Lab 6: Terrain Descriptions, Viewshed Analysis, and Watershed Delineation

Due Dec 1

Introduction

 

This lab has six short tasks. Task 1 lets you create a contour layer, a vertical profile, and a shaded-relief layer from a DEM. In Task 2, you will derive a slope layer, an aspect layer, and a surface curvature layer from DEM data. Task 3 covers viewshed analysis and the effect of the viewpoint's height offset on the view shed. Task 4 creates a cumulative view shed by using two viewpoints, one of which is added through on-screen digitizing. Task 5 covers the steps for deriving watersheds from a DEM. Task 6 focuses on the derivation of point-based watersheds and the importance of snapping points of interest to the stream channel.

 

You will use Spatial Analyst, 3D Analyst, and ArcToolbox to perform the tasks in this lab. Viewshed tools are available through Spatial Analyst, 3D Analyst, and ArcToolbox in ArcGIS. Watershed tools are available through Spatial Analyst and ArcToolbox. You will alternate the use of Spatial Analyst and ArcToolbox for different tasks. Type the answers to all questions at the end of each task, attach your map, and submit them to the instructor.

 

Before you start, copy the Lab6 folder in I:\Students\Instructors\Geoffrey_Duh\GEOG4593\ to your workspace in C:\Users.

 

 

Task 1. Use DEM for Terrain Mapping

 

Instructions

What you need: plne, an elevation raster; and streams.shp, a stream shapefile. The elevation raster plne is imported from a USGS 7.5-minute DEM. The shapefile streams.shp shows major streams in the study area.

 

Create a contour layer

  1. Launch ArcMap. Add plne to ArcMap. Make sure that both the Spatial Analyst and 3D Analyst extensions are checked in the Tools menu and their toolbars are checked in the View menu.
  2. Click the Spatial Analyst dropdown arrow, point to Surface Analysis, and select Contour. In the Contour dialog, select plne for the input surface, enter 100 (meters) for the contour interval and 800 (meters) for the base contour, and save the output features as ctour.shp. Click OK to run the analysis.
  3. ctour appears in the map. Select Properties from the context menu of ctour. On the Labels tab, check the box to label features in this layer and select CONTOUR for the label field. Click OK. The contour lines are now labeled. (To remove the contour labels, right-click ctour and uncheck Label Features.)

 

Create a vertical profile

  1. Add streams.shp to ArcMap. Now select a stream by attribute for the vertical profile. Open the attribute table of streams. Click the Options dropdown arrow and choose Select By Attributes. Enter the following SQL statement in the expression box: "USGH_ID" = 167. Click Apply. Close the streams attribute table. Zoom in on the selected stream.
  2. Click the Interpolate Line tool on the 3D Analyst toolbar. Use the mouse pointer to digitize points along the selected stream. Double-click the last point to finish digitizing. A rectangle with handles appears around the digitized stream.
  3. Click the Create Profile Graph tool on the 3D Analyst toolbar. A vertical profile appears with a default title and subtitle. Right-click the title bar of the graph and select Properties. Change the title to “Stream Profile” and the subtitle to “Distance from Confluence”. Get a screen shot of the graph and paste it into your report.
  4. The digitized stream becomes a graphic element on the map. You can delete it by using the Select Elements tool to first select it. To unselect the stream, choose Clear Selected Features from the Selection menu.

 

Create a hillshade layer

  1. Click the Spatial Analyst dropdown arrow, point to Surface Analysis, and select Hillshade. In the Hillshade dialog, select plne for the Input surface. Take the default values of 315 for the azimuth, 45 for the altitude, 1 for the Z factor, and 30 for the output cell size. Opt for a temporary output raster. Click OK to run the operation. Get a screen shot of the hillshade and paste it into your report. Please clearly label the azimuth and altitude you used.
  2. Try different values of azimuth and altitude to see how these two parameters affect hill shading. Create another hillshade with 135 for the azimuth and 45 for the altitude. Get a screen shot of the hillshade and paste it into your report. Clearly label the azimuth and altitude you used.

