Appendices

Topographic Data Collection

Survey Library

Each GCD project contains a survey library where you load all your topographic surveys for use in change detection analyses. Conceptually, GCD software considers each DEM to be a separate "survey". The survey library is intended to house all your surveys through time for a given study site. Different study sites on the same river, for example, will have non-overlapping spatial extents and should be placed in separate GCD Projects. [survey_library_contents]

Survey Method

Below is a list of common topographic survey methods and brief descriptions of how these methods work.

• Total Station Survey
• CHaMP Protocol Survey
• rtkGPS Survey
• Aerial Photogrammetry
• Aerial LiDAR
• Terrestrial Laser Scanning
• Single Beam SONAR
• Multi Beam SONAR
• Bathymetric LiDAR
• Hybrid Approach
• Undefined

Total Station Survey

A total station topographic survey requires two people. One person operates the survey base instrument while the other person places a stadia rod and prism along topographic features. For best results major grade breaks in the streambed and bank topography should be captured Lane et al. 1994, Brasington et al. 2003, ). In addition, to point data, breaklines should be used to capture distinct features such as top-of-bank, edge-of-water, and bankfull indicators. For this specific total station survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

CHaMP Protocol Survey

The total station topographic survey should adhere to the Topographic Point Collection Method Guidelines to generate a high resolution (10 cm) DEM of the site. A total station topographic survey requires two people. One person operates the survey instrument while the other person places a stadia rod and prism along topographic features. For best results major grade breaks in the streambed and bank topography should be captured Lane et al. 1994, Brasington et al. 2003, ). In addition to point data, breaklines should be used to capture distinct features such as top-of-bank, edge-of-water, and bankfull indicators. The spatial information collected during the total station survey should have been transformed and projected with the CHaMP Transformation Tool (CTT) using three benchmarks [benchmarks] (collected from handheld GPS) established by the surveyors and compass-derived orientation. The CTT transforms vector data in a manner that preserves the high relative accuracy of a total station survey. See (Wheaton et al. 2012) and the ReadMe page for more background. For this specific survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

rtkGPS Survey

Real-time kinematic (rtk) GPS units are a type of survey-grade GPS, and consist of a mobile unit and a base station. Individual data point positional accuracy is improved from differential calculations sent from the base station to the rover via radio. Average reported vertical precision of individual survey-grade GPS points is lower than total stations, with reported values ranging from 0.01 m to 0.08 m (Brasington et al. 2003, Wheaton et al. 2010a). RtkGPS is a proven ground based method for obtaining high resolution topographic data in support of geomorphic change detection. The advantages of rtkGPS surveys include much less cost relative to aerial data collection methods, a greater sampling efficiency relative to total stations, and arguably requires less skill to operate than a total station or TLS. Data can be collected day or night and in most weather conditions. Data is high precision, with built-in quality control and data has real world coordinates which eliminates the need to establish control prior to a survey. RtkGPS surveys are limited at sites with dense vegetation making total station a more universal approach (Bangen 2013). Like total station surveying rtkGPS, relies on surveyor judgment to select appropriate point sampling locations. This allows the surveyor to increase point density in areas with greater topographic complexity or to detail features of interest and decrease point density in areas with homogeneous topography and of less geomorphic interest (Brasington et al. 2000, Bangen 2013). For this specific rtk survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Aerial Photogrammetry

Aerial photogrammetry is the traditional airborne method for obtaining high resolution topographic data. In order to generate topographic data 60 percent overlapping stereo pairs are required. Aerial triangulation, orthorectification and DEM generation is also required. The quality of the resulting topographic data can vary greatly depending on the platform, camera, ground control and post processing of the imagery. For this specific aerial photogrammetry survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Aerial LiDAR

During the past decade aerial LiDAR (also called airborne laser scanning) has become the standard airborne method for obtaining high resolution topographic data. With this approach measurements are made by a laser beam pulse emitted from a sensor (Large and Heritage 2009, Lefsky et al. 2002). The time elapsed from when the emitted signal hits an object, is reflected and returns to the instrument sensor is recorded. As the speed and wavelength of the emitted laser pulse is known, the elapsed return time is used to derive the distance to the object that reflected the pulse. In contrast to total station and rtkGPS sampling methods that employ surveyor judgment, aerial LiDAR collects data uses a constant, pre-set emitted pulse rate. Few studies have field validated the accuracy of aerial LiDAR in river environments, but those that have reported point root mean square error of 0.18 m in foodplains, 0.19 m on cobble bars, 0.53 m on sand bars and 3.26 m on bedrock outcrops in a confined gorge (Notebaert et al. 2009, Bowen and Waltermire 2002, Reusser and Bierman 2007). For this specific aerial LiDAR survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Terrestrial Laser Scanning

