A Tutorial for Viewshed Delineation Using SAGA and LIDAR Data

Developed by Vern W. Cimmery

May 2016
Updated December 2018

I. Overview
This tutorial is about the use of LIDAR derived elevation and vegetation height data to support
viewshed analysis. Viewshed analysis is a technique for delineating the areas of the landscape
visible from one or more observation points of interest.

There are two general strategies guiding the placement of observation points. One is to define
observation points for sensitive features. For example, a recreation planner wants to find out
what can be seen from popular hiking trails, campgrounds, scenic highways in an area. Is the
transmission power line on the valley side visible, and, if so, how visible? A second strategy is to
establish observation points related to a specific project, e.g., a transmission corridor or land fill.
What is the viewshed for the project? Are there visually sensitive features that fall within the
projects seen-area?

LIDAR data provides an additional component for viewshed analysis, top surface elevations that
can be used to assess the screening affect of vegetation or buildings. Vegetation screening as a
viewshed factor has been cost prohibitive until the availability of low-cost LIDAR data. Large-
scale projects related to commercial development could contract for blocking data to be
compiled using photogrammetric techniques and tools. However, small-scale projects, such as
county land-use planning or corridor-level impact assessment, generally were conducted
including viewshed analysis on a worst-case qualifier. Since it is usually infeasible to collect and
include blocking data representing vegetation and man-made structures (due to the high-cost),
viewsheds are calculated using DEM's that do not account for vegetation. The delineated
viewsheds that result are worst-case situation where no view screening is considered.

This tutorial presents three hypothetical scenarios involving the delineation of viewsheds.

1. The first scenario involves a proposed addition to the Sequim Bay State Park located on the
western shore of Sequim Bay just east of the town of Sequim, Washington. A set of observation
points are defined on a recreation trail as it meanders through the proposed addition to the park
and in the park itself. Given the existing vegetation, how much of Highway 101 is visible from
the trail? How much of Sequim Bay is visible from the trail? If park management plants
strategically located tree vegetation, can the amount of Highway 101 visible be reduced over
time? Can vegetation be removed to encourage the visibility of the Sequim Bay water area to the
east?

2. The second scenario involves comparing the before and after viewsheds related to a timber
harvest clear-cut. An option for the planned harvest is to leave a thin strip of vegetation along the
western and northern boundaries of the harvest area. A set of observer points are located along
Wood Road. This road parallels the western edge of the area between 200 and 400 feet from the
area. Another set of observation points are defined along Highway 101 that roughly parallels the

1
northern boundary of the clearcut about 1/4 mile away. The Discovery Bay Trail follows along
the Highway 101 right-of-way. A set of points is located along the recreation trail, also. How
much of the post-harvest clear-cut area is visible from the several sets of observer points? If the
thin strip of vegetation along the western and northern boundaries of the harvest area is harvested
as part of the planned timber sale, how much of the post-harvest area can be viewed?

3. Delineating the accumulative viewshed for an 8 mile portion of a major recreation hiking trail.
Observation points are defined every 250 feet along the trail. This stretch of the Discovery Bay
Trail (DBT) has one beginning point 2 or 3 miles east of Sequim Bay, passes around the south
end of the Bay, and has another ending point about 1 mile west of the Bay. The (hypothetical)
"Support the Trail Action Committee" (STAC) addresses a set of visual issues related to open
space land use, timber harvesting, and land development.

Here is an alphabetical list of the SAGA tools applied in this tutorial.

Grid - Calculus/Grid Calculator

Grid - Gridding/Shapes to Grid
Grid - Tools/Clip Grids [interactive]
Grid - Tools/Mirror Grid
Grid - Tools/Mosaicking
Grid - Tools/Reclassify Grid Values
Import/Export/ESRI E00/Import ESRI E00 File
Projection - Proj4/Coordinate Transformation (Grid List)
Projection - Proj4/Set Coordinate Reference System
Shapes - Grid Tools/Clip Grid with Polygon
Shapes - Lines/Line Dissolve
Shapes - Lines/Line Properties
Shapes - Points/Convert Lines to Points
Shapes - Polygons/Polygon Clipping
Shapes - Tools/Copy Selection to New Shapes Layer
Shapes - Tools/Create New Shapes Layer
Shapes - Tools/Merge Layers
Shapes - Tools/Select by Attributes... (String Expression)
Shapes - Tools/Shapes Buffer
Shapes - Tools/Split Table/Shapes by Attribute
Spatial and Geostatistics - Grids/Zonal Grid Statistics
Table - Calculus/Field Calculator (Shapes)
Terrain Analysis - Lighting, Visibility/Visibility (points)
Terrain Analysis - Profiles/Profiles from Lines

The tutorial is developed using the System for Automated Geoscientific Analysis (SAGA),
version 5.0.0, 64-bit operating on a laptop. As you go through this tutorial, if you encounter
mistakes please let me know. My e-mail contact address is kapcimmery at hotmail dot com.

The following guides may be useful to understanding SAGA and are available in the download
section at the SAGA website sourceforge.net/projects/saga-gis/Files:

2
Cimmery, Vern (2010). User Guide for SAGA(Version 2.0.5), Vol. 1. 393 pp.

Cimmery, Vern (2010). User Guide for SAGA(Version 2.0.5), Vol. 2. 336 pp.

Cimmery, Vern (2015). SAGA: Basic Table and Vector Support. 275 pp.

Cimmery, Vern (2018). User Guide for SAGA (Version 5.0.0), 506 pp.

Conventions
Input, output, and option tool parameters are one of two types: mandatory or optional. A
parameter including "<<" or ">>" in the field is mandatory. A parameter that includes "<" or ">"
is optional. Rather than repeat this information when describing tool parameters I am providing
this explanation at the beginning.

Many parameters are what I call "toggle check box" parameters. That is, the value field to their
right contains a check box used to toggle between an on and off condition. Move your mouse
pointer into the check box and press the left mouse button and it toggles the status of the box
between on and off. A check mark in the box indicates, normally, that the parameter is toggled
on while the lack of a check mark indicates the parameter is toggled off. Again, rather than
repeat this explanation every time a check box parameter is encountered, I will most often
indicate that it is a toggle check box type parameter and assume the reader understands how they
operate. I may occasionally repeat the explanation as a reminder.

Grid maps are members of a specific grid system defined by a spatial extent and cell size. All of
the grid maps within the same grid system have the same grid cell size and spatial extent. When
a grid map is brought into a SAGA work session, if the grid system it is a member did not
already exist for the session, it is now listed along with any other loaded grid systems. Tools
involving grid maps generally have an input parameter for grid system that must be chosen first
before a member grid map can be selected for a grid input parameter. In this tutorial, a specific
grid system is defined for the overall project area and referred to as the Sequim Bay project area
grid system. A sub-area project area grid system is defined for each of the three tutorial
scenarios.

Many tools are executed using an 'Okay' button that is part of the tool parameters window. Also,
there are many dialog windows that include buttons like 'Okay', 'Yes', and 'No. You move the
mouse pointer over the 'Okay' button and press the left mouse button. The tool then executes
using the current parameter settings or returns a response to the GUI. This explanation may not
be repeated for every tool description or dialog window. Sometimes I will refer to clicking on the
'Okay' button to execute the tool as a reference to this more detailed procedure. This same
convention is true for 'Yes' and 'No' buttons.

The default for mandatory output parameters is often "<create>". The tool produces the expected
data layer or table output. The name of the output may follow from the name of the input data
layer, a combination of the input data layer name plus the tool process, or from the parameter
name, for example "Result" or "Filter". Another option that is often available on the drop-down

3
list is "<not set>". The tool executes but does not produce an output. When an existing data layer
is chosen for the output parameter from the drop-down list of options, the output from the tool
overwrites the current data values of the chosen data layer for output.

The coordinate system used for the project area of this tutorial is the Washington State Plane
Coordinate System, south zone.

Every grid map has a no data value. The most common no data value is -99999 followed by -1
and 0. Often, grid cells outside of an irregular shaped project area but within the spatial extent of
the grid system contain the no data value. In other cases, the grid cells for features of a grid map,
for example, roads, will contain valid data values and all other grid cells (i.e., non-road grid
cells) of the map contain a no data value. If you have a question about what the no data value is
for a grid map you can display the settings for the grid map in the 'Object Properties' window.
Navigate to the 'Data' tab area of the Manager window and click on the grid map of interest to
make it active. The entries for the settings of the active grid map are now displayed in the
'Settings' tab area of the 'Object Properties' window. Near the top of the display, in the "General"
section, is the setting "No Data". The no data value for the grid map displays in the value field to
the right of this setting.

The no data value used in a grid map is stored as part of the grid map. When you load the grid
map into a work session, the no data value is part of the data included. SAGA tools that create
grid map output have a no data value used for the output as part of the tool program. It is
possible, depending on a tool option, that one option supports the use of -99999 for an output
grid map and another option supports the use of 0 for an output grid map. For example, the
'Output Values' parameter of the Grid - Gridding/Shapes to Grid tool supports the 0 no data
value for the "data / no data" option and -99999 for the "index number" and "attribute" options.

II. Background
The three scenarios explored in this tutorial take place in Clallam County, Washington. Figure II-
1 displays the location of this county.

4
 Figure II-1. The location of Clallam County, Washington, USA.

Clallam County is located in western Washington state on the Olympic Peninsula. The three
scenarios focus on a small area on the eastern edge of the county around Sequim Bay.

5
 Figure II-2. The location of the three tutorial project areas.

Figure II-2(a) displays a portion of eastern Clallam County and the relative locations of the three
scenarios for the tutorial. The location of a portion of the Discovery Bay Trail (displayed in
green) is shown going east-west around the southern tip of Sequim Bay.

Figure II-2(b) displays the proposed addition (red cross-hatching) to the Sequim Bay State Park
(filled in red). The state park is about 90 acres in size and the addition is about 43 acres. Notice
that the Discovery Bay Trail (in green) and Highway 101 (the black line) go through the park
and the addition.

The yellow filled area in Figure II-2(c) is a planned timber harvest area. I am going to call the
timber harvest area the "Creek Timber Sale". In the text it is often be referred to as the TS. This
area is about 25.6 acres in size. You can see that it is just to the south of the bottom of Sequim
Bay and south of both the Discovery Bay Trail and Highway 101. The road immediately to the
west of the area is Wood Road.

6
Figure II-2(d) provides a slightly zoomed in area of the portion of the Discovery Bay Trail
featured in the third section of this tutorial. The Discovery Bay Trail is a planned non-motorized,
multi-user transportation/recreational trail system that spans over 120 miles beginning in Port
Townsend, Washington. It will pass through the cities of Sequim, Port Angeles, Forks and
terminate on the Pacific coast near LaPush, Washington. The portion that is the subject of this
scenario is 8.3 miles in length.

The three scenarios used to illustrate how SAGA can be applied for viewshed delineation are
hypothetical. The Sequim Bay State Park does exist but there is not a proposed addition to the
park. Much of the land area around the park is in residential land use. The Creek Timber Sale
area is hypothetical. The planned harvest area is within land designated as public timberland.
This scenario has been created only for the purpose of illustrating a role for viewshed delineation
in a hypothetical timber sale process. And last, the 8.3 mile length of hiking trail is a part of the
much longer 120 mile Discovery Bay Trail. This trail exists and it is quite popular. The trail
takes advantage of highway right-of-ways, an abandoned railroad bed, and new trail
construction. A delineated viewshed, if it is not already available, should be an integral part of
visual management for the trail and a tool for evaluating impacts and issues.

Viewshed Delineation
The delineation of a viewshed requires two inputs. One is an observation point. The second is an
elevation model. An elevation model, at a minimum, provides elevations for the ground or
topography.

In very general terms, a viewshed is that portion of the landscape that can be seen from an
observation point. A viewshed can also be considered as two distinct inter-mixed areas, areas
that can be seen and areas that are blocked from being seen.

Seen areas of the viewshed are not blocked from an observation point view by topographic
features (like hills and mountains). The analysis can be enhanced by incorporating vegetation
height and man-made structures as add-ons to ground elevations where they occur. This means
that in addition to topographic features blocking a view, vegetation and buildings can also block
a view.

Historically, viewshed delineation that relies on an elevation model for the ground or
topography, is often characterized as identifying a "worst-case" delineation. Unless the
application area is in an extremely arid un-populated region, vegetation and/or buildings exist
between the observation point and the landscape that could block areas of the landscape from
view from the observation point. If this data is not included, thus the reference to a "worst-case"
delineation because some areas identified as seen are actually blocked from being seen by
existing vegetation or buildings.

As you will see in this tutorial, LIDAR can provide height data for vegetation and man-made
structures (in addition to elevation for topography or ground). This allows for more accurate
delineation of viewsheds when this data is available. One must remember, however, that unlike
topography that changes on a geologic time-clock, vegetation can experience significant change
from forest to crop conversion, timber-harvest, fire, etc., over short time-periods.

7
Introduction to the LIDAR Data
LIDAR is an acronym for Light Detection and Ranging. A laser sends out a beam of light pulses.
Knowing the speed of light, the return time back to the sensor for the reflected pulses is used to
calculate the distance the pulses have traveled. The time is also used to determine whether the
pulse is ground reflected or non-ground (vegetation canopy, e.g.) reflected. The two common
sensor products produced by LIDAR sensing are a "top surface" and "bare earth" elevation or
digital elevation models. This is a simplification of the LIDAR product, as different height levels
vegetation strata are sensed.

"Top surface" is focused on returns from vegetation canopy and man-made structures. "Bare
earth" consists of pulses passing through canopy cover and reflected from the ground surface.
The elevation and canopy data values and location data for the two products are collected
simultaneously.

The Sequim Bay tutorial area includes a portion of the large water area of Sequim Bay, a
saltwater bay indentation along the Pacific Ocean extension of the Strait of Juan de Fuca. There
are also several fresh-water lakes and ponds in the area. LIDAR pulses tend to be absorbed by
water and result with weak or no return when they hit water.

Preparing the LIDAR Data Files

The LIDAR data files used in this tutorial are available from the Puget Sound LIDAR
Consortium (PSLC) website. The PSLC is an informal group of local agency staff and Federal
research scientists devoted to developing public-domain high-resolution LIDAR topography and
derivative products for the Puget Sound region. The organization was formed in the fall of 1999,
with initial participants from Kitsap County, Kitsap PUD, City of Seattle, Puget Sound Regional
Council, NASA, and the USGS. The PSLC is a major facilitator of the use of LIDAR data in the
state of Washington. The reference for the LIDAR files used in this tutorial is:

LiDAR Bare Earth/Top Surface DEMs (computer files). 2001. The Woodlands, TX: Terrapoint.
Available: Puget Sound LiDAR Consortium, Seattle, WA.
http://pugetsoundlidar.ess.washington.edu/index.htm

The LIDAR data is available on a quarter quadrangle basis. A quarter quadrangle is 1/4 of a 7.5-
minute quadrangle and, at this latitude, is about 12.5 square miles in size.

I used the location of the 8.3 mile length of the Discovery Bay Trail (Figure II-2(d) to identify
the quarter quadrangles needing downloaded. There are six: q48123a12, q48122a81, q48123a14,
q48122a83, q47123h12, and q47122h81. There are two .e00 format files per quarter quadrangle;
one is for bare earth and the second one is for top surface.

The downloaded files are in the .zip format to optimize storage space and facilitate downloading.
The files are unzipped resulting in 12 .e00 files accessible by the SAGA Import/Export/ESRI
E00/Import ESRI E00 File tool. The tool imports the unzipped .e00 files into SAGA converting
them to the SAGA grid map format. Figure II-3 displays the parameters window for the
Import/Export/ESRI E00/Import ESRI E00 File tool.

8
Figure II-3. The parameters window for the Import/Export/ESRI E00/Import ESRI E00 File
tool.

The entry in the value field to the right of the 'File' parameter is a list of six bare earth .e00 files.
They are saved in a storage folder different from the one used for the six top surface files. The
six bare earth files are imported with a single execution of the tool into SAGA. A second
execution of this tool imports the .e00 files for top surface.

After importing the .e00 files I review the new grid maps. When I display the first one I find that
for some reason an extremely large no data value (and minimum data value) has been
interpreted. When I display a data layer it displays as one solid color due to a large data range.
Realizing the large negative no data/minimum value is an issue, I go to the 'Settings' tab in the
'Object Properties' window for the data file and change the 'Minimum' from the large negative
value to 0 and click with the mouse pointer on the 'Apply' button at the bottom of the window.
This corrected the display as the color display now looks as expected. The large negative no
data/minimum value no longer affects the data display.

The .e00 files imported with a north down orientation. The Grid - Tools/Mirror Grid tool (using
"vertically" for the 'Method' parameter) is used to convert them to north up orientation. I save the
flipped, re-oriented grid maps using the original names prefixed with the letter "F"; for example,
'q47122h81be' and 'q47123h12be' are saved as 'Fq47122h81be' and 'Fq47123h12be', etc.

A third issue resulted possibly due to how the no data value in the .e00 file is interpreted by
SAGA. It looks like SAGA interprets the no data value as a data value. I would like the no data
value to be -99999. An extremely large no data value is being used and it appears there are some
extraneous very large negative data values causing gaps to exist when adjacent quarter
quadrangles are viewed together. These extraneous data values exist outside of valid data values
but within the spatial extent of the data layer. The Grid - Tools/Reclassify Grid Values tool is
used to fix this issue. Figure II-4 displays the parameters window for my work-around.

9
Figure II-4. The Grid - Tools/Reclassify Grid Values tool work-around for large negative
data values.

The invalid negative values within the range of the 'minimum value' and 'maximum value'
parameters are recoded to the no data value -99999. This reclassification is applied to each of the
twelve imported quarter quadrangle files. The new reclassified grid maps have their original
names prefixed with the letters "FR"; for example, 'FRq47122h81be', 'FRq47123h12be', etc.

Figure II-5 displays an example of six adjacent quarter quadrangles before the work-around for
the no data value and after.

10
Figure II-5. Comparing six quarter quadrangles before and after applying the no data
work-around.

Gaps between adjacent grid maps is easily discerned in the left map view window. Applying the
described work around resulted in the map view window on the right. You can also notice color
miss-matching between quadrangles. This is due to different color display settings for each
quarter-quadrangle and their data ranges.

Two mosaics are created using the Grid - Tools/Mosaicking tool. One mosaic consists of the six
bare earth files and the second one is for the six top surface files. The Grid - Tools/Mosaicking
tool parameters window used to create the bare earth mosaic is displayed in Figure II-6.

