www.whiteboxgeo.com | support@whiteboxgeo.com

WhiteboxTools Version 2.3.0
Dr. John B. Lindsay © 2017-2023
Geomorphometry and Hydrogeomatics Research Group
University of Guelph
Guelph, Canada
March 25, 2023


Sponsored by:

Introduction

WhiteboxTools is an advanced geospatial data analysis platform created by Prof. John Lindsay at the University of Guelph's Geomorphometry and Hydrogeomatics Research Group (GHRG). The project began in January 2017 and quickly evolved in terms of its analytical capabilities. The WhiteboxTools homepage contains more project information about the platform and the software download site.

Project Highlights

• Contains more than 480 tools for processing various types of geospatial data.
• Many tools operate in parallel, taking full advantage of your multi-core processor.
• Written in the safe and cross-platform systems programming language Rust and compiled to highly efficient native code.
• Small stand-alone application with no external dependencies, making installation as easy as downloading the 8Mb zip file and decompressing it.
• Simple yet powerful Python scripting interface that allows users to develop custom scripted workflows.
• Embed WhiteboxTools functions into hetergenous scripting environments along with ArcPy, GDAL, and other geoprocessing libraries.
• Serves as an analytical back-end for other GIS and remote sensing software (e.g. the QGIS Whitebox for Processing plugin).
• Permissive MIT open-source license allows for ready integration with other software.
• Transparent software philosopy allows for easy source code inspection and rapid innovation and development.
• Powerful extension products are available to further enhance WhiteboxTools' capabilites in areas such as DEM processing, LiDAR data analysis, remote sensing, spatial hydrology, and crop yield data analysis.

What can WhiteboxTools do?

WhiteboxTools can be used to perform common geographical information systems (GIS) analysis operations, such as cost-distance analysis, distance buffering, and raster reclassification. Remote sensing and image processing tasks include image enhancement (e.g. panchromatic sharpening, contrast adjustments), image mosaicking, numerous filtering operations, simple classification (k-means clustering), and common image transformations. WhiteboxTools also contains advanced tooling for spatial hydrological analysis (e.g. flow-accumulation, watershed delineation, stream network analysis, sink removal), geomorphometric analysis (e.g. common terrain indices such as slope, curvatures, wetness index, hillshading; hypsometric analysis; multi-scale topographic position analysis), and LiDAR data processing. LiDAR point clouds can be interrogated (LidarInfo, LidarHistogram), segmented, tiled and joined, analyzed for outliers, interpolated to rasters (DEMs, intensity images), and ground-points can be classified or filtered.

The WhiteboxTools Extensions extend the platform's capabilities even further.

WhiteboxTools is not a cartographic or spatial data visualization package; instead it is meant to serve as an analytical backend for other data visualization software, mainly GIS (e.g. Whitebox GAT and QGIS).

WhiteboxTools vs. Whitebox Geospatial Analysis Tools (GAT)

Although WhiteboxTools was first developed with to serve as a source of plugin tools for the Whitebox Geospatial Analysis Tools (GAT) open-source GIS project, the tools contained in the library are stand-alone and can run outside of the larger Whitebox GAT project. See Interacting With WhiteboxTools From the Command Prompt for further details. There have been a large number of requests to call Whitebox GAT tools and functionality from outside of the Whitebox GAT user-interface (e.g. from Python automation scripts). WhiteboxTools is intended to meet these usage requirements. For example, a WhiteboxTools plug-in for QGIS is available.

In this manual, WhiteboxTools refers to the standalone geospatial analysis library, a collection of tools contained within a compiled binary executable command-line program and the associated Python scripts that are distributed alongside the binary file (e.g. whitebox_tools.py and wb_runner.py). Whitebox Geospatial Analysis Tools and Whitebox GAT refer to the GIS software, which includes a user-interface (front-end), point-and-click tool interfaces, and cartographic data visualization capabilities. Importantly, WhiteboxTools and Whitebox GAT are related but separate projects.

Why is it named WhiteboxTools?

The project name WhiteboxTools clearly takes it inspiration from the related project Whitebox GAT. However, the name Whitebox is intended to convey opposition to a 'black box' system, one for which only the inputs and outputs may be observed and the internal workings may not be scrutinized. WhiteboxTools is inspired by the concept of open-access software, the tenants of which were described by Lindsay (2014)¹. Open-access software can be viewed as a complementary extension to the traditional open-source software (OSS) model of development. The concept of open access has been previously defined in the context of publishing scholarly literature in a way that removes financial, legal, and technical access barriers to knowledge transfer. Lindsay (2014) argued that the stated goals of reducing barriers associated with knowledge transfer applies equally to the software used in research. Open-access software is distinct from other OSS in that it has an explicitly stated design goal of reducing barriers to the transfer of knowledge to the user community. Direct insight into the workings of algorithm design and implementation allows for educational opportunities and increases the potential for rapid innovation, experimentation with algorithms, and community-directed development. This is particularly important in geomatics because many geospatial algorithms are complex and are strongly affected by implementation details. Also, there are often multiple competing algorithms for accomplishing the same task and the choice of one method over another can greatly impact the outcome of a workflow.

All OSS allow users the opportunity to download the source code and inspect the software’s internal workings. However, traditional OSS often does not lend itself to end-user source code inspection. Open-access software, by comparison, is designed from the project's inception in a way that reduces the barriers that typically discourage end-users from examining the algorithm and implementation details associated with specific software artifacts. WhiteboxTools attempts to address some of the barriers to knowledge transfer by allowing users to view the source code associated with each tool directly (e.g. --viewcode=ExtendVectorLines). This functionality removes the need to download separate, and often large, project source code files and eliminates the requisite familiarity with the project to identify the code sections related to the operation of the tool of interest. The viewcode flag is the embodiment of a design philosophy that is intended to empower the user community. Each tool included in the library has been written in a way to isolate the major functionality within a single file, thereby easing the task of interpreting the code (traditional coding style would split complex code among numerous files). This design goal is also why the developers have chosen to exclude external libraries commonly found in other similarly software (e.g. GDAL), thereby simplifying the installation process and code interpretation. This approach has the potential to encourage further community involvement and experimentation with geospatial analysis techniques.

Setting Up WhiteboxTool

WhiteboxTools is a stand-alone executable command-line program with no actual installation. Download the appropriate file for your system from the WhiteboxTools homepage and decompress the folder. Pre-compiled binaries can be downloaded for MS Windows, MacOS, and Linux operating systems. Depending on your operating system, you may need to grant the WhiteboxTools executable file execution privileges before running it.

Information about how to install any of the WhiteboxTools extension products can be found on www.whiteboxgeo.com or by watching this YouTube video.

If you intend to use WhiteboxTools from the command prompt (i.e. a terminal application), you may wish to add the directory containing the WhiteboxTools executable file to your system PATH variable. Doing so will greatly simplify usage of the library from your terminal. Instructions for doing this depend on your operating system and can be found on the Internet.

If you intend to use the Python programming interface for WhiteboxTools you will need to have Python 3 installed. Please note that the Python interface will not work correctly with Python 2. If your system has Python 2 as the default Python version, it is possible to install Python 3 alongside your current version. You may need to use the python3 command in place of the usual python if this is the case.

Building WhiteboxTools From Source Code

Most users rely on the pre-compiled versions of WhiteboxTools and will never need to compile the software from its source code. There are two circumstances under which you may find that you need to build your own binary executable: 1) if you need a binary compiled for a platform other than the project-supported operating systems, and 2) if you require a bleeding-edge feature or bug-fix only available in the development branch. It is likely that WhiteboxTools will work on a wider variety of operating systems and architectures than those of the distributed pre-compiled binaries. If you do not find your operating system/architecture in the list of available WhiteboxTool binaries, then compilation from source code will be necessary. WhiteboxTools can be compiled from the source code with the following steps:

1. Install the Rust compiler; Rustup is recommended for this purpose. Further instruction can be found at this link.

The proper way to install Rustup depends on your operating system and instructions can be found on the Rust install page. For Unix-type OSs (including linux and MacOS), the recommended install command is:

curl https://sh.rustup.rs -sSf | sh

After installing Rustup, install the Rust compiler and tools (including the Cargo package manager and build tool):

rustup install stable

Note, you may need to install a linker in addition to the Rust compiler (e.g. MS Visual C++ 2015 Build Tools on MS Windows; XCode on MacOS).

2. Download the WhiteboxTools source code. To download the code, click the green Clone or download button on the GitHub repository site.

3. Decompress the zipped download file.

4. Open a terminal (command prompt) window and change the working directory to the whitebox_tools sub-folder, which is contained within the decompressed downloaded Whitebox GAT folder:

>> cd /path/to/folder/whitebox_tools/

5. Finally, use the rust package manager Cargo, which will be installed alongside Rust, to compile the executable:

>> cargo build --release

Depending on your system, the compilation may take several minutes. When completed, the compiled binary executable file will be contained within the whitebox_tools/target/release/ folder. Type ./whitebox_tools --help at the command prompt (after changing the directory to the containing folder) for information on how to run the executable from the terminal.

The '>>' is shorthand used in this document to denote the command prompt and is not intended to be typed.

Be sure to follow the instructions for installing Rust carefully. In particular, if you are installing on Microsoft Windows, you must have a linker installed prior to installing the Rust compiler (rustc). The Rust webpage recommends either the MS Visual C++ 2015 Build Tools or the GNU equivalent and offers details for each installation approach.

Using WhiteboxTools

Users interact with the WhiteboxTools platform either from a command prompt (i.e. terminal), through the Python, R language, or Nim interfaces, or from one of the available graphical user interfaces (GUI) applications.

Interfacing With Python

Important note: all of the following material assumes the user system is configured with Python 3. The code snippets below are not guaranteed to work with older versions of the language.

By combining the WhiteboxTools library with a high-level scripting language, such as Python, users are capable of creating powerful stand-alone geospatial applications and workflow automation scripts. In fact, WhiteboxTools functionality can be called from many different programming languages. However, given the prevalent use of the Python language in the geospatial fields, the library is distributed with several resources specifically aimed at Python scripting. This section focuses on how Python programming can be used to interact with the WhiteboxTools library.

If you use the Python package manager PIP, you may install WhiteboxTools at the command prompt with pip install whitebox. The PIP package is maintained by Prof. Qiusheng Wu. There is also an Anaconda package, which can be installed with conda install -c conda-forge whitebox_tools , although it is unclear if this package is regularly updated to reflect the latest versions of WhiteboxTools.

Using the whitebox_tools.py script

Interacting with WhiteboxTools from Python scripts is easy. To begin, each script must start by importing the WhiteboxTools class, contained with the whitebox_tools.py script; a new WhiteboxTools object can then be created:

from WBT.whitebox_tools import WhiteboxTools

wbt = WhiteboxTools()

Depending on the relative location of the WhiteboxTools directory and the script file that you are importing to, the import statement may need to be altered slightly. In the above script, it is assumed that the folder containing the WhiteboxTools files (including the whitebox_tools Python script) is named WBT (Line 1) and that the calling script that is importing WhiteboxTools is located in the parent directory of WBT. See An Example WhiteboxTools Python Project for more details on project set-up. The use of wbt to designate the WhiteboxTools object variable in the above script (Line 3) is just the convention used in this manual and other project resources. In fact, any variable name can be used for this purpose.

The WhiteboxTools class expects to find the WhiteboxTools executable file (whitebox_tools.exe on Windows and whitebox_tools on other platforms) within the same directory (WBT) as the whitebox_tools.py script. If the binary file is located in a separate directory, you will need to set the executable directory as follows:

wbt.set_whitebox_dir('/local/path/to/whitebox/binary/')

Individual tools can be called using the convenience methods provided in the WhiteboxTools class:

# This line performs a 5 x 5 mean filter on 'inFile.tif':
wbt.mean_filter('/file/path/inFile.tif', '/file/path/outFile.tif', 5, 5)

Each tool has a cooresponding convenience method. The listing of tools in this manual includes information about each tool's Python convienience method, including default parameter values. Parameters with default values may be optionally left off of function calls. In addition to the convenience methods, tools can be called using the run_tool() method, specifying the tool name and a list of tool arguments.

source = "source.tif"
cost = "cost.tif"
out_accum = "accum.tif"
out_backlink = "backlink.tif"
args = []
args.append("--source='{}'".format(source))
args.append("--cost='{}'".format(cost))
args.append("--out_accum='{}'".format(out_accum))
args.append("--out_backlink='{}'".format(out_backlink))
self.run_tool('cost_distance', args)

Each of the tool-specific convenience methods collect their parameters into a properly formated list and then ultimately call the run_tools() method. Notice that while internally whitebox_tools.exe uses CamelCase (e.g. MeanFilter) to denote tool names, the Python interface of whitebox_tools.py uses snake_case (e.g. mean_filter), according to Python style conventions. The only exceptions are tools with names that clash with Python keywords (e.g. And(), Not(), and Or()).

The return value can be used to check for errors during operation:

if wbt.ruggedness_index('/path/DEM.tif', '/path/ruggedness.tif') != 0:
    # Non-zero returns indicate an error.
    print('ERROR running ruggedness_index')

If your data files tend to be burried deeply in layers of sub-directories, specifying complete file names as input parameters can be tedius. In this case, the best option is setting the working directory before calling tools:

from whitebox_tools import WhiteboxTools

wbt = WhiteboxTools()
wbt.set_working_dir("/path/to/data/") # Sets the working directory

# Setting the following to True enables tools to output DEFLATE compressed GeoTIFFs.
# You only need to do this once, if you wish all tools to compress their raster
# outputs.
wbt.set_compress_rasters(True) 

# Because the working directory has been set, file arguments can be
# specified simply using file names, without paths.
wbt.d_inf_flow_accumulation("DEM.tif", "output.tif", log=True)

An advanced text editor, such as VS Code or Atom, can provide hints and autocompletion for available tool convenience methods and their parameters, including default values (Figure 1).

Sometimes it can be useful to print a complete list of available tools:

print(wbt.list_tools()) # List all tools in WhiteboxTools

The list_tools() method also takes an optional keywords list to search for tools:

# Lists tools with 'lidar' or 'LAS' in tool name or description.
print(wbt.list_tools(['lidar', 'LAS']))

To retrieve more detailed information for a specific tool, use the tool_help() method:

print(wbt.tool_help("elev_percentile"))

tool_help() prints tool details including a description, tool parameters (and their flags), and example usage at the command line prompt. The above statement prints this report:

ElevPercentile
Description:
Calculates the elevation percentile raster from a DEM.
Toolbox: Geomorphometric Analysis
Parameters:

Flag               Description
-----------------  -----------
-i, --input, --dem Input raster DEM file.
-o, --output       Output raster file.
--filterx          Size of the filter kernel in the x-direction.
--filtery          Size of the filter kernel in the y-direction.
--sig_digits       Number of significant digits.

Example usage:
>>./whitebox_tools -r=ElevPercentile -v --wd="/path/to/data/" --dem=DEM.tif
>>-o=output.tif --filterx=25

A note on default parameter values

Each tool contains one or more parameters with default values. These will always be listed after any input parameters that do not have default values. You do not need to specify a parameter with a default value if you accept the default. That is, unless you intend to specify an input value different from the default, you may leave these parameters off of the function call. However, be mindful of the fact that Python assigns values to parameters based on order, unless parameter names are specified.

Consider the Hillshade tool as an example. The User Manual gives the following function definition for the tool:

hillshade(
dem,
output,
azimuth=315.0,
altitude=30.0,
zfactor=1.0,
callback=default_callback)

The dem and output parameters do not have default values and must be specified every time you call this function. Each of the remaining parameters have default values and can, optionally, be left off of calls to the hillshade function. As an example, say I want to accept the default values for all the parameters except altitude. I would then need to use the named-parameter form of the function call:

wbt.hillshade(
"DEM.tif",
"hillshade.tif",
altitude=20.0)

If I hadn't specified the parameter name for altitude, Python would have assumed that the value 20.0 should be assigned to the third parameter, azimuth.

Handling tool output

Tools will frequently print text to the standard output during their execution, including warnings, progress updates and other notifications. Sometimes, when users run many tools in complex workflows and in batch mode, these output messages can be undesirable. Most tools will have their outputs suppressed by setting the verbose mode to False as follows:

wbt.set_verbose_mode(False)

Alternatively, it may be helpful to capture the text output of a tool for custom processing. This is achieved by specifying a custom callback function to the tool's convenience function:

# This callback function suppresses printing progress updates,
# which always use the '%' character. The callback function
# approach is flexible and allows for any level of complex
# interaction with tool outputs.
def my_callback(value):
    if not "%" in value:
        print(value)

wbt.slope('DEM.tif', 'slope_raster.tif', callback=my_callback)

Every convienience function takes an optional callback as the last parameter. The default callback simply prints tool outputs to the standard output without any additional processing. The default callback itself can be overridden, instead of having to set callbacks in convienience functions individually:

wbt.set_default_callback(my_callback)

Callback functions can serve as a means of cancelling operations:

def my_callback(value):
    if user_selected_cancel_btn: # Assumes a 'Cancel' button on a GUI
        print('Cancelling operation...')
        wbt.cancel_op = True
    else:
        print(value)

wbt.breach_depressions('DEM.tif', 'DEM_breached.tif', callback=my_callback)

Additional functions in whitebox_tools.py

The whitebox_tools.py script provides several other functions for interacting with the WhiteboxTools library, including:

# Print the WhiteboxTools help...a listing of available commands
print(wbt.help())

# Print the WhiteboxTools license
print(wbt.license())

# Print the WhiteboxTools version
print("Version information: {}".format(wbt.version()))

# Get the toolbox associated with a tool
tb = wbt.toolbox('lidar_info')

# Retrieve a JSON object of a tool's parameters.
tp = wbt.tool_parameters('raster_histogram')

# Opens a browser and navigates to a tool's source code in the
# WhiteboxTools GitHub repository
wbt.view_code('watershed')

# Use this function to specify whether output GeoTIFF rasters should be 
# compressed using the DEFLATE compression method.
wbt.set_compress_rasters(True) 

For a working example of how to call functions and run tools from Python, see the whitebox_example.py Python script, which is distributed with the WhiteboxTools library.

Additional resources for using WhiteboxTools' Python interface can be found on the Tutorials site of the WhiteboxTools home page. This site contains in-depth tutorials on topics such as, 'Interpolating LiDAR data'.

An Example Python Project

In this section, we will create a Python project that utilizes the WhiteboxTools library to interpolate a LiDAR point-cloud, to process the resulting digital elevation model (DEM) to make it suitable for hydrological applications, and to perform a simple flow-accumulation operation. I suggest using an advanced coding text editor, such as Visual Studio Code or Atom, for this tutorial, but Python code can be written using any basic text editor.

Begin by creating a dedicated project directory called FlowAccumExample and copy WhiteboxTools binary file (i.e. the compressed file downloaded from the Geomorphometry & Hydrogeomatics Research Group website) into this folder. Using the decompression software on your computer, decompress (i.e. an operation sometimes called unzipping) the file into the newly created FlowAccumExample directory. You will find the compressed file contains a folder with contents similar to the following:

The folder contains a number of files, including the WhiteboxTools executable file, the whitebox_tools.py python script, the WhiteboxTools Runner (wb_runner.py; see below), and this user manual. It is likely that the folder has a name that reflects the operating system and architecture that the binary file was compiled for (e.g. WhiteboxTools_darwin_amd64). Rename this directory to WBT. Also note, depending on your decompression software, it may be the case that the contents of the WBT folder itself contains a sub-directory that actually holds these files. If this is the case, be sure to move the contents of the sub-directory into the WBT parent directory.

Using your text editor, create a new Python script file, called FlowAccumulation.py within the FlowAccumExample directory. We will begin by importing the WhiteboxTools class from the whitebox_tools.py script contained within the WBT sub-directory. Unfortunately, Python's module system is only able to import classes and function definitions declared in external Python scripts if these external files are contained somewhere on the Python path or in the directory containing the script file into which you are importing. This is important because based on the project structure that we have established, the whitebox_tools.py script is actually contained within a sub-directory of the FlowAccumExample directory and is therefore not directly accessible, unless you have previously installed the script on the Python path. Another, perhaps easier solution to this problem is to create a file named __init__.py (those are two leading and trailing underscore characters) within the FlowAccumExample directory. The presence of this empty file will make Python treat the WBT directory as containing packages, in this case, the whitebox_tools package. For more information, see the Python documentation on modules and packages.

At this stage, you should have a project directory structure like the following:

Many operating systems will disallow the execution of files that are downloaded directly from the Internet. As such, it is possible that you will need to explicitly give the whitebox_tools.exe permission to execute on your computer (Note: here we are referring to the compiled WhiteboxTools binary file and not the similarly named Python script whitebox_tools.py also contained in the folder). The procedure for doing this depends on your specific operating system. On MacOS, for example, this is usually achieved using the 'Security & Privacy' tab under 'System Preferences'. To test whether whitebox_tools.exe has permission to run on your system, double-click the file. If the file is configured to execute, a command terminal will automatically open and the WhiteboxTools help documentation and a listing of the available tools will be printed. If this does not occur, you likely need to give the file permission to execute.

Using your text editor, you may now add the following lines to the FlowAccumulation.py file.

from WBT.whitebox_tools import WhiteboxTools

wbt = WhiteboxTools()

In the import statement, WBT is a reference to the package folder containing the WhiteboxTools files; whitebox_tools is a reference to the whitebox_tools.py script contained with this package folder; and WhiteboxTools is a reference to the WhiteboxTools class contained within this script file. Please note that if you named your directory containing the WhiteboxTools files something other than WBT, you would need to alter the import statement accordingly.

Download the St. Elis Mountains and Gulf of Alaska sample data set (StElisAk.laz) from the WhiteboxTools section of the site and decompress the zip file. This file contains a LiDAR point cloud that has been previously filtered to remove points associated with non-ground returns, mainly trees (Figure 4). Create a sub-directory within the project folder called 'data' and copy StElisAk.laz into the folder.

Now we can complete our flow accumulation analysis with the following code:

import os
from WBT.whitebox_tools import WhiteboxTools

wbt = WhiteboxTools()

# Set the working directory, i.e. the folder containing the data,
# to the 'data' sub-directory.
wbt.set_working_dir(os.path.dirname(os.path.abspath(__file__)) + "/data/")

# When you're running mulitple tools, the outputs can be a tad
# chatty. In this case, you may want to suppress the output by
# setting the verbose mode to False.
# wbt.set_verbose_mode(False)

# Interpolate the LiDAR data using an inverse-distance weighting
# (IDW) scheme.
print("Interpolating DEM...")
wbt.lidar_idw_interpolation(
i="StElisAk.laz",
output="raw_dem.tif",
parameter="elevation",
returns="last",
resolution=1.0,
weight=1.0,
radius=2.5
)

# The resulting DEM will contain NoData gaps. We need to fill
# these in by interpolating across the gap.
print("Filling missing data...")
wbt.fill_missing_data(
i="raw_dem.tif",
output="dem_nodata_filled.tif",
filter=11
)

# This DEM will contain grid cells that have no lower neighbours.
# This condition is unsuited for flow-path modelling applications
# because these operations assume that each interior cell in the
# DEM has at least one downslope neighour. We'll use an operation
# called depression breaching to 'fix' the elevations within the
# DEM to enforce continuous flow.
print("Performing flow enforcement...")
wbt.breach_depressions(
dem="dem_nodata_filled.tif",
output="dem_hydro_enforced.tif"
)

# Lastly, perform the flow accumulation operation using the
# D-infinity flow algorithm.
print("Performing flow accumulation...")
wbt.d_inf_flow_accumulation(
dem="dem_hydro_enforced.tif",
output="flow_accum.tif",
log=True
)

print("Complete!")

