Bathymetry Acquisition - Technologies and Strategies · The main environmental considerations include the depths, coverage, turbidity, temporal variations and seabed bottom type

www.crcsi.com.au

Investigating shallow water bathymetry acquisition technologies, survey considerations and strategies N. D. Quadros

Report prepared for the Commonwealth Government of Australia, Department of Climate Change and Energy Efficiency

Bathymetry Acquisition - Technologies and Strategies

http://www.crcsi.com.au/

www.crcsi.com.au 2

Document Attributes

File name File owner File Location

CRSI UDEM2 Project4 Report Stage 2 ND Quadros ...Coastal and Business Projects\ Projects\04_2011_Bathy_UNA\Mngmt

Document Control

Version Status & revision notes Author Date Approved by Date

0.1 First Draft ND Quadros 21/12/2012 G Kernich 29/01/2012

0.2 Second Draft ND Quadros 29/01/2012 C Fraser 15/03/2013 1.0 Final Publication ND Quadros 03/04/2013 G Kernich 11/04/2013


www.crcsi.com.au 3

Acknowledgements The CRC for Spatial Information (CRCSI) would like to acknowledge the funding and support provided by the Australian Department of Climate Change and Energy Efficiency. A special acknowledgement to Fugro LADS, Pelydryn and EOMAP for their significant inputs into the report. The CRCSI would like to also thank the following individuals for their input into this report:

• Fugro LADS - Mark Sinclair, Hugh Parker and Nigel Townsend

• Pelydryn - Andy Waddington

• EOMAP - Magnus Wettle

• Members of the Intergovernmental Committee for Surveying and Mapping (ICSM) Bathymetry Working Group

• Royal Australian Navy (RAN) LADS - Richard Mortimer

• James Cook University - Rob Beaman

• Optech Australia - David Collison

• Airborne Hydrography AB (AHAB) - Swante Welander

• Riegl - Martin Pfennigbauer

• Australian Hydrographic Service - Doug White

• Commonwealth Scientific and Industrial Research Organisation (CSIRO) - Norm Campbell

• Deakin University - Daniel Ierodiaconou

• Defence Science and Technology Organisation (DSTO) - Julian Vrbancich

• Department of Sustainability and Environment Victoria - Christina Ratcliff

• Department of Science, Information Technology, Innovation and Arts Queensland (DSITIA) - Ramona Dalla Pozza

• Land Information New Zealand (LINZ) - Stuart Caie

• Office of Environment and Heritage New South Wales - Bruce Coates

• Geoimage - David Brady


www.crcsi.com.au 4

Executive Summary Australian Government Departments have recently committed to significant investments in bathymetric LiDAR acquisition for the modelling of coastal processes. This project has been initiated by the Cooperative Research Centre for Spatial Information (CRCSI) and Department of Climate Change and Energy Efficiency (DCCEE) to better understand the requirements for the near-shore bathymetry collection, and to outline the strategies which can be employed to satisfy user needs.

The first stage of this research identified user needs and challenges by analysing a questionnaire distributed to bathymetry users in Australia and New Zealand. User concerns on bathymetry extent, quality and accessibility were all discussed within the first research report. This research complements the first stage by outlining alternative technologies, sensors and strategies to acquire bathymetry. To address this research this report is divided into the following five chapters:

1. A description of current bathymetric sensor technologies

2. Project and environmental factors which impact bathymetric survey technologies

3. Recent sample bathymetric surveys conducted with multiple sensors

4. An example strategy for a complex, large area, near-shore bathymetric survey

5. The development of a standard bathymetric LiDAR specification template

The most recent development in bathymetric LiDAR is the new range of so-called topo-bathy sensors. These lower power sensors complement the traditional bathymetric LiDAR by acquiring a higher point density onshore and in depths less than 10m. The most suitable technologies to supplement bathymetric and topo-bathy LiDAR are satellite imagery and maritime vessel based multi-beam echo sounders (MBES), of which the latter may be integrated with a terrestrial laser scanner. A number of other technologies exist, however they are still in the research phase or do not have a distinct advantage over LiDAR, satellite imagery nor MBES.

When initiating a bathymetric survey a number of project and environmental considerations should be assessed to select the most appropriate technology and sensor. The main project considerations include the required accuracy, point spacing, vertical datum, coverage, budget, timelines, accessibility and supplementary datasets. The main environmental considerations include the depths, coverage, turbidity, temporal variations and seabed bottom type. All these factors affect the choice of technology and sensor in different ways. Knowledge and experience of how each technology and sensor are impacted by these factors should be utilised when planning a bathymetric survey.

For large area, near-shore bathymetry acquisition we propose a hierarchy of technologies to produce the most effective survey. The hierarchy is broadly based on a trade-off between acquisition time and cost efficiency, against the measurement density and accuracy. The most cost effective technology over a large area (which is first in the hierarchy) is satellite image derived bathymetry, provided the necessary conditions exist. However, it is less accurate, has a reduced depth penetration , and is not as suitable for as many applications as bathymetric LiDAR and MBES. Bathymetric (and topo-bathy) LiDAR is the second technology in the hierarchy. Airborne LiDAR traverses the littoral zone and produces bathymetry which has both the accuracy and coverage required for coastal modelling. Bathymetric LiDAR is less cost effective than satellite image derived bathymetry, and also produces gaps in areas of high turbidity. The third technology in the hierarchy


www.crcsi.com.au 5

is maritime vessel-based MBES. MBES has the slowest rate of coverage and is the least cost effective over large, near-shore areas. However, unlike the previous two technologies it is not as affected by turbidity and adverse seafloor conditions. This enables MBES to be acquired strategically within data gaps following a bathymetric LiDAR and satellite imagery acquisition.

For large area, near-shore bathymetry surveys, there are numerous options once multi-sensor bathymetry acquisitions are considered. Single sensor surveys can be used, however they do not necessarily present the best value to end users. A multi-sensor approach is able to take advantage of the suitability of each sensor for particular near-shore environments, and when used together are able to complement each other to provide an optimal, value solution for a survey area.

A multi-sensor bathymetry acquisition strategy is presented in the report for the Queensland coast south of Cooktown. The strategy is based on minimising the acquisition cost, whilst providing data that would be suitable for a number of applications, thereby providing value for money. The proposed strategy recommends deriving bathymetry from satellite imagery, followed by a bathy LiDAR acquisition, and then an MBES survey in critical areas and LiDAR data gaps.

To assist with commissioning a bathymetric LiDAR project for coastal and environmental applications specifications are attached in Appendix B. These specifications are similar to those developed for topographic LiDAR in Australia and New Zealand. The specifications have been reviewed and approved by bathymetric LiDAR operators in Australia and New Zealand. The specifications uphold the International Hydrographic Organisations (IHO) standards and provide a mechanism to commission a bathymetric LiDAR survey.


www.crcsi.com.au 6

Table of Contents

1 Introduction .................................................................................................................................... 8 1.1 Background ............................................................................................................................. 8

1.2 Aim and Outcomes .................................................................................................................. 9

2 Bathymetric Acquisition Technologies .......................................................................................... 10 2.1 LiDAR Bathymetry ................................................................................................................. 10

2.2 Satellite and Airborne Derived Bathymetry .......................................................................... 13

2.2.1 Satellite and Aerial Imagery .......................................................................................... 13 2.2.2 Hyperspectral Imagery .................................................................................................. 16 2.2.3 Algorithms to Derive Bathymetry from Imagery .......................................................... 16 2.2.4 Satellite Radar ............................................................................................................... 18

2.3 Maritime Vessel Bathymetry ................................................................................................ 19

2.3.1 Multi-Beam and Single-Beam Echo Sounders ............................................................... 19 2.3.2 Side-Scan Sonar and Sub-bottom Profilers ................................................................... 21

2.4 Specialised Technologies ...................................................................................................... 22

2.4.1 Airborne Electro-Magnetic Bathymetry (AEMB) .......................................................... 22 2.4.2 Autonomous Underwater Vehicles (AUV) .................................................................... 23 2.4.3 Satellite LiDAR ............................................................................................................... 23

2.5 Bathymetry Profiling ............................................................................................................. 23

3 Shallow Water Bathymetric Survey Considerations ..................................................................... 24 3.1 Project Considerations .......................................................................................................... 24

3.1.1 Extent and Internal Coverage ....................................................................................... 24 3.1.2 Accuracy, Object Detection and Point Spacing ............................................................. 26 3.1.3 Vertical Datums ............................................................................................................. 29 3.1.4 Budget and Timelines .................................................................................................... 29 3.1.5 Supplementary Datasets ............................................................................................... 31

3.2 Environmental Considerations .............................................................................................. 32

3.2.1 Minimum, Maximum and Average Depths ................................................................... 32 3.2.2 Turbidity Impacts and Temporal Variations .................................................................. 34 3.2.3 Sea State and Seabed Bottom Type .............................................................................. 35 3.2.4 Environmental Changes - Tide, Water Flow, Seasons, Wind and Daylight ................... 37

3.3 Project and Environment Considerations Conclusion .......................................................... 39

4 LiDAR Bathymetry Survey Strategies ............................................................................................ 40 4.1 Elevation Acquisition of a Turbid Bay Using Multiple Technologies ..................................... 40

4.2 A Shallow Water Dataset Surveyed by Multiple MBES Sensors ........................................... 43

4.3 Progressive Statewide Bathymetry in Critical Areas ............................................................. 44

4.4 One Capture Using Both Topographic and Bathymetric LiDAR ............................................ 47

4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors ......................... 48

5 Future Bathymetry Survey Strategy .............................................................................................. 50 5.1 Recommendations For a Queensland Large Area, Near-Shore Bathymetry Survey ............ 50


www.crcsi.com.au 7

6 Bathymetric LiDAR Acquisition Specifications .............................................................................. 58 6.1 Specification Development ................................................................................................... 58

7 References .................................................................................................................................... 60

Appendix A - Summary of Current LiDAR Sensors ................................................................................ 63 Fugro LADS Mk 3 Bathymetric LiDAR ............................................................................................ 63 Optech SHOALS 3000, CZMIL and ALTM Aquarius Bathymetric LiDAR ........................................ 63 AHAB Hawk Eye II and Chiroptera Bathymetric LiDAR ................................................................. 66 Riegl VQ-820-G Bathymetric LiDAR ............................................................................................... 68

Appendix B - Bathymetric LiDAR Specifications .................................................................................... 69


www.crcsi.com.au 8

1 Introduction

1.1 Background Bathymetric data has traditionally been acquired by hydrographers for nautical charting purposes, and it continues to be acquired predominantly by maritime vessels. In recent years, near-shore bathymetry has been used within an array of new applications. Near-shore bathymetry, along with river, lake and estuarine bathymetry, has rapidly been gaining importance, particularly in regards to coastal hazard and habitat conservation.

Significant progress was made during the early 1990s in near-shore bathymetric surveying with the development of airborne bathymetric Light Detection and Ranging (LiDAR) systems. These systems allow for the rapid acquisition of shallow water (<30-50m depth) bathymetry. Airborne surveys provide a natural complement to maritime vessel surveys, with the former being suited to clear, shallow water and the latter being largely restricted to deeper waters, due to the physical limitations of safely and efficiently operating maritime vessels in shallow water.

Bathymetric LiDAR is increasingly being collected to derive elevation for applications such as storm surge modelling, coastal inundation and vulnerability assessments. Bathymetry also supports less obvious tasks such as marine habitat classification, which takes advantage of the LiDAR pulse intensity. The addition of the water column makes bathymetric LiDAR systems more vulnerable than their topographic counterparts to the adverse impacts of environmental effects, which can lead to data gaps, reduced data coverage and quality. For example, measurements tend to fail when water clarity is poor (e.g. high turbidity), which is often the case in the critical inter-tidal zone due to suspended sediments and breaking waves.

The Cooperative Research Centre for Spatial Information (CRCSI) and Department of Climate Change and Energy Efficiency (DCCEE) commissioned this report to inform the growing interest and investment in bathymetric LiDAR surveys around Australia. Recently, the Victorian and Western Australian Governments have undertaken significant near-shore bathymetric surveys. New South Wales and Queensland have started with smaller operations, with the desire to expand into large surveys in upcoming years.

This report highlights the importance of choosing the most suitable bathymetric LiDAR sensor for a survey and its associated environment. It also recognises the limitations of existing bathymetric LiDAR systems and therefore provides a review of alternative and supporting acquisition technologies. The report takes into consideration different project requirements, coastal environments, and suggests possible multi-sensor survey strategies.

This report is the second stage to the bathymetry research within the Urban Digital Elevation Modelling (UDEM) project. The first stage of the research identified a list of bathymetry users in Australia and New Zealand, and their respective bathymetric needs and challenges. This research provides an overview of bathymetry acquisition alternatives to ensure that the most suitable technology is employed. Example projects and bathymetric LiDAR specifications are provided to assist in the planning of future bathymetric LiDAR surveys.

This report is intended to be used to guide future shallow water bathymetry collection in Australia.


www.crcsi.com.au 9

1.2 Aim and Outcomes The aim of the reported investigation has been to evaluate the applicability of alternative bathymetric mapping technologies to supplement, or in some cases replace, bathymetric LiDAR surveys.

To address the above aim, this report includes:

- A review of the currently available bathymetric LiDAR sensors. Each bathymetric LiDAR sensor has unique characteristics. The review focuses on the advantages and disadvantages of each sensor. (Chapter 2 and Appendix A)

- A detailed listing of alternative bathymetric acquisition techniques. In light of the limitations of bathymetric LiDAR, the advantages of each technique as an alternative or supplement to bathymetric LiDAR are presented. The details listed for each technique include key features, availability, cost, environmental suitability, accuracy, efficiency and effectiveness. (Chapter 2)

- A list of environmental and project features which impact upon the technology/sensor selection, as well as upon design and strategy for bathymetry acquisition. (Chapter 3)

- An outline of multi-sensor bathymetry acquisition projects from Australia, New Zealand and overseas. The positive lessons learnt from each project are provided to guide similar projects in the Australian region. (Chapter 4)

- A strategy and recommendations for a large area, near-shore bathymetric survey. The recommended survey strategy is exemplified via a project within Queensland, Australia. (Chapter 5)

- Specification standards for the collection of bathymetric LiDAR. (Chapter 6 and Appendix B)


www.crcsi.com.au 10

2 Bathymetric Acquisition Technologies The near-shore environment impacts each bathymetric survey technology in different ways. It is important to understand each technology and its nuances so the most suitable technology is engaged to optimise a survey. Each technology has different sensors available, with unique characteristics that can impact upon a survey. This chapter provides a brief description of each bathymetry acquisition technology. The review analyses each sensor's suitability to near-shore surveys, and as a supplement to bathymetric LiDAR.

2.1 LiDAR Bathymetry Bathymetric LiDAR (or bathy LiDAR) has been gathered since the development of combined Global Positioning System (GPS) and Inertial Measurement Unit (IMU) systems in the early 1990s. For a number of years there were two main sensors, Fugro LADS and Optech SHOALS, used for bathy LiDAR surveys. Since then, Airborne Hydrography AB (AHAB) and the National Aeronautics and Space Administration (NASA) have established bathy LiDAR sensors, and more recently Riegl have entered the market. All companies have released new sensors in the past year, with a number of new developments attached to each sensor. This chapter provides an overview of each bathy LiDAR sensor, and highlights some of the technical and operational differences between the various sensors.

All LiDAR systems operate on the principle of measuring the time a laser pulse takes to travel from a transmitter to a surface and back to the receiver. Airborne LiDAR systems require a differential GPS (DGPS) or Precise Point Positioning (PPP) solution to position the aircraft before computing the location of the measured surface. In addition an IMU is used to orient the aircraft and sensor.

Airborne LiDAR systems have been typically divided into topographic LiDAR (topo LiDAR) and bathy LiDAR systems. However, in the past year new LiDAR systems have been developed which bridge the divide between topo and bathy LiDAR systems. These new systems, which efficiently measure both topographic and bathymetric elevations, are termed topo-bathy LiDAR systems. This report focuses on bathymetry acquisition, and will therefore include both bathy LiDAR and topo-bathy LiDAR systems in its review.

Bathy LiDAR systems acquire elevations using a high-powered green (532nm) laser. The 532nm wavelength is optimal for penetrating the water column and it provides the best chance of measuring the seafloor. The wavelength is the crucial difference between bathy LiDAR and topo LiDAR systems, as the latter use an infra-red wavelength of 1064nm, which is unable to penetrate water bodies. Some bathy LiDAR systems use an infra-red wavelength to determine the height of the water surface. However, this is not the case in all systems.

All current bathy LiDAR systems can measure both topographic data and bathymetry. The topographic data gathered from a bathy LiDAR system tends to be of a lower quality and density compared to that provided by topo LiDAR. The depth range of bathy LiDAR systems is typically 2-3 times the Secchi Depth, which is a measure of the transparency of the water and is related to turbidity. To obtain the Secchi Depth a small black and white patterned disk is lowered into the water and when it can no longer be seen, the depth is recorded. The bathy LiDAR depth limit is typically 25-40m in Australian coastal waters, however in clear waters deeper observations have been acquired.


www.crcsi.com.au 11

Figure 1 - Airborne bathy LiDAR diagram depicting a measurement of the seafloor using a 532nm green laser.

Traditionally, advances in bathy LiDAR involved increases in laser power to obtain greater depth penetration. This has limitations due to the maintenance of eye-safe operation. As the power of the laser pulse is increased, likewise the laser footprint is increased to maintain eye safety. This requirement limits the advantages of endlessly increasing the laser power for bathymetric surveys.

In the past couple of years there has been a shift away from high laser power, towards lower laser power, narrower transmitted beams, more frequent measurements and a smaller receiver field-of-view (FOV) (Dewberry 2012). These changes have resulted in topo-bathy LiDAR systems that are similar to topo LiDAR due to their power requirements and footprint diameter, except that they use a green laser. These topo-bathy LiDAR systems are primarily focussed on acquiring topographic data and bathymetry in and surrounding very shallow coastal waters, rivers and lakes.

The smaller footprint topo-bathy LiDAR sensors have a shorter laser pulse and a narrow FOV, which are beneficial in coastal submerged environments in the determination of topographic data under short vegetation (USGS 2012). The small receiver FOV rejects ambient light and scattered photons from the water column and bottom-reflected backscatter (Feygels et al. 2003), thereby ensuring a higher contrast for the detection of the bottom return signal.

Table 1 provides a summary of the currently available bathy and topo-bathy LiDAR sensors. Older systems still exist, however these have been superseded by at least one of the sensors listed in the table.


www.crcsi.com.au 12

Fugro LADS Mk3

Optech SHOALS 3000

Optech SHOALS 1000T

Optech CZMIL

Optech ALTM Aquarius

AHAB HawkEye IIB

AHAB Chiroptera

Riegl VQ-820-G

Typical Sensor Environment

Bathy Bathy Bathy Topo-Bathy Topo-Bathy Bathy Topo-Bathy Topo-Bathy

Origin Australia Canada Canada Canada Canada Sweden Sweden Austria Year Released 2011 2010 2005 2011 2011 2009 2012 2011 Laser Wavelength/s

Green 532nm

Green 532nm Infra-Red 1064nm



Green 532nm



Green 532nm

Laser Energy Per Pulse

7mJ@532 4mJ@532

4mJ@532

3mJ@532 0.1mJ@532 3mJ@532 0.1 mJ@532 0.02mJ @532

Measurement Frequency

1.5kHz@532 3kHz@532 1kHz@532 10kHz@532 70kHz @1064

33-70kHz @532

4kHz@532 128kHz @1064

36kHz@532 400kHz @1064

200kHz @532

Nominal Laser Footprint @ Water Surface

3m@532 2m@532 2m@532 2m@532 0.3-0.6m @532

4.m@532 0.5m@1064

1.5m@532 0.2m@1064

0.6m@532

Nominal Flying Height

400-700m AGL

300-400m AGL

300-400m AGL

Nominal 400m AGL

300-600m AGL

250-500m AGL

250-600m AGL

Nominal 600m AGL

Swath Width (as a function of point spacing or altitude)

585m @8x5m 360m @5x5m 125m @2.5x2.5m

160m @2x2m 300m @3x3m

60m @2x2m 130m @3x3m

291m up to 0.93 x AGL

160m-260m @400m AGL 100m @250m AGL

300m @400m AGL

400m

Typical Point Spacings

2x2m -8x5m 2x2m - 5x5m 2x2m - 5x5m 2x2m 0.4x0.4m - 1x1m

0.5x0.5m - 3.5x3.5m

0.4x0.4m - 1 x 1m

0.2x0.2m - 0.8x0.8m

Minimum Water Depth

~0.5m ~0.3 m ~0.3 m ~0.1 m ~0.1 m ~0.5 m ~0.1 m ~0.1 m

Typical Maximum Water Depth

~60m 2.5-3 x Secchi depth

~50m 2-2.5 x Secchi depth

~50m 2-2.5 x Secchi depth

~50m 2.5-3 x Secchi depth

~20m 1 x Secchi depth

~40m 2-3 x Secchi depth



Commercial Opportunities in Australia and Pacific

Fugro LADS operated

No Commercial Operations

SHOALS 1000T was used in Northern Australia by Fugro Pelagos

None Currently

ALTM Gemini operators - AAM Pty Ltd and Photo-Mapping Services

AAM Pty Ltd - Pelydryn operated. BLOM has operated in Victoria.

None Currently

Fugro LADS operated

Table 1 - Summary of the bathy and topo-bathy LiDAR systems. The laser wavelengths 532nm and 1064nm are the green bathymetry laser and infra-red topographic/sea surface laser respectively. Specifications supplied by Fugro LADS,

Optech, AHAB and Riegl. Note: The NASA EAARL-B topo-bathy LiDAR sensor specifications were not available at the time of publication.

Each bathy and topo-bathy LiDAR sensor exhibit characteristics which impact upon bathymetry capture. All of the sensors have unique capabilities which provide advantages and disadvantages depending upon the survey requirements. Appendix A - Summary of Current LiDAR Sensors presents each of the LiDAR sensors shown in Table 1 in more detail.


www.crcsi.com.au 13

2.2 Satellite and Airborne Derived Bathymetry 2.2.1 Satellite and Aerial Imagery Satellite imagery provides bathymetry estimates which are not reliable enough to be used for navigation purposes. However, they do provide a cost effective option for bathymetry over large areas. These products are suitable for a range of environmental and scientific applications.

The same techniques are used to extract bathymetry from aerial imagery as from high resolution satellite imagery. The most significant difference between these two sources of data is the spatial resolution of the imagery. Typically aerial imagery is a higher resolution than satellite derived imagery. This higher resolution also equates to higher acquisition costs and restricted coverage. As such, more research has been conducted into satellite image derived bathymetry which does not have any mobilisation requirements.

Imagery derived bathymetry is not directly measured, it is inferred, and as such the bathymetry is estimated, with a lower accuracy than LiDAR or multibeam echo sounders. Satellite derived bathymetry is used by a number of research organisations and private companies. The depths to which the imagery is used is limited by light attenuation. Depending on water clarity, depths derived from aerial or satellite imagery are limited to 25-30m because of light penetration issues (Collet et al., 2000).

Deriving near-shore bathymetry from imagery is based on the principle that different wavelengths of light are attenuated by water to differing degrees depending upon the water depth. This principle is demonstrated practically by inferring lighter areas in the image to shallower water. The bathymetry is derived by establishing the relationship between water depth and pixel values. This relationship can be established by two main approaches (Collet et al., 2000). The first is a physical approach which attempts to take into account the parameters of the physical process affecting pixel values. The second is an empirical approach which requires points of known bathymetry for the purposes of image calibration. The physical approach requires data related to water composition, nature of the sea bottom, and atmospheric conditions amongst others, whereas the empirical approach requires seafloor classification and measured water depths for control (Collet et al., 2000). For the empirical approach, the accuracy of the depth control data will be reflected in the heights derived from the imagery.

As with bathy LiDAR, there are various camera sensors available to derive bathymetry from satellite imagery. Table 2 lists some of the satellites from which bathymetry can be derived. Each of these satellites has characteristics which impact upon the imagery's suitability for deriving bathymetry. The number of wavelengths (or bands), especially in the blue-green spectrum enables better penetration of the water column and typically more accurate bathymetry. The satellite processing company GRAS has used WorldView2 imagery to compute bathymetry. The additional wavelengths available in the blue-green spectrum with WorldView2 have enabled a depth penetration of 10-15m deeper than from other satellites (GRAS 2011).

