A comparison of Landsat 8, RapidEye and Pleiades products for improving empirical predictions of satellite-derived bathymetry

Satellite derived bathymetry (SDB) enables rapid mapping of large coastal areas through measurement of optical penetration of the water column. The resolution of bathymetric mapping and achievable horizontal and vertical accuracies vary but generally, all SDB outputs are constrained by sensor type, water quality and other environmental conditions. Efforts to improve accuracy include physics-based methods (similar to radiative transfer models e.g. for atmospheric/vegetation studies) or detailed in-situ sampling of the seabed and water column, but the spatial component of SDB measurements is often under-utilised in SDB workflows despite promising results suggesting potential to improve accuracy significantly. In this study, a selection of satellite datasets (Landsat 8, RapidEye and Pleiades) at different spatial and spectral resolutions were tested using a log ratio transform to derive bathymetry in an Atlantic coastal embayment. A series of non-spatial and spatial linear analyses were then conducted and their influence on SDB prediction accuracy was assessed in addition to the significance of each model's parameters. Landsat 8 (30 m pixel size) performed relatively weak with the non-spatial model, but showed the best results with the spatial model. However, the highest spatial resolution imagery used – Pleiades (2 m pixel size) showed good results across both non-spatial and spatial models which suggests a suitability for SDB prediction at a higher spatial resolution than the others. In all cases, the spatial models were able to constrain the prediction differences at increased water depths.


Introduction
This Report of Survey applies to data collected by Tenix LADS Corporation during survey operations between 13 May and 16 June 2008 off the north and west coasts of Ireland, including North Galway Bay, north coast of the Aran Islands, Tralee Bay, Blacksod Bay and Lough Foyle, using the LADS Mk II LiDAR system; and provided to the Geological Survey of Ireland (GSI) as an enclosure to this report under contract number LADS_08_01.
In addition data was also collected by a subcontractor, BLOM Aerofilms, using the Hawkeye II LiDAR system in Donegal Harbour, Sligo Bay, Greatman's and Cashla Bay. The collection of this data whilst referred to in this report, will be fully detailed in a separate report by BLOM Aerofilms.
Hawkeye data collected in Greatman's Bay and Cashla Bay has been merged with LADS data collected along the north coast of Galway Bay and is delivered and reported on under this report. Separate products for the Hawkeye II areas of Sligo and Donegal Bay will be delivered separately with the BLOM Aerofilm's report.

Purpose
The survey was conducted as part of the Integrated Mapping for Sustainable Development of Ireland's Marine Resource (INFOMAR) project, which is a follow on strategy from the Irish National Seabed Survey (INSS). Data collected and supplied under this contract will be used by the Geological Survey of Ireland (GSI) and the Irish Marine Institute (MI) to provide sustainable management of marine based activities including: • Mineral, oil and gas exploration and exploitation • Fishing and offshore aquaculture • Marine safety • Coastal zone management • Coastal engineering and erosion • Renewable energy development