To run the above script, open a terminal (command prompt), cd to the script containing folder, and run the following command:

>>python FlowAccumulation.py

If Python 3 is not your default Python version, substitute python3 for python in the above command line. The final D-infinity flow accumulation raster can be displayed in any GIS software of choice and should look similar to Figure 5.

Interfacing With R

In addition to the Python interface, the WhiteboxTools library is also accessible from an R language package. R is a common programming language used within the statistical and scientific computing communities and the R WhiteboxTools package targets these groups. Prof. Qiusheng Wu, at Binghamton University (SUNY) maintains the R package.

Installation

WhiteboxTools is available on R-Forge and can be installed with the command:

install.packages("whitebox", repos="http://R-Forge.R-project.org")

You can alternatively install the development version of the R package whitebox from the GitHub repository as follows:

if (!require(devtools)) install.packages('devtools')
devtools::install_github("giswqs/whiteboxR")

You will also need to make sure your machine is able to build packages from source. See Package Development Prerequisites for the tools needed for your operating system.

Usage

A complete list of functions available in the whitebox R package can be found within the GitHub repository. A comprehensive demonstration, complete with detailed examples, is also available from this site.

About WhiteboxTools

library(whitebox)

# Prints the whitebox-tools help...a listing of available commands
print(wbt_help())

# Prints the whitebox-tools license
print(wbt_license())

# Prints the whitebox-tools version
print(wbt_version())

# Prints the toolbox for a specific tool.
print(wbt_toolbox())

# List all available tools in whitebox-tools
print(wbt_list_tools())

# Lists tools with 'lidar' in tool name or description.
print(wbt_list_tools("lidar"))

# Prints the help for a specific tool.
print(wbt_tool_help("lidar_info"))

# Retrieves the tool parameter descriptions for a specific tool.
print(wbt_tool_parameters("slope"))

# View the source code for a specific tool on the source code repository.
print(wbt_view_code("breach_depressions"))

How to run tools?

Tool names in the whitebox R package can be called using the snake_case (e.g. lidar_info). A comprehensive list of all available function tools can be found on the package repository site. For example:

library(whitebox)

# Set input raster DEM file
dem <- system.file("extdata", "DEM.tif", package="whitebox")

# Run tools
feature_preserving_denoise(dem, "./smoothed.tif", filter=9)
breach_depressions("./smoothed.tif", "./breached.tif")
d_inf_flow_accumulation(dem, "./flow_accum.tif", verbose_mode=FALSE)

Interfacing With Nim

Nim is a statically compiled programming language that has Python-like syntax (e.g. significant whitespace) and c-level efficiency. It has been used by Python programmers who are looking for a faster, more efficient language but don't want to give up the elegant and terse syntax of Python.

wbt_nim is a Nim-based application programming interface (API) used to call WhiteboxTools functionality from Nim programs. The documentation for wbt_nim can be found on GitHub. To use this API, simply copy the source file into your Nim project and import wbt_nim. You will need to ensure that the WhiteboxTools binary, pre-compiled for your operating system, is also within the same directory. Otherwise, use the setExecutableDirectory function to tell wbt_nim where the WhiteboxTools binary is located on your system.

The Nim interface is very similar to the Python API:

import wbt_nim
import options, strformat, strutils

proc main() =
    # Create a new WhiteboxTools object
    var wbt = newWhiteboxTools()

    # Tell the wbt object where to find the WhiteboxTools executable.
    # If you don't do this, it assumes that it is in the same directory as 
    # your Nim code.
    wbt.setExecutableDirectory("/Users/johnlindsay/Documents/programming/whitebox-tools/")

    # Set the working directory
    let wd = "/Users/johnlindsay/Documents/data/LakeErieLidar/"
    wbt.setWorkingDirectory(wd)

    # Set the verbose mode. By default it is 'true', which prints all output
    # from WBT. If you need to make it less chatty, set it to false.
    wbt.setVerboseMode(false)

    # By default, all GeoTiff outputs of tools will be compressed. You can 
    # modify this with the following:
    wbt.setCompressRasters(false)

    # Print out the version of WBT:
    echo(wbt.getVersionInfo())

    # WhiteboxTools is open-access software. If you'd like to see the source 
    # code for any tool, simply use the following:
    discard wbt.viewCode("balanceContrastEnhancement")

    # To get a brief description of a tool and it's parameters:
    echo(wbt.getToolHelp("breachDepressionsLeastCost"))

    # If you'd like to see more detailed help documentation:
    discard wbt.viewHelpPage("breachDepressionsLeastCost")
    # This will open the default browser and navigate to the relevant tool help.

    # Here's an example of how to run a tool:
    if wbt.hillshade(
        dem="90m_DEM.tif",
        output="tmp1.tif",
        azimuth=180.0,
        altitude=45.0,
        zFactor=1.0
    ) != 0:
        echo("Error while running hillshade.")

    # If you haven't previously set the working directory, you need to include
    # full file path names.

    # You can capture tool output by creating a custom callback function
    proc myCallback(value: string) =
        if not value.contains("%"):
            echo(value)
        else:
            let s = value.replace("%", "").strip()
            echo(fmt"{s} percent")

    wbt.setCallback(myCallback)
    
    # And to reset the default callback, which just prints to stdout
    wbt.setDefaultCallback()

main()

Whitebox Runner

There is an application file (executable) contained within the WhiteboxTools WBT directory called 'whitebox_runner'. This application is intended to provide a very basic user-interface, known as the Whitebox Runner or simply WbRunner, for running the tools contained within the WhiteboxTools library. The user-interface is cross-platform and has been compiled for Windows, Linux, and macOS.

To launch the runner application, you should simply need to double-click the executable file. The application has a dark-mode theme (seen above) as well as a light-mode theme (seen below). Like WhiteboxTools itself, WbRunner is developed using pure Rust code and is a native application. It is therefore very performant, especially when compared with the previous version of the Runner, which was a Python interface application.

Even if you typically use other Whitebox front-ends, WbRunner can be useful at times. For example, unlike many of the other front-end applications, the WbRunner is always up-to-date with your current version of the WhiteboxTools back-end. You may always access the latest tools in Whitebox using the WbRunner. It is also the preferred way to install the Whitebox Toolset Extension. Furthermore, because WbRunner is developed in-house at Whitebox Geospatial Inc., unlike many of the other Whitebox front-ends, the tools interface with the front-end exactly as they are intended to without having to compromise for limitations on front-end design. For example, tools output directly to an output window embedded in the tool's dialog. Progress bars update as the tool is running. The help button links to the appropriate tool documentation in the user manual and the cancel button works as expected (except for plugin tools, which cannot be directly cancelled).

QGIS Plugin

IMPORTANT The WhiteboxTools for QGIS plugin is considered to be legacy code since the release of the next-generation Whitebox Workflows for QGIS. Users should prefer the Whitebox Workflows for QGIS plugin instead, which shares the same functionality with significant improvements in operation (e.g., inline help documentation, sample data, etc.).

WhiteboxTools functionality can be accessed conveniently through the popular open-source geospatial software QGIS. Historically, the QGIS WhiteboxTools plugin was developed by Alexander Bruy.

Please note, as of February 2022, Alex Bruy, the original developer of the QGIS WhiteboxTools plugin, is no longer distributing any of his numerous QGIS open-source plugins, in protest of the war in Ukraine. As such, we are currently developing the WhiteboxTools plugin for QGIS in-house at Whitebox Geospatial Inc.

The Whitebox for QGIS plugin works QGIS v3 but cannot be installed on the earlier v2 series. The minimum QGIS version supported by the plugins is currently 3.18.0. If you are using an older version of QGIS, you will need to update before the following procedure for installation will work.

Installation of the Plugin

For detailed instructions on how to install and configure the Whitebox plugin for QGIS, please watch this short video. Note that installing the Whitebox plugin does not automatically install the open-source WhiteboxTools back-end, nor any of the commercial extension products. You will need to visit the Whitebox Geospatial homepage and download the WhiteboxTools binary executable before installing the QGIS plugin. The Whitebox Toolset Extension (WTE) can also be downloaded from the site, although a valid license is required to run the extension.

ArcGIS Plugin

WhiteboxTools functionality can also be accessed conveniently through the ArcGIS. This WhiteboxTools front-end has been developed and is maintained by Prof. Qiusheng Wu, of Binghamton University (SUNY). This front-end is available from the Gihub repository. The plugin works for ArcGIS 10.6 and ArcGIS Pro; other version of ArcGIS have not been tested for support. Detailed instructions on installing and setting-up the ArcGIS toolbox can be found on the GitHub site.

Command-Line Interface

WhiteboxTools is a command-line program and can be run either by calling it from a terminal application with appropriate commands and arguments, or, more conveniently, by calling it from a script. The following commands are recognized by the WhiteboxTools library:

Generally, the Unix convention is that single-letter arguments (options) use a single hyphen (e.g. -h) while word-arguments (longer, more descriptive argument names) use double hyphens (e.g. --help). The same rule is used for passing arguments to tools as well. Use the --toolhelp argument to print information about a specific tool (e.g. --toolhelp=Clump).

Tool names can be specified either using the snake_case or CamelCase convention (e.g. lidar_info or LidarInfo).

The following is an example of calling the WhiteboxTools binary executable file directly from the command prompt:

>>./whitebox_tools --wd='/Users/johnlindsay/Documents/data/' ^
--run=DevFromMeanElev --input='DEM clipped.tif' ^
--output='DEV raster.tif' -v

Notice the quotation marks (single or double) used around directories and filenames, and string tool arguments in general. After the --run flag, used to call a tool, a series of tool-specific flags are provided to indicate the values of various input parameters. Note that the order of these flags is unimportant. Use the '-v' flag (run in verbose mode) to force the tool to print output to the command prompt. Please note that the whitebox_tools executable file must have permission to be executed; on some systems, this may require setting special permissions. Also, the above example uses the forward slash character (/), the directory path separator used on unix based systems. On Windows, users should use the back slash character (\) instead. Also, it is sometimes necessary to break (^) commands across multiple lines, as above, in order to better fit with the documents format. Actual command prompts (>>) should be contained to a single line.

WhiteboxTools Extensions

Whitebox Geospatial Inc. currently offers one extension for WhiteboxTools, known as the Whitebox Toolset Extension. This product contains plugins that extend the functionality of the open-core of WhiteboxTools. The list of the plugin tools and there descriptions is found here. While managed separately from the open-core, the extension easily integrate into your current WhiteboxTools environment. Please contact support@whiteboxgeo.com for further details, or visit www.whiteboxgeo.com for current pricing information.

If you are using the WhiteboxTools Python API, you may install any of the Whitebox Extensions with the following script:

from WBT.whitebox_tools import WhiteboxTools

wbt = WhiteboxTools()

wbt.install_wbt_extension()

After running the above script, Python will prompt you, from the command line terminal application, which extension you would like to install. Options include 'gte' (General Toolset Extension), 'lidar' (LiDAR and Remote Sensing Extension), 'dem' (DEM and Spatial Analysis Extension), and 'agri' (Agriculture Extension). Once you specify the extension you would like installed, the application will download and install the appropriate plugin tools (this requires an Internet connect), and then prompt the user whether they would like to activate a license key. This is a necessary step before you can use the newly installed plugin tools.

If you are using the WhiteboxTools Runner, you may instead press the 'Install Whitebox Extension' button at the bottom of the Settings panel.

Tools Reference

The WhiteboxTools platform currently contains approximately 485 tools organized into thematic toolboxes and including tools both within the WhiteboxTools open-core and the various extensions. The thematic toolboxes include:

• Data Tools
• Geomorphometric Analysis
• GIS Analysis
• Hydrological Analysis
• Image Analysis
• LiDAR Analysis
• Machine Learning Tools
• Mathematical and Statistical Analysis
• Precision Agriculture
• Stream Network Analysis
• Whitebox Utilities

To retrieve detailed information about a tool's input arguments and example usage, either use the toolhelp command from the terminal, or the wbt.tool_help('tool_name') function from the whitebox_tools.py script. The following is a complete listing of available tools, with brief descriptions, tool parameter, and example usage.

The Tool Index located at the end of the user manual contains a complete alphabetical listing of the available tools.

Data Tools

• AddPointCoordinatesToTable
• CleanVector
• ConvertNodataToZero
• ConvertRasterFormat
• CsvPointsToVector
• ExportTableToCsv
• FixDanglingArcs
• JoinTables
• LinesToPolygons
• MergeTableWithCsv
• MergeVectors
• ModifyNoDataValue
• MultiPartToSinglePart
• NewRasterFromBase
• PolygonsToLines
• PrintGeoTiffTags
• RasterToVectorLines
• RasterToVectorPoints
• RasterToVectorPolygons
• ReinitializeAttributeTable
• RemovePolygonHoles
• RemoveRasterPolygonHoles
• SetNodataValue
• SinglePartToMultiPart
• VectorLinesToRaster
• VectorPointsToRaster
• VectorPolygonsToRaster

AddPointCoordinatesToTable

This tool modifies the attribute table of a vector of POINT ShapeType by adding two fields, XCOORD and YCOORD, containing each point's X and Y coordinates respectively.

Parameters:

Python function:

wbt.add_point_coordinates_to_table(
    i, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=AddPointCoordinatesToTable -v ^
--wd="/path/to/data/" --input=points.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 25/09/2018

Last Modified: 12/10/2018

CleanVector

This tool can be used to remove all features in Shapefiles that are of the null ShapeType. It also removes line features with fewer than two vertices and polygon features with fewer than three vertices.

Parameters:

Python function:

wbt.clean_vector(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=CleanVector -v --wd="/path/to/data/" ^
-i=input.shp -o=output.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 30/06/2019

Last Modified: 27/05/2020

ConvertNodataToZero

This tool can be used to change the value within the grid cells of a raster file (--input) that contain NoData to zero. The most common reason for using this tool is to change the background region of a raster image such that it can be included in analysis since NoData values are usually ignored by by most tools. This change, however, will result in the background no longer displaying transparently in most GIS. This change can be reversed using the SetNoDataValue tool.

See Also: SetNoDataValue, IsNoData

Parameters:

Python function:

wbt.convert_nodata_to_zero(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ConvertNodataToZero -v ^
--wd="/path/to/data/" --input=in.tif -o=NewRaster.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 11/07/2017

Last Modified: 12/10/2018

ConvertRasterFormat

This tool converts raster data from one format to another. It determines input/output raster formats based on extensions, but due to file extension naming collisions, it would be good to add user hints. For example, the extension 'grd' could belong to a SurferAscii or a Surfer7Binary. This is more important for distinguishing output files since input files can be read and distiguishing features idenfitied from the file structure. At the moment, this tool does not support user hints however.

Parameters:

Python function:

wbt.convert_raster_format(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ConvertRasterFormat -v ^
--wd="/path/to/data/" --input=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: July 6, 2017

Last Modified: 12/10/2018

CsvPointsToVector

This tool can be used to import a series of points contained within a comma-separated values (*.csv) file (--input) into a vector shapefile of a POINT ShapeType. The input file must be an ASCII text file with a .csv extensions. The tool will automatically detect the field data type; for numeric fields, it will also determine the appropriate length and precision. The user must specify the x-coordinate (--xfield) and y-coordiante (--yfield) fields. All fields are imported as attributes in the output (--output) vector file. The tool assumes that the first line of the file is a header line from which field names are retrieved.

See Also: MergeTableWithCsv, ExportTableToCsv

Parameters:

Python function:

wbt.csv_points_to_vector(
    i, 
    output, 
    xfield=0, 
    yfield=1, 
    epsg=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=CsvPointsToVector -v ^
--wd="/path/to/data/" -i=points.csv -o=points.shp --xfield=0 ^
--yfield=1 --epsg=4326 

Source code on GitHub

Author: Prof. John Lindsay

Created: 07/08/2019

Last Modified: 28/01/2020

ExportTableToCsv

This tool can be used to export a vector's attribute table to a comma separated values (CSV) file. CSV files stores tabular data (numbers and text) in plain-text form such that each row corresponds to a record and each column to a field. Fields are typically separated by commas within records. The user must specify the name of the vector (and associated attribute file), the name of the output CSV file, and whether or not to include the field names as a header column in the output CSV file.

See Also: MergeTableWithCsv

Parameters:

Python function:

wbt.export_table_to_csv(
    i, 
    output, 
    headers=True, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ExportTableToCsv -v ^
--wd="/path/to/data/" -i=lines.shp -o=output.csv --headers 

Source code on GitHub

Author: Dr. John Lindsay

Created: 24/04/2018

Last Modified: 18/10/2019

FixDanglingArcs

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool can be used to fix undershot and overshot arcs, two common topological errors, in an input vector lines file (--input). In addition to the input lines vector, the user must also specify the output vector (--output) and the snap distance (--snap). All dangling arcs that are within this threshold snap distance of another line feature will be connected to the neighbouring feature. If the input lines network is a vector stream network, users are advised to apply the RepairStreamVectorTopology tool instead.

See Also: RepairStreamVectorTopology, CleanVector

Parameters:

Python function:

wbt.fix_dangling_arcs(
    i, 
    output, 
    dist="", 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=FixDanglingArcs --input=streams.shp ^
--output=streams_snapped.shp --dist=2.0 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 03/08/2021

Last Modified: 03/08/2021

JoinTables

This tool can be used to join (i.e. merge) a vector's attribute table with a second table. The user must specify the name of the vector file (and associated attribute file) as well as the primary key within the table. The primary key (--pkey flag) is the field within the table that is being appended to that serves as the identifier. Additionally, the user must specify the name of a second vector from which the data appended into the first table will be derived. The foreign key (--fkey flag), the identifying field within the second table that corresponds with the data contained within the primary key in the table, must be specified. Both the primary and foreign keys should either be strings (text) or integer values. Fields containing decimal values are not good candidates for keys. Lastly, the names of the field within the second file to include in the merge operation can also be input (--import_field). If the --import_field field is not input, all fields in the attribute table of the second file, that are not the foreign key nor FID, will be imported to the first table.

Merging works for one-to-one and many-to-one database relations. A one-to-one relations exists when each record in the attribute table corresponds to one record in the second table and each primary key is unique. Since each record in the attribute table is associated with a geospatial feature in the vector, an example of a one-to-one relation may be where the second file contains AREA and PERIMETER fields for each polygon feature in the vector. This is the most basic type of relation. A many-to-one relation would exist when each record in the first attribute table corresponds to one record in the second file and the primary key is NOT unique. Consider as an example a vector and attribute table associated with a world map of countries. Each country has one or more more polygon features in the shapefile, e.g. Canada has its mainland and many hundred large islands. You may want to append a table containing data about the population and area of each country. In this case, the COUNTRY columns in the attribute table and the second file serve as the primary and foreign keys respectively. While there may be many duplicate primary keys (all of those Canadian polygons) each will correspond to only one foreign key containing the population and area data. This is a many-to-one relation. The JoinTables tool does not support one-to-many nor many-to-many relations.

See Also: MergeTableWithCsv, ReinitializeAttributeTable, ExportTableToCsv

Parameters:

Python function:

wbt.join_tables(
    input1, 
    pkey, 
    input2, 
    fkey, 
    import_field=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=JoinTables -v --wd="/path/to/data/" ^
--i1=properties.shp --pkey=TYPE --i2=land_class.shp ^
--fkey=VALUE --import_field=NEW_VALUE 

Source code on GitHub

Author: Prof. John Lindsay

Created: 07/10/2018

Last Modified: 22/11/2018

LinesToPolygons

This tool converts vector polylines into polygons. Note that this tool will close polygons that are open and will ensure that the first part of an input line is interpreted as the polygon hull and subsequent parts are considered holes. The tool does not examine input lines for line crossings (self intersections), which are topological errors.

See Also: PolygonsToLines

Parameters:

Python function:

wbt.lines_to_polygons(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=LinesToPolygons -v ^
--wd="/path/to/data/" -i=input.shp -o=output.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/09/2018

Last Modified: 12/10/2018

MergeTableWithCsv

This tool can be used to merge a vector's attribute table with data contained within a comma separated values (CSV) text file. CSV files stores tabular data (numbers and text) in plain-text form such that each row is a record and each column a field. Fields are typically separated by commas although the tool will also support seimi-colon, tab, and space delimited files. The user must specify the name of the vector (and associated attribute file) as well as the primary key within the table. The primary key (--pkey flag) is the field within the table that is being appended to that serves as the unique identifier. Additionally, the user must specify the name of a CSV text file with either a *.csv or *.txt extension. The file must possess a header row, i.e. the first row must contain information about the names of the various fields. The foreign key (--fkey flag), that is the identifying field within the CSV file that corresponds with the data contained within the primary key in the table, must also be specified. Both the primary and foreign keys should either be strings (text) or integer values. Fields containing decimal values are not good candidates for keys. Lastly, the user may optionally specify the name of a field within the CSV file to import in the merge operation (--import_field flag). If this flag is not specified, all of the fields within the CSV, with the exception of the foreign key, will be appended to the attribute table.

Merging works for one-to-one and many-to-one database relations. A one-to-one relations exists when each record in the attribute table corresponds to one record in the second table and each primary key is unique. Since each record in the attribute table is associated with a geospatial feature in the vector, an example of a one-to-one relation may be where the second file contains AREA and PERIMETER fields for each polygon feature in the vector. This is the most basic type of relation. A many-to-one relation would exist when each record in the first attribute table corresponds to one record in the second file and the primary key is NOT unique. Consider as an example a vector and attribute table associated with a world map of countries. Each country has one or more more polygon features in the shapefile, e.g. Canada has its mainland and many hundred large islands. You may want to append a table containing data about the population and area of each country. In this case, the COUNTRY columns in the attribute table and the second file serve as the primary and foreign keys respectively. While there may be many duplicate primary keys (all of those Canadian polygons) each will correspond to only one foreign key containing the population and area data. This is a many-to-one relation. The JoinTables tool does not support one-to-many nor many-to-many relations.