Table 2 provides a list of satellites and parameters which affect the suitability of a sensor to derive bathymetry. In the table, the spectral resolution is the number of wavelengths measured by the sensor, which in turn effects its ability to distinguish between different surface types. The radiometric resolution is the sensor's ability to discriminate small differences in the area that corresponds to a single raster pixel. The spatial resolution of the sensor effects the amount of bathymetric detail that can be derived from within the image. The swath width relates to the


www.crcsi.com.au 14

coverage efficiency of each sensor. The higher resolution satellites tend to have a smaller swath. The cost of each satellite's imagery varies, and depending upon the application and budget, the various factors in Table 2 will determine which satellite's imagery is most suitable for a project. A guide for the imagery costs are provided by Geoimage per scene or km2, and are dependent upon whether the imagery is archived or a new capture. The provider supplied metric accuracy of the unrectified orthoimagery is also listed. The only pertinent factor not provided which may influence the sensor choice is the signal-to-noise ratio. Each sensor should have the signal-to-noise ratio calculated on a scene-by-scene basis, therefore it is not relevant to list as a satellite parameter as it is variable.

Spectral Resolution

Radiometric Resolution

Spatial Resolution

Swath Width

Cost Per...

Cost (Archive)

Cost (New)

Metric Accuracy

ENVISAT MERIS 15 12 bits 300m 1100km Scene Mid N/A 70-225m MODIS 36 12 bits 250m 2330km Scene Low - 30-50m LANDSAT7 ETM+ 7 8 bits 30m 180km Scene Low N/A 25-250m EO1 HYPERION 220 12 bits 30m 7.75km Scene Mid Mid 15-25m ALOS AVNIR-2 4 8 bits 10m 70km Scene Low N/A 20m SPOT 5 4 8 bits 10m 60km Scene High High 15-25m IKONOS 4 11 bits 4m 11.3km Km2 Low Low 15m QUICKBIRD 4 11 bits 2.6m 18km Km2 Mid Low 17-23m PLEIADES 4 11 bits 2m 20km Km2 Mid Low 4.5-7.5m GEOEYE1 4 11 bits 2m 15.5km Km2 Mid Mid 3.5-6.5m WORLDVIEW2 8 11 bits 2m 17.7km Km2 High High 3.5-6.5m

Table 2 - Satellite imagery parameter comparison. Costs per scene per scene - Low $0-$1000, Mid $1000-$5000, High Above $5000. Costs per km2 (archive) - Low $10-$12, Mid $12-20, High Above $20. Costs per km2 (new capture) - Low

$20-$25, Mid 25-$30, High Above $30. Costs are based on imagery costs, no processing applied. Accuracies are based on vendor supplied information. Data provided by Geoimage.

MERIS, MODIS, Landsat and Hyperion have a low spatial resolution and should only be used for regional or national bathymetry derivation. The detail provided by these products is coarse.

WorldView2 is the most suitable for deriving detailed near-shore bathymetry as it contains more bands and a high resolution, however it is one of the most expensive sources of imagery, limiting its usability. WorldView2 alleviates some of the concerns over spatial versus spectral resolution identified by researchers (Mumby and Edwards 2002, Capolsini et al. 2003 and Hochberg et al. 2003). As seen in Table 2, till WorldView2 was launched the higher spatial resolution satellites contained less spectral bands than the lower resolution satellites. This resulted in a trade-off when choosing a satellite other than WorldView2. If cost and coverage are factors WorldView2 may not be an option for a project. In these cases a sensor such as Landsat or SPOT may provide the optimal balance for spatial and spectral resolution depending upon the budget, and mission requirements.

Satellite derived bathymetry has a significant cost and acquisition advantage over maritime vessel or airborne platforms. Much less time and costs are required to derive bathymetry for substantially larger areas with accompanying resolution and accuracy tradeoffs. The efficiency, cost and area covered by each satellite differs depending upon the satellite's parameters, particularly its swath. As an example Sagar and Wettle have derived bathymetry from the ALOS Advanced Visible and Near Infrared Radiometer (AVNIR-2) sensor. With a swath width of 70km, each scene covers 4,900 km2 (Sagar and Wettle 2010). To survey the same area with bathy LiDAR would take at least five months of airborne survey and ten months of data processing. For an area such as the Great Barrier Reef which covers more than 340,000km2 satellite derived bathymetry provides a natural complement to maritime and airborne bathymetry platforms in less critical areas.


www.crcsi.com.au 15

A comparison of satellite derived bathymetry with bathy LiDAR was conducted by CSIRO on multispectral WorldView2 data over the Marmion Marine Park (just north of Perth, Australia). This is but one example of analysing WorldView2 for bathymetry derivation. Table 3 summarises the results from the report published by Campbell et al. (2012) which analysed the relationship between the LiDAR depth values and the WorldView2 bathymetry derived using the Lyzenga algorithm (as described in Table 4). The R2 value is the coefficient of determination which represents how well the derived bathymetry matches the LiDAR data. The RMSE represents the accuracy of the derived bathymetry.

Coastal Blue

Blue Green Yellow R2 RMSE

0.45 2.88 0.12 3.64 0.30 3.24 0.40 3.00 0.30 3.26 0.35 3.12 0.41 3.00 0.45 2.89

Table 3 - Four WorldView2 wavelengths and their significance to deriving bathymetry in the Campbell et al 2012 study. The correlation is shown as the R2 value and the root mean square error (RMSE) is provided.

The combination of the blue, green and yellow bands provided the optimal solution for deriving the bathymetry using the Lyzenga algorithm in Marmion Marin Park. In this study the coastal blue band added very little to the regression analysis. The blue and green bands accounted for most of the variation, and with the yellow band adding very little to the correlation.

In the Campbell et al. (2012) study three seafloor bottom types were identified from the bathy LiDAR reflectance data: sand, mixed bottom and reef. The satellite derived bathymetry in sand showed the strongest R2 correlation (0.53) with the bathy LiDAR, and an RMSE 2.31m. The mixed bottom showed a lower R2 correlation (0.35), with an RMSE of 2.99m. The reef showed poorer results with a R2 correlation of 0.11 and RMSE 3.50m. The relationship for the sand areas is better than the mixed areas, while the predicted depths appear to be poorly correlated for the reef areas.

This study demonstrates that the accuracy and quality of satellite derived bathymetry is not only dependent on depth, but also on the seafloor composition. The study also provides an example of the contribution of each wavelength to the WorldView2 imagery derived bathymetry. The optimal band combination is important for the efficiency of deriving bathymetry from imagery. The contribution of each band to the bathymetry will vary depending upon the water conditions.

To summarise, satellite imagery provides bathymetry estimates which are not intended for navigation purposes. In areas where sediment concentration is too high for optical data to provide results, alternative technologies, such as Synthetic Aperture Radar (SAR) data can be used to indicate shallow water shoal locations. This is addressed further in section 2.2.4 Satellite Radar.


www.crcsi.com.au 16

2.2.2 Hyperspectral Imagery Hyperspectral imagery is more complex than multispectral imagery. The derivation of bathymetry from hyperspectral imagery is still in its infancy. The increased number of spectral bands used by hyperspectral sensors enables the discrimination between different components of the water column and seabed, however this extra complexity has generally restricted its usage to the research sector. The additional spectral bands enable a more accurate estimation of water depth and bottom type than is possible from the multispectral sensors discussed in 2.2.1 Satellite and Aerial Imagery. Although, hyperspectral imagery is still primarily acquired using airborne acquisition and so does not have the advantages associated with satellite imagery.

The first hyperspectral sensor in space, Hyperion, has been used for mapping shallow water benthic habitat (Kutser and Jupp, 2002). Hyperion on Earth Observing One (EO-1) was designed as a technical demonstration instrument with a short lifetime. It has already collected data for a longer period than was originally planned with the mission being close to an end. Airborne hyperspectral instruments can also provide the spectral and spatial resolutions needed for deriving bathymetry. However, the cost of acquisition is higher, to the point of limiting the usage of hyperspectral airborne imagery in mapping large coastal areas.

In 2008 a hyperspectral sensing survey using the HyMap system was conducted in Jervis Bay, located about 180km south of Sydney, Australia. HyMap is an airborne hyperspectral remote sensing instrument that collects 126 spectral bands from the visible to the shortwave infrared wavelength regions (0.45 to 2.5um). Like multispectral imagery the HyMap radiance data needs to be atmospherically corrected to surface reflectance. Following surface sun glint corrections, the remote sensing reflectance is transformed to subsurface reflectance (Jing 2010). The HyMap results demonstrated that water depths up to 20m can be extracted from hyperspectral imagery and showed a good correlation with bathymetry from conventional hydrographical surveys (Jing 2010).

Increased developments and launches of satellite-based hyperspectral sensors will make this acquisition technique more feasible for large area bathymetry processing. As an airborne technique, the advantages for deriving bathymetry from hyperspectral sensors are limited compared to other available airborne technologies. There are no easily identifiable depth, coverage or environmental advantages to airborne hyperspectral imagery derived bathymetry over bathy LiDAR.

2.2.3 Algorithms to Derive Bathymetry from Imagery As bathymetry derived from imagery is estimated rather than directly measured, it is worth discussing the influence and impact of the algorithms used to model depths. A number of different algorithms have been derived to determine bathymetry from imagery. For hyperspectral and multispectral imagery-erived bathymetry, it is just as important to select the most appropriate algorithm, as it is to select the most suitable image sensor.

Mobley (2009) compared six common algorithms used to derive bathymetry, following a workshop held in Brisbane, Australia in February 2009. The tests were sponsored by the U.S. Office of Naval Research (ONR-Global) and the Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO). Participants compared the results obtained by applying algorithms to a common set of images.


www.crcsi.com.au 17

The six algorithms were applied to two different airborne hyperspectral images, one from a PHILLS sensor flown near Lee Stocking Island (LSI) in the Bahamas and one from a CASI sensor flown over Moreton Bay (MB) in eastern Australia. Each investigator applied their algorithm to both images and sent the results to an independent third party (S. Phinn at the University of Queensland) for comparison with the ground-truth measurements at each site (Mobley 2009). A summary of the algorithms and test results is provided in Table 4. Even though the tests were performed on hyperspectral imagery the results are still relevant to multispectral imagery-derived bathymetry.

Algorithm Algorithm Description Area Accuracy to LiDAR Bathymetry

Processor Speed

Pixels Processed Per Second

HOPE (Hyper-spectral Optimization Process Exemplar)

This method is an implementation of the semi-analytical, non-linear search algorithm developed by Lee et al. (1998, 1999). The model retrieves five parameters: phytoplankton absorption at 440 nm, CDOM absorption at 440 nm, particulate backscatter at 550 nm, bottom reflectance at 550 nm, and bottom depth.

LSI RMSE = 1.12 R2=0.85

2.66GHz 156.39

MB RMSE = 3.17 R2=0.78

2.66GHz 157.01

BRUCE (Bottom Reflectance Un-mixing Computation of the Environment)

This inversion technique incorporates the HOPE model with a modification to the bottom reflectance parameterization (Klonowski et al. 2007).

LSI RMSE = 0.86 R2=0.91

2.40GHz 10.43

MB RMSE = 2.11 R2=0.80

2.40GHz 15.70

SAMBUCA (Semi-Analytical Model for Bathymetry, Un-mixing, and Concentration Assessment)

SAMBUCA is an implementation of the HOPE algorithm, modified to (1) retrieve water-column concentrations of chlorophyll-a, CDOM, and non-algal-particles, (2) account for more than one substratum cover type, and (3) estimate the contribution of the substratum to the remote sensing signal (Brando et al. 2009).

LSI RMSE = 1.30 R2=0.85

16 process-ors

0.11

MB RMSE = 0.96 R2=0.91

16 process-ors

0.38

SMLUT (Spectrum-Matching and Look-Up-Table)

This method is based on spectrum matching by searching through a pre-computed database of remote-sensing reflectance spectra, as described in Mobley et al. (2005).

LSI RMSE = 1.14 R2=0.88

2.00GHz 326.38

MB RMSE = 4.71 R2=0.38

2.00GHz 642.32

ALLUT (Adaptive Linearized Look-Up Trees)

This algorithm facilitates spectrum-matching inversion of any radiative transfer model parameterized by a set of real-valued and integer parameters. The method used here is identical to that described in Hedley et al. (2009), but in addition includes a local linear gradient calculation.

LSI RMSE = 2.36 R2=0.81

3.00GHz 61.53

MB RMSE = 2.24 R2=0.78

3.00GHz 92.97

LYZENGA - Common empirical, multi-spectral technique

This empirical, multi-spectral technique developed by Lyzenga (1978) can retrieve bathymetry in areas of constant water clarity and homogenous benthos/substrate composition. Although limited to retrieval of bathymetry under restrictive environmental conditions and therefore much less general than the above techniques, bathymetry retrieved by the Lyzenga algorithm was included in this comparison study because of its historical importance and continued widespread use under certain conditions.

LSI RMSE = 1.68 R2=0.72

- -

MB RMSE = 3.12 R2=0.65

- -

Table 4 - Results from a test which compared six different algorithms used to derived bathymetry from imagery. The tests were performed on two datasets near Lee Stocking Island (LSI) in the Bahamas and Moreton Bay (MB) in Australia

(Mobley 2009).


www.crcsi.com.au 18

Lyzenga is a generic empirical algorithm as opposed to the other five algorithms which are non-empirical. HOPE, BRUCE and SAMBUCA are semi-analytical approaches which use a specific code to derive the bathymetry. SMLUT and ALLUT are non-empirical, look-up table approaches to deriving bathymetry. Of the six algorithms in Table 4 SAMBUCA had the highest correlation with the bathy LiDAR and lowest RMSE. However, it was also the slowest to derive the bathymetry.

It is important to note that the algorithms and processes to derive bathymetry from imagery are not limited to the six provided in the study. The study is provided to demonstrate the variability in algorithms to derive bathymetry from imagery, and the importance of using an algorithm which is the most suitable to the imagery, project area, timeframes and budget. Technical expertise and experience is required to select the most appropriate processing technique to derive bathymetry from imagery.

2.2.4 Satellite Radar Similar to multispectral and hyperspectral imagery which infer depth, radar does not directly measure bathymetry; it infers depths from changes in the sea surface. This enables radar to provide a potential solution in turbid environments where other remote sensing techniques are unsuccessful. It also has the benefit of being unaffected by cloud cover. Radar produces relative bathymetry, rather than absolute depths. The technique is particularly suited to areas of sandbanks and shoals where there are often frequent or continuous changes in bathymetry.

Radar bathymetry determination is based on its ability to measure the change in height and roughness of the sea surface. The roughness changes as tidal currents approach a shoal, as they have to accelerate to flow over the shoal. As shown in Figure 2, this effect results in a zone of divergence of the surface currents on the side of the shoal facing the current, which in turn leads to a reduction of the surface roughness and a darker tone on the radar image. The rougher water enhances the radar backscatter giving a brighter zone on the radar image (Huang et al., 2001; Robinson, 2004).

Figure 2 - Technique to derive satellite radar bathymetry (Quadros 2009)

Practical implementation of this form of bathymetric measurement requires knowledge of the tidal currents and the wind, as the wind speed and direction affects the roughness modulation. There are several uncertainties inherent in the measurement and manipulation of satellite radar altimetry observations used to estimate ocean depth. These make radar derived bathymetry notoriously difficult to determine and inherently unreliable compared to other technologies. Radar is one of the least frequently used technologies employed to determine near-shore bathymetry and the technique is not currently reliable enough to be used as a supplementary technology in bathymetry gaps caused by turbidity.


www.crcsi.com.au 19

2.3 Maritime Vessel Bathymetry 2.3.1 Multi-Beam and Single-Beam Echo Sounders Multi-beam echo sounders (MBES) are the most efficient and reliable sensor to gather bathymetry from a maritime vessel platform. Single-beam echo sounders (SBES) only generate a single pulse at nadir and do not have the swath coverage of MBES. Notably, SBES can provide spot heights to calibrate other bathymetry acquisition techniques such as satellite derived bathymetry. However, the focus of this section is on MBES due to its efficiency and reliability.

MBES systems are operated by a number of commercial, government and research institutions. As shown in Figure 3, MBES sensors produce measurements either side of a vessel. The high-performance systems have wide-angle swaths that cover an area up to seven times the water depth; although it is more typically twice the water depth (Ozcoasts 2012). As the water depth decreases the swath width also decreases, and the coverage efficiency of the survey is reduced. In order to generate a complete coverage in shallower water, the survey lines need to be closer together.

MBES have the advantage of producing a higher measurement density than airborne sensors. This results in detailed measurements of features which span more than a decimetre. The increased point density makes MBES more suitable to small object detection than airborne sensing. It is standard to record backscatter (reflected energy) from MBES surveys, which is similar to the LiDAR reflectivity/intensity, as a measurement of the strength of the return signal.

The accuracy of MBES systems is dependent on the correction applied for the vessel and sensor position and motion, especially in higher sea states. The success of the MBES solution is dependent not only on the MBES sensor, but also on the inertial sensor (vessel motion), DGPS sensor (vessel position), acoustic refraction correction (speed of sound in the water column) and operator (management and processing). Errors in any of these factors will be reflected in the bathymetry.

Figure 3 - Diagram with airborne bathy LiDAR and maritime vessel MBES

Comparisons between currently available systems have not been widely published. MBES systems comprise the MBES sensor, an inertial sensor for vessel orientation and a GNSS receiver for positioning the vessel. Hydro International has published a comparison of technical specifications for 23 MBES sensors1, 21 inertial motion sensors2 and 61 GNSS receivers3. The MBES sensors listed by 1 http://www.hydro-international.com/productsurvey/id28-Multibeam_Echosounders,_JulyAugust.html 2 http://www.hydro-international.com/productsurvey/id31-Inertial_Navigation_Systems_INS,_JanuaryFebruary.html


www.crcsi.com.au 20

Hydro International are those most widely used for depths of less than 500m. The inertial motion sensors and GNSS receivers are manufactured by a large number of companies, with Hydro International only providing a selection of those on the market.

Each manufacturer offers a range of MBES models at different costs and specifications. For a bathymetric survey the MBES beam spacing and frequency need to be considered. The depth of operations also needs to be considered, with some newer units having multi-frequency options which are useful across a range of depths.

When commissioning a MBES survey the data acquisition quality should be specified in line with IHO (International Hydrographic Organisation) standards. For multi-use purposes, backscatter data should always be logged, processed and delivered. Different MBES sensors produce different quality backscatter results, which may need to be corrected for several parameters.

Some considerations for commissioning a MBES survey include:

• The vessel plan with sensor offsets to indicate the quality of measurements.

• The processing of backscatter data with corrections.

• The expected percentage overlap between run lines.

• The proposed survey speed to indicate the density of data along track.

• Accessibility and vessel safety

• The required IHO standard for data quality.

• The vertical positioning corrections applied to the GNSS positions, and whether corrections are made using tide gauge measurements or modelled tides.

• A description of the frequency at which sound velocity corrections are measured or cast.

• The vessel capacity in regards to the daily/weekly schedule for acquisition. Larger vessels will be able to operate for longer periods, however the draught will generally be greater resulting in less accessibility to the very near-shore.

MBES surveys can generate bathymetry to the same or better resolution and accuracy as bathy LiDAR. These factors ease the MBES data integration process to produce a seamless, multi-source elevation model. Within a large area, near-shore survey a MBES is ideally used following the bathy LiDAR acquisition to provide measurements within the LiDAR data gaps. The MBES is able to penetrate gaps within the bathy LiDAR data caused by turbidity, seafloor absorption and maximum depth limitations. The most significant limitation of a MBES survey to filling LiDAR gaps is vessel access. Therefore a MBES would be unable to fill-in very shallow (<2m) depths and gaps in hazardous areas.

3 http://www.hydro-international.com/productsurvey/id32-GNSS_Receivers,_MarchApril.html


www.crcsi.com.au 21

2.3.2 Side-Scan Sonar and Sub-bottom Profilers Traditionally, side-scan sonar and sub-bottom profilers have not been used for gathering bathymetry. Side-scan sonar has been primarily used to detect small, off-nadir bottom objects, and sub-bottom profilers have been used to measure the depths of geomorphic layers beneath the seafloor.

Side-scan sonar uses oblique acoustic images to provide a wide coverage of the seafloor. By "looking across" the seafloor, protrusions are easily identified. A recent development has provided an exception for using side-scan sonar to generate bathymetry. Interferometric side-scan sonar (ISSS) uses the phase measurement at multiple receivers to determine the angle from which the acoustic return originates. This angle of origin, in combination with the range, provides bathymetry (NOAA 2012). ISSS is still not widely used, however it can be compared MBES due to its swath coverage.

ISSS has the advantage of a greater swath width than MBES in shallow waters less than 10m. The increased swath enables higher acquisition rates in shallow waters and allows vessels to survey hazardous features from a greater distance. However the two downsides of ISSS include data noise and a lower data density ("gap") at nadir. A 60%-75% overlap is required between adjacent lines for complete coverage of the nadir gaps. If the nadir gap could be eliminated then the efficiency of an entire ISSS operation would increase dramatically (Brisette 2006).

Data noise and artefacts have been an issue with ISSS sensors. However, in comparisons with two MBES datasets, a GeoAcoustics GeoSwath ISSS sensor still met the IHO Order 1 accuracy requirements (see Table 5) (Gostnell 2005). The standard deviations over 5m2 regions of point data in relatively flat sections of seafloor tended to be three to five times higher for the GeoSwath than for the MBES systems. The ISSS datasets are inherently noisy requiring more post-processing and making them difficult to clean (Gostnell 2005). In a separate assessment of the Geoswath ISSS sensor artefacts in the dataset were in the order of 10cm in amplitude (NOAA 2012).

Sub-bottom profilers use a single pulse to measure the seafloor and bathymetry sub-surface. A portion of the incident energy is reflected from the sediment-water interface, whereas the remainder is transmitted deeper into the substrate (McQuillan et al., 1984, Stoker et al., 1997). The profiler records changes in the acoustic impedance between different substrates. Sub-bottom profilers are not efficient at recording bathymetry, as the sensor is primarily designed for sub-surface measurements. The bathymetry retrieved is similar to SBES as it does not have the swath coverage that is produced by MBES and side-scan sonar.

Side-scan sonar and sub-bottom profilers are generally not recommended for large area bathymetry surveys unless there are additional requirements to collect bottom objects or sub-surface data. ISSS could be used in combination with MBES to infill bathy LiDAR gaps. The combination of the two sensors will maximise possible coverage, utilising the coverage advantage of ISSS, with the accuracy and consistent point spacing of the MBES system, although the use of the two sensors has to be weighed against the higher cost.


www.crcsi.com.au 22

2.4 Specialised Technologies 2.4.1 Airborne Electro-Magnetic Bathymetry (AEMB) Airborne LiDAR's limited effectiveness in turbid waters and the surf zone has given rise to alternative techniques such as AEMB. AEMB transmits an electrical current to create a primary electromagnetic field below the aircraft. This electrical current in turn creates a secondary current in the ground below the aircraft. The flow of this secondary current depends on the electrical resistance of the ground and/or water (Vrbancich, 2006). This electrical resistance can be used to compute the depth of the water. The main difficulty with this technology is unscrambling the contributions of water, sediment and ground in the resistance of the secondary current. The contributions of these factors differ significantly along the coast (Vrbancich and Fullagar, 2006; Zollinger et al., 1986). AEMB is has the distinct advantage of being unaffected by turbidity (water clarity) and surf (breaking waves) (Vrbancich, 2006; Wolfgram and Vrbancich, 2006).

Figure 4 - SeaTEM AEMB system with an 18m wingspan. Courtesy of Vrbancich 2012.

In 1982, Morrison and Becker were the first to consider the use of AEM systems for mapping water depths (Vrbancich 2012). Since then, interpreted water depths obtained from AEMB data have resulted in absolute water depth accuracies of 1–2m for depths between 10m and 30m, and 0.3m–0.5m in water shallower than 10m (Vrbancich 2012). A study in Broken Bay, New South Wales (NSW) using the SeaTEM system shown in Figure 4, compared depths derived from AEMB and marine seismic data. This study showed good agreement between the systems, both achieving depths penetrating up to 60–70m below the seafloor. The comparison showed that AEMB agreed to the marine seismic data to within 5–10m (Vrbancich 2012).

The resolution of the AEMB data is determined by the electro-magnetic footprint. The point spacing of the AEMB data is generally at 10m along track and 50m between lines. The elevation models produced by Vrbancich have had a 10m grid cell size.

AEMB is still considered as a research sensor and is not commercially available for bathymetry acquisition. The technique has the potential to supplement bathy LiDAR in turbid areas without the accessibility restrictions affecting maritime vessels.


www.crcsi.com.au 23

2.4.2 Autonomous Underwater Vehicles (AUV) AUVs have been used to collect targeted, high-resolution bathymetry. AUVs provide a tool for collecting datasets which are complementary to more traditional sources. AUVs provide a platform to obtain quantitative information in more detail and beyond the depths accessible by traditional methods. AUVs have the advantages of an increased mapping resolution, simultaneous recording from additional onboard sensors and they are the only solution for some difficult survey areas e.g. deep-water, under ice survey and certain naval applications.