11
Figure II-6. The parameters entries for the Grid - Tools/Mosaicking tool to create a bare
earth mosaic.

The six quarter-quadrangle bare earth grid maps are chosen for the '>> Input Grids' input
parameter. Based on the spatial extents for the input grid maps, the spatial definition of the
output mosaic is displayed in the 'Target Grid System' section. The "user defined" option for the
'Target Grid System' is used. This means the output of the tool defines a new grid system using
the spatial extent and cell size of the mosaic grid map. Notice that the entry for the 'Cellsize'
parameter is 6. This is the original cell size resolution of the input bare earth grid maps. The bare
earth mosaic is named 'QNbeElev1'. This output file name is entered for the 'Name' parameter.

The Grid - Tools/Mosaicking tool is executed a second time with the six top surface grid maps.
This time the entry for the 'Target Grid System' parameter is "grid or grid system". The output
mosaic is created using the already defined grid system from the first execution of the tool and
cell size for the 'QNbeElev1' mosaic grid map. Thus, the top surface mosaic (named 'QNtsVeg1')
becomes a part of the same grid system as the 'QNbeElev1' grid map.

Reviewing the 'QNtsVeg1' and 'QNbeElev1' grid maps I notice several water areas with the no
data value (-99999). These returns result from the way LIDAR is absorbed and reflected from
water. I replace these no data areas after applying a coordinate transformation.

The LIDAR quarter-quadrangle files imported to SAGA use the Washington State Plane
Coordinate System (WSPC) north zone, FIPS 4601. The dataset in this tutorial uses a similar
coordinate system except it is the one for the south zone, FIPS 4602. The coordinate reference
system information used for the bare earth and top surface mosaics has not been included as part
of the description of the two mosaic layers. The Projection - Proj4/Set Coordinate Reference
System tool is used to link the coordinate system information to the 'QNbeElev1' and 'QNtsVeg1'

12
grid maps. The Projection - Proj4/Coordinate Transformation (Grid List) tool uses this linked
descriptive information as input to transform the coordinate system from the north zone to the
south zone version.

The parameter window for the Projection - Proj4/Set Coordinate Reference System tool is
displayed in Figure II-7.

Figure II-7. The Projection - Proj4/Set Coordinate Reference System tool parameters
window.

The two grid layers for bare earth and top surface are chosen for the input parameter '> Grids'. A
list of the available coordinate systems for the 'Projected Coordinate Systems' parameter is
viewed by moving the mouse pointer into the value field to the right of the parameter name.
When I press the left mouse button a drop down, scrollable list of coordinate systems displays.
The "NAD83(HARN) / Washington North (ftUS)" system is chosen on this list. When the tool
executes, the characteristics of the WSPCS north zone coordinate system become available to be
displayed as part of the 'Description' tab information in the 'Object Properties' windows for the
data layers and also available to provide variables input for other georeferencing tools.

The Projection - Proj4/Coordinate Transformation (Grid List) tool is used to convert the
coordinates from the north zone to south zone WSPCS. The parameters window for this tool
(Figure II-8) is very similar to the parameters window for the Set Coordinate Reference System
tool.

13
Figure II-8. The Projection - Proj4/Coordinate Transformation (Grid List) tool parameters
window.

The "NAD83(HARN) / Washington South (ftUS)" coordinate system is chosen for the 'Projected
Coordinate Systems' parameter. The bare earth 'QNbeElev1' and top surface 'QNtsVeg1' layers
are entered for the input '>> Source' parameter. The target grid system is "user defined". This
means that the spatial boundaries for a new grid system are defined by the 8 parameters listed
below the 'Target Grid System' parameter.
14
The two output grid maps use the same names as the input maps. I re-name the output grid maps
'QSbeElev2a' and 'QStsVeg2a' and save them. The execution of this tool defines a new grid
system based on the transformed coordinates, WSPCS south zone. The new grid system is:

6; 5441x 10096y; 1012356x 967992y

A third LIDAR derived grid map, for vegetation height, may be needed. Vegetation height is the
difference between top surface and bare earth elevations. The Grid - Calculus/Grid Calculator is
the tool that creates this new layer.

Figure II-9 displays the parameters window for the Grid - Calculus/Grid Calculator tool.

Figure II-9. The parameter entries for creating the vegetation height grid map.

A simple subtraction formula, "g1-g2", is entered for the 'Formula' parameter. The two input grid
maps, 'QStsVeg2a' and 'QSbeElev2a', are entered for the input parameter '> Grids'. The "g1" and
"g2" in the formula are variables for the two input grid maps. The elevation data values for bare
earth (variable "g2", 'QSbeElev2a') are subtracted from the elevation data values for top surface
(variable "g1", 'QStsVeg2a'). The output grid map is named 'QSVegHt2a'. This name is entered
in the value field to the right of the 'Name' parameter. The mouse pointer is clicked on the 'Okay'
button and the tool executes creating the new grid map 'QSVegHt2a'. The new grid map is saved.

Figure II-10 displays three map view windows for bare earth, top surface, and vegetation height
grid maps.

15
(a) Bare earth (b) Top surface (c) Vegetation height
Figure II-10. The bare earth, top surface, and vegetation height grid maps.

The white areas (no data values of -99999) displayed in Sequim Bay on both the bare earth and
top surface layers is due to the way LIDAR reflects from water. There are more no data values in
the top surface grid map compared to the bare earth. On the top surface I also note several small
no data areas outside of Sequim Bay. These are freshwater lakes and ponds.

The spatial extent of the grid maps containing data is a result of the downloaded LIDAR .e00
layers. Before addressing the no data areas, I am going to trim away some area to the north and
the south. If you look back at Figure II-2 you can see the scenario locations for the tutorial. The
area involved for the portion of the Discovery Bay Trail (Figure II-2d) is the larger of the three.

I use the Grid - Tools/Clip Grids [interactive] tool to clip the grid map areas not needed. Before
executing the tool, the map window in Figure II-10(a) is displayed in the work area. The
parameters window for the Grid - Tools/Clip Grids [interactive] tool displays in Figure II-11.

16
 Figure II-11. The parameters window for the Grid - Tools/Clip Grids [interactive] tool.

This tool is used to clip one or more existing grid maps that are members of the same grid
system. The grid system defined by the Projection - Proj.4/Coordinate Transformation (Grid
List) tool is chosen for the 'Grid system' input parameter. This is the system the 'QSbeElev2a',
'QStsVeg2a' and 'QSVegHt2a' data layers are a part. These are the three grid maps to be clipped
and are selected for the '>> Grids' input parameter. They are the bare earth, top surface, and
vegetation height grid maps.

There is one option parameter: Run Once. This is a toggle check box parameter. If this option is
turned on (a check mark is in the toggle box), when I define a rectangle (as described below), the
tool execution stops. When this option is turned off the tool remains running and a new grid
system and clipped grid maps are defined each time the user defines a new rectangle in the map
view window.

Tool execution begins when the mouse pointer is clicked on the 'Okay' button in the parameters
window.

This is an interactive tool. I use the 'Action' tool from the toolbar to interact with the tool while it
is executing. The 'Action' tool is applied in the map view window displayed in the work area. I
am using the map view window displayed in Figure II-10(a). The grid map displayed in the map
view window is used for orientation and to recognize where to define the rectangle. A
rectangular area is defined by moving the mouse pointer to a corner of the desired rectangle in
the map view window. Keeping the mouse button depressed, the mouse pointer is dragged to an
opposite corner, and the mouse button is let up. This highlights and defines a rectangular area
that becomes the spatial extent defining a sub-area of the active grid system. The 'Clip to Extent'
dialog window in Figure II-12 displays.

17
 Figure II-12. The 'Clip to Extent' dialog window.

The four parameters 'Left', 'Right', 'Bottom', and 'Top' define the four sides of the interactively
defined rectangle. The 'Columns' and 'Rows' parameters display the number of columns and rows
definition using the original grid cell size of 6'. I accept the definition by clicking the mouse
pointer on the 'Okay' button.

A new grid system definition, using the parameters, is defined and appears in the "Grids" section
of the 'Data' tab area of the Manager window.

6; 5018x 6144y; 1012356x 977274y

The three input grid maps have been clipped by the new grid system definition and are listed as
members of the new grid system in the 'Data' tab area of the Manager window.

The interactive tool is still running. I quit the tool by moving the mouse pointer to the
'Geoprocessing' title in the Menu Bar, pressing the left mouse button. A drop-down list of
options displays. Near the bottom, the tool execution history for the work session is listed. The
names of the tools in the history list are grayed out except for the first one. The first tool listed in
this section is the "Clip Grids [interactive]" tool and a check is showing to the left of its' name.
The check indicates the tool is currently executing. I move the mouse pointer over the tool name
and press the left mouse button. A 'Tool Execution' dialog window displays with the question:
Shall execution be stopped? I click on the 'Yes' button in the window and the tool execution
stops. The check mark disappears.

18
Figure II-13 shows the three grid maps in Figure II-10 after trimming away portions of the data
to the north and south.

Figure II-13. Trimmed versions of the bare earth, top surface, and vegetation height grid
maps.

The Grid - Tools/Clip Grids [interactive] tool defines a new grid system based on the
interactively defined rectangle. Comparing the map view windows in Figures II-10 and II-13 the
differences are obvious. The portion of the existing grid maps within the rectangle is retained and
grid cells outside of the rectangle are clipped away. I re-save the three grid maps in the new grid
system as layers 'QSbeElev2', 'QStsVeg2', and 'QSVegHt2'.

A new version of the grid maps is going to be created that has the water grid cells of Sequim Bay
(including the no data layer grid cells within the bay) recoded to the sea level elevation of 0.

I have a polygon shape data layer for Sequim Bay ('SequimBay'). One of the attributes in the
attribute table linked to this data layer is "Bay". The data value for this attribute is 1.

It is an easy process to use the 'SequimBay' data layer to update the clipped versions of the three
grid maps 'QSbeElev2', 'QStsVeg2' and 'QSVegHt2'. I use the Grid - Calculus/Grid Calculator
tool. Before using this tool for this purpose, I need to create a grid map version of the
'SequimBay' polygon shapes data layer using the Grid - Gridding/Shapes to Grid tool. Figure II-
14 displays the parameters window for this tool.

19
Figure II-14. Creating the Sequim Bay grid map using the Grid - Gridding/Shapes to Grid
tool.

The 'SequimBay' data layer is chosen for the input parameter '>> Shapes'. The "attribute" option
is used for the 'Output Values' parameter. When the "attribute" option is used, the 'Attribute'
parameter is activated for the parameters window. The attribute "Bay", mentioned earlier, is
chosen for the 'Attribute' parameter. This attribute is the source of data values for grid cells of the
output grid map.

The tool creates a grid map conversion of the polygon data layer that will be part of the grid
system defined for the tutorial. This grid system is the one just created using the Clip Grids
[interactive] tool. This system is chosen for the 'Grid System' parameter. The default "<create>"
option is used for the output parameter '<< Target Grid'.

I re-name the default output 'SequimBay [Bay]' to 'SequimBayGR' and save it. The data value 1
from the attribute "Bay" is the data value for any grid cell falling within the polygon defining the
Sequim Bay. Because of how the Grid Calculator tool operates, the data value used for this
conversion is important for its' value, whether it is 0 or 1 or 99, etc. It is the only value, besides
the no data value -99999, stored for grid cells in the layer.

Figure II-15 displays the parameters window for the Grid - Calculus/Grid Calculator tool. The
parameter entries are for updating the Sequim Bay water elevations to 0 (sea level).

20
 Figure II-15. Updating Sequim Bay elevation data values to sea level.

The grid system for the tutorial is chosen for the 'Grid system' parameter. The 'SequimBayGR'
and 'QSbeElev2' grid maps are chosen for the input parameter '> Grids'. The default entry for the
'<< Result' output parameter "<create>" is accepted.

You will notice the references "g1" and "g2" used in the formula entered in the value field to the
right of the 'Formula' parameter. These references are variables in the formula for the two input
grid maps. The formula checks if the grid cell data value on the Sequim Bay grid map is 1. If the
value is 1 it means the grid cell is a cell defining Sequim Bay. The value 0, representing sea level
elevation, is output to the corresponding grid cell of the output grid map. If the grid cell is a no
data grid cell, i.e., it is outside of Sequim Bay, than the data value for the grid cell of the
'QSbeElev2' grid map is output to the corresponding grid cell of the output grid map.

Usually a no data value in any grid map produces a no data value for the corresponding grid cell
of an output grid map. The grid cells of the 'SequimBayGR' that are not Sequim Bay water
contain the no data value -99999 and would therefore produce no data grid values on the output
grid map. However, you see that a check mark exists in the toggle check box for the 'Use
NoData' parameter. This allows for the default role of the no data value to be ignored and the
data values of grid cells outside of Sequim Bay to be treated as valid data values. Thus, data
values that are outside of Sequim Bay on the bare earth grid map are output to their
corresponding grid cells on the output grid map. The name for the output grid map is entered for
the 'Name' parameter, "QSbeElev3". I save the new data layer.

I execute the Grid - Calculus/Grid Calculator tool two more times updating the Sequim Bay
elevation values for the 'QStsVeg2', and 'QSVegHt2' grid maps. The output grid maps are named
'QStsVeg3' and 'QSVegHt3' and saved.
21
Several areas of the top surface and vegetation height data layers, outside of Sequim Bay, also
contain the no data value -99999. These happen to be freshwater lakes and ponds and there
elevations are above sea level. These no data values will be replaced by data values from the bare
earth grid map.

The Grid - Calculus/Grid Calculator tool is applied with the 'QStsVeg3' (variable "g1" in the
formula) and 'QSbeElev3' (variable "g2" in the formula) grid maps for the '> Grids' input
parameter. The formula entered is "ifelse(eq(g1,-99999),g2,g1)". The two input grid maps both
contain the no data value -99999 for areas outside of the tutorial study area. Only the top surface
data layer has no data values in grid cells within the study area. So the formula replaces the no
data values falling within the tutorial study area on the top surface grid map with the elevations
for the corresponding grid cells of the bare earth grid map. The name for the output grid map
"QStsVeg4" is entered for the 'Name' parameter. I execute the tool and the new grid map is
created. I save it to a storage folder.

No changes have been made to the bare earth grid map 'QSbeElev3' but to reduce any potential
confusion I re-name it to 'QSbeElev4' and save it.

Recall that the vegetation height data layer is based on the difference in elevations between the
bare earth and the top surface layers. Due to the no data values in the 'QSVegHt3' layer, a new
updated vegetation height grid map is created using the Grid - Calculus/Grid Calculator tool.
The equation used is "g1-g2" where variable "g1" represents the 'QStsVeg4' layer and variable
"g2" the 'QSbeElev4' layer. I execute the tool and the new grid map is created. The output layer
is named 'QSVegHt4' and saved to a storage folder.

The three new layers are 'QSbeElev4', 'QStsVeg4', and 'QSVegHt4'. Each one is saved as part of
the tutorial grid system. They have been trimmed and the elevation values for Sequim Bay are
set to sea level. Elevations for freshwater lakes and ponds are set to their elevation values on the
bare earth data layer. The grid system is 6; 5018x 6144y; 1012356x 977274y. This grid system is
referred to as the Sequim Bay project area grid system.

It is of value to point out some major differences between data resulting from LIDAR sensing
and more readily available U.S. Geologic Survey (USGS) digital elevation (DEM) data. Most of
the USGS DEM's have a spatial resolution of 10 or 30 meters or more. Many of these DEM's are
compiled via a high-resolution scanning of the contour separations of topographic quadrangles.

LIDAR data, due to its sensing process, is extremely dense. This tutorial uses LIDAR data that
has been post-processed into a 6' grid cell raster format. For every 10 meter grid cell of a
standard 10-meter DEM, there are approximately 25 LIDAR elevation points. This is a huge
amount of data at a 6' resolution. This tutorial will retain this resolution level. The obvious
alternative, if analysis objectives support a more general resolution level, is to re-sample the data
to a larger grid cell level.

Rather than re-sample the data, each of the three viewshed analyses uses a subset of the Sequim
Bay grid system project area. The same grid cell size is used in each. The first processing step for

22
each scenario is to define a sub-area referred to as the project area grid system. It is a portion of
the Sequim Bay project area grid system that includes a suitable area for analysis around the
theme of the scenario, e.g., the state park and proposed addition. A second sub-area is defined for
the Creek timber sale. This sub-area is referred to as the Creek timber sale or Creek TS project
area. A third project area is defined including the 8.3 mile length of the Discovery Bay Trail.
This sub-area is referred to as the DBT project area.

Recall the three issues related to the imported .e00 data files. One was an apparent extremely
large negative value interpreted as the no data value. A second one was some invalid large
negative values perceived as valid data. A third issue was the orientation of the data files after
their initial import to SAGA. Fortunately, SAGA includes tools, functions, and tools that provide
the flexibility to correct unexpected issues. It is likely that the SAGA import tool is misreading
the no data value associated with the .e00 data file or the .e00 data format of the data file is not
representing the no data value correctly. Regardless, these three issues are corrected.

III. Vegetation management for Sequim Bay State Park.

A 43 acre addition is proposed adjacent to the northern boundary of the Sequim Bay State Park.
Figure III-1 displays a zoomed in view of the addition and the state park.

Figure III-1. A zoomed in view of Sequim Bay State Park and the proposed addition.

The area in red is the existing Sequim Bay State Park. The red cross-hatched area on its northern
boundary is the proposed addition. The blue area to the east is Sequim Bay. The green linear
feature that meanders through the park and the addition is the Discovery Bay Trail (DBT). The
heavy black line that curves through the area is Highway 101. This major highway experiences a
lot of use.

23
There are two visual impact concerns. The vegetation within the state park tends to block any
view of Sequim Bay from the DBT. One of the visual management plan objectives is to open up
views to the bay as opportunities emerge. For example, the plan suggests that strategic removal
of diseased trees be considered as well as salvage of storm-damaged trees. The viewshed for the
DBT within the addition is unknown. What is the viewshed for the DBT in the proposed
addition? Secondly, is there a visual impact to users of the DBT from Highway 101? If Highway
101 is within the seen-area of the DBT viewshed, are there vegetation management activities that
will minimize the visual impact?

Summary of Steps
1. Define a state park project area grid system.
2. Define observer points for the DBT sections in the state park and the park addition.
3. Preparing the top surface layer.
4. Preparing viewshed grid maps for the DBT sections in the state park and in the proposed park
addition.
5. Analyze the viewsheds relative to visual issues.

The first step is to define a sub-area of the much larger Sequim Bay grid system to serve as the
project area grid system for the state park and vicinity. Next, a series of observer points on the
DBT must be defined for the segment of the DBT within the park and for the segment within the
addition. Third, adjust elevations within the DBT right-of-way (ROW) as needed. Fourth, the
viewsheds for the DBT within the park and the proposed addition must be delineated. Fifth, the
viewsheds must be analyzed for potential views of Sequim Bay and opportunities to minimize
the visual impact from Highway 101.