Terrestrial Laser Scanning (TLS; also called ground-based LiDAR) units can collect approximately 2,000 to 100,000 points per second with sub-centimeter accuracy in the vertical and horizontal direction. The spatial resolution of the point clouds, allow for three-dimensional representation of complex features ranging from individual trees to pebbles. In the field of geomorphology, TLS point clouds have been used to: extract grain size on dry exposed gravel bars, document sand-bar evolution, measure changes in glacial outwash barforms, and delineate stream biotopes. TLS point precision over short distances and on un-submerged bar surfaces is approximately 0.002 m. However, TLS has mean error of approximately 0.007 on exposed bedrock, 0.069 m in grass, 0.075 in broad leaf vegetation, and 0.256 m in submerged areas is likely. The latter finding highlights a disadvantage of discrete return TLS, which is its inability to produce reliable returns from below the water surface (Heritage and Hetherington 2007). While capable of efficiently capturing sub geomorphic unit, or grain, scale topography over large areas, the high capital expenditure of TLSs relative to other ground-based methods (e.g. total station, rtkGPS) has limited their application (Bangen 2013). For this TLS survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Single and Multi Beam SONAR

Single beam and multi beam SONAR sensors measure water depth with high precision X and Y position determined by a mounted rtkGPS or total station unit. Therefore, the precision and accuracy limitations of rtkGPS and total station apply to single beam and multi beam SONAR surveys. Depending on water depth and sampling frequency, multi beam SONAR can provide high point resolution datasets of the instream channel similar to that achievable with TLS (Ferrari and Collins 2006). For this specific SONAR survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Bathymetric LiDAR

Bathymetric LiDAR operates similarly to airborne LiDAR, with one notable exception. Bathymetric LiDAR transmits two light waves, one in the infrared and one in the green spectrum. The infrared band is quickly absorbed and is therefore used to detect the water surface, while the green band is used as the optimum color to achieve maximum penetration in shallow water and is used to delineate the river bed. For this specific Bathymetric LiDAR survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Hybrid Approach

A hybrid approach is advantageous because ground-based methods can be limited in their practicality due to the sheer size of some floodplains and the time required to survey their subtle topography (Bangen 2013). Rayburg et al. (2009) found that aerial LiDAR point densities averaging 4 points/m2 coverage on floodplains was sufficient to generate 10 cm resolution DEMs that captured subtle variations in micro-topography such as abandoned paleo channels and active overbank channels. Therefore, when feasible, researchers should consider conducting hybrid surveys designed to exploit the strengths of different methods and to increase overall sampling efficiency and data accuracy such as exploiting aerial LiDAR or aerial photogrammetry datasets on the channel margins and floodplain while using rtkGPS or imagery derived depths in the wetted channel (Bangen 2013). Here is a link to a video tutorial that describes working with hybrid surveys in GCD. For this specific hybrid survey a total of [number_of_survey_points] XYZ points were collected and are displayed in [raw_survey_data].

Undefined

No survey method was specified by the user.

Acknowledgements

The original Matlab versions were coded by Joe Wheaton (Utah State University Department of Watershed Sciences) and James Brasington (University of Canterbury) with financial support from the University of Southampton School of Geography, Aberystwyth University Institute for Geography and Earth Sciences. This version of the GCD is currently under development by ESSA Technologies. The underlying C++ library that boasts most of the core functionality was coded by Chris Garrard (Utah State University RSGIS Lab) , under the direction of Joe Wheaton and with generous financial support from ICRRR as part of GCD 4. The current development team consists of Philip Bailey (North Arrow Research), Nick Ochoski (ESSA) , Frank Poulsen (ESSA) , Matthew Reimer, James Hensleigh and Joe Wheaton (USU). Current funding for the Geomorphic Change Detection Software development (GCD 5) is being provided by the USGS’s Grand Canyon Monitoring & Research Center.