The application of AUVs in large area bathymetry acquisition is limited due to their slow rate of coverage. Within the context of large areas surveys and multi-sensor approaches, they can be used to acquire detailed bathymetry at depth, and access areas which may not be available to maritime vessels. These factors should be considered at the end of a bathymetry survey for areas which may require additional coverage.

2.4.3 Satellite LiDAR The National Aeronautics and Space Administration (NASA) have flown the Multiple Altimeter Beam Experimental LiDAR (MABEL) system at high altitudes as a demonstration prelude to the space borne Advanced Topographic Laser Altimeter System (ATLAS) LiDAR sensor on the ICESat-2 mission scheduled for 2016. These LiDAR sensors use photon counting to produce a point cloud. Unlike conventional LiDAR, which uses millions of photons to make a single distance measurement, photon counting LiDAR precisely records the time of flight of individual photons (NASA 2012).

The MABEL LiDAR sensor operates at both the 532nm (green) and 10634nm (infra-red) wavelengths. The green wavelength is the same as that used by bathy LiDAR sensors except with different parameters, such as a larger, 10m footprint. Although the MABEL (and potentially ATLAS) sensor uses the green wavelength it is primarily used to detect clouds, not the seafloor (Yang 2010). Clouds affect LiDAR measurements through particle forward scattering, which increases the photon-path length and makes the surface appear farther from the satellite (Yang 2010). The detection of clouds enables corrections to be made to the infra-red LiDAR measurements.

The ICESat-2 mission is primarily focussed on land ice, sea ice and vegetation measurements (Birkett 2011). However, surface waters measurements are possible over the ocean, coast and continental waters (Birkett 2011). The recent developments in space borne LiDAR operations may have potential for bathymetric applications. It is worth monitoring future developments in this field as the technology may become more applicable for bathymetry acquisition.

2.5 Bathymetry Profiling Bathymetry profiling is the most localised technique for acquiring bathymetry. Advances in DGPS and echo-sounders have enabled greater complexities and efficiencies for profiling than was traditionally available. Bathymetry profiling is confined to on-ground measurements using DGPS, and can be supplemented with a SBES to obtain depths beyond 2m. The technique is inexpensive for obtaining localised bathymetry.

A number of examples of bathymetry profiling exist, however one particular example from the River Murray Operations Unit (RMOU) of the South Australian Water Corporation summarises a number of different strategies for obtaining bathymetry profiles.

The RMOU requires regular surveys and monitoring at the River Murray Mouth and the Coorong area of South Australia (Heinz Burlik, 2004). The mouth is surveyed to monitor sand movement and


www.crcsi.com.au 24

the survey data is then used to calculate volumes for dredging and beach degradation or build-up. For these surveys the RMOU implemented a variety of different mechanisms to obtain elevations (Heinz Burlik, 2004):

• A backpack fitted with a DGPS was used to either walk or cycle the beach and waterline.

• An aluminium survey boat with echo-sounders and DGPS to survey water depths ranging from 0.5m to a maximum of 5m, averaging about 2m.

• A six wheel-drive amphibious all-terrain vehicle in shallow areas to a depth of 2m. The addition of an echo sounder extends the capabilities beyond 2m.

• A jet-ski was used in the surf zone and the shallow offshore beach areas. These areas were inaccessible to the boat and the amphibious all-terrain vehicle.

• A paddle ski enabled small surveys in very shallow muddy areas, narrow creeks and around in-water structures that are impossible to approach with conventional craft.

These options could be used strategically to calibrate, control and quality assure a large bathymetry survey. Particularly in regard to calibrating satellite derived bathymetry in areas where other sources of bathymetry are impractical or not available.

3 Shallow Water Bathymetric Survey Considerations Understanding bathymetric survey considerations are vital to selecting the optimal acquisition strategy. This section focuses on the various considerations, their impact on the acquisition technologies and survey strategies. The information in this section is slanted more towards large area surveys, particularly those involving LiDAR and multiple technologies, rather than small surveys which would only require a single maritime vessel. Also, the focus is on surveys which are conducted in water depths less than 50m, as the majority of surveys beyond this depth are primarily undertaken with MBES. The considerations for these surveys are slanted towards airborne LiDAR acquisition supplemented with complementary technologies. The considerations are divided into project and environmental considerations. Project considerations are independent of survey location, where as environmental considerations are dependent on the survey location and area.

3.1 Project Considerations 3.1.1 Extent and Internal Coverage The size of the survey area has a major impact on the feasibility of the bathymetry acquisition technologies. The impact the survey extent has on the chosen survey strategy is also partly related to the budget, timelines and survey depths. Satellite and airborne platforms are suited to larger surveys than maritime vessels, as the mobilisation cost of aircraft can be considerably more than that of a vessel. The smaller the survey, the higher will be the percentage of the project cost devoted to mobilisation. If the survey involves depths less than 5m then remote sensing is the only option, unless traditional ground survey techniques are used, as maritime vessels operating an MBES cannot efficiently access depths less than 5m. It should be noted that around ports and harbours terrestrial laser scanners operated from a maritime vessel with MBES may provide a complete seamless coverage above the waterline, however very shallow waters will still be an issue.


www.crcsi.com.au 25

Figure 5 - Seamless bathymetry and topographic data produced by HawkEye II bathy LiDAR. Narrow gaps remain in the very shallow water. Courtesy of Pelydryn.

Survey coverage is the term used to explain the "completeness" of the survey or the number of gaps in the survey. The required coverage will impact on the combination of survey technologies used in large near-shore surveys. In the Western Port Bay survey strategy to be reviewed in Section 4.1, the gaps remaining after the topo and bathy LiDAR surveys were filled in with a MBES vessel survey. This is the most logical sequence of surveys for any large near-shore bathymetry project. Whether aerial imagery or LiDAR is used to derive bathymetry, gaps in the data can be filled by maritime vessel surveys. In areas too shallow for a maritime vessel to approach, a topo-bathy system can be used. If this is unsuccessful, the only way to gather depths would be by traditional means, with a survey pole or amphibious vehicle. If areas cannot be filled by survey, data interpolation can be used, however this is not always suitable, especially if object detection is a priority.

For environmental risk and modelling, 100% coverage is not a requirement. However, large gaps in the survey can cause issues for the modelling if interpolation is not reliable. If problem gaps exist after a bathy LiDAR survey, they should be covered with an MBES survey using the appropriate specifications.

For topo-bathy LiDAR surveys, the performance is still being tested against bathy LiDAR systems. Topo-bathy systems can measure bathymetry in very shallow waters (<5m), but whether this is more reliable than a bathy sensor in moderately turbid waters is yet to be validated. The coverage may differ due to the higher pulse rate, smaller footprint diameter and narrower FOV of the topo-bathy systems. A topo-bathy sensor can measure to depths of around 10m. A decision selecting either a bathy and topo-bathy sensor is dependent upon the depth requirements of the survey, rather than the extent. Both LiDAR systems are suited to large survey extents.

To maximise bathymetry coverage a bathy and topo-bathy system can be used concurrently. However, the added coverage and data should be balanced against the added survey cost. The two LiDAR sensors are complementary if the survey requires bathymetry coverage from the waterline down to depths greater than 10m. An example of the results from the two LiDAR systems is provided in Section 4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors.


www.crcsi.com.au 26

3.1.2 Accuracy, Object Detection and Point Spacing Object detection and point spacing are closely interwoven within a bathymetric survey. Both survey requirements are dependent upon the sensor footprint size and measurement frequency. Accuracy generally refers to the vertical accuracy of each point.

The horizontal accuracy, although relevant, is discussed less frequently. For a well designed bathy LiDAR system the horizontal error of the LiDAR beam at the water surface is not expected to exceed ±0.20m RMSE (Maune 2007). Simulations of beam scattering through the water column have estimated the horizontal error at ±0.32m RMSE per 10m depth (Maune 2007). The magnitudes of these horizontal errors are acceptable for all the IHO standards listed in Table 5, up to the depth limits of bathy LiDAR.

Point spacing refers to the horizontal distance between measurements, whereas object detection relates to the maximum undetected cube size that could exist within a survey. This is essentially a measure of the gap between measurement footprints. Spacing factors are fundamental within all project specifications, however they tend to have less of an impact on the selection of bathymetric acquisition technology than other requirements.

Bathymetry accuracy, object detection and point spacing requirements alone, do not generally provide enough information to discriminate between using LiDAR or a maritime vessel for the survey (unless a point density of more than 1 point per m2 is required). The technology selection will more likely be decided by the environmental conditions and the extent of the survey. However, the main impact these three factors have on a survey is the cost.


www.crcsi.com.au 27

The International Hydrographic Organisation (IHO) sets standards for these three factors within bathymetric surveys for nautical charting. The IHO standards listed in Table 5 can be used to commission a survey.

ORDER Special 1a 1b 2 3 Typical Areas Harbours, berthing

areas, and associated critical channels with minimum under-keel clearances.

Harbours, harbor approach channels, recommended tracks and some coastal areas with depths up to 100m. Where under-keel clearance is less critical but features of concern to surface shipping may exist.

Harbours, harbor approach channels, recommended tracks and some coastal areas with depths up to 100m. Where under-keel clearance is not considered to an issue for the type of shipping.

Areas not described in Special Order and Order 1, or areas up to 200 m water depth, where a general description of the seafloor is considered adequate.

Offshore areas not described in Special Order, and Orders 1 and 2

Horizontal Accuracy (95% Confidence)

2 m 5 m + 5% of depth 5 m + 5% of depth 20 m + 5% of depth 150 m + 5% of depth

Depth accuracy for Reduced Depths

(95% Confidence)(1)

a = 0.25 m b = 0.0075

a = 0.5 m b = 0.013

a = 0.5 m b = 0.013

a = 1.0 m b = 0.023

Same as Order 2

100 % Bottom Search

Compulsory Required in selected areas

May be required in selected areas

May be required in selected areas

Not applicable

Search Requirement

Full seafloor search required

Full seafloor search required

Full seafloor search not required



System Detection Capability

Cubic features > 1 m Cubic features > 2 m in depths up to 40 m; 10% of depth beyond 40 m

Not applicable Not applicable Not applicable

Position of Natural Coastline

10 m at 95% confidence level



Same as Order 1 Same as Order 1

Position of Topographical Features




Same as Order 1 Same as Order 1

Tidal Observations Error not to exceed 5cm at 95% confidence

Error not to exceed 10cm at 95% confidence

Error not to exceed 10 cm at 95% confidence

Same as Order 1 No tides applied in depths > 200 m

Bottom Sampling 10 times line spacing; denser in anchorages

10 times line spacing 10 times line spacing Same as Order 1 No tides applied in depths > 200 m

Table 5 - IHO Bathymetry Survey Standards for Nautical Charting

(1) To calculate the error limits for depth accuracy the corresponding values of a and b listed above have to be introduced into the

formula (a2 + (b *d)2) with: a - constant depth error, i.e. the sum of all constant errors; b*d - depth dependent error; b - factor of depth dependent error; d - depth

For most bathymetry acquisition techniques it is impractical to obtain a vertical accuracy better than 0.5m @ 95% confidence. For surveys with an accuracy requirement of 0.5m @ 95% confidence the only options are airborne LiDAR and/or maritime vessel (including AUV). If the accuracy requirements are relaxed to 2m @ 95% confidence, other remote sensing techniques may be utilised.

For object detection, bathy LiDAR surveys are capable of meeting up to IHO Order 1A. Requirements for the detection of objects less than 2m are difficult and expensive for bathy LiDAR, as they generally require multiple passes of the aircraft. These surveys should generally use maritime vessels which capture more points per area. Topo-bathy sensors are able to produce more points per area than bathy sensors. This provides an additional technology to obtain better object detection, however the sensors can only be employed within their depth limits. Object detection is also dependent on full coverage (no data gaps), and is limited by depth limitations and point spacing


www.crcsi.com.au 28

requirements determined by each sensor. Small object detection in very shallow water can be supplemented by aerial photography.

For environmental risk and modelling, the bathy LiDAR standard survey accuracy is the IHO Order 1B, with no requirement for object detection. Current bathy LiDAR systems enable the point spacing to be set between 2x2m to 8x5m (5x5m for Order 1B). However, the two commercial bathy LiDAR systems which operate in Australia (Fugro LADS Mk3 and AAM-Pelydryn HawkEyeIIB) have different point spacing capabilities. The LADS Mk3 system typically operats at a 5x5m point spacing in Australian surveys. If the application employing the bathymetry data supports a relaxation of the point spacing to 8x5m, the cost of the survey (excluding mobilisation) can be reduced. The HawkEyeIIB system typically surveys at a point spacing of 4x4m. There are no significant advantages, such as cost, of lowering the resolution requirements if this sensor is used within a project. Both systems need to operate at a higher resolution than 2.5x2.5m, with reflies, to obtain the IHO Order 1A standard for object detection.

Topo-bathy LiDAR systems are able to take advantage of a higher pulse rate to obtain more points per area than the bathy LiDAR systems. For an example of data from the two systems see Figure 6. The smaller laser footprint and higher point density from the topo-bathy sensors results in better object detection in very shallow waters (<5m). The vertical accuracies of topo-bathy and bathy LiDAR systems are independent of the selected sensor. However, the horizontal accuracy of topo-bathy systems is improved due to the smaller footprint and the accuracy of the resulting DEM due to the high point density.

Figure 6 - Concurrent bathy LiDAR survey by Fugro LADS Mk3 and Riegl VQ-820-G in Mourillon, France 2012. The lower,

sparser measurements obtained by LADS Mk3 and higher, denser measurements (without gaps) by Riegl VQ-820-G.


www.crcsi.com.au 29

3.1.3 Vertical Datums The required vertical datums for the bathymetric data need to be considered before planning a survey. A vertical datum is the reference surface to which all depths/heights are referred. In MBES and bathy LiDAR surveys, depths have commonly been referenced to tide gauge and/or tide model, and a Global Navigation Satellite System (GNSS) solution for horizontal position. The depths are provided against either a tidal or geodetic datum. In Australia, the most commonly used datums for bathymetric data are the Australian Height Datum (AHD), mean sea level (MSL) and lowest astronomical tide (LAT). Each one of these vertical datums is used for different purposes. AHD is typically used for topographic surveys and bathymetric surveys which traverse the coastal zone. MSL closely approximates AHD and is frequently used within environmental and scientific applications. LAT is used for maritime navigation and nautical charting.

In recent years GNSS positions using DGPS or PPP have enabled ellipsoid heights to be reliably produced from airborne and maritime vessel based surveys. Ellipsoid heights are produced without the need for tidal observations and/or modelling. Ellipsoids also have the advantage of being fixed in space and time, unlike tidal datums which vary between epochs and along the coast. The consistency of ellipsoid referenced bathymetry has advantages. The most significant being the ability to resurvey using the same datum to compare bathymetry captured at different epochs. The derivation of ellipsoid heights and depths for bathymetric surveys also allows a survey to be conducted without tidal observations and modelling. In areas which have an established geoid model and do not have an operating tide gauge this can save time and cost by removing the requirement for a temporary tide gauge to be installed.

Recent bathy and topo LiDAR surveys by the Victorian Department of Sustainability and Environment (DSE) and the CRCSI along the Australian coast have required bathymetry and topographic data to be delivered referenced to both AHD and the GRS80 ellipsoid. The AHD product allows for integration into current datasets, whilst the ellipsoid dataset enables a consistent elevation product to be used for future comparison and datum transformations. A parallel research project at the CRCSI addresses issues with vertical datums in the near-shore. For more information on this research access: http://www.crcsi.com.au/Research/Commissioned-Research/UDEM-for-CC

3.1.4 Budget and Timelines The survey budget and timelines will obviously have an impact on the chosen bathymetric acquisition technology, as will the other project considerations. The higher the budget the more points, greater accuracy, larger extent and greater coverage may be obtained from the survey. The timeline may also be shortened by adding additional resources to the project.

Project costs will be reduced if the area to be surveyed is regular and elongated, rather than irregular, which can require many vehicle turns. These projects will have a lower the cost per area, as the survey/flight lines will be more efficient. The remoteness of a survey to an airport or port will also affect the cost. A bathy LiDAR survey is around four times the cost of a topo LiDAR survey for an equivalent large area. Bathy LiDAR surveys are around AU$600 per km2 for the acquisition and processing of large, regular surveys at IHO Order 1B. However, the cost is generally around AU$1200 per km2 for most near-shore surveys with an extent around 300km2. The cost of topo LiDAR surveys have been as low as AU$70 per km2 for large, regular areas. However a typical survey of around 300km2 will be priced around AU$150 per km2. Topo-bathy surveys will cost midway between a bathy and topo LiDAR survey. The market for these surveys is only new, however the cost is expected to be around $250-$500 per km2.


http://www.crcsi.com.au/Research/Commissioned-Research/UDEM-for-CC

www.crcsi.com.au 30

Airborne LiDAR surveys are cheaper than MBES surveys per area. Muane (2007) estimated the cost benefit of bathy LiDAR over MBES at a ratio of 8:1, excluding mobilisation and transit costs. The cost for MBES surveys in the US has varied between $4400 per km2 for NOAA surveys, and up to $13,300 per km2 for contract surveys (Muane 2007). The main two qualitative factors which will vary the MBES survey costs are the logistics and hazard risk (Muane 2007). The cost of surveys per area need to also account for the mobilisation, which will be a lower percentage of the project cost, as the survey area is increased.

The rate of survey will frequently determine the most suitable acquisition technology. Satellite imagery has by far the fastest acquisition rate for a large area survey, however there are obvious downsides in terms of the other project considerations, such as accuracy and coverage. Airborne survey has the next highest rate of acquisition, followed by maritime vessel based surveys. Table 6 shows the rate of survey for SBES, MBES and bathy LiDAR for three different projects. This information is provided as a guide for the rate of acquisition. In Western Port Bay (see section 4.1 Elevation Acquisition of a Turbid Bay Using Multiple Technologies) a MBES survey of 240km2 took around 60 days to complete. A similar sized survey using bathy LiDAR would take less than a week of flying. However, the use of MBES in Western Port Bay was necessary due to turbidity conditions which prevented bathy LiDAR soundings.

Single-Beam Echo Sounder (SBES)

Multi-Beam Echo Sounder (MBES)

Bathymetric LiDAR (5x5m point spacing)

Example Survey Beaufort Sea Cape Jaffa Mackerel Islands Hours of Survey 5 hours 3.5 hours 5.5 hours Surveyed Depths 80-150m 9-21m 0-45m Average Speed 21km/hr 12km/hr 324km/hr Number of Points 86,112 41,130,000 10,304,000 Area 16.5km2 2.6km2 349km2 Area Coverage per hour 3.3km2/hr 0.75km2/hr 63km2/hr Survey Lines 89.2km (200m spacing) 33.24km 1033km Survey Lines per hour 17km/hr 9.5km/hr 187km/hr

Table 6 - The different rates of survey based on acquisition sensor

Of the three technologies listed in Table 6, SBES is the least expensive sensor and easiest to setup and mobilise for a survey. However, SBES does not have swath coverage and only gathers depths directly below the survey vessel. MBES gathers more data via the swath coverage and can obtain full coverage of the seafloor, achieving more detail than airborne LiDAR. MBES is generally acquired at a slower vessel speed compared to SBES. Bathymetric LiDAR gathers data at the fastest rate, however it records less detail than MBES and is much more expensive to mobilise. The sample coverage rate of 63km2 per hour is on the high side, as the coverage rate for bathy LiDAR usually varies between 20-78km2 per hour (Maune, 2007) depending upon the specifications.

Surveys can be completed more rapidly by using additional vehicles and sensors however this is not always practical and will often add to the cost. It also becomes more expensive if the weather is not conducive to acquisition and standby rates have to be paid.

The ideal survey strategy for environmental modelling applications (using bathy LiDAR followed by MBES in the gaps) is not always possible if there is a time restriction, as this strategy would depend on the LiDAR survey being completed before the MBES survey is initiated. The turnaround time for LiDAR surveys depends on a number of factors including the survey area, environment and quality (manual validation) of the output LiDAR data. Projects generally require one week of acquisition per


www.crcsi.com.au 31

150km2 of survey. The processing time for LiDAR data is at least six weeks after the completion of acquisition. Twelve weeks for a 300km2 survey in ideal conditions with standard products and an average amount of point validation is generally used, if time constraints permit. As the classification, bottom detection and development of new products are increased the processing time is also increased. The necessity of these products must be weighed against the survey schedule before contracting.

3.1.5 Supplementary Datasets - Including Video Imagery, Aerial Imagery, Hyperspectral Imagery and Reflectivity/Backscatter

All airborne systems have the ability to collect supplementary concurrent datasets to complement the LiDAR bathymetry. All systems can capture aerial imagery of varying resolution and quality. Some of the systems capture downward video during the flight and have the capability to capture hyperspectral imagery.

Maritime vessels which use MBES to collect bathymetry also have the ability to capture a range of other datasets from different sensors. This includes video tow, temperature, seismic, magnetic data and others, depending upon the equipment and storage capacity on the vessel.

The downside of capturing concurrent imagery with LiDAR is that it limits acquisition to daytime operations. It is important to note that image quality will not be the same as from a dedicated aerial imagery capture. If the same quality as a dedicated aerial imagery capture is required with the LiDAR, the acquisition times will be limited given that high quality imagery requires no cloud, limited sun glint and shadows. If the expectation of the imagery quality is lowered, concurrent imagery can be cost effective as part of the LiDAR capture.

Figure 7 - Aerial imagery draped over concurrent HawkEye II bathy LiDAR. Courtesy of Pelydryn.

Topo-bathy systems are less restrictive than bathy systems when conducting concurrent imagery acquisition, as their power consumption is lower, which can be a limitation with airborne capture. Hyperspectral imagery requires more power than standard RGB imagery, which cannot always be afforded, depending upon the power requirements of the bathy LiDAR system and power capacity of the aircraft.


www.crcsi.com.au 32

The two active bathy LiDAR systems in Australia are the Fugro LADS Mk3 and AAM-Pelydryn Hawk-EyeII. Different cameras are used for image capture in these two systems. The Fugro LADS system captures concurrent aerial photography with a Redlake ES2020 camera. The output pixel size from the camera at a typical altitude is 0.4m. The hyperspectral camera used concurrently with the Fugro LADS system is a Hyspex VNIR-1600 which operates between 400 - 1000nm and can capture imagery at a typical spatial resolution of 27cm x 54cm. The AAM-Pelydryn system captures concurrent aerial photography with a number of operating cameras however it does not currently acquire hyperspectral imagery concurrently with LiDAR. AAM-Pelydryn can include hyperspectral imagery captured from a different aircraft using a CASI 1500 which is operated from 380-1050nm at a 75cm resolution.

Reflectivity, as it is known from bathy LiDAR systems, or the intensity from topo LiDAR systems, or the backscatter from MBES systems, is a measure of the strength of the return signal. The intensity value is a function of the wavelength used (commonly 532nm for bathy LiDAR and 1064nm for topo LiDAR). It is an important attribute to include within bathymetric datasets. The addition of this information is inexpensive, as it is gathered whilst measuring the bathymetry, however the information is invaluable for modelling seafloor bottom types and benthic habitats. Although, it is not a supplementary dataset in its own right, it is an important inclusion within any bathymetry deliverable.

3.2 Environmental Considerations 3.2.1 Minimum, Maximum and Average Depths The depths within the survey extent play a major role in the selection of the acquisition technology. Depths less than 3-5m are hazardous to maritime vessels. Depths less than 20m are relatively expensive for maritime vessels to record as the survey lines are close together for a full coverage. MBES survey efficiency is poor in complex shallow waters. For an MBES survey in the 20m depth range the survey lines are in the vicinity of 50m apart for complete coverage, this can be extremely expensive for large areas. Shallow depths also lead to increased hazard risk..

The cost of MBES surveys decrease dramatically as the depth and MBES swath width increases. For surveys beyond 20m depth, maritime vessels are a much more viable option. Depending upon the conditions, surveys deeper than 30-40m would frequently be beyond the depth range of LiDAR and imagery. LiDAR may be able to obtain depths beyond 40m in some clear water environments however there is a high risk of no bottom detection in most waters of this depth. Therefore, any survey beyond 40m depth will generally require a maritime vessel with a MBES.

For depths less than 40m bathy LiDAR systems are a viable option. The maximum depth obtained from bathy LiDAR is dependent upon a number of factors, such as turbidity and seabed bottom type (addressed in more detail in the following sections). A bathy LiDAR system's minimum water depth differs depending upon the laser footprint, laser power, measurement frequency, receiver sensitivity and processing techniques.