Survey Areas
Three survey areas were initially nominated in the contract; these areas were located inside Donegal Bay, Galway Bay and Sligo Bay. One area, Sligo Bay, was conducted solely using the Hawkeye system and will be reported separately; the remaining two areas were planned to be surveyed in a joint operation involving both the LADS and Hawkeye systems. Hawkeye was tasked to survey the intertidal areas of Donegal Harbour at the head of Donegal Bay and Greatman's Bay and Cashla Bay running off the north west coast of Galway Bay, whilst the LADS system was tasked to survey the intertidal areas of north eastern Galway Bay and all coastal areas and deep water areas with the nominated areas of Donegal Bay and Galway Bay.
Whilst the Hawkeye operations in Donegal Harbour were successfully completed in April, algae blooms in May and June resulted in the remaining parts of Donegal Bay being unsuitable for LiDAR survey operations by the time the LADS system commenced survey operations in mid May. As a result LADS operations in Donegal Bay were cancelled and alternative areas were nominated by GSI.
Two alternative areas, Tralee Bay and Blacksod Bay, were nominated. During the course of the survey an additional area, Lough Foyle, was also nominated for operations and the Galway Bay area was extended to include the northern shore of the Aran Islands.
All areas were required to be flown at between 100% and 200% coverage. Coverage flown was varied to maintain the required sounding density in coastal areas and to also allow coverage in The LADS Aircraft VH-LCL arrived at Galway at 1600 Tuesday 20 May. After a full set off checks on the airborne system was completed, a static positional check on all aircraft systems was conducted during the afternoon of 21 May.
Flying operations were commenced with a flight to Galway Bay on 22 May, during which data was successfully collected in all parts of the Galway Bay survey area; weather conditions in Donegal Bay at this time were not suitable due to strong north easterly winds and low cloud. A second successful flight to Galway Bay was achieved on 24 May, however a third flight on 25 May was aborted as the strong north easterly winds generated high levels of turbulence throughout both the Donegal and Galway Bay survey areas. A successful fourth flight was achieved on 27 May to Galway Bay before weather conditions closed in again.
On 28 May LADS surveyors recovered the tide gauge at Mullagmore Harbour, Donegal Bay, and installed it at Feint Harbour in Tralee Bay during the morning of 29 May. This provided an alternative to the Donegal and Galway Bay areas during the periods of poor weather. During the afternoon of 29 May the alternative Tralee Bay survey area was activated with a successful fifth flight completed and extended via a refuelling stop at Kerry airport. A sixth flight was achieved in Galway Bay on 31 May in relatively calm conditions. During this flight operations were attempted in Donegal Bay but had to be aborted due to the presence of an algae bloom caused by warming water temperatures within the bay. A seventh flight was aborted on 01 June as a large bank of sea fog moved into the Tralee and Galway Bay survey areas.
A successful eighth flight was flown to Tralee and Galway Bays on 02 June, working around a storm front that moved through the area during the afternoon. The ninth flight on 03 June was aborted due to poor water quality in Donegal Bay and strong winds and rough sea conditions in Galway and Tralee Bays. Further operations in Donegal Bay were officially cancelled after this flight, with efforts now to be concentrated in Galway Bay, Tralee Bays and extended into Blacksod Bay.
Flight approvals from the United Kingdom for operations in Northern Ireland were received at the start of June. As the next neap tide window opportunity for operations in Lough Foyle was mid June the deployment was extended.

Survey Datum
All depths are reduced to Malin Head (MH) as defined by observed tides connected to local survey marks in Galway Bay, Blacksod Bay and Tralee Bay. All depths in Lough Foyle are reduced to Belfast Lough (BL) as defined by observed tides connected to local survey marks. A separate digital dataset of all areas reduced to Lowest Astronomical Tide (LAT) was provided for delivery to the United Kingdom Hydrographic Office (UKHO) for charting purposes.

Tides
Observed tides were used to reduce all soundings to sounding datum Observed tides were supplied from a combination of temporary gauges installed by Tenix LADS Corporation surveyors; temporary gauges installed by GSI staff and permanently installed tide gauges operated as part of the Irish National Tide Gauge network. Separate tide models were established for each of the survey areas, a total of 12 tide stations were incorporated into 24 tide areas.
Full details of the tidal model and the algorithms used for this survey are provided in Annex E.

Bathymetry
Main lines of sounding were conducted using 5x5 metre laser spot spacing with a swath width of 240 metres, at an aircraft ground speed of 175 knots. Main lines of sounding were flown at 110 and 220 metre spacing which provided 100% or 200% coverage of the seabed.
Main lines of sounding were orientated along the longest side of each defined survey polygon, with at least one cross line flown at right angles to the main lines in each survey polygon.
Data quality is dependent on the signal to noise ratio of the laser return from the seabed. The quality of data collected during this survey varied from area to area. In general, very good data quality was achieved in flat or shallow sandy areas, whilst the steeper rock areas had increased levels of noise, possibly due to high levels of seaweed and kelp.
A full description of the results achieved in each study area is provided in Section 4.

Footprint of the Laser Beam
At the sea surface the footprint of the laser beam is approximately 2.5 metres in diameter. As the beam passes through the water column it diverges slightly due to scattering.