References

Bangen SG. 2013. Comparison of Topographic Surveying Techniques in Streams. Unpublished Masters Thesis Utah State University, Logan, Utah, 151 pp. Available at: http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=2532&context=etd

Bouwes N, Moberg J, Weber N, Bouwes B, Bennett S, Beasley C, Jordan CE, Nelle P, Polino S, Rentmeester S, Semmens B, Volk C, Ward MB and White J. 2011. Scientific Protocol for Salmonid Habitat Survyes within the Columbia Habitat Monitoring Program, Prepared by the Integrated Status and Effectiveness Monitoring Program and published by Terraqua, Inc., Wauconda, WA, 118 pp. http://www.pnamp.org/sites/default/files/champhabitatprotocol_20110125.pdf

Bowen, Z.H., Waltermire, R.G., 2002. Evaluation of light detection and ranging for measuring river corridor topography. JAWRA 38 (1), 33-41. DOI: 10.1111/j.1752-1688.2002.tb01532.x.

Brasington J, Rumsby BT, Mcvey RA. 2000. Monitoring and modelling morphological change in a braided gravel-bed river using high resolution GPS-based survey. Earth Surface Processes and Landforms 25(9): 973–990. DOI: 10.1002/1096-9837(200008)25:9<973::AID-ESP111>3.0.CO;2-Y.

Brasington, J., Langham, J., Rumsby, B., 2003. Methodological sensitivity of morphometric estimates of coarse fluvial sediment transport. Geomorphology 53 (3-4), 299-316. DOI: 10.1016/S0169-555X(02)00320-3.

Ferrari, R., Collins, K., 2006. Reservoir survey and data analysis. In: Yang, C. (Ed.), Erosion and Sediment Manual. Technical Service Center, Bureau of Reclamation, Denver, CO, pp. 1-65. http://www.riversimulator.org/Resources/USBR/ErosionSedimentManual.pdf.

Heritage, G., Hetherington, D., 2007. Towards a protocol for laser scanning in fuvial geomorphology. Earth Surface Processes and Landforms 32 (1), 66-74. DOI: 10.1002/esp.1375.

Lane, S., Richards, K., Chandler, J., 1994. Developments in monitoring and modelling smallscale river bed topography. Earth Surface Processes and Landforms 19 (4), 349-368. DOI: 10.1002/esp.3290190406.

Lane, S.N., Westaway, R.M., Murray Hicks, D., 2003. Estimation of erosion and deposition volumes in a large, gravel-bed, braided river using synoptic remote sensing. Earth Surface Processes and Landforms 28 (3), 249-271. DOI: 10.1002/esp.483.

Large, A.R.G., Heritage, G., 2009. Laser scanning: Evolution of the discipline. In: Heritage, G., Large, A.R.G. (Eds.), Laser Scanning for the Environmental Sciences. Wiley Blackwell, West Sussex, UK, pp. 1-20. https://www.google.com/search?tbm=bks&hl=en&q=ISBN+978-1-4051-5717-9+(hardcover+%3A+alk.+paper)

Lefsky, M.A., Cohen, W.B., Parker, G.G., Harding, D.J., 2002. Lidar remote sensing for ecosystem studies. Bioscience 52 (1), 19-30. DOI: 10.1641/0006-3568(2002)052[0019:LRSFES]2.0.CO;2.

Notebaert, B., Verstraeten, G., Govers, G., Poesen, J., 2009. Qualitative and quantitative applications of LiDAR imagery in fuvial geomorphology. Earth Surface Processes and Landforms 34 (2), 217-231. DOI: 10.1002/esp.1705.

Rayburg, S., Thoms, M., Neave, M., 2009. A comparison of digital elevation models generated from different data sources. Geomorphology 106 (3-4), 261-270. DOI: 10.1016/j.geomorph.2008.11.007.

Reusser, L., Bierman, P., 2007. Accuracy assessment of LiDAR-derived DEMs of bedrock river channels: Holtwood Gorge, Susquehanna River. Geophysical Research Letters 34 (23), 1-6. DOI: 10.1029/2007GL031329.

Wheaton JM. 2008. Uncertainty in Morphological Sediment Budgeting of Rivers. Unpublished PhD University of Southampton, Southampton, 412 pp. Available at: http://www.joewheaton.org/Home/research/projects-1/morphological-sediment-budgeting/phdthesis.

Wheaton JM, Brasington J, Darby SE, Merz JE, Pasternack GB, Sear DA and Vericat D. 2010a. Linking Geomorphic Changes to Salmonid Habitat at a Scale Relevant to Fish. River Research and Applications. 26: 469-486. DOI: 10.1002/rra.1305.