In bathy LiDAR systems the sea surface and seabed both contribute a return signal. An example can be seen below from the Fugro LADS Mk3 system. In this example the sea surface return (left) and seabed return (right) are clearly separated. In shallower waters, the returns from the sea surface and seabed are closer together, which can make them indistinguishable.


www.crcsi.com.au 33

Figure 8 - Raw waveform in deeper water on benchmark line 1.0.1 in position 26 40 32.39S 153 11 49.48E from the Fugro

LADS Mk3 system

Bathy LiDAR systems use various techniques to separate the return from the surface and seabed. In the LADS Mk3 system, a polarizer is used on the laser receiver to suppress the polarized surface return from the sea surface reflection, as opposed to the randomly polarized reflections from the seabed.

In the Fugro LADS processing software, turning points and points of inflection are identified on the leading edge of the returned laser waveform to detect a seabed return that is partially merged with the surface return (see waveform in Figure 9). In very shallow water, the trailing edge of the combined surface / subsurface return may be used, and the pulse width is used to estimate the depth of the return from the shallow seabed. The figure below shows the return signal from the sea surface and seabed in shallow water.

Figure 9 - Raw laser waveform in shallow water in a depth of 2.0m 48E from the Fugro LADS Mk3 system

Topo-bathy systems are better able to distinguish the sea surface from the seabed in extremely shallow water (<0.5m). This is shown in Table 1, where the minimum depth is listed for each LiDAR system. The Riegl topo-bathy sensor is also able to record detailed points mapping the water surface at the time of survey. The importance of extremely shallow water depth measurements will vary depending upon the survey area and application. If obtaining measurements in these depths is


www.crcsi.com.au 34

critical, an appropriate sensor should be used, or repeat surveys at different times in the tide cycle and supplementary data acquired as needed. In depths less than 1m where the seabed has poor reflectivity, depths may not be measured from any LiDAR system, as the seabed will not be distinguished from the surface return.

Topo-bathy LiDAR systems will record more measurements in very shallow depths, but this comes at a depth penetration cost. Their maximum depths are limited to 10-20m, which is primarily as a result of the lower laser energy per pulse (shown in Table 1), compared to the 30-40m maximum depth range for higher power bathy LiDAR systems. Depending upon the minimum, maximum and average depths within the required survey area, individual LiDAR systems and survey technologies will be better suited.

3.2.2 Turbidity Impacts and Temporal Variations Turbidity has an impact on the coverage of surveys conducted by remote sensing. This includes imagery derived bathymetry and airborne LiDAR. Turbidity is caused by sediment in the water column. As the turbidity increases the depth measuring sensor is increasingly blocked and the likelihood of no bottom detection is increased.

Water clarity has the highest impact on the coverage performance of bathy LiDAR systems. The maximum depth penetration of bathy LiDAR systems is 2 to 3 times the secchi disk depth. The secchi disk depth is greatly affected by turbidity. Waters which have a secchi depth less than 2 metres are generally not suitable for a bathy LiDAR survey, as they are considered too turbid.

In estuary environments higher turbidity and shallower depths are frequently encountered. Bathy LiDAR systems are yet to be fully exploited in these areas, as they have been mainly used over large coastal areas where the water is relatively clear. The results of a trial bathy LiDAR survey on the Sunshine Coast in Queensland are consistent with achieving good coverage in areas where the secchi disk depths are greater than 1.5 metres during on-ground reconnaissance. In areas where secchi disk depths were less than 1.5 metres during the on-ground reconnaissance, little or no coverage was achieved.

Turbidity management is one of the most important aspect of a successful airborne LiDAR survey. A turbidity management plan should always be developed before project initiation to ensure problems areas are managed. The following factors need to be considered in regards to their impact on turbidity when planning a bathy or topo-bathy LiDAR survey:

• Optimal season for the survey;

• Historical and predicted weather including wind strength, turbulence, cloud height, rainfall, sea state and swell;

• Observed impact of spring and neap tides, high and low water, flood and ebb streams;

• Possible biological effects, including plankton blooms;

• Runoff from rivers, river discharge volumes, effects of flooding, farming practices;

• Monitoring of human activity, ports, dredging, ship movements;

• Sampled seabed types, their impact on water clarity, seabed vegetation and kelp growth.


www.crcsi.com.au 35

Turbidity factors vary temporally. Areas in and near estuaries can change daily, or even hourly, depending upon the factors listed above. In the Victorian Trial project in 2007 the Lakes Entrance estuary was flown at 200% on a single day. Even within this survey the repeated survey lines had noticeable differences in the areas of successful coverage.

In very shallow areas which are highly turbid and poorly reflective, it may not be possible to measure the depth. Poor coverage within bathy LiDAR surveyed lakes along the Sunshine Coasts, Queensland, resulted from a combination of high turbidity and a silty, soft seabed. The laser pulse was either fully scattered and absorbed in the water column, or the low return from the poorly reflective seabed was unable to be separated from the surface return, or it was a combination of these effects.

Bathy and topo-bathy LiDAR systems are designed for different depths. Topo-bathy systems have been designed to gather bathymetry down to 10m depth. The coverage capabilities of these systems are currently being tested in rivers, lakes, estuaries and very shallow coastal water. Depending upon the turbidity within the survey extent, a different level of coverage may be achieved from a bathy or topo-bathy LiDAR system. The topo-bathy system has a higher point density for very shallow areas and may therefore achieve a more suitable coverage. Choosing the most suitable system for the project is vital, including the carrying out of a concurrent survey using both types of LiDAR systems to minimise the risk of no coverage in moderately turbid areas, although high turbidity will adversely affect both systems.

MBES are not as adversely affected by turbidity in shallow water as airborne LiDAR. Gaps which exist in the LiDAR derived bathymetry due to turbidity may be filled in with MBES bathymetry. The importance of full coverage, depth limitations and accessibility of maritime vessels needs to be considered against the cost of the survey before employing the MBES. If additional coverage is required in turbid areas after an imagery derived bathymetry or LiDAR survey, then MBES should be considered.

3.2.3 Sea State and Seabed Bottom Type Sea state relates to the swell of the sea. The rise and fall (or roughness) of the sea has an effect on the accuracy of the derived bathymetry, particularly from LiDAR. It also limits the conditions for bathymetry surveys as a high sea state will increase the sea surface error, and for maritime vessels provide a safety hazard. The conditions that come with a high sea state will delay all forms of survey, as aircraft and maritime vessels are placed on stand-by. A high sea state also impacts surveys by raising the turbidity in environments with a fine sedimentary seabed. A high sea state will also cause quenching for an MBES which prevents measurements. Quenching is a reduction in the transmission or reception of the sound energy from absorption and scattering by air bubbles entrapped around the sonar dome.

Sea surface errors due to the swell and roughness are variable and dependant on the angle of incidence of the LiDAR beam at the air/sea boundary, the depth of water and sea state. Bathy LiDAR surveys are generally only acquired in Sea States 1 and 2 which have waves below 0.6m.

The opposite conditions to a high sea state are mirror-like (still) conditions which can also have an adverse impact on bathy LiDAR coverage. Mirror-like conditions may result in more reflection from the sea-surface resulting in less energy reaching the seabed. In addition, the return from the sea surface at nadir may saturate the receiver despite the attenuation of the polarizer. This may result in subdued shallow seabed returns hidden within saturated surface returns. A development to overcome the limitation of mirror-like conditions is to pitch the sensor platform slightly forward to


www.crcsi.com.au 36

produce a rectilinear off-nadir scan. In deeper areas more gain (laser power) is applied to overcome this effect, so that the effect is limited to less than 2-3m depth.

Seafloor type impacts the signal received by bathy and topo-bathy LiDAR systems. The effect is recorded in the return signal reflectivity. The transmitted energy is determined by the laser power and the received energy is measured by the receiver. The difference between the transmitted and received energy is the energy lost. Energy is lost due to:

• the passage of the laser through the atmosphere to and from the water surface; • the surface reflection from the pulse entering the water • the scattering and absorption on the passage down and back up through the water column • through bottom absorption and isotropic reflection (uniform in all directions) from the

seabed • internal surface reflection from the pulse exciting the water column; and • losses at the receiver.

If all the losses are estimated or modelled, the bottom absorption can be determined. The bottom reflectivity can be calculated knowing the bottom absorption loss. The bottom absorption loss changes with seafloor type. The mapping of the bottom absorption loss enables the identification of seafloor type boundaries. White sand is an excellent reflective surface, returning significant energy and producing strong waveforms, with large signal to noise ratios. Silt and other poorly reflective surfaces return less energy. As a result some bathy LiDAR systems apply an automatic gain control (automatically increase the sensitivity)to the receiver to detect weaker returned signals from the seabed. The maximum gain that can be applied is limited by the environmental conditions, particularly turbidity. In turbid environments the backscatter in the water column is amplified to noise by increasing the gain, which can prevent the seabed from being detected. As a result, in clear water the seabed type does not significantly affect the performance of the system, as more gain is applied in areas of low reflectivity. However, in deep areas and turbid areas, there is a limit to the maximum gain, which will leave gaps in areas of low reflectivity.

The most difficult to detect bottom types are darker and softer, such as silty and muddy bottom types. Brighter and harder surface are easier to detect. Bright surfaces have a strong reflection, and hard surfaces have a clearly timed return signal. Sampling the bottom type in a pre-survey reconnaissance will provide information on difficult survey areas due to bottom.

Seafloor vegetation which absorbs/blocks the LiDAR signal can have a detrimental impact on bathy LiDAR surveys. Seagrasses and kelp are the two most common vegetation types to impact bathy LiDAR surveys. The growth of kelp can be seasonal. In the French survey (outlined in section 4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors) three different types of kelp were present during summer. Each type of kelp, shown in Figure 10 and Figure 11, grows within a particular depth and is at its highest stage of growth in late summer. This type of kelp is to be avoided during the LiDAR acquisition by optimising the timing of the survey.


www.crcsi.com.au 37

Figure 10 - Kelp (himanthallia) growing between 0 and 3m depth in France during summer. Image courtesy of Parc

Naturel Marine Iroise France and Fugro LADS.

Figure 11 - Kelp (laminaria digitata) growing between 0 and 5m depth in France during summer. Image courtesy of Parc

Naturel Marine Iroise France and Fugro LADS.

3.2.4 Environmental Changes - Tide, Water Flow, Seasons, Wind and Daylight Environmental changes have a significant impact on the success of bathymetry surveys. In many coastal environments the timing of the survey is pivotal to the data coverage and time taken to complete a survey. The following environmental changes are listed along with their potential effects on the survey:

High and Low Tide - Low tide affects near-shore maritime vessels by reducing access to very shallow areas and reducing the swath width of MBES systems. Aerial and satellite imagery is best acquired at low tide as the reduced depth gives the optimal water penetration. Bathy LiDAR can be acquired at both low and high tides depending upon the requirements of the survey. Low tide will increase the maximum coverage of the LiDAR, except in estuaries where the turbidity levels increase. High tide will shift the LiDAR shallow water gap further into the foreshore and can result in better coverage in shallow water (<5m) areas. The shallow water gap will be larger in the bathy LiDAR, compared to the topo-bathy LiDAR data. The shallow water discrimination in the bathy LiDAR systems is larger than 0.3m for most systems, whereas the topo-bathy systems can discriminate the bathymetry in depths of as little as 0.1m.


www.crcsi.com.au 38

In the Western Port bathy LiDAR survey (described in section 4.1 Elevation Acquisition of a Turbid Bay Using Multiple Technologies) the presence of seagrass was a problem in shallow water because of the difficulty in differentiating between the sea surface and the seafloor reflections. This problem was minimised by collecting data at high tide so that the seafloor was clearly separated from the sea surface.

Spring and Neap Tides - Neap tides provide better conditions than spring tides for bathymetry surveys. Spring tides have the highest tidal range which results in more water movement. In bays and constricted coastal areas this has the potential to produce poorer conditions, especially when the currents increase turbidity. Although, spring tides will expose more land during low tide for the topographic elevations.

Flood and Ebb Tides - In moderately turbid coastal areas the water conditions are clearer on the flood (incoming) tide than on the ebb (outgoing) tide. Near-shore areas can contain significant turbidity from sediment which flows into an area on the ebb stream, for these areas the flood tide should be used as it brings clear water from offshore. In these situations the spring tide may bring in more clear water than a neap tide.

Seasonal Weather: Summer, Winter, Wet and Dry - Seasonal weather dictates a number of factors for bathy LiDAR acquisition. Not only does it affect flying conditions, and therefore the number of stand-by days, it also affects water conditions. A survey performed in the optimal season can make significant savings on time and budget, as well as producing better quality products. The main seasonal factors are composed of winds, temperature, cloud-cover and rain-fall. These factors in turn affect the currents, turbidity, vegetation, sea state and river discharge. The times where these factors have less influence on the survey are most optimal for bathymetry acquisition.

Onshore and Offshore Winds - Strong winds increase the sea state, can cause a water spray above the surface and can reduce the water clarity. Strong tail winds can also cause a problem if the aircraft is near its stall speed. The direction of the winds will also have an impact in increasing turbidity. Localized areas of turbidity can be experienced due to strong winds blowing straight into the mouth of a bay or estuary.

Day and Night Operations - Operations in the coastal zone are generally conducted by day in order to capture digital imagery and actively monitor the water conditions. At night added noise from the sun is no longer present in the LiDAR return signal and the day filter can be removed from the optics which reduces losses in the receiving system. Operations at night may have maximum depths up to approximately 15 - 25% greater than during the day. Generally greater depth performance at night is only achieved in areas where the water is clear and in turbid areas there seems to be little improvement in operating the system by night.

Air Traffic and Restrictions - Air restrictions are common around major cities and sensitive sites. Restrictive flying conditions should be considered during project planning. Optimal times for flying can save on project time and budget. There can also be minimum height restrictions for the aircraft which may prevent operations in certain areas at certain times, or other ATC (Air Traffic Control) restrictions.

Surf and White Water - Surf and white water increase the suspended sediment in the water column as well as reflecting and attenuating the LiDAR pulse at the surface. These affects often result in gaps in the data coverage. These gaps can be filled by re-flying the area in calmer conditions and/or


www.crcsi.com.au 39

at different states of the tide, as the depth of water and direction of tidal stream may affect the disposition of breaking waves along the coast.

3.3 Project and Environment Considerations Conclusion Bathymetry acquisition is impacted by a number of environmental factors; in the air, water and on the ground. Experience in anticipating and handling these factors is critical to a successful project. Elements of the project constraints, such as time and budget, will impact of the choice of sensor and success of the project. Most environmental factors should be managed by the data acquisition organisation, however gathering background information and selecting the most suitable provider and sensor to acquire the project will reduce a significant portion of risk and delays within the survey.

The factors and considerations presented in this chapter provide an overview for commissioning a survey. Knowledge and experience relating to each of these factors is critical to survey design. Key aspects of the project involve balancing competing factors such as:

• Spatial resolution vs survey extent

• Laser power vs point spacing

• Survey flexibility vs supplementary datasets

• Optimal conditions vs stringent product delivery dates

Recognising the optimal balance between competing factors will make a critical difference to the success of a bathymetric survey.


www.crcsi.com.au 40

4 LiDAR Bathymetry Survey Strategies This section of the report focuses on example bathy LiDAR strategies implemented by several different organisations. It presents the purpose of these projects, why the survey strategy was used, the successes and lessons learnt during the project. It is through these examples that some of the project and environmental considerations can be appreciated.

4.1 Elevation Acquisition of a Turbid Bay Using Multiple Technologies Organisation: Victorian Department of Sustainability and Environment (DSE) Project: Future Coasts Program Location: Western Port Bay, Victoria

The Western Port survey was initiated by the Future Coasts Program in Victoria. This program was established to help Victoria better understand and plan for the risks associated with sea level rise and storm surge. Part of this program involved an investment in the acquisition of coastal elevation data along the Victorian coast. One of the most difficult areas to survey was Western Port Bay. Western Port contains extensive mudflats around French Island, and to the north-east of Phillip Island. The muddy bottom creates a highly turbid marine environment. The area is economically significant due, in part, to the surrounding property values and the shipping accessing the Port of Hastings. It is also environmentally significant as it contains a Ramsar (internationally significant) wetland to the north of French Island. These factors contributed to the investment in survey techniques which could achieve near-to-complete elevation coverage of Western Port. The Western Port survey presents an example of how multiple technologies can be used to complete a bathymetry survey in a complex near-shore environment.

The first survey undertaken in the Western Port region was a topo LiDAR survey of the surrounding mainland area. The Western Port part of this survey is small portion of a greater Port Phillip LiDAR project. This was closely followed by the Wonthaggi topo LiDAR project to the south-east of Western Port, and the Greater Melbourne topo LiDAR project which included Phillip Island. These projects completed the surrounding topographic elevation for Western Port.

The next part of the project tackled the more complex areas of bathymetry, the littoral zone and mudflats. The Victorian Government captured their whole coast with bathymetric LiDAR in 2008-2009. Part of this project involved the capture of Western Port bathymetry. However, during the project planning it was deemed that the areas to the north, west and east of French Island would be too turbid for the bathy LiDAR sensor to obtain successful measurements. As such the bathy LiDAR sensor flew only the southern portion (south of French Island) of Western Port (see the light blue coverage in Figure 12). Even in this less turbid area of the bay significant gaps in the LiDAR existed between Phillip Island and French Island.


www.crcsi.com.au 41

Figure 12 - Map of the various surveys used to complete the mapping of bathymetry in Western Port Bay. Light Blue is bathy LiDAR data. Dark blue is MBES data. The remaining areas were captured with topo LiDAR.

Whilst the bathy LiDAR survey was being undertaken, the French Island and Western Port mudflat topo LiDAR project was being acquired (see the yellow coverage in Figure 12) This survey was flown by the topo LiDAR sensor at extreme low tides. This involved capturing the majority of the project during one low tide window and waiting eight months before a similar low tide window exposed the mudflats once again. Having the time and patience to complete the mudflat survey with a topo LiDAR sensor at extreme low tide resulted in less risk and a higher quality product. This completed a significant part of the elevation capture, however more than half the bathymetry was yet to be acquired. The final survey was a MBES bathymetry survey in depths greater than 4m (see the darker blue coverage in Figure 12). The MBES survey was commissioned to the same accuracy and resolution specifications as the bathy LiDAR. This enabled the seamless integration of elevation datasets without a change in quality. The MBES survey data was captured with a pole mounted Geo Acoustics Geoswath 250kHz sensor. All the surveys combined yielded the coverage shown in Figure 13. Small gaps still exist in turbid areas shallower than 4m however the vast majority of Western Port has been surveyed.

This survey yielded successful elevation coverage for Western Port. The major beneficiaries include the bathymetry project partners (Port of Hastings and DSE Biodiversity and Ecosystems Services) who are using the bathymetry within their own projects. A successful aspect of the project was the sequential acquisition of elevation data, particularly with regard to bathymetry. Waiting until the


www.crcsi.com.au 42

mudflat topo LiDAR and the bathy LiDAR surveys were completed before commissioning the MBES survey was an advantage. Once the LiDAR survey was completed the gaps were analysed and depths estimated along the elevation boundaries. The MBES survey was then planned and commissioned to complete the elevation model with enough overlap so that the datasets could be merged. This minimised the MBES acquisition which at shallow depths (all less than 20m) was the most expensive component of the survey based on the area covered. The completed multi-source elevation model for Western Port Bay is shown in Figure 13. The remaining gaps at the end survey of the survey are located in hazardous or inaccessible turbid areas. These areas will remain difficult to acquire for most bathymetric technologies.

One of the issues with the project was the differences in elevation between surveys which resulted from changes in the seabed. These changes were highlighted in the join between the MBES and bathy LiDAR data to the west of Phillip Island. The differences between the MBES and bathy LiDAR data were generally no more than 0.2m, however to the west of Phillip Island differences of up to 5m were evident due to strong tidal currents creating dynamic sand waves on the seabed. These differences pose a risk to any non-concurrent survey, and cause issues for data integration. These issues need to be accounted for in the metadata and data descriptions.

Figure 13 - The Western Port Bay multi-source DEM. Image supplied by the Department of Sustainability and Environment Victoria.


www.crcsi.com.au 43

4.2 A Shallow Water Dataset Surveyed by Multiple MBES Sensors Organisation: Land Information New Zealand (LINZ) Project: Common Dataset for Shallow Water Survey 2012 Location: Wellington Harbour, New Zealand

The Common Dataset is a feature of the Shallow Survey Conferences. The Common Dataset for these conferences contain multiple bathymetric surveys conducted by different sensors. The surveys are all acquired over a common area. The purpose is to provide manufacturers with a chance to showcase their latest systems, and for hydrographic scientists to view the latest data and perform comparisons between sensors and datasets.

The Common Dataset for Shallow Survey 2012 was located in Wellington Harbour. The surveys were performed with four MBES systems between May 2010 and April 2011. A number of key sites were identified for the Common Dataset. To provide an accurate comparison between systems the surveys were required to achieve at least IHO SP-44 Ed 5 Special Order depth accuracy and feature detection, as well as survey coverage of 800x800m.

The common sites analysed in this report are Wellington Wreck (see Figure 14) and Barrets Reef. These sites were surveyed by Reson, R2Sonic, Kongsberg and Geoacoustics. The differences between the bathymetry for each of the sites are shown in Table 7 and Table 8. The Geoacoustics survey had a 12-13m offset from the other MBES datasets which is assumed to be a datum issue, as the standard deviation of the differences was low. The remaining datasets were within a mean difference of 1.36m. The standard deviation of the differences between the datasets was generally better than 0.4m. Based on these comparisons it would seem that the datum establishment is still critical to providing quality bathymetry. Mean differences of more than 1m in bathymetry should not be expected in an IHO SP-44 Ed 5 Special Order survey. Especially to , known bathymetry heights which can play a crucial role in confirming the datum establishment of the survey.

Figure 14 - Kongsberg MBES bathymetry of Wellington Wreck, New Zealand


www.crcsi.com.au 44

System Reson R2Sonic Kongsberg Geoacoustics Min 4.67m

Max 20.57m Mean 11.81m StdDev 0.24m

Min 6.91m Max 17.39m Mean 12.91m StdDev 0.22m


Kongsberg Min -6.94m Max 7.63m Mean -1.36m StdDev 0.21m

Min -4.01m Max 4.48m Mean -0.25m StdDev 0.16m

R2Sonic Min -5.01m Max 8.14m Mean -1.11m StdDev 0.13m

Table 7 - Matrix of height differences between the MBES surveys on the Wellington Wreck, New Zealand

System Reson R2Sonic Kongsberg Geoacoustics Min 2.94m

Max 24.45m Mean 12.04m StdDev 0.31m



Kongsberg Min -7.75m Max 9.07m Mean -0.86m StdDev 0.44m

Min -10.16m Max 9.84m Mean Mean -0.21m StdDev 0.39m

R2Sonic Min -9.72m Max 8.87m Mean -0.66m StdDev 0.17m

Table 8 - Matrix of height differences between the MBES surveys on Barrets Reef, Wellington Harbour, New Zealand

4.3 Progressive Statewide Bathymetry in Critical Areas Organisation: New South Wales Office of Environment and Heritage (OEH) Project: Natural Disasters Mitigation Program (NDMP) Location: New South Wales Near-Shore

Survey budgets do not always enable a single season, statewide collection of bathymetry. More commonly a survey budget is available either opportunistically through grant schemes or on a year-to-year allocation to enable staged, progressive projects. In NSW bathymetric data has been acquired as an initial trial and then progressively to map the Central Coast using LiDAR technology to support projects covering tsunami modelling, coastal erosion and inundation risk assessments, climate change adaptation and marine habitat mapping. The resultant elevation and modelling products are being made available for whole-of-government use. In 2008 the Office of Environment and Heritage commissioned a feasibility trial of bathymetric LiDAR and derived spatial products for areas between Avoca Beach and The Entrance, and from Swansea to Nobbys Head, as shown in Figure 15. The trial covered both near-shore and estuarine environments with a range of depths, water clarity, waves and benthic habitats. Additional high priority areas were consequently extended and captured in mid-2011 along the NSW coast. These areas involved Norah Head to Swansea, Byron Bay and Port Stephens. In late 2011 these surveys were extended to include Byron Bay to Tweed Heads. All surveys have been captured using the Fugro LADS bathy LiDAR sensor. In the near future it is envisioned that the surveyed areas will be further extended in priority areas depending on the availability of resources.


www.crcsi.com.au 45

Figure 15 - NSW bathy LiDAR survey areas. Newcastle region surveys are shown on the left. Byron Bay and Tweed Heads surveys are shown to the right.