See Also: JoinTables, ReinitializeAttributeTable, ExportTableToCsv

Parameters:

Python function:

wbt.merge_table_with_csv(
    i, 
    pkey, 
    csv, 
    fkey, 
    import_field=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MergeTableWithCsv -v ^
--wd="/path/to/data/" -i=properties.shp --pkey=TYPE ^
--csv=land_class.csv --fkey=VALUE ^
--import_field=NEW_VALUE
>>./whitebox_tools -r=MergeTableWithCsv ^
-v --wd="/path/to/data/" -i=properties.shp --pkey=TYPE ^
--csv=land_class.csv --fkey=VALUE 

Source code on GitHub

Author: Prof. John Lindsay

Created: 11/10/2018

Last Modified: 09/03/2020

MergeVectors

Combines two or more input vectors of the same ShapeType creating a single, new output vector. Importantly, the attribute table of the output vector will contain the ubiquitous file-specific FID, the parent file name, the parent FID, and the list of attribute fields that are shared among each of the input files. For a field to be considered common between tables, it must have the same name and field_type (i.e. data type and precision).

Overlapping features will not be identified nor handled in the merging. If you have significant areas of overlap, it is advisable to use one of the vector overlay tools instead.

The difference between MergeVectors and the Append tool is that merging takes two or more files and creates one new file containing the features of all inputs, and Append places the features of a single vector into another existing (appended) vector.

This tool only operates on vector files. Use the Mosaic tool to combine raster data.

See Also: Append, Mosaic

Parameters:

Python function:

wbt.merge_vectors(
    inputs, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MergeVectors -v --wd="/path/to/data/" ^
-i='polys1.shp;polys2.shp;polys3.shp' -o=out_file.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 01/10/2018

Last Modified: 12/10/2018

ModifyNoDataValue

This tool can be used to modify the value of pixels containing the NoData value for an input raster image. This operation differs from the SetNodataValue tool, which sets the NoData value for an image in the image header without actually modifying pixel values. Also, SetNodataValue does not overwrite the input file, while the ModifyNoDataValue tool does. This tool cannot modify the input image data type, which is important to note since it may cause an unexpected behaviour if the new NoData value is negative and the input image data type is an unsigned integer type.

See Also: SetNodataValue, ConvertNodataToZero

Parameters:

Python function:

wbt.modify_no_data_value(
    i, 
    new_value="-32768.0", 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ModifyNoDataValue -v ^
--wd="/path/to/data/" --input=in.tif --new_value= -999.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 08/09/2019

Last Modified: 08/09/2019

MultiPartToSinglePart

This tool can be used to convert a vector file containing multi-part features into a vector containing only single-part features. Any multi-part polygons or lines within the input vector file will be split into separate features in the output file, each possessing their own entry in the associated attribute file. For polygon-type vectors, the user may optionally choose to exclude hole-parts from being separated from their containing polygons. That is, with the --exclude_holes flag, hole parts in the input vector will continue to belong to their enclosing polygon in the output vector. The tool will also convert MultiPoint Shapefiles into single Point vectors.

See Also: SinglePartToMultiPart

Parameters:

Python function:

wbt.multi_part_to_single_part(
    i, 
    output, 
    exclude_holes=True, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiPartToSinglePart -v ^
--wd="/path/to/data/" -i=input.shp -o=output.shp ^
--exclude_holes 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/09/2018

Last Modified: 16/06/2020

NewRasterFromBase

This tool can be used to create a new raster with the same coordinates and dimensions (i.e. rows and columns) as an existing base image, or the same spatial extent as an input vector file. The user must specify the name of the base file (--base), the value that the new grid will be filled with (--value flag; default of nodata), and the data type (--data_type flag; options include 'double', 'float', and 'integer'). If an input vector base file is used, then it is necessary to specify a value for the optional grid cell size (--cell_size) input parameter.

See Also: RasterCellAssignment

Parameters:

Python function:

wbt.new_raster_from_base(
    base, 
    output, 
    value="nodata", 
    data_type="float", 
    cell_size=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=NewRasterFromBase -v ^
--wd="/path/to/data/" --base=base.tif -o=NewRaster.tif ^
--value=0.0 --data_type=integer
>>./whitebox_tools ^
-r=NewRasterFromBase -v --wd="/path/to/data/" --base=base.tif ^
-o=NewRaster.tif --value=nodata 

Source code on GitHub

Author: Dr. John Lindsay

Created: 11/07/2017

Last Modified: 27/08/2021

PolygonsToLines

This tool converts vector polygons into polylines, simply by modifying the Shapefile geometry type.

See Also: LinesToPolygons

Parameters:

Python function:

wbt.polygons_to_lines(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=PolygonsToLines -v ^
--wd="/path/to/data/" -i=input.shp -o=output.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 04/09/2018

Last Modified: 12/10/2018

PrintGeoTiffTags

This tool can be used to view the tags contained within a GeoTiff file. Viewing the tags of a GeoTiff file can be useful when trying to import the GeoTiff to different software environments. The user must specify the name of a GeoTiff file and the tag information will be output to the StdOut output stream (e.g. console). Note that tags that contain greater than 100 values will be truncated in the output. GeoKeys will also be interpreted as per the GeoTIFF specification.

Parameters:

Python function:

wbt.print_geo_tiff_tags(
    i, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=PrintGeoTiffTags -v ^
--wd="/path/to/data/" --input=DEM.tiff 

Source code on GitHub

Author: Dr. John Lindsay

Created: March 2, 2018

Last Modified: March 2, 2018

RasterToVectorLines

This tool converts raster lines features into a vector of the POLYLINE ShapeType. Grid cells associated with line features will contain non-zero, non-NoData cell values. The algorithm requires three passes of the raster. The first pass counts the number of line neighbours of each line cell; the second pass traces line segments starting from line ends (i.e. line cells with only one neighbouring line cell); lastly, the final pass traces any remaining line segments, which are likely forming closed loops (and therefore do not have line ends).

If the line raster contains streams, it is preferable to use the RasterStreamsToVector instead. This tool will use knowledge of flow directions to ensure connections between stream segments at confluence sites, whereas RasterToVectorLines will not.

See Also: RasterToVectorPolygons, RasterToVectorPoints, RasterStreamsToVector

Parameters:

Python function:

wbt.raster_to_vector_lines(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RasterToVectorLines -v ^
--wd="/path/to/data/" -i=lines.tif -o=lines.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 09/10/2018

Last Modified: 12/10/2018

RasterToVectorPoints

Converts a raster data set to a vector of the POINT shapetype. The user must specify the name of a raster file (--input) and the name of the output vector (--output). Points will correspond with grid cell centre points. All grid cells containing non-zero, non-NoData values will be considered a point. The vector's attribute table will contain a field called 'VALUE' that will contain the cell value for each point feature.

See Also: RasterToVectorPolygons, RasterToVectorLines

Parameters:

Python function:

wbt.raster_to_vector_points(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RasterToVectorPoints -v ^
--wd="/path/to/data/" --input=points.tif -o=out.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 25/09/2018

Last Modified: 12/10/2018

RasterToVectorPolygons

Converts a raster data set to a vector of the POLYGON geometry type. The user must specify the name of a raster file (--input) and the name of the output (--output) vector. All grid cells containing non-zero, non-NoData values will be considered part of a polygon feature. The vector's attribute table will contain a field called 'VALUE' that will contain the cell value for each polygon feature, in addition to the standard feature ID (FID) attribute.

See Also: RasterToVectorPoints, RasterToVectorLines

Parameters:

Python function:

wbt.raster_to_vector_polygons(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RasterToVectorPolygons -v ^
--wd="/path/to/data/" --input=points.tif -o=out.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 18/02/2020

Last Modified: 05/03/2020

ReinitializeAttributeTable

Reinitializes a vector's attribute table deleting all fields but the feature ID (FID). Caution: this tool overwrites the input file's attribute table.

Parameters:

Python function:

wbt.reinitialize_attribute_table(
    i, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ReinitializeAttributeTable -v ^
--wd="/path/to/data/" -i=input.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 04/09/2018

Last Modified: 12/10/2018

RemovePolygonHoles

This tool can be used to remove holes from the features within a vector polygon file. The user must specify the name of the input vector file, which must be of a polygon shapetype, and the name of the output file.

Parameters:

Python function:

wbt.remove_polygon_holes(
    i, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RemovePolygonHoles -v ^
--wd="/path/to/data/" --input=polygons.shp ^
--output=no_holes.shp 

Source code on GitHub

Author: Dr. John Lindsay

Created: 26/09/2018

Last Modified: 12/10/2018

RemoveRasterPolygonHoles

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool can be used to remove holes in raster polygons. Holes are areas of background values (either zero or NoData), completely surrounded by foreground values (any value other than zero or NoData). Therefore, this tool can somewhat be considered to be the raster equivalent to the vector-based RemovePolygonHoles tool. Users may optionally remove holes less than a specified threshold size (--threshold), measured in grid cells. Hole size is determined using a clumping operation, similar to what is used by the Clump tool. Users may also optionally specify whether or not to included 8-cell diagonal connectedness during the clumping operation (--use_diagonals).

Some GIS professionals have previously used a closing operation to lessen the extent of polygon holes in raster data. A closing is a mathematical morphology operation that involves expanding the raster polygons using a dialation filter (MaximumFilter), followed by a dialation filter (MinimumFilter) on the resulting image. While this common image processing technique can be helpful for reducing the prevalance of polygon holes, it can also have considerable impact on non-hole features within the image. The RemoveRasterPolygonHoles tool, by comparison, will only affect hole features and does not impact the boundaries of other polygons at all. The following image compares the removal of polygon holes (islands in a lake polygon) using a closing operation (middle) calculated using an 11x11 convolution filter and the output of the RemoveRasterPolygonHoles tool. Notice how the convolution operation impacts the edges of the polygon, particularly in convex regions, compared with the RemoveRasterPolygonHoles.

Here is a video that demonstrates how to apply this tool to a classified Sentinel-2 multi-spectral satellite imagery data set.

See Also: Closing, RemovePolygonHoles, Clump, GeneralizeClassifiedRaster

Parameters:

Python function:

wbt.remove_raster_polygon_holes(
    i, 
    output, 
    threshold=3, 
    use_diagonals=True, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=RemoveRasterPolygonHoles -i=input.tif ^
-o=output.tif --threshold=25 --use_diagonals 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 29/10/2022

Last Modified: 29/10/2022

SetNodataValue

This tool will re-assign a user-defined background value in an input raster image the NoData value. More precisely, the NoData value will be changed to the specified background value and any existing grid cells containing the previous NoData value, if it had been defined, will be changed to this new value. Most WhiteboxTools tools recognize NoData grid cells and treat them specially. NoData grid cells are also often displayed transparently by GIS software. The user must specify the names of the input and output rasters and the background value. The default background value is zero, although any numeric value is possible.

This tool differs from the ModifyNoDataValue tool in that it simply updates the NoData value in the raster header, without modifying pixel values. The ModifyNoDataValue tool will update the value in the header, and then modify each existing NoData pixel to contain this new value. Also, SetNodataValue does not overwrite the input file, while the ModifyNoDataValue tool does.

This tool may result in a change in the data type of the output image compared with the input image, if the background value is set to a negative value and the input image data type is an unsigned integer. In some cases, this may result in a doubling of the storage size of the output image.

See Also: ModifyNoDataValue, ConvertNodataToZero, IsNoData

Parameters:

Python function:

wbt.set_nodata_value(
    i, 
    output, 
    back_value=0.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=SetNodataValue -v --wd="/path/to/data/" ^
-i=in.tif -o=newRaster.tif --back_value=1.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 10/09/2017

Last Modified: 14/03/2023

SinglePartToMultiPart

This tool can be used to convert a vector file containing single-part features into a vector containing multi-part features. The user has the option to either group features based on an ID Field (--field flag), which is a categorical field within the vector's attribute table. The ID Field should either be of String (text) or Integer type. Fields containing decimal values are not good candidates for the ID Field. If no --field flag is specified, all features will be grouped together into one large multi-part vector.

This tool works for vectors containing either point, line, or polygon features. Since vectors of a POINT ShapeType cannot represent multi-part features, the ShapeType of the output file will be modified to a MULTIPOINT ShapeType if the input file is of a POINT ShapeType. If the input vector is of a POLYGON ShapeType, the user can optionally set the algorithm to search for polygons that should be represented as hole parts. In the case of grouping based on an ID Field, hole parts are polygon features contained within larger polygons of the same ID Field value. Please note that searching for polygon holes may significantly increase processing time for larger polygon coverages.

See Also: MultiPartToSinglePart

Parameters:

Python function:

wbt.single_part_to_multi_part(
    i, 
    output, 
    field=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=SinglePartToMultiPart -v ^
--wd="/path/to/data/" -i=input.shp -o=output.shp ^
--field='COUNTRY' 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/09/2018

Last Modified: 12/10/2018

VectorLinesToRaster

This tool can be used to convert a vector lines or polygon file into a raster grid of lines. If a vector of one of the polygon ShapeTypes is selected, the resulting raster will outline the polygons without filling these features. Use the VectorPolygonToRaster tool if you need to fill the polygon features.

The user must specify the name of the input vector (--input) and the output raster file (--output). The Field Name (--field) is the field from the attributes table, from which the tool will retrieve the information to assign to grid cells in the output raster. Note that if this field contains numerical data with no decimals, the output raster data type will be INTEGER; if it contains decimals it will be of a FLOAT data type. The field must contain numerical data. If the user does not supply a Field Name parameter, each feature in the raster will be assigned the record number of the feature. The assignment operation determines how the situation of multiple points contained within the same grid cell is handled. The background value is the value that is assigned to grid cells in the output raster that do not correspond to the location of any points in the input vector. This value can be any numerical value (e.g. 0) or the string 'NoData', which is the default.

If the user optionally specifies the --cell_size parameter then the coordinates will be determined by the input vector (i.e. the bounding box) and the specified Cell Size. This will also determine the number of rows and columns in the output raster. If the user instead specifies the optional base raster file parameter (--base), the output raster's coordinates (i.e. north, south, east, west) and row and column count will be the same as the base file. If the user does not specify either of these two optional parameters, the tool will determine the cell size automatically as the maximum of the north-south extent (determined from the shapefile's bounding box) or the east-west extent divided by 500.

See Also: VectorPointsToRaster, VectorPolygonsToRaster

Parameters:

Python function:

wbt.vector_lines_to_raster(
    i, 
    output, 
    field="FID", 
    nodata=True, 
    cell_size=None, 
    base=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=VectorLinesToRaster -v ^
--wd="/path/to/data/" -i=lines.shp --field=ELEV -o=output.tif ^
--nodata --cell_size=10.0
        >>./whitebox_tools ^
-r=VectorLinesToRaster -v --wd="/path/to/data/" -i=lines.shp ^
--field=FID -o=output.tif --base=existing_raster.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 18/04/2018

Last Modified: 22/10/2019

VectorPointsToRaster

This tool can be used to convert a vector points file into a raster grid. The user must specify the name of the input vector and the output raster file. The field name (--field) is the field from the attributes table from which the tool will retrieve the information to assign to grid cells in the output raster. The field must contain numerical data. If the user does not supply a field name parameter, each feature in the raster will be assigned the record number of the feature. The assignment operation determines how the situation of multiple points contained within the same grid cell is handled. The background value is zero by default but can be set to NoData optionally using the --nodata value.

If the user optionally specifies the grid cell size parameter (--cell_size) then the coordinates will be determined by the input vector (i.e. the bounding box) and the specified cell size. This will also determine the number of rows and columns in the output raster. If the user instead specifies the optional base raster file parameter (--base), the output raster's coordinates (i.e. north, south, east, west) and row and column count will be the same as the base file.

In the case that multiple points are contained within a single grid cell, the output can be assigned (--assign) the first, last (default), min, max, sum, mean, or number of the contained points.

See Also: VectorPolygonsToRaster, VectorLinesToRaster

Parameters:

Python function:

wbt.vector_points_to_raster(
    i, 
    output, 
    field="FID", 
    assign="last", 
    nodata=True, 
    cell_size=None, 
    base=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=VectorPointsToRaster -v ^
--wd="/path/to/data/" -i=points.shp --field=ELEV -o=output.tif ^
--assign=min --nodata ^
--cell_size=10.0
        >>./whitebox_tools ^
-r=VectorPointsToRaster -v --wd="/path/to/data/" -i=points.shp ^
--field=FID -o=output.tif --assign=last ^
--base=existing_raster.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 19/04/2018

Last Modified: 19/04/2023

VectorPolygonsToRaster

Converts a vector containing polygons into a raster.

Parameters:

Python function:

wbt.vector_polygons_to_raster(
    i, 
    output, 
    field="FID", 
    nodata=True, 
    cell_size=None, 
    base=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=VectorPolygonsToRaster -v ^
--wd="/path/to/data/" -i=lakes.shp --field=ELEV -o=output.tif ^
--nodata --cell_size=10.0
        >>./whitebox_tools ^
-r=VectorPolygonsToRaster -v --wd="/path/to/data/" ^
-i=lakes.shp --field=ELEV -o=output.tif ^
--base=existing_raster.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 17/04/2018

Last Modified: 18/10/2019

Geomorphometric Analysis

• AccumulationCurvature
• Aspect
• AssessRoute
• AverageHorizonDistance
• AverageNormalVectorAngularDeviation
• BreaklineMapping
• CircularVarianceOfAspect
• ContoursFromPoints
• ContoursFromRaster
• Curvedness
• DemVoidFilling
• DevFromMeanElev
• DiffFromMeanElev
• DifferenceCurvature
• DirectionalRelief
• DownslopeIndex
• EdgeDensity
• ElevAbovePit
• ElevPercentile
• ElevRelativeToMinMax
• ElevRelativeToWatershedMinMax
• EmbankmentMapping
• ExposureTowardsWindFlux
• FeaturePreservingSmoothing
• FetchAnalysis
• FillMissingData
• FindRidges
• GaussianCurvature
• GaussianScaleSpace
• GeneratingFunction
• Geomorphons
• Hillshade
• HorizonAngle
• HorizonArea
• HorizontalExcessCurvature
• HypsometricAnalysis
• HypsometricallyTintedHillshade
• LocalHypsometricAnalysis
• LocalQuadraticRegression
• MapOffTerrainObjects
• MaxAnisotropyDev
• MaxAnisotropyDevSignature
• MaxBranchLength
• MaxDifferenceFromMean
• MaxDownslopeElevChange
• MaxElevDevSignature
• MaxElevationDeviation
• MaxUpslopeElevChange
• MaximalCurvature
• MeanCurvature
• MinDownslopeElevChange
• MinimalCurvature
• MultidirectionalHillshade
• MultiscaleCurvatures
• MultiscaleElevationPercentile
• MultiscaleRoughness
• MultiscaleRoughnessSignature
• MultiscaleStdDevNormals
• MultiscaleStdDevNormalsSignature
• MultiscaleTopographicPositionImage
• NumDownslopeNeighbours
• NumUpslopeNeighbours
• Openness
• PennockLandformClass
• PercentElevRange
• PlanCurvature
• Profile
• ProfileCurvature
• RelativeAspect
• RelativeTopographicPosition
• RemoveOffTerrainObjects
• RingCurvature
• Rotor
• RuggednessIndex
• SedimentTransportIndex
• ShadowAnimation
• ShadowImage
• ShapeIndex
• SkyViewFactor
• SkylineAnalysis
• Slope
• SlopeVsAspectPlot
• SlopeVsElevationPlot
• SmoothVegetationResidual
• SphericalStdDevOfNormals
• StandardDeviationOfSlope
• StreamPowerIndex
• SurfaceAreaRatio
• TangentialCurvature
• TimeInDaylight
• TopoRender
• TopographicHachures
• TopographicPositionAnimation
• TotalCurvature
• Unsphericity
• VerticalExcessCurvature
• Viewshed
• VisibilityIndex
• WetnessIndex

AccumulationCurvature

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the accumulation curvature from a digital elevation model (DEM). Accumulation curvature is the product of profile (vertical) and tangential (horizontal) curvatures at a location (Shary, 1995). This variable has positive values, zero or greater. Florinsky (2017) states that accumulation curvature is a measure of the extent of local accumulation of flows at a given point in the topographic surface. Accumulation curvature is measured in units of m^-2.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary PA (1995) Land surface in gravity points classification by a complete system of curvatures. Mathematical Geology 27: 373–390.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

TangentialCurvature, ProfileCurvature, MinimalCurvature, MaximalCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.accumulation_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=AccumulationCurvature -i=DEM.tif ^
-o=output.tif --log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 11/01/2022

Last Modified: 28/11/2022

Aspect

This tool calculates slope aspect (i.e. slope orientation in degrees clockwise from north) for each grid cell in an input digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output raster image. The Z conversion factor is only important when the vertical and horizontal units are not the same in the DEM, and the DEM is in a projected coordinate system. When this is the case, the algorithm will multiply each elevation in the DEM by the Z conversion factor.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

Reference:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

See Also: Slope, TangentialCurvature, PlanCurvature, ProfileCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.aspect(
    dem, 
    output, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=Aspect -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 12/01/2022

AssessRoute

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool assesses the variability in slope, elevation, and visibility along a line vector, which may be a footpath, road, river or any other route. The user must specify the name of the input line vector (--routes), the input raster digital elevation model file (--dem), and the output line vector (--output). The algorithm initially splits the input line vector in equal-length segments (--length). For each line segment, the tool then calculates the average slope (AVG_SLOPE), minimum and maximum elevations (MIN_ELEV, MAX_ELEV), the elevation range or relief (RELIEF), the path sinuosity (SINUOSITY), the number of changes in slope direction or breaks-in-slope (CHG_IN_SLP), and the maximum visibility (VISIBILITY). Each of these metrics are output to the attribute table of the output vector, along with the feature identifier (FID); any attributes associated with the input parent feature will also be copied into the output table. Slope and elevation metrics are measured along the 2D path based on the elevations of each of the row and column intersection points of the raster with the path, estimated from linear-interpolation using the two neighbouring elevations on either side of the path. Sinuosity is calculated as the ratio of the along-surface (i.e. 3D) path length, divided by the 3D distance between the start and end points of the segment. CHG_IN_SLP can be thought of as a crude measure of path roughness, although this will be very sensitive to the quality of the DEM. The visibility metric is based on the Yokoyama et al. (2002) Openness index, which calculates the average horizon angle in the eight cardal directions to a maximum search distance (--dist), measured in grid cells.

Note that the input DEM must be in a projected coordinate system. The DEM and the input routes vector must be also share the same coordinate system. This tool also works best when the input DEM is of high quality and fine spatial resolution, such as those derived from LiDAR data sets.

Maximum segment visibility:

Average segment slope:

For more information about this tool, see this blog on the WhiteboxTools homepage.

