Suitability modeling workflow using ArcGIS Desktop—Lesson 4: Locating and connecting regions

The goal of a suitability model is to identify the best locations for the thing you are modeling—in your case, the best bobcat patches. To maintain genetic diversity and to allow for movement between the different populations, the patches need to be connected with the best network of wildlife corridors.

In this lesson you will identify the best five patches for bobcats, which will be derived from the final suitability surface created in Lesson 3: Weighting and combining data, and you will then connect the patches with the optimum network of wildlife corridors.

This lesson is divided into four sections:

1. Opening and editing the ModelBuilder model
2. Using the Locate Regions tool to identify regions
3. Using the Raster Calculator tool to combine regions into one layer
4. Using the Cost Connectivity tool to connect regions with an optimum network

ArcGIS Spatial Analyst extension is required for this analysis and you must ensure that it is enabled. See Enabling the Spatial Analyst extension.

This lesson continues on the data that you downloaded in lesson 1. If you have not downloaded the data, see the first section in Lesson 1: Exploring and deriving data.

Opening and editing the ModelBuilder model

1. If the SuitabilityModeling map document was closed after lesson 3, reopen it by double-clicking the SuitabilityModeling.mxd file in the SuitabilityModeling folder.

   

2. Open the SuitabilityModel model that you started building in lessons 1, 2, and 3 by right-clicking the model and clicking Edit.

   

Using the Locate Regions tool to identify regions

The Locate Regions tool will be used to identify the optimal arrangement of habitat patches from the suitability surface.

Note:

The Locate Regions tool is only available in ArcMap 10.5 or ArcGIS Pro 1.3 or newer releases. If you are using ArcMap 10.4.1 or lower, you will not be able to actually run the Locate Regions tool in this lesson but the output has been created for you, therefore, you will be able to complete all of the steps in this lesson. See the ArcGIS Pro lesson for the steps on how to use the Locate Regions tool.

1. For the purposes of this lesson for ArcMap, it is assumed that you are using version 10.4.1 or lower. Therefore, the output NewRegions layer from the Locate Regions tool has been calculated for you and is in the Data.gdb geodatabase in the SuitabilityModeling folder. These regions were determined from the Locate Regions tool using the following parameters with the inclusion of an existing region:
   • Input raster: FinalSuitability
   • Total area: 50
   • Area units: Square miles
   • Output raster: NewRegions
   • Number of regions: 5
   • Region shape: Circle
   • Evaluation: Highest average
   • Region minimum area: 5
   • Region maximum area: 14
   • Minimum distance between regions: 4
   • Maximum distance between regions: 13
   • Distance units: Miles
   • Input raster or feature of existing regions: ExistingRegion
2. Add the ExistingRegion and NewRegions layers to the display to view them relative to the suitability surface created in lesson 3. To do so, browse to the ExistingRegion and NewRegions layers in Data.gdb within the SuitabilityModeling folder.

   

3. Drag the ExistingRegion and NewRegions layers from the Data.gdb geodatabase into the table of contents. Place the ExistingRegion and NewRegions layers on top of the FinalSuitability surface, and turn off all other layers.

   

4. Change the symbology of the NewRegions layer by clicking the color square to the left of the 0 (zero) value in the NewRegions legend. Choose No color from the drop-down list of Fill color options.

   

5. Explore the NewRegions and ExistingRegion layers relative to the FinalSuitability surface. The Locate Regions tool calculated the optimum spatial arrangement of regions with the highest average suitability value (while accounting for an existing region), in accordance with the parameters outlined above.

   

Using the Raster Calculator tool to combine regions into one layer

Before the new and existing regions can be connected by an optimum, least-cost network, they need to be put into a single layer. In this section, you will use the Raster Calculator tool to merge the ExistingRegion layer with the NewRegions layer to create a single layer of all the regions—all the patches that will be available to the bobcats.

1. Add the ExistingRegion and NewRegions layers to the SuitabilityModel ModelBuilder model, placing them to the right of the FinalSuitability layer. Leave a gap between the FinalSuitability and NewRegions layers to indicate where the Locate Regions tool goes.

   

2. The ExistingRegion layer has a value of 1 for cells in the existing region and NoData everywhere else. If you add the existing regions to the new regions, any locations that are assigned NoData in ExistingRegion will receive NoData in the merged output. Since you want to union the two datasets, you need to change the NoData cell locations in ExistingRegion to 0. Then the two rasters can be successfully added together. To reclassify the NoData values in the ExistingRegion layer, click the ArcToolbox icon on the Standard toolbar. Browse to the Reclassify tool in the Reclass toolset of the Spatial Analyst Tools toolbox.

   

3. Drag the Reclassify tool into the SuitabilityModel and place it to the right of the ExistingRegion layer.

   

4. Using the Connect tool in the ModelBuilder taskbar, connect the ExistingRegion layer to the Reclassify tool. Choose the Input raster option.

   

5. Double-click the Reclassify tool to open the tool's dialog box. Type ExistingRegion_new in the Output raster parameter to name the output reclassified raster. The model automatically saves the file into the Output.gdb geodatabase. Enter the following new values in the table provided within the Reclassify tool's dialog box.

   Old values	New values
   1	6
   NoData	0

   To summarize the input and output layers for the Reclassify tool, specify the parameters as follows:

   • Input raster: ExistingRegion
   • Reclass field: Value
   • Output raster: ExistingRegion_new

   

Each of the regions in NewRegions is uniquely numbered from 1 to 5, with all other cells assigned 0. In this step you are reclassifying the 1 values (the existing region) in ExistingRegion to 6, and all other locations, the NoData values, to 0. When you later add the two layers you will get values 1 through 6 for the existing and new regions, with all locations with no patches being assigned 0.
6. Click OK to apply the changes you have made and close the Reclassify tool's dialog box.
7. Once the dialog box is closed, right-click the Reclassify tool and click Run to execute the tool.