Depth Benchmarks
A benchmark line was surveyed north of the Aran Islands at the start of the survey; however the highly variable water clarity and weather conditions meant that no meaningful data was collected over the benchmarks on subsequent sorites. In addition, the wide geographical separation of each of the five survey areas meant that there was little opportunity to obtain repeat observations at the one location. As a result no meaningful benchmark comparisons were observed, and historical values for the system accuracy have been adopted and used in the calculation of the overall vertical accuracy of the depths.
The overlap between lines was consistent with the LADS Mk II system operating correctly, that the tidal model was consistent and that tides were correctly applied.

13
Report of Survey tended to be shallower than the surrounding sandy areas. Least depths in the rocky kelp areas may not have been achieved, despite being flown at 200%, Full coverage was achieved inside Greatman's Bay and Cashla Bay using the Hawkeye II system, with overlap between the LADS and Hawkeye systems datasets achieved.

North Coast Aran Islands
This area was flown at 5x5m spot spacing at 200%. Depths were achieved from drying 50 metres to depths of approximately 15 metres relative to Malin Head. The required survey area was fully covered.
Excellent coverage was achieved in the shoaler areas less than 10 metres deep. Deep areas showed signs of increased levels of noise on the waveforms near the seabed and as such the data is slightly noisier in nature. This increased level of noise was reflected in the cross line comparisons.

Tralee Bay
This area was flown at 5x5m spot spacing at 100% to 200%. Depths were achieved from drying 50 metres to depths of 10 to 15 metres relative to Malin Head.
Coverage was achieved throughout the main part of Tralee Bay in depths less than 10 metres, including inside Fenit Harbour and Barrow Harbour. Parts of two lines, one on the north side and one on the south side of the intertidal harbour east of Derrymore Point had to be rejected during the processing due to poor quality leaving striped holes in the dataset. Likewise parts of three lines in the intertidal areas of the northern part of Barrow Harbour and part of one line over the Fahamore Peninsula also had to be removed. Data collected along the northern edge of the survey area in the vicinity of the Seven Hogs and Rough Point in the north west, and Muckaghmore Rock in the north, and west of the Fahamore Peninsula was of a slightly sparse and noisy nature due to near seabed turbidity caused by swell and large kelp areas. Full seabed coverage was not achieved inside the kelp areas where the data was generally shoaler in nature and lesser depths may still exist. This area was outside the required survey area and so further effort into improving coverage in this area was not made.

Lough Foyle
This area was flown at 5x5m spot spacing at 200%. Depths were achieved from drying 50 metres to depths of 10 metres inside the Lough, and 30 metres seaward of the Lough. All depths were reduced relative to Belfast Lough. Operations were suspended once turbid conditions were encountered in the southern and south eastern parts of Lough Foyle. It may be possible to extend coverage further to the south east during an appropriate neap tide window at a later time.

Blacksod Bay
This area was flown at 5x5m spot spacing at 100%. Depths were achieved from drying 50 metres to depths of 25 metres relative to Malin Head. Operations in Blacksod Bay were optimised to maximise coverage in depths to 10 metres as required by the contract and so deeper coverage was not achieved in the southern part of the bay where it may have been possible to obtain deeper depths.
Only one flight was available for operations in Blacksod Bay, thus the survey of Blacksod Bay was progressed from the south to the north until the completion of the sortie. The northern part of the bay remains unsurveyed, but could easily be surveyed by LADS at a later time.
15 Report of Survey

Navigational Aids
Many navigation beacons exist within the survey area marking channels into Tralee Bay and Galway Harbour. Where detected by the system they have been retained, tagged with the appropriate S-57 tag and listed at Annex K. Due to the small size of some of these beacons, not all beacons were detected by the system.
There were also many minor jetties, buoys and moorings detected, however these have not been attributed in the data. The geo-referenced digital imagery provides excellent coverage for the purpose of identifying these features.
The light characteristics of the beacons have not been checked, nor are they indicated on the sheet drawings. Contours around beacons have been removed from the drawings.

Wrecks and Obstructions
No charted wrecks exist within the survey areas. No wrecks were identified in the data.
Two charted obstructions exist in the Lough Foyle survey area; these features have been identified in the data. Full details are provided at Annex J.