1. Define a State Park Project Area Grid System.

The map view window in Figure III-1 displays the area that will serve as the project area for the
state park and proposed park addition. Between a half-mile and a mile of Sequim Bay east of the
park is visible. Over a half-mile of area to the west of Highway 101 is included. These
boundaries should be adequate for addressing the issues described above.

A major reason for reducing the project area to a sub-area of the much larger Sequim Bay grid
system is due to the large volume of data created by the LIDAR process. The raw data collected
by this sensor is extremely dense. Generalizing the data to a 6' grid cell resolution reduces the
volume but still results with a large amount of information. This has an impact on software and
system performance. If the detailed resolution is not necessary for an application it is best to re-
sample the data using a grid system based on a larger grid cell size that would be more
appropriate.

The interactive tool Grid - Tools/Clip Grids [interactive] is used to define a project area that is a
sub-area of the larger Sequim Bay grid system. The map view window in Figure III-1 is
displayed in the work area. These are the data layers, in order, displayed in the map view
window: ParkAddition01, Highway101, DiscBayTrl, SequimBay, ClallamWater, SequimBaySP,
and ClallamRoads.

Figure III-2 displays the parameters window for the Grid - Tools/Clip Grids [interactive] tool.

24
Figure III-2. The parameters window for the Grid - Tools/Clip Grids [interactive] tool used
to define a new project area.

This tool is used to clip one or more existing grid maps that are members of the same grid
system. The Sequim Bay grid system that was defined during the processing of the downloaded
LIDAR files is chosen for the 'Grid system' parameter. Three grid maps, members of the selected
grid system, are chosen to be clipped and are identified for the '>> Grids' input parameter. These
are the bare earth, top surface, and vegetation height grid maps.

There is one option parameter: Run Once. This is a toggle box parameter. If this option is turned
on (a check mark entered in the toggle box), once you define a rectangle (as described below),
the tool execution stops. When this option is turned off the tool remains running and a new grid
system and clipped grid maps are defined each time the user defines a new rectangle in the map
view window.

I click with the mouse pointer on the 'Okay' button in the parameters window and the tool begins
execution.

This is an interactive tool. The 'Action' tool from the toolbar is used to interact with the tool
while it is executing. The 'Action' tool is applied in the map view window (Figure III-1)
displayed in the work area. The data layers displayed in the map view window serve as
background and orientation for defining a rectangle. A rectangular area is defined by moving the
mouse pointer to a point near a desired corner. Pressing the mouse button and holding it down,
the mouse pointer is dragged to near a corner point diagonally opposite. The mouse button is
released. This highlights and defines a rectangular area that becomes the spatial extent defining a
sub-area of the Sequim Bay grid system to be the project area grid system for the state park and
park addition. The 'Clip to Extent' dialog window in Figure III-3 displays.

25
 Figure III-3. The 'Clip to Extent' dialog window for a new project area extent.

The four parameters 'Left', 'Right', 'Bottom', and 'Top' define the four sides of the interactively
defined rectangle. The 'Columns' and 'Rows' parameters display the number of columns and rows
definition using the original grid cell size of 6'. I accept the definition by clicking the mouse
pointer on the 'Okay' button.

A new grid system definition, using the parameters, is created and appears in the "Grids" section
of the 'Data' tab area of the Manager window.

6; 1375x 1009y; 1017420x 994296y

This new grid system is referred to as the state park project area grid system. The three grid maps
chosen for the input parameter '< Grids' are clipped by the interactively defined rectangle and
become members of the new grid system. The default output names for the new layers are the
same as the input names. I re-name them 'SPTu1beElev5', 'SPTu1tsVeg5', and 'SPTu1VegHt5'.

The interactive tool is still running. I quit the tool by moving the mouse pointer to the
'Geoprocessing' title in the Menu Bar, pressing the left mouse button. A drop-down list of
options displays. Near the bottom is the tool execution history for the work session. The names
of the tools in the history list are grayed out except for the first one. The first tool listed in this
section is the "Clip Grids [interactive]" tool and a check is showing to the left of its' name. The
check indicates the tool is currently executing. I move the mouse pointer over the tool name and
press the left mouse button. A 'Tool Execution' dialog window displays with the question: Shall
execution be stopped? I click on the 'Yes' button in the window and the tool execution stops. The
check mark disappears.

26
The line shapes data layer 'DiscBayTrl' contains line objects identifying the location of the DBT.
Two new line shapes data layers are needed for the portions of the DBT that lie within the park
addition and the state park. Two sets of observer points are going to be defined later, one for the
section of the DBT within the state park and a second set for the park addition. A viewshed for
the state park and a viewshed for the park addition is developed from the two sets of observer
points.

Polygon shapes data layers exist for both the state park and the proposed addition to the state
park that can be used to clip out the DBT sections. These are the 'SequimBaySP' and
'ParkAdditon01' data layers. These data layers are used as input to the Shapes -
Polygons/Polygon Clipping tool to create the two new line shapes data layers for the DBT from
the 'DiscBayTrl' line shapes data layer. Figure III-4 displays the parameters window for this tool
using the proposed park addition polygon to clip the DBT.

Figure III-4. Creating the park addition DBT line shapes data layer.

The 'ParkAddition01' data layer contains a single polygon object defining the proposed park
addition. This data layer is chosen for the input parameter '>> Clip Features'. The DBT line
shapes data layer 'DiscBayTrl' is chosen for the second input parameter '>> Input Features'. The
defaults for the two options (check marks in the toggle boxes) are accepted. I click with the
mouse pointer on the 'Okay' button to execute the tool.

The default name for the output data layer is 'DiscBayTrl [clipped]'. I re-name it 'SPtu1DBTadd'
and save it. This new layer contains a single line object.

For your information, the two options, 'Dissolve Clip Features' and 'Multiple Input Features', do
not have any effect on our output. The 'Dissolve Clip Features' would make a difference if the
clip features data layer has two or more polygons with internal common boundaries. The

27
'ParkAddition01' data layer has a single polygon object. When the option is off, the overall
coverage of two or more polygons act as a cookie cutter on the input features and internal
common boundaries define boundaries of intersection on the output data layer. When the option
is on, the internal common boundaries do not create new polygons on the output data layer. The
'ParkAddition01' layer contains a single polygon object.

The 'Multiple Input Features' option is used when two or more input polygon shapes data layers
are chosen for the '>> Input Features' parameter. The data layer chosen for the '>> Clip Features'
parameter is applied to each input features layer producing a clipped output data layer for each
input layer.

The tool is executed a second time to create a line shapes data layer for the portion of the DBT
within the state park boundary. This time the 'SequimBaySP' data layer is the input for the
parameter '>> Clip Features'. The 'DiscBayTrl' data layer is again used for the second input
parameter '>> Input Features'. I re-name the default output line shapes data layer 'SPtu1DBTsp'
and save it. This new data layer contains 6 line objects for the DBT.

Two new line shapes data layers now exist. The 'SPtu1DBTadd' layer contains a single line
object for the DBT that exists in the proposed park addition. The 'SPtu1DBTsp' layer contains
six line objects for the portion of the DBT that exists within the state park.

2. Define Observer Points for the DBT Sections in the State Park and the Park Addition.
An observer point is a point location from which areas of the landscape are identified that are in
view and not blocked by topography, man-made structures or vegetation. I want to define an
observation point every 50 feet along the DBT within the state park and proposed addition.

The Shapes - Points/Convert Lines to Points tool defines point objects at specific distance
intervals along line objects of a line shapes data layer. One of the parameters for this tool is the
insertion method. There are three choices: per line segment, per line, and from line center'.

When the "per line segment" method is chosen, the vertices defining the line objects are
interpreted as end points of line segments. A line segment is defined as a line connecting two
adjacent vertices. A point object is defined for each vertex and new point objects, within the
length of two adjacent vertices, are inserted using the insertion distance entered for the 'Insertion
Distance' parameter. In this example, the entry for the 'Insertion Distance' parameter is 50. If the
length between two adjacent vertices is greater than 50 feet but less than 100, a single new point
object is defined between the vertices 50 feet from one of them.

The second option for the 'Insertion' parameter is "per line". In this case a line object is viewed as
a single line, regardless of the placement of vertices defining line segments for the line object. A
new point object is defined at the distance interval specified by the entry for the 'Insertion
Distance'. The vertices do not define new point objects, the exception being the two end vertices
of the line object. Point objects are defined on the basis of the distance interval starting at one of
the line end points.

28
The third option is "from line center". Point objects are defined at specific distance intervals
starting at the mid-point of the line object toward each end point of the line object. The end
points are not defined as point objects on the output.

The "per line" option produces point objects (observer points) at 50 foot space intervals along the
length of the trail if it is a single line object. The portion of the trail in the 'SPtu1DBTadd' data
layer is a single line object but the portion of the trail in the 'SPtu1DBTsp' data layer consists of
six line objects. The six line objects can be logically defined to act as a single one using the
Shapes - Lines/Line Dissolve tool.

Figure III-5 displays the parameters window for the Shapes - Lines/Line Dissolve tool.

Figure III-5. The parameter entries for the Shapes - Lines/Line Dissolve tool to combine the
six polygon objects of the DBT in the state park.

The 'SPtu1DBTsp' data layer is chosen for the input parameter '>> Lines'. The "Name" attribute
in the attribute table linked to the data layer has the same entry for each line object. This is the
attribute used to combine the six objects into a single one. The default "<create>" is used for the
output parameter '<< Dissolved Lines'. The default name for the output data layer is
'SPtu1DBTsp [Dissolved]'. I re-name it 'SPtu1DBTsp2' and save it.

The two DBT line shapes data layers each now contain a single line object. Figure III-6 displays
the parameters window for the Shapes - Points/Convert Lines to Points tool.

29
 Figure III-6. Defining observation points for the state park DBT.

The parameter entries in Figure III-6 define a point object every 50' along the length of a line
object. The line shape data layer for the DBT within the state park created by the Shapes -
Lines/Line Dissolve tool, 'SPtu1DBTsp2', is chosen for the input parameter '>> Lines'. The
default "<create>" option is used for the output parameter '<< Points'. The default name for the
output point shapes data layer uses the name of the input layer.

The toggle box 'Insert Additional Points' parameter is turned on (the box contains a check mark).
When the parameter is off, the tool produces a point object for each vertex that defines the line
objects including the end point vertices of the line objects of the input data layer.

With the 'Insert Additional Points' parameter turned on, two additional parameters are displayed:
'Insertion' and 'Insertion Distance'. The 'Insertion' parameter determines how additional point
objects are generated. The three options are: per line segment, per line, and from line center.
These options were described earlier.

The second option for the 'Insertion' parameter, "per line", is chosen. In this case the line object
is viewed as a single line, regardless of the placement of vertices defining line segments defining
the line object. A new point object is defined at the distance interval specified by the entry for
the 'Insertion Distance'. The distance interval entered is 50'. The vertices do not define new point
objects, the exception being the two end vertices of the line object.

The toggle box parameter 'Add Point Order' is used when the numerical order of the point
objects in the input line object needs to be captured as an output in the attribute table linked to
the output point shapes data layer.

30
I execute the tool and save the default output point shapes data layer 'SPtu1DBTsp2' to
'SPtu1Sp2Pts' and save it.

The Shapes - Points/Convert Lines to Points tool is executed a second time with the following
parameter entries:

'>> Lines' input parameter, "SPtu1DBTadd"

'<< Points' output parameter, "<create>"
'Insert Additional Points' toggle box parameter, turned on
'Insertion' parameter, "per line"
'Insertion Distance' parameter, 50

The default 'SPtu1DBTadd' point shapes data layer is re-named 'SPtu1AddPts' and saved.

The two point shapes data layers defining observer points (point objects) for the state park and
park addition are 'SPtu1Sp2Pts' and 'SPtu1AddPts'.

When applying the Shapes - Lines/Line Dissolve and Shapes - Points/Convert Lines to Points
tools as described above, in the output you may notice a few places where the distance interval
between adjacent point objects is less than the value entered for the 'Insertion Distance'
parameter. Relative to delineating a viewshed this discrepancy is probably not an issue unless the
distance interval has a statistical purpose. Hopefully, the problem is corrected in a future release
of SAGA.

3. Preparing the Top Surface Layer.

The bare earth grid map is based on LIDAR returns interpreted as ground elevations. The top
surface grid map is created with LIDAR returns interpreted to represent a surface above the
ground, such as vegetation over-story and/or structures. The difference between bare earth and
top surface elevation for the same grid cell indicates vegetation height or the height of a
structure.

Top surface elevations are used as input to the Terrain Analysis - Lighting, Visibility/Visibility
(points) tool. The other input is a set of observation points. Each observation point is going to be
6' above the ground (roughly representing eye-level). Because the elevation input is top surface,
without making adjustments to the top surface elevation input model this means the observation
points are 6' above the vegetation or the top surface elevation. The observers need to be
positioned above the ground or bare earth and not above the vegetation height.

Before the 'SPtu1tsVeg5' grid map (top surface elevations) can serve as the source of elevation
input, the top surface elevation values within the DBT right-of-way (ROW) need to be replaced
with bare earth elevations. Otherwise, as noted above, the observer points are processed as being
above the vegetation rather than above the ground. This would not produce a viewshed that takes
into consideration the effect of vegetation.

The ROW width for the DBT is not known but a value of 6' seems reasonable. Much of the time
this trail has a paved surface but there are also lengths where the surface is hard-packed dirt or

31
gravel. The first step in replacing the top surface elevations in the DBT ROW with bare earth
elevations is to define a polygon buffer that represents the DBT ROW. The Shapes -
Tools/Shapes Buffer tool can define this polygon. The parameters window for the tool is
displayed in Figure III-7.

Figure III-7. Creating a 6' wide buffer for the DBT using the Shapes - Tools/Shapes Buffer
tool.

The line shape data layer for the DBT in the Sequim Bay project area is named 'DiscBayTrl'.
This layer is chosen for the input parameter '>> Shapes'. If the buffer width is included as an
attribute in the attribute table linked to the input data layer, the attribute can be chosen for the
'Buffer Distance' parameter. This is particularly useful if the buffer width or distance can vary for
the data layer objects. Since this is not the case with the DBT I can enter a buffer width for the
'Default' parameter. The width entered is half of the desired final polygon width as the buffer is
defined on both sides of a line object. The "<create>" option is chosen for the output parameter
'<< Buffer'. The default settings are accepted for the options parameters. I move the mouse
pointer to the 'Okay' button and click.

The tool executes creating an output polygon shapes data layer named 'DiscBayTrl [Buffer]'.
This layer is re-named to 'DiscBayTrlBuffer'. Before re-naming the layer the data value for the
"ID" attribute in the linked attribute table is changed from 0 to 1.

The 'DiscBayTrl' data layer includes much more of the DBT than what is located in the state park
and the proposed addition. This is not a problem. This shapes layer is going to be converted to a
grid map using the Grid - Gridding/Shapes to Grid tool. One of the input parameters for this tool

32
is the grid system the converted layer is to become a part. The portion of the buffer polygon,
located outside of the extent of the input grid system, is truncated by the tool.

A grid map version of the 'DiscBayTrlBuffer' is used with the Grid Calculus tool to identify grid
cells of the top surface ('SPtu1tsVeg5') layer that fall within the DBT ROW. The grid map
version is created by the Grid - Gridding/Shapes to Grid tool. The 'DiscBayTrlBuffer' shapes
data layer is selected for the input parameter '>> Shapes'. The "attribute" option for the 'Output
Values' parameter and the "ID" attribute for the 'Attribute' parameter are chosen as entries. The
project area grid system for the 'Grid System' parameter is selected from the list of available grid
systems and the default "<create>" for the output parameter '<< Target Grid' is used. The default
name for the output grid map is 'DiscBayTrlBuffer [ID]'. I re-name the layer
'DiscBayTrlBufferGR' and save it.

The Grid - Calculus/Grid Calculator tool is used to replace the elevation data values within the
DBT ROW on the 'SPtu1tsVeg5' (top surface) data layer with elevation values of corresponding
grid cells on the 'SPtu1beElev5' (bare earth) data layer. The parameters window is displayed in
Figure III-8.

Figure III-8. The Grid - Calculus/Grid Calculator tool parameters window for updating
ROW elevations.

The state park and addition area grid system is chosen for the 'Grid system' parameter. Three grid
maps are entered for the input parameter '> Grids'. They include the one for the DBT ROW (the
3' ROW on either side of the trail, a total 6' wide ROW), the bare earth DEM 'SPtu1beElev5' and
the top surface DEM 'SPtu1tsVeg5'. The order of the input grid maps must match to how the
variables are used in the equation. Note that the grid cells that are not DBT ROW grid cells

33
contain the no data value -99999. The default entry "<create>" is used for the output parameter
'<< Result'.

The equation "ifelse(eq(g1,1),g2,g3)" is entered for the 'Formula' parameter. The references "g1",
"g2", and "g3" are variables representing the input grid maps corresponding to their listing order.
The output grid map is named "SPtu1tsBuffVeg5". This name is entered for the 'Name'
parameter.

The Grid Calculator tool has two choices for dealing with no data values in grid maps. The
default is that any grid cell containing a no data value on any input grid map produces a no data
value for the corresponding grid cell of the output layer. This means that since all grid cells
outside of the DBT ROW on the 'DiscBayTrlBufferGR' grid map contain the no data value these
grid cells produce no data grid cell values for the output grid map. When the 'UseNoData'
parameter is turned on the no data value is interpreted as a valid data value by the equation and a
valid data value is output to the corresponding grid cell of the output data layer. The toggle check
box to the right of the 'Use NoData' parameter is where you make this choice. If a check mark is
entered in the toggle box it means the option is turned on and the no data value is allowed to
produce a valid grid cell value for the corresponding grid cell of the output data layer. This is the
option used.

The tool is executed using the 'Okay' button and produces the new 'SPtu1tsBuffVeg5' grid map.
Remember the only difference between this data layer and the 'SPtu1tsVeg5' data layer is that the
grid cells of the DBT ROW contain bare earth elevations from the 'SPtu1beElev5' data layer. All
the grid cells outside of the DBT ROW still have top surface elevation values.

4. Creating Viewshed Grid Maps for the DBT within the State Park and Proposed Park
Addition.
The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is one of two SAGA tools that
produces a viewshed. This tool accepts multiple observation points contained in a point shapes
data layer and produces an aggregate viewshed for all of the input observation points.

The other SAGA tool used for creating a viewshed is the Terrain Analysis - Lighting,
Visibility/Visibility (single point) [interactive] tool. One or more observation points are
interactively defined with the 'Action' tool and a viewshed is produced. This is an interactive
tool. If the 'Multiple Observer' parameter is turned off, a new viewshed is created, replacing the
last one, each time the user interactively locates another observation point. Using this mode the
user can change the location of the observation point interactively and re-run the viewshed
delineation. When you turn on the 'Multiple Observer' parameter, each time you locate a new
observation point, the first viewshed is updated with the results of each subsequent observer
point. The tool is active until the user explicitly stops the tool execution.