 

Questions

  1. What is the elevation range along the vertical profile? Does the range correspond to the readings on ctour.shp? (Paste the stream profile graph here)
  2. Does the hillshade layer look darker or lighter with a lower altitude (e.g., 10)? (Paste the hillshade maps here)

 

Task 2. Derive Slope, Aspect, and Curvature from DEM

 

Instructions

 

Derive a slope layer

  1. Click Show/Hide ArcToolbox Window to open the ArcToolbox window in ArcMap. Double-click the Slope tool in the Spatial Analyst Tools/Surface toolset. Select plne for the input raster, specify plne_slope for the output raster, select PERCENT_RISE for the output measurement, and click OK to execute the command.
  2. plne_slope is a continuous raster. You can divide plne_slope into slope classes. Double-click Reclassify in the Spatial Analyst Tools/Reclass toolset. In the Reclassify dialog, select plne_slope for the input raster and click on Classify. In the next dialog, change the number of classes to 5 (you will need to change the classification method to equal interval, then change the number of classes to 5, and then change the classification method back to manual), enter 10, 20, 30, and 40 as the first four break values, and click OK. Save the output raster as slope_reclass in the Reclassify dialog, before running the command. Get a screen shot of the slope_reclass map and paste it into your report. Label the map as “Reclassified Slope of plne.”

 

Derive an aspect layer

  1. Double-click the Aspect tool in the Spatial Analyst Tools/Surface toolset. Select plne for the input raster, specify plne_aspect for the output raster, and click OK.
  2. plne_aspect shows an aspect layer with the eight principal directions and flat area. But it is actually a continuous raster. You can verify the statement by checking the layer properties. To create an aspect raster with the eight principal directions, you need to reclassify plne_aspect.
  3. Double-click the Reclassify tool in the Spatial Analyst Tools/Reclass toolset. Select plne_aspect for the input raster and click on Classify. In the Classification dialog, make sure that the number of classes is 10. Then click the first cell under Break Values and enter -1. Enter 22.5, 67.5, 112.5, 157.5, 202.5, 247.5, 292.5, 337.5, and 360 in the following nine cells. Click OK to dismiss the Classification dialog.
  4. The old values in the Reclassify dialog are now updated with the break values you have entered. Now you have to change the new values. Click the first cell under new values and enter -1. Click and enter 1, 2, 3, 4, 5, 6, 7, 8, and 1 in the following 9 cells. The last cell has the value of 1 because the cell (337.5 to 360°) and the second cell (-1 ° to 22.5°) make up the north aspect. Change the name of the output raster to aspect_rec before running the command. aspect_rec is an integer aspect raster with the eight principal directions and flat (-1). Get a screen shot of the aspect_rec map and paste it into your report. Label the map as “Reclassified Aspect of plne.”

 

Derive a surface curvature layer

  1. Double-click the Curvature tool in the Spatial Analyst Tools/Surface toolset. Select plne for the input raster, specify plne_curv for the output raster, and click OK.
  2. A positive cell value in plne_curv indicates that the surface at the cell location is upwardly convex. A negative cell value indicates that the surface at the cell location is upwardly concave. The ArcGIS Desktop Help further suggests that the curvature output value should be within -0.5 to 0.5 in a hilly area and within -4 to 4 in rugged mountains. The elevation data set plne is from the Priest Lake area in North Idaho, a mountainous area with steep terrain. Therefore, it is no surprise that plne_curv has cell values ranging from -6.89 to 6.33.
  3. Right-click plne_curv and select Properties. On the Symbology tab, change the show type to Classified and then click on Classify. In the Classification dialog, first select 6 for the number of classes and then enter the following break values: -4, -0.5, 0, 0.5, 4, and 6.34. Return to the Properties dialog, select a diverging color ramp (e.g., red-to-yellow-to-green), and click OK.
  4. Through the color symbols, you can now tell upwardly convex cells in plne_curv from upwardly concave cells. Add streams.shp to ArcMap, if necessary. Check cells along the stream channels. Many of these cells should have symbols indicating upwardly concave. Get a screen shot of the plne_curv map and paste it into your report. Label the map as “Curvature of plne.”

 

Questions

  1. What is the range of percent slope values in plne_slope?
  2. The value with the largest number of cells on aspect_rec is -1. Please speculate what these cells were. (Paste the DEM derivative maps here. You can put them on the same page.)

 

Task 3. Perform Viewshed Analysis

 

Instructions

You will use the lookout.shp, a lookout point shapefile, to run a viewshed analysis. The lookout point shapefile contains a viewpoint. You will need the hillshade map of plne to better visualize the terrain. You first run the viewshed analysis without specifying any parameter value. Then you add 15 meters to the height of the viewpoint to increase the viewshed coverage.

 

  1. Right-click Hillshade of plne and select Properties. On the Display tab, enter 30% transparent. The 30% transparency allows Hillshade of plne to be superimposed with other layers.

 

  1. Now run a viewshed analysis. Click the Spatial Analyst dropdown arrow, point to Surface Analysis, and select Viewshed. Make sure that the input surface is plne and the observer point is from lookout. Opt for a temporary output raster. Click OK to run the viewshed analysis.

