Final Project

Goals and Background

          For the final project in GIS II, I decided to designated potential areas where a wildlife corridor could be built underneath a road particularly for the Blanding's turtle in Wisconsin. Protecting at-risk species from extinction helps preserve local biodiversity, which in turn, strengthens the health of the local ecosystems. The Blanding's turtle is a protected species and a species of concern in Wisconsin, threatened by habitat fragmentation from encroaching road development.
          Starting with the entire state, the analysis will narrow in to one county for further analysis of suitable roads for the installation of a wildlife corridor.

Methods

          The process started with reprojecting the land cover raster data and Blanding's turtle distribution raster data to NAD 1983 HARN Wisconsin (US feet). The ecological systems field of the land cover data was reclassified to designate all wetland related land types as 1 (the rest as 0). Then the Blanding's turtle distribution data was reclassified to a single value of 1. The two reclassified rasters were multiplied in a Boolean raster operator and overlayed in a raster of likely turtle habitat. The original land cover raster was reclassified to designate grassland and meadow-like habitats, likely nesting habitat, to a value of 1.
          Polygons of Wisconsin counties were used as zones for zonal statistics for the sum of cells valued as turtle habitat. Using the identity tool, it turns out that Marinette County contained the most turtle habitat. The turtle habitat raster and nesting site raster were subsequently clipped to a polygon of Marinette County. A Marinette road shapefile was added to the display and processed through a 1620 m (maximum distance the turtle will travel from habitat to nesting sites) Euclidean Distance tool. The distance from 0 to 114 m (area of relatively high turtle traffic) was then reclassified to a value of 1 and then converted into a polygon. Then, the road data was intersected to create points where the lines crossed. Afterward, the points were used to split the road data where it intersects the polygon. It was then possible to select road sections that were contained within the 144 m polygon. Lastly, the selected roads that were merely slivers or did not significantly cross swaths of turtle habitat were deleted as they would not make likely candidate for the location of a highly utilized corridor. Figure 1 displays the results.

Results

          Plenty of roads sections make suitable locations for a potential wildlife corridor for the Blanding's turtle. For some reason, not all of the roads were selected that were within the 114 m polygon. This might have something to due with problems with intersecting or splitting the roads. Turtle habitat adjacent to selected roads could hypothetically be surveyed for the presence of turtle population to determine the validity of the selection. 

Figure 1: Map of potentially suitable site to install a wildlife corridor.

Sources

Data

Wisconsin Land Cover - U.S. Geological Survey Gap Analysis Program, 2013, GAP/LANDFIRE National Terrestrial Ecosystems 2011: U.S. Geological Survey. https://gis1.usgs.gov/csas/gap/viewer/land_cover/Map.aspx

Blanding’s Turtle Distribution – U.S. Geological Survey Gap Analysis Program, 2016, U.S. Geological Survey Gap Analysis Program Species Distribution Models: U.S. Geological Survey. https://gapanalysis.usgs.gov/species/data/download/

Marinette County Roads – Marinette County Land Information Office, 2017, Roads Marinette County, WI 2017: UW – Madison. http://maps.sco.wisc.edu/opengeoportal/

Wisconsin Counties – Price, H. M., 2016, Counties: Mastering ArcGIS -  McGraw-Hill Education.

Bibliography

Bury, R. B., & Germano, D. J. (2002). Differences in Habitat Use by Blanding's Turtles, Emydoidea blandingii, and Painted Turtles, Chysemys picta, in the Nebraska Sandhills. The American Midland Naturalist , 241-244.

Joyal, L. A., McCollough, M., & Hunter Jr. , M. L. (2001). Landscape Ecology Approaches to Wetland Species Conservation: A Case Study of Two Turtle Species in Southern Maine. Conservation Biology , 1755-1762.

Proulx, C. L., Fortin, G., & Blouin-Demers, G. (2014). Blanding's Turtles (Emydoidea blandingii) Avoid Crossing Unpaved and Paved Roads. Journal of Herptology , 267-271.

Ross, D. A., & Anderson, R. K. (1990). Habitat Use, Movements, and Nesting of Emydoidea blandingi in Central Wisconsin. Journal of Herptology , 6-12.

Lab 4

Goals and Background

          The objective of Lab 4 was to implement the basic knowledge and skills learned in two ESRI courses introducing the concepts of Network Analysis. The topology and connectivity of networks allow extensive analyses to to be performed to solve a variety of applied problems, from finding the shortest route to point A to point B to minimizing distribution cost for a business. To apply network analysis to UW - Eau Claire, a simple question was asked: What is the shortest route to get from the Haas Fine Arts Center to the L.E. Phillips Science Hall? This is a path that students take almost every day. In 2017, construction began on Garfield avenue and had disrupted the natural flow of pedestrian traffic (Figure 1). The goal was to use Network Analysis to figure out if the route that students had to take during the start of construction was longer of shorter than the route that opened up after the bulk of the construction was finished.