One of the differences between the two versions of the visibility tools relates to how the height
of the observer above the ground is specified. Due to the nature of the interactive tool, the height
for the observation point is entered as a data value for the 'Height' parameter. The same height
value is used for each interactively defined observer point. However, the version supporting
observation points defined as point objects in a point shapes data layer, requires that the attribute

34
table linked to the point shapes data layer include an attribute for the height of the observation
point above the ground. This approach means that the observer height to adjusted for each
observer point.

For this tutorial, the point shapes data layers are 'SPtu1Sp2Pts' and 'SPtu1AddPts' that were
developed earlier. The height above the ground for the observers (the point objects in the two
data layers) is the same for each point. It is 6 feet.

I use the table function "Add Field" to add a new attribute to the attribute tables linked to the
'SPtu1Sp2Pts' and 'SPtu1AddPts' data layers. I assign the data type "8 byte floating point
number" for the new fields. The attribute name is "HEIGHT".

The Table - Calculus/Table Calculator (Shapes) tool is used to enter the data value 6 for the
"HEIGHT" attribute for each of the 77 point objects of the 'SPtu1Sp2Pts' layer and the 22 point
objects of the 'SPtu1AddPts' data layer. Figure III-9 displays the parameters window for this tool.

Figure III-9. Entering 6 for the "HEIGHT" attribute.

The "f3" in the formula "f3 + 6" is how you reference, within the formula, an attribute in the
attribute table of the input shapes data layer. The number is the position of the attribute in the
table. I have chosen the 'SPtu1AddPts' point shapes data layer for the input parameter '>>
Shapes'. The third attribute ("f3") in the attribute table is the "HEIGHT" attribute. The
"<create>" option is used for the output parameter '< Result'. The output shapes data layer uses
the same name as the input. This data layer is saved using the same layer name.

The tool is executed a second time using the 'SPtu1Sp2Pts' for the input parameter '>> Shapes'.
The output data layer is re-named using the name of the input data layer. The "HEIGHT"
attributes in both attribute tables linked to the two layers now contain the data value 6. The
viewsheds grid maps can now be created.

35
Figure III-10 displays the parameters window for the Terrain Analysis - Lighting,
Visibility/Visibility (points) tool.

Figure III-10. The Terrain Analysis - Lighting, Visibility/Visibility (points) tool parameters
window.

The TS project area grid system is chosen for the 'Grid system' parameter. The
'SPtu1tsBuffVeg5' elevation layer is chosen for the input '>> Elevation' parameter. The
"<create>" option is used for the output parameter '<< Visibility'.

The DBT observation points in the proposed addition are the point objects in the 'SPtu1AddPts'
data layer. The "HEIGHT" attribute containing the height above the ground for the observation
points is chosen for the 'Height' parameter.

There are four options available for the 'Unit' parameter: Visibility, Shade, Distance, and Size.
The "Visibility" option is chosen for delineating a viewshed. I move the mouse pointer over the
'Okay' button and press the left mouse button and the tool executes using these parameter entries.
I re-name the default output grid map 'SPaddVisiTS' and save it.

The tool is executed a second time using the 'SPtu1Sp2Pts' point shapes data layer for the input
parameter '>> Points'. These points are the observation points on the section of the DBT located
in the state park. The output grid map is re-named 'SPspVisiTS' and saved to a storage folder.

5. Analyze the Viewsheds Relative to Visual Issues.

The questions introduced earlier regarding the state park, the proposed park addition, the DBT,
and Highway 101, were:

36
What is the viewshed for the DBT in the proposed addition?
Is there a visual impact to users of the DBT from Highway 101?
If Highway 101 is within the seen-area of the DBT viewshed, are there vegetation management
activities that will minimize the visual impact?

Figure III-11. Comparing the viewsheds for the park addition (in red) and the state park
(in yellow).

The map view window in Figure III-11 displays the DBT viewsheds for the proposed park
addition and the state park. The park addition is outlined in yellow. The viewshed for the park
addition is displayed in yellow. The state park is displayed with a red outline and the viewshed
displayed in red. One of the obvious differences between the two viewsheds is the lack of a view
of Sequim Bay from the portion of the DBT located in the state park. The bay is visible from
several observer points of the DBT in the park addition.

37
Is there a visual impact to users of the DBT from Highway 101?

The SAGA 3-D display tool is used to visually explore areas of Highway 101 in the state park
and the proposed park addition. The exploration started at the south end of the state park. In
Figure III-12 Highway 101 is displayed by the heavy black line and the DBT by the thin green
line.

Figure III-12. A zoomed-in section of Highway 101 falls within the DBT viewshed for the
state park.

38
The 'SPtu1tsBuffVeg5' grid map is used for the '>> Elevation' parameter for the 3D display. The
top map view windows, in the figure, display zoomed in views of a composite of data layers
including a topographic relief as backdrop and the red tinted viewshed for the DBT in the state
park. The thin green line is the DBT and the red dots are observation points. The black line is
Highway 101. The map view window on the bottom is a 3D map view of the area enclosed by
the black outlined rectangular box in the map view windows on the top.

The 3-D view is looking in a northerly direction over the DBT before it makes a small angle turn
to the right. You can see that the observation points just prior to the angle turn define a part of
the viewshed in which Highway 101 is visible (where highway 101 crosses over the red area).
This is the only area of Highway 101 that is visible from the DBT within the state park. As the
zoom view moves north of the highway and DBT, no other area of the highway shows up within
the DBT viewshed.

The data layers displayed in Figure III-13 are: SPtu1Sp2Pts, DiscBayTrl, SPspVisiTS,
Highway101, and DiscTrlEastReliefW2.

39
Figure III-13. Viewing a portion of the park addition viewshed (in yellow) along Highway
101 and the DBT.

The upper map view windows are zoomed in views of a composite of data layers including a
topographic relief as backdrop and the yellow tinted viewshed for the DBT in the park addition.
The thin green line is the DBT and the yellow dots are observation points along the DBT. The
black line is Highway 101. The black outlined rectangle identifies the approximate area of
coverage of the 3-D view window in the lower portion of the figure.

The 3-D view is looking in a northerly direction over the DBT as it angles from the southeast to
the northwest in the park addition. There appears to be topography or vegetation that serves to

40
block viewing Highway 101 from the DBT in the park addition. Notice that the seen-area in the
viewshed approaches the edge of the ROW for the highway but does not overlap the highway.
This is especially noticeable in the 3-D view. As the zoom view moves along the highway and
DBT further to the north, no area of the highway shows up within the DBT viewshed.

One approach to get a look at elevation and visibility detail is to use a profile. I create a new line
shapes data layer using the Shapes - Tools/Create New Shapes Layer tool. The new layer is
named 'SampleProfiles'. Using the "Add Shape" tool a short profile (line object) starting just to
the west of Highway 101 and ending to the east of the DBT is defined. In Figure III-14 the
profile line object is displayed in red.

Figure III-14. Defining a profile.

The Terrain Analysis - Profiles/Profiles from Lines tool is applied to create a profile using the
line object (in red) displayed in Figure III-14. Figure III-15 displays the parameters window for
this tool.

41
Figure III-15. The parameters window for the Terrain Analysis - Profiles/Profiles from
Lines tool.

The source for elevations for the profile is the 'SPtu1tsBuffVeg5' grid map. This layer is chosen
for the '>> DEM' parameter. Multiple grid data layers can be chosen for the input '> Values'
parameter. These data values are graphed for the profile. The 'SPtu1tsBuffVeg5', 'SPtu1tsVeg5',
'SPtu1beElev5', and 'SPtu1VegHt5' grid maps are chosen for this parameter. The "Diagram"
function is used as described below to graph the data values for these four data layers for the
profile. I move the mouse pointer over the 'Okay' button and click on it. The tool executes. A
point shapes data layer named 'SPtu1tsBuffVeg5 [Profile]' is created.

Graphs of the profile grid map data values for the four layers along the profile are created using
the "Diagram" function. This function is accessed in the 'Data' tab area of the Manager window.
The mouse pointer is moved over the point shapes data layer name 'SPtu1tsBuffVeg5 [Profile]'
and the right mouse button pressed. A pop-up menu of options displays. I move the mouse
pointer down to the "Attribute" option and another pop-up menu displays. I choose the
"Diagram" option. A properties window displays.

The following entries are made in this window. The default "Lines" option is used for the
'Display Type' parameter. Ground distance is going to be displayed as X-Axis values. The mouse
pointer is moved into the value field to the right of the 'Values' parameter in the X-Axis section
of parameters. A drop-down list of options displays. The mouse pointer is moved over the
"DIST" option and the left mouse button pressed. In the bottom portion of the window is a
section titled "Attributes'. These are toggle box parameters. The four grid maps chosen for the '>
Values' parameter for the tool execution are listed at the bottom. I want the diagram to have
graphs of the data values for these four layers. When I move the mouse pointer into one of the
toggle boxes and press the left mouse button, a check mark appears in the check box, and a color
selection box opens in the next row down. A different display color can be chosen using the

42
color selection box. I change the color and follow the same process for the next two data layers
in the list. Figure III-16 displays the diagram produced with these entries.

Figure III-16. Comparing bare earth and top surface elevations with the adjusted top
surface elevations along the profile.

The horizontal black line (c), approximately in the middle of the diagram in Figure III-16, is a
plot of the bare earth data values along the profile. The lower line (d), in olive green, is height
values for vegetation. You can see that the tallest vegetation is a little over 90 feet in height. If
you add the vegetation height data values to the bare earth vegetation values the result is the
elevation values for the top surface. The lime green plot (b) is for the elevation values for the top
surface layer. Comparing the bottom plot (d) to the upper plot (the lime green line) you can see
the relationship between the vegetation height and top surface elevation.

The green line (b) is a plot of top surface elevations. The red line (a) is identical for elevation
values except in the area of the DBT. Thus, in the diagram the top surface elevation graph is on
top of the red line except in the area of the DBT where the red line shows. The red line is
difficult to see but it is visible approximately at the 200' data value of the X-axis (e).

I can see that a very slight increase in elevation occurs between the DBT (f) and Highway 101
(g), between the X-axis values 124.801 and 132.601. The same area of the slight elevation
increase is mostly short vegetation (probably meadow grass and short shrubs).

If Highway 101 is within the seen-area of the DBT viewshed, are there vegetation management
activities that will minimize the impacts?

I can see one area of Highway 101 that falls within the viewshed of the DBT in the state park
(see Figure III-12). There are a variety of options to consider.

Tall-growing tree species can be planted up to and along the eastern edge of the Highway 101
ROW. Existing vegetation in that area can be encouraged and any further reduction discouraged.
Additional tree species and shrubbery can be planted just to the north of where the DBT makes

43
the slight right angle turn. Using this option would not affect ROW management for Highway
101 and might be more attractive to park management.

IV. Timber Harvest Clearcut Viewshed Options.

The Creek Timber Sale (TS) is in the planning stage. The visual impact from the clear-cut
harvest of the vegetation is being analyzed. In particular, the visual impact for users of nearby
travel routes (Wood Road and Highway 101) and the popular outdoor recreation trail, the
Discovery Bay Trail. The harvest method is clear-cut which means that all of the vegetation
within the TS area is removed.

The TS area is 25.6 acres in size. A narrow swath of vegetation along the 1900' long west and
600' long north sides of the area is being evaluated for a role as a potential visual block for the
post-harvest clear-cut area. This swath of vegetation varies between 50 and 100 feet wide and is
4.25 acres in area. The tallest vegetation is 136' and the mean is 53'. Approximately 68% of the
vegetation in the swath is in the height range of 22 to 84 feet.

Three sets of observer points are defined; one for each of the two roads and a third one for the
recreation trail. Three viewsheds, one for each group of observer points, are delineated for the TS
area and another set of three for the TS including the removal of the border vegetation as part of
the harvest.

A simple comparison of viewshed acres is made to identify the visual impact differences for the
three observer point groups relative to the TS area and when the border vegetation is also
harvested.

Figure IV-1 displays a zoomed in view of the Creek TS showing its relationship to adjacent
transportation routes and the recreation trail.

44
 Figure IV-1. A zoomed in view of the Creek TS areas.

The red filled area is the TS area. The smaller area displayed in yellow is the swath of vegetation
being analyzed to determine its effectiveness as a visual block.

Wood Road is located parallel and within 130 to 400 feet west of the Creek TS. Highway 101
(the thick black line) and the Discovery Bay Trail (DBT) (the thinner green line) are seen just
below the bottom of Sequim Bay. They are both about 1/4 of a mile to the north of the sale area.

Summary of Steps
1. Define a timber sale project area grid system.
2. Define sets of observer points for the DBT, Highway 101, and Wood Road.
3. Prepare a top surface elevation layer.
4. Prepare viewshed grid maps for the TS area for the DBT, Highway 101, and Wood Road
groups of observer points.
5. Analyze the viewsheds relative to visual issues.

1. Define a Timber Sale Project Area Grid System.

The map view window in Figure IV-1 displays the area that will serve as the project area for the
Creek TS. Visual impact is related to observation points to the west and north of the TS. The

45
project area grid system is a sub-area of the much larger Sequim Bay grid system and includes
landscape from the southeast of the TS, to just west of Wood Road, and to the south shore of
Sequim Bay.

A major reason for reducing the project area to a sub-area of the larger Sequim Bay grid system
is due to the large volume of data created by the LIDAR process. The raw data collected by this
sensor is extremely dense. Generalizing the data to a 6' grid cell resolution reduces the volume
but still results in a large volume of information. This has an impact on software and system
performance. If the detailed resolution is not necessary for an application it is best to re-sample
the data using a grid system based on a larger grid cell size that would be more appropriate.

The interactive tool Grid - Tools/Clip Grids [interactive] is used to define a project area that is a
sub-area of the larger Sequim Bay grid system. The map view window in Figure IV-1 is
displayed in the work area. The data layers 'AddOn1', 'HarvestArea1', 'Highway101',
'DiscBayTrl', 'ClallamRoads', and 'SequimBay' are displayed in the map view window.

Figure IV-2 displays the parameters window for the Grid - Tools/Clip Grids [interactive] tool.

Figure IV-2. Parameter entries using the Grid - Tools/Clip Grids [interactive] tool to define a
project area.

This tool is used to clip one or more existing grid maps that are members of the same grid
system. The Sequim Bay grid system that was defined during the processing of the downloaded
LIDAR files is chosen for the 'Grid system' parameter (6; 5018x 6144y; 101235x 977274y).
Three grid maps, members of the selected grid system, are chosen to be clipped and are
identified for the '>> Grids' input parameter. These are the bare earth ('QSbeElev4'), top surface
('QStsVeg4'), and vegetation height ('QSVegHt4') grid maps.

The 'Run Once' parameter uses a toggle box for turning the option on or off. When I enter a
check mark in the toggle box the option is turned on. This means that after a rectangular area is

46
defined, as described below, the tool stops running. If this option is turned off, after defining a
rectangle, the tool continues to run. Additional rectangles can be created to define additional grid
systems; each new defined rectangle replaces the previous one. The user must explicitly quit the
tool.

The tool executes when the mouse pointer is clicked on the 'Okay' in the parameters window.

This is an interactive tool. The 'Action' tool from the toolbar is used to interact with the tool
while it is running. The 'Action' tool is applied in the map view window displayed in the work
area. The data layers displayed in the map view window are used for orientation and to assist in
defining a rectangle. A rectangular area is defined by moving the mouse pointer to a location for
a corner. I press the mouse button, hold it down, and drag the mouse pointer to another corner
that is diagonally opposite, and release the mouse button. This highlights and defines a
rectangular area that becomes the spatial extent for a sub-area of the Sequim Bay grid system to
be the project area grid system for the Creek TS. The 'Clip to Extent' dialog window in Figure
IV-3 displays.

Figure IV-3. The 'Clip to Extent' dialog window for a new project area.

The four parameters 'Left', 'Right', 'Bottom', and 'Top' show the coordinates defining the four
sides of the interactively defined rectangle. The 'Columns' and 'Rows' parameters display the
number of columns and rows definition using the original grid cell size of 6'. The definition is
accepted by clicking the mouse pointer on the 'Okay' button.

47
A new grid system definition, using the parameters, is defined and the definition appears in the
'Data' tab area of the Manager window.

6; 478x 655y; 1027272x 987066y

This new grid system is referred to as the project area grid system for the Creek TS. The three
grid maps chosen for the input parameter '< Grids' are clipped by the interactively defined
rectangle and become members of the new grid system. The default names for the new layers are
the same as the input names. The three new grid maps are re-named 'TSTu2beElev5',
'TSTu2tsVeg5', and 'TSTu2VegHt5' and saved in a directory.

The interactive tool is still running. The 'Run Once' parameter is in off status. The tool execution
is stopped by moving the mouse pointer to the 'Geoprocessing' title in the Menu Bar and pressing
the left mouse button. A drop-down list of options displays. Near the bottom of the options, the
tool execution history for the work session displays. The names of the tools in the history list are
grayed out except for the first one. The first tool listed in this section is the "Clip Grids
[interactive]" tool and a check is showing to the left of its' name. The check indicates the tool is
currently executing. The mouse pointer is moved over the tool name and the left mouse button
pressed. A 'Tool Execution' dialog window displays with the question: Shall execution be
stopped? When I click on the 'Yes' button in the window, the tool execution stops. The check
mark disappears.

Line shapes data layers exist for Highway 101 and the DBT. Highway 101 is defined by a series
of line objects as is the DBT. Wood Road is a single line object in the 'ClallamRoads' data layer.
The 'ClallamRoads' data layer is the road network layer for Clallam County. Three new line
shape data layers need to be created; one for the portion of Highway 101 that falls within the
Creek TS project area grid system, a second layer for the portion of the DBT that falls within the
grid system, and a third one for Wood Road.

The Shapes - Polygon/Polygon Clipping tool is applied to create the two new line shapes data
layers for Highway 101 and the DBT. Before I can apply the tool I need to create a polygon
shapes data layer that contains a polygon object representing the grid system extent for the
project area. I use the Shapes - Grid Tools/Grid System Extent tool for this objective. The
parameters window for the tool displays in Figure IV-4.

48
 Figure IV-4. The parameters window for the Shapes - Grid Tools/Grid System Extent tool.

The grid system for the timber sale project area is chosen for the input 'Grid system' parameter.
The default "<create>" option is used for the output '<< Extent' parameter and the default is also
used for the 'Border' option. The default output is named 'Grid System Extent'. I re-name it to
'TSgsextent' and save it.

The Shapes - Polygons/Polygon Clipping tool is used to create project area data layers for the
DBT and highway 101. The parameters window for this tool displays in Figure IV-5.