The bathymetric data is being combined with existing topographic LiDAR data to create a seamless land /sea DEM. The seamless coastal DEM will underpin sea level rise, storm surge, inundation and coastal erosion modelling to identify how the coastline is likely to change over time. The DEM will also form the base layer for the proposed coastal information system being developed by NSW OEH.

The progressive acquisition of high resolution elevation data is part of a long term state wide strategy to assist local councils assess the vulnerability of the NSW coast to coastal hazards and climate change at a local scale. An immediate area of concern has been the coastal township of Kingscliff, located in the Tweed Shire area. As seen in Figure 16, the Kingscliff foreshore has experienced severe erosion placing infrastructure at risk. In response the Tweed Shire Council (TSC) has undertaken emergency works, to provide protection for parts of the beach and avoid any further damage to buildings located in close proximity to the coastline. To support the TSC, OEH has collected bathy LiDAR as part of its acquisition program. This data is being used to help develop a better understanding of the coastal processes and coastal hazards along this area of coastline.


www.crcsi.com.au 46

Figure 16 - Cable and piping infrastructure currents protruding from the erosion escarpment adjacent to the Kingscliff Holiday Park. Photo: M. Daley 8 November 2011, NSW Coastal Panel Report 2011.

The Kingscliff program provides an example of the value of obtaining additional elevation data to supplement the bathy LiDAR data. Along Kingscliff, additional RTK GPS beach surveys and acoustic bathymetry in the surf zone using a jet ski based platform have been undertaken to supplement the bathy LiDAR data. The NSW Government has also collected terrestrial LiDAR for this section of the coast as part of a state wide data acquisition program. The combination of these four data sources provides a seamless topo-bathy digital elevation model for the site.

Figure 17 - Contours and profiles generated from the Kingscliff RTK GPS survey used to provide coverage in the bathy LiDAR gaps. Kingscliff is located in the Byron Bay to Tweed Heads bathy LiDAR survey.

The bathy LiDAR survey data will also be used in conjunction with a study on the distribution, extent and structure of seabed habitats on the continental shelf of NSW. This study involved collating and analysing existing broad scale bathymetric and marine datasets from previous single-beam and swath acoustic surveys. More information on this study can be found at


www.crcsi.com.au 47

http://www.environment.nsw.gov.au/research/SeabedHabMap.htm. The data was combined with around 100km2of acquired swath acoustic data collected using the NSW OEH’s) interferometric sidescan sonar system. These datasets have provided high-resolution maps of the seabed bathymetry and habitats.

The NSW Government has also evaluated the accuracy of the bathy LiDAR, by comparing the elevation to new and existing surveys conducted with MBES. The bathy LiDAR showed a high correlation with overlapping survey data. An anomalous area in Port Stephens was investigated by repeat maritime vessel survey and then by a further bathy LiDAR acquisition. NSW OEH concluded that the bathymetry differences between surveys were due to highly mobile sand shoals in the outer bay.

Comparative wave propagation modelling using the high density bathy LiDAR dataset and the existing course resolution hydrographic charts was undertaken in 2012. It found that the quality of the bathymetric information can have a significant effect on modelling results in terms of water depths and velocity as well as the extent of inundation. Similar studies have been performed in Queensland on the Maroochydore catchment and coast.

As demonstrated in this section, the NSW OEH was able to obtain LiDAR bathymetry to support a number of applications. Using the bathy LiDAR to underpin multiple projects enables high priority areas to be identified for survey and maximises the data's usage, thereby the value of the acquired bathymetry. Assessing these priority areas on an annual basis and then scheduling them for acquisition is a one of the major successes of the NSW acquisition program.

Another success of the program, which reduced project cost, was that the NSW survey leveraged a nearby bathy LiDAR survey in Queensland in late 2011. By having the survey funding available, NSW was able to take advantage of opportunistic survey timings to save on mobilisation costs. NSW also had several survey areas available which enabled the LiDAR provider to mobilise at the most suitable site on any particular day. Cost reduction and increased efficiency could have been achieved for this survey if more alternate/plan B survey areas were made available to the bathy LiDAR provider. Occasionally weather conditions and water clarity hindered progress on some of the survey areas and at these times additional alternate survey areas would have enabled greater data collection and less standby costs.

4.4 One Capture Using Both Topographic and Bathymetric LiDAR Organisation: Department of Climate Change and Energy Efficiency (DCCEE) Project: Pacific-Australia Climate Change Science and Adaptation Program (PACCSAP) Location: Tongatapu and Lifuka Islands, Kingdom of Tonga

The Kingdom of Tonga survey involved collecting topo and bathy LiDAR over the two islands shown in Figure 18 within the same project. Even though the flights were non-concurrent the overall project designed enabled the seamless integration of the bathymetry and topographic data gathered from each LiDAR system. The project funding was provided by the Australian DCCEE to assist the Government of Tonga in managing risks posed by sea level rise. The survey was flown by AAM-Pelydryn using a HawkEyeII bathy LiDAR sensor and an Optech Orion topographic LiDAR sensor. Concurrent aerial imagery was gathered over the whole project.

A seamless bathymetric and topographic elevation model was produced for the survey area. The seamless product used common survey points and an analysis of foreshore elevation differences for


http://www.environment.nsw.gov.au/research/SeabedHabMap.htm

www.crcsi.com.au 48

the integration. The ability of the project manager to control both aspects of the survey within the same project eased the integration process. The project was able to plan the ground control component of the survey to suit both sensors and elevation products.

The geodetic infrastructure on the islands was limited and establishing the vertical datums on the islands was an important part of the project. Establishing a tide gauge in a safe and secure location was paramount to the project. Ground control should also be linked to permanent marks which can be found by following surveys. Both these aspects of the ground control could have been improved for follow up work.

The acquisition of topographic data and bathymetry within a single project was hugely advantageous for the creation of a seamless elevation model spanning the coast. The once-off mobilisation of equipment and personnel resulted in significant budget savings.

Figure 18 - Lifuka Island (left) and Tongatapu Island (right) bathymetric and topographic surveys, Kingdom of Tonga

4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors Organisation: Department of Finistére and the Regional Council Provence-Alpes-Coté D'Azur Project: Litto3D Project - Integrated Coastal Zone Management Location: Brittany, France Galway

The French bathymetric LiDAR project carried out between February and May 2012 provides a unique example of a survey conducted with two complimentary bathymetric LiDAR sensors. The dataset produced by the LiDAR sensors has contributed to the Litto3D project. The Litto3D national program aims to produce a continuous land-sea DEM on the French coastal fringe. The program covers all the French mainland and territories, including Guadeloupe, Martinique, Reunion, Guyana, Mayotte and Saint Pierre, and Miquelon. The coastal strip is surveyed to the 10m depth contour at a minimum and up to 6 miles offshore, and on land to an altitude of 10m, and at least 2km inland. In the north Atlantic and Mediterranean the coastal strip may be extended. In total, the program covers approximately 45,000km2. In France, the Litto3D will become the common basis for management applications of the integrated coastal zone. The bathymetry has been collected to support the following range of applications:

• Knowledge and management of the coastal environment

• Prevention of risks i.e. flooding


www.crcsi.com.au 49

• Economic development i.e. port and tourism

• Exploitation of resources

• Studies and scientific research

• Defence and national security

The concurrent bathy LiDAR sensors provided high density point clouds in the dynamic near-shore area, enabling a more rigorous integration with the adjacent topographic LiDAR data. The greater number of overlapping LiDAR points and the increased ground penetration provided more data for integrating the topographic and bathymetric datasets.

The survey involved the Fugro LADS bathy sensor and the Riegl VQ-820-G topo-bathy sensor acquiring data concurrently (at the same time in the same aircraft). As discussed previously, these two sensors have different specifications and are designed for different depths. The biggest technical difference is that the Riegl has a smaller laser footprint and more frequent measurements, as shown in Figure 19. This results in the Riegl sensor being better suited to shallower waters and topographic measurements.

Figure 19 - Profile showing the different point densities of the bathy and topo-bathy LiDAR systems. Fugro LADS Mk3 LiDAR points (yellow) and Riegl VQ-820-G points (purple). Sea surface points (red) are provided by the Riegl sensor.

Using the two sensors concurrently achieved the best coverage result possible using airborne LiDAR. Both sensors were optimised for different depth conditions so that the overall coverage was maximised. Having one console to operate both sensors eased the data acquisition and weight of equipment onboard the aircraft. The downside of the two sensors is that the flying height is not necessarily optimised for both sensors. For this survey the flying height was a trade off between the Riegl VQ-820-G sensor, LADS Mk3 sensor, cloud heights and elevated terrain.

The survey included an area on the Mediterranean and an area in Brittany. Each of these areas was large, with a number of sections, enabling a choice of optimal survey areas within each section. A downside of the two survey areas was the transit time of three flying hours. This distance impacted the survey as the transit between areas is less time on task. The large overall survey did have some cost savings as it enabled a larger onsite office and server to be setup to create efficiencies for the data processing and transfer from the field office.


www.crcsi.com.au 50

5 Future Bathymetry Survey Strategy The research presented within this report and the User Needs Analysis should be used to inform a strategy for collecting bathymetry over large, complex, shallow water areas. Typically, multi-sensor options are not considered when developing a strategy for acquiring bathymetry. Organisations tend to consider one technology at a time without an overall multi-sensor approach. This chapter is a culmination of all previous research, providing a reasoning for a suggested multi-sensor bathymetry acquisition strategy. To demonstrate this concept Queensland, Australia has been used as an example. The Queensland example provides a guide for a recommended cost-effective, large area, near-shore (LANS) bathymetry acquisition.

5.1 Recommendations For a Queensland Large Area, Near-Shore Bathymetry Survey The Queensland coast provides a suitable example for collecting LANS bathymetry. In a trial survey, the Queensland Government invested in bathy LiDAR acquisition along the Sunshine Coast. Also, in the preceding User Needs Analysis Report the Queensland coast was identified as a priority area due to the numerous organisations interested in near-shore bathymetry. By investing in the Sunshine Coast trial, the Queensland Government has shown support for bathy LiDAR acquisition as an efficient and cost effective method of collecting bathymetry over large areas. However to acquire bathy LiDAR for the whole 7,000km of the Queensland mainland near-shore (<20m depth) would cost in excess of AU$70M. This is an unrealistic budget, and a more cost effective solution needs to be provided if a statewide or LANS capture is to occur. An additional reason for providing an alternate, cost effective solution are the unfavourable water clarity conditions for bathy LiDAR along sections of the Queensland coast.

In 2012 the Department of Science, Information Technology, Innovation and the Arts (DSITIA), Queensland leveraged the Sunshine Coast bathy LiDAR survey to release a report for collecting near-shore bathymetry. The overall aim of the report was to provide a recommendation for building a bathymetry library along the Queensland coast to complement the recently acquired high-resolution topo LiDAR. The study investigated the existing near-shore bathymetry, and outlined the priorities for the most effective and economical method for the collection of near-shore bathymetry for a broad range of users (Dalla Pozza 2012).

When it comes to LANS bathymetry acquisition there are a hierarchy of bathymetry acquisition technologies. The hierarchy is broadly based on a trade-off between acquisition time and cost efficiency, against the measurement density and accuracy. The first and most cost effective technology is satellite derived bathymetry. Satellite derived bathymetry can cover a large area quickly and cheaply compared to other acquisition technologies. However, its downside is that it is less accurate, and is more limited in its depth penetration than airborne LiDAR or MBES. The second technology in the hierarchy is bathy and topo-bathy LiDAR. Airborne LiDAR takes more time and planning than satellite derived bathymetry. For large area surveys LiDAR can be used strategically due to the cost of acquisition. The upside of LiDAR is that it is more detailed and accurate than satellite derived bathymetry. The third technology in the hierarchy is maritime vessel based MBES bathymetry. MBES has the slowest coverage rate in the near-shore of the three, and is the least cost effective over large, near-shore areas. However, unlike the previous two technologies it is least affected by turbidity and adverse seafloor conditions. This enables MBES to acquire bathymetry within data gaps following the LiDAR and satellite imagery acquisition.


www.crcsi.com.au 51

For LANS bathymetry surveys there are numerous options once multi-sensor approaches are considered. Table 9 presents seven options for acquiring LANS bathymetry. The first five options use one type of acquisition technology. Options 6 and 7 provide two multi-sensor approaches to LANS bathymetry. These options are presented as "value" strategies to acquiring LANS bathymetry.

Option Survey Technique Detail and Applications Cost and Timeframes Comparative Pros and Cons 1 Landsat Satellite

Bathymetry Only 30m resolution. Cannot be used for nautical charting.

Cheapest and quickest option to obtain bathymetry.

Coarsest bathymetry, therefore cannot be reliably used for localised wave modelling.

2 WorldView2 Satellite Bathymetry Only

2m bathymetry resolution. Cannot be used for nautical charting.

Expensive satellite imagery. Quicker and cheaper than bathy LiDAR.

More detail and better imagery penetration than SPOT imagery. More expensive to purchase. Cheaper than obtaining a bathy LiDAR survey.

3 Bathy LiDAR Only 8x5m (or 5x5m) bathy LiDAR point spacing. Can be used in nautical charts.

More expensive than satellite imagery. More time required for data acquisition and processing.

Highly accurate bathymetry with a high likelihood of good coverage. Expensive to obtain.

4 Topo-Bathy and Bathy LIDAR Only

0.5x0.5m topo-bathy LiDAR point spacing. Topo-bathy can be used to detect detailed coastal infrastructure.

More data processing and slightly higher budget required than a bathy LiDAR only option.

More detail for the near-shore topographic data and depths less than 10m than the bathy LiDAR only option.

5 Martime Vessel MBES Only

Highest level of detail. Can be used for detailed bottom object detection and highest order surveys.

Most expensive option to cover a LANS bathy survey. Significant amount of time required for survey.

Highly accurate and detailed survey. Impractical timeframes for LANS acquisition. No access to depths less than 5m without a small vessel.

6 WorldView2 Satellite Bathymetry followed by Bathy LiDAR followed by Maritime Vessel MBES

Mixed-source 4m bathymetry grid. Higher accuracy LiDAR data over critical areas. LiDAR areas can be used for nautical charting.

Approximately half the cost of a bathy LiDAR only option. Timeframe is longer than a single technology option as the surveys should be run sequentially.

Accurate LiDAR bathymetry in important coastal areas. Supplementary bathymetry in less critical areas for regional hydro-modelling. Cheaper than obtaining bathy LiDAR for the whole area. Processing if easier as a consistent resolution bathy DEM produced.

7 Landsat Satellite Bathymetry followed by WorldView2 Satellite Bathymetry followed by Bathy LiDAR followed by Martime Vessel MBES

Mixed-source 30m bathymetry grid. Separate 2m grid over townships and critical areas. LiDAR areas can be used for nautical charting.

Cheaper than Option 6 as the Landsat imagery is easier to process and free to obtain. Timeframes would be similar to Option 6.

Multi-resolution DEM makes processing more difficult than Option 6. Lower quality bathymetry derived from Landsat imagery. Cheaper to produce than Option 6.

Table 9 - A list of potential strategies for acquiring LANS bathymetry

Single sensor LANS surveys can be acquired however they do not necessarily present the best value to bathymetry acquisition or to the end users. A multi-sensor approach is able to take advantage of the suitability of each sensor to particular near-shore environments, and when used together are able to complement each other to provide an optimal, value solution across the survey area. Options 6 and 7 in Table 9 provide examples of two multi-sensor strategies to acquiring LANS bathymetry. These options are discussed in more detail using the Queensland coast as an example.


www.crcsi.com.au 52

For Queensland, a mixed accuracy, multi-sensor derived bathymetry grid is proposed. The recommendation is to capture the Queensland coast south of Cooktown using either archived or newly acquired satellite imagery. This area of the Queensland coast is the most populated and the main area of interest (AOI) for the Queensland Government (Dalla Pozza 2012). The AOI south of Cooktown reduces the coastline length for the survey from 7,000km to 4,000km. The total area of for this length of coast till the 20m depth contour is approximately 29,000km2.

The difference between Options 6 and 7 is in the satellite derived bathymetry. Option 6 uses WorldView2 imagery for the entire AOI. This option is substantially more cost effective than using airborne LiDAR for the same coverage. However this option is expensive compared to other satellite sources such as Landsat and SPOT. Option 7 provides a lower quality product, more cost effective solution by utilising cheaper satellite imagery options.

As outlined in 2.2.1 Satellite and Aerial Imagery, the Digital Globe WorldView2 satellite is the most suitable satellite imagery for deriving bathymetry. However, the downside of using WorldView2 is the cost of imagery and processing time due to the increased spectral bands and resolution. To reduce the cost of Option 6 archived satellite imagery from WorldView2 can be used, which is cheaper than newly acquired satellite imagery. The multi-spectral imagery from WorldView2 has the advantage of providing better water penetration than other satellite imagery. Imagery from the WorldView2 satellite could achieve depths down to around 30m in clear conditions. The bathymetric products from this imagery would be created at a 2m grid resolution. The total cost to produce the first pass bathymetry derived from WorldView2 satellite imagery would be around $3.75M for Option 6 based on EOMAP provided costings. The cost to purchase the WorldView2 imagery from archive would be $1.25M, the remaining budget would be dedicated to processing the imagery. The Queensland AOI shown in Figure 20 is covered by at least one already existing WorldView2 scene the most recent of which was archived on 27 September 2012. As the AOI is captured by a single scene this will ensure consistency of imagery and conditions within the AOI.

Figure 20 - Area of Interest (AOI) is covered by a single WorldView2 scene as shown for 27 September 2012. Courtesy of EOMAP.


www.crcsi.com.au 53

To reduce the project cost Option 7 uses Landsat satellite imagery to derive bathymetry for the entire AOI. WorldView2 could then be used to supplement the coarser imagery in areas of high interest and/or variable bathymetry. If the Landsat satellite imagery is used and the WorldView2 supplemented in 10-20% of the project area the satellite imagery derived bathymetry cost estimate would be reduced to around $900K for Option 7. However, the grid resolution would be reduced from 2m to 30m due to the coarser imagery. This is a significant reduction in the level of detail contained in the bathymetry product.

The satellite bathymetry coverage needs to be established first so that the depth limits and gaps are identified. The time of year for capture is critical and the optimal season should be chosen for satellite and airborne capture. For most of the Queensland coast the optimal time is the end of the dry season between October and November. This may vary for some locations towards the north of the survey where cyclones and monsoonal rains occur December-March and strong south-easterly winds dominate the dry season from April-November. A desktop study, and even field reconnaissance, needs to occur for critical areas which may have unique, localised conditions affecting water clarity and/or seabed reflectivity.

One of the advantages of obtaining satellite imagery as a first step for deriving near shore bathymetry is that it is an efficient method for identifying areas of poor water clarity and/or seabed reflectivity. The satellite imagery will highlight areas of poor visibility, these areas supplemented by field reconnaissance, may be identified as being unsuitable for bathy LiDAR capture. Therefore, they should not be flown by the bathy LiDAR sensor, and should only be obtained by a following MBES survey if deemed critical to the final bathymetry model.

The bathy LiDAR survey should follow the satellite imagery shortly after processing completion. Ideally the bathy LiDAR survey can be acquired at the same time of year, the following year from the satellite imagery capture. This will result in each capture occurring in the optimal season and similar conditions will be experienced during the survey, with seabed changes minimised.

Ideally, the bathy LiDAR survey should be captured in a single season if funding and weather permits. The overall acquisition strategy should be outlined to the LiDAR contractor so the areas can be captured piecemeal on an opportunistic basis to reduce cost, if capturing in a single season is not feasible. This would be a similar strategy to the NSW Government acquisition presented in the previous chapter.

A topo-bathy sensor may be used over critical near-shore lakes and rivers as required. A concurrent acquisition using a topo-bathy and bathy LiDAR system would minimise the risk of no capture whilst maximising data density and coverage. Although, using both systems would increase the cost of LiDAR capture by approximately 30%.

The bathy LiDAR should be captured in critical areas along the coast and as cross-strip areas to calibrate and quality assure the satellite derived bathymetry. The critical areas for initial bathy LiDAR acquisition highlighted by the Queensland Government include the townships of Cairns, Townsville, Mackay and Bundaberg. These major regional centres with large residential settlements and infrastructure in low lying coastal areas that regularly experience severe tropical cyclones therefore extremely vulnerable to storm tide inundation.


www.crcsi.com.au 54

Figure 21 - Priority bathymetry acquisition highlighted by the Queensland Climate Change Centre for Excellence

The area of interest for each of the four townships till the 20m depth contour is shown in Figure 22 and the associated acquisition costs in Table 10. The total cost for the four townships is estimated at $3.75M. These costs are estimated based on 100% coverage (flying once over each area). Realistically some of these areas will be unfavourable for bathy LiDAR capture, and therefore will require multiple flights (>100% coverage), or will not be captured by bathy LiDAR at all.

Township Area of Interest Cost (@$500/km2) Cairns 391km2 $195K

Townsville 2097km2 $1050K

Mackay 2097km2 $370K

Bundaberg 4279km2 $2140K Table 10 - The estimated cost to survey each township using bathy LiDAR at 5x5m point spacing.

Before the bathy LiDAR is captured in each township the turbidity and seafloor reflectance should be observed for areas of concern. These areas could be identified as low visibility in the satellite imagery. Bottom sampling and turbidity measurements should be acquired preceding and during the survey to identify the optimal conditions and feasibility of capture. If the area is deemed to be unfavourable all year round it can be removed from the survey scope, or identified for a MBES survey.

Cairns

Townsville

Mackay

Bundaberg


www.crcsi.com.au 55

The bathy LiDAR cross-strips used to calibrate the satellite imagery derived bathymetry should be capture for the 4,000km (29,000km2 AOI) of coast from Cooktown to the NSW-QLD border. North and west of Cooktown is less populated and therefore provides less value for collecting bathy LiDAR. The bathy LiDAR cross-strips should be continued, where possible, onshore up to 1km and over stable, open ground areas so that tie points between overlapping topo LiDAR datasets can be established and used to assess the integration of these datasets. It is proposed that the cross-strips be flown 20km apart, however these can be further separated to reduce acquisition costs. The cross-strips can be flown seawards to between the 20m and 30m depth contour depending upon the depth penetration of the satellite imagery. Or at slightly more cost and to maximise the survey opportunity the cross-strips may also be flown seaward till the bathymetry can no longer be identified by the sensor.

To increase the bathy LiDAR coverage and provide a stronger survey solution a single survey flight line along the coast should be considered joining 3 or more cross-strips as shown in Figure 23. The cross strips flown 20km apart are estimated to cost $560K and the along coast flights $600K as outlined in this strategy.

Figure 22 - Maps of the Bundaberg (top left), Cairns (top right), Mackay (bottom left) and Townsville (bottom right) bathy LiDAR survey areas till the 20m depth contour.


www.crcsi.com.au 56

Figure 23 - Flight plan for the Queensland Bathy LiDAR capture. Flight lines across the coast (orange) 20km apart. Flight lines along the coast (red) should be placed around 15m depth. The 20m and 30m depth contours are shown in blue.

The bathy LiDAR should be flown at a point spacing close to the satellite derived bathymetry grid. If WorldView2 is used for the entire area (Option 6) this would require a 2-4m point spacing for the bathy LiDAR. If Option 7 is used or the bathy LiDAR cost needs to be reduced, a point spacing of up to 10m may be used. The Stage 1 User Needs Analysis of this research indicated that a 10m resolution is adequate for a large number of LANS users in environmental and modelling applications. Currently, the Fugro LADS system provides the most cost effective, coarse point spacing at 5m along track and 8m across track. However, other systems may be able to match the cost of this sparse point density if given the opportunity. If coarser, more cost effective point spacing is provided in the future these may be utilised.

The final location of the bathy LiDAR cross-strips should be identified by the provider, and should avoid both gaps in the satellite imagery and sand wave areas, as these are dynamic providing poor calibration sites. There needs to be flexibility in the cross-strips, allowing for a change in the location along the coast of up to 5km. The flexibility should aim to maximise overlap with the satellite derived bathymetry, minimise adverse survey conditions and provide optimal overlap with the topo LiDAR. If conditions for acquisition are not suitable within the survey period, then the bathy LiDAR survey should not persist in trying to acquire bathymetry problem areas. Either an alternative area can be selected or the area should be removed from the bathy LiDAR scope for the acquisition period.

Once the gaps in the LiDAR and satellite bathymetry are known, areas can be scheduled for a MBES survey or a follow-up bathy LiDAR survey if conditions improve. The MBES component of the survey should be performed last and should only be used in areas critical to the survey. This includes port and township areas, areas of environmental significance, as well as significant gaps in the LiDAR cross-strips which pose problems for calibrating the satellite derived bathymetry. For Queensland


www.crcsi.com.au 57

bathymetry, gaps along the high priority towns of should be considered for acquiring MBES bathymetry. The MBES acquisition can occur in a staged process as weather and funding permits.