Coastline and Topography
Within the survey area, soundings were collected to approximately 50 metres above the drying line.
The topographic data has been checked for consistency; topographic data with anomalies, such as 16 Report of Survey errors in datum, has been rejected and removed from the dataset. Individual topographic points have not been validated. The accuracy of the topographic dataset has been assessed to be better than ±1m (95% confidence).
Man made features such as buildings, roads, bridges and breakwaters were also surveyed and are present in the topographic data.

Features Recommended for Further Investigation
During the survey a number of features were identified inside Lough Foyle which appear on the datasets as shoals and correspond to charted obstructions. The return from the laser of these features was strong enough to indicate a feature exists and they have been retained in the dataset. No additional examination of these features was conducted. Features which fall into this category have been tabulated in Annex J.

Hawkeye II Operations
A separate report will be provided to cover the Hawkeye operations. A brief description of the Hawkeye system and processes is provided at Annex N.

Quality Control of Hawkeye data
A comprehensive review and quality control process was conducted by LADS to quality control data supplied by BLOM Aerofilms and incorporated into the Galway Bay survey areas. This process has been described in Annex N.

Merging of LADS / Hawkeye Datasets
LADS Galway Bay data was merged with the Hawkeye Greatman's Bay and Cashla Bay data to create a merged data set for use in the generation of the required products. For the creation of the merged data file, the fields of 'Eastings', 'Northings' and 'Depths' were extracted from the Hawkeye raw data files for the Malin Head datum reference, and from the LADS data, also to the Malin Head datum reference. The output files were then merged using an AWK script within UNIX. This created an ASCII XYZ file, which was imported into Fledermaus and visualized to confirm correspondence.
This merged LADS / Hawkeye data set was used in the generation of sheet LADS 04, and to create the average depth grid for this sheet.
The Hawkeye LAT data for Greatman's Bay and Cashla Bay has not been merged and will be supplied separately.

Digital Aerial Mosaic
During flying operations digital photographs were taken at 1 Hz (1 second) intervals using a downward looking geo-referenced two mega pixel digital camera. These images were then processed to provide 40cm per pixel resolution. Full coverage was achieved.
Images captured in the main survey areas were then joined together using a semi-automated processing technique to provide a total series of 47 geo-referenced aerial mosaics in 8km tiles, which have been provided as digital data.
These individual tiled mosaics have then been joined together in ArcGIS to form overview mosaics at a reduced resolution for each individual survey area. Full details and a copy of each overview mosaic are provided at Annex M.

Relative Reflectance Data
Relative reflectance data was been generated from the accepted data in each of the main survey areas.
These files have been generated as ASCII files with the suffix RR1. The file format consists of the beam footprint position in easting and northing then a reflectivity value between 0 and 255. Full details on how the reflectance data was generated area at Annex O.

Interim Data
An initial set of interim digital data of Galway Bay and Tralee Bay was provided prior to the departure of the survey team on 14 June. A follow on set of interim digital data of Blacksod Bay and Lough Foyle was provided on 31 July 08.

Final Digital Data
An initial set of final digital data for Galway Bay, Blacksod Bay and Lough Foyle along with digital copies of the associated fairsheets for this data was delivered on 14 November 2008.

Report of Survey
Final delivery of all final approved processed data for all survey area, including those previously delivered, accompanies this report; this data has been provided in the following formats: • 4m shoal bias clashed CARIS .CAF export format for each sheet relative to Malin Head (relative to Belfast Lough for Lough Foyle data), • 4m shoal bias clashed simplified ASCII x,y,z format for each sheet relative to Malin Head (relative to Belfast Lough for Lough Foyle data). Note for Sheet 04 this data contains merged LADS / Hawkeye data, • A 6m average grid DTM, ASCII x,y,z format file for each sheet relative to Malin Head (relative to Belfast Lough for Lough Foyle data), Note for Sheet 04 this data contains merged LADS / Hawkeye data, • A MicroStation .DGN file for each sheet, • A .PDF file of each sheet, • Relative Reflectance data set in .RR format for each sheet, • 4m shoal bias clashed CARIS .CAF export format for each sheet relative to Lowest Astronomical Tide (LAT).
• A Digital Aerial Mosaic of each survey area in GeoTIFF format, • 8 km Tiled Digital Aerial Mosaics at 40cm resolution.
All digital data has been supplied on a USB Hard Disk, with full details of the supplied data provided at Annex A