Wheaton JM, Brasington J, Darby SE and Sear D. 2010b. Accounting for uncertainty in DEMs from repeat topographic surveys: Improved sediment budgets. Earth Surface Processes and Landforms. 35(2): 136-156. DOI: 10.1002/esp.1886.

Wheaton JM, Garrard C, Volk C, Whitehead K and Bouwes N. 2012. A Simple, Interactive GIS Tool for Transforming Assumed Total Station Surveys to Real World Coordinates - The CHaMP Transformation Tool. Submitted to Computers & Geosciences.42: 28-36. DOI: DOI: 10.1016/j.cageo.2012.02.003.

About this Report

What is this report?

This is an on demand Geomorphic Change Detection (GCD) analysis report. The report is user generated at either the GCD 5.x software ArcGIS toolbar menu, or as part of a command line GCD analysis. The report is interactive and dynamic and provides a transparent accounting of the data inputs, processing options and outputs generated during a particular run of the GCD 5.x software. This report helps to determine whether or not CHaMP Protocol (Bouwes et al. 2011) techniques were followed correctly, helps to ensure that values fall within acceptable ranges and tolerance limits, and helps troubleshoot where within the workflow errors may have been introduced. Consequently, the metadata with this report will significantly streamline the topographic survey and change detection QA/QC process and will help to synthesize and analyze the GCD 5.x results.

What does GCD software do?

GCD software was developed primarily for morphological sediment budgeting (i.e. change detection) in rivers. High resolution topographic data that can support change detection comes from a variety of platforms including: total station, rtkGPS, aerial photogrammetry, aerial LiDAR, terrestrial laser scanning, single beam sonar, multi beam sonar, bathymetric LiDAR or a hybrid approach that combines two or more of these techniques. The volumetric change in storage is calculated from the difference in surface elevations from digital elevation models (DEMs) derived from repeat topographic surveys (DEM Differencing). DEM differencing is the simple act of subtracting the elevations of an older DEM from the elevations of a newer DEM on a cell by cell basis:

DEMs are a model of 'true' topography, and like all models, contain some degree of error. Individual DEM errors are inevitably propagated into other analyses, such as the DEMs of Difference (DoD). GCD 5.x software provides a suite of tools that quantify the uncertainty in individual DEMs and propagates them through to the DoD (Brasington et al. 2003 and Wheaton et al. 2010a). This helps distinguish 'real' net change from noise inherit in DEMs. The software program also provides ways for determines significance of uncertainty on DoDs and sediment budgets, calculates changes in storage sediment budgets (with +/- vol.) and quantitatively interprets and spatially segregates sediment budgets using different types of masks. Previous DEM error propagation studies (e.g. Brasington et al. 2003, Lane et al. 2003, Wheaton et al. 2010a) have demonstrated that a critical threshold can be applied to DoDs to detect differences at a chosen probabilistic confidence interval (Wheaton et al. 2010a, Bangen 2013).

Geomorphic Change Detection Software

Produced by Joe Wheaton, North Arrow Research & ESSA Technologies Copyright © 2010-2014 Wheaton

GNU License

Developer can be contacted at Joe.Wheaton@usu.edu or Joe Wheaton, Department of Watershed Sciences, Utah State University, 5210 Old Main Hill, Logan, UT 84322-5210, USA. Source code can be acquired from developer by request.

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

Source Code & Other Versions

The latest GCD ArcGIS Add-in was made to extend the functionality of GCD 4 and as an alternative to the Matlab Code in earlier versions, which required users to have a Matlab License, the Matlab Fuzzy Logic Toolbox installed, and a limited knowledge of Matlab. That version was made available to accompany a paper published in Earth Surface Processes and Landforms (Wheaton et al., 2010b), the Wheaton (2008) thesis, and the Wheaton et al. (2010a) RRA paper. That code was provided as supplemental information with the ESPL paper so that readers could test or extend the code as they see fit for their purpose. The main library for this version is available upon request for educational, research and non commercial purposes. However, as it is a C++ library and Visual Studio.NET. The GCD 6 Add-in to ArcGIS is freeware, coded in VisualBasic.Net. We are not currently providing this source code as it leverages ArcObjects and is not easy to adapt without breaking. It is not anticipated that many/any users will be wanting to get under the bonnet. If you really do, call us.

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.