An interferometric side-scan bathymetric sonar system may be used to augment or in-conjunction with the MBES survey. In shallow waters (<10 m) the interferometric side-scan is able to provide superior coverage to the MBES enabling survey efficiencies. The critical areas should still be acquired by MBES due to the technology's point spacing consistency, high reliability and accuracy.

The total cost estimates of the bathymetry acquisition solutions and their individual survey components for the Queensland coast are provided in Table 11. These estimates assume modest turbidity is present during the survey, and does not allow for reflies across data gaps.

Solution Survey Technique Cost Survey the whole of the Queensland mainland coast (<20m depth)

Bathymetric LiDAR $70M

Survey the Queensland mainland coast from Cooktown to the NSW Border (<20m depth)

Bathymetric LiDAR $20-30M


WorldView 2 Satellite Derived Bathymetry

$3.75M


Landsat Satellite Derived Bathymetry

$250K

LiDAR cross strips flown every 20km to control the satellite derived bathymetry

Bathymetric LiDAR $560K

The addition of an along coast flight to join the LiDAR cross strips flown to control the satellite derived bathymetry

Bathymetric LiDAR +$600K

The addition of a topo-bathy LiDAR sensor to the survey

Topo-Bathy LiDAR +30% of cost

Total cost for Option 6 in Table 9 excluding the MBES survey

WorldView 2 Satellite Derived Bathymetry + Bathymetric LiDAR

$5M

Total cost for Option 7 in Table 9 excluding the MBES survey

Landsat Satellite Derived Bathymetry + Bathymetric LiDAR

$2M

Total cost for Option 6 in Table 9 including LiDAR coverage for the townships of Cairns, Townsville, Mackay and Bundaberg (excluding MBES)

WorldView 2 Satellite Derived Bathymetry + Bathymetric LiDAR

$8-9M

Estimated cost of contracting MBES at 100% coverage in depths between 5-20m

Multibeam Echo Sounder ~$2500 per km2

Table 11 - Costs of the bathymetry solution for Queensland

The recommended solution for the Queensland coast would be Option 6. This solution provides a balance between cost and data suitability for multiple applications. This solution requires WorldView2 derived bathymetry for the coast south of Cooktown out to a depth of around 20m. If the satellite derived bathymetry is successful for the majority of the Queensland coast, LiDAR bathymetry cross-strips and along coast flights would be completed, along with bathy LiDAR coverage of the major townships in moderate to low risk areas. The total cost of the project is estimated to be $8-9M, as opposed to $20-30M if only bathy LiDAR was used.


www.crcsi.com.au 58

6 Bathymetric LiDAR Acquisition Specifications In recent years bathy LiDAR is being acquired by more agencies, and for many new applications and purposes. For topographic LiDAR, the increased usage and acquisition required standard national specifications to ease the utility of elevation datasets across jurisdictional boundaries. The national base specification for topographic LiDAR ensured a set of consistent products across Australia. The equivalent benefits of a standard specification for bathy LiDAR should now be realised.

Bathy LiDAR has a number of characteristics which differentiate it from topographic LiDAR. These characteristics mean that the standard topographic LiDAR specifications cannot be easily adapted for bathy LiDAR. As such, through this project, standard bathy LiDAR specifications have been developed in a format similar to the topographic LiDAR specifications, however with the inclusion of factors which are critical to the commissioning of bathy LiDAR projects.

6.1 Specification Development In 2008 the Australian Intergovernmental Committee for Surveying and Mapping (ICSM) Elevation Working Group released Version 1.0 of the Guidelines for Digital Elevation Data acquisition. A key part of this development included the standard topographic LiDAR specifications. Use of the specifications has ensured that the LiDAR point cloud data and derived products can be easily integrated and ingested into the National Elevation Framework Data Portal (NEDF-Portal), providing increased discoverability and access to the broader user community (ICSM LiDAR template).

The standard topographic and proposed bathymetry LiDAR specifications provide a minimum base specification and are not intended to limit development of more specialised products. Nor are they intended to inhibit industry development and innovation(ICSM LiDAR template). Interested users of the specifications are encouraged to contribute to the ongoing development of the document.

The bathy LiDAR specifications are not intended to provide new survey standards, much rather to provide product, documentation and delivery. The International Hydrographic Organisation (IHO) has developed the S-44 standards for hydrographic surveys. The 5th and most recent edition of these standards was released in 2008. These standards identify a number of minimum standards for each order of hydrographic surveys. The standards for each order of survey are listed Table 5 in 3.1.2 Accuracy, Object Detection and Point Spacing. The bathy LiDAR specifications uphold the IHO survey standards and the document applies these standards to bathy LiDAR surveys.

The bathy LiDAR specifications provided in Appendix B - Bathymetric LiDAR Specifications have been developed in consultation with Fugro LADS and AAM-Peldryn who operate the two active sensors in Australia and the Pacific region. The specifications will also be presented for review at the next meeting of the Intergovernmental Committee on Surveying and Mapping (ICSM) Bathymetry Working Group for comment, approval and distribution. The ICSM bathymetry working group is the bathymetry equivalent to the elevation working group which approved the standard specifications for topographic LiDAR.


www.crcsi.com.au 59

The viability of the provided bathy LiDAR specifications have been used on a number of bathy LiDAR projects performed by both Fugro LADS and AAM-Pelydryn. These project include:

• Sunshine Coast, Queensland

• Lifuka and Tongatapu Islands, Tonga

• Espiritu Santo and Efate Islands, Vanuatu

• Apia, Samoa

For reference the bathymetric LiDAR specifications are provided in Appendix B - Bathymetric LiDAR Specifications.


www.crcsi.com.au 60

7 References

Birkett C., Markus T. and Neumann T. 2011. The ICESat-2 Mission Laser Altimetry of Ice, CLouds and Land Elevation.... and also Ocean, Coastal and Continental Waters. Ocean Surface Topography Science Team Meeting 2011 (San Diego, US) 16-18 Oct. 2011

Brisette M.B. 2006. Mine Detection Using Swath Bathymetric Sonars: Tools and Techniques. Undersea Defence Technology Conference 2006 (Hamburg, Germany) 27-29 June 2006

Campbell N., Collings S., and Allen S. 2012. WorldView2 June 2012 - LiDAR Depths vs Log Spectral Bands. Commonwealth Scientific and Industrial Research Organisation. pp. 16

Capolsini P., Andréfouët S., Rion S.C. and Payri. C. 2003. A comparison of Landsat ETM+, SPOT HRV, Ikonos, ASTER and airborne MASTER data for coral reef habitat mapping in South Pacific Islands. Canadian Journal of Remote Sensing Vol 29

Collet C., Provost J.-N., Rostaing P., Perez P. and Bouthemy P. 2000. SPOT satellite data analysis for bathymetric mapping. Proceedings of the International Conference on Image Processing, pp. 464-467 vol.3.

Dalla Pozza R. 2012. Queensland Coastal Risk and Bathymetric LiDAR. Department of Science, Information Technology, Innovation and the Arts Report, Queensland Government. pp. 22

Dewberry 2012. Final Report of the National Enhanced Elevation Assessment. United States Geological Service. pp. 817

Feygels V.I., Wright C.W. , Kopilevich Y.I., and Surkov A.I. 2003. Narrow-field-of-view bathymetrical lidar: theory and field test. Proceedings of the Ocean Remote Sensing and Imaging II. Vol 5155 pg. 1-11

Gostnell C. 2005. Efficacy of an interferometric sonar for hydrographic surveying: Do interferometers warrant an in depth examination? National Atmospheric and Oceanic Division (NOAA). pp. 35

GRAS 2010. Mapping bathymetry with satellite images. http://www.grasdk.com/News/2011/10/08/MappingBathymetryWithSatelliteImages.aspx (accessed 1 Sep. 2012)

Hochberg E.J., Atkinson M.J. and Andréfouët. S. 2003. Spectral reflectance of coral reef bottom-types worldwide and implications for coral reef remote sensing. Remote Sensing of Environment. Vol 85 pg. 159-173

Huang W.G., Fu B., Zhou C.B., Yang J.S., Shi A.Q. and Li D.L. 2001. Shallow water bathymetric surveys by spaceborne synthetic aperture radar. Geoscience and Remote Sensing Symposium, 2001. IGARSS '01. IEEE 2001 International, pp. 2810-2812 vol.6.

ICSM 2011. ICSM LiDAR Acquisition Specifications and Tender Template. Intergovermental Committee on Surveying and Mapping (ICSM). pp. 34


www.crcsi.com.au 61

IHO 2008. Standards for Hydrographic Surveys Special Publication No. 44 5th Edition. International Hydrographic Office (IHO), Monaco. February 2008. pp. 36

Jing E. 2010. Detection of coastal bathymetry using hyperspectral imagery. Oceans 2010 IEEE. (Sydney, Australia) 24-27 May 2010

Kutser T. and Jupp D.L.B 2002. Mapping coral reef benthic habitat with a hyperspec-tral space borne sensor. Proceedings of Ocean Optics XVI. (Santa Fe, US) 18 -22 November 2002

Kutser T., Vahtmae E. and Martin G. 2006. Assessing suitability of multispectral satellites for mapping benthic macroalgal cover in turbid coastal waters by means of model simulations. Estuarine. Coastal and Shelf Science Vol 67 Pg 521-529

Maune D.F. 2007. Digital Elevation Model Technologies and Applications: The DEM Users Manual, 2nd Edition. The American Society for Photogrammetry and Remote Sensing (Bethesda Maryland, US) pp. 655

McQuillin R., Bacon M., and Barclay W. 1984. An introduction to seismic interpretation: reflection seismics in petroleum exploration, 2nd Ed. Graham & Trotman Ltd. (London, UK)

Minghelli-Roman A., Goreac A., Mathieu S., Spigai M. and Gouton P. 2009. Comparison of bathymetric estimation using different satellite images in coastal sea water. International Journal of Remote Sensing Vol 30 No 21 10 Nov 2009 Pg 5737-5750

Mobley C.D. 2009. Algorithm Comparison for Shallow-Water Remote Sensing. Office of Naval Research Report. pp. 6

NASA 2012. MABEL Flies her Maiden Voyage. http://www.nasa.gov/topics/earth/features/mabel-maiden.html (accessed 11 Oct. 2012)

NOAA 2012. Phase Differencing Bathymetric Sonar. http://www.nauticalcharts.noaa.gov/csdl/PDBS.html (accessed 10 Oct. 2012)

OzCoasts 2012. Multibeam and Swath Echo Sounders. http://www.ozcoasts.gov.au/geom_geol/toolkit/Tech_CA_mbe.jsp (accessed 27 July 2012)

Quadros, N. 2009. A New Approach to Delineating the Littoral Zone for an Australian Marine Cadastre. The University of Melbourne. pp.236

Mumby P.J., A.J. Edwards. 2002. Mapping marine environments with IKONOS imagery: enhanced spatial resolution can deliver greater thematic accuracy. Remote Sensing of Environment. Vol 82 pg. 248-257

Robinson I.S. 2004. Measuring the Oceans from Space: The Principles and Methods of Satellite Oceanography. Praxis Publishing (Chichester, UK) pp. 720.

S. Sagar and M. Wettle 2010. Mapping the fine-scale shallow water bathymetry of the Great Barrier Reef using ALOS AVNIR-2 data. Oceans 2010 IEEE. (Sydney, Australia) 24-27 May 2010

Stoker M.S., Pheasant J.B. and Josenhans, H. 1997. Seismic methods and interpretation. Glaciated Continental Margins: An Atlas of Acoustic Images. Chapman and Hall (London, UK) pp. 315


www.crcsi.com.au 62

USGS 2012. LiDAR for Science and Research Management. http://ngom.usgs.gov/dsp/tech/eaarl/index.php (accessed 12 Dec. 2012)

Wolfgram P. and Vrbancich, J., 2006. Layered-Earth Inversions of AEM Bathymetry Data

Incorporating Aircraft Attitude and Bird Offset. Australian Earth Science Convention 2006 (Melbourne, Australia).

Vrbancich J., 2006. Mapping Turbid Coastal Waters. Position Magazine, No. 26, pp. 64-69 Vrbancich J. 2012. Airborne Electromagnetic Bathymetry and Estimation of Bedrock Topography in

Broken Bay, Australia. Geophysics. Vol. 77 No. 4. pg. 1-15 Vrbancich J. and Fullagar, P., 2006. Remote Characterisation of Shallow Seafloor Using Airborne TEM.

Australian Earth Science Convention 2006 (Melbourne, Australia). Yang Y., Marshak A., Varnai T., Wiscombe W. and Yang P. 2010. Uncertainties in Ice-Sheet Altimetry

From a Spaceborne 1064-nm Single-Channel Lidar Due to Undetected Thin Clouds. IEEE Transactions on Geoscience and Remote Sensing. Vol 48 No 1 pg. 250-259

Zollinger R., Morrison H.F., Lazenby P.G. and Becker, A., 1986. Airborne Electromagnetic Bathymetry.

54th Annual International Meeting - Society of Exploration Geophysicists (Atlanta, US).


www.crcsi.com.au 63

Appendix A - Summary of Current LiDAR Sensors Fugro LADS Mk 3 Bathymetric LiDAR Source: H. Parker and M. Sinclair, personnel communications, August 2012

Fugro LADS, who operate the LADS Mk3 system, have a vast amount of survey experience in Australia. Fugro LADS also makes frequent trips to North America and Europe. The majority of Australian bathymetric LiDAR projects have been flown with a LADS sensor. The Royal Australian Navy also acquires LiDAR bathymetry using LADS, although from a LADS MkII sensor operated from a de Havilland Dash8 aircraft.

Recent developments in the Fugro LADS Mk3 system include the acquisition of ellipsoid referenced bathymetry from DGPS rather than tidally derived depths, the addition of hyperspectral imagery to compliment the LiDAR bathymetry, and the delivery of files in LAS format with relevant point classifications. However, the most significant recent development is that Fugro LADS are currently trialling the topo-bathy Riegl VQ-820-G LiDAR sensor concurrently with the bathy LADS Mk3 LiDAR sensor for the capture of shallow water bathymetry. The results from the initial tests are presented in 4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors.

LADS Mk3 Features

• The LADS Mk3 system was released in June 2011 designed, built, operated and continually enhance by Fugro LADS.

• LADS Mk3 has a powerful laser which can reach depths of 80m, although depths are typically limited to 20-35m in Australian coastal waters.

• LADS Mk3 measures both topographic and bathymetric elevations with only a green laser.

• As opposed to the previous LADS sensors, the LADS Mk3 is not fixed to one aircraft. It can be moved into different survey aircraft.

• Concurrent supplementary technologies with the LADS Mk3 sensor include 40cm RGB imagery captured with a Redlake Mega Plus II ES2020 camera, hyperspectral imagery captured with a Hyspex VNIR-1600 camera and the Riegl VQ-820-G topo-bathy LiDAR.

Figure 24 - Fugro LADS Mk3 laser. Courtesy of Fugro LADS.

Optech SHOALS 3000, CZMIL and ALTM Aquarius Bathymetric LiDAR Source: D. Collison, personnel communications, 7 September 2012


www.crcsi.com.au 64

Optech have been designing and building bathymetric LiDAR sensors in Canada since the early 1990s. However, they do not operate any of their own systems commercially. Optech have delivered a number of bathymetric LiDAR sensors around the world to customers who include the US Navy, US Army Corps of Engineers, Japanese Coast Guard, Swedish Navy and United Arab Emirates Survey Department. Their most relevant commercial system is the SHOALS 1000T bathy LiDAR sensor which was delivered to Fugro Pelagos in 2005 for commercial operation. The 1000T was used in Australia off the coast of the Northern Territory in a Royal Australian Navy project spanning 2007-08. The Optech SHOALS 1000T has since been upgraded to the SHOALS 3000 which managed to increase the measurement frequency of the previous system by three fold.

Recent developments at Optech include the topo-bathy CZMIL and Airborne Laser Terrain Mapper’s (ALTM) Aquarius LiDAR sensors. Like other topo-bathy sensors, the CZMIL sensor has been designed to perform better than the SHOALS bathy sensor in shallow, turbid water. The sensor has increased the size of the receiver aperture to double the spatial resolution and also better detect the seafloor in shallow turbid environments. CZMIL uses beam-splitting and a segmented detector to acquire more measurements from a single laser pulse. An additional part of the design was in the integration of the hardware and software to provide a combined mapping solution which creates products by combining the information from three sensors; LiDAR, RGB imagery and hyperspectral imagery.

The ALTM Aquarius topo-bathy LiDAR sensor has been designed to collect shallow and inland water depths to around 10m. This sensor is used to compliment the Optech ALTM Gemini topo LiDAR system. Operators in Australia, such as PhotoMapping Services and AAM Pty Ltd, who use the ALTM Gemini system may use the Aquarius sensor simply by replacing the laser head to collect shallow water measurements.

SHOALS 3000 and 1000T Features

• Collects water, shoreline and topographic (ground elevation) data simultaneously, integrating land and water measurements in the same data set

• Optech’s REA software enables detailed analysis of the seafloor and water column, including seafloor reflectance information and water clarity information

• Fully integrated flight planning and data processing software, with automated data processing and manual 3D editing

CZMIL Features

• The Optech CZMIL system was released for operation in May 2011

• CZMIL has been designed as a system which integrates its three sensors into an automated processing package. The design produces a suite of marine products, which include coastal topographic data, benthic classification and water column characterisations.

• A new circular scanning pattern is reported to provide better penetration in turbid and surf waters as it provides two opportunities to measure the seafloor (one fore and one aft). The two viewing angles should provide additional coverage.

• Five CZMIL sensors have been ordered from several national governments however a commercial provider has not yet purchased a system.


www.crcsi.com.au 65

• Supplementary technologies include RGB imagery captured with an Optech T-4800 16MP camera and hyperspectral imagery captured with a CASI 1500 camera.

Figure 25 - Optech CZMIL System. Courtesy of Optech Inc.

Aquarius Features

• Commercial providers who operate the ALTM Gemini topo LiDAR system can replace the head with a topo-bathy ALTM Aquarius sensor.

• There is much less investment for a commercial provider with an ALTM Gemini topo system to invest in an Aquarius system compared to other bathy and top-bathy LiDAR sensors.

• The laser is not as powerful as bathy LiDAR sensors however its measurement rate is significantly higher; taking more readings per area.

• The sensor can be operated in topographic mode to capture up to four discrete points per pulse.


www.crcsi.com.au 66

AHAB Hawk Eye II and Chiroptera Bathymetric LiDAR Source: S. Welander, personnel communications, 6 September 2012

Airborne Hydrography AB (AHAB) develops airborne LiDAR systems for hydrographic and topographic surveys in Jönköping, Sweden. The HawkEye II bathy-topo LiDAR system was developed in 2002 for deepwater penetrating LiDAR, whereas the Chiroptera bathy-topo LiDAR system is designed for higher density near-shore surveys. The HawkEye II sensor was originally developed in the 1990s with a focus on object detection and submarine hunting in the Baltic Sea. Since 2002 AHAB has focused on bathymetry for nautical charts and marine environment surveying.

The Hawk Eye II system was designed for operation in a low cost aircraft such as the AeroCommander 690 or Cessna 206. When it was released the HawkEye system had the lowest weight in comparison to LADS and SHOALS, and it used lower power (24V, <50 amps). The lower power meant a higher laser sounding density than other bathy LiDAR systems. It was the first system to enable the seamless survey of land and sea floor in a single mission without the need for additional flights and without changing the system configuration between topographic and hydrographic survey modes. However, all bathy LiDAR systems can now acquire topographic data seamlessly. An example of the Hawk Eye II system used in-conjunction with a topographic LiDAR system is discussed in section 4.4 One Capture Using Both Topographic and Bathymetric LiDAR.

Figure 26 - HawkEye II sensor installed in an aircraft. Courtesy of Pelydryn.

The Chiroptera bathy-topo LiDAR system does not compete with the HawkEye II sensor for depth penetration. The sensor is one of the new generations of shallow water bathy-topo LiDAR systems. The Chiroptera system is more effective than the traditional bathy LiDAR systems for gathering elevations. The shallow depth penetration, down to a maximum depth of around 15m, is due to the short pulse length, the high sensitivity and fast response time of the receiver system. AHAB recommends the Chiroptera system for near-shore surveys and high precision surveys of infrastructure and objects, including shallow-water coastlines and inland waters. The full waveform is saved, thus enabling further analysis of the seafloor, such as submerged vegetation analysis.


www.crcsi.com.au 67

Figure 27 - Chiroptera LiDAR sensor. Courtesy of Airborne Hydrography AB.

HawkEyeIIB Features

• The compact size of the LiDAR system allows installation in smaller size fixed wing aircraft. It is a medium strength bathy LiDAR system with a higher sounding frequency, enabling depth penetration down to 50m in clear, non-turbid waters.

• It was the first LiDAR sensor to use "pixelated discrimination" which enables improvements in the point density as the receiver is divided into four sections. This enables a higher density of measurements (four points per pulse) and better identification of bottom objects.

• The nominal 1kHz sounding frequency is effectively multiplied by four when the spot return signals are received. This gives an effective operational frequency of 4kHz and a large optical aperture compared with other LiDAR systems.

Chiroptera Features

• The Chiroptera was released in mid-2012 and weighs less than 80kg. It measures topographic heights with an infra-red laser and bathymetric depths with a green laser.

• The Chiroptera uses the unique oblique LiDAR rotating scanner and illuminates objects from multiple angles minimising the shadowed areas in the dataset. It is more effective in high sea states compared to other systems.

• Concurrent supplementary technologies include various camera options such as DigiCAM 50MP RGB imagery or UI-2280SE 5MP RGB imagery, and can also be integrated with hyperspectral cameras.


www.crcsi.com.au 68

Riegl VQ-820-G Bathymetric LiDAR Source: M. Pfennigbauer, personnel communications, 11 September 2012

The Riegl, like other topo-bathy LiDAR sensors, is designed for shallow waters. It has been successfully trialled in shallow seabeds, rivers and lakes. The Riegl laser has key differences to the other LiDAR sensors. The low beam divergence and the short laser pulses result in a higher net measurement rate and therefore a higher point density. The high point density and narrow beam on the water surface enables the derivation of detailed bathymetric and topographic features, whilst creating a detailed imprint of the water surface for additional applications.

One of the Riegl sensors is operated in Australia and the Pacific by Fugro LADS. The Fugro LADS Riegl sensor has been flown concurrently with the Fugro LADS sensor in France. Initial trials have yielded positive results which have maximised coverage by utilising two LiDAR sensors in the one aircraft. This trial is further discussed in section 4.5 Integrated LiDAR Acquisition Using Two Concurrent Bathy LiDAR Sensors.

Riegl VQ-820-G Features

• Designed for both land and hydrographic surveys with the green laser producing linear scan lines with a rotating mirror backward/forward looking (depending on the integration) at an angle of 21 degrees from nadir.

• Multiple target capabilities, online waveform processing, and optional access to the full waveform data.

• High spatial resolution due to a high measurement rate, low beam divergence, and short laser pulse.

• Compact instrument designed for a helicopter, light aircraft or UAV.

• The highest net measurement rate and point densities for a bathymetry capable system.

• The new surface modeller is capable of generating a detailed model of the water surface including waves by exploiting the high point density. This leads to highly accurate point clouds after refraction correction.


www.crcsi.com.au 69

Appendix B - Bathymetric LiDAR Specifications

See following page.


www.crcsi.com.au 70

<Location> Bathymetric LiDAR Acquisition

QUOTE REQUESTED BY: <Company Name> Request for Quote No: <Quote Number>

Date issued: <Date>

Requested by: <Contract Manager>

Telephone No: <Contact Number>

TENDER SPECIFICATION <MONTH YEAR>


www.crcsi.com.au 71

Preface Digital elevation data which describes Australia’s landforms and seabed is crucial for addressing issues relating to the impacts of climate change, disaster management, water security, environmental management, urban planning and infrastructure design. In recent years dramatic developments in LiDAR technology and industry capabilities have revolutionised our ability to address these issues at the local level. However, inconsistent and diverse product specifications, and variable data quality are often making it difficult to integrate datasets to address regional, state and national issues. In order to optimise investment and the utility of both existing and future data collections there is a need for a national base specification which defines a consistent set of minimum products which ensure compatibility across projects and States.