LADS Mk II Digital Surveying System
The LADS Mk II hydrographic survey system comprises two main sub-systems: the Airborne System (AS) used for acquiring raw bathymetric data, and the Ground System (GS) which is used to plan operations, calculate depth values from the raw data, provide tools which allow the hydrographic surveyor to validate processed depth values, apply tidal corrections, generate fairsheets and digital survey data and conduct general survey management. Other tools required for quality control activities, in particular contouring and 3-D visualisation software complement these two sub-systems. GPS logging and data processing hardware and software are also provided.
All sounding data are acquired by the AS which is mounted in the LADS Mk II DeHavilland Dash-8 fixed wing aircraft.
The GS software is supported by the UNIX operating system and operates on a ground based server.
Prior to a sortie, planning information is passed from the GS to the AS on disk. During the sortie, logged raw sounding, position and airborne system data is logged on digital linear tape. This is processed on the GS at the completion of each sortie.
The primary Quality Control tools used during this survey were: Post-processed GPS positioning is accomplished with Ashtech logging software (datalogr) and post-processed using Novatel GrafNAV software.

B.1 Equipment
This section provides a description of the LADS Mk II Airborne System (AS) and the Ground System (GS).

B.1.1 Airborne System
A laser, scanner, optical system, photo-multiplier tube and conditioning electronics collect the raw sounding signal. These items are mounted on a stabilised platform controlled via servo systems using information from an Attitude and Heading Reference System (AHRS) mounted on the platform. Aircraft position information is obtained from the Global Positioning System. Three computers, linked via an FDDI optic fibre network, control and monitor the AS operations. These computers are:

•
The System Control Computer (SCC) for operator interface, logging and overall system coordination.

•
The Navigation System and Support (NSS) computer for position monitoring and control.

•
The Laser Control and Acquisition (LCA) computer for control of the scanner and laser and digitisation of raw sounding data. The LCA also synchronises overall AS timing.
AS system time is synchronised with GPS time and all data acquired for logging is appropriately time stamped at the point of acquisition then passed to the SCC to be written to digital linear tape.
Ancillary equipment includes: • A downward looking video camera and VCR to provide images below the aircraft and a forward looking video camera.
• A downward looking Redlake MegaPlus II ES 2020 digital camera to capture digital imagery below the aircraft.
• Systems for temperature control of equipment.
• VHF transceiver and aircraft intercom.
The operator interface allows the operator to monitor the quality of sounding, position and other data in order to set appropriate system parameters and control the sequence of sortie operations.

B.1.2 Sounding equipment
Soundings in the LADS Mk II system are obtained by the transmission of laser pulses from the aircraft through a scanning system and detecting return signals from land, the sea surface, the water body and the seabed. The transmitting and receiving components are housed on a stabilised platform that compensates for aircraft pitch and roll. The return signals are electronically amplified and conditioned prior to being digitised and logged.
The primary sounding components of the AS are: • Laser. A Nd: Yag laser producing infrared energy at a wavelength of 1064nm at 990 pulses per second of which 900 pulses are used for sounding purposes.
• Optical Coupler. The optical coupler is used to split the infrared beam. Part of the IR beam is transmitted vertically to nadir on the sea surface. The other part of the split beam is frequency doubled to produce green laser pulses of wavelength 532nm. The green pulses are transmitted onto the mirror of the scanner.
• Scanning System. The scanning mirror is oscillated in both the major (across track) and minor (along track) axes. The required scan pattern is generated by controlling software. All possible patterns are listed in B.9, Sounding Patterns section.
• Optical Receivers. The IR and green return signals are detected by two separate receivers. The IR return from the surface of the sea is used to establish a height datum. The IR receiver is a solid state detector producing an electronic signal from the IR return. The green return comprises energy returned from the surface, subsurface and seabed and is used to determine water depth. The green return is transmitted via the scanner into a photomultiplier tube. The electronic output of the two return signals are electronically mixed prior to digitisation.
• Attitude and Heading Reference System (AHRS). The AHRS is a laser gyro inertial navigation system providing platform attitude information to the platform servo system that in turn maintains platform stability. The AHRS also reports platform attitude to the LCA computer and provides height data.
• LCA computer. This controls the laser and scanner operations and digitises (8 bits at 500MHz) appropriate sections of the composite electronic red/green return signal along with platform attitude data and other system parameters. This digital information is passed to the System Control Computer (SCC) where it is logged to digital linear tape.
• Waveform Display. This CRT display presents the operator with sounding waveforms as digitised and is used by the operator to check data quality during acquisition.
Tenix LADS Corporation Pty Ltd.
B-4 Report of Survey