Figure 1: Garfield Avenue construction zone on Lower Campus.

Method

          Before beginning to create the network dataset, enable the Network Analysis extension in the Customize tab. In the ArcCatalog window, navigate to and right-click the UWEC routing dataset. Select to create a New Network Dataset. In the New Network Dataset window, choose uwec_paths as the edges that will be participating in the network. Run through the steps in setting up the network dataset. Select to model turns. Use elevation fields. Don't use traffic data with the network dataset. In the Evaluators tab, change the value for uwec_paths to shape-length. Set the travel mode to walk and distance impedance to meters. Finish and build the network dataset. 

          Once the network dataset is built, add it to the viewing window. Open the Network Analysis toolbar. Make sure that it is set to the new dataset. Select New Route under the Network Analysis toolbar drop down window. Open the Network Analysis window from the toolbar and dock it below the Table of Contents. Right-click Stops and choose Load Locations. Load the uwec_stops feature and use objectID field and the default value of 1 to select the first stop near the Haas Fine Arts Center. Afterwords, go the Route Properties and in the Analysis Settings tab, set the impedance to length (meters). Finally solve the route. This gives the route that students take now (Figure 2). To analyze the other route, right-click Barriers and choose Load Locations. Load the uwec-barrier feature and use the objectID field and the default value of 1 to select the first barrier. Afterwards, solve the route again (Figure 3).

Results

          The first route that the Network Analysis generated had a length of approximately 429.9 meters (Figure 2). If the average pedestrian walking speed in 5 km/hr, then the travel time for Route 1 is approximately 5.16 minutes. The second route that the Network Analysis generated had a length of approximately 517.4 meters and a travel time of approximately 6.21 minutes. The shortest route was Route 1, the route that was opened up recently for students. It was approximately 87.5 meters shorter and 1.05 minutes shorter to traverse. Students lost a little over a minute going to classes when construction blocked major sidewalks in 2017.

Figure 2: Route 1 generated the shorter and current route that students take.

Figure 3: Route two generated the previous route students took during the bulk of the construction.

Sources

UW- Eau Claire (n.d.) Garfield Avenue Redesign. Retrieved from https://www.uwec.edu/facprojects/garfield.htm.

Curtis, C. (2018) UWEC_paths.

Curtis, C. (2018) UWEC_stops.

Curtis, C. (2018) UWEC_barrier.

ESRI (2018) World Topographic Map. Retrieved from https://www.arcgis.com/home/item.html?id=30e5fe3149c34df1ba922e6f5bbf808f.

Lab 3

Goals and Background

          The goal of Lab 3 was to learn how to delineate watersheds and understand the concepts behind the analysis. Watersheds are geographically and environmentally important because the water within coverage at low points within the area and exit at a single point, forming rivers and lakes. This network of water sources make pollution particularly troublesome as it can effect all water sources downstream from it's point of origin. Therefore, delineating watershed is of special interest of land and water managers in order to help monitor the amount and quality of water networks within different watersheds.

Methods

          To begin Lab 3, download the Adirondack Park Boundary shapefile from the New York Stat Clearinghouse and unzip the data to your geodatabase. Open ArcMap and notice that the projection for Adirondack Park Boundary feature class is in NAD 1983 UTM Zone 18N in meters. All other features are going to be reprojected to this projection. But first, open the Buffer tool from the ToolBox within Analysis Tools > Proximity. Create a 20 km buffer around the park boundary, setting Dissolve to All. This will create smoother watersheds later in the analysis. 

          Use the Reproject tool from Data Management Tools > Projections and Transformation and reproject the hydrology feature class to the same projection as the park boundary. Utilize Import Projection to easily accomplish this. Use the Clip tool to clip the reprojected hydrology feature class to the original park boundary layer. 

          From Add Data From ArcGIS Online, add a raster called 30-arc-second DEM of North America. Since the DEM has a different project than the layers in the data frame, a window will appear. Click Transformations and set the transformation as convert from GCS_WGS_1984 to NAD 1983. Clip the DEM to the park boundary buffer and check Input Features for Clipping Geometry. Remove the original DEM as it's not needed anymore. Use the Project Raster tool to reproject the clipped DEM to the same projection as the park boundary layer. Use the same method as with the hydrology layer, but include the WGS_1984 to NAD_1983 transformation, choosing bilinear resampling method, and set the X and Y output cell size to 60 m. Once finished, display data that is only in the UTM projection.

