private organization websites. The only information I was able to find was from the Illinois Department of Agriculture, which gave the county, city, address, and confirmation date of the emerald ash borer positive sites. The data contained information for the entire state, so I was able to extract the points only in Cook County. I created my own excel file for the data but problems began to arise when I looked through the actual addresses. For a large portion of the site data the listed address was only a street. While for some locations it wasn't a huge problem since the street was only three blocks long, it was a major problem for longer streets. For example, the address for the Chicago point is listed as State Street even though the street runs for about 13 miles. I desperately wanted to be able to plot this data on my map, so I decided to do a lot more research. For addresses only listed as a single street, as opposed to an intersection, I searched through local press releases and news articles, most of which usually contained an intersection.

The Illinois Department of Agriculture's data sometimes had discrepancies included misspelled street names, incorrect street suffixes, and locations labeled to be in Cook County when actually they lie outside county limits. Eventually I was able to create my own excel file with the city name, intersection address, zip code, and confirmed date. I wanted to add a delimited text file, but needed to convert my location points into coordinates so that all the points could be accurately plotted at the same time. I used a geocoder website called GPS Visualizer and was able to copy and paste all my data into their tool and convert it to a txt file. From there I copied that information into a gedit txt file, saved it, and then opened a delimited text file which plotted all the points (in red) perfectly on my GIS Model. I could zoom in on a point and using the identify feature tool on the point I could see the data I originally entered in the excel file. I verified the locations of several points by selecting the All Lines layer and then clicking on the streets where the point was located. After confirming the locations, I deselected the All Lines file since it was no longer necessary to view. After that step I needed to find a shapefile that contained the data for all of the forest preserves in Cook County since this was a major variable in my project. Finding this shapefile also proved difficult, but I eventually found it on a private company's website called Green Map System. I was able to download the shapefile for all the forest preserves in the county (dark green) and also downloaded the shapefile for Chicago's public parks (light purple). At this point of my project I had my entire model completed. The next step was to analyze what exactly I had created and what this information meant in terms of my original project idea. I looked at the proximity of emerald ash borer positive sites to forest preserves to see which areas are at higher risk for infestations making the assumption that emerald ash borers are being spread through flight and natural occurrences (wind, rain) and not by human vectors, which will disregard street routes. The model also assumes emerald ash borers will land on the closest available ash tree. Using the measuring tool, I was able to determine that the forest preserve at the highest risk for an emerald ash borer infestation is Region 3, with the closest points within Leyden Township limits. The second highest risk forest preserve is Region 4, with the closest points in New Trier, Niles, and Evanston townships.

Conclusion

In conclusion, there were some setbacks and advances in creating my GIS model. However, I was able to use the data I found to create a model showing an invasive species problem and how distance can be used to determine areas that are at the highest risk for infestation. I believe that this model could be used as a starting point for further mapping as needed by local governments and state agencies that work with managing forest preserves and natural areas. Researchers could add further data such as forest preserve tree surveys and the number of infested trees at a positive site to better determine high risk areas in associated with the proximity of positive sites. Additionally, county and state agencies could use a more information-rich version of this model to see where they should focus more resources and work more closely with local governments to create conservation projects and give further aid to protect preserves that are at high risk of developing an emerald ash borer infestation.

Sources and Further Reading

D. J. Nowak. "Urban forest structure: The state of chicago's urban forest." http://www.na.fs.fed.us/fhp/eab/pubs/chicago_ash/chic_ ash.shtm. USDA Forest Service.

How to identify emerald ash borer. http://www.emeraldashborer.info/identifyeab.cfm, 2010. USDA Forest Service.

J. R. Muirhead, B. Leung, C. Van Overdijk, D. W. Kelly, K. Nandakumar, K. R. Marchant, and H. J. MacIsaac. "Modelling local and long-distance dispersal of invasive emerald ash borer agrilus planipennis (coleoptera) in North America." Diversity and Distributions, 12:71-79, 2006.

K. F. Kovacs, R. G. Haight, D. G. McCullough, R. J. Mercader, N. W. Siegert, and A. M. Liebhold. "Cost of potential emerald ash borer damage in u.s. communities, 2009-2019." Ecological Economics, 69:569-578, 2010.

"Wisconsin's emerald ash borer information source: Why is eab important." https://onlineservices.datcp.wi.gov/eab/article.jsp?topicid=13, 2010. Wisconsin Department of Natural Resources.