See Also: SplitVectorLines, Openness

Parameters:

Python function:

wbt.assess_route(
    routes, 
    dem, 
    output, 
    length="", 
    dist=20, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=AssessRoute --routes=footpath.shp ^
--dem=DEM.tif -o=assessedRoutes.shp --length=50.0 --dist=200 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 16/05/2021

Last Modified: 16/05/2021

AverageHorizonDistance

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the spatial pattern of average distance to the horizon based on an input digital elevation model (DEM). As such, the index is a measure of landscape visibility. In the image below, lighter areas have a longer average distance to the horizon, measured in map units.

The user must specify an input DEM (dem), the azimuth fraction (az_fraction), the maximum search distance (max_dist), and the height offset of the observer (observer_hgt_offset). The input DEM should usually be a digital surface model (DSM) that contains significant off-terrain objects. Such a model, for example, could be created using the first-return points of a LiDAR data set, or using the lidar_digital_surface_model tool. The azimuth fraction should be an even divisor of 360-degrees and must be between 1-45 degrees.

The tool operates by calculating horizon angle (see horizon_angle) rasters from the DSM based on the user-specified azimuth fraction (az_fraction). For example, if an azimuth fraction of 15-degrees is specified, horizon angle rasters would be calculated for the solar azimuths 0, 15, 30, 45... A horizon angle raster evaluates the vertical angle between each grid cell in a DSM and a distant obstacle (e.g. a mountain ridge, building, tree, etc.) that obscures the view in a specified direction. In calculating horizon angle, the user must specify the maximum search distance (max_dist), in map units, beyond which the query for higher, more distant objects will cease. This parameter strongly impacts the performance of the function, with larger values resulting in significantly longer processing-times.

The observer_hgt_offset parameter can be used to add an increment to the source cell's elevation. For example, the following image shows the spatial pattern derived from a LiDAR DSM using observer_hgt_offset = 0.0:

Notice that there are several places, plarticularly on the flatter rooftops, where the local noise in the LiDAR DEM, associated with the individual scan lines, has resulted in a noisy pattern in the output. By adding a small height offset of the scale of this noise variation (0.15 m), we see that most of this noisy pattern is removed in the output below:

This feature makes the function more robust against DEM noise. As another example of the usefulness of this additional parameter, in the image below, the observer_hgt_offset parameter has been used to measure the pattern of the index at a typical human height (1.7 m):

Notice how at this height the average horizon distance becomes much farther on some of the flat rooftops where a guard wall prevents further viewing areas at shorter observer heights.

The output of this function is similar to the Average View Distance provided by the Sky View tool in Saga GIS. However, for a given maximum search distance, the Whitebox tool is likely faster to compute and has the added advantage of offering the observer's height parameter, as described above.

See Also:

SkyViewFactor, HorizonArea, SkylineAnalysis, Openness, TimeInDaylight, HorizonAngle

Parameters:

Python function:

wbt.average_horizon_distance(
    dem, 
    output, 
    az_fraction=5.0, 
    max_dist=9999.0, 
    observer_hgt_offset=0.05, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=AccumulationCurvature -i=DEM.tif ^
-o=output.tif --log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 09/04/2024

Last Modified: 09/04/2024

AverageNormalVectorAngularDeviation

This tool characterizes the spatial distribution of the average normal vector angular deviation, a measure of surface roughness. Working in the field of 3D printing, Ko et al. (2016) defined a measure of surface roughness based on quantifying the angular deviations in the direction of the normal vector of a real surface from its ideal (i.e. smoothed) form. This measure of surface complexity is therefore in units of degrees. Specifically, roughness is defined in this study as the neighborhood-averaged difference in the normal vectors of the original DEM and a smoothed DEM surface. Smoothed surfaces are derived by applying a Gaussian blur of the same size as the neighborhood (--filter).

The MultiscaleRoughness tool calculates the same measure of surface roughness, except that it is designed to work with multiple spatial scales.

Reference:

Ko, M., Kang, H., ulrim Kim, J., Lee, Y., & Hwang, J. E. (2016, July). How to measure quality of affordable 3D printing: Cultivating quantitative index in the user community. In International Conference on Human-Computer Interaction (pp. 116-121). Springer, Cham.

Lindsay, J. B., & Newman, D. R. (2018). Hyper-scale analysis of surface roughness. PeerJ Preprints, 6, e27110v1.

See Also: MultiscaleRoughness, SphericalStdDevOfNormals, CircularVarianceOfAspect

Parameters:

Python function:

wbt.average_normal_vector_angular_deviation(
    dem, 
    output, 
    filter=11, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=AverageNormalVectorAngularDeviation -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=roughness_mag.tif ^
--out_scale=roughness_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 26/01/2019

Last Modified: 03/09/2020

BreaklineMapping

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool can be used to map breaklines in an input digital elevation model (DEM; --input). Breaklines are locations of high surface curvature, in any direction, measured using Curvedness. Curvedness values are log-transformed using the resolution-dependent method proposed by Shary et al. (2002). Breaklines are coincident with grid cells that have log-transformed curvedness values exceeding a user-specified threshold value (--thresshold). While curvedness is measured within the range 0 to infinity, values are typically lower. Appropriate values for the threshold parameter are commonly in the 1 to 5 range. Lower threshold values will result in more extensive breakline mapping and vice versa. The algorithm will vectorize breakline features and the output of this tool (--output) is a line vector. Line features that are less than a user-specified length (in grid cells; --min_length), will not be output.

Watch the breakline mapping video for an example of how to run the tool.

References:

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

Curvedness

Parameters:

Python function:

wbt.breakline_mapping(
    dem, 
    output, 
    threshold=2.0, 
    min_length=3, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=BreaklineMapping -i=input.tif ^
-o=output.tif --threshold=0.8 --min_length=5 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 13/01/2022

Last Modified: 13/01/2022

CircularVarianceOfAspect

This tool can be used to calculate the circular variance (i.e. one minus the mean resultant length) of aspect for an input digital elevation model (DEM). This is a measure of how variable slope aspect is within a local neighbourhood of a specified size (--filter). CircularVarianceOfAspect is therefore a measure of surface shape complexity, or texture. It will take a value of 0.0 for smooth sites and near 1.0 in areas of high surface roughness or complex topography.

The local neighbourhood size (--filter) must be any odd integer equal to or greater than three. Grohmann et al. (2010) found that vector dispersion, a related measure of angular variance, increases monotonically with scale. This is the result of the angular dispersion measure integrating (accumulating) all of the surface variance of smaller scales up to the test scale. A more interesting scale relation can therefore be estimated by isolating the amount of surface complexity associated with specific scale ranges. That is, at large spatial scales, the metric should reflect the texture of large-scale landforms rather than the accumulated complexity at all smaller scales, including microtopographic roughness. As such, this tool normalizes the surface complexity of scales that are smaller than the filter size by applying Gaussian blur (with a standard deviation of one-third the filter size) to the DEM prior to calculating CircularVarianceOfAspect. In this way, the resulting distribution is able to isolate and highlight the surface shape complexity associated with landscape features of a similar scale to that of the filter size.

This tool makes extensive use of integral images (i.e. summed-area tables) and parallel processing to ensure computational efficiency. It may, however, require substantial memory resources when applied to larger DEMs.

References:

Grohmann, C. H., Smith, M. J., & Riccomini, C. (2010). Multiscale analysis of topographic surface roughness in the Midland Valley, Scotland. IEEE Transactions on Geoscience and Remote Sensing, 49(4), 1200-1213.

See Also: Aspect, SphericalStdDevOfNormals, MultiscaleRoughness, EdgeDensity, SurfaceAreaRatio, RuggednessIndex

Parameters:

Python function:

wbt.circular_variance_of_aspect(
    dem, 
    output, 
    filter=11, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=CircularVarianceOfAspect -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=roughness_mag.tif ^
--out_scale=roughness_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 26/01/2019

Last Modified: 03/09/2020

ContoursFromPoints

This tool creates a contour coverage from a set of input points (--input). The user must specify the contour interval (--interval) and optionally, the base contour value (--base). The degree to which contours are smoothed is controlled by the Smoothing Filter Size parameter (--smooth). This value, which determines the size of a mean filter applied to the x-y position of vertices in each contour, should be an odd integer value, e.g. 3, 5, 7, 9, 11, etc. Larger values will result in smoother contour lines.

See Also: ContoursFromRaster

Parameters:

Python function:

wbt.contours_from_points(
    i, 
    output, 
    field=None, 
    use_z=False, 
    max_triangle_edge_length=None, 
    interval=10.0, 
    base=0.0, 
    smooth=5, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ContoursFromPoints -v ^
--wd="/path/to/data/" -i=points.shp --field=HEIGHT ^
-o=contours.shp --max_triangle_edge_length=100.0 ^
--interval=100.0 --base=0.0 --smooth=11 

Source code on GitHub

Author: Dr. John Lindsay

Created: 26/04/2020

Last Modified: 24/06/2021

ContoursFromRaster

This tool can be used to create a vector contour coverage from an input raster surface model (--input), such as a digital elevation model (DEM). The user must specify the contour interval (--interval) and optionally, the base contour value (--base). The degree to which contours are smoothed is controlled by the Smoothing Filter Size parameter (--smooth). This value, which determines the size of a mean filter applied to the x-y position of vertices in each contour, should be an odd integer value, e.g. 3, 5, 7, 9, 11, etc. Larger values will result in smoother contour lines. The tolerance parameter (--tolerance) controls the amount of line generalization. That is, vertices in a contour line will be selectively removed from the line if they do not result in an angular deflection in the line's path of at least this threshold value. Increasing this value can significantly decrease the size of the output contour vector file, at the cost of generating straighter contour line segments.

See Also: RasterToVectorPolygons

Parameters:

Python function:

wbt.contours_from_raster(
    i, 
    output, 
    interval=10.0, 
    base=0.0, 
    smooth=9, 
    tolerance=10.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ContoursFromRaster -v ^
--wd="/path/to/data/" --input=DEM.tif -o=contours.shp ^
--interval=100.0 --base=0.0 --smooth=11 --tolerance=20.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/02/2020

Last Modified: 04/03/2020

Curvedness

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the curvedness (Koenderink and van Doorn, 1992) from a digital elevation model (DEM). Curvedness is the root mean square of maximal and minimal curvatures, and measures the magnitude of surface bending, regardless of shape (Florinsky, 2017). Curvedness is characteristically low-values for flat areas and higher for areas of sharp bending (Florinsky, 2017). The index is also inversely proportional with the size of the object (Koenderink and van Doorn, 1992). Curvedness has values equal to or greater than zero and is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Raw curvedness values are often challenging to visualize given their range and magnitude, and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Koenderink, J. J., and Van Doorn, A. J. (1992). Surface shape and curvature scales. Image and vision computing, 10(8), 557-564.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

ShapeIndex, MinimalCurvature, MaximalCurvature, TangentialCurvature, ProfileCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.curvedness(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=Curvedness -i=DEM.tif -o=output.tif ^
--log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 13/01/2022

Last Modified: 28/11/2022

DemVoidFilling

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool implements a modified version of the Delta Surface Fill method of Grohman et al. (2006). It can fill voids (i.e., data holes) contained within a digital elevation model (dem) by fusing the data with a second DEM (fill) that defines the topographic surface within the void areas. The two surfaces are fused seamlessly so that the transition from the source and fill surfaces is undetectable. The fill DEM need not have the same resolution as the source DEM.

The algorithm works by computing a DEM-of-difference (DoD) for each valid grid cell in the source DEM that also has a valid elevation in the corresponding location within the fill DEM. This difference surface is then used to define offsets within the near void-edge locations. The fill surface elevations are then combined with interpolated offsets, with the interpolation based on near-edge offsets, and used to define a new surface within the void areas of the source DEM in such a way that the data transitions seamlessly from the source data source to the fill data. The image below provides an example of this method.

The user must specify the mean_plane_dist parameter, which defines the distance (measured in grid cells) within a void area from the void's edge. Grid cells within larger voids that are beyond this distance from their edges have their vertical offsets, needed during the fusion of the DEMs, set to the mean offset for all grid cells that have both valid source and fill elevations. Void cells that are nearer their void edges have vertical offsets that are interpolated based on nearby offset values (i.e., the DEM of difference). The interpolation uses inverse-distance weighted (IDW) scheme, with a user-specified weight parameter (weight_value).

The edge_treatment parameter describes how the data fusion operates at the edges of voids, i.e., the first line of grid cells for which there are both source and fill elevation values. This parameter has values of "use DEM", "use Fill", and "average". Grohman et al. (2006) state that sometimes, due to a weakened signal within these marginal locations between the area of valid data and voids, the estimated elevation values are inaccurate. When this is the case, it is best to use fill elevations in the transitional areas. If this isn't the case, the "use DEM" is the better option. A compromise between the two options is to average the two elevation sources.

References:

Grohman, G., Kroenung, G. and Strebeck, J., 2006. Filling SRTM voids: The delta surface fill method. Photogrammetric Engineering and Remote Sensing, 72(3), pp.213-216.

See Also: FillMissingData

Parameters:

Python function:

wbt.dem_void_filling(
    dem, 
    fill, 
    output, 
    mean_plane_dist=20, 
    edge_treatment="use DEM", 
    weight_value=2.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=DemVoidFilling --dem=DEM.tif ^
--fill=fill_DEM.tif --output=dem_fused.tif --mean_plane_dist=25 ^
--weight_value=1.0 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 24/05/2022

Last Modified: 09/02/2023

DevFromMeanElev

This tool can be used to calculate the difference between the elevation of each grid cell and the mean elevation of the centering local neighbourhood, normalized by standard deviation. Therefore, this index of topographic residual is essentially equivalent to a local z-score. This attribute measures the relative topographic position as a fraction of local relief, and so is normalized to the local surface roughness. DevFromMeanElev utilizes an integral image approach (Crow, 1984) to ensure highly efficient filtering that is invariant with filter size.

The user must specify the name (--dem) of the input digital elevation model (DEM), the name of the output file (--output), and the size of the neighbourhood in the x and y directions (--filterx and --filtery), measured in grid size.

While DevFromMeanElev calculates the deviation from mean elevation (DEV) at a single, user-defined scale, the MaxElevationDeviation tool can be used to output the per-pixel maximum DEV value across a range of input scales.

See Also: DiffFromMeanElev, MaxElevationDeviation

Parameters:

Python function:

wbt.dev_from_mean_elev(
    dem, 
    output, 
    filterx=11, 
    filtery=11, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=DevFromMeanElev -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif ^
--filter=25 

Source code on GitHub

Author: Dr. John Lindsay

Created: 21/06/2017

Last Modified: 30/01/2020

DiffFromMeanElev

This tool can be used to calculate the difference between the elevation of each grid cell and the mean elevation of the centering local neighbourhood. This is similar to what a high-pass filter calculates for imagery data, but is intended to work with DEM data instead. This attribute measures the relative topographic position. DiffFromMeanElev utilizes an integral image approach (Crow, 1984) to ensure highly efficient filtering that is invariant with filter size.

The user must specify the name (--dem) of the input digital elevation model (DEM), the name of the output file (--output), and the size of the neighbourhood in the x and y directions (--filterx and --filtery), measured in grid size.

While DevFromMeanElev calculates the DIFF at a single, user-defined scale, the MaxDifferenceFromMean tool can be used to output the per-pixel maximum DIFF value across a range of input scales.

See Also: DevFromMeanElev, MaxDifferenceFromMean

Parameters:

Python function:

wbt.diff_from_mean_elev(
    dem, 
    output, 
    filterx=11, 
    filtery=11, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=DiffFromMeanElev -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif ^
--filter=25 

Source code on GitHub

Author: Dr. John Lindsay

Created: 25/06/2017

Last Modified: 30/01/2020

DifferenceCurvature

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the difference curvature from a digital elevation model (DEM). Difference curvature is half of the difference between profile and tangential curvatures, sometimes called the vertical and horizontal curvatures (Shary, 1995). This variable has an unbounded range that can take either positive or negative values. Florinsky (2017) states that difference curvature measures the extent to which the relative deceleration of flows (measured by kv) is higher than flow convergence at a given point of the topographic surface. Difference curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary PA (1995) Land surface in gravity points classification by a complete system of curvatures. Mathematical Geology 27: 373–390.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

ProfileCurvature, TangentialCurvature, Rotor, MinimalCurvature, MaximalCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.difference_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=DifferenceCurvature -i=DEM.tif ^
-o=output.tif --log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 11/01/2022

Last Modified: 28/11/2022

DirectionalRelief

This tool calculates the relief for each grid cell in a digital elevation model (DEM) in a specified direction. Directional relief is an index of the degree to which a DEM grid cell is higher or lower than its surroundings. It is calculated by subtracting the elevation of a DEM grid cell from the average elevation of those cells which lie between it and the edge of the DEM in a specified compass direction. Thus, positive values indicate that a grid cell is lower than the average elevation of the grid cells in a specific direction (i.e. relatively sheltered), whereas a negative directional relief indicates that the grid cell is higher (i.e. relatively exposed). The algorithm is based on a modification of the procedure described by Lapen and Martz (1993). The modifications include: (1) the ability to specify any direction between 0-degrees and 360-degrees (--azimuth), and (2) the ability to use a distance-limited search (--max_dist), such that the ray-tracing procedure terminates before the DEM edge is reached for longer search paths. The algorithm works by tracing a ray from each grid cell in the direction of interest and evaluating the average elevation along the ray. Linear interpolation is used to estimate the elevation of the surface where a ray does not intersect the DEM grid precisely at one of its nodes. The user must specify the name of an input DEM raster file, the output raster name, and a hypothetical wind direction. Furthermore, the user is able to constrain the maximum search distance for the ray tracing. If no maximum search distance is specified, each ray will be traced to the edge of the DEM. The units of the output image are the same as the input DEM.

Ray-tracing is a highly computationally intensive task and therefore this tool may take considerable time to operate for larger sized DEMs. This tool is parallelized to aid with computational efficiency. NoData valued grid cells in the input image will be assigned NoData values in the output image. The output raster is of the float data type and continuous data scale. Directional relief is best displayed using the blue-white-red bipolar palette to distinguish between the positive and negative values that are present in the output.

Reference:

Lapen, D. R., & Martz, L. W. (1993). The measurement of two simple topographic indices of wind sheltering-exposure from raster digital elevation models. Computers & Geosciences, 19(6), 769-779.

See Also: FetchAnalysis, HorizonAngle, RelativeAspect

Parameters:

Python function:

wbt.directional_relief(
    dem, 
    output, 
    azimuth=0.0, 
    max_dist=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=DirectionalRelief -v ^
--wd="/path/to/data/" -i='input.tif' -o=output.tif ^
--azimuth=315.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 07/07/2017

Last Modified: 03/09/2020

DownslopeIndex

This tool can be used to calculate the downslope index described by Hjerdt et al. (2004). The downslope index is a measure of the slope gradient between a grid cell and some downslope location (along the flowpath passing through the upslope grid cell) that represents a specified vertical drop (i.e. a potential head drop). The index has been shown to be useful for hydrological, geomorphological, and biogeochemical applications.

The user must specify the name of a digital elevaton model (DEM) raster. This DEM should be have been pre-processed to remove artifact topographic depressions and flat areas. The user must also specify the head potential drop (d), and the output type. The output type can be either 'tangent', 'degrees', 'radians', or 'distance'. If 'distance' is selected as the output type, the output grid actually represents the downslope flowpath length required to drop d meters from each grid cell. Linear interpolation is used when the specified drop value is encountered between two adjacent grid cells along a flowpath traverse.

Notice that this algorithm is affected by edge contamination. That is, for some grid cells, the edge of the grid will be encountered along a flowpath traverse before the specified vertical drop occurs. In these cases, the value of the downslope index is approximated by replacing d with the actual elevation drop observed along the flowpath. To avoid this problem, the entire watershed containing an area of interest should be contained in the DEM.

Grid cells containing NoData values in any of the input images are assigned the NoData value in the output raster. The output raster is of the float data type and continuous data scale.

Reference:

Hjerdt, K.N., McDonnell, J.J., Seibert, J. Rodhe, A. (2004) A new topographic index to quantify downslope controls on local drainage, Water Resources Research, 40, W05602, doi:10.1029/2004WR003130.

Parameters:

Python function:

wbt.downslope_index(
    dem, 
    output, 
    drop=2.0, 
    out_type="tangent", 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=DownslopeIndex -v --wd="/path/to/data/" ^
--dem=pointer.tif -o=dsi.tif --drop=5.0 --out_type=distance 

Source code on GitHub

Author: Dr. John Lindsay

Created: July 17, 2017

Last Modified: 12/10/2018

EdgeDensity

This tool calculates the density of edges, or breaks-in-slope within an input digital elevation model (DEM). A break-in-slope occurs between two neighbouring grid cells if the angular difference between their normal vectors is greater than a user-specified threshold value (--norm_diff). EdgeDensity calculates the proportion of edge cells within the neighbouring window, of square filter dimension --filter, surrounding each grid cell. Therefore, EdgeDensity is a measure of how complex the topographic surface is within a local neighbourhood. It is therefore a measure of topographic texture. It will take a value near 0.0 for smooth sites and 1.0 in areas of high surface roughness or complex topography.

The distribution of EdgeDensity is highly dependent upon the value of the norm_diff used in the calculation. This threshold may require experimentation to find an appropriate value and is likely dependent upon the topography and source data. Nonetheless, experience has shown that EdgeDensity provides one of the best measures of surface texture of any of the available roughness tools.

See Also: CircularVarianceOfAspect, MultiscaleRoughness, SurfaceAreaRatio, RuggednessIndex

Parameters:

Python function:

wbt.edge_density(
    dem, 
    output, 
    filter=11, 
    norm_diff=5.0, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=EdgeDensity -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif --filter=15 --norm_diff=20.0 ^
--num_iter=4 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/01/2019

Last Modified: 03/09/2020

ElevAbovePit

This tool will calculate the elevation of each grid cell in a digital elevation model (DEM) above the nearest downslope pit cell or grid edge cell, depending on which is encountered first during the flow-path traverse. The resulting image is therefore a measure of relative landscape position. The user must specify the names of a D8 flow pointer grid, a DEM file, and the output file. The flow pointer grid must be derived using the D8 flow algorithm.

See Also: ElevationAboveStream

Parameters:

Python function:

wbt.elev_above_pit(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ElevAbovePit -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 13/07/2017

Last Modified: 12/10/2018

ElevPercentile

Elevation percentile (EP) is a measure of local topographic position (LTP). It expresses the vertical position for a digital elevation model (DEM) grid cell (z₀) as the percentile of the elevation distribution within the filter window, such that:

EP = count_i∈C(z_i > z₀) x (100 / n_C)

where z₀ is the elevation of the window's center grid cell, z_i is the elevation of cell i contained within the neighboring set C, and n_C is the number of grid cells contained within the window.

