As the final lab of the semester and the GIS 1 class in whole, we were tasked with applying all the skills we learned over the semester to each create our own final project. The final project would involve answering a question each of us formulated by looking at several collections of data provided to us. Using data provided from the Wisconsin DNR, I was able to gather data on reported invasive plant species locations within state forests. However, this data was largely limited to only the state forests of Wisconsin, while the nearby county forests remained largely unsurveyed. From this, I determined my research question. What areas of county forests were vulnerable to infestation from nearby state forests? The goal of the question was to identify these vulnerable areas in relation to the nearby infested areas so they could be properly monitored. This data would ideally be available to anyone on the state or county level that was interested in the preservation of natural biodiversity in these forests. Hopefully, the local DNR could use it to effectively survey and mange the local county forests.
Data and Source:
All of the data used in this lab was provided by the Wisconsin DNR. The data used includes a state forest feature class, a county forest feature class, a roads feature class, the reported invasive plant species locations within state forests, and the surrounding county and state lines. The problem with much of this data is that its to focused to be viewed at a statewide level. In order for an accurate and visible map to be generated, the map needed to focus on one specific county. Additionally, this specific county needed to county both designated state and county forests, which counties within Wisconsin didn't have. For this reason, Jackson County of Wisconsin was selected to be the area of interests. Additionally, estimating area for the infected locations presented difficulties, as these points lacked true values for the area infested, opting for rough ranges and estimates of area. Because of this, a 100 meter buffer was performed on the reported invasive plant locations to estimate area.
Methods:
In order for the any of the data to be used, it first needed to be cut to the Jackson County borders. This was done for the county forests, state forests, roads, and invasive species locations (SF Invasive) in order to create a focus area and to speed up load times. Afterwards, a 100 meter buffer was performed on the SF Invasive feature class to estimate the area infected. This buffer was intersected with the state forests feature class. Afterwards, the dissolve tool was used to remove internal boundaries, creating a feature class showing the area with reported invasive plant species in state forests. Additionally, the total area infested 100 meter buffer was also intersected with the county forests, as several of the locations reported did in fact lie beyond the boundaries of state forests and within the boundaries of the county forests. From this intersect, a feature class was created showing the county forests that were already reported as infected.
From previous knowledge of invasive species, it was believed that roads could assist in unintentional transportation. In order to verify this, a spacial query was performed to see if a significant portion of reported locations lay within 100 meters of roads. Indeed, a significant portion did. Because of this, it was necessary to take roads into consideration. A two mile buffer was performed on the SF Invasive feature class in order to create a logical range outside of the state forests where more invasive plants may exist. The resulting feature class was run through the buffer tool to decrease processing time. Then, the two mile buffer was intersected with a 100 meter buffer of the road feature class, as it was previously determined that a majority of invasive species locations lay within 100 meters of a road. This was then intersected with the county forests feature class to create a logical search area within county forests. The county area already known to be infected was erased from this using the erase tool, creating the final resulting area within county forests that was suspect of infection and should be checked.
With this completed, a map of Jackson County was created in order to highlight these suspect areas, showing them in relation to the county forests, state forests, roads, and the areas already known to be infected. A smaller reference map was also created to show Jackson County in relation to the rest of the state of Wisconsin. Furthermore, a data flow model was created showing the full process, minus the original clipping to scale, used to create the feature classes present in the final product.
Results:The resulting map shows that much of the county forest areas should in fact be surveyed for invasive plant species. According to the map, all of the area around the roads leading into the eastern county forest from the west should be surveyed, as well as as the areas around the roads leading into the central county forest from the north, east, and west. Small patches of these areas already show infestation. This is relatively unsurprising, given the presence of invasive plant species throughout the state forests of Jackson county.
Post-Project Evaluation:
Overall, I believe that this project served as an excellent way to creatively put together all of the GIS information and map building skills learned over the course of the semester. It provides a fairly accurate visual representation of the suspected presence of invasive plant species within state and county forests. If asked to repeat this project, I would wish to perform it with all the counties from which information was readily made available, as the project was limited in scope in order to prevent it from becoming overwhelming for a single individual. Additionally I faced challenges in creating realistic representations of areas already infected and for creating a reasonable search window for the areas suspect of infection. This is due to the fact that the area infected was based on an estimate I largely estimated and the 2 mile search window was arbitrarily decided after several buffers were performed. If this project were to be performed in the future, I would prefer to gather accurate area measurements for areas infested with invasive species and not have to base it around a estimated buffer performed on a series of roughly 850 points.
Sources:
Data Source: Wisconsin DNR
Hupy, C. (2016). Lab 4: Mini-Final Project. Eau Claire, Wisconsin.