In late 2008, the Australian Intergovernmental Committee for Surveying and Mapping (ICSM) Elevation Working Group released Version 1.0 of the Guidelines for Digital Elevation Data (topographic) acquisition. In late 2012, the ICSM Bathymetry Working Group released Version 1.0 of the Guidelines for Digital Bathymetry Data acquisition. The intent of these specifications and tender templates is to further improve on the quality, consistency, utility and compatibility of data being captured by government and commercial off-the-shelf (COTS) products increasingly being offered by the private sector. Moreover, the specifications and tender template provide opportunities for greater collaborative investment across all levels of government, and capacity to reduce tender and compliance costs for investors and providers.

Use of these specifications will also ensure that primary LiDAR point cloud data and derived products can be easily integrated and ingested into the National Elevation Framework Data Portal (NEDF-Portal), providing increased discoverability and access to the broader user community.

The specifications have drawn on recent experience across all levels of Australian government, consultation with LiDAR data providers, and the U.S. Center for LiDAR Information, Coordination and Knowledge (CLICK). They provide a minimum base specification and are not intended to limit development of more specialised products. Nor are they intended to inhibit industry development and innovation. We therefore encourage interested users, investors, researchers and suppliers to contribute to ongoing development. If you wish to make a submission aimed at improving this document or require technical support, please email [email protected]. For further related information please visit the following sites:

http://www.anzlic.org.au/nedf.html

http://www.icsm.gov.au/icsm/elevation/index.html

http://www.ga.gov.au/topographic-mapping/elevation/index.jsp

http://nedf.ga.gov.au


Bathymetric LiDAR Acquisition for the <Location> – <Month> <Year>

www.crcsi.com.au 72

<LOGO>

<LOGO>

<Location> Bathymetric LiDAR Acquisition

Contents

2 Project Brief ................................................................................................................... 73

3 General Project Requirements ....................................................................................... 73

4 Project Details and Timeframe ....................................................................................... 73

5 Project Area ................................................................................................................... 73

6 Product Summary of Key Deliverables ........................................................................... 74

7 General Bathymetric LiDAR Specifications..................................................................... 75

8 LiDAR Point Cloud Specifications .................................................................................. 80

9 Bathymetric LiDAR Derivative Data Specifications ......................................................... 84

10 Data Supply Specifications ............................................................................................. 86

11 Project Planning and Reporting Specifications ............................................................... 88

12 Bathymetric LiDAR Quality Assurance Specifications .................................................... 90

Attachment A – Project Area Maps and Available Geodetic Control Points ............................ 91

Attachment C – NEDF Metadata Specifications ..................................................................... 99

Attachment D - Submission of Quotation ............................................................................. 100

Attachment E - Quotation Template ..................................................................................... 101

Attachment F - Ownership/licensing of Foreground Intellectual Property ............................. 102

Attachment G – Statement of Compliance ........................................................................... 103



www.crcsi.com.au 73

<LOGO>

2 Project Brief Provide an overview of the background to the project and the project objectives.

3 General Project Requirements Provide a summary of the required project deliverables and any specific issues that must be addressed in the project.

4 Project Details and Timeframe Provide a summary of the overall project timeframes and specific milestone dates. This should include dates relating to the Tender process, data acquisition, product delivery and reporting. Any requirements for any staged delivery of services and products should also be specified.

Lodgement of the tender by email (20 MB LIMIT) or received by post by <Time and Date>. <Company Name> reserves the right to not accept any tender lodged after the closing date. Direct any further enquiries or questions to <Contract Manager>.

Email: <Contact E-mail>

Phone: <Contact Phone>

Post: <Contract Manager>

<Contract Manager Title>

<Address> The following timeline for the project is required:

Issue of Tender: <Date> Tender Closing: <Date> Award of Tender: <Date> Data Acquisition: <Date> - <Date> Final delivery of data and derived products, including reports:

<Date>

5 Project Area Provide an overview map of the project area and any detailed maps or diagrams as an attachment. Describe the overall characteristics (biophysical, cultural, climatic, etc) which may affect data acquisition, processing or validation. For example, the nature of the bathymetry, vegetation that may impact on responses, access for validation. Provide a digital file in shapefile format depicting the extent of the project and other relevant features.



www.crcsi.com.au 74

<LOGO>

6 Product Summary of Key Deliverables

Product Types Key Deliverables Format Resolution Product Specification Notes

Bathymetric LiDAR Delivery date for the unclassified LAS product by <Date>. Delivery date for the classified LAS and DEM products by <Date>. All other datasets at project completion.

- Unclassified LiDAR points

- Classified LiDAR points

- LiDAR reflectivity Tiles and mosaic

- Digital Elevation model (DEM)

- Contours - Aerial Photography

Mosaic

LAS LAS ECW (mos) Geotiff (tiles) ESRI Grid ESRI Shape ECW (mos.) Geotiff (tiles)

-

Ancillary Information - Flight trajectories - Tidal data - Survey marks - Coverage Tiles

ESRI Shape Excel (xls) ESRI Shape ESRI Shape

-

Delivery Date

Project Reports Delivery date for all products and project completion by <Date>.

- Project plan - Pre-survey QA plan - Post-survey SA report - Progress reports - Data delivery reports - Final project report

including maps

Word (doc) Word (doc) Word (doc) Word (doc) Word (doc) Word (doc)

Ten days post quote acceptance Before survey commencement Before product generation Weekly throughout project Attached to every delivery At conclusion of project

Metadata - Metadata statements XML Attached to every data delivery



www.crcsi.com.au 75

<LOGO>

7 General Bathymetric LiDAR Specifications

General Guidelines

Description

1 Extent Description of the survey area with reference to detailed diagram(s) provided as Attachment A and available in digital (shape file) format. The defined survey area should be buffered by a minimum of 100m.

2 Date of Capture

1. LiDAR: specific capture window requirements.

2. Field Data: specific requirements relative to LiDAR acquisition.

3 Delivery Dates Tender process, data acquisition, product delivery and reporting. Any requirements for any staged delivery of services and products should also be specified. The tenderer must provide realistic timeframes with appropriate justifications if the final date is altered.

4 Gap Minimisation

Where possible, the provider should collect data during periods of low turbidity eg. calm weather and low swell. If feasible, gaps shall be filled by re-flying under different conditions. The provider should exercise judgement when considering 200% coverage* or re-flying at a later date to achieve adequate coverage. The re-fly allowance needs to be indicated for each survey area in the tender response. *200% coverage is defined as half the standard flight line separation

5 Coverage For the off-shore coverage assessment the provider will ensure that a minimum of XX% of all off-shore 100m x 100m tiles have a minimum coverage of XXX soundings for each of the survey areas. The appropriate re-fly allowance should be made to satisfy this requirement.

Where a 5m resolution is used, 100% coverage represents at least 400 soundings per hectare (100 x100m) tile. Tiles intersecting the land and survey boundary are removed for assessment purposes.

If the provider is unable to meet this requirement for each survey area sufficient evidence for the whole of the survey period in the form of “no bottom detection” soundings, photography, swell or other factors shall be supplied in the project report as a justification for not meeting the requirement.

6 Fundamental Bathymetric LiDAR Spatial Accuracy Requirements

The fundamental spatial accuracy of the offshore component of the bathymetric survey must conform to the <IHO Order 1B> standard as published in the Standards for Hydrographic Surveys, Special Publication No. 44. February 2008 Ed 5. Onshore the fundamental spatial accuracy of the survey must conform to the following standard:

a. Fundamental Vertical Accuracy (FVA) i. <= ±25cm. 95% confidence interval (1.96 x RMSE)

b. Fundamental Horizontal Accuracy (FHA) i. <= ±2.0m. 95% confidence interval (1.73 x RMSE)

The tender response should outline the proposed methodology to confirm that these project specifications have been met. Project metadata must include results of accuracy testing.

7 Object Detection Requirements

Object detection requirements can be altered. Typically, IHO Order 1B is used which does not require object detection. If object detection is required the point spacing will need to be less than 4x4m depending upon the LiDAR sensor. Object detection is not required as the survey is to <IHO Order 1B> standard



www.crcsi.com.au 76

<LOGO>

as published in the Standards for Hydrographic Surveys, Special Publication No. 44. February 2008 Ed 5.

8 Horizontal Datum

The Geocentric Datum of Australia 1994 (GDA94).

9 Map Projection The coordinate system for all deliverables is the Map Grid of Australia (MGA).

10 Vertical Datum Orthometric: All deliverables specified below as orthometric will be referenced to the Australian Height Datum (AHD) – as determined by the published heights of local survey control marks within or adjacent to the project extent. Ellipsoid: All deliverables specified below as ellipsoidal will be in terms of the GDA94 reference frame. The source of the ellipsoidal height control shall be explained in the ‘Post-Survey Spatial Accuracy Report’. Tidal: All deliverables specified below as tidal referenced will be referenced to the lowest astronomical tide (LAT) – as determined by the published heights at local tide gauges within or adjacent to the project extent.

11 Survey Control 1. All survey control data used or derived from this contract must be supplied to ensure independent Quality Assurance (QA) of the survey operations, and for possible inclusion in the State’s survey control infrastructure. It is therefore essential that all primary ground stations are permanently marked in accordance with the appropriate State system.

2. The primary ground control and check point surveys must be referenced to the local datum specified above comprising State survey control marks with “established” GDA94 coordinates and/or “accurate AHD” heights as defined in the relevant State Surveying regulation.

3. Survey to establish new primary control shall use techniques to achieve a minimum standard of:

a. Horizontal: Class B b. Vertical: Class B or LD.

As described in the ICSM Standards and Practices for Control Surveys (SP1) Version 1.7.

4. The survey control needs to be outlined in the tender response and in more detail in the Pre-Survey QA Plan. The tide management will form part of the survey control.

5. The analysed differences with overlapping topographic LiDAR and the survey integration points should form part of the control. Areas which will be tested for differences should be highlighted in the pre-survey QA plan.

Elevation data must be validated and corrected for systematic errors to ensure accuracy specifications are met. Documentation must describe how this has been achieved. Refer to the Quality Assurance Section for specific deliverables in relation to this topic.

12 Tide and Turbidity Management

The tender response and Pre-Survey QA Plan are to include a tide and turbidity management plan. The final project report is to include as a minimum for the survey:

a. Tide Model Diagram b. Tide Station Details (position, LAT / MSL difference, status (observer /

monitor / dummy)) c. Prediction Constituents



www.crcsi.com.au 77

<LOGO>

d. All observed tides e. Observed and predicted turbid areas

If the survey technique does not require the recording and derivation of the above tidal information. An explanation must be provided on the survey technique and which of the above attributes were and were not required for the survey.

13 Benchmark Survey

The benchmark areas are flown to confirm the bathymetric LiDAR system performance and to provide data which will enable the appropriate quality assurance.

A benchmark is a small area surveyed as part of a calibration procedure and repeat surveys should be undertaken to randomise systematic errors. Off-shore benchmark areas should be in areas of smooth seabed with no more than one metre change in depth across the benchmark. If required, on-shore benchmark (lake, river and estuary) areas should be chosen strategically depending upon the terrain and turbidity conditions. The on-shore benchmark area should give an insight into the overall turbidity within the on-shore survey area.

Independent of a main benchmark a sub surface feature (shoal) should be identified as early in the survey as is practical and used to verify sounding repeatability and accuracy in both the horizontal and vertical dimension.

The feature benchmarks should be re-flown regularly during the survey. The least depths and positions of the features shall be analysed to confirm the repeatability and accuracy of the survey. If the survey includes an on-shore component, part of the benchmark survey needs to include a survey line in the rivers and estuaries so that the turbidity can be monitored and an appropriate time can be chosen for the estuary and river survey. The daily monitoring of coverage over the benchmark lines should be assessed in real-time for decision-making and reporting. Depending upon the daily survey operations different benchmark surveys can be flown at the start of operations. All benchmark survey lines should be identified within the tender response. Within the first day of successful operations all benchmark surveys must be flown. The results/success of penetration within the benchmark survey over on-shore areas must be included in the progress report during acquisition.

14 Bathymetric LiDAR Sensor Requirements

The bathymetric LiDAR sensor must be capable of: a. detecting the seafloor to depths of at least 40m. b. recording the backscatter from each pulse.

15 Collection

Requirements The survey design must plan on a sounding density of <XxXm>, which will require at least 1 valid sounding in each <Xm x Xm> bin. A higher sounding density option may be presented. Flight line overlap must be 10% or greater. The relative vertical accuracy of adjacent flight lines must be within ± 5cm @ 95% confidence interval Crosslines are to be flown to determine tidal and datum issues across the survey areas. Crossline comparison statistics are to be reported in the Post-Survey Spatial Accuracy Report.



www.crcsi.com.au 78

<LOGO>

16 Environmental Condition for Data Capture

Where possible environmental conditions should reflect the gap minimisation principle. The provider should collect data during periods of low turbidity eg. calm weather, low river discharge and low swell. The survey should be cloud and fog free between the aircraft and the ground. The primary data product is the bathymetry with any optional products including aerial imagery having less priority in survey timing. If the optional products are included within the final project scope the weather should be taken into account, but not at the expense of the primary objectives or timeframes. Estuaries and Rivers Pre-mobilisation secchi depths and/or transmissometer observations and bottom sampling shall be obtained for the estuaries, lakes and rivers. Turbidity observations shall be conducted at fortnightly intervals during the acquisition window and shall take account of different conditions and changes in the riverbed. Repeat observations at different states of the tide (annotating results accordingly) can be used. The bathymetry for the rivers/estuaries/lakes needs to be surveyed during times of low turbidity. On days when the bathymetry is collected within the rivers/estuaries/lakes turbidity observations need to be taken on the ground. The turbidity observations and bottom sampling need to be included in the final project report and correlated against the final coverage. In addition to on ground observations, a section of the rivers, lakes and estuaries needs to be included in the benchmark survey so that an appropriate time can be chosen for the data capture (see 12 - Benchmark Survey for more details).

17 Optional Products

Aerial Photography

1. The aerial photography should be metric digital and coincident with the bathymetric LiDAR.

2. The collection of aerial photography will be secondary to the collection of LiDAR data. Whilst the atmospheric conditions should take into account the quality of the aerial photography it should not do so at the expense of LiDAR capture.

3. The aerial photography should be supplied as an ECW mosaic for the full extent of data acquired and as geotiff tiles using the same tile index used for the LiDAR delivery.

4. The RFQ response should identify the achievable image resolution based on the flying height required to meet LiDAR acquisition specifications. A resolution better than 50cm is required.

5. The spectral range of the imagery should be three band: R,G,B.

6. The aerial photography should be corrected using a georeferenced solution. The accuracies for a georeferenced product are considered “relaxed” in relation to fully orthorectified products (see spatial accuracy below). The rectification process may not necessarily follow a complete orthorectification process work flow, however, providers should clearly outline the processing steps in their proposal, including the use of control and camera corrections.

7. The spatial accuracy of the aerial photography should be ±4 x GSD RMSE (1 sigma or 68%).

8. There should be no gaps between imagery from adjacent flight lines.

9. Colour balancing and colour matching between frames is not expected.

10. Contrast and brightness adjustment of each image is not expected.

11. Frame selection should minimise noticeable exposure patches, vertical



www.crcsi.com.au 79

<LOGO>

height displacement and seam lines between ground features.

12. For all image products, ensure the data values are set to the range of 0-255, with the value 0 or 255 reserved for null image data.

13. The inclusion of aerial photography within the scope of the project will depend on cost and impact on the LiDAR. The RFQ response should include any potential impact on the LiDAR collection.

14. If, after collection, the photography is affected by significant cloud the contracting authority may remove the product from the scope of the project. Therefore, the costs need to be split between collection and supply.



www.crcsi.com.au 80

<LOGO>

8 LiDAR Point Cloud Specifications

Deliverables Specifications

1 Unclassified Point Cloud

1. All returns, all collected points, fully calibrated and adjusted to specified vertical datum, by swath. 1 file per swath, 1 swath per file, (file size not to exceed 2GB).

2. Fully compliant LAS v1.2 (or v1.3), point record format with all standard attributes including:

a. Intensity values (native radiometric resolution) for topographic LiDAR or backscatter values for the bathymetric LIDAR.

b. Return number. c. Georeferencing information in all LAS file headers. d. GPS times recorded as adjusted GPS time, at a precision

sufficient to allow unique timestamps for each pulse. 3. Optionally, include LAS v1.3 deliverables with waveform data are to use

external “auxiliary” files with the extension “.wdp” for the storage of waveform packet data. See the LAS v1.3 specification for additional information).

4. Data is to be provided in the following Vertical Datums: a. Orthometric (AHD) b. Ellipsoidal (GRS80).

5. File naming as per Attachment B.

2 Classified Point Cloud

1. All returns, all collected points, fully calibrated and adjusted to specified vertical datum, and classified as specified below.

2. Fully compliant LAS v1.2 (or v1.3), point record format with all standard attributes including:

a. Backscatter values for the bathymetric LIDAR. b. Optionally, Return number. c. Georeferencing information in all LAS file headers. d. GPS times recorded as adjusted GPS time, at a precision

sufficient to allow unique timestamps for each pulse. e. ASPRS/LAS “Overlap” classification (Class=12) shall not be

used. ALL points not identified as “Withheld” are to be classified.

3. Optionally, include LAS v1.3 deliverables with waveform data are to use external “auxiliary” files with the extension “.wdp” for the storage of waveform packet data. See the LAS v1.3 specification for additional information)

4. Data is to be provided in the following Vertical Datums: a. Orthometric (AHD) b. Ellipsoidal (GRS80) c. <Tidal (LAT)>

4. Tiled delivery, as per Data Supply Specifications below. 5. File naming as per Attachment B.



www.crcsi.com.au 81

<LOGO>

3 LiDAR Point Cloud Classification Scheme

1. All classified point cloud data must adhere to the following modified ASPRS classification scheme.

2. The minimum number of point classes to be delivered according to this scheme is defined by the Classification Level specified below.

Number Point class Description 0 Unclassified Created, never classified

1 Default Unclassified

2 Ground Bare ground

3 Low vegetation 0 – 0.3m (essentially sensor ‘noise’)

4 Medium vegetation 0.3 – 2m

5 High vegetation 2m >

6 Buildings, structures Buildings, houses, sheds, silos etc.

7 Low / high points Spurious high/low point returns (unusable)

8 Model key points Reserved for ‘model key points’ only

9 Water Any point in water

10 Bridge Any bridge or overpass

11 Not used Reserved for future definition

12 Overlap points Flight line overlap points

13 Bathymetry – Underwater Seabed

Seabed

14 Bathymetry – Non-Seabed

Non-seabed and undefined points

15 Bathymetry – Shoals High points subset from seabed (13) points

16-31 not used Reserved for future definition

3. Class 1 (default) are points which have been subjected to a classification

process but emerged in an undefined state. Class 0 have never been subjected to a classification process. This definition is necessary to maintain compatibility with common LiDAR processing suites.

4. When a simple ground/non-ground classification has been applied, all non-ground points will be allocated to Class 1.

5. Class 8 “model key points” is actually a subset of class 2 and so is created as a separate product.

6. Class 15 “bathymetry shoal points” is a subset of class 13.



www.crcsi.com.au 82

<LOGO>

Bathymetric LiDAR Point Cloud Classification Levels

Once the Fundamental Spatial Accuracy requirements have been achieved, significant errors in the vertical accuracy of the classified point cloud are likely to be caused by incorrect classification. LiDAR Point Cloud Classification Levels have been introduced to provide greater transparency in the overall quality of the LiDAR products, particularly within non-bare-ground/bathymetry land cover and seabed types, to ensure products are “fit-for-purpose”. It is expected that classification of the point cloud data will be carried out to achieve known minimum accuracy levels for ground data. The onus for reaching the required accuracy lies with the data supplier. Independent assessments may also be carried out by the Contracting Authority. Classification accuracy requirements may be relaxed to accommodate areas where the Contract Authority agrees classification to be particularly difficult. Undefined All points are allocated classes 0 (unclassified) or 1 (default) by LiDAR processing software with no classification algorithms or pre-validation applied. If classification is not required no levels will be selected in the required classes.

Classification Accuracy Required: unspecified.

Level 1. Automated and Semi-Automated Classification.

Semi-automated processing of the validated point cloud data into the following classes: 1 (non-ground points above waterline), 2 (ground points above waterline), 13 (seabed points) and 14 (non-seabed and undefined points within the water). At Level 1 the data processing involves the following stages:

1. Automatic data processing of the run lines 2. Any pre-validation of the data and initial cleaning of the data by

survey personnel 3. Validation of the data and checking of the data by a Hydrographic

Surveyor / Degree Surveyor 4. Visualisation of the data 5. Approval of the data

Classification Accuracy Required: 98% for seabed points only (minimum), 90% for other specified classes.

Anomalies

Large (>2m) anomalies (errors) within all classification levels must not exist. If clear errors are identified in the QA/QC process these must be corrected.

Shoal Identification.

Level 1 classified data is further enhanced, using automated and manual methods, to include class (15) high seabed points (shallow depth). These are a subset of class (13) seabed and identifies shoal points within the bathymetry dataset. The search radius to define the shoal points is 20m.

Level 2. Detailed Classification and Correction.

Detailed classification and correction of all specified classes including classes: 3-5 (vegetation) and 6 (buildings/structures), which are taken from 1 (non-ground above waterline) points. This may include all or a subset of classes listed in section 3. When specified, each class must achieve the required classification accuracy.

Classification Accuracy Required: 99% for seabed points, 98% for all other specified classes.



www.crcsi.com.au 83

<LOGO>

5 Required Point Cloud Classification Level

1. The following point cloud classification levels are required as a minimum standard for new acquisitions under the NEDF:

Number Point class Required

Classes Classification Level Required

L1 L2

0 Unclassified

1 Default

2 Ground

3 Low vegetation

4 Medium vegetation

5 High vegetation

6 Buildings, structures

7 Low / high points

8 Model key points

9 Water

10 Bridge

11 Not used

12 Overlap points

13 Bathymetry – Underwater Seabed

14 Bathymetry – Underwater Non-Seabed

15 Bathymetry – Underwater Shoals

16-31 Other As specified

Examples Class Example Features 1 vegetation, buildings and bridges (anything not ground) 2 roads, bare ground, sandy beach (only ground points) 13 reef, sand and rocks on the seabed (underwater) 14 sediment, kelp and erroneous measurements in water column 15 highest point on a reef (underwater)



www.crcsi.com.au 84

<LOGO>

9 Bathymetric LiDAR Derivative Data Specifications

Deliverables Specifications 1 Backscatter

Image 1. 5m grid backscatter image 2. Mosaic generated using a scaled 8 bit value representing reflectance of

seabed from each valid sounding. 3. Any changes in gain settings made to the backscatter should be

compensated for in the final product. The backscatter product must be void of any systematic changes to the backscatter values.

4. ECW format using 5:1 compression. 5. Tiled delivery, as per Data Supply Specifications below. 6. File naming as per Attachment B.

2

Digital Elevation Model (DEM) (orthometric)

1. 5m grid Digital Elevation Model (DEM) 2. The DEM should be generated from the LiDAR mass point data classified

as 2 (ground points above waterline) and 13 (seabed points), so that it defines the seafloor and ground surface.

3. All non-ground objects such as sea grass and man-made structures shall be removed from the DEM by classifying the points as either 1 (non-ground points above waterline) and 14 (non-seabed and undefined points within the water).

4. DEM is to be generated from the irregular spaced data. 5. The DEM generation should employ a Point to TIN and TIN to Raster

process with Natural Nearest Neighbour interpolation. 6. DEM interpolation should be performed in areas containing equal to or

less than 4 no bottom detections or in the area equivalent of 10x10m. 7. Void areas (i.e., areas outside the project boundary but within any tiling

scheme) shall be coded using a unique “NODATA” value 8. ESRI floating point GRID format. 9. Tiled delivery, as per Data Supply Specifications below. 10. File naming as per Attachment B.



www.crcsi.com.au 85

<LOGO>

3 Contours (orthometric)

1. The contour interval is to be 0.5m, with every 5m contour interval being assigned as a major contour line.

2. The contours are to be supplied as a single file for each of the survey areas.

3. All contours are to be provided in ESRI Shapefile Format. 4. All files must have projection details attached. 5. Each Contour file should contain the attribute “ALTITUDE” that carries the

elevation value relative to AHD for each contour. Elevations below AHD must be negative.

6. The contour data set should adhere to the following Topology Rules: a. Contours must not Intersect b. Contours must not have dangles unless at the edge of the

data set. 7. Contours are to be visually cartographic type contours. 8. A minimal number of vertices are to be used in defining the contour line,

but without destroying the “natural look” into a computer generated set of unnatural contour lines. Smoothing may be applied.

9. Distance between adjoining vertices would generally not be less than 15 metres, unless depicting a sharp change in direction. The contour data should have a smooth look when displayed/printed at a scale of 1:250.

