Lab 2: ArcGIS Spatial Analyst Extension Tutorial

Lab Overview

This lab demonstrates the capacities of ArcGIS's Spatial Analyst extension. The extension has many modeling and analyzing capabilities including finding suitable locations, performing distance and cost-of-travel analyses, identifying the best paths between two locations and performing statistical analyses, to name a few. In the case of this lab, data from the town of Stowe in Vermont is used and analyzed to find the best location to build a new school to relieve the onset pressure of other school due to a population increase in the region.

1. Exercise 1: Preparing for Analysis

1.1 Organizing your data

In this section, a new ArcGIS map was opened and it was connected to the Spatial Analyst folder that was created to hold the data of the exercise. The New File Geodatabase icon was clicked and a geodatabase for the exercise named "Scratch.gdb" was created. These two steps can be seen, respectfully, in figure 1 and 2.

1.2 Setting your workspace

Within the Geoprocessing menu, within the Environments tab, the current workspace is set to the Stowe.gdb and the scratch workspace is set to the scratch.gdb as shown in figure 3.

1.3 Adding data to your ArcMap session

The data is added from the Catalog window and dragging over the data from the Stowe file within the Spatial Analyst folder.

1.4 Saving your map document

The map in the figure below containing all of the inputted data was saved.

2. Exercise 2: Accessing the extension and exploring the data

2.1 Checking out an ArcGIS Spatial Analyst extension license

Customize > Extensions > Spatial Analyst.

2.2 Add the Spatial Analyst Toolbar

Customize > Toolbars > Spatial Analyst.

2.3 Creating a hillshade

Hillshades are a raster seen as a shaded relief. To retrieve this tool, the following must be followed: Search > Tools > Hillshade (Spatial Analyst). Elevation was set as the Input raster and a z factor equal to 0.3048. The resulting map can be seen below.

2.4 Displaying and exploring data

The Properties tab for the features can be opened to change the colour or symbols of the map components. The transparency of the layers can also be changed to better view the map features.

2.5 Select features on the map

Features on the maps can be selected by opening Attribute tables.

2.6 Identifying features on the map

ArcMAP has an identifier tool that can be used to examine the regions on the map. This is represented by an icon consisting of a blue circle with an "i" inside of it as selected in the image below.

2.7 Examining a histogram

On the Spatial Analyst toolbar the land use layer was selected and the histogram pictogram was clicked to render the table below.

3. Exercise 3: Finding a site for a new school

This exercise works toward the creation of a suitability model that can evaluate the optimal locations where a new should be built.

3.1 Creating a new toolbox

To create a new tool box, you must:

Open Catalog window, navigate to appropriate folder, right click folder, click New>Toolbox, and rename the toolbox. This toolbox was renamed as Site Analysis Tools. It was right clicked and New > Model was selected in which the model of this exercise was built.

3.2 Creating a new model

Models are built by connecting tools on a ModelBuilder window. The Site Analysis Tools toolbox was clicked then New>Model was selected. This opened the ModelBuilder window. The model Environments properties including the Processing extent's and Raster Analysis cell size's values were set to those of the elevation data since it represents the largest input in this scenario. The model was named FindSchool. The environmental settings of the model including its extent and cell size were set to "same as layer elevation" since elevation has the largest cell size.

3.3 Deriving datasets

The slope from the elevation dataset, the distance from recreation sites from the rec_sites dataset and the distance from existing schools from the schools dataset were derived from the exercise data. Slope and Euclidean distances tools were dragged from the Spacia Analyst Tools toolbox Surface and distance toolsets respectively. These were connected within the mode, missing information was entered for the elevation data such as a z factor and all outputs were renamed appropriately. The results of the model once run are shown below.

Figure 11. Distance from schools, distance from recreation sites and slope variance.

3.4 Reclassifying datasets

In order to combine the data of slope, land use, distance to recreation sites, and distance to schools, the locations must first be rated on a similar suitability scale system, such as a rating of 1-10, with 10 being the most suitable. The Weighted Overlay tool can be used to select the best combinations. However, the data ranges of each dataset must be set to discrete integers before being assigned a suitability rating and before the tool can use them. High ratings are assigned to low slopes, close proximities to recreation sites and far distances from existing schools. The model thus far with its functions to reclassify the datasets is seen in figure 12 below. Figure 13 show the reclassified data.

Figure 13. Reclassified data with most suitable regions in green and least suitable regions in red.

3.5 Weighting and combining datasets

In this stage of the exercise, the reclassified slope and distances will be combined with the land use to find the optimal locations to put a new school. Wetland and water areas will be restricted along with slopes of suitability equal to 4 or less. A further distance from existing schools and close to recreational sites is more highly regarded than the other parameters. The Weighted Overlay tool will be used for these operations as seen in figure 14. These are the importance ratings for each factor:

Reclassed distance to rec_sites: 50%

Reclassed distance to schools: 25%

Reclassed slope: 13%

landuse: 12%

Tool is found under Spacial Analyst toolbox > overlay toolset

The values to exclude such as a slope less than 30% (suitability of 1, 2, and 3) are given a Scale Value of "Restricted".

3.6 Selecting optimal sites

The pixels with a value of 9 are the most optimal. The size of the site must also be considered. Thus, optimal sites would have many pixels valued at 9 bunched together. The Con tool will be used to extract the optimal sites from the map. The Majority Filter tool is used to refine the optimal areas where a minimum of 8 neighboring pixels were selected as optimal sites from the Optimal areas raster. In the map, the optimal area pixels are white and the filtered optimal areas are in yellow. It can see that they are lesser in amount.

Figure 16. Code and output of Optimal and Filtered Optimal areas.

3.7 Selecting the best site

The next step is to filter out the locations that are far from the town's road network. Locations that are intersected with roads must be selected. The raster data must be converted to a polygon to begin. This is done with the Raster to Polygon tool. Then, the Select Layer By Location tool is used to select the polygon area that intersects the roads. This selected area is highlighted in the figure below. Then, the Select Layer By Attribute tool is used to select a location with over 10 acres in area. The Copy Features tool is then used to save a final_site file in the Stowe folder.

4. Exercise 4: Finding an alternate access route

This stage of the lab works on finding the best route for a new access road to the optimal school location.

4.1 Creating a new model

A model must be created to find the best route while considering the land slope and the land type. The parameters were set to the same as the previous model with the extends set to those of the elevation layer.

4.2 Creating the cost dataset

The model will evaluate the cost of travelling over the landscape, while attributing higher costs to travelling over sloped land as well as the costs to build roads on more more less difficult landscapes. The slopes are reclassified and the higher sloped classes are assigned a value of 10 being the most costly. The Landuse categories were assigned values as well to represent the more costly to least costly land to build a road on. This can be seen in figure 18 below. The average cost surface value map is seen in figure 19.

4.3 Finding the least costly path

The Cost Distance tool and the Cost Path tool are used in this section to determine the path of least cost to travel to the new site. Figure 20 shows the code used.

Figure 20. Model used for determining the lowest cost for a path.

4.4 Raster to Polyline

This step changes the raster path seen in the image above to a polyline for easier viewing of the pathway. The raster to polyline tool is used as seen below. The resultant map is in figure 23.

5. Conclusion

This exercise explores several tools within the Spatial Analyst extension. Two models were successfully built within ArcGIS to both find the best site for a school and find the least costly path. Skills like creating a geodatabase and properly saving files for ArcMAP were developed.