EP is unsigned and expressed as a percentage, bound between 0% and 100%. Quantile-based estimates (e.g., the median and interquartile range) are often used in nonparametric statistics to provide data variability estimates without assuming the distribution is normal. Thus, EP is largely unaffected by irregularly shaped elevation frequency distributions or by outliers in the DEM, resulting in a highly robust metric of LTP. In fact, elevation distributions within small to medium sized neighborhoods often exhibit skewed, multimodal, and non-Gaussian distributions, where the occurrence of elevation errors can often result in distribution outliers. Thus, based on these statistical characteristics, EP is considered one of the most robust representation of LTP.

The algorithm implemented by this tool uses the relatively efficient running-histogram filtering algorithm of Huang et al. (1979). Because most DEMs contain floating point data, elevation values must be rounded to be binned. The --sig_digits parameter is used to determine the level of precision preserved during this binning process. The algorithm is parallelized to further aid with computational efficiency.

Neighbourhood size, or filter size, is specified in the x and y dimensions using the --filterx and --filteryflags. These dimensions should be odd, positive integer values (e.g. 3, 5, 7, 9, etc.).

References:

Newman, D. R., Lindsay, J. B., and Cockburn, J. M. H. (2018). Evaluating metrics of local topographic position for multiscale geomorphometric analysis. Geomorphology, 312, 40-50.

Huang, T., Yang, G.J.T.G.Y. and Tang, G., 1979. A fast two-dimensional median filtering algorithm. IEEE Transactions on Acoustics, Speech, and Signal Processing, 27(1), pp.13-18.

See Also: DevFromMeanElev, DiffFromMeanElev

Parameters:

Python function:

wbt.elev_percentile(
    dem, 
    output, 
    filterx=11, 
    filtery=11, 
    sig_digits=2, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ElevPercentile -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif --filter=25 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 02/04/2019

ElevRelativeToMinMax

This tool can be used to express the elevation of a grid cell in a digital elevation model (DEM) as a percentage of the relief between the DEM minimum and maximum values. As such, it provides a basic measure of relative topographic position.

See Also: ElevRelativeToWatershedMinMax, ElevationAboveStream, ElevAbovePit

Parameters:

Python function:

wbt.elev_relative_to_min_max(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ElevRelativeToMinMax -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/07/2017

Last Modified: 12/10/2018

ElevRelativeToWatershedMinMax

This tool can be used to express the elevation of a grid cell in a digital elevation model (DEM) as a percentage of the relief between the watershed minimum and maximum values. As such, it provides a basic measure of relative topographic position. The user must specify the names of DEM (--dem) and watersheds (--watersheds) raster files.

See Also: ElevRelativeToMinMax, ElevationAboveStream, ElevAbovePit

Parameters:

Python function:

wbt.elev_relative_to_watershed_min_max(
    dem, 
    watersheds, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ElevRelativeToWatershedMinMax -v ^
--wd="/path/to/data/" --dem=DEM.tif --watersheds=watershed.tif ^
-o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/07/2017

Last Modified: 12/10/2018

EmbankmentMapping

This tool can be used to map and/or remove road embankments from an input fine-resolution digital elevation model (--dem). Fine-resolution LiDAR DEMs can represent surface features such as road and railway embankments with high fidelity. However, transportation embankments are problematic for several environmental modelling applications, including soil an vegetation distribution mapping, where the pre-embankment topography is the contolling factor. The algorithm utilizes repositioned (--search_dist) transportation network cells, derived from rasterizing a transportation vector (--road_vec), as seed points in a region-growing operation. The embankment region grows based on derived morphometric parameters, including road surface width (--min_road_width), embankment width (--typical_width and --max_width), embankment height (--max_height), and absolute slope (--spillout_slope). The tool can be run in two modes. By default the tool will simply map embankment cells, with a Boolean output raster. If, however, the --remove_embankments flag is specified, the tool will instead output a DEM for which the mapped embankment grid cells have been excluded and new surfaces have been interpolated based on the surrounding elevation values (see below).

Hillshade from original DEM:

Hillshade from embankment-removed DEM:

References:

Van Nieuwenhuizen, N, Lindsay, JB, DeVries, B. 2021. Automated mapping of transportation embankments in fine-resolution LiDAR DEMs. Remote Sensing. 13(7), 1308; https://doi.org/10.3390/rs13071308

See Also: RemoveOffTerrainObjects, SmoothVegetationResidual

Parameters:

Python function:

wbt.embankment_mapping(
    dem, 
    road_vec, 
    output, 
    search_dist=2.5, 
    min_road_width=6.0, 
    typical_width=30.0, 
    max_height=2.0, 
    max_width=60.0, 
    max_increment=0.05, 
    spillout_slope=4.0, 
    remove_embankments=False, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=EmbankmentMapping -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif ^
--search_dist=1.0 --min_road_width=6.0 --typical_width=30.0 ^
--max_height=2.0 --max_width=60.0 --max_increment=0.05 ^
--spillout_slope=4.0 --remove_embankments=true 

Source code on GitHub

Author: Dr. John Lindsay and Nigel Van Nieuwenhuizen

Created: 21/09/2020

Last Modified: 05/10/2020

ExposureTowardsWindFlux

This tool creates a new raster in which each grid cell is assigned the exposure of the land-surface to a hypothetical wind flux. It can be conceptualized as the angle between a plane orthogonal to the wind and a plane that represents the local topography at a grid cell (Bohner and Antonic, 2007). The user must specify the names of the input digital elevation model (--dem) and output file (--output), as well as the dominant wind azimuth (--azimuth) and a maximum search distance (--max_dist) used to calclate the horizon angle. Notice that the specified azimuth represents a regional average wind direction.

Exposure towards the sloped wind flux essentially combines the relative terrain aspect and the maximum upwind slope (i.e. horizon angle). This terrain attribute accounts for land-surface orientation, relative to the wind, and shadowing effects of distant topographic features but does not account for deflection of the wind by topography. This tool should not be used on very extensive areas over which Earth's curvature must be taken into account. DEMs in projected coordinate systems are preferred.

Algorithm Description:

Exposure is measured based on the equation presented in Antonic and Legovic (1999):

cos(E) = cos(S) sin(H) + sin(S) cos(H) cos(Az - A)

Where, E is angle between a plane defining the local terrain and a plane orthogonal to the wind flux, S is the terrain slope, A is the terrain aspect, Az is the azimuth of the wind flux, and H is the horizon angle of the wind flux, which is zero when only the horizontal component of the wind flux is accounted for.

Exposure images are best displayed using a greyscale or bipolar palette to distinguish between the positive and negative values that are present in the output.

References:

Antonić, O., & Legović, T. 1999. Estimating the direction of an unknown air pollution source using a digital elevation model and a sample of deposition. Ecological modelling, 124(1), 85-95.

Böhner, J., & Antonić, O. 2009. Land-surface parameters specific to topo-climatology. Developments in Soil Science, 33, 195-226.

See Also: RelativeAspect

Parameters:

Python function:

wbt.exposure_towards_wind_flux(
    dem, 
    output, 
    azimuth="", 
    max_dist="", 
    zfactor="", 
    callback=default_callback
)

Command-line Interface:

-./whitebox_tools -r=ExposureTowardsWindFlux -dem=input.tif ^
-o=exposure.tif --azimuth=35.0 --max_dist=500.0 

Source code on GitHub

Author: Whitebox Geospatial Inc. (c)

Created: 20/07/2021

Last Modified: 20/07/2021

FeaturePreservingSmoothing

This tool implements a highly modified form of the DEM de-noising algorithm described by Sun et al. (2007). It is very effective at removing surface roughness from digital elevation models (DEMs), without significantly altering breaks-in-slope. As such, this tool should be used for smoothing DEMs rather than either smoothing with low-pass filters (e.g. mean, median, Gaussian filters) or grid size coarsening by resampling. The algorithm works by 1) calculating the surface normal 3D vector of each grid cell in the DEM, 2) smoothing the normal vector field using a filtering scheme that applies more weight to neighbours with lower angular difference in surface normal vectors, and 3) uses the smoothed normal vector field to update the elevations in the input DEM.

Sun et al.'s (2007) original method was intended to work on input point clouds and fitted triangular irregular networks (TINs). The algorithm has been modified to work with input raster DEMs instead. In so doing, this algorithm calculates surface normal vectors from the planes fitted to 3 x 3 neighbourhoods surrounding each grid cell, rather than the triangular facet. The normal vector field smoothing and elevation updating procedures are also based on raster filtering operations. These modifications make this tool more efficient than Sun's original method, but will also result in a slightly different output than what would be achieved with Sun's method.

The user must specify the values of three key parameters, including the filter size (--filter), the normal difference threshold (--norm_diff), and the number of iterations (--num_iter). Lindsay et al. (2019) found that the degree of smoothing was less impacted by the filter size than it was either the normal difference threshold and the number of iterations. A filter size of 11, the default value, tends to work well in many cases. To increase the level of smoothing applied to the DEM, consider increasing the normal difference threshold, i.e. the angular difference in normal vectors between the center cell of a filter window and a neighbouring cell. This parameter determines which neighbouring values are included in a filtering operation and higher values will result in a greater number of neighbouring cells included, and therefore smooother surfaces. Similarly, increasing the number of iterations from the default value of 3 to upwards of 5-10 will result in significantly greater smoothing.

Before smoothing treatment:

After smoothing treatment with FPS:

For a video tutorial on how to use the FeaturePreservingSmoothing tool, please see this YouTube video.

Reference:

Lindsay JB, Francioni A, Cockburn JMH. 2019. LiDAR DEM smoothing and the preservation of drainage features. Remote Sensing, 11(16), 1926; DOI: 10.3390/rs11161926.

Sun, X., Rosin, P., Martin, R., & Langbein, F. (2007). Fast and effective feature-preserving mesh denoising. IEEE Transactions on Visualization & Computer Graphics, (5), 925-938.

Parameters:

Python function:

wbt.feature_preserving_smoothing(
    dem, 
    output, 
    filter=11, 
    norm_diff=15.0, 
    num_iter=3, 
    max_diff=0.5, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=FeaturePreservingSmoothing -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif --filter=15 ^
--norm_diff=20.0 --num_iter=4 

Source code on GitHub

Author: Dr. John Lindsay

Created: 23/11/2017

Last Modified: 03/09/2020

FetchAnalysis

This tool creates a new raster in which each grid cell is assigned the distance, in meters, to the nearest topographic obstacle in a specified direction. It is a modification of the algorithm described by Lapen and Martz (1993). Unlike the original algorithm, Fetch Analysis is capable of analyzing fetch in any direction from 0-360 degrees. The user must specify the name of an input digital elevation model (DEM) raster file, the output raster name, a hypothetical wind direction, and a value for the height increment parameter. The algorithm searches each grid cell in a path following the specified wind direction until the following condition is met:

Z_test >= Z_core + DI

where Z_core is the elevation of the grid cell at which fetch is being determined, Z_test is the elevation of the grid cell being tested as a topographic obstacle, D is the distance between the two grid cells in meters, and I is the height increment in m/m. Lapen and Martz (1993) suggest values for I in the range of 0.025 m/m to 0.1 m/m based on their study of snow re-distribution in low-relief agricultural landscapes of the Canadian Prairies. If the directional search does not identify an obstacle grid cell before the edge of the DEM is reached, the distance between the DEM edge and Zcore is entered. Edge distances are assigned negative values to differentiate between these artificially truncated fetch values and those for which a valid topographic obstacle was identified. Notice that linear interpolation is used to estimate the elevation of the surface where a ray (i.e. the search path) does not intersect the DEM grid precisely at one of its nodes.

Ray-tracing is a highly computationally intensive task and therefore this tool may take considerable time to operate for larger sized DEMs. This tool is parallelized to aid with computational efficiency. NoData valued grid cells in the input image will be assigned NoData values in the output image. Fetch Analysis images are best displayed using the blue-white-red bipolar palette to distinguish between the positive and negative values that are present in the output.

Reference:

Lapen, D. R., & Martz, L. W. (1993). The measurement of two simple topographic indices of wind sheltering-exposure from raster digital elevation models. Computers & Geosciences, 19(6), 769-779.

See Also: DirectionalRelief, HorizonAngle, RelativeAspect

Parameters:

Python function:

wbt.fetch_analysis(
    dem, 
    output, 
    azimuth=0.0, 
    hgt_inc=0.05, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=FetchAnalysis -v --wd="/path/to/data/" ^
-i='input.tif' -o=output.tif --azimuth=315.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 07/07/2017

Last Modified: 03/09/2020

FillMissingData

This tool can be used to fill in small gaps in a raster or digital elevation model (DEM). The gaps, or holes, must have recognized NoData values. If gaps do not currently have this characteristic, use the SetNodataValue tool and ensure that the data are stored using a raster format that supports NoData values. All valid, non-NoData values in the input raster will be assigned the same value in the output image.

The algorithm uses an inverse-distance weighted (IDW) scheme based on the valid values on the edge of NoData gaps to estimate gap values. The user must specify the filter size (--filter), which determines the size of gap that is filled, and the IDW weight (--weight).

The filter size, specified in grid cells, is used to determine how far the algorithm will search for valid, non-NoData values. Therefore, setting a larger filter size allows for the filling of larger gaps in the input raster.

The --no_edges flag can be used to exclude NoData values that are connected to the edges of the raster. It is usually the case that irregularly shaped DEMs have large regions of NoData values along the containing raster edges. This flag can be used to exclude these regions from the gap-filling operation, leaving only interior gaps for filling.

See Also: SetNodataValue

Parameters:

Python function:

wbt.fill_missing_data(
    i, 
    output, 
    filter=11, 
    weight=2.0, 
    no_edges=True, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=FillMissingData -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif --filter=25 ^
--weight=1.0 --no_edges 

Source code on GitHub

Author: Dr. John Lindsay

Created: 14/06/2017

Last Modified: 12/10/2018

FindRidges

This tool can be used to identify ridge cells in a digital elevation model (DEM). Ridge cells are those that have lower neighbours either to the north and south or the east and west. Line thinning can optionally be used to create single-cell wide ridge networks by specifying the --line_thin parameter.

Parameters:

Python function:

wbt.find_ridges(
    dem, 
    output, 
    line_thin=True, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=FindRidges -v --wd="/path/to/data/" ^
--dem=pointer.tif -o=out.tif --line_thin 

Source code on GitHub

Author: Dr. John Lindsay

Created: 04/12/2017

Last Modified: 18/10/2019

GaussianCurvature

This tool calculates the Gaussian curvature from a digital elevation model (DEM). Gaussian curvature is the product of maximal and minimal curvatures, and retains values in each point of the topographic surface after its bending without breaking, stretching, and compressing (Florinsky, 2017). Gaussian curvature is measured in units of m^-2.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

TangentialCurvature, ProfileCurvature, PlanCurvature, MeanCurvature, MinimalCurvature, MaximalCurvature

Parameters:

Python function:

wbt.gaussian_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=GaussianCurvature -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/01/2022

Last Modified: 15/01/2022

GaussianScaleSpace

This tool uses the fast Gaussian approximation algorithm to produce scaled land-surface parameter (LSP) measurements from an input DEM (--dem). The algorithm iterates over scales defined by an initial scale (--sigma), a step size (--step) and a number of scales (--num_steps)/. After smoothing the input DEM to the target sigma, a 3x3 window is used to calculate a variety of LSPs (--lsp). LSP options include local derivatives (Elevation, Slope, Aspect, Eastness, Northness, Mean curvature, Plan curvature, Profile curvature, Tangential curvature and Total curvature), Hillshade, and Difference from mean elevation, all as defined in Wilson (2018), and Anisotropy of topographic position (Newman et al., 2018), and Ruggedness (Riley et al., 1999). An initial sigma value of 0 will compute the LSP without Gaussian smoothing. The step size can be and positive number, however, sigam values < 0.5 and duplicated scales are skipped due to a minimum filter size limitation, or the scale discretization of the fast gaussian approximation.

The LSP values are then transformed to z-scores using the population of values at a single scale, are are evaluated to identify the optimal scale defined as the maximum absolute z-score for each cell. The final outputs are three rasters: the first containing the z-score at the optimal scale (z_opt), the sigma value at the optimal scale (g_opt), and the LSP value at the optimal scale (v_opt). These all need to be specified using the (--output_zscore), (--output_scale), and (--output) flags respectively. Additionally, a vector file of points (--points) can optionally be provided to generate scale signatures for the provided point locations.

Due to the use of the integral image, edge effects can be problematic; especially when 'NoData' values are found. It is recommended that 'NoData' holes filled during pre-processing. Irregular shaped data (i.e., non-rectangular DEMs) are buffered with a crude check for 'NoData' values at the filter edge in the 8 cardinal directions to buffer the edges. This should be adequate for most data, additional buffer masks may be required.

Reference:

Wilson, J. P. (2018). Environmental-applications-of-digital-terrain-modeling. Wiley Blackwell. Newman, D. R., Lindsay, J. B., & Cockburn, J. M. H. (2018). Measuring Hyperscale Topographic Anisotropy as a Continuous Landscape Property. Geosciences, 8(278). https://doi.org/10.3390/geosciences8080278

Riley, S. J., DeGloria, S. D., and Elliot, R. (1999). Index that quantifies topographic heterogeneity.Intermountain Journal of Sciences, 5(1-4), 23-27.

See Also: MaxDifferenceFromMean, MaxAnisotropyDev, ProfileCurvature, TangentialCurvature, RuggednessIndex

Parameters:

Python function:

wbt.gaussian_scale_space(
    dem, 
    output, 
    output_zscore, 
    output_scale, 
    points=None, 
    sigma=0.5, 
    step=0.5, 
    num_steps=10, 
    lsp="Slope", 
    z_factor=None, 
    callback=default_callback
)

Command-line Interface:

-./whitebox_tools -r=GaussianScaleSpace --dem=DEM.tif ^
--output=slope.tif --output_zscore=slope_z.tif ^
--output_scale=slope_scale.tif --sigma=0.5 --step=1.0 ^
--num_steps=100 --lsp='Slope' 

Source code on GitHub

Author: Daniel Newman

Created: 01/01/2021

Last Modified: 01/01/2021

GeneratingFunction

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the generating function (Shary and Stepanov, 1991) from a digital elevation model (DEM). Florinsky (2016) describes generating function as a measure for the deflection of tangential curvature from loci of extreme curvature of the topographic surface. Florinsky (2016) demonstrated the application of this variable for identifying landscape structural lines, i.e. ridges and thalwegs, for which the generating function takes values near zero. Ridges coincide with divergent areas where generating function is approximately zero, while thalwegs are associated with convergent areas with generating function values near zero. This variable has positive values, zero or greater and is measured in units of m^-2.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Raw generating function values are often challenging to visualize given their range and magnitude, and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

This tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems, however, this tool cannot use the same 3x3 polynomial fitting method for equal angle grids, also described by Florinsky (2016), that is used by the other curvature tools in this software. That is because generating function uses 3rd order partial derivatives, which cannot be calculated using the 9 elevations in a 3x3; more elevation values are required (i.e. a 5x5 window). Thus, this tool uses the same 5x5 method used for DEMs in projected coordinate systems, and calculates the average linear distance between neighbouring cells in the vertical and horizontal directions using the Vincenty distance function. Note that this may cause a notable slow-down in algorithm performance and has a lower accuracy than would be achieved using an equal angle method, because it assumes a square pixel (in linear units).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Koenderink, J. J., and Van Doorn, A. J. (1992). Surface shape and curvature scales. Image and vision computing, 10(8), 557-564.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

Shary P. A. and Stepanov I. N. (1991) Application of the method of second derivatives in geology. Transactions (Doklady) of the USSR Academy of Sciences, Earth Science Sections 320: 87–92.

See Also:

ShapeIndex, MinimalCurvature, MaximalCurvature, TangentialCurvature, ProfileCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.generating_function(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=GeneratingFunction -i=DEM.tif ^
-o=output.tif --log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 11/01/2022

Last Modified: 28/11/2022

Geomorphons

This tool can be used to perform a geomorphons landform classification based on an input digital elevation model (--dem). The geomorphons concept is based on line-of-sight analysis for the eight topographic profiles in the cardinal directions surrounding each grid cell in the input DEM. The relative sizes of the zenith angle of a profile's maximum elevation angle (i.e. horizon angle) and the nadir angle of a profile's minimum elevation angle are then used to generate a ternary (base-3) digit: 0 when the nadir angle is less than the zenith angle, 1 when the two angles differ by less than a user-defined flatness threshold (--threshold), and 2 when the nadir angle is greater than the zenith angle. A ternary number is then derived from the digits assigned to each of the eight profiles, with digits sequenced counter-clockwise from east. This ternary number forms the geomorphons code assigned to the grid cell. There are 3⁸ = 6561 possible codes, although many of these codes are equivalent geomorphons through rotations and reflections. Some of the remaining geomorphons also rarely if ever occur in natural topography. Jasiewicz et al. (2013) identified 10 common landform types by reclassifying related geomorphons codes. The user may choose to output these common forms (--forms) rather than the the raw ternary code. These landforms include:

One of the main advantages of the geomrophons method is that, being based on minimum/maximum elevation angles, the scale used to estimate the landform type at a site adapts to the surrounding terrain. In principle, choosing a large value of search distance (--search) should result in identification of a landform element regardless of its scale.

An experimental feature has been added to correct for global inclination. Global inclination biases the flatness threshold angle becasue it is measured relative to the z-axis, especially in locally flat areas. Including the --residuals flag "flattens" the input by converting elevation to residuals of a 2-d linear model.

Reference:

Jasiewicz, J., and Stepinski, T. F. (2013). Geomorphons — a pattern recognition approach to classification and mapping of landforms. Geomorphology, 182, 147-156.

See Also: PennockLandformClass

Parameters:

Python function:

wbt.geomorphons(
    dem, 
    output, 
    search=50, 
    threshold=0.0, 
    fdist=0, 
    skip=0, 
    forms=True, 
    residuals=False, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=Geomorphons -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif --search=50 --threshold=0.0 ^
--tdist=0.0 --forms 

Source code on GitHub

Author: Daniel Newman

Created: 21/09/2018

Last Modified: 07/06/2022

Hillshade

This tool performs a hillshade operation (also called shaded relief) on an input digital elevation model (DEM). The user must specify the name of the input DEM and the output hillshade image name. Other parameters that must be specified include the illumination source azimuth (--azimuth), or sun direction (0-360 degrees), the illumination source altitude (--altitude; i.e. the elevation of the sun above the horizon, measured as an angle from 0 to 90 degrees) and the Z conversion factor (--zfactor). The Z conversion factor is only important when the vertical and horizontal units are not the same in the DEM, and the DEM is in a projected coordinate system. When this is the case, the algorithm will multiply each elevation in the DEM by the Z conversion factor. If the DEM is in the geographic coordinate system (latitude and longitude), the following equation is used:

zfactor = 1.0 / (111320.0 x cos(mid_lat))

where mid_lat is the latitude of the centre of the raster, in radians.