10. Contour lines are to be long continuous line strings, rather than separate lengths

11. As per the DEM, all non-ground objects such as vegetation and man-made structures shall be removed from the contour data, so that contours define the sea surface.

12. Contours must have an accuracy such that 95% of “well defined” points along the contour line are to have a value that must not differ to the "true" ground surface by any more then half of the contour interval.



www.crcsi.com.au 86

<LOGO>

10 Data Supply Specifications

Deliverables Specifications 1 File naming See Attachment B for NEDF file naming conventions.

2 Coordinate Origins for Gridded Data.

The origin of all gridded data must be placed on a whole metre coordinate value that will align with the zero (0) origin of the UTM/MGA

3 Data Tiling

1. All standard data sets should be supplied as single files where possible and tiled to manageable file sizes if necessary as below:

a. 1km x 1km tiles based on MGA94 coordinates with origins that align with the zero (0) origin of the Map Grid of Australia.

b. Larger tile sizes which maximise workflow efficiency will be considered.

c. The origin of the tile must be placed on a whole metre coordinate value of the south west corner of each tile. e.g. 426000mE_7243000mN

2. A Tile Index is to be provided by the contractor in ESRI shape file format. The tile name as specified above must be included as an attribute in the Tile Index file.

3. File naming as per Attachment B.

4 GPS data for occupations of base-stations

1. GPS data for all base station occupations in excess of 6 hours is to be provided in RINEX V1.2 format (Receiver Independent Exchange Format).

2. GPS observation log sheets which include the following details: a. Survey mark id b. Occupation time & date c. Antenna height measurements d. Instrument /antenna types & serial numbers

The GPS observation log sheets should be provided in pdf format or Excel spreadsheet if data is captured digitally.

6 Data Delivery Reports

1. A delivery report describing the contents of the data supplied with every data delivery (interim, staged, final). The delivery report must also contain reference to the metadata supplied within the delivery.

7 Metadata 1. For each supplied data product a complete metadata statement consistent with the ANZLIC Metadata Profile (Version 1.1) must be provided in XML format. The ANZMET Lite metadata tool will be used to validate all XML records.

http://www.osdm.gov.au/Metadata/ANZLIC+Metadata+Profile/default.aspx

2. In addition, the NEDF Metadata Profile and Tool will be used to provide additional LiDAR specific metadata. The NEDF Metadata tool reads an XML metadata record created by ANZMET Lite. The tool will be made available by the Contract Authority.

3. The list of additional NEDF metadata required is provided in Attachment C.

4. Metadata must be provided with every delivery including interim, partial and final deliveries.

5. The job will not be signed off by Contract Authority until the metadata is satisfactorily supplied.





www.crcsi.com.au 87

<LOGO>

8 Delivery Media 1. Data should be delivered on DVD or External Hard Drive (USB or FireWire). External hard drives will be retained by the Contract Authority.

2. Data deliveries should be clearly labelled with name of Service Provider, date of supply and list of contents.

9 Report Formats All reports are to be provided in Word (.doc) format, Excel spreadsheet (.xls) or appropriate digital format approved by the Contract Authority.



www.crcsi.com.au 88

<LOGO>

11 Project Planning and Reporting Specifications

Deliverables Specifications 1 Project Plan Project plan detailing work breakdown structure, agreed data capture plans,

project milestones, data delivery formats, delivery schedules and progress reporting schedules etc within 10 days of the acceptance of the Contractor’s quote. It is expected that the Plan will include a turbidity and tidal plan for the bathymetric survey.

2 Pre-Survey Quality Assurance Plan

The contractor shall prepare and submit to the Contracting Authority a Quality Assurance Plan for the bathymetric LiDAR survey that conforms to an identified management system and generally complies with ISO 9001. The plan must address the organisation and management of the project, work procedures, environmental considerations, safety and risk control and test procedures. The Plan must also detail the procedures to be used in verifying that the deliverables meet the required specification including:

• The procedures and methodologies to be used to verify that the deliverables meet the required specifications.

• Details of proposed calibration checks and methodology to be used to establish both reference stations and ground test sites.

• Proposed flight plans • Tide and Turbidity Management Plan to address successful capture of

bathymetric data ensuring that water conditions are optimal. • Details of proposed tide gauge sites (if required) and bathymetric

LiDAR survey control • Any other details that the contractor deems relevant or are requested

within these specifications. The pre-survey QA plans must be submitted and accepted prior to the commencement of the survey.



www.crcsi.com.au 89

<LOGO>

3 Post-Survey Spatial Accuracy Report

Acceptance of the Post-Survey Spatial Accuracy Report and related information is required before point classification and other product derivation is to proceed. The absolute and relative accuracy of the data, both horizontal and vertical, and relative to known control, shall be verified prior to classification and subsequent product development. This validation is measured against the Fundamental Spatial Accuracy. For terrestrial measurements these are confirmed against benchmarks, overlapping topographic LiDAR and TIPs in clear, open ground areas. A detailed report of this validation is a required deliverable. For the offshore bathymetric LIDAR the spatial accuracy is linked to the benchmark surveys and any other bathymetric control areas. The report for the bathymetric LiDAR survey will include the following:

• Flight trajectories as specified below. • Details of system calibration checks. • Results of relative (flight run) matching and details of any

adjustments made. • Source of primary ellipsoidal height control. • Details of ellipsoid to orthometric corrections applied including any

final adjustment to AHD supplemental to the standard Geoid correction.

• Results of vertical and horizontal accuracy validation. • All survey control coordinates, site id and check point comparisons

in both Excel spreadsheet and ESRI shape file formats. • Digital photographs of all survey and check sites, with the site id

included in the filename. The bearing of the photo direction should also be included.

• Relationship of the collected bathymetric LiDAR data to previously collected topographic LiDAR data at integration sites.

• Other related information.

4 Flight Trajectories

All flight trajectories used for the capture of the delivered LiDAR data will be supplied in ESRI Shape files. The shape file table’s must include the date of capture, local start time, local end time and which reference station was used for each trajectory. The shapefile must also include the height of tide recorded at a relevant local/temporary tide gauge relative to MSL and AHD.

5 Progress Reports

The contractor, as a minimum will report by email each fortnight. The report should contain a summary of progress, delivery and implementation, and details of any problems encountered and remedial action taken. The report should also address the planned activities for the two weeks ahead, regardless of whether successful capture has been achieved. The initial report should contain details of mobilisation progress; and during the acquisition window the turbidity and tide monitoring should be detailed.

6 Project Report The Project Report should comprise a technical discussion addressing how each of the contract specifications has been met, a statement of consistency with any specified standards, results of independent accuracy and validation tests, metadata statements and extra-ordinary issues that may have affected the nature or delivery of the project. All aspects of the project operations must be adequately reported. All images shown in the project report must also be supplied as high-resolution jpeg images.



www.crcsi.com.au 90

<LOGO>

12 Bathymetric LiDAR Quality Assurance Specifications

Description Specifications 1 Fundamental

Spatial Accuracy Validation (FSA)

Vertical Accuracy Validation 1. The fundamental vertical accuracy of the point cloud dataset will be

determined by a combination of: a. Terrestrial check points located only in open, relatively flat

terrain, where there is a very high probability that the bathymetric LiDAR sensor will have detected the ground surface.

b. Repeatability of observations over the benchmark run and features.

c. Differences to the topographic LiDAR on the foreshore and the TIP areas.

d. Differences to any overlapping bathymetry supplied by the contracting authority.

Horizontal Accuracy Validation 2. The onus for reaching the required accuracy lies with the data supplier.

Independent accuracy assessments may also be carried out by the Contracting Authority.

3. Independent testing of horizontal accuracy for LiDAR products is not required as part of this base specification. Instead data producers are required to report on the expected horizontal accuracy of elevation products as determined from system and sensor calibration studies.

4. In the above circumstances a “compiled to meet” statement of horizontal accuracy at 95 percent confidence should be reported.

5. As an alternative, the producer may demonstrate compliance through analysis of distinct features which are identifiable in the elevation data (e.g. fences) or backscatter images with other data sources such as imagery with known horizontal accuracy.

2 Interpolation Consistency Validation

All products derived from the LiDAR mass point data as tiles will show no edge artefacts or mismatch. A quilted appearance in the overall project surfaces, whether caused by differences in processing quality or character between tiles, swaths, lifts, or other non-natural divisions, will be cause for rejection of the entire deliverable.



www.crcsi.com.au 91

<LOGO>

Attachment A – Project Area Maps and Available Geodetic Control Points

1. Project Area Maps

The maps of the are shown on the following pages. The maps show the areas required for the bathymetric LiDAR capture.

The approximate areas of capture are as follows:

<Location> Bathymetric LiDAR XXX km²



www.crcsi.com.au 92

<LOGO>

Attachment B - File Naming Conventions

The following naming conventions have been developed to provide easy ingestion into the NEDF-Portal. The NEDF-Portal utilises the following file naming conventions for spatial and attribute searching, with the “_” used to separate each component of the file name. It is therefore a critical element of the process. There are a number software tools available for renaming existing data files. One used regularly in is the Bulk Rename Utility which can be downloaded from http://www.bulkrenameutility.co.uk/Main_Intro.php. Importantly, ESRI GRIDS cannot be renamed using this tool. Geoscience Australia can make an ESRI GRID renaming tool available by contacting [email protected].

NEDF Data Naming Conventions The NEDF Portal uses 2 types of spatial searching. For ESRI GRIDs it uses the dataset itself to undertake geoprocessing, and for rapid searching it uses the spatial extent of datasets by incorporating the extents into the name of the file. A single file image mosaic is named in a similar manner to a tiled dataset only with the added flexibility of defining tiles of any width and height in addition to traditional square tile. The Portal also uses “_” as a delimiter so it is crucial that you only use these where specified. Using this naming system allows files of any type to be spatially indexed and catalogued. For example, in addition to LiDAR tile and mosaic products, you can also catalogue project reports, pictures or any other reference information and retrieve them through the Portal. The following file naming conventions have been developed to achieve national consistency, to improve dataset management, and to minimise data transfer and ingest costs for both producers and users. Intensity imagery, or other forms of imagery provided This image will generally cover the entire extent of the survey and uses the following filename convention in ECW or geoTIFF format as specified.

Naming Convention for LiDAR intensity or other forms of imagery: ProjectNameYYYY-INT-GSD_xxxyyyy_zz_wwww_hhhh.ecw ProjectName SunshineBathy

etc A meaningful description of the total survey area of interest. Do not use “_” as part of the Project Name. The ProjectName to include either Topo or Bathy to distinguish data type

YYYY 2011 Year of survey INT/RGB -INT Intensity image file identifier. Use RGB for 3 band natural

colour imagery or RGBI for 4 band infrared GSD -002 Ground sampling distance or resolution of image in metres. xxxyyyy _4806558

(480,000mE) (6558,000mN)

Easting and northing value (whole kilometre) of the south- west corner of the tile. A single “_” must be used to separate the remaining file name components.

zz _01 UTM zone of the file wwww _0020 Width of the dataset or tile in whole kilometres hhhh _0050 Height of dataset or tile in whole kilometres For example: SunshineBathy2011-INT-002_4806558_01_0020_0050.ecw


http://www.bulkrenameutility.co.uk/Main_Intro.php

mailto:[email protected]


www.crcsi.com.au 93

<LOGO>

LiDAR Unclassified Point Cloud in LAS Format All LiDAR point cloud data are to be delivered fully compliant LAS v1.2 (or v1.3), Point Record Format

Naming Convention for LiDAR point clouds: ProjectNameYYYY-UNC-DAT-SWT_xxxyyyy_zz_wwww_hhhh.las ProjectName SunshineBathy

etc A meaningful description of the total survey area of interest. Do not use “_” as part of the Project Name. ProjectName to include either Topo or Bathy to distinguish data type

YYYY 2011 Year of survey UNC -UNC Unclassified point cloud. Fully calibrated and adjusted to

specified datum DAT -ELL Ellipsoidal heights (GRS80) SWT -1..n Swath number (1 file per swath) xxxyyyy _4806558

(480,000mE) (6558,000mN)


zz _01 UTM zone of the file wwww _0002 Width of the tile in whole kilometres hhhh _0002 Height of the tile in whole kilometres For example: SunshineBathy2011-RAW-ELL-001_4806558_01_0002_0002.las

LiDAR Classified Point Cloud in LAS Format All LiDAR point cloud data are to be delivered fully compliant LAS v1.2 (or v1.3), Point Record Format.

Naming Convention for LiDAR point clouds: ProjectNameYYYY-CL-DAT_xxxyyyy_zz_wwww_hhhh.las ProjectName SunshineBathy


YYYY 2011 Year of survey CL -C2 classification level. DAT -ELL or AHD Specified vertical datums. Ellipsoidal (ELL) or Orthometric

(AHD) xxxyyyy _4806558

(480,000mE) (6558,000mN)


zz _01 UTM zone of the file wwww _0002 Width of the tile in whole kilometres hhhh _0002 Height of the tile in whole kilometres For example: SunshineBathy2011-C3-AHD_4806558_01_0002_0002.las

LiDAR Classified Point Cloud Model Key Points in LAS Format



www.crcsi.com.au 94

<LOGO>

Model Key points (MKP) are a generalised subset of the original mass points and represent the minimum number of points required to determine the shape of the ground. The filename convention is identical to that above with “-MKP” appended to the classification level: Naming Convention: ProjectNameYYYY-CL-MKP-DAT_xxxyyyy_zz_wwww_hhhh.las ProjectName SunshineBathy


YYYY 2011 Year of survey CL-MKP -C2-MKP classification level and Model Key Point identifiers. DAT -ELL or AHD Specified vertical datums. Ellipsoidal (ELL) or Orthometric

(AHD) xxxyyyy _4806558

(480,000mE) (6558,000mN)


zz _01 UTM zone of the file wwww _0002 Width of the tile in whole kilometres hhhh _0002 Height of the tile in whole kilometres For example: SunshineBathy2011-CL2-MKP-AHD_4806558_01_0002_0002.las

ESRI GRID Format ESRI GRID’s have the following constraints which require specific naming conventions:

a. Names cannot be more than 13 characters b. Names must start with a letter

Due to these constraints the following folder and filenaming convention for ESRI GRIDs must be used for both projected and geographic units. It is also important to note that each individual ESRI GRID must be stored within a standardised folder structure consistent with the following convention to provide appropriate project information to easily associate the ESRI GRID’s with the other files from which they may have been derived. Separate folder structures for the projections are required in addition to each Product Type specified (e.g. DEM, DSM). All ESRI GRIDS must also have all necessary projection definitions populated.



www.crcsi.com.au 95

<LOGO>

Naming Convention for tiled MGA ESRI GRIDS: txxxyyyyssppp

t = surface type.

e Surface type

• s – digital Surface model (DSM) • e – digital Elevation model (DEM) • f - canopy Foliage model (CFM) • c - Canopy elevation model (CHM) • h – Hydro digital elevation model (DEMH) • b – Bathymetry • m – Bathymetry and terrain elevations • t – Derived terrain variables (add as necessary)

xxxyyyy 6458595

(645,000mE)

(8,595,000mN)

• Easting and northing value (whole kilometre) of the south- west corner of the tile.

ss 01 Tile size (km) (square tile)

• 01 – one kilometre • 02 – two kilometre • 05 - five kilometre • 10 – 10 kilometre • _5 (represents half a kilometre)

ppp 001 Ground sampling distance (GSD) or pixel size

• 0_5 - half a metre • 001 – one metres • 002 – two metres etc

For Example: e645859501001



www.crcsi.com.au 96

<LOGO>

Naming Convention for Mosaic (TMG) ESRI GRIDS: txxxxxxxyyppp

t = surface type.

e Surface type

• s – digital Surface model (DSM) • e – digital Elevation model (DEM) • f - canopy Foliage model (CFM) • c - canopy elevation model (CHM) • h – Hydro digital elevation model (DEMH) • b – Bathymetry • m – Bathymetry and terrain elevations • t – Derived terrain variables (add as necessary)

xxxxxxx SunshineLidar A meaningful description of the total survey area and or

sensor, dataset version etc.

yy 11 Year of Survey

ppp 010 Ground sampling distance (GSD) or pixel size in metres (UTM)

UTM

• 0_5 - half a metre • 001 – one metres • 002 – two metres etc

For Example: eSunshineLidar11010



www.crcsi.com.au 97

<LOGO>

Naming conventions for other files The following naming conventions should be used for other file types and formats that may be specified as deliverables.

Naming Convention for all other TMG files: ProjectNameYYYY-SSSS-PPPP-GSD_xxxyyyy_zz_wwww_hhhh.asc ProjectName SunshineLidar A meaningful description of the total survey area of

interest. Do not use “_” as part of the Project Name YYYY 2011 Year of survey SSSS-PPPP -DEM-GRID Surface type.

• DSM • DEM • HDEM • CHM • CFM • Bathymetry (BAT) • Mixed (MIX). Bathymetry and terrain elevations • TTT (Other terrain variables e.g. slope (SLP). Add

as necessary. Product type

• Mass points (MASS) • Breaklines (BRK) • TIN (TIN) • GRID (GRID) • Contours (CON) • Cross Sections (CROSS) • Imagery (BIL, TIF, IMG, ECW etc) • Other

Use additional field width and more characters if required. GSD -010 Ground sampling distance or resolution of product where

appropriate. Where GSD is not required producers can extend the surface type and product description field.

xxxyyyy _4806558 (480,000mE) (6558,000mN)


zz _01 MGA zone of the file wwww _0020 Width of the dataset or tile in whole kilometres hhhh _0050 Height of dataset or tile in whole kilometres ext File extension according to format conventions

• LAS • xyz ascii format for easting, northing, elevation,

intensity • asc – ESRI ascii GRID format • shp • dxf etc

For example: Sunshine-DEM-GRID-010_4806558_56_0020_0050.asc



www.crcsi.com.au 98

<LOGO>

Naming Convention for all other TGD files: ProjectNameYYYY-SSSS-PPPP-GSD_xxxxyyy_wwww_hhhh.ext ProjectName SunshineLidar A meaningful description of the total survey area of

interest. Do not use “_” as part of the Project Name YYYY 2011 Year of survey SSSS-PPPP -DEM-CON Surface type.

• DSM • DEM • HDEM • CHM • CFM • Bathymetry (BAT) • Mixed (MIX). Bathymetry and terrain elevations • TTT (Other terrain variables e.g. slope (SLP). Add

as necessary. Product type

• Mass points (MASS) • Breaklines (BRK) • TIN (TIN) • GRID (GRID) • Contours (CON) • Cross Sections (CROSS) • Imagery (BIL, TIF, IMG, ECW etc) • Other

Use additional field width and more characters if required. GSD 20cm Ground sampling distance or resolution of product where

appropriate. Where GSD is not required producers can extend the surface type and product description field.

xxxxyyy 1185324 (118.5E, 32.4S)

Lower left longitude and latitude ( to 1 decimal place) A single “_” must be used to separate the remaining file name components.

wwww _0015 (1.5deg)

Width of the dataset or tile in whole degrees (including 1 decimal place)

hhhh _0028 (2.8deg)

Height of dataset or tile in whole degrees (including 1 decimal place)

ext shp File extension according to format conventions • LAS • xyz ascii format for easting, northing, elevation,

intensity • asc – ESRI ascii GRID format • shp • dxf etc

For example: SunshineLidar2011-DEM-CON20cm_1185324_0015_0028.shp



www.crcsi.com.au 99

<LOGO>

Attachment C – NEDF Metadata Specifications For each supplied data product a complete metadata statement consistent with the current ANZLIC standard (http://www.anzlic.org.au/infrastructure_metadata.html) is required. Additional metadata specific to LiDAR data is also required.

These metadata may be entered via the ANZMET Lite facility for the general description and via the NEDF Metadata entry tool for the LiDAR-specific data. The two tools are integrated to produce one comprehensive entry. The NEDF Metadata Tool is available by contacting [email protected].

Figure 1 - NEDF Metadata Entry facility


http://www.anzlic.org.au/infrastructure_metadata.html

mailto:[email protected]


www.crcsi.com.au 100

<LOGO>

Attachment D - Submission of Quotation

The following information must be submitted:

a) Details on how the LiDAR work is to be undertaken, including methodology, equipment being used, system calibration, sensor parameters (e.g. pulse rate, footprint size and other relevant technical data), data processing etc;

b) Diagrams of the proposed survey area and flight lines for LiDAR including cross strips for each of the options proposed.

c) Diagrams of the proposed survey area and location of planned ground control and check points, and the origin of points (e.g. field measurement for this project or state survey control) for each of the options proposed.

d) Description of the processes to produce the specified data products and how the specified accuracies will be met.

e) Pros and cons of including the optional products within the project scope, and any specification recommendations.

f) Technical qualifications and relevant experience of the company, project team members and project manager in undertaking airborne LiDAR surveys.

g) Gantt chart or table describing tasks, milestones, deliverables and timeframes in weeks from the day of receipt of purchase order.

h) Statement of compliance against specified deliverables and specifications. Tenderers are to use the statement of compliance template provided in Attachment C.

i) A schedule of service charges against deliverables and submission dates for each of the options proposed.

j) Prices submitted need to be valid for 60 days after the date this offer closes.

k) In addition to the specifications requested, proponents may also wish to offer alternative solutions which could offer cost or time savings to the project.

The criteria for assessing quotations will be:

• Ability to meet the project’s milestones and deadlines for final acceptance of contract deliverables;

• Previous experience and performance of the contractor in relation to bathymetric LiDAR capture and processing;

• The information, options and methodology presented in response to the list of items above;

• Price.




<LOGO>

Attachment E - Quotation Template

The following template must be completed as shown below:

<Company Name>_Quotation_Template.xls




<LOGO>

Attachment F - Ownership/licensing of Foreground Intellectual Property Ownership and licensing arrangements in relation to Foreground IP will be as follows: Note to tenderers: The Contracting Agency placing the Official Order will indicate its required arrangement in relation to ownership or licensing of IP, using the categories below.

[Tick one] Category Description

A Ownership of Foreground IP vests in the Agency submitting the Official Order (Agency). No limits as to use, exploitation, reproduction, adaptation or sublicensing of Foreground IP.

B Ownership of Foreground IP vests in the Contractor. The Contractor grants a non-commercial, perpetual, irrevocable, royalty-free, worldwide, non-exclusive licence (including a right of sub-license) for the Foreground IP to be used, reproduced (including by displaying on a secure network at full resolution and on a public website, for viewing only), adapted and exploited by the licensee and persons and companies undertaking services for, on behalf of, or in collaboration with the licensee. The license may be granted to one or all of the following:

• Australian Government Departments, agencies, authorities and companies (including the Australian Defence Force);

• State and Territory government departments, agencies, authorities and companies; and

• Local/municipal government departments, agencies, authorities and companies, and Natural Resource Management Regional Bodies (as defined by the Australian Government in association with State and Territory Governments)

C Ownership of Foreground IP vests in the Contractor. The Contractor grants a non-commercial, perpetual, irrevocable, royalty-free, worldwide, non-exclusive licence (including a right of sublicense) for the Foreground IP to be used, reproduced (including by displaying on a secure network at full resolution and on a public website, for viewing only), adapted and exploited by the licensee and persons and companies undertaking services for, on behalf of, or in collaboration with the licensee. The license may be granted to Australian Government Departments, agencies, authorities and companies (including the Australian Defence Force);




<LOGO>

Attachment G – Statement of Compliance Tenderers are to state the level of compliance of it’s Tender Response to each Deliverable by inserting one of the following terms against each Deliverable in the appropriate space provided in the table below.

COMPLIES means the requirement or performance standard to be met by the Deliverables to be provided, that the offer shall provide the requirement or standard.

PARTIALLY COMPLIES means the requirement or performance standard can only be met subject to certain conditions. Where this is the case and the tenderer is prepared to make good on the condition, requirement or performance standard the tenderer must explain the technical and cost impact of proposed modifications.

DOES NOT COMPLY means that the requirement or performance standard of the clause is not met by the offer.

COMPLIES WITH ALTERNATIVE means that the tenderer's method, system or process either does not require the feature or the tenderer's method, system or process fully complies in a manner different to that described.

IMPORTANT

In each case where a tenderer’s response is Complies, Partially Complies, Does not Comply or Complies with Alternative the Tenderer is to provide as a separate attachment to their Tender, clarification identifying how the respective response complies, partially complies, does not comply or complies with an alternative including where appropriate, identifying what if any, cost impacts such responses would have on tendered prices.

Compliance - General Specifications:

Description or Deliverable

Statement of Compliance (Complete response using terms indicated above)

Comments or Tenderer’s Reference (including reference to alternatives, modifications or information supporting compliance)

response response


Documents

Bathymetry Acquisition - Technologies and Strategies · The main environmental considerations include the depths, coverage, turbidity, temporal variations and seabed bottom type