B.1.3 Position equipment
The centre of the scanning mirror is the survey reference point on the aircraft. The GPS antenna is positioned relative to this point as described in B.7, Laybacks.
The signal from the antenna is split and fed to two independent GPS receivers: an Ashtech GG24 single frequency GPS receiver used for real-time aircraft position fixing and track keeping and an Ashtech Z12 dual frequency GPS receiver is used to compute post-processed KGPS positions. The data from this receiver is independently logged and post-processed as described in Annex C.
The output of the real-time GPS receiver is fed to the NSS to: The NSS passes the received GPS and derived information to the SCC computer for logging.

B.2 Sortie control
A sortie plan is generated on the GS to transfer survey information to the AS. The sortie plan contains spheroidal, grid and magnetic variation parameters and a list of survey objectives including the line number, start/end coordinates and coordinates for navigation checks. During the course of the sortie, the airborne operator amends the sequence of execution to suit local conditions and can amend the scan pattern parameters for the survey lines to suit survey requirements.
The SC computer controls the sequence of survey operations by: • planning all required flight paths and communicating these to the NSS • transmitting required parameters for scan patterns, aircraft altitude, etc. to the LCA • initiating the starting and stopping of system operations, via commands sent to the LCA and NSS at specific waypoints on the run-in and run-out of survey lines.
The operator may abort and restart the sortie operations at any time and the sequence of objectives may be amended at any time. Scan patterns can be amended on all lines except the executing objective. A display of the planned survey line and received GPS data is situated in the cockpit and used to advise the pilots of required aircraft configurations. The display provides an indication of cross track error with required and actual values for altitude and ground speed.
Aircraft position during survey acquisition is under automatic control of the NSS via the aircraft autopilot. Aircraft turns are under pilot control assisted by the display. Aircraft altitude and speed are under pilot control, and communication between the operator and pilots is via the aircraft intercom system.
The management of survey operations can be impacted by both low cloud and high ground in the survey area. LADS Mk II is able to operate at different survey heights so that adequate clearances can be maintained while surveying and survey activities can continue below low cloud ceilings. Survey altitudes at 200ft increments are available from 1200 to 2200 feet (366 to 671m). Altitudes must be constant for the duration of a survey line but may be varied from line to line by the AS operator during the course of a sortie.
During daytime operations a narrow band green filter is used to filter out other light frequencies from the photomultiplier tube. This filter has a slight attenuating effect on the laser returns, which reduces the maximum depth performance. This filter can be removed once the ambient sunlight levels drop which results in improved performance at night.
Glassy sea conditions may result in very strong IR surface returns that can saturate the IR receiver causing a loss of surface datum. The AS monitors the IR surface return performance and advises the operator if IR saturation occurs. The operator can activate an attenuator that provides correct IR surface return amplitudes to be fed to the IR receiver. Should sea surface conditions change which may result in lower IR return amplitudes the AS informs the operator to deactivate the attenuator.
The laser is designed to be eye safe in accordance with the following standards: The laser power can be reduced by a further factor of four using a built-in attenuator. The operator may activate/deactivate the attenuator at any time.

B.3 Ancillary equipment
A video camera is positioned on the stabilised platform and directed downward at nadir. A calibrated graticule is superimposed on the camera image to provide the operator with a scan width and distance reference. The image, graticule and other relevant system information including position and time are presented to the operator and recorded throughout a sortie.
A forward-looking video camera is also provided to assist the AS operator for the purpose of evaluating the conditions ahead of the aircraft. A digital imagery system provides geo-referenced imagery. This system comprises of a RedLake MegaPlus II ES 2020 digital camera, a Matrox 4sight M frame grabber and a Matrox embedded computer running Windows EP embedded operating systems. Images are taken at one-second intervals with a 1600x1200 resolution and a 2-megapixel interline-transfer camera head and controller. At the end of each sortie, the images are copied to the LADS GS using a removable hard drive.