Figure IV-5. The Shapes - Polygons/Polygon Clipping tool parameters entries for creating
line shapes data layers.

49
The line shapes data layers 'DiscBayTrl' and 'Highway 101' are chosen for the input parameter
'>> Input Features'. The 'TSgsextent' polygon shapes data layer recently created by the Shapes -
Grid Tools/Grid System Extent tool is chosen for the '>> Clip Features' input parameter. There
are two options. I am using the defaults for the two toggle box parameters.

The tool is executed using the parameter entries in Figure IV-5. The default output line shapes
data layers are named 'DiscBayTrl [clipped]' and 'Highway101 [clipped]'. I re-name the layers
'TSdiscbaytrl' and 'TShighway101' and save them.

A line shapes data layer for Wood Road is also needed. This new data layer can be created by
selecting the line object for Wood Road on the 'ClallamRoads' line shapes data layer and using
the Shapes - Tools/Copy Selection to New Shapes Layer to copy the selection to a new layer.

The 'ClallamRoads' line shapes data layer is displayed in a map view window. This is the data
layer for the road network for Clallam County. I zoom in on the Creek TS area. Making sure the
'ClallamRoads' layer is active, I use the 'Action' tool to select the Wood Road line object. When
it is highlighted, it displays in red. I execute the Shapes - Tools/Copy Selection to New Shapes
Layer tool to copy the highlighted line object (i.e., Wood Road), into a new line shapes data
layer. The new layer is re-named 'TSwoodroad' and saved.

In summary, a grid system for the timber sale project area has been defined. The grid system has
been used to create project area versions of the LIDAR elevation data layers. The bare earth grid
map is named 'TSTu2beElev5, the top surface layer is named 'TSTu2tsVeg5' and the vegetation
height layer is named 'TSTu2VegHt5'. In addition, the grid system extent has been used to define
two line shapes data layers for Highway 101 and the DBT within the project area. These are the
'TShighway101' and 'TSdiscbaytrl' layers. A third line shapes data layer was created for Wood
Road named 'TSwoodroad'.

2. Define Sets of Observer Points for the DBT, Highway 101, and Wood Road.
A viewshed is going to be defined for sets of observer points on each of the three linear features.
The viewsheds for Highway 101 and the DBT may be quite similar because these routes are
parallel and near each other. Differences will result due to the placement of observation points.

The Shapes - Tools/Create New Shapes Layer tool is used to create a blank point shapes data
layer for each of the three linear features. The parameters window for the tool is displayed in
Figure IV-6.

50
Figure IV-6. Creating new point shapes data layers with the Shapes - Tools/Create New
Shapes Layer tool.

In this example, the name for the new shapes data layer, 'TShighway101Pts', is entered for the
'Name' parameter. I choose the "Point" option for the 'Shape Type' parameter. I need to define an
attribute in the linked attribute table named "HEIGHT". This attribute stores the height of the
observer above the ground. The "HEIGHT" attribute is used later with the Terrain Analysis -
Lighting, Visibility/Visibility (points) tool for delineating viewsheds.

I move the mouse pointer over the ellipsis in the value field to the right of the 'Attributes'
parameter and press the left mouse button. The 'Attribute' dialog window displays. I edit the
second attribute entry in the window changing the entry for the 'Name' parameter to "HEIGHT".
I change the "string" entry for the 'Type' parameter to "4 byte floating point number". I return to
the tool parameters window by clicking the mouse pointer on the 'Okay' button. I click on the
'Okay' button on the tool parameters window and the tool creates the new, empty point shapes
data layer. I save it.

The Shapes - Tools/Create New Shapes Layer tool is executed two more times creating point
shapes data layers named 'TSdiscbaytrlPts' and 'TSwoodroadPts'.

The process for defining observation points is the same for each of the three data layers. The
'TSTu2tsVeg5' top surface elevation grid map displays in a map view window in the work area.
The elevation data for this grid map identify the height of the vegetation cover (or man-made
structures) above sea-level. In the same map view window the line shape data layer the observer
points are to be defined for is displayed. In the 'Settings' tab area in the 'Object Properties'
window for the line shapes data layer the value for the 'Default Size' parameter (sets the width of
the display of the line) is increased from 0 to 2. This will make it easier to see the linear feature.

51
In the same map view window the empty point shapes data layer created for capturing the
observation points for the linear feature displays.

In addition to this map view window, a 3-D view window is opened, using the 'TSTu2tsVeg5'
top surface elevation layer for the '>> Elevation' parameter. The 3D tool on the tool bar develops
the 3-D view window.

Five observation points are defined along the DBT. The active data layer is the 'TSdiscbaytrlPts'
point shapes data layer. This layer is made active by clicking on the data layer name in the 'Data'
tab area of the Manager window. With this layer active, I edit two parameters in the 'Settings' tab
area for the layer.

When a point object is defined, the location of the point object is to be on a line object of the
'TSdiscbaytrl' line shapes data layer. In other words, the defined observer points are to be on the
trail. There are two parameters in the 'Settings' tab area in the 'Object Properties' window for the
'TSdiscbaytrlPts' point shapes data layer that support this requirement. The 'TSdiscbaytrl' line
shapes data layer is selected to be the entry for the '>> Snap to...' parameter. The 'Snap Distance'
parameter default entry is changed from 10 to 20. These units are screen pixels. When the mouse
pointer is used to locate an observer point, if the point defined is not on the DBT but within 20
screen pixels of the DBT, the point object location will snap to the nearest location on the DBT.

The "Add Shape" option available for shapes data layers is used. The 'Action' tool on the toolbar
is chosen and the mouse pointer moved within the map view window that displays the
'TSTu2tsVeg5' grid map and the 'TSdiscbaytrlPts' point shapes data layer. I press the right mouse
button and a pop-up list of edit options displays. The second one in the list, "Add Shape", is
chosen. I move the mouse pointer to where I want to add an observer location (a point object)
and press the left mouse button. This temporarily defines a point object at that coordinate
location. If it is within the value for the 'Snap Distance' parameter, its location snaps to the
nearest line object. If I am not happy with the location I can move the mouse pointer to a better
location and re-press the left mouse button, moving the location of the temporary point object.
When I am okay with the location and want to save it, I press the right mouse button and a dialog
window displays with the question "Edit Selected Shapes". A check mark displays to the left of
the question when a tool is active. I can press either the left or right mouse buttons and another
dialog window displays: "Apply Changes?". There are two choices, selected with either the
"Yes" or "No" buttons. I choose "Yes".

Most of the new observation point objects are located so their (estimated) view to the TS is not
blocked by vegetation. The 3-D view window is used as a visual aide to see if the point object
placement is free from being blocked by vegetation. As the 5 point locations are defined, I also
enter the value 6 for the "HEIGHT" attribute. This is the height the observer point is above the
ground. This attribute is used later when the Terrain Analysis - Lighting, Visibility/Visibility
(points) tool is executed to delineate the viewshed.

The process outlined above is also followed to define observation points for Highway 101 and
Wood Road. Each time I make sure to edit the two parameters in the 'Setting' tab area of the
appropriate line shapes data layer to "snap" the observer points to the line feature. Five point

52
objects are defined on the 'TShighway101Pts' layer and 9 point objects on the 'TSwoodroadPts'
layer. Figure IV-7 displays the three point shapes data layer.

Figure IV-7. Map and 3-D view windows of the observation points.

Observation points on the DBT are displayed in yellow, on Highway 101 the points are in red,
and on Wood Road the points are displayed in green. The TS is displayed in a reddish tint and
the potential add-on to the TS area is in a yellow-tint. The 3-D view is looking from Sequim Bay
south toward the TS area. Remember the potential add-on is the border vegetation along the west
and north sides of the TS.

3. Prepare a Top Surface Elevation Layer.

Top surface elevation is one of the two inputs for the Terrain Analysis - Lighting,
Visibility/Visibility (points) tool. The other input is a set of observation points. Each observation
point is going to be 6' above the ground (roughly representing eye-level). This tool delineates the
aggregate viewshed for the group of input observation points.

Elevations for top surface are provided in the 'TSTu2tsVeg5' grid map. These elevations are for
the vegetation surface above the ground elevation. For example, if a grid cell has a bare earth
elevation of 100' and the top surface elevation for the same grid cell is 150', it means a vegetation
canopy (or building roof) exists 50' above the bare earth elevation. The height of the observer
point is added to the elevation entry for the grid cell the observer point falls in. In this example,
since I am using the top surface elevation layer this means the observer point is 6' above the
vegetation cover. This is not desired. The observer point should be 6' above the ground, not
above the vegetation. Placing the observer above the vegetation does not produce a viewshed
that considers the blocking effect of vegetation. The height above the ground for all observer

53
points is 6'. One solution is to replace the top surface elevations within the linear features right-
of-ways with bare earth elevation values.

A buffer polygon can be created that represents the ROW for a linear feature. The ROW width
for Highway 101 appears to be 75'. The ROW width for Wood Road is 35' and a reasonable
ROW for the DBT is 6'. A polygon shape data layer can be created that contains the ROW
polygon objects for the linear features. These ROW polygons are created using the Shapes -
Tools/Shapes Buffer tool.

The Grid - Calculus/Grid Calculator tool is used to identify the grid cells of the 'TSTu2tsVeg5'
grid map that fall within the linear features' ROWs. Before the polygon shapes data layers for the
ROW's of the linear features can be used with this tool they must be converted to grid maps
using the Grid - Gridding/Shapes to Grid tool.

In summary, replacing the top surface elevations within the linear features' ROW's involves three
steps:

(1) Create a ROW polygon for each linear feature.

(2) Convert the ROW polygon shapes data layers to grid maps.
(3) Replace the top surface elevation values with the bare earth elevation data values within the
ROW's.

The parameters window for the Shapes - Tools/Shapes Buffer tool is displayed in Figure IV-8.

Figure IV-8. Creating a 6' wide buffer for the DBT.

54
The entries in the parameters window in Figure IV-8 are for creating a buffer polygon for the
DBT. The line shapes data layer for the DBT in the TS project area is named 'TSdiscbaytrl'. This
layer is chosen for the input parameter '>> Shapes'.

The attribute table linked to the input line shapes data layer can include an attribute identifying a
buffer width. For example, if the buffer width varied for the line objects, this attribute would
capture that variation. The attribute could be used to supply to the tool the buffer width for each
line object. This is not the case here with the DBT, so the buffer width is entered for the 'Default'
parameter. The width entered, 3, is half of the final polygon width as the buffer is defined on
both sides of an object boundary.

The "<create>" option is chosen for the output parameter '<< Buffer'. I use the default settings
for the options parameters. I move the mouse pointer to the 'Okay' button and click. The tool
executes creating an output polygon shapes data layer named 'TSdiscbaytrl [Buffer]'. I re-name it
to 'TSdiscbaytrlBuff''. Before I re-named the layer I changed the data value for the "ID" attribute
from 0 to 1.

The Shapes - Tools/Shapes Buffer is executed two more times creating similar buffers for the
'TShighway101' data layer and the 'TSwoodroad' layer. I use an entry of 37.5 for the 'Default'
parameter for the Highway 101 buffer and 17.5 for the 'Default' parameter for the Wood Road
buffer. After changing the data value for the "ID" attribute in the attribute tables linked to the
output data layers from 0 to 1, I re-name the layers 'TShighway101Buff' and 'TSwoodroadBuff''
and save them.

There are now three polygon shapes data layers, each consisting of a single polygon object for
the ROW buffer of the linear features. The data layers are 'TSdiscbaytrlBuff,
'TShighway101Buff' and 'TSwoodroadBuff''. The next step is to convert these data layers to a
grid map of the project area grid system.

The easiest way to accomplish this is to merge the three layers into one polygon shapes data
layer containing three polygon objects, one for each ROW buffer area. This means that rather
than creating a grid map for each linear feature I create a grid map for the three features after
they are merged into a single polygon shapes data layer (containing three polygon objects).

Another consideration relates to the TS polygon shapes data layer 'HarvestArea1'. The harvest
area polygon is used to select the vegetation area of the top surface elevation grid map that
changes as a result of the clear-cut harvest. Thus it is used to identify the top surface elevations
that must be replaced with bare earth elevations, simulating the harvest (removal) of existing
vegetation. I can include the single polygon object of the 'HarvestArea1' data layer in this merge
along with the ROW buffer areas.

The Merge Layer tool works with any of the four types of shapes data layer types: line, points,
multipoints, or polygon. However, it will not merge shapes layers of different types. The tool
merges two or more data layers of the same type. The parameters window for the Shapes -
Tools/Merge Layer tool displays in Figure IV-9.

55
Figure IV-9. Merging four polygon shapes data layers using the Shapes - Tools/Merge Layer
tool.

Four polygon shapes data layers are chosen for the input parameter '>> Layers'. Note that the
tool assumes that the input layers are all of the same data layer type. Input layers are selected
from a list of shapes data layers available in the 'Data' tab area of the Manager window. The tool
does not check to see if the layers chosen are all of the same type. The tool decides the type of
shapes data layer output from the first data layer in the list regardless of whether the other data
layers in the list are of the same type. If they are not of the same type they are ignored and not
part of the output.

The three ROW buffer polygon layers ('TSdiscbaytrlBuff', 'TShighway101Buff',

'TSwoodroadBuff') and the TS harvest area layer ('HarvestArea1') are chosen and listed as input
for the input parameter '>> Layers'.

The default output name is 'Merged Layers'. The 'Okay' button is clicked on and the tool
executes.

The output polygon shapes data layer is re-named 'TSrowRecode1' and saved. A "1" is included
in the layer name because a second version of this layer is needed, one that includes the
'AddOn1' polygon data layer. This second data layer includes the polygon for the narrow strip of
vegetation along the west and north sides of the TS. So this second merged data layer is used to
evaluate the viewshed difference when the narrow strip of vegetation along the west and north
sides of the TS is also harvested.

The Shapes - Tools/Merge Layer tool is executed a second time using the same entries for the
input parameter '>> Layers' as the first time with the 'AddOn1' layer added to the list. The default
'Merged Layers' output layer is re-named to 'TSrowRecode2' and saved.

These two shapes layers are going to be converted to grid data layers using the Grid -
Gridding/Shapes to Grid tool. One of the parameters for this tool is to enter the grid system the
converted layer becomes a part.

56
Another of the inputs for the Grid - Gridding/Shapes to Grid tool is to identify an attribute in the
attribute table linked to the input shapes data layer that is the source of data values for grid cells
of the output grid map. The attribute tables linked to the 'TSrowRecode1' and 'TSrowRecode2'
data layers each have an attribute named "ID". This is the attribute that is going to be used to
provide output data values.

I need to make sure that the data value 1 is entered for the "ID" attribute for each of the records
in the two attribute tables. Looking at the two attribute tables I find that a couple records have 0's
for the "ID" attribute. I manually key in the value 1 to replace the 0's and re-save the layers.
Once this is done, the tool can be executed. The parameters window for the Grid -
Gridding/Shapes to Grid tool is displayed in Figure IV-10.

Figure IV-10. The parameter entries for converting the 'TSrowRecode1' shapes data layer
to a grid map.

The entries in the parameters window in Figure IV-10 are for the first of two executions of this
tool. The polygon shapes data layer 'TSrowRecode1' is chosen for the input parameter '>>
Shapes'. The option parameter 'Output Values' has the "attribute" option selected. This means
that the 'Attribute' parameter becomes visible and available in the parameters window. You can
see that the "ID" attribute is chosen for the 'Attribute' parameter. This is the attribute that
provides the data values for the output grid map. The "grid or grid system" option is chosen for
the 'Target Grid System' parameter. The grid system for the TS project area is chosen for the
value field to the right of the 'Grid System' parameter. The default "<create>" option is used for

57
the output parameter '<< Target Grid'. I move the mouse pointer over the 'Okay' button and press
the left mouse button and the tool executes.

The output grid map is named 'TSrowRecode1 [ID]'. It is re-named 'TSrowRecode1GR' and
saved. The tool is executed a second time choosing for input the 'TSrowRecode2' data layer. The
'TSrowRecode2 [ID]' output grid map is re-named to 'TSrowRecode2GR' and saved.

The top surface elevation adjustments to bare earth elevations for linear features and the TS area
result in a new version of the 'TSTu2tsVeg5' layer to be named 'TSTu2tsVeg6'. The effect of
harvesting the 4.25 acres of vegetation bordering the timber sale on its west and north sides
results with another version of the top surface elevation data layer to be named 'TSTu2tsVeg7'.
The Grid - Calculus/Grid Calculator tool is used for making these elevation changes and new
grid maps. The parameters window for the tool is displayed in Figure IV-11.

Figure IV-11. Using the Grid - Calculus/Grid Calculator tool to edit top surface elevations
within the ROW and TS.

Three grid maps are chosen for the input parameter '> Grids'. They are 'TSrowRecode1GR',
'TSTu2beElev5' and 'TSTu2tsVeg5'. The default entry "<create>" is used for the output
parameter '<< Result'. The equation entered for the 'Formula' parameter refers to the input grid
maps with variables "g1", "g2" and "g3". The equation checks the 'TSrowRecode1GR' data layer
(variable "g1") to see if the grid cell value is 1. If it is 1 then the grid cell is a ROW grid cell or
TS area grid cell so the condition is true. When true, the elevation data value in the
corresponding grid cell of the bare earth elevation 'TSTu2beElev5' data layer (variable "g2") is
output to the corresponding grid cell of the output grid map. If the value is not 1, the condition is
false and the elevation data value for the same grid cell of the top surface 'TSTu2tsVeg5' layer
(variable "g3") is output to the corresponding grid cell of the output grid map.

58
The name for the output grid map, 'TSTu2tsVeg6', is entered for the 'Name' parameter. The grid
cells of the 'TSrowRecode1GR' data layer that are not ROW or TS area grid cells contain the no
data value -99999. When a grid cell containing the no data value is encountered, the no data
value is automatically output to the corresponding grid cell of the output grid map regardless
data values of corresponding grid cells of other input layers. The 'Use NoData' parameter is an
option that allows the no data value to be considered a valid data value. The equation can then
produce a valid data value for the output grid data. This is a toggle box parameter. A check mark
in the box indicates the option is on. I make sure this option is on.

The tool is executed and the new grid map 'TSTu2tsVeg6' is created. I save it.

59
Figure IV-12. Zoomed in areas of 2-D and 3-D map view windows for the 'TSTu2tsVeg5'
and 'TSTu2tsVeg6' grid maps.

The graphics in Figure IV-11 clearly reveal the differences between the original top surface grid
map, (a) and (c) and the modified version on the right, (b) and (d). In (b) and (d) notice the
apparent absence of vegetation for the TS (outlined in black) compared to its presence (a) and
(c). The two graphics in the lower half of the figure, (c) and (d), provide a 3-D perspective of the
before and after.