          To delineate watersheds, flow directions need to be calculated for each cell. In the Spatial Analyst Tools > Hydrology, select the Flow Direction tool. Use the reprojected DEM as the input surface raster. Next, sinks need to be removed so that the water flow won't be disrupted falsely. Use the Fill tool (also in the Hydrology category) to fill sinks in the reprojected DEM. Determine flow direction for the filled DEM. Water accumulation areas need to be determined. Open the Flow Accumulation tool (Hydrology category) and use your second flow direction output as input. Lastly, a source raster is needed to create a threshold to determine the minimum number of cells that flow into any cell before it is designated as a stream cell. Open the Con tool from Spatial Analyst Tools > Conditional. Choose your water accumulation output as your input conditional raster. Set Type to Value > 50,000 and use 1 as your input true raster value. Label it as net_50k and run the tool. Open the Stream Link tool (Hydrology) to assign unique identifiers to each stream reach. Use net_50k as your input stream raster and use your second flow direction output as your input flow direction raster. Label it as source and run the tool. Open the Stream to Feature tool (Hydrology) to create vector streams using the source raster as your input stream raster and your second flow direction output as your input flow direction raster. 

          Finally, to delineate watersheds, open the Watershed tool (Hydrology). Use you second flow direction output as your input flow direction raster and your source raster as your input raster. Run the tool. Clip the output to the park boundary, checking Input Features for Clipping Geometry. Add the clipped hydrology layer to compare than generated watersheds to the stream locations (Figure 1). 

Results

          Comparing the results from the watershed delineation from a DEM with a cell size of 60 m (Figure 1) to a DEM with a cell size of 120 m (Figure 2), the differences are quit evident. Designating a larger cell size for the same DEM will simplify the raster, cascading its effects to the delineation. This creates less and oversimplified watershed areas. 

          The Vector Streams created from the Methods section are much more simplified versions of the hydrology layer (Figure 3). They follow the general trends of the rivers, being based of the flow direction, water accumulation, and an arbitrarily defined threshold value. 

Figure 1: Watershed delineation with a 50,000 cell threshold derived from a 60 cell size DEM.

Figure 2: Watershed delineation with a 50,000 cell threshold derived from a 120 m cell size DEM.

Figure 3: Vector streams from the watershed analysis compared to the hydrology feature class.

Sources

Barge, J. (n.d.). Adirondack Park Boundary [Downloaded Data]. Retrieved from http://gis.ny.gov/gisdata/inventories/details.cfm?DSID=303.

 National Aeronautics and Space Administration (NASA), the United Nations Environment Programme/Global Resource Information Database (UNEP/GRID), the U.S. Agency for International Development (USAID), the Instituto Nacional de Estadistica Geografica e Informatica (INEGI) of Mexico, the Geographical Survey Institute (GSI) of Japan, Manaaki Whenua Landcare Research of New Zealand, and the Scientific Committee on Antarctic Research (SCAR). 30-arc-second digital elevation model (DEM) of North America. Retrieved from ArcGIS Online.

Cornell University. hydrology. Retrieved from https://cugir.library.cornell.edu.

Lab 2

Goal and Background

          The main purpose of Lab 2 was to familiarize with georeferencing and digitizing techniques from GIS I in order to solve spatial referencing problems and analyze spatial data in preparation or new concepts in GIS II.

Methods

          Open ArcMap and set up the georeferencing toolbar from the Customize tab. Bring the Centerlines_clip shapefile into ArcMap so that the data frame projects on the fly for that projection. Then add the Topographic base map from ESRI. Next, bring in the scanned 1878 map of Eau Claire, WI which will be georeferenced to the Centerlines_clip as its reference. Using the georeferencing toolbar, Fit to Display the scanned map and shift and scale the map so that it relatively matches the base map. Adding transparency to the scanned map will make this process easier. Add control points by clicking the scanned map first and then to its corresponding place in the reference image or feature class. Control points should be equally dispersed across the target image, but in this case, that might not be entirely possible. Easily distinguishable features between the target and reference image were clustered around the center of the image. Add control points until the scanned map lines up with the reference image until it is satisfactory. The RMS might be high, but considering the 140-year gap between the reference and target image, if it visibly looks alright, it'll have to do. Set a transformation. A first-order transformation might be the most appropriate choice since the warping of the other orders aren't necessary for a scanned map and the exact coordinates aren't known for a spline transformation (if there's enough control points). Update georeferencing when finished. Set the transparency to 50% to inspect how well the georeferenced image fits the base map (Figure 1).