   

8. Add the ExistingRegion_new layer to the map document by right-clicking the green ExistingRegion_new layer in the model and clicking Add To Display.

   

9. Visually explore the ExistingRegion_new output layer. The color scheme clearly displays that the Reclassify tool transformed the existing patch locations to 6 and all the NoData cells to a value of 0.

   

This reclassification allows you to now add the existing and new regions together to identify all patches available to bobcats. To do so, you will use the Raster Calculator tool. The Raster Calculator tool will be used to create a mathematical expression allowing you to add the existing region and new regions resulting in a single layer of bobcat patches.
10. Click the ArcToolbox icon on the Standard toolbar. Browse to the Raster Calculator tool in the Map Algebra toolset of the Spatial Analyst Tools toolbox.

    

11. Drag the Raster Calculator tool into the SuitabilityModel model and place it to the right of the ExistingRegion_new and NewRegions layers.

    

12. Double-click the Raster Calculator tool icon in the SuitabilityModel model to open the tool's dialog box. Enter the following expression, which combines the ExistingRegion_new and NewRegions layers into a single layer. Within the expression, once the two layers are combined (using the + operator), the expression then sets all zero values (where there are neither existing nor new patches) to NoData. Use the following expression in the Raster Calculator:

    SetNull(("%ExistingRegion_new%" + "%NewRegions%"), "%NewRegions%", "Value = 0")

    Type Regions in the Output raster parameter to name the output raster of available bobcat patches.

    

13. Click OK to apply the Map Algebra expression and close the Raster Calculator tool's dialog box.
14. Once the dialog box is closed, right-click the Raster Calculator tool and click Run to execute the tool.

    

15. Add the output Regions layer to the map document by right-clicking the green Regions layer in the model and clicking Add To Display.

    

16. Turn off all other layers in the table of contents except the FinalSuitability layer. Each region now has a unique value, and they are ready to be connected with the Cost Connectivity tool.

    

Using the Cost Connectivity tool to connect regions with an optimum network

At this stage of the workflow, the optimum spatial arrangement of regions has been found. Now you will connect the regions with a network of optimum paths using the Cost Connectivity tool. The Cost Connectivity tool requires two inputs: the regions to be connected and a cost surface. A cost surface identifies how costly it is to move across a landscape. For more information on cost surfaces, see the Understanding cost-distance analysis case study.

1. Add the CostSurface layer to the table of contents from the Data.gdb geodatabase within the SuitabilityModeling folder.

   

2. Choose a red-to-green color ramp for the CostSurface layer to help portray each location's cost. Double-click the CostSurface layer in the table of contents and change the Color Scheme under the Symbology tab. Select the Stretched display of values from the left panel before choosing a red-to-green color ramp from the drop-down options. Check the Invert check box near the bottom of the window to align high-cost values with red and low-cost values with green. Click OK once the color ramp has been selected to apply it to the layer.

   

3. Visually explore the cost surface by clicking in different locations with the Identify tool , found on the Tools toolbar. According to the legend, green areas are less costly and easier for travel, whereas red areas are costly to travel through.

   

4. Add the CostSurface layer to the SuitabilityModel model by clicking the SuitabilityModel model. Drag the CostSurface layer into the model and place it above the Regions layer.

   

5. Now that both required inputs for the Cost Connectivity tool are in the SuitabilityModel, the Cost Connectivity tool can be added to the model. Click the ArcToolbox icon on the Standard toolbar. Browse to the Cost Connectivity tool in the Distance toolset of the Spatial Analyst Tools toolbox.

   

6. Drag the Cost Connectivity tool into the SuitabilityModel model and place it to the right of the CostSurface and Regions layers.

   

7. Using the Connect tool in the SuitabilityModel ModelBuilder model taskbar, connect the CostSurface layer to the Cost Connectivity tool. Make the connection by clicking the CostSurface layer first, and then click the Cost Connectivity tool. Choose the Input cost raster option.

   

8. With the Connect tool still turned on in ModelBuilder, connect the Regions layer to the Cost Connectivity tool. Choose the Input raster or features region data option.

   

9. Double-click the Cost Connectivity tool in the SuitabilityModel model to open the tool's dialog box. Type OptimumNetwork in the Output feature class parameter. To summarize the parameters of the Cost Connectivity tool, specify the parameters as follows:
   • Input raster or feature region data: Regions
   • Input cost raster: CostSurface
   • Output feature class: OptimumNetwork

   

10. Click OK to apply the changes and close the Cost Connectivity tool's dialog box.
11. Right-click the Cost Connectivity tool and click Run to execute the tool.

    

Note:

Currently, the Cost Connectivity tool requires an Advanced license when multi-cell regions or polygon sources are entered. If you have a Basic or Standard license you will receive an error message and you will not be able to add the results to the map.

12. Right-click the output OptimumNetwork layer (green) and click Add To Display to add the output optimum network to the map document.

    

13. Save the SuitabilityModeling model by clicking the Save button in the ModelBuilder taskbar. Close the model once it is saved.
14. Order the layers in the table of contents so that the OptimumNetwork layer is on top of the Regions and CostSurface layers.

    

15. Explore the OptimumNetwork layer by clicking the individual paths with the Identify tool , found on the Tools toolbar. Identified are the total cost of the path (PATHCOST) and the source and destination regions.

    

16. Save the map document by clicking the Save button on the Standard toolbar before closing the SuitabilityModeling ArcMap document.

Congratulations! You have successfully created a network of wildlife corridors that connect an optimal arrangement of wildlife patches.