 

  1. Viewshed of lookout separates visible areas from not visible areas. Open the attribute table of Viewshed of lookout. The table shows the cell counts for the visibility classes of 0 (not visible) and 1 (visible).

 

  1. Suppose the viewpoint has a physical structure that adds a height of 15 meters. You can use the field of OFFSETA to include this height in viewshed analysis. Open ArcToolbox. Double-click the Add Field tool in the Data Management Tools/Fields toolset. Select lookout for the input table, enter OFFSETA for the field name, and click OK. Double-click the Calculate Field tool in the Data Management Tools/Fields toolset. Select lookout for the input table, select OFFSETA for the field name, enter 15 for the expression, and click OK. Open the attribute table of lookout to make sure that the offset is set up correctly.

 

  1. Follow Step 2 to run another viewshed analysis with the added height of 15 meters to the viewpoint. The result should show an increase of visible areas.

 

Questions

  1. Insert maps showing the visible areas from viewpoints with and without added height.
  2. What area percentage of plne is visible from the viewpoint?
  3. What area percentage of plne is visible from the viewpoint with the added height?

 

 

Task 4. Create a New Lookout Shapefile for Viewshed Analysis

 

Instructions

You will digitize one more lookout location before running a view shed analysis. The output from the analysis represents a cumulative viewshed.

 

1.      Select Copy from the context menu of lookout in the table of contents. Select Paste Layer(s) from the context menu. The copied shapefile is also named lookout. Right-click the top lookout, and select Properties. On the General tab, change the layer name from lookout to newpoints.

 

2.      Make sure that the Editor toolbar is available. Click Editor's dropdown arrow and select Start Editing. The task is to Create New Feature and the target is newpoints.

 

3.      Next add a new viewpoint. To find suitable viewpoint locations, you can use Hillshade of plne as a guide and the Zoom In tool for close-up looks. You can also use plne and the Identify tool to find elevation data. When you are ready to add a viewpoint, click the Sketch Tool first and then click the intended location of the point. The new viewpoint should be complementary to the other point and be located on a vantage point. The new viewpoint has an OFFSETA value of 0. Open the attribute table of newpoints. Click the cell of OFFSETA for the new point and enter 15. Also, to separate the two viewpoints, enter the ID values of 1 and 2 respectively. Click the Editor menu and select Stop Editing. Save the edits. You are ready to use newpoints for view shed analysis.

 

4.      Click the Spatial Analyst dropdown arrow, point to Surface Analysis, and select Viewshed. Make sure that the input surface is plne and the observer points are from newpoints. Opt for a temporary output raster. Click OK to run the operation.

 

5.      Viewshed of new points shows visible and not visible areas. The visible areas represent the cumulative viewshed. Portions of the view shed are visible to only one viewpoint, whereas others are visible to both viewpoints. The attribute table of Viewshed of new points provides the cell counts of visible from one point and visible from two points.

 

6.      To save newpoints as a shapefile, right-click newpoints, point to Data, and select Export Data. In the Export Data dialog, specify the path and name of the output shapefile.

 

Questions

  1. Insert a map showing the visible areas from one or both locations in the newpoints layer. Make sure your map is clearly labeled.
  2. What area percentage of plne is visible from newpoints? Report the increase in viewshed from one to two viewpoints?

 

 

Task 5. Delineate Watersheds

 

Instructions

You will use a new data set for this task. What you need: emidalat, an elevation raster; and emidastrm.shp, a stream shapefile. You will delineate areawide watersheds using a “filled (or depressionless)” elevation raster, which is converted from a DEM, as the data source. emidastrm.shp serves as a reference. ArcToolbox has a Hydrology toolset in the Spatial Analyst toolbox that has tools for completing this Task. This exercise shows you only the standard procedures for watershed delineation. You can go to online to download other enhanced watershed delineation procedures mentioned in Baker et al. 2006 (this week’s reading). For example, the AGREE algorithm is available as an AML at the Center for Research in Water Resources, The University of Texas at Austin.

 

1.      Insert a new data frame in ArcMap. Rename the new data frame Watershed, and add emidalat and emidastrm.shp to the new frame. If necessary, click Show/Hide ArcToolbox Window to open the ArcToolbox window.

 

2.      First check if there are any sinks in emidalat. Sinks may result in an erroneous watershed delineation. Double-click the Flow Direction tool in the Spatial Analyst Tools/Hydrology toolset. Select emidalat for the input surface raster, enter temp_flowd for the output flow direction raster, and click OK. Double-click the Sink tool. Select temp _flowd for the input flow direction raster, specify sinks for the output raster, and click OK. Inspect the sinks layer and use the online help to figure out what its cell values represent. Answer Question 1 below.