The hillshade value (HS) of a DEM grid cell is calculate as:

HS = tan(s) / [1 - tan(s)²]^0.5 x [sin(Alt) / tan(s) - cos(Alt) x sin(Az - a)]

where s and a are the local slope gradient and aspect (orientation) respectively and Alt and Az are the illumination source altitude and azimuth respectively. Slope and aspect are calculated using Horn's (1981) 3rd-order finate difference method.

Reference:

Gallant, J. C., and J. P. Wilson, 2000, Primary topographic attributes, in Terrain Analysis: Principles and Applications, edited by J. P. Wilson and J. C. Gallant pp. 51-86, John Wiley, Hoboken, N.J.

See Also: HypsometricallyTintedHillshade, MultidirectionalHillshade, Aspect, Slope

Parameters:

Python function:

wbt.hillshade(
    dem, 
    output, 
    azimuth=315.0, 
    altitude=30.0, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=Hillshade -v --wd="/path/to/data/" ^
-i=DEM.tif -o=output.tif --azimuth=315.0 --altitude=30.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 03/09/2020

HorizonAngle

This tool calculates the horizon angle (Sx), i.e. the maximum slope along a specified azimuth (0-360 degrees) for each grid cell in an input digital elevation model (DEM). Horizon angle is sometime referred to as the maximum upwind slope in wind exposure/sheltering studies. Positive values can be considered sheltered with respect to the azimuth and negative values are exposed. Thus, Sx is a measure of exposure to a wind from a specific direction. The algorithm works by tracing a ray from each grid cell in the direction of interest and evaluating the slope for each location in which the DEM grid is intersected by the ray. Linear interpolation is used to estimate the elevation of the surface where a ray does not intersect the DEM grid precisely at one of its nodes.

The user is able to constrain the maximum search distance (--max_dist) for the ray tracing by entering a valid maximum search distance value (in the same units as the X-Y coordinates of the input raster DEM). If the maximum search distance is left blank, each ray will be traced to the edge of the DEM, which will add to the computational time.

Maximum upwind slope should not be calculated for very extensive areas over which the Earth's curvature must be taken into account. Also, this index does not take into account the deflection of wind by topography. However, averaging the horizon angle over a window of directions can yield a more robust measure of exposure, compensating for the deflection of wind from its regional average by the topography. For example, if you are interested in measuring the exposure of a landscape to a northerly wind, you could perform the following calculation:

Sx(N) = [Sx(345)+Sx(350)+Sx(355)+Sx(0)+Sx(5)+Sx(10)+Sx(15)] / 7.0

Ray-tracing is a highly computationally intensive task and therefore this tool may take considerable time to operate for larger sized DEMs. Maximum upwind slope is best displayed using a Grey scale palette that is inverted.

Horizon angle is best visualized using a white-to-black palette and rescaled from approximately -10 to 70 (see below for an example of horizon angle calculated at a 150-degree azimuth).

See Also: TimeInDaylight

Parameters:

Python function:

wbt.horizon_angle(
    dem, 
    output, 
    azimuth=0.0, 
    max_dist=100.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=HorizonAngle -v --wd="/path/to/data/" ^
-i='input.tif' -o=output.tif --azimuth=315.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 07/07/2017

Last Modified: 03/09/2020

HorizonArea

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates horizon area, i.e., the area of the horizon polygon centered on each point in an input digital elevation model (DEM). Horizon area is therefore conceptually related to the viewhed and visibility_index functions. Horizon area can be thought of as an approximation of the viewshed area and is therefore faster to calculate a spatial distribution of compared with the visibility index. Horizon area is measured in hectares.

The user must specify an input DEM (dem), the azimuth fraction (az_fraction), the maximum search distance (max_dist), and the height offset of the observer (observer_hgt_offset). The input DEM should usually be a digital surface model (DSM) that contains significant off-terrain objects. Such a model, for example, could be created using the first-return points of a LiDAR data set, or using the lidar_digital_surface_model tool. The azimuth fraction should be an even divisor of 360-degrees and must be between 1-45 degrees.

The tool operates by calculating horizon angle (see horizon_angle) rasters from the DSM based on the user-specified azimuth fraction (az_fraction). For example, if an azimuth fraction of 15-degrees is specified, horizon angle rasters would be calculated for the solar azimuths 0, 15, 30, 45... A horizon angle raster evaluates the vertical angle between each grid cell in a DSM and a distant obstacle (e.g. a mountain ridge, building, tree, etc.) that obscures the view in a specified direction. In calculating horizon angle, the user must specify the maximum search distance (max_dist), in map units, beyond which the query for higher, more distant objects will cease. This parameter strongly impacts the performance of the function, with larger values resulting in significantly longer processing-times.

With each evaluated direction, the coordinates of the horizon point is determined, using the azimuth and the distance to horizon, with each point then serving as a vertex in a horizon polygon. The shoelace algorithm is used to measure the area of each horizon polgon for the set of grid cells, which is then reported in the output raster.

The observer_hgt_offset parameter can be used to add an increment to the source cell's elevation. For example, the following image shows the spatial pattern derived from a LiDAR DSM using observer_hgt_offset = 0.0:

Notice that there are several places, plarticularly on the flatter rooftops, where the local noise in the LiDAR DEM, associated with the individual scan lines, has resulted in a noisy pattern in the output. By adding a small height offset of the scale of this noise variation (0.15 m), we see that most of this noisy pattern is removed in the output below:

As another example, in the image below, the observer_hgt_offset parameter has been used to measure the pattern of the index at a typical human height (1.7 m):

Notice how at this height a much larger area becomes visible on some of the flat rooftops where a guard wall prevents further viewing areas at shorter observer heights.

See Also: SkyViewFactor, AverageHorizonDistance, SkylineAnalysis, Openness, TimeInDaylight, HorizonAngle

Parameters:

Python function:

wbt.horizon_area(
    dem, 
    output, 
    az_fraction=5.0, 
    max_dist=9999.0, 
    observer_hgt_offset=0.05, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=HorizonArea -i=dem.tif -o=output.tif ^
--az_fraction=5.0 --max_dist=250.0 --observer_hgt_offset=1.75 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 09/04/2024

Last Modified: 09/04/2024

HorizontalExcessCurvature

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the horizontal excess curvature from a digital elevation model (DEM). Horizontal excess curvature is the difference of tangential (horizontal) and minimal curvatures at a location (Shary, 1995). This variable has positive values, zero or greater. Florinsky (2017) states that horizontal excess curvature measures the extent to which the bending of a normal section tangential to a contour line is larger than the minimal bending at a given point of the surface. Horizontal excess curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary PA (1995) Land surface in gravity points classification by a complete system of curvatures. Mathematical Geology 27: 373–390.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

TangentialCurvature, ProfileCurvature, MinimalCurvature, MaximalCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.horizontal_excess_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=HorizontalExcessCurvature -i=DEM.tif ^
-o=output.tif --log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 11/01/2022

Last Modified: 28/11/2022

HypsometricAnalysis

This tool can be used to derive the hypsometric curve, or area-altitude curve, of one or more input digital elevation models (DEMs) (--inputs). A hypsometric curve is a histogram or cumulative distribution function of elevations in a geographical area.

See Also: SlopeVsElevationPlot

Parameters:

Python function:

wbt.hypsometric_analysis(
    inputs, 
    output, 
    watershed=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=HypsometricAnalysis -v ^
--wd="/path/to/data/" -i="DEM1.tif;DEM2.tif" ^
--watershed="ws1.tif;ws2.tif" -o=outfile.html 

Source code on GitHub

Author: Dr. John Lindsay

Created: 30/01/2018

Last Modified: 12/10/2018

HypsometricallyTintedHillshade

This tool creates a hypsometrically tinted shaded relief (Swiss hillshading) image from an input digital elevation model (DEM). The tool combines a colourized version of the DEM with varying illumination provided by a hillshade image, to produce a composite relief model that can be used to visual topography for more effective interpretation of landscapes. The output (--output) of the tool is a 24-bit red-green-blue (RGB) colour image.

The user must specify the name of the input DEM and the output image name. Other parameters that must be specified include the illumination source azimuth (--azimuth), or sun direction (0-360 degrees), the illumination source altitude (--altitude; i.e. the elevation of the sun above the horizon, measured as an angle from 0 to 90 degrees), the hillshade weight (--hs_weight; 0-1), image brightness (--brightness; 0-1), and atmospheric effects (--atmospheric; 0-1). The hillshade weight can be used to increase or subdue the relative prevalence of the hillshading effect in the output image. The image brightness parameter is used to create an overall brighter or darker version of the terrain rendering; note however, that very high values may over-saturate the well-illuminated portions of the terrain. The atmospheric effects parameter can be used to introduce a haze or atmosphere effect to the output image. It is intended to reproduce the effect of viewing mountain valley bottoms through a thicker and more dense atmosphere. Values greater than zero will introduce a slightly blue tint, particularly at lower altitudes, blur the hillshade edges slightly, and create a random haze-like speckle in lower areas. The user must also specify the Z conversion factor (--zfactor). The Z conversion factor is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z conversion factor. If the DEM is in the geographic coordinate system (latitude and longitude), the following equation is used:

zfactor = 1.0 / (111320.0 x cos(mid_lat))

where mid_lat is the latitude of the centre of the raster, in radians.

See Also: Hillshade, MultidirectionalHillshade, Aspect, Slope

Parameters:

Python function:

wbt.hypsometrically_tinted_hillshade(
    dem, 
    output, 
    altitude=45.0, 
    hs_weight=0.5, 
    brightness=0.5, 
    atmospheric=0.0, 
    palette="atlas", 
    reverse=False, 
    zfactor=None, 
    full_mode=False, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=HypsometricallyTintedHillshade -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif --altitude=45.0 ^
--hs_weight=0.3 --brightness=0.6 --atmospheric=0.2 ^
--palette=arid --full_mode 

Source code on GitHub

Author: Dr. John Lindsay

Created: 09/07/2020

Last Modified: 03/09/2020

LocalHypsometricAnalysis

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the hypsometric integral from the elevation distribution contained within the local neighbourhood surrounding each grid cell in an input (--input) DEM. The hyspometric integral (HI) is the area under the hypsometric curve, which is a plot that relates elevation and area. This plot is a cumulative distribution function with elevation expressed as a proportion of maximum elevation and area expressed as the proportion of the area above. Hypsometry, or area-altitude analysis, is commonly used by geomorphologists and geologists to characterize the erosional history of drainage basins. The HI, ranging between 0 and 1, expresses the volume of land that lies above the lowest point within an area, and thus has not been eroded. Relatively low HI values are indicative of more strongly eroded surfaces.

Some researchers (e.g. Pérez‐Peña et al., 2009) have demonstrated the usefulness of applying hypsometry in a spatially distributed fashion, rather than aggregated by basins, as it is typically applied. While Pérez‐Peña et al. (2009) characterized spatial distributions of HI using coarse grids overlayed overtop a digital elevation model (DEM), this tool uses a filter-based approach instead. Each grid cell in the input DEM (--input) has an individual HI calculation based on the elevation distribution within a moving kernel. HI values are calculated using the elevation-relief ratio method described by Pike and Wilson (1971).

In actuality, the tool uses a multi-scale approach, much like many of the other tools within the Geomorphometric Analysis toolbox (e.g. MaxElevationDeviation, MultiscaleStdDevNormals), such that the neighbourhood size is varied according to a range defined by user-specified input parameters. The HI that is reported within each grid cell in the output raster is the minimum HI value measured for each of the tested scales, defined by lower (r_L) and upper (r_U) ranges.

HI_min=min{HI(r):r=r_L...r_U},

In this way, it represents a heterogenous, locally scale optimized map of HI distributions. A nonlinear scale sampling interval is used by this tool to ensure that the scale sampling density is higher for short scale ranges, where there is often greater variability in HI values, and coarser at longer tested scales, such that:

r_i = r_L + [step × (i - r_L)]^p

Where ri is the filter radius for step i and p is the nonlinear scaling factor (--step_nonlinearity) and a step size (--step) of step.

There are two outputs generated from this tool, including the HI_min raster (--out_mag) and the r_min scale raster (--out_scale).

References:

Pérez‐Peña, J. V., Azañón, J. M., Booth‐Rea, G., Azor, A., and Delgado, J. (2009). Differentiating geology and tectonics using a spatial autocorrelation technique for the hypsometric integral. Journal of Geophysical Research: Earth Surface, 114(F2).

Pike, R. J., and Wilson, S. E. (1971). Elevation-relief ratio, hypsometric integral, and geomorphic area-altitude analysis. Geological Society of America Bulletin, 82(4), 1079-1084.

See Also: HypsometricAnalysis, MaxElevationDeviation, MultiscaleStdDevNormals

Parameters:

Python function:

wbt.local_hypsometric_analysis(
    i, 
    out_mag, 
    out_scale, 
    min_scale=4, 
    step=1, 
    num_steps=10, 
    step_nonlinearity=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=LocalHypsometricAnalysis ^
--input=DEM.tif --out_mag=HI_magnitude.tif ^
--out_scale=HI_scale.tif --min_scale=4 --step=1 --num_steps=50 ^
--step_nonlinearity=1.1 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 07/11/2021

Last Modified: 05/12/2021

LocalQuadraticRegression

This tool is an implementation of the constrained quadratic regression algorithm using a flexible window size described in Wood (1996). A quadratic surface is fit to local areas of input DEM (--dem), defined by a filter size (--filter) using least squares regression. Note that the model is constrained such that it must pass through the cell at the center of the filter. This is accomplished by representing all elevations relative to the center cell, and by making the equation constant 0.

Surface derivatives are calculated from the coefficients of the local quadratic surface once they are known. These include: Slope, Aspect, Profile convexity, Plan convexity, Longitudinal curvature, Cross-sectional curvature, and Minimum profile convexity, all as defined in Wood (1996). The goodness-of-fit (r-squared) of each local quadratic model is also returned.

Due to the fact that large filter sizes require long processing times, and that fitting the surface is the most time consuming part of the algorithm, all LSPs are output every time this tool is run. The content of each output is described by the suffixes of the output file names.

Reference:

Wood, J. (1996). The Geomorphological Characterisation of Digital Elevation Models. University of Leicester.

See Also: Aspect, Slope, PlanCurvature, ProfileCurvature

Parameters:

Python function:

wbt.local_quadratic_regression(
    dem, 
    output, 
    filter=3, 
    callback=default_callback
)

Command-line Interface:

-./whitebox_tools -r=LocalQuadraticRegression --dem=DEM.tif ^
--output=out_ras.tif --filter=15 

Source code on GitHub

Author: Daniel Newman

Created: 22/06/2020

Last Modified: 22/06/2020

MapOffTerrainObjects

This tool can be used to map off-terrain objects in a digital surface model (DSM) based on cell-to-cell differences in elevations and local slopes. The algorithm works by using a region-growing operation to connect neighbouring grid cells outwards from seed cells. Two neighbouring cells are considered connected if the slope between the two cells is less than the user-specified maximum slope value (--max_slope). Mapped segments that are less than the minimum feature size (--min_size), in grid cells, are assigned a common background value. Note that this method of mapping off-terrain objects, and thereby separating ground cells from non-ground objects in DSMs, works best with fine-resolution DSMs that have been interpolated using a non-smoothing method, such as triangulation (TINing) or nearest-neighbour interpolation.

See Also: RemoveOffTerrainObjects

Parameters:

Python function:

wbt.map_off_terrain_objects(
    dem, 
    output, 
    max_slope=40.0, 
    min_size=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MapOffTerrainObjects -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif ^
--max_diff=1.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/07/2020

Last Modified: 27/07/2020

MaxAnisotropyDev

Calculates the maximum anisotropy (directionality) in elevation deviation over a range of spatial scales.

Parameters:

Python function:

wbt.max_anisotropy_dev(
    dem, 
    out_mag, 
    out_scale, 
    max_scale, 
    min_scale=3, 
    step=2, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxAnisotropyDev -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=DEVmax_mag.tif ^
--out_scale=DEVmax_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dan Newman and John Lindsay

Created: January 26, 2018

Last Modified: 12/10/2018

MaxAnisotropyDevSignature

Calculates the anisotropy in deviation from mean for points over a range of spatial scales.

Parameters:

Python function:

wbt.max_anisotropy_dev_signature(
    dem, 
    points, 
    output, 
    max_scale, 
    min_scale=1, 
    step=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxAnisotropyDevSignature -v ^
--wd="/path/to/data/" --dem=DEM.tif --points=sites.shp ^
--output=roughness.html --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dan Newman and John Lindsay

Created: 27/03/2018

Last Modified: 12/10/2018

MaxBranchLength

Maximum branch length (Bmax) is the longest branch length between a grid cell's flowpath and the flowpaths initiated at each of its neighbours. It can be conceptualized as the downslope distance that a volume of water that is split into two portions by a drainage divide would travel before reuniting.

If the two flowpaths of neighbouring grid cells do not intersect, Bmax is simply the flowpath length from the starting cell to its terminus at the edge of the grid or a cell with undefined flow direction (i.e. a pit cell either in a topographic depression or at the edge of a major body of water).

The pattern of Bmax derived from a DEM should be familiar to anyone who has interpreted upslope contributing area images. In fact, Bmax can be thought of as the complement of upslope contributing area. Whereas contributing area is greatest along valley bottoms and lowest at drainage divides, Bmax is greatest at divides and lowest along channels. The two topographic attributes are also distinguished by their units of measurements; Bmax is a length rather than an area. The presence of a major drainage divide between neighbouring grid cells is apparent in a Bmax image as a linear feature, often two grid cells wide, of relatively high values. This property makes Bmax a useful land surface parameter for mapping ridges and divides.

Bmax is useful in the study of landscape structure, particularly with respect to drainage patterns. The index gives the relative significance of a specific location along a divide, with respect to the dispersion of materials across the landscape, in much the same way that stream ordering can be used to assess stream size.

See Also: FlowLengthDiff

Reference:

Lindsay JB, Seibert J. 2013. Measuring the significance of a divide to local drainage patterns. International Journal of Geographical Information Science, 27: 1453-1468. DOI: 10.1080/13658816.2012.705289

Parameters:

Python function:

wbt.max_branch_length(
    dem, 
    output, 
    log=False, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxBranchLength -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 09/07/2017

Last Modified: 18/10/2019

MaxDifferenceFromMean

Calculates the maximum difference from mean elevation over a range of spatial scales.

Parameters:

Python function:

wbt.max_difference_from_mean(
    dem, 
    out_mag, 
    out_scale, 
    min_scale, 
    max_scale, 
    step=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxDifferenceFromMean -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=DEVmax_mag.tif ^
--out_scale=DEVmax_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 28/08/2018

Last Modified: 12/10/2018

MaxDownslopeElevChange

This tool calculates the maximum elevation drop between each grid cell and its neighbouring cells within a digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output (--output) raster name.

See Also: MaxUpslopeElevChange, MinDownslopeElevChange, NumDownslopeNeighbours

Parameters:

Python function:

wbt.max_downslope_elev_change(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxDownslopeElevChange -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=out.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 11/07/2017

Last Modified: 12/10/2018

MaxElevDevSignature

Calculates the maximum elevation deviation over a range of spatial scales and for a set of points.

Parameters:

Python function:

wbt.max_elev_dev_signature(
    dem, 
    points, 
    output, 
    min_scale, 
    max_scale, 
    step=10, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxElevDevSignature -v ^
--wd="/path/to/data/" --dem=DEM.tif --points=sites.tif ^
--output=topo_position.html --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: March 1, 2018

Last Modified: 12/10/2018

MaxElevationDeviation

This tool can be used to calculate the maximum deviation from mean elevation, DEVmax (Lindsay et al. 2015) for each grid cell in a digital elevation model (DEM) across a range specified spatial scales. DEV is an elevation residual index and is essentially equivalent to a local elevation z-score. This attribute measures the relative topographic position as a fraction of local relief, and so is normalized to the local surface roughness. The multi-scaled calculation of DEVmax utilizes an integral image approach (Crow, 1984) to ensure highly efficient filtering that is invariant with filter size, which is the algorithm characteristic that allows for this densely sampled multi-scale analysis. In this way, MaxElevationDeviation allows users to estimate the locally optimal scale with which to estimate DEV on a pixel-by-pixel basis. This multi-scaled version of local topographic position can reveal significant terrain characteristics and can aid with soil, vegetation, landform, and other mapping applications that depend on geomorphometric characterization.

The user must specify the name (--dem) of the input digital elevation model (DEM). The range of scales that are evaluated in calculating DEVmax are determined by the user-specified --min_scale, --max_scale, and --step parameters. All filter radii between the minimum and maximum scales, increasing by step, will be evaluated. The scale parameters are in units of grid cells and specify kernel size "radii" (r), such that:

d = 2r + 1

That is, a radii of 1, 2, 3... yields a square filters of dimension (d) 3 x 3, 5 x 5, 7 x 7...

DEV is estimated at each tested filter size and every grid cell is assigned the maximum DEV value across the evaluated scales.

The user must specify the file names of two output rasters, including the magnitude (DEVmax) and a second raster the assigns each pixel the scale at which DEVmax is encountered (DEVscale). The DEVscale raster can be very useful for revealing multi-scale landscape structure.

Reference:

Lindsay J, Cockburn J, Russell H. 2015. An integral image approach to performing multi-scale topographic position analysis. Geomorphology, 245: 51-61.

See Also: DevFromMeanElev, MaxDifferenceFromMean, MultiscaleElevationPercentile

Parameters:

Python function:

wbt.max_elevation_deviation(
    dem, 
    out_mag, 
    out_scale, 
    min_scale, 
    max_scale, 
    step=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxElevationDeviation -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=DEVmax_mag.tif ^
--out_scale=DEVmax_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: July 20, 2017

Last Modified: 12/10/2018

MaxUpslopeElevChange

This tool calculates the maximum elevation difference in the upslope direction between each grid cell and its neighbouring cells within a digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output (--output) raster name.

See Also: MaxDownslopeElevChange

Parameters:

Python function:

wbt.max_upslope_elev_change(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaxUpslopeElevChange -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=out.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/01/2022

Last Modified: 27/01/2022

MaximalCurvature

This tool calculates the maximal curvature from a digital elevation model (DEM). Maximal curvature is the curvature of a principal section with the highest value of curvature at a given point of the topographic surface (Florinsky, 2017). The values of this curvature are unbounded, and positive values correspond to ridge positions while negative values are indicative of closed depressions (Florinsky, 2016). Maximal curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

MinimalCurvature, TangentialCurvature, ProfileCurvature, PlanCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.maximal_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MaximalCurvature -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/01/2022

Last Modified: 15/01/2022

MeanCurvature

This tool calculates the mean curvature from a digital elevation model (DEM). Mean curvature is the average of any mutually orthogonal normal sections, such as profile and tangential curvature (Wilson, 2018). This variable has an unbounded range that can take either positive or negative values. Florinsky (2017) states that mean curvature represents the two accumulation mechanisms of gravity-driven substances, convergence and relative deceleration of flows, with equal weights. Mean curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

Wilson, J. P. (2018). Environmental applications of digital terrain modeling. John Wiley & Sons.