60
The tool is executed a second time replacing the entry 'TSrowRecode1GR' layer with the
'TSrowRecode2GR' in the grid map input list. The three grid maps chosen for the input
parameter '> Grids' are 'TSrowRecode2GR', 'TSTu2beElev5' and 'TSTu2tsVeg5'. The default
entry "<create>" is used for the output parameter '<< Result'. The equation entered for the
'Formula' parameter is "ifelse(eq(g1,1),g2,g3)".

The name for the output grid map, 'TSTu2tsVeg7', is entered for the 'Name' parameter. The 'Use
NoData' parameter option is on. I execute the tool and the new grid map 'TSTu2tsVeg7' is
created. I save it.

In summary, the 'TSTu2tsVeg6' top surface elevation grid map has top surface elevations in the
linear feature ROW's and TS area replaced with bare earth elevations. Data values outside of
these specific areas retain their original top surface elevations. The 'TSTu2tsVeg7' grid map is
similar adjusting top surface elevations in the TS and the additional border vegetation area
adjacent to the west and north sides of the TS.

4. Prepare Viewshed Grid Maps for the TS Area for the DBT, Highway 101, and Wood
Road Groups of Observer Points.
The first set of viewshed grid maps are based on the 'TSTu2TSveg6' top surface elevation grid
map. The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is used with the three
point shapes data layers and the 'TSTu2tsVeg6' to create an aggregate viewshed for the observer
points in each one of the point shapes data layers. Below is a list identifying the input parameters
and output grid maps associated with this tool.

Observation
Points Elevations Viewshed
TSdiscbaytrlPts TSTu2TSVeg6 TSdbt6Visi
TShighway101Pts TSTu2TSVeg6 TS1016Visi
TSwoodroadPts TSTu2TSVeg6 TSwood6Visi

An alternative approach to using three point shapes data layers for the observer points is to
merge them into a single layer. This results with a single composite viewshed grid map. It would
not be possible to consider the source of differences between the three sets of observer points.
Thus, using three point shapes data layers, one for each set of linear feature observers, was felt to
be a better approach.

Figure IV-13 displays the parameters window for the tool.

61
Figure IV-13. The parameters window for the Terrain Analysis - Lighting,
Visibility/Visibility (points) tool.

The TS grid system is chosen for the parameter 'Grid system'. The 'TSTu2tsVeg6' top surface
grid map is chosen for the '>> Elevation' input parameter. The default "<create>" option is used
for the output parameter '<< Visibility'. The output is a grid map named 'Visibility'.

The observation points are the point objects of the 'TSdiscbaytrlPts' data layer. The "HEIGHT"
attribute in the attribute table linked to this layer identifies the observation height for each of the
point objects. In our case, this height value does not change for any of the observation points.
The observer is assumed to be 6' above the ground.

There are four options available for the 'Unit' parameter: Visibility, Shade, Distance, and Size.
The "Visibility" option is chosen for delineating a viewshed. I move the mouse pointer over the
'Okay' button and press the left mouse button and the tool executes using these parameter entries.
I re-name the default output grid map 'TSdbt6Visi' and save it.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is executed two more times
using the other two observation point shapes data layers 'TShighway101Pts' and
'TSwoodroadPts' with the same elevation data layer 'TSTu2tsVeg6'. The viewshed grid maps
created are 'TS1016Visi' for the Highway 101 observation points and 'TSwood6Visi' for the
Wood Road observation points.

The second set of viewshed grid maps are based on the 'TSTu2TSveg7' top surface elevation grid
map. The only difference between the two sets of viewsheds is that this second set includes the

62
narrow band of vegetation along the west and north sides of the TS as part of the harvest, thus
removing it as a potential shield for visual impact.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is used with the three point
shapes data layers and the 'TSTu2tsVeg7' to create an aggregate viewshed for the observer points
in each of the shapes data layers. Below is a list identifying the input parameters and output grid
maps from executing the tool.

Observation
Points Elevations Viewshed
TSdiscbaytrlPts TSTu2TSVeg7 TSdbt7Visi
TShighway101Pts TSTu2TSVeg7 TS1017Visi
TSwoodroadPts TSTu2TSVeg7 TSwood7Visi

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool was executed three times using
the inputs listed above. The new aggregate viewshed grid maps produced and saved are
'TSdbt7Visi', 'TS1017Visi' and 'TSwood7Visi'.

5. Analyze the Viewsheds Relative to Visual Issues.

The primary objective of this analysis is to determine whether the narrow border of vegetation
along the west and north sides of the TS acts as a visual barrier for a resulting clear-cut. How
much of the post-harvest clear-cut area is visible from the several sets of observer points when
the border vegetation is retained versus when the border vegetation is harvested?

Figure IV-14 displays six map view windows providing a visual comparison of the clear-cut
results when including and not including the narrow strip of vegetation along the west and north
sides of the TS.

63
Figure IV-14. Comparing the viewsheds for the three observer groups when the vegetation
barrier is present (on the left) versus when it is removed (on the right).

64
The map view windows in Figure IV-14 display the viewsheds for the TS. The map view
windows on the right in the figure display the viewsheds after the narrow strip of vegetation
along the west and north sides is harvested as part of the TS. You can see the TS boundary
displayed in black in the lower half of each map view window. The narrow vegetation strip along
the western and northern boundaries of the TS is also displayed. The blue area to the north is
Sequim Bay.

The top two map view windows are viewsheds based on the 5 observation points defined for the
DBT. The observer points display as yellow circles. The viewshed displays in a yellow tint.
Areas not displayed in yellow are not seen from the observer points.

The middle two map view windows displaying the viewsheds in a red tint show what can be seen
from the 5 observation points along Highway 101. These observation points are displayed as red
circles. Again, areas not displayed in red are not visible from the observation points.

The lower two map view windows use a green tint to display the areas of the landscape that can
be seen from the 9 observation points along Wood Road to the west of the TS. The observation
points displays as lime green circles.

Visually comparing the clear-cut with the vegetation barrier versus the clear-cut without the
barrier you can see there are differences. The vegetation barrier does act, to a degree, as a visual
shield for the three groups of observation points.

65
Figure IV-15. A zoomed in view of the viewsheds for the three observer groups within the
TS and vegetation barrier areas.

66
Any difference in viewsheds, for the same observer group, between the TS with the barrier and
without the barrier, is due to the removal of the barrier. Nothing else changed between the
creation of the two viewsheds. Figure IV-15 displays map view windows zoomed in on the TS
and the vegetation barrier.

The blocking effect of the strip of border vegetation is very apparent when you enlarge this area
on each of the map view windows for the TS viewsheds. Many grid cells within the strip of
border vegetation fall in the seen-areas of the viewsheds. In the TS viewshed for the DBT
observers, a few grid cells along the south and southeast border of the TS are visible. This is the
same for the Highway 101 observers plus a handful near the middle of the sale area. There are
several small areas of visible grid cells scattered throughout the viewshed for Wood Road.

There are differences when the narrow strip of border vegetation is harvested.

The Spatial and Geostatistics - Grids/Zonal Grid Statistics tool can be used to quantify the
difference in seen areas between the pairs of viewsheds for an observer group. This tool is used
with each viewshed grid map chosen for the '>>Zone Grid' input parameter. The output 'Zonal
Statistics' table lists the count of grid cells containing each data value, 0 and 1. Knowing that
there are 1210 grid cells in an acre it is relatively easy to produce the acre statistics. Table IV-1
displays a summary of this data.

Table IV-1: Comparing Viewshed Acres

The viewshed acres in Table IV-1 show an increase in the seen-area, when the vegetation barrier
is harvested, of 6.2 acres for the DBT group of observers, an increase of 8 acres for the Highway
101 observers, and an increase of 14.3 acres for the Wood Road group of observers.

The Shapes - Grid Tools/Clip Grid with Polygon tool can be used to create grid maps for the
portions of the viewsheds that fall within the TS polygon. Then the Spatial and Geostatistics -
Grids/Zonal Grid Statistics tool can be used with these new grid maps and a table similar to
Table IV-1 developed using only the TS area of 25.6 acres. The results are in Table IV-2. Figure
IV-16 displays the parameters window for the Grid - Grid Tools/Clip Grid with Polygon tool.

67
 Figure IV-16. The parameters window for the Grid - Tools/Clip Grid with Polygon tool.

The TS project area grid system is chosen for the 'Grid system' parameter. You can see that the
six viewshed grid maps produced earlier make up the input list for the '>> Input' parameter. The
polygon shape data layer for the TS is chosen for the input parameter '>> Polygons'. This data
layer consists of a single polygon defining the TS boundary. The tool is executed and a new,
clipped version (clipped to the TS polygon), of each of the six input grid maps is created. They
are re-named 'TSH1016Visi', 'TSH1017Visi', ... TSHwood7Visi' and saved.

The Spatial and Geostatistics - Grids/Zonal Grid Statistics tool is used to capture the number of
grid cells seen (grid data values of 1) versus the number of grid cells not seen (grid data values of
0). I, again, manually convert the numbers to acres and build the second table.

Table IV-2: Comparing Timber Sale Viewshed Acres

Table IV-2 provides a much better accounting of the blocking affect of the narrow strip of
vegetation along the west and north sides of the TS.

The visual impact from DBT observers is for recreation hikers and bicyclists. Using the pair of
viewsheds for this group, you can see that if the strip of vegetation along the north side of the TS
was retained the clear-cut would be mostly blocked from view from this group of observers. If
you add in the last 400' of the northwest portion of the strip the view would also be blocked from
view by travelers of Highway 101. Users of the first 800' of the northern portion of Wood Road
would also benefit from this retention. However, most of the vegetation blocking for Wood Road
is from the strip of vegetation along the western edge of the TS. It would have to all be retained
in order to avoid any visual impact to users of Wood Road.

68
The first 1/4 mile of Wood Road, off of Highway 101, is along a mineral extraction area. This
mining area lies between the TS and Wood Road. It is doubtful there would be any visual impact
to users of the road. It appears the road acts as access to the mineral extraction area and to
timbered areas further inland.

V. Viewshed Delineation for an 8 Mile Portion of the Discovery Bay Trail.

An 8.28 mile section of the Discovery Bay Trail (DBT) is being evaluated by the "Support the
Trail Action Committee" (STAC). STAC represents a range of concerned citizens including
residents along the trail, real estate developers, outdoor recreationists, and timber companies.
STAC members appear to agree that vegetation management of open space and undeveloped
land parcels within 1/2 mile of the DBT needs to be addressed.

DBT hikers and bicyclists enjoy vistas of Sequim Bay and would like more vistas. Real estate
developers see opportunities for development of parcels currently in the undeveloped land
classification. DBT users would like vegetation along the trail adjacent to these parcels to shield
trail users from the developments. Some undeveloped land parcels are timbered but are not being
managed as commercial timber stands. Over the years these parcels have experienced storm
damage. There are timber salvage opportunities for these parcels.

Here are the four primary concerns of the STAC.

(1) Parcels for real estate development need to be shielded by vegetation from view of DBT
users.
(2) Visual impact to DBT users from parcels to be salvage harvested.
(3) Open space parcels need protection from real estate development and salvage timber
harvests.
(4) Protection of the foreground zone of the DBT.

The STAC has reviewed the Clallam County land use map with local real estate developers and
timber company sale planners and identified parcels of interest. Figure V-1 displays this map.

69
 Figure V-1. The Discovery Bay Trail (in green).

Land use classes are displayed in color in the map view window in Figure V-1. Light green is for
the timber land use class, medium blue is for residential, olive green is used for undeveloped
land, red for parks, and there are others displayed. Sequim Bay is displayed in bright blue. The
lime green linear feature is the DBT. The heavier black line that parallels the DBT is Highway
101. The eastern portion of the DBT in the map view window shares the Highway 101 ROW.
The lime green line representing the DBT overlaps on the heavy black line representing Highway
101.

Summary of Steps
1. Define a DBT project area grid system.
2. Define observer points for the DBT.
3. Preparing the top surface layer.
4. Prepare a viewshed grid map for the DBT.
5. Analyze the viewsheds relative to visual issues.

70
The first step is to define a sub-area of the much larger Sequim Bay grid system to serve as the
project area grid system for the DBT. Next, an observer point every 250' along the DBT must be
defined. Third, adjust elevations on the top surface elevation layer within the DBT ROW as
needed. Fourth, delineate the viewshed for the DBT. Fifth, the STAC concerns must be analyzed.

1. Define a DBT Project Area Grid System.

The map view window in Figure V-1 displays the area that will serve as the project area for the
DBT. You can see the DBT (displayed in lime green) enters the map view window on both sides
of the window and follows a route south around the bottom of Sequim Bay.

As noted earlier in this tutorial, one of the reasons for reducing the project area to a sub-area of
the much larger Sequim Bay grid system is due to the large volume of data created by the
LIDAR process. The raw data collected by this sensor is extremely dense. Re-sampling the data
to a 6' grid cell resolution reduces the volume but still results in a voluminous amount of data.
This can have an impact on software and system performance. If the detailed resolution is not
necessary for an application it is best to re-sample the data using a grid system based on a larger
grid cell size that would be more appropriate.

The interactive tool Grid - Tools/Clip Grids [interactive] is used to define this sub-area of the
larger Sequim Bay grid system. The map view window in Figure V-1 is displayed in the work
area. These are the data layers, in order, displayed in the map view window: DiscBayTrl2,
Highway101, ClallamRoads, SequimBay, and LandUseDBT.

Figure V-2 displays the parameters window for the Grid - Tools/Clip Grids [interactive] tool.

Figure V-2. Clipping the bare earth, top surface, and vegetation height grid maps using the
Grid - Tools/Clip Grids [interactive] tool.

This tool clips one or more existing grid maps that are members of the same grid system, the grid
system chosen for the 'Grid system' parameter. The Sequim Bay grid system that was defined

71
during the processing of the downloaded LIDAR files is chosen for the 'Grid system' parameter
(6; 5018x 6144y; 101235x 977274y). Three grid maps, members of the selected grid system, are
chosen to be clipped and are identified for the '>> Grids' input parameter. These are the bare
earth ('QSbeElev4'), top surface ('QStsVeg4'), and vegetation height ('QSVegHt4') grid maps.

The 'Run Once' parameter uses a toggle box for turning it on or off. When you enter a check
mark in the toggle box the option is turned on. After defining a rectangle, as described below, the
tool stops running. If this option off, after defining a rectangle, the tool continues to run. The
user may re-define a rectangle and must explicitly quit the tool.

I click with the mouse pointer on the 'Okay' button in the parameters window and the tool begins
execution.

This is an interactive tool. The 'Action' tool, on the toolbar, is used to interact with the tool while
it is executing. The 'Action' tool is applied in the map view window displayed in the work area.
A rectangular area is defined by moving the mouse pointer to a location near a corner. The left
mouse button is pressed and held down. The mouse pointer is dragged to near a corner point
diagonally opposite. The mouse button is released. This highlights and defines a rectangular area
that becomes the spatial extent defining a sub-area of the Sequim Bay grid system to be the
project area grid system for the DBT. The 'Clip to Extent' dialog window in Figure V-3 displays.

Figure V-3. The sub-area definition.

72
The four parameters 'Left', 'Right', 'Bottom', and 'Top' define the four sides of the interactively
defined rectangle. The 'Columns' and 'Rows' parameters display the number of columns and rows
definition using the original grid cell size of 6'. I accept the definition by clicking the mouse
pointer on the 'Okay' button.

A new grid system definition, using the parameters, is defined and appears in the 'Data' tab area
of the Manager.

6; 4842x 4472y; 1013418x 983382y

This new grid system is referred to as the project area for the DBT. The three grid maps chosen
for the input parameter '< Grids' are clipped by the spatial extent of the grid system and become
members of the new grid system. The default names for the new layers are the same as the input
names. They are re-named 'DBT3beElev8', 'DBT3tsVeg8', and 'DBT3VegHt8' and saved.

The interactive tool is still running. Execution of the tool is stopped by moving the mouse
pointer to the 'Geoprocessing' title in the Menu Bar and pressing the left mouse button. A drop-
down list of options displays. Near the bottom, the tool execution history for the work session, is
displayed.

The names of the tools in the history list are grayed out except for the first one, the "Clip Grids
[interactive]" tool. A check is showing to the left of its' name. The check indicates the tool is
currently executing. I move the mouse pointer over the tool name and press the left mouse
button. A 'Tool Execution' dialog window displays with the question: Shall execution be
stopped? I click on the 'Yes' button in the window and the tool execution stops. The check mark
disappears.

In summary, this step has resulted in the definition of a grid system for the DBT project area.
The three grid maps for bare earth ('DBT3beElev8'), top surface ('DBT3tsVeg8') and vegetation
height ('DBT3VegHt8') were defined as part of the new grid system.

2. Define Observer Points for the DBT.

An observer point is a point location from which areas of the landscape are identified that are in
view and not blocked from the observer by topography, man-made structures or vegetation. An
observation point is to be defined every 250 feet along the section of the DBT. The data layer
'DiscBayTrl2' is the line shapes data layer for the Discovery Bay trail.

The Shapes - Points/Convert Lines to Points tool defines point objects at specific distance
intervals along line objects of a line shapes data layer. One of the parameters for this tool is the
insertion method. There are three choices: per line segment, per line, and from line center'.

When the "per line segment" option is chosen, the vertices defining the line objects are
interpreted as end points of line segments. A point object is defined for each vertex and new
point objects, within the length of two adjacent vertices, are inserted using the insertion distance
entered for the 'Insertion Distance' parameter. In this example, the entry for the 'Insertion
Distance' parameter is 250. If the length between two adjacent vertices is greater than 250 feet

73
but less than 500, a single new point object is defined between the vertices 250 feet from one of
them.

The second option for the 'Insertion' parameter is "per line". In this case a line object is viewed as
a single line, regardless of the placement of vertices defining line segments defining the line
object. A new point object is defined at the distance interval specified by the entry for the
'Insertion Distance'. The vertices do not define new point objects, the exception being the two
end vertices of the line object. Point objects are defined on the basis of the distance interval
starting at one of the line end points.

The third option is "from line center". Point objects are defined at specific distance intervals
starting at the mid-point of the line object toward each end point of the line object.

The "per line" option produces point objects (observer points) at 250 foot space intervals along
the length of the trail if it is a single line object. The 'DiscbayTrl2' line shapes data layer is made
up of 14 line objects defining a 8.26 mile portion of the DBT. The 14 line objects can to be
aggregated to a single one using the Shapes - Lines/Line Dissolve tool.

Figure V-4 displays the parameters window for the Shapes - Lines/Line Dissolve tool.

Figure V-4. The parameters window for the Shapes - Lines/Line Dissolve tool.