 

3.      This step fills the sinks in emidalat. Double-click the Fill tool. Select emidalat for the input surface raster, specify emidafill for the output surface raster, and click OK.

 

4.      You will use emidafill for the rest of Task 5. Double-click the Flow Direction tool. Select emidafill for the input surface raster, and specify flowdirection for the output flow direction raster. Run the command. Inspect the flowdirection layer and use the online help to figure out what its cell values represent. Answer Question 2 below.

 

5.      Next create a flow accumulation raster. Double-click the Flow Accumulation tool. Select flowdirection for the input flow direction raster, enter flowaccumu for the output accumulation raster, and click OK. Inspect the flowaccumu layer and use the online help to figure out what its cell values represent. Answer Question 3 below.

 

6.      Next create a source raster, which will be used as the input layer for watershed delineation. Creating a source raster involves two steps. First select from (or threshold) flowaccumu those cells that have more than 500 cells flowing into them. Double-click the Con tool in the Spatial Analyst Tools/Conditional toolset. Select flowaccumu for the input conditional raster, enter 1 for the constant value, specify net for the output raster, and enter Value > 500 for the expression. (You can also click the SQL button to set the expression. Make sure to have a space before and after >.) Run the command. Second, we need to assign a unique value to each section of net between junctions (intersections). Go back to the Hydrology toolset. Double-click the Stream Link tool. Select net for the input stream raster, select flowdirection for the input flow direction raster, and specify source for the output raster. Run the command. Use unique values as the symbology of source so that you can see individual stream links.

 

7.      Now you have the necessary inputs for watershed delineation. Double-click the Watershed tool. Select flowdirection for the input flow direction raster, select source for the input raster, specify watershed for the output raster, and click OK. Change the symbology of watershed to that of unique values so that you can see individual watersheds. Answer Question 4 below.

 

Questions

1.      How many sinks does emidalat have? Describe where these sinks are located.

2.      If a cell in flowdirection has a value of 64, what is the cell's flow direction?

3.      What is the range of cell values in flowaccumu?

4.      How many watersheds are in watershed?

5.      If the flow accumulation threshold were changed from 500 to 1000, would it increase, or decrease, the number of watersheds?

 

 

Task 6. Derive Upstream Contributing Areas at Pour Points

 

Instructions

In this task, you will derive a specific watershed (i.e., upstream contributing area) for each point in pourpoints.shp. As of ArcGIS, Spatial Analyst and ArcToolbox do not share the same data analysis environment. Therefore all operations in Task 6 are performed using Spatial Analyst. You may not get the same result if you mix operations in Spatial Analyst and ArcToolbox.

 

1.      Add pourpoints.shp to ArcMap.

 

2.      First convert pourpoints to a raster. Select Options from the Spatial Analyst dropdown menu. On the General tab, select your lab5 folder for the working directory. On the Extent tab, select flowaccumu for the analysis extent. On the Cell Size tab, select flowaccumu for the analysis cell size. Click the Spatial Analyst dropdown arrow, point to Convert, and select Feature to Raster. Select pourpoints for the input features, enter pourptgd for the output raster, and click OK to convert.

 

3.      Select Raster Calculator from the Spatial Analyst dropdown menu. Enter the following expression in the expression box:

watershed([flowdirection], [pourptgd])

(pourptgd may appear as pourptgd-pourptgd; ignore the discrepancy.) Click Evaluate. The command creates Calculation as a temporary raster. Inspect the Calculation layer and its attribute table. Answer Question 1 below.

 

4.      Zoom in on a pour point. The pour point is not right on the stream link raster created in previous task. It is the same with the other points. This is why the pour points generated very small watersheds in Step 3. ArcGIS has a SnapPour command, which can snap a pour point to the cell with the highest flow accumulation value within a search distance. Use the Measure tool to measure the distance between the pour point and the nearby stream segment. A snap distance of 90 meters (3 cells) should place the pour points onto the stream channel.

 

5.      Select Raster Calculator from the Spatial Analyst dropdown menu. Enter the following expression in the expression box:

snappour([pourptgd], [f1owaccumu), 90)

Click Evaluate. The command creates Calculation2 as a temporary grid.

 

6.      Run the Watershed command again. Enter the following expression in the Raster Calculator's expression box:

watershed([f1owdirection], [Calculation2])

Click Evaluate. Calculation3 should have many more cells for each snapped pour point. Inspect the Calculation3 layer and its attribute table. Answer Question 2 below.

 

Questions

1.      How many cells are associated with each of the original pour points?

2.      How many cells are associated with each of the new pour points?