TangentialCurvature, ProfileCurvature, PlanCurvature, GaussianCurvature, MinimalCurvature, MaximalCurvature

Parameters:

Python function:

wbt.mean_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MeanCurvature -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/11/2021

Last Modified: 12/01/2022

MinDownslopeElevChange

This tool calculates the minimum elevation drop between each grid cell and its neighbouring cells within a digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output (--output) raster name.

See Also: MaxDownslopeElevChange, NumDownslopeNeighbours

Parameters:

Python function:

wbt.min_downslope_elev_change(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MinDownslopeElevChange -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=out.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 11/07/2017

Last Modified: 12/10/2018

MinimalCurvature

This tool calculates the minimal curvature from a digital elevation model (DEM). Minimal curvature is the curvature of a principal section with the lowest value of curvature at a given point of the topographic surface (Florinsky, 2017). The values of this curvature are unbounded, and positive values correspond to hills while negative values are indicative of valley positions (Florinsky, 2016). Minimal curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

MaximalCurvature, TangentialCurvature, ProfileCurvature, PlanCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.minimal_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MinimalCurvature -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/01/2022

Last Modified: 15/01/2022

MultidirectionalHillshade

This tool performs a hillshade operation (also called shaded relief) on an input digital elevation model (DEM) with multiple sources of illumination. The user must specify the name of the input DEM (--dem) and the output hillshade image name (--output). Other parameters that must be specified include the altitude of the illumination sources (--altitude; i.e. the elevation of the sun above the horizon, measured as an angle from 0 to 90 degrees) and the Z conversion factor (--zfactor). The Z conversion factor is only important when the vertical and horizontal units are not the same in the DEM, and the DEM is in a projected coordinate system. When this is the case, the algorithm will multiply each elevation in the DEM by the Z conversion factor. If the DEM is in the geographic coordinate system (latitude and longitude), the following equation is used:

zfactor = 1.0 / (111320.0 x cos(mid_lat))

where mid_lat is the latitude of the centre of the raster, in radians. The Z conversion factor can also be used used to apply a vertical exageration to further emphasize landforms within the hillshade output.

The hillshade value (HS) of a DEM grid cell is calculate as:

HS = tan(s) / [1 - tan(s)²]^0.5 x [sin(Alt) / tan(s) - cos(Alt) x sin(Az - a)]

where s and a are the local slope gradient and aspect (orientation) respectively and Alt and Az are the illumination source altitude and azimuth respectively. Slope and aspect are calculated using Horn's (1981) 3rd-order finate difference method.

Lastly, the user must specify whether or not to use full 360-degrees of illumination sources (--full_mode). When this flag is not specified, the tool will perform a weighted summation of the hillshade images from four illumination azimuth positions at 225, 270, 315, and 360 (0) degrees, given weights of 0.1, 0.4, 0.4, and 0.1 respectively. When run in the full 360-degree mode, eight illumination source azimuths are used to calculate the output at 0, 45, 90, 135, 180, 225, 270, and 315 degrees, with weights of 0.15, 0.125, 0.1, 0.05, 0.1, 0.125, 0.15, and 0.2 respectively.

Classic hillshade (Azimuth=315, Altitude=45.0)

Multi-directional hillshade (Altitude=45.0, Four-direction mode)

Multi-directional hillshade (Altitude=45.0, 360-degree mode)

See Also: Hillshade, HypsometricallyTintedHillshade, Aspect, Slope

Parameters:

Python function:

wbt.multidirectional_hillshade(
    dem, 
    output, 
    altitude=45.0, 
    zfactor=None, 
    full_mode=False, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultidirectionalHillshade -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif ^
--altitude=30.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 19/07/2020

Last Modified: 03/09/2020

MultiscaleCurvatures

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates several multiscale curvatures and curvature-based indices from an input DEM (--dem). There are 18 curvature types (--curv_type) available, including: accumulation curvature, curvedness, difference curvature, Gaussian curvature, generating function, horizontal excess curvature, maximal curvature, mean curvature, minimal curvature, plan curvature, profile curvature, ring curvature, rotor, shape index, tangential curvature, total curvature, unsphericity, and vertical excess curvature. Each of these curvatures can be measured in non-multiscale fashion using the corresponding tools available in either the WhiteboxTools open-core or the Whitebox extension.

Like many of the multi-scale land-surface parameter tools available in Whitebox, this tool can be run in two different modes: it can either be used to measure curvature at a single specific scale or to generate a curvature scale mosaic. To understand the difference between these two modes, we must first understand how curvatures are measured and how the non-multiscale curvature tools (e.g. ProfileCurvature) work. Curvatures are generally measured by fitting a mathematically defined surface to the elevation values within the local neighbourhood surrounding each grid cell in a DEM. The Whitebox curvature tools use the algorithms described Florinsky (2016), which use the 25 elevations within a 5 x 5 local neighbouhood for projected DEMs, and the nine elevations within a 3 x 3 neighbourhood for DEMs in geographic coordinate systems. This is what determines the scale at which these land-surface parameters are calculated. Because they are calculated using small local neighbourhoods (kernels), then these algorithms are heavily impacted by micro-topographic roughness and DEM noise. For example, in a fine-resolution DEM containing a great deal of micro-topographic roughness, the measured curvature value will be dominated by topographic variation at the scale of the roughness rather than the hillslopes on which that roughness is superimposed. This mis-matched scaling can be a problem in many applications, e.g. in landform classification and slope failure modelling applications.

Using the MultiscaleCurvatures tool, the user can specify a certain desired scale, larger than that defined by the grid resolution and kernel size, over which a curvature should be characterized. The tool will then use a fast Gaussian scale-space method to remove the topographic variation in the DEM at scales less than the desired scale, and will then characterize the curvature using the usual method based on this scaled DEM. To measure curvature at a single non-local scale, the user must specify a minimum search neighbourhood radius in grid cells (--min_scale) greater than 0.0. Note that a minimum search neighbourhood of 0.0 will replicate the non-multiscale equivalent curvature tool and any --min_scale value > 0.0 will apply the Gassian scale space method to eliminate topographic variation less than the scale of the minimum search neighbourhood. The base step size (--step), number of steps (--num_steps), and step nonlinearity (--step_nonlinearity) parameters should all be left to their default values of 1 in this case. The output curvature raster will be written to the output magnitude file (--out_mag). The following animation shows several multiscale curvature rasters (tangential curvature) measured from a DEM across a range of spatial scales.

Alternatively, one can use this tool to create a curvature scale mosaic. In this case, the user specifies a range of spatial scales (i.e., a scale space) over which to measure curvature. The curvature scale-space is densely sampled and each grid cell is assigned the maximum absolute curvature value (for the specified curvature type) across the scale space. In this scale-mosaic mode, the user must also specify the output scale file name (--out_scale), which is an output raster that, for each grid cell, specifies the scale at which the maximum absolute curvature was identified. The following is an example of a scale mosaic of unsphericity for an area in Pole Canyon, Utah (min_scale=1.0, step=1, num_steps=50, step_nonlinearity=1.0).

Scale mosaics are useful when modelling spatial distributions of land-surface parameters, like curvatures, in complex and heterogeneous landscapes that contain an abundance of topographic variation (micro-topography, landforms, etc.) at widely varying spatial scales, often associated with different geomorphic processes. Notice how in the image above, relatively strong curvature values are being characterized for both the landforms associated with the smaller-scale mass-movement processes as well as the broader-scale fluvial erosion (i.e. valley incision and hillslopes). It would be difficult, or impossible, to achieve this effect using a single, uniform scale. Each location in a land-surface parameter scale mosaic represents the parameter measured at a characteristic scale, given the unique topography of the site and surroundings.

The properties of the sampled scale space are determined using the --min_scale, --step, --num_steps (greater than 1), and --step_nonlinearity parameters. Experience with multiscale curvature scales spaces has shown that they are more highly variable at shorter spatial scales and change more gradually at broader scales. Therefore, a nonlinear scale sampling interval is used by this tool to ensure that the scale sampling density is higher for short scale ranges and coarser at longer tested scales, such that:

r_i = r_L + [step × (i - r_L)]^p

Where ri is the filter radius for step i and p is the nonlinear scaling factor (--step_nonlinearity) and a step size (--step) of step.

In scale-mosaic mode, the user must also decide whether or not to standardize the curvature values (--standardize). When this parameter is used, the algorithm will convert each curvature raster associated with each sampled region of scale-space to z-scores (i.e. differenced from the raster-wide mean and divided by the raster-wide standard deviation). It it usually the case that curvature values measured at broader spatial scales will on the whole become less strongly valued. Because the scale mosaic algorithm used in this tool assigns each grid cell the maximum absolute curvature observed within sampled scale-space, this implies that the curvature values associated with more local-scale ranges are more likely to be selected for the final scale-mosaic raster. By standardizing each scaled curvature raster, there is greater opportunity for the final scale-mosaic to represent broader scale topographic variation. Whether or not this is appropriate will depend on the application. However, it is important to stress that the sampled scale-space need not span the full range of possible scales, from the finest scale determined by the grid resolution up to the broadest scale possible, determined by the spatial extent of the input DEM. Often, a better approach is to use this tool to create multiple scale mosaics spanning this range, thereby capturing variation within broadly defined scale ranges. For example, one could create a local-scale, meso-scale, and broad-scale curvature scale mosaics, each of which would capture topographic variation and landforms that are present in the landscape and reflective of processing operating at vastly different spatial scales. When this approach is used, it may not be necessary to standardize each scaled curvature raster, since the gradual decline in curvature values as scales increase is less pronounced within each of these broad scale ranges than across the entirety of possible scale-space. Again, however, this will depend on the application and on the characteristics of the landscape at study.

Raw curvedness values are often challenging to visualize given their range and magnitude, and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

See Also: GaussianScaleSpace, AccumulationCurvature, Curvedness, DifferenceCurvature, GaussianCurvature, GeneratingFunction, HorizontalExcessCurvature, MaximalCurvature, MeanCurvature, MinimalCurvature, PlanCurvature, ProfileCurvature, RingCurvature, Rotor, ShapeIndex, TangentialCurvature, TotalCurvature, Unsphericity, VerticalExcessCurvature

Parameters:

Python function:

wbt.multiscale_curvatures(
    dem, 
    out_mag, 
    curv_type="ProfileCurv", 
    out_scale=None, 
    min_scale=0, 
    step=1, 
    num_steps=1, 
    step_nonlinearity=1.0, 
    log=True, 
    standardize=False, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=MultiscaleCurvatures --dem=dem.tif ^
--curv_type='Unsphericity' --out_mag=curv.tif --min_scale=10.0 ^
--log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 28/11/2022

Last Modified: 28/11/2022

MultiscaleElevationPercentile

This tool calculates the most elevation percentile (EP) across a range of spatial scales. EP is a measure of local topographic position (LTP) and expresses the vertical position for a digital elevation model (DEM) grid cell (z₀) as the percentile of the elevation distribution within the filter window, such that:

EP = count_i∈C(z_i > z₀) x (100 / n_C)

where z₀ is the elevation of the window's center grid cell, z_i is the elevation of cell i contained within the neighboring set C, and n_C is the number of grid cells contained within the window.

EP is unsigned and expressed as a percentage, bound between 0% and 100%. This tool outputs two rasters, the multiscale EP magnitude (--out_mag) and the scale at which the most extreme EP value occurs (--out_scale). The magnitude raster is the most extreme EP value (i.e. the furthest from 50%) for each grid cell encountered within the tested scales of EP.

Quantile-based estimates (e.g., the median and interquartile range) are often used in nonparametric statistics to provide data variability estimates without assuming the distribution is normal. Thus, EP is largely unaffected by irregularly shaped elevation frequency distributions or by outliers in the DEM, resulting in a highly robust metric of LTP. In fact, elevation distributions within small to medium sized neighborhoods often exhibit skewed, multimodal, and non-Gaussian distributions, where the occurrence of elevation errors can often result in distribution outliers. Thus, based on these statistical characteristics, EP is considered one of the most robust representation of LTP.

The algorithm implemented by this tool uses the relatively efficient running-histogram filtering algorithm of Huang et al. (1979). Because most DEMs contain floating point data, elevation values must be rounded to be binned. The --sig_digits parameter is used to determine the level of precision preserved during this binning process. The algorithm is parallelized to further aid with computational efficiency.

Experience with multiscale EP has shown that it is highly variable at shorter scales and changes more gradually at broader scales. Therefore, a nonlinear scale sampling interval is used by this tool to ensure that the scale sampling density is higher for short scale ranges and coarser at longer tested scales, such that:

r_i = r_L + [step × (i - r_L)]^p

Where ri is the filter radius for step i and p is the nonlinear scaling factor (--step_nonlinearity) and a step size (--step) of step.

References:

Newman, D. R., Lindsay, J. B., and Cockburn, J. M. H. (2018). Evaluating metrics of local topographic position for multiscale geomorphometric analysis. Geomorphology, 312, 40-50.

Huang, T., Yang, G.J.T.G.Y. and Tang, G., 1979. A fast two-dimensional median filtering algorithm. IEEE Transactions on Acoustics, Speech, and Signal Processing, 27(1), pp.13-18.

See Also: ElevationPercentile, MaxElevationDeviation, MaxDifferenceFromMean

Parameters:

Python function:

wbt.multiscale_elevation_percentile(
    dem, 
    out_mag, 
    out_scale, 
    sig_digits=3, 
    min_scale=4, 
    step=1, 
    num_steps=10, 
    step_nonlinearity=1.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleElevationPercentile -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=roughness_mag.tif ^
--out_scale=roughness_scale.tif --min_scale=1 --step=5 ^
--num_steps=100 --step_nonlinearity=1.5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/12/2019

Last Modified: 22/12/2019

MultiscaleRoughness

Calculates surface roughness over a range of spatial scales.

Parameters:

Python function:

wbt.multiscale_roughness(
    dem, 
    out_mag, 
    out_scale, 
    max_scale, 
    min_scale=1, 
    step=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleRoughness -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=roughness_mag.tif ^
--out_scale=roughness_scale.tif --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 26/02/2018

Last Modified: 03/09/2020

MultiscaleRoughnessSignature

Calculates the surface roughness for points over a range of spatial scales.

Parameters:

Python function:

wbt.multiscale_roughness_signature(
    dem, 
    points, 
    output, 
    max_scale, 
    min_scale=1, 
    step=1, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleRoughnessSignature -v ^
--wd="/path/to/data/" --dem=DEM.tif --points=sites.shp ^
--output=roughness.html --min_scale=1 --max_scale=1000 ^
--step=5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 27/02/2018

Last Modified: 03/09/2020

MultiscaleStdDevNormals

This tool can be used to map the spatial pattern of maximum spherical standard deviation (σ_{s max}; --out_mag), as well as the scale at which maximum spherical standard deviation occurs (r_max; --out_scale), for each grid cell in an input DEM (--dem). This serves as a multi-scale measure of surface roughness, or topographic complexity. The spherical standard deviation (σ_s) is a measure of the angular spread among n unit vectors and is defined as:

σ_s = √[-2ln(R / N)] × 180 / π

Where R is the resultant vector length and is derived from the sum of the x, y, and z components of each of the n normals contained within a filter kernel, which designates a tested spatial scale. Each unit vector is a 3-dimensional measure of the surface orientation and slope at each grid cell center. The maximum spherical standard deviation is:

σ_{s max}=max{σs(r):r=r_L...r_U},

Experience with roughness scale signatures has shown that σ_{s max} is highly variable at shorter scales and changes more gradually at broader scales. Therefore, a nonlinear scale sampling interval is used by this tool to ensure that the scale sampling density is higher for short scale ranges and coarser at longer tested scales, such that:

r_i = r_L + [step × (i - r_L)]^p

Where ri is the filter radius for step i and p is the nonlinear scaling factor (--step_nonlinearity) and a step size (--step) of step.

Use the SphericalStdDevOfNormals tool if you need to calculate σ_s for a single scale.

Reference:

JB Lindsay, DR Newman, and A Francioni. 2019 Scale-Optimized Surface Roughness for Topographic Analysis. Geosciences, 9(322) doi: 10.3390/geosciences9070322.

See Also: SphericalStdDevOfNormals, MultiscaleStdDevNormalsSignature, MultiscaleRoughness

Parameters:

Python function:

wbt.multiscale_std_dev_normals(
    dem, 
    out_mag, 
    out_scale, 
    min_scale=1, 
    step=1, 
    num_steps=10, 
    step_nonlinearity=1.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleStdDevNormals -v ^
--wd="/path/to/data/" --dem=DEM.tif --out_mag=roughness_mag.tif ^
--out_scale=roughness_scale.tif --min_scale=1 --step=5 ^
--num_steps=100 --step_nonlinearity=1.5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 05/06/2019

Last Modified: 03/09/2020

MultiscaleStdDevNormalsSignature

Calculates the surface roughness for points over a range of spatial scales.

Parameters:

Python function:

wbt.multiscale_std_dev_normals_signature(
    dem, 
    points, 
    output, 
    min_scale=1, 
    step=1, 
    num_steps=10, 
    step_nonlinearity=1.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleStdDevNormalsSignature -v ^
--wd="/path/to/data/" --dem=DEM.tif --points=sites.shp ^
--output=roughness.html --min_scale=1 --step=5 --num_steps=100 ^
--step_nonlinearity=1.5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 20/06/2019

Last Modified: 03/09/2020

MultiscaleTopographicPositionImage

This tool creates a multiscale topographic position (MTP) image (see here for an example) from three DEV_max rasters of differing spatial scale ranges. Specifically, MultiscaleTopographicPositionImage takes three DEV_max magnitude rasters, created using the MaxElevationDeviation tool, as inputs. The three inputs should correspond to the elevation deviations in the local (--local), meso (--meso), and broad (--broad) scale ranges and will be forced into the blue, green, and red colour components of the colour composite output (--output) raster. The image lightness value (--lightness) controls the overall brightness of the output image, as depending on the topography and scale ranges, these images can appear relatively dark. Higher values result in brighter, more colourful output images.

The user may optionally specify a hillshade image. When specified, the hillshade will be used to provide a shaded-relief overlaid on top of the coloured multi-scale information, providing a very effective visualization. Any hillshade image may be used for this purpose, but we have found that multi-directional hillshade (MultidirectionalHillshade), and specifically those derived using the 360-degree option, can be most effective for this application. However, experimentation is likely needed to find the optimal for each unique data set.

The output images can take some training to interpret correctly and a detailed explanation can be found in Lindsay et al. (2015). Sites within the landscape that occupy prominent topographic positions, either low-lying or elevated, will be apparent by their bright colouring in the MTP image. Those that are coloured more strongly in the blue are promient at the local scale range; locations that are more strongly green coloured are promient at the meso scale; and bright reds in the MTP image are associated with broad-scale landscape prominence. Of course, combination colours are also possible when topography is elevated or low-lying across multiple scale ranges. For example, a yellow area would indicated a site of prominent topographic position across the meso and broadest scale ranges.

Reference:

Lindsay J, Cockburn J, Russell H. 2015. An integral image approach to performing multi-scale topographic position analysis. Geomorphology, 245: 51-61.

See Also: MaxElevationDeviation

Parameters:

Python function:

wbt.multiscale_topographic_position_image(
    local, 
    meso, 
    broad, 
    output, 
    hillshade=None, 
    lightness=1.2, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=MultiscaleTopographicPositionImage -v ^
--wd="/path/to/data/" --local=DEV_local.tif --meso=DEV_meso.tif ^
--broad=DEV_broad.tif -o=output.tif --lightness=1.5 

Source code on GitHub

Author: Dr. John Lindsay

Created: 19/07/2017

Last Modified: 30/01/2020

NumDownslopeNeighbours

This tool calculates the number of downslope neighbours of each grid cell in a raster digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output (--output) raster name. The tool examines the eight neighbouring cells for each grid cell in a the DEM and counts the number of neighbours with an elevation less than the centre cell of the 3 x 3 window. The output image can therefore have values raning from 0 to 8. A raster grid cell with eight downslope neighbours is a peak and a cell with zero downslope neighbours is a pit. This tool can be used with the NumUpslopeNeighbours tool to assess the degree of local flow divergence/convergence.

See Also: NumUpslopeNeighbours

Parameters:

Python function:

wbt.num_downslope_neighbours(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=NumDownslopeNeighbours -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 12/10/2018

NumUpslopeNeighbours

This tool calculates the number of upslope neighbours of each grid cell in a raster digital elevation model (DEM). The user must specify the name of the input DEM (--dem) and the output (--output) raster name. The tool examines the eight neighbouring cells for each grid cell in a the DEM and counts the number of neighbours with an elevation less than the centre cell of the 3 x 3 window. The output raster can therefore have values ranging from 0 to 8, although in a DEM that has been hydrologically conditioned (i.e. depressions and flats removed), the values of the output will not exceed seven. This tool can be used with the NumDownslopeNeighbours tool to assess the degree of local flow divergence/convergence.

See Also: NumDownslopeNeighbours

Parameters:

Python function:

wbt.num_upslope_neighbours(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=NumUpslopeNeighbours -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 12/10/2018

Openness

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the Yokoyama et al. (2002) topographic openness index from an input DEM (--input). Openness has two viewer perspectives, which correspond with positive and negative openness outputs (--pos_output and --neg_output). Positive values, expressing openness above the surface, are high for convex forms, whereas negative values describe this attribute below the surface and are high for concave forms. Openness is an angular value that is an average of the horizon angle in the eight cardinal directions to a maximum search distance (--dist), measured in grid cells. Openness rasters are best visualized using a greyscale palette.

Positive Openness:

Negative Openness:

References:

Yokoyama, R., Shirasawa, M., & Pike, R. J. (2002). Visualizing topography by openness: a new application of image processing to digital elevation models. Photogrammetric engineering and remote sensing, 68(3), 257-266.

See Also: Viewshed, HorizonAngle, TimeInDaylight, Hillshade

Parameters:

Python function:

wbt.openness(
    i, 
    pos_output, 
    neg_output, 
    dist=20, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=Openness --input=DEM.tif ^
--pos_output=positive_openness.tif ^
--neg_output=negative_openness.tif --dist=500 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 30/03/2021

Last Modified: 30/03/2021

PennockLandformClass

Tool can be used to perform a simple landform classification based on measures of slope gradient and curvature derived from a user-specified digital elevation model (DEM). The classification scheme is based on the method proposed by Pennock, Zebarth, and DeJong (1987). The scheme divides a landscape into seven element types, including: convergent footslopes (CFS), divergent footslopes (DFS), convergent shoulders (CSH), divergent shoulders (DSH), convergent backslopes (CBS), divergent backslopes (DBS), and level terrain (L). The output raster image will record each of these base element types as:

The definition of each of the elements, based on the original Pennock et al. (1987) paper, is as follows:

Where PROFILE is profile curvature, GRADIENT is the slope gradient, and PLAN is the plan curvature. Note that these values are likely landscape and data specific and can be adjusted by the user. Landscape classification schemes that are based on terrain attributes are highly sensitive to short-range topographic variability (i.e. roughness) and can benefit from pre-processing the DEM with a smoothing filter to reduce the effect of surface roughness and emphasize the longer-range topographic signal. The FeaturePreservingSmoothing tool offers excellent performance in smoothing DEMs without removing the sharpness of breaks-in-slope.