The 'DiscBayTrl2' data layer is chosen for the input parameter '>> Lines'. The "Name" attribute
in the attribute table linked to the data layer contains the same entry for each object of the layer.
This is the attribute used to aggregate the 14 objects into a single one. The default "<create>" is
used for the output parameter '<< Dissolved Lines'. The default name for the output data layer is
'DiscBayTrl2 [Dissolved]'. I re-name it 'DiscBayTrl3' and save it.

74
The 'DiscBayTrl3' line shapes data layer, for the DBT project area, contains a single line object.
Figure V-5 displays the parameters window for the Shapes - Points/Convert Lines to Points tool.

Figure V-5. Defining observation point locations on the Discovery Bay Trail line shapes
data layer using the Shapes - Points/Convert Lines to Points tool.

The parameter entries in Figure V-5 define a point object every 250' along the length of a line
object. The line shape data layer for the DBT created by the Shapes - Lines/Line Dissolve tool,
'DiscBayTrl3', is chosen for the input parameter '>> Lines'. The default "<create>" option is used
for the output parameter '<< Points'. The default name for the output point shapes data layer uses
the name of the input layer.

The toggle box 'Insert Additional Points' parameter is turned on (the box contains a check mark).
When the parameter is off, the tool produces a point object for each vertex that defines the line
objects including the end point vertices of the line objects of the input data layer.

When the 'Insert Additional Points' parameter is on, two additional parameters are displayed:
'Insertion' and 'Insertion Distance'. The 'Insertion' parameter determines how additional point
objects are to be generated. The three options are: per line segment, per line, and from line
center. These options were described earlier.

The second option for the 'Insertion' parameter, "per line", is chosen. In this case the line object
is viewed as a single line, regardless of the placement of vertices defining line segments defining
the line object. A new point object is defined at the distance interval specified by the entry for

75
the 'Insertion Distance'. The distance interval entered is 250'. The vertices do not define new
point objects, the exception being the two end vertices of the line object.

The toggle box parameter 'Add Point Order' is used when the numerical order of the point
objects in the input line object needs to be captured as an output in the attribute table linked to
the output point shapes data layer.

I execute the tool and save the default output point shapes data layer 'DiscBayTrl3' to
'DiscBayTrl3Pts' and save it. This data layer contains 176 point objects.

When applying the Shapes - Lines/Line Dissolve and Shapes - Points/Convert Lines to Points
tools as described above you may notice a few places where the distance interval between
adjacent point objects is less than the value entered for the 'Insertion Distance' parameter.
Related to delineating a viewshed, this discrepancy is probably not an issue unless the distance
interval is determined on a statistical basis. Hopefully, the problem is corrected in a future
release of SAGA. The problem appears to happen due to the way the tool "aggregates" the line
objects into what appears to be a multi-line line object rather than a single line object.

3. Preparing the Top Surface Layer.

The bare earth grid map is based on LIDAR returns interpreted as representing the bare earth or
ground. The top surface grid map is based on LIDAR returns interpreted to represent a surface
above the ground, such as vegetation over-story or structure tops. The difference between bare
earth and top surface elevations for the same grid cell is vegetation height or the height of a
building or structure.

Related to viewshed delineation, the topography defined by bare earth elevations can cause the
view from an observer to be blocked. When a top surface elevation layer is input, the topography
defined by the elevations and the additional elevation of vegetation or buildings can cause the
view of an observer to be blocked. Using top surface elevation as input allows for the blocking
effect of vegetation and buildings to be a factor. This is not the case with bare earth or USGS
DEM's.

Top surface elevations are going to be used as input with the Terrain Analysis - Lighting,
Visibility/Visibility (points) tool for defining a viewshed. The other input is a set of observation
points. Each observation point is going to be 6' above the ground (roughly representing eye-
level). Without making adjustments to the top surface elevation input model this means the
observation points are 6' above the vegetation. The intent is that the observers be positioned
above the ground or bare earth, not above the vegetation height.

Before the top surface 'DBT3tsVeg8' grid map can be used for the elevation input, the elevation
values within the DBT right-of-way (ROW) must be bare earth elevations and not elevation
values that include vegetation height. Otherwise, as noted above, the observer points are going to
be 6 feet above the vegetation negating the potential for vegetation blocking or shielding for a
ground observer. An alternative to adjusting the elevations within the DBT ROW would be to
adjust the elevations only where the observers are located.

76
I am not sure what the ROW width is for the DBT but six feet seems reasonable. The first step to
replace the top surface elevations in the DBT ROW with bare earth elevations is to define a
polygon buffer for the DBT ROW. The Shapes - Tools/Shapes Buffer tool can define this
polygon. The parameters window for the tool is displayed in Figure V-6.

Figure V-6. The Shapes - Tools/Shapes Buffer tool parameters window.

The line shape data layer for the DBT in the project area is named 'DiscBayTrl3'. This layer is
chosen for the input parameter '>> Shapes'. A buffer width is entered for the 'Default' parameter.
The width entered is 3 and is half of the final polygon width as the buffer is defined on both sides
of a line object. The "<create>" option is chosen for the output parameter '<< Buffer'. The default
settings are accepted for the options parameters. I click the mouse pointer on the 'Okay' button.

The tool executes creating an output polygon shapes data layer named 'DiscBayTrl3 [Buffer]'.
Before re-naming the layer I changed the data entry for the "ID" attribute from 0 to 1. I re-name
it to 'DiscBayTrl3Buffer'.

A grid data layer of the 'DiscBayTrl3Buffer' is used in the Grid Calculus tool to identify grid
cells of the 'DBT3tsVeg8' layer that fall within the DBT ROW. The grid map version is created
with the Grid - Gridding/Shapes to Grid tool. I choose the 'DiscBayTrl3Buffer' shapes data layer
for the input parameter '>> Shapes'. I choose the "attribute" option for the 'Output Values'
parameter and select the "ID" attribute for the 'Attribute' parameter. The data value 1 is used for
this attribute in the attribute table. I choose the DBT project area grid system (6; 4842x 4472y;
1013418x 983382y) for the 'Grid System' parameter and the default "<create>" for the output
parameter '<< Target Grid'. The default name for the output grid map is 'DiscBayTrl3Buffer
[ID]'. I re-name the layer 'DiscBayTrl3BufferGR' and save it.

77
The Grid - Calculus/Grid Calculator tool is used to replace the elevation data values within the
DBT ROW on the 'DBT3tsVeg8' grid map with elevation values of the 'DBT3beElev8' grid map.
The parameters window is displayed in Figure V-7.

Figure V-7. The Grid - Calculus/Grid Calculator tool parameter entries to create the
'DBT3tsBuffVeg8' grid map.

The DBT project area grid system is chosen for the 'Grid system' parameter. Three grid maps are
entered for the input parameter '> Grids'. They include the grid map for the DBT ROW (the 3'
ROW on either side of the trail, a total 6' wide ROW), the bare earth DEM 'DBT3beElev8' and
the top surface DEM 'DBT3tsVeg8'. The default entry "<create>" is used for the output
parameter '<< Result'. The equation "ifelse(eq(g1,1),g2,g3)" is entered for the 'Formula'
parameter. The references "g1", "g2", and "g3" are variables representing the input grid maps
corresponding to their listing order. The name for the output grid map is entered for the 'Name'
parameter: 'DBT3tsBuffVeg8'.

The Grid Calculator tool has two choices for dealing with no data values in grid maps. All of the
grid cells outside of the DBT ROW of the 'DiscBayTrl3BuffGR' grid map contain the no data
value -99999 and all the ROW grid cells contain the valid data value 1. The default is that any
grid cell containing the no data value also produces a no data value for the corresponding grid
cell of the output grid map. The 'Use NoData' parameter provides an option for the tool to
interpret no data values as data values rather than no data values. When this option is turned on,
the no data value is treated just like a regular data value by the tool and by the equation. I click
with the mouse pointer on the toggle box for the 'Use NoData' option and a check mark appears
in the box. This means the option is in the on status.

78
I click on the 'Okay' button and the tool produces the new 'DBT3tsBuffVeg8' grid map. I save it
to a storage folder. Remember the only difference between this data layer and the 'DBT3tsVeg8'
data layer is that the grid cells of the DBT ROW contain bare earth elevations from the
'DBT3beElev8' data layer. All other elevations are top surface elevations.

4. Prepare a Viewshed Grid Map for the DBT.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is one of two SAGA tools that
produces a viewshed grid map. This tool accepts multiple observation points and produces an
aggregate viewshed grid map for all of the input observation points.

The other SAGA tool used for creating a viewshed is the Terrain Analysis - Lighting,
Visibility/Visibility (single point) [interactive] tool. One of the differences between the Visibility
(points) and Visibility (single point) [interactive] tools is how the height of the observer above
the ground is entered. Due to the nature of the interactive tool, the height for the single
observation point is entered as a data value for a 'Height' parameter. The same height value is
used for each observer. However, the version supporting multiple observation points, defined as
point objects in a point shapes data layer, requires that the attribute table linked to the point
shapes data layer include an attribute for the height of the observation point above the ground.
The height for the observation points can vary. For this tutorial example, the point shapes data
layer is 'DiscBayTrl3Pts'.

The height of our observers above ground, 6', is the same for each point. The table function "Add
Field" is used to add a second attribute to the attribute table linked to the 'DiscBayTrl3Pts' named
"HEIGHT". I use the data type "8 byte floating point number" for the new field. The attribute
name is "HEIGHT". I use the Table - Calculus/Field Calculator (Shapes) tool to enter the data
value 6 for the "HEIGHT" attribute for each of the 176 point objects of the 'DiscBayTrl3Pts'
layer. Figure V-8 displays the parameters window for this tool.

79
Figure V-8. Parameter entries entering data for the "HEIGHT" attribute of the
'DiscBayTrl3Pts' shapes data layer.

The "f2" in the formula "f2 + 6" is how you reference within the formula an attribute in the
attribute table of the input shapes data layer. The 2 is the position of the attribute in the table. I
have chosen the ' DiscBayTrl3Pts' point shapes data layer for the input parameter '>> Shapes'.
The second attribute ("f2") in the attribute table is the "HEIGHT" attribute. I use the "<not set>"
option for the output parameter '< Result'. The output goes to the "HEIGHT" attribute in the
existing attribute table linked to the 'DiscBayTrl3Pts' data layer. The data layer is re-saved.

The point objects of the ' DiscBayTrl3Pts' are the observation points used with the Terrain
Analysis - Lighting, Visibility/Visibility (points) tool for delineating an aggregate viewshed for
the DBT. Using this application with other projects that use a small grid cell size, I have
discovered that point shapes data layers containing 20 or less point objects for observation points
are more practical than a single layer containing 176 point objects. I did use the 'DiscBayTrl3Pts'
layer as input in a test. After 12 hours of execution the tool had not finished its task. In
anticipation of the need to use data layers containing 20 or less point objects I am going to create
these layers now.

I add a new attribute named "Group" to the attribute table linked to the ' DiscBayTrl3Pts' data
layer. I add it as the last attribute in the table and use the "signed 8 byte integer" data type. I
decide to create 11 groups, most of the groups contain 17 objects, and one has 6. Record 1 of the
attribute table is for the first point object on the east end of the trail. Record 2 is for the adjacent
point object following it, and so on. The 176th record is for the point object on the west end of
the trail.

I manually enter a 1 for the "Group" attribute for the first 17 records, a 2 for records 18 through
35, a 3 for records 36 through 53, etc., until I have made an entry for the "Group" attribute for
each of the 176 records.

80
The Shapes - Tools/Split Table/Shapes by Attribute tool is used to create a new point shapes data
layer for each group, 1 through 11. Figure V-9 displays the parameters window for this tool.

Figure V-9. Creating point shapes data layers using the "Group" attribute.

The 'DiscBayTrl3Pts' point shapes data layer is chosen for the input parameter '>> Table /
Shapes' and the new attribute "Group" is identified for the 'Attribute' parameter. There are 11
unique values for the "Group" attribute. The tool generates a new point shapes data layer for each
unique value. The default data layer names include references to the "Group" data value. For
example, the output default data layer for group 1 is named 'DiscBayTrl3Pts [Group = 1]'. I re-
name these default data layers to 'DiscBayTrlgrp01' through 'DiscBayTrlgrp11'. I save these new
point shapes data layers. I am now ready to create an aggregate viewshed for the DBT.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool uses the 'DBT3tsBuffVeg8' grid
map to provide the elevations for the viewshed delineation. There will be 11 execution of the
tool, the difference in each one is the entry for the '>> Points' input parameter. A different group
point shapes data layer is used for each execution. An aggregate viewshed grid map is developed
using the Grid - Calculus/Grid Calculator tool. This is described later.

Figure V-10 displays the parameters window for the Terrain Analysis - Lighting,
Visibility/Visibility (points) tool.

81
Figure V-10. The Terrain Analysis - Lighting, Visibility/Visibility (points) tool parameters
window.

The DBT project area grid system is chosen for the 'Grid system' parameter. The
'DBT3tsBuffVeg8' elevation layer is selected for the input '>> Elevation' parameter. The
"<create>" option is used for the output parameter '<< Visibility'.

The observation points are the point objects of the 'DiscBayTrlgrp01' data layer. The "HEIGHT"
attribute containing the height above the ground for the observation points is chosen for the
'Height' parameter.

There are four options for the 'Unit' parameter: Visibility, Shade, Distance, and Size. I choose the
"Visibility" option. This option is used to delineate a viewshed. I execute the tool using these
parameter entries.

The default output grid map is named 'Visibility'. Once this output is produced by the tool (it
takes about 1 hour of processing time on my laptop) I re-name the data layer 'DBTvis001' and
save it. I re-name each of the 'Visibility' data layers using the same naming convention. This tool
is executed eleven times resulting with new data layers 'DBTvis001' through 'DBTvis011'.

The viewshed grid maps each have two data values. Grid cells that can be seen from any of the
point objects in the group contain a data value of 1. This means the grid cell is in the seen-area.
Grid cells that cannot be seen contain the data value 0. The goal is to merge the 11 group-based
viewsheds into a single aggregate viewshed where seen grid cells contain the data value 1 and

82
not seen grid cells contain the data value 0. Developing this aggregate viewshed is a two step
process using the Grid - Calculus/Grid Calculator tool.

The first step is to merge the 11 viewsheds (grid maps 'DBTvis001' through 'DBTvis011')
together into a single aggregate grid map. The easiest method is to use an add equation for the
'Formula' parameter in the Grid Calculator tool. The grid map produced will have data values 1
and up to 11 depending on the amount of overlap between the viewsheds. The second step is to
convert data values 1 and greater, of this first preliminary aggregate viewshed, to 1's, also using
the Grid Calculator.

Figure V-11 displays the parameters window for the Grid - Calculus/Grid Calculator tool.

Figure V-11. The parameter entries for the Grid - Calculus/Grid Calculator tool to create
the 'DBTvisiInterim' grid map.

The entries displayed for the parameters in Figure V-11 produce an interim grid map that is an
aggregate of the 11 group viewsheds. These viewshed grid maps are chosen for the input
parameter '> Grids'. The default "<create>" option is used for the output parameter '<< Result'.
The name for the output grid map, "DBTvisiInterim", is entered for the 'Name' parameter.

The equation used for the 'Formula' parameter is a simple addition equation. The references "g1"
through "g11" are variables each of the 11 grid maps chosen for the input. I execute the tool and
the new data layer 'DBTvisiInterim' is created and saved.

The data value range for the output grid map 'DBTvisiInterim' is from 0 to 11. This means there
are some portions of viewsheds on each input grid map that overlap. Figure V-12 displays the
parameters for the Grid - Calculus/Grid Calculator tool used to change data values between 1
and 11 to 1.

83
Figure V-12. The parameters window for the Grid - Calculus/Grid Calculator tool to edit
data values and create the 'DBTvisiAggregate' grid map.

The interim grid map created by the Grid Calculator 'DBTvisiInterim' is chosen for the input
parameter '> Grids'. The default "<create>" entry is used for the output parameter '<< Result'.
The name for the output grid map is 'DBTvisiAggregate' and this name is entered for the 'Name'
parameter.

The equation used, "ifelse(gt(g1,0),1,0)", checks a grid cell data value of the input grid map to
see if the value is greater than 0 "(gt(g1,0)". If this condition is true, the tool outputs a data value
of 1 to the corresponding grid cell of the output grid map. If the value being checked is not
greater than 0, in this case the only other data value it can be is 0, than the condition is false and
a data value of 0 is output to the corresponding grid cell of the output grid map. I move the
mouse pointer to the 'Okay' button and press the left mouse button. The tool executes and
produces the new grid map 'DBTvisiAggregate'. I save the new layer.

Note that the Grid - Tools/Reclassify Grid Values could also have been used to recode the data
values 1 through 11 to the new data value 1.

Figure V-13 displays the 'DBTvisiAggregate' viewshed for the DBT project area grid system.

84
 Figure V-13. The aggregate viewshed of the DBT for the DBT project area.

The reddish tinted areas (and the purple shade on Sequim Bay) are the seen-areas of the
aggregate viewshed. The DBT observation points are displayed with the small yellow circles.

5. Analyze the Viewsheds Relative to Visual Issues.

Here are the four primary concerns of the STAC introduced at the beginning of this section of
the tutorial.

(1) Parcels for real estate development need to be shielded by vegetation from view of DBT
users.
(2) Visual impact to DBT users from parcels to be salvage harvested.
(3) Open space parcels need protection from real estate development and salvage timber
harvests.
(4) Protection of the foreground zone of the DBT.

There are three land use categories of primary interest to the STAC. These are open space,
timber, and undeveloped. The parcels in these land use categories can be selected from the
85
county land use layer and placed into their own polygon shapes data layer. The Shapes -
Tools/Select by Attributes... (String Expression) tool is used to select the polygon objects, by land
use class, for this new polygon shapes data layer. Figure V-14 displays the parameters window
for this tool.

Figure V-14. Selecting the "OpenSpace" land use type on the 'LandUseDBT' data layer
with the Shapes - Tools/Select by Attributes... (String Expression) tool.

The Shapes - Tools/Select by Attributes... (String Expression) tool is executed three times, once
for each land use class (open space, timber, and undeveloped). The parameters for the first
execution are displayed in Figure V-14.

The 'LandUseDBT' polygon shapes data layer is chosen for the input parameter '>> Shapes'. The
"GENLU" attribute is chosen for the 'Attribute' parameter. The text entries for this attribute are
the land use categories including the three of interest: "OpenSpace", "Timber", and "Undevel".
The "OpenSpace" attribute name is entered for the 'Expression' parameter. This is the first
execution of the tool in a series of three executions so the "new selection" option is used for the
'Method' parameter. The 'Okay' button is clicked on with the mouse pointer and the tool
executes.