          The second part of the lab involving creating data to analyze the differences in water area between 1878 and 2018. First, create a new geodatabase. From the Toolbox, open the Create Feature Class tool. Set its location to the new geodatabase, name it hydro_1878, set its geometry type to polygon, and set its coordinate system to the same one as Centerlines_clip. Bring in the study area shapefile into the data frame. Open the Editor toolbar and click Start Editing. Edit the feature class hydro_1878. Using polygons, digitize the Chippewa River, Eau Claire River, and Half Moon Lake within the study area using the georeferenced 1878 scanned map as a reference. Save edits and click Stop Editing. Now create another feature class and call it hydro_2018. Start editing the hydro_2018 feature class. Digitize the Chippewa River, Eau Claire River, and Half Moon Lake within the study area using the ESRI Topographic base map as a reference. Save edits and stop editing. Open the new feature classes' attribute table and use Statistics on the Shape_Area field to compare the total area of water between 1878 and 2018. Visually compare the differences by displaying the feature classes together (Figure 2).

Results

          Figure 1 shows the 1878 scanned map georeference and displayed with a 50% transparency over a topographic base map. 

Figure 1: Map of the 1878 Eau Claire map georeferenced over a modern base map.

          Figure 2 shows the differences between water features in 1878 and 2018. Visually, there appears to be more water area in 2018, but in reality, there was more water area in 1878, given by the statistics in the attribute table.

Figure 2: Map of the water area differences in Eau Claire, WI between 1878 and 2018.

Sources

David Ramsey Map Collection (2018). Eau Claire and Medford. Retrieved  from https://www.davidrumsey.com/luna/servlet/detail/RUMSEY~8~1~4181~480085#.

Eau Claire County (2014) Master_Centerlines. Retrieved from Caitlyn Curtis.

ESRI (2018). Wold Topographic Map. Retrieved from https://www.arcgis.com/home/item.html?id=30e5fe3149c34df1ba922e6f5bbf808f.

Lab 1

Goal and Background

          The purpose of Lab 1 was to revisit key concepts and basics of GIS I and to familiarize with the software before learning new skills and techniques of GIS II.

Methods

          Within ArcCatalog, selecting the Connect to Folder icon will allow the user to bring in the desired directory to the Catalog Tree, where the desired folders and contents are easily accessible on the left side of ArcCatalog. The contents of the folder can be examined by clicking the + icon next to the folders. Connect to the Lab 1 folder and activate Countries94. In the Preview tab, the data Counties94 is shown in its original form. Preview other data within the Lab 1 folder.

          Open a blank map in ArcMap, where the data can be displayed and manipulated in layers. Open the ArcCatalog tab within ArcMap and drag the shapefile WorldCities into the display screen in ArcMap. Clicking on the colored box in the Table of Contents can allow the user to change the colors of the data.

          Under File, select Open. Navigate to the Lab 1 folder and select the Redlands map (it's a mxd. file). The map is displayed in ArcMap. Checking and unchecking the layers in the Table of Contents will display and remove layers on the display screen. Try displaying Railroads and Streets. Right click any of the layers and select Properties. Information about the layer and be viewed and altered in ArcMap, like its geographic coordinate and projected coordinate system. 

          The Redlands map has a stored bookmark. From the Bookmark tab, select ESRI. The display zooms in on the ESRI feature, the location of ESRI. ESRI and street names are now visible because the map creator put a scale limit to these two features. In the Toolbar, click on the Identity tool. Click on New York Street. A window opens with all the attribute information on New York Street. Clicking on the name of the street in the Identity window will flash the feature on the display screen. Selecting All Layers from the layer dropdown in the Identify window will display all attributes of all the layers that New York Street is contained in. Identify more features and review their attributes.

          Return to the original extent by clicking on Original in the Bookmarks tab. Right click Railroads and open its Attribute Table. Review its attribute fields and the number of rows. Do the same for the Donut Shops and Streets layer. 

          Open a new blank map in ArcMap. From the Add Data icon, navigate to the Lab 1 folder and add the Erie shapefile. Open its attribute table and take note of its attribute fields. Open the layers Properties window and go to its Symbology tab. Select Quantities and then Graduated Colors. For its Field, select Persons. Change the color ramp to an appropriate choice. Some data should be normalized. For example, if an attribute can be affected by how much land there is, it's a good idea to normalize it by Area. Normalize Person by Area to get a population density (Figure 1). From the Properties window, go to the Labels tab. Select Person as a Label Field. Right clicking the Erie layer in the Table of Contents and selecting Label Features will show the labels in the display screen. In the Symbology tab, create a graduated color scheme for another attribute and decide if normalization is appropriate (Figure 2).

          

Results

           The attributes Persons and Households were both normalized by Area to show more useful data - population density (Figure 1) and household density (Figure 2). Both variable show very similar patterns in which tracts have high and low densities. This comes to no surprise as both variables should go hand-in-hand.

Figure 1: Population density for Erie County.

Figure 2: Household density for Erie County.

Sources

Curtis, Caitlyn (2018) Lab 1 Data [Downloaded Data].