Reference:

Pennock, D.J., Zebarth, B.J., and DeJong, E. (1987) Landform classification and soil distribution in hummocky terrain, Saskatchewan, Canada. Geoderma, 40: 297-315.

See Also: FeaturePreservingSmoothing

Parameters:

Python function:

wbt.pennock_landform_class(
    dem, 
    output, 
    slope=3.0, 
    prof=0.1, 
    plan=0.0, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=PennockLandformClass -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif --slope=3.0 ^
--prof=0.1 --plan=0.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 12/07/2017

Last Modified: 03/09/2020

PercentElevRange

Percent elevation range (PER) is a measure of local topographic position (LTP). It expresses the vertical position for a digital elevation model (DEM) grid cell (z₀) as the percentage of the elevation range within the neighbourhood filter window, such that:

PER = z₀ / (z_max - z_min) x 100

where z₀ is the elevation of the window's center grid cell, z_max is the maximum neighbouring elevation, and z_min is the minimum neighbouring elevation.

Neighbourhood size, or filter size, is specified in the x and y dimensions using the --filterx and --filteryflags. These dimensions should be odd, positive integer values (e.g. 3, 5, 7, 9, etc.).

Compared with ElevPercentile and DevFromMeanElev, PER is a less robust measure of LTP that is susceptible to outliers in neighbouring elevations (e.g. the presence of off-terrain objects in the DEM).

References:

Newman, D. R., Lindsay, J. B., and Cockburn, J. M. H. (2018). Evaluating metrics of local topographic position for multiscale geomorphometric analysis. Geomorphology, 312, 40-50.

See Also: ElevPercentile, DevFromMeanElev, DiffFromMeanElev, RelativeTopographicPosition

Parameters:

Python function:

wbt.percent_elev_range(
    dem, 
    output, 
    filterx=3, 
    filtery=3, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=PercentElevRange -v ^
--wd="/path/to/data/" -i=DEM.tif -o=output.tif --filter=25 

Source code on GitHub

Author: Dr. John Lindsay

Created: 25/06/2017

Last Modified: 30/01/2020

PlanCurvature

This tool calculates the plan (or contour) curvature from a digital elevation model (DEM). Plan curvature is the curvature of a contour line at a given point on the topographic surface (Florinsky, 2017). This variable has an unbounded range that can take either positive or negative values. Positive values of the index are indicative of flow divergence while negative plan curvature values indicate flow convergence. Thus plan curvature is similar to tangential curvature, although the literature suggests that the latter is more numerically stable (Wilson, 2018). Plan curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

Wilson, J. P. (2018). Environmental applications of digital terrain modeling. John Wiley & Sons.

TangentialCurvature, ProfileCurvature, MeanCurvature, GaussianCurvature, MinimalCurvature, MaximalCurvature

Parameters:

Python function:

wbt.plan_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=PlanCurvature -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 01/06/2017

Last Modified: 12/01/2022

Profile

This tool can be used to plot the data profile, along a set of one or more vector lines (--lines), in an input (--surface) digital elevation model (DEM), or other surface model. The data profile plots surface height (y-axis) against distance along profile (x-axis). The tool outputs an interactive SVG line graph embedded in an HTML document (--output). If the vector lines file contains multiple line features, the output plot will contain each of the input profiles.

If you want to extract the longitudinal profile of a river, use the LongProfile tool instead.

See Also: LongProfile, HypsometricAnalysis

Parameters:

Python function:

wbt.profile(
    lines, 
    surface, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=Profile -v --wd="/path/to/data/" ^
--lines=profile.shp --surface=dem.tif -o=profile.html 

Source code on GitHub

Author: Dr. John Lindsay

Created: 21/02/2018

Last Modified: 12/10/2018

ProfileCurvature

This tool calculates the profile (or vertical) curvature, or the rate of change in slope along a flow line, from a digital elevation model (DEM). It is the curvature of a normal section having a common tangent line with a slope line at a given point on the surface (Florinsky, 2017). This variable has an unbounded range that can take either positive or negative values. Positive values of the index are indicative of flow acceleration while negative profile curvature values indicate flow deceleration. Profile curvature is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

PlanCurvature, TangentialCurvature, MeanCurvature, GaussianCurvature, MinimalCurvature, MaximalCurvature

Parameters:

Python function:

wbt.profile_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=ProfileCurvature -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/062017

Last Modified: 15/01/2022

RelativeAspect

This tool creates a new raster in which each grid cell is assigned the terrain aspect relative to a user-specified direction (--azimuth). Relative terrain aspect is the angular distance (measured in degrees) between the land-surface aspect and the assumed regional wind azimuth (Bohner and Antonic, 2007). It is bound between 0-degrees (windward direction) and 180-degrees (leeward direction). Relative terrain aspect is the simplest of the measures of topographic exposure to wind, taking into account terrain orientation only and neglecting the influences of topographic shadowing by distant landforms and the deflection of wind by topography.

The user must specify the name of a digital elevation model (DEM) (--dem) and an azimuth (i.e. a wind direction). The Z Conversion Factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM, and the DEM is in a projected coordinate system. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor.

Reference:

Böhner, J., and Antonić, O. (2009). Land-surface parameters specific to topo-climatology. Developments in Soil Science, 33, 195-226.

See Also: Aspect

Parameters:

Python function:

wbt.relative_aspect(
    dem, 
    output, 
    azimuth=0.0, 
    zfactor=None, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RelativeAspect -v --wd="/path/to/data/" ^
--dem=DEM.tif -o=output.tif --azimuth=180.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 17/06/2017

Last Modified: 03/09/2020

RelativeTopographicPosition

Relative topographic position (RTP) is an index of local topographic position (i.e. how elevated or low-lying a site is relative to its surroundings) and is a modification of percent elevation range (PER; PercentElevRange) and accounts for the elevation distribution. Rather than positioning the central cell's elevation solely between the filter extrema, RTP is a piece-wise function that positions the central elevation relative to the minimum (z_min), mean (μ), and maximum values (z_max), within a local neighbourhood of a user-specified size (--filterx, --filtery), such that:

RTP = (z₀ − μ) / (μ − z_min), if z₀ < μ

OR

RTP = (z₀ − μ) / (z_max - μ), if z₀ >= μ

The resulting index is bound by the interval [−1, 1], where the sign indicates if the cell is above or below than the filter mean. Although RTP uses the mean to define two linear functions, the reliance on the filter extrema is expected to result in sensitivity to outliers. Furthermore, the use of the mean implies assumptions of unimodal and symmetrical elevation distribution.

In many cases, Elevation Percentile (ElevPercentile) and deviation from mean elevation (DevFromMeanElev) provide more suitable and robust measures of relative topographic position.

Reference:

Newman, D. R., Lindsay, J. B., and Cockburn, J. M. H. (2018). Evaluating metrics of local topographic position for multiscale geomorphometric analysis. Geomorphology, 312, 40-50.

See Also: DevFromMeanElev, DiffFromMeanElev, ElevPercentile, PercentElevRange

Parameters:

Python function:

wbt.relative_topographic_position(
    dem, 
    output, 
    filterx=11, 
    filtery=11, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RelativeTopographicPosition -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif ^
--filter=25 

Source code on GitHub

Author: Dr. John Lindsay

Created: 06/06/2017

Last Modified: 30/01/2020

RemoveOffTerrainObjects

This tool can be used to create a bare-earth DEM from a fine-resolution digital surface model. The tool is typically applied to LiDAR DEMs which frequently contain numerous off-terrain objects (OTOs) such as buildings, trees and other vegetation, cars, fences and other anthropogenic objects. The algorithm works by finding and removing steep-sided peaks within the DEM. All peaks within a sub-grid, with a dimension of the user-specified maximum OTO size (--filter), in pixels, are identified and removed. Each of the edge cells of the peaks are then examined to see if they have a slope that is less than the user-specified minimum OTO edge slope (--slope) and a back-filling procedure is used. This ensures that OTOs are distinguished from natural topographic features such as hills. The DEM is preprocessed using a white top-hat transform, such that elevations are normalized for the underlying ground surface.

Note that this tool is appropriate to apply to rasterized LiDAR DEMs. Use the LidarGroundPointFilter tool to remove or classify OTOs within a LiDAR point-cloud.

Reference:

J.B. Lindsay (2018) A new method for the removal of off-terrain objects from LiDAR-derived raster surface models. Available online, DOI: 10.13140/RG.2.2.21226.62401

See Also: MapOffTerrainObjects, TophatTransform, LidarGroundPointFilter

Parameters:

Python function:

wbt.remove_off_terrain_objects(
    dem, 
    output, 
    filter=11, 
    slope=15.0, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RemoveOffTerrainObjects -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=bare_earth_DEM.tif ^
--filter=25 --slope=10.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 06/06/2017

Last Modified: 07/08/2020

RingCurvature

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the ring curvature, which is the product of horizontal excess and vertical excess curvatures (Shary, 1995), from a digital elevation model (DEM). Like Rotor, ring curvature is used to measure flow line twisting. Ring curvature has values equal to or greater than zero and is measured in units of m^-2.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary PA (1995) Land surface in gravity points classification by a complete system of curvatures. Mathematical Geology 27: 373–390.

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

Rotor, MinimalCurvature, MaximalCurvature, MeanCurvature, GaussianCurvature, ProfileCurvature, TangentialCurvature

Parameters:

Python function:

wbt.ring_curvature(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=RingCurvature -i=DEM.tif -o=output.tif ^
--log 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 11/01/2022

Last Modified: 28/11/2022

Rotor

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the spatial pattern of rotor, which describes the degree to which a flow line twists (Shary, 1991), from a digital elevation model (DEM). Rotor has an unbounded range, with positive values indicating that a flow line turns clockwise and negative values indicating flow lines that turn counter clockwise (Florinsky, 2017). Rotor is measured in units of m^-1.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor. Curvature values are often very small and as such the user may opt to log-transform the output raster (--log). Transforming the values applies the equation by Shary et al. (2002):

Θ' = sign(Θ) ln(1 + 10ⁿ|Θ|)

where Θ is the parameter value and n is dependent on the grid cell size.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Shary PA (1991) The second derivative topographic method. In: Stepanov IN (ed) The Geometry of the Earth Surface Structures. Pushchino, USSR: Pushchino Research Centre Press, 30–60 (in Russian).

Shary P. A., Sharaya L. S. and Mitusov A. V. (2002) Fundamental quantitative methods of land surface analysis. Geoderma 107: 1–32.

See Also:

RingCurvature, ProfileCurvature, TangentialCurvature, PlanCurvature, MeanCurvature, GaussianCurvature, MinimalCurvature, MaximalCurvature

Parameters:

Python function:

wbt.rotor(
    dem, 
    output, 
    log=False, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=Rotor -i=DEM.tif -o=output.tif --log

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 08/01/2022

Last Modified: 28/11/2022

RuggednessIndex

The terrain ruggedness index (TRI) is a measure of local topographic relief. The TRI calculates the root-mean-square-deviation (RMSD) for each grid cell in a digital elevation model (DEM), calculating the residuals (i.e. elevation differences) between a grid cell and its eight neighbours. Notice that, unlike the output of this tool, the original Riley et al. (1999) TRI did not normalize for the number of cells in the local window (i.e. it is a root-square-deviation only). However, using the mean has the advantage of allowing for the varying number of neighbouring cells along the grid edges and in areas bordering NoData cells. This modification does however imply that the output of this tool cannot be directly compared with the index ranges of level to extremely rugged terrain provided in Riley et al. (1999)

Reference:

Riley, S. J., DeGloria, S. D., and Elliot, R. (1999). Index that quantifies topographic heterogeneity. Intermountain Journal of Sciences, 5(1-4), 23-27.

See Also: RelativeTopographicPosition, DevFromMeanElev

Parameters:

Python function:

wbt.ruggedness_index(
    dem, 
    output, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=RuggednessIndex -v ^
--wd="/path/to/data/" --dem=DEM.tif -o=output.tif 

Source code on GitHub

Author: Dr. John Lindsay

Created: 22/06/2017

Last Modified: 03/09/2020

SedimentTransportIndex

This tool calculates the sediment transport index, or sometimes, length-slope (LS) factor, based on input specific contributing area (A_s, i.e. the upslope contributing area per unit contour length; --sca) and slope gradient (β, measured in degrees; --slope) rasters. Moore et al. (1991) state that the physical potential for sheet and rill erosion in upland catchments can be evaluated by the product R K LS, a component of the Universal Soil Loss Equation (USLE), where R is a rainfall and runoff erosivity factor, K is a soil erodibility factor, and LS is the length-slope factor that accounts for the effects of topography on erosion. To predict erosion at a point in the landscape the LS factor can be written as:

LS = (n + 1)(A_s / 22.13)ⁿ(sin(β) / 0.0896)^m

where n = 0.4 (--sca_exponent) and m = 1.3 (--slope_exponent) in its original formulation.

This index is derived from unit stream-power theory and is sometimes used in place of the length-slope factor in the revised universal soil loss equation (RUSLE) for slope lengths less than 100 m and slope less than 14 degrees. Like many hydrological land-surface parameters SedimentTransportIndex assumes that contributing area is directly related to discharge. Notice that A_s must not be log-transformed prior to being used; A_s is commonly log-transformed to enhance visualization of the data. Also, A_s can be derived using any of the available flow accumulation tools, alghough better results usually result from application of multiple-flow direction algorithms such as DInfFlowAccumulation and FD8FlowAccumulation. The slope raster can be created from the base digital elevation model (DEM) using the Slope tool. The input images must have the same grid dimensions.

Reference:

Moore, I. D., Grayson, R. B., and Ladson, A. R. (1991). Digital terrain modelling: a review of hydrological, geomorphological, and biological applications. Hydrological processes, 5(1), 3-30.

See Also: StreamPowerIndex, DInfFlowAccumulation, FD8FlowAccumulation

Parameters:

Python function:

wbt.sediment_transport_index(
    sca, 
    slope, 
    output, 
    sca_exponent=0.4, 
    slope_exponent=1.3, 
    callback=default_callback
)

Command-line Interface:

>>./whitebox_tools -r=SedimentTransportIndex -v ^
--wd="/path/to/data/" --sca='flow_accum.tif' ^
--slope='slope.tif' -o=output.tif --sca_exponent=0.5 ^
--slope_exponent=1.0 

Source code on GitHub

Author: Dr. John Lindsay

Created: 02/07/2017

Last Modified: 30/01/2020

ShadowAnimation

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool creates an interactive animated GIF of shadows based on an input digital surface model (DSM). The shadow model is based on the modelled positions of the sun throughout a user-specified date (--date) sampling at a regular interval (--interval), in minutes. Similar to the TimeInDaylight tool, this tool uses calculated horizon angle (HorizonAngle) values and a solar position model to determine which grid cells are located in shadow areas due to distant obsticles. The calculation of horizon angle, requires the user input a maximum search distance parameter (--max_dist).

The location parameter (--location) should take the form Lat/Long/UTC-offset, e.g. 43.5448/-80.2482/-4/. Note, the location need only be approximate; the postion of the central location of the input DSM raster should suffice.

The output (--output) of this tool is an HTML file, containing the interactive GIF animation. Users are able to zoom and pan around the displayed DEV animation. The DSM may be rendered in one of several available palettes (--palette) suitable for visualization topography. The user must also specify the image height (--height) in the output file, the time delay (--delay, in milliseconds) used in the GIF animation, and an optional label (--label), which will appear in the upper lefthand corner. Note that the output is simply HTML, CSS, javascript code, and a GIF file, which can be readily embedded in other documents.

Users should be aware that the outut GIF can be very large in size, depending on the size of the input DEM file. To reduce the file size of the output, it may be desirable to coarsen the input DEM resolution using image resampling (Resample).

The following is an example of what the output of this tool looks like. Click the image for an interactive example.

For more information about this tool, see this blog on the WhiteboxTools homepage.

See Also: ShadowImage, TimeInDaylight, HorizonAngle, LidarDigitalSurfaceModel

Parameters:

Python function:

wbt.shadow_animation(
    i, 
    output, 
    palette="atlas", 
    max_dist="", 
    date="21/06/2021", 
    interval=15, 
    location="43.5448/-80.2482/-4", 
    height=600, 
    delay=250, 
    label="", 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=ShadowAnimation -i=dsm.tif ^
-p='high_relief' -o=shadow_animation.html --max_dist=500 ^
--date='21/06/2021' --interval=20 --location='43.55/ -80.25/ ^
-4' --height=620 --delay=200 --label='Shadow Animation' 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 01/05/2021

Last Modified: 01/05/2021

ShadowImage

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool generates a raster of shadow areas based on an input digital surface model (DSM). This shadow model is based on the calculated positions of the sun throughout a user-specified date (--date), sampling at a regular interval (--interval), in minutes. Similar to the TimeInDaylight tool, this tool uses calculated horizon angle (HorizonAngle) values and a solar position model to determine which grid cells are located in shadow areas due to distant obsticles. The calculation of horizon angle, requires the user input a maximum search distance parameter (--max_dist).

The user must specify the date (--date), time (--time), and location (--location) of the input DSM. The date should have the format DD/MM/YYYY, e.g. 27/11/1976. The time should have the format HH::MM, e.g. 03:15AM or 14:30. The location parameter should take the form Lat/Long/UTC-offset, e.g. 43.5448/-80.2482/-4/. Note, the location need only be approximate; the postion of the central location of the input DSM raster should suffice.

The output (--output) of this tool is a raster. If a palette (--palette) is chosen, then the output raster will be a colour composite image, containing a hysometrically tinted (i.e. elevation coloured) shadow model. The DSM may be rendered in one of several available palettes (--palette) suitable for visualization topography. If the palette is set to 'none', the output image will not be a colour composite, but rather, will be a 16-bit integer raster, and should be displayed in a grey-scale palette.

The following is an example of what the output of this tool looks like.

For more information about this tool, see this blog on the WhiteboxTools homepage.

See Also: ShadowAnimation, TimeInDaylight, HorizonAngle, LidarDigitalSurfaceModel, HypsometricallyTintedHillshade

Parameters:

Python function:

wbt.shadow_image(
    i, 
    output, 
    palette="soft", 
    max_dist="", 
    date="21/06/2021", 
    time="13:00", 
    location="43.5448/-80.2482/-4", 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=ShadowImage -i=dsm.tif ^
-p='high_relief' -o=shadow_animation.html --max_dist=500 ^
--date='21/06/2021' --time='14:30' --location='43.55/ -80.25/ ^
-4' 

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 08/06/2021

Last Modified: 13/06/2021

ShapeIndex

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the shape index (Koenderink and van Doorn, 1992) from a digital elevation model (DEM). This variable ranges from -1 to 1, with positive values indicative of convex landforms, negative values corresponding to concave landforms (Florinsky, 2017). Absolute values from 0.5 to 1.0 are associated with elliptic surfaces (hills and closed depressions), while absolute values from 0.0 to 0.5 are typical of hyperbolic surface form (saddles). Shape index is a dimensionless variable and has utility in landform classification applications.

Koenderink and vsn Doorn (1992) make the following observations about the shape index:

• Two shapes for which the shape index differs merely by sign represent complementary pairs that will fit together as ‘stamp’ and ‘mould’ when suitably scaled;

• The shape for which the shape index vanishes - and consequently has indeterminate sign - represents the objects which are congruent to their own moulds;

• Convexities and concavities find their places on opposite sides of the shape scale. These basic shapes are separated by those shapes which are neither convex nor concave, that are the saddle-like objects. The transitional shapes that divide the convexities/concavities from the saddle-shapes are the cylindrical ridge and the cylindrical rut.

The user must specify the name of the input DEM (--dem) and the output raster (--output). The The Z conversion factor (--zfactor) is only important when the vertical and horizontal units are not the same in the DEM. When this is the case, the algorithm will multiply each elevation in the DEM by the Z Conversion Factor.

For DEMs in projected coordinate systems, the tool uses the 3rd-order bivariate Taylor polynomial method described by Florinsky (2016). Based on a polynomial fit of the elevations within the 5x5 neighbourhood surrounding each cell, this method is considered more robust against outlier elevations (noise) than other methods. For DEMs in geographic coordinate systems (i.e. angular units), the tool uses the 3x3 polynomial fitting method for equal angle grids also described by Florinsky (2016).

References:

Florinsky, I. (2016). Digital terrain analysis in soil science and geology. Academic Press.

Florinsky, I. V. (2017). An illustrated introduction to general geomorphometry. Progress in Physical Geography, 41(6), 723-752.

Koenderink, J. J., and Van Doorn, A. J. (1992). Surface shape and curvature scales. Image and vision computing, 10(8), 557-564.

Curvedness, MinimalCurvature, MaximalCurvature, TangentialCurvature, ProfileCurvature, MeanCurvature, GaussianCurvature

Parameters:

Python function:

wbt.shape_index(
    dem, 
    output, 
    zfactor=1.0, 
    callback=default_callback
)

Command-line Interface:

>> ./whitebox_tools -r=ShapeIndex -i=DEM.tif -o=output.tif

Source code is unavailable due to proprietary license.

Author: Whitebox Geospatial Inc. (c)

Created: 13/01/2022

Last Modified: 28/11/2022

SkyViewFactor

Note this tool is part of a WhiteboxTools extension product. Please visit Whitebox Geospatial Inc. for information about purchasing a license activation key (https://www.whiteboxgeo.com/extension-pricing/).

This tool calculates the sky-view factor (SVF) from an input digital elevation model (DEM) or digital surface model (DSM). The SVF is the proportion of the celestial hemisphere above a point on the earth's surface that is not obstructed by the surrounding land surface. It is often used to model the diffuse light that is received at the surface and has also been applied as a relief-shading technique (Böhner et al., 2009; Zakšek et al., 2011).

The user must specify an input DEM (dem), the azimuth fraction (az_fraction), the maximum search distance (max_dist), and the height offset of the observer (observer_hgt_offset). The input DEM should usually be a digital surface model (DSM) that contains significant off-terrain objects. Such a model, for example, could be created using the first-return points of a LiDAR data set, or using the lidar_digital_surface_model tool. The azimuth fraction should be an even divisor of 360-degrees and must be between 1-45 degrees.