The tool is executed and all parcels having the general land use category of "OpenSpace" for the
"GENLU" attribute are selected and highlighted in the data layer. The second and third
executions differ in that the general land use categories used are "Timber" and "Undevel". The
other difference is that the "add to current selection" option is used for the 'Method' parameter.
This means that the polygons previously selected remain selected, and the polygons selected with
the new parameters are added in to those already selected.

86
Following the third execution of the Shapes - Tools/Select by Attributes... (String Expression)
tool, the Shapes - Tools/Copy Selection to New Shapes Layer tool is used to copy the selected
land use parcel polygons into a new polygon shapes data layer. The new layer is named
'LandUseOpUnTb' and saved.

The STAC has been successful in having a foreground zone defined for the DBT. The zone is 1/8
mile (660') either side of the DBT. A polygon object representing this zone is in the
'DBTforeground' polygon shapes data layer. The intent is for the zone to be a criteria for various
activities specific to proximity to the DBT.

Figure V-15 displays a zoomed in area of the DBT project area.

Figure V-15. The 660' foreground zone along the DBT.

Figure V-15 displays two zoomed in areas in the DBT project area. The heavy yellow linear
feature across the middle of the map view window is the DBT. Land parcels (on the
'LandUseOpUnTb' data layer) tinted in green are in timber land use, areas tinted in red are
undeveloped, and areas tinted in blue are in open space land areas. The foreground zone of the
DBT is displayed using yellow backward diagonals.

The area on the right is a zoomed in area near the middle of the right side of the left map view
window. The STAC sees an opportunity for using GIS tools to assess an area of common
interest. You can see an open space land parcel (tinted in blue) near the middle of the right map

87
display window. It occupies about 1/4 of an adjacent area designated as undeveloped.
Immediately to the west is a parcel in timber land use. Figure V-16 displays this area.

Figure V-16. An area of focus for the STAC.

The DBT viewshed grid map 'DBTvisiAggregate' is displayed with a yellow tint in the map view
window on the right. The open space parcel (the blue tinted parcel near the middle) is adjacent to
timber land use on the west and north sides and undeveloped land use on the east and south
sides. Roads display in black. There is not road access to the open space parcel as there is for the
adjacent parcels.

A local timber company wants to harvest the portion of the timber land use parcel on the west
side of the open space parcel and undeveloped parcel south to near the DBT. The northern
boundary of this harvest area would be an extension of the northern boundary of the open space
parcel extended to the west joining the small undeveloped parcel (located about 660' to the west).
In addition, a local land developer is interested in developing the undeveloped land parcel on the
east and south of the open space parcel but not quite all the way to the DBT.

Viewing the two map view windows in Figure V-16, STAC members are very interested in
looking at viewsheds using an observer point interval of 125 feet as compared to 250 feet used
for creating the 'DBTvisiAggregate' viewshed data layer.

The STAC decides on the following tasks:

88
(1) Identify observation points every 125 feet along the section of the DBT immediately to the
south of the several land parcels of interest.
(2) Define a "before" viewshed using these observation points.
(3) Harvest and/or remove vegetation cover from the parcel to the west of the open
space/undeveloped parcel and from the undeveloped parcel.
(4) Define an "after" viewshed using the same observation points but with the top surface
elevation layer adjusted for the harvest and/or vegetation cover removal.
(5) Compare the before and after viewsheds.

(1) Identify observation points every 125 feet along the section of the DBT immediately to the
south of the several land parcels of interest.

A polygon object needs to be defined that can be used to clip the portion of the DBT
immediately to the south of the land parcels being evaluated. First a new polygon shapes data
layer is created, the polygon object is defined, and the polygon shapes data layer is used with the
Shapes - Polygons/Polygon Clipping tool to produce a line object for the section of the DBT.

The Shapes - Tools/Create New Shapes Layer tool is used to create a new polygon shape data
layer named 'DBTeast'. The 'DBTeast' polygon shapes data layer is displayed in a map view
window in the work area. In the same map view window is also displayed these data layers:
DiscBayTrl2, ClallamRoads, DBTforeground, LandUseOpUnTb and DiscTrlEastReliefW2. It
happens that these are the same layers in the map view window in Figure V-16. The reason for
displaying these layers is to support the identification of the boundaries for a new polygon, one
to be used to "clip" the section of the DBT.

The 'DBTeast' polygon shapes data layer is made the active layer and the 'Action' tool is chosen
from the Tool Bar. The 'Action' tool (mouse pointer) is moved within the map view window (I
have zoomed in to the same area of Figure V-16) and press the left mouse button. I press the
right mouse button and a pop-up list of edit options displays. I choose the "Add Shape" option. I
screen digitize a polygon (approximating a rectangle) so that the section of the DBT I want to
clip falls within the polygon area. I actually make the rectangle slightly larger than needed.
When I am happy with the polygon definition I press the right mouse button and exit from the
"Add Shape" mode. When the dialog window with the question "Apply changes?" appears I click
on the 'Yes' button. I go to the 'Data' tab area of the Manager window and save the modified
'DBTeast' data layer.

A polygon object on the shapes data layer named 'DBTeast' now exists that can be used to create
a new line shapes data layer containing a single line object for a section of the DBT.

Figure V-17 displays the parameters window for the Shapes - Polygons/Polygon Clipping tool.

89
 Figure V-17. The parameters window for the Shapes - Polygons/Polygon Clipping tool.

The new polygon shapes data layer 'DBTeast' is chosen for the '>> Clip Features' input parameter
and the line shapes data layer for the DBT, 'DiscBayTrl3', is chosen for the other input parameter
'>> Input Features'. I execute the tool and a new line shapes data layer named 'DiscBayTrl3
[clipped]' is produced. I re-name it 'DBTsection' and save it. This new line shapes data layer has
a single line object for the section of the DBT falling within the 'DBTeast' polygon.

The Shapes - Points/Convert Lines to Points tool is used to create a point shapes data layer
containing the needed observation points. Figure V-18 displays the parameters window for the
tool.

90
Figure V-18. The parameter entries for the Shapes - Points/Convert Lines to Points tool to
create observer point objects along the DBT.

The new line shapes data layer for the section of the DBT, 'DBTsection', is chosen for the input
parameter '>> Lines'. The "<create>" option is used with the output parameter '<< Points'. The
'Insert Additional Points' parameter is a toggle check box. I move the mouse pointer into the box
and press the left mouse button. A check mark appears indicating the option is now in "on"
status. At the same time two new option parameters, 'Insertion' and 'Insertion Distance', appear in
the parameters window. The option "per line" is used for the 'Insertion' parameter and the
distance value 125' is entered for the insertion distance in the value field to the right of the
'Insertion Distance' parameter label. The tool is executed using these parameter entries and a new
point shapes data layer using the layer name 'DBTsection' appears in the list of shapes data layers
in the 'Data' tab area of the Manager. This new layer contains 27 point objects.

I need to add an attribute for "HEIGHT" to the new data layer. I will be using the Terrain
Analysis - Lighting, Visibility/Visibility (points) tool for defining the before and after viewsheds.
One of the tool requirements is that the point shapes data layer providing the observation points
also include an attribute in its linked attribute table for observer height above the ground. I create
the new attribute "HEIGHT" in the attribute table linked to the data layer using the table column
tool "Add Field". I choose the data type "8 floating point number". Once the new attribute is
available I manually key the data value 6 into the attribute field for the 27 records or point
objects. It is faster to manually key this data in, however, in the case where there were many
records I could use the Table - Calculus/Field Calculator (Shapes) tool as I have in other
sections of this tutorial.

I re-name the data layer to 'DBTsectionPts' and save it.

91
(2) Define a "before" viewshed using these observation points.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool requires two input data layers
for delineating an aggregate viewshed. One is the point shapes data layer ('DBTsectionPts')
containing point objects for observation points that was just created and the second is an
elevation data layer.

The top surface elevation grid map 'DBT3tsBuffVeg8' is going to be chosen for the elevation
data input. You will recall observers are 6' above the ground. Using a ROW of 6' (3' on either
side of the DBT) the top surface elevations within the DBT ROW have been replaced by the bare
earth elevations of the 'DBT3beElev8' grid map. The 'DBT3tsBuffVeg8' grid map provides the
elevation data for the Terrain Analysis - Lighting, Visibility/Visibility (points) tool.

I choose the DBT project area grid system for the 'Grid system' parameter for the tool. I make the
following parameter entries: the 'DBT3tsBuffVeg8' grid map for the '>> Elevation' input
parameter, the "<create>" option for the output parameter '<< Visibility', the 'DBTsectionPts'
point shapes data layer for the input parameter '>> Points' and the "HEIGHT" attribute for the
'Height' parameter. I execute the tool.

The output grid map default name is 'Visibility'. I re-name it 'BeforeVisi1'. Figure V-19 displays
a map view window that includes the 'BeforeVisi1' aggregate viewshed displayed in a yellow
tint. The same map view window also displays the land use parcels (open space in blue tint,
timber in green tint, and undeveloped in red tint), the DBT, and the observation points (blue
circles) on the DBT. The background for the map view window is topographic relief.

92
 Figure V-19. The 'BeforeVisi1' aggregate viewshed.

(3) Harvest and/or remove vegetation cover from the parcel to the west of the open
space/undeveloped parcel and from the undeveloped parcel.

The process for this step is very similar to the one used for replacing the DBT ROW elevation
values on the top surface elevation layer with elevation values from the bare earth elevation
layer. Rather than creating a ROW polygon, I create a polygon shapes data layer with one
polygon object that defines the area for vegetation removal. This new shapes data layer is named
'VegMod'.

93
The Shapes - Tools/Create New Shapes Layer tool is used to create the new data layer and
include an attribute for a floating point data value in the linked attribute table. The attribute is
named "VEG". The data value 1 is going to be entered for the attribute.

The 'Action' tool from the toolbar and the earlier described on-screen digitizing tools, e.g., "Add
Shape", etc., are used to define a polygon object for the new data layer. The polygon boundaries
for the polygon object are to be aligned to parcel boundaries. The '>> Snap to...' parameter in the
'Settings' tab area of the 'Object Properties' window for the 'VegMod' layer is edited. The
'LandUseOpUnTb' polygon shapes data layer is identified as the background layer. It is used to
align digitized points of new polygon objects to existing polygon objects of the background
layer. The default value "10" is accepted for the 'Snap Distance' parameter. One polygon object
is digitized for the vegetation removal area on the new polygon shapes data layer named
'VegMod'. Figure V-20 displays this polygon object using a heavy red outline.

94
 Figure V-20. The vegetation removal polygon object.

Here is a text description of this polygon object. The south boundary of the new polygon
parallels the DBT (about 135' to the north) for almost 700'. At that point the DBT angles to the
southwest, the polygon boundary continues in a west direction. The south portion of the polygon
falls within the foreground zone. Between the south boundary of the polygon and the DBT there
exists a narrow swath of vegetation that is not to be removed. The intent is for this swath of
vegetation to provide a default level of vegetation shielding for the observers on the DBT.

The same map view window (Figure V-20) also displays the land use parcels (open space in blue
tint, timber in green tint, and undeveloped in red tint), the DBT (in yellow), the observation
points (blue circles) on the DBT, the foreground zone (yellow back diagonals) and the vegetation
removal polygon (red outline). The background for the window is topographic relief. A lot of
map information is displayed in this map view window. Note that the observer points (in blue)
display over the DBT (in yellow). The DBT parallels the highway.

95
The Grid - Calculus/Grid Calculator tool is used to develop a new version of the
'DBT3tsBuffVeg8' layer that has the top surface elevation values changed to the bare earth
elevation data values for the area of vegetation removal. This simulates the harvesting of the
vegetation. Before the 'VegMod' polygon shapes data layer can be used in this process, the layer
must be converted to a grid map using the Grid - Gridding/Shapes to Grid tool.

In the parameters window for the Grid - Gridding/Shapes to Grid tool the 'VegMod' polygon
shapes data layer is selected for the '>> Shapes' input parameter. The entry for the 'Output
Values' option parameter is set to "attribute". The 'Attribute' parameter then becomes available in
the parameters window. The "VEG" attribute is selected for the 'Attribute' parameter. The "grid
or grid system" option is chosen for the 'Target Grid System' parameter and the grid system for
the DBT project area is chosen for the 'Grid System' parameter. The default "<create>" option is
used for the '<< Grid' output parameter. The tool is executed and the output grid map is named
'VegMod [VEG]'. It is re-named to 'VegModGR' and saved. Grid cells that are in the parcel area
that will have vegetation removed have a grid cell value of 1. All other grid cells have the no
data value -99999.

Figure V-21 displays the parameters window for the Grid - Calculus/Grid Calculator tool.

Figure V-21. The parameters window for the Grid - Calculus/Grid Calculator tool to create
the 'DBT3tsBuffVeg9' grid map.

The entries for the parameters in the parameters window for the Grid Calculator are very similar
to the ones used with the DBT ROW earlier. The grid maps for the input parameter '> Grids'
include the 'VegModGR', 'DBT3beElev8' (also used earlier with the ROW), and the
'DBT3tsBuffVeg8' layer. The equation entered for the 'Formula' parameter is identical. The
output grid map is to be named 'DBT3tsBuffVeg9'.

96
The tool is executed and the output grid map 'DBT3tsBuffVeg9' saved.

(4) Define an "after" viewshed using the same observation points but with the top surface
elevation layer adjusted by the harvest and/or vegetation cover removal.

The Terrain Analysis - Lighting, Visibility/Visibility (points) tool is executed to create the "after"
aggregate viewshed. The 'DBT3tsBuffVeg9' grid map is used for the input parameter '>>
Elevation' and the 'DBTsectionPts' will again provide the observation points. The "HEIGHT"
attribute in the attribute table linked to the point shapes data layer provides the observer height
above the ground for each observer. I execute the tool and the default output grid map 'Visibility'
is produced. I re-name the layer 'AfterVisi1' and save it.

Figure V-22. The 'AfterVisi1' aggregate viewshed.

97
Figure V-22 displays a map view window that includes the 'AfterVisi1' aggregate viewshed
displayed in a yellow tint. The same map view window also displays the land use parcels (open
space in blue tint, timber in green tint, and undeveloped in red tint), the DBT (in yellow), the
observation points (blue circles) on the DBT, the vegetation removal polygon (red outline), and
the foreground zone (light yellow back diagonals). The background for the window is
topographic relief.

(5) Compare the before and after viewsheds.

Comparing Figures V-19 and V-22 one can discern some differences. More grid cells are in the
seen-area in the "after" view shed than in the "before" viewshed. The overall differences between
the two viewsheds can be quantified and the differences specific to the open space parcel can be
emphasized.

The Spatial and Geostatistics - Grids/Zonal Grid Statistics tool is used to quantify the difference
in seen areas between the before and after viewsheds for the DBT observer group having 27
observation points. Table V-1 displays a summary of this data.

Table V-1: Comparing "Before" Viewshed Acres and "After" Viewshed Acres

The differences in Table V-1 are directly attributable to removal of vegetation. The locations and
number of observation points did not change between the "before" and "after" viewsheds. The
only change was the replacement of top surface vegetation elevations with bare earth elevations
to simulate vegetation removal for the "after" aggregate viewshed.

The STAC is more interested in the open space land parcel. The parcel is 5.1 acres in size. I can
select the parcel on the 'LandUseOpUnTb' polygon shapes data layer, copy it to its' own data
layer, and use it to clip the 'BeforeVisi1' and 'AfterVisi1' aggregate viewshed grid maps. Then I
can build a table, similar to Table V-2, for comparing acres before and after for the one open
space land use parcel.

I use the 'Action' tool from the toolbar to select and highlight the specific open space land parcel
polygon on the 'LandUseOpUnTb' polygon shapes data layer. Then I execute the Shapes -
Tools/Copy Selection to New Data Shapes Layer tool. I name this new polygon shapes data layer
'OpenSpc1' and save it. The Shapes - Grid Tools/Clip Grid with Polygon tool uses the
'OpenSpc1' polygon shapes data layer to clip the 'BeforeVisi1' and 'AfterVisi1' grid maps. In that
tool, the DBT project area grid system is chosen for the 'Grid system' parameter and the
'BeforeVisi1' and 'AfterVisi1' grid maps chosen for the '>> Input' parameter. The 'OpenSpc1'
polygon shape data layer is selected for the input parameter '>> Polygons'. This is the new
polygon shapes data layer having the single polygon object representing the open space land use
parcel.

98
The Clip Grid with Polygon tool results with a new grid system whose spatial extent is based on
the spatial extent of the polygon shapes data layer chosen for the '>> Polygons' input parameter. I
re-name the default output grid maps 'BeforeVisi1' and 'AfterVisi1' to 'BeforeVisi2' and
'AfterVisi2' and save them.

I use the Spatial and Geostatistics - Grids/Zonal Grid Statistics tool to collect the grid cell counts
for the 'BeforeVisi2' and 'AfterVisi2' grid maps. I entered the grid cell counts into a spread sheet
and calculated the acres. The total acres for the parcel is 5.09. Table V-2 displays the acre
results.

Table V-2: Comparing the Open Space Parcel Viewshed Acres

Using the data in Table V-2 you can see that about 5% of the open space parcel was visible in the
"before" viewshed and 21% is visible in the "after" viewshed.

STAC members have explored the DBT viewshed and how it can be used to address a variety of
issues that exist. The example involving the open space parcel is very similar to a large number
of similar situations STAC members have discussed. They believe that SAGA tools, tools, and
functions can really help them address many of these issues and support their decision-making
process.

References
Carter et al., 2012. Lidar 101: An Introduction to LIDAR Technology, Data, and Applications.
National Oceanic and Atmospheric Administration (NOAA), Coastal Services Center. 76 pp.

Conrad, O., Bechtel, B., Bock, M., Dietrich, H.,Fischer, E., Gerlita, L., Wehberg, J., Wichmann,
V., and Boehner, J. (2015): System for Automated Geoscientific Analyses (SAGA) v. 2.1.4.
Geosci. Model Dev., 8, 1991 - 2007, doi:10.5194/gd-8-1991-2015.

Environmental Systems Research Institute, 2011. Lidar Analysis in ArcGIS 10 for Forestry
Applications. An Esri White Paper. 53 pp.

LiDAR Bare Earth/Top Surface DEMs (computer files). 2001. The Woodlands, TX: Terrapoint.
Available: Puget Sound LiDAR Consortium, Seattle, WA.
http://pugetsoundlidar.ess.washington.edu/index.htm

VIEWIT: computation of seen-areas, slope, and aspect for land-use planning. Michael R. Travis,
Gary H. Elsner, Wayne D. Iverson, and Christine G. Johnson. Pacific Southwest Forest and
Range Experiment Station, Forest Service, U.S. Department of Agriculture, Berkeley, California.
USDA Forest Service, General Technical Report PSw-11 / 1975. 70 pages.

99