B.4 Operator interface
The operator monitors and controls system operation from the console. The following key information is provided to monitor system performance: • Sortie Information. The Sortie ID, spheroid and grid in use and available survey objectives are displayed. Sortie objective information includes the scan pattern set for the objective and estimated time to complete the objective.
• Objective Information. The Objective ID, selected scan pattern, required speed and altitude pertaining to the current objective being executed and objective status such as time to completion are presented.
• Waveform Display. This display is a CRT on which is displayed each of the mixed red/green sounding return signals as digitised by the LCA (the traces are overlayed). The operator continually assesses this display to determine data quality.
Tenix LADS Corporation Pty Ltd. B-6 Report of Survey • Depth Profile. A depth profile determined from nadir soundings is available to the operator with an associated confidence factor. As the algorithm is limited by real-time considerations these depths and confidences are only indicative.
• Aircraft Position, Speed, Altitude and Cross Track Error. A number of displays including a copy of the pilot display are available to the operator to determine the aircraft position and performance parameters. Speed and altitude are continually monitored and the pilot informed of deviations from the desired values.
• GPS status. The operator is provided with the data from the GPS receiver including number of satellites, satellite altitudes and azimuths, S/N ratio and which satellites are being used.
• Equipment Status. System status and performance parameters are available to the operator including laser power and temperature, dynamic gain values, AHRS status and scanner performance.
Items controlled by the operator for sortie execution and data acquisition are:

B.5 Depth and topographic mode
During normal bathymetric survey mode (Depth Mode) LADS Mk II determines the depth of water with the height datum being determined from the reflected IR laser signal, GPS height and AHRS height. When over land this IR signal is not valid and the height datum is obtained from the GPS and AHRS.
This ancillary height datum allows LADS Mk II to measure topographic heights. The topographic height range is dependent on the depth range being used. Digital Image Horizontal Accuracy: +/-5m (95% confidence).

B.7 Laybacks
All laybacks are measured relative to the survey reference position on the aircraft which is the centre of the scanning mirror. The GPS antenna used for position determinations in the AS is positioned on the upper side of the aircraft fuselage forward and to the left (facing forward) of the sounding reference position. The signal from this antenna is passed to a splitter, one signal going to the GPS receiver in the Navigation System computer and the other passes to the GPS airborne logger.
Offsets are from the sounding reference point to the antenna with the following axis and sign convention assuming the aircraft is level: The Airborne System obtains a position fix every 0.5 seconds.

B.8.2 Navigation update
While executing a survey line under AS control navigation correction is passed to the aircraft autopilot every 0.5 seconds.

B.8.3 Post-Processed GPS
The GPS airborne and base logging stations log position information from GPS satellites at 0.5 second intervals.

B.8.4 Sounding rates
LADS Mk II obtains depth soundings in a rectilinear pattern where the sounding density is variable (see Table 1) but sounding rate is invariant.
For all sounding patterns the soundings are grouped into one second frames made up of 18 scan lines. Each of the 18 scan lines contain 50 laser pulses, of which 48 pulses are used for depth sounding. The outermost laser pulses are not used for depth sounding. This provides an effective sounding rate of 864 soundings per second.

B.9 Sounding patterns
LADS Mk II has variable scan pattern functionality as detailed in the following

B.10 Ground System
The GS provides the facilities for all LADS survey management tasks from initial mission planning through to production of fairsheets and deliverable digital data.
The primary functions are: • Mission planning. This includes the specification of the total survey area, spheroid and grid, survey sub-areas, line spacing, swath widths, survey lines to cover the sub-area, individual survey lines, cross lines, tidal areas and navigation check points.
• Sortie planning. A sortie plan is the specification of a series of survey objectives to be executed by the AS. Survey lines and navigation check objectives are selected by the operator and written to disk along with grid and spheroidal information.
• Sortie processing. This function calculates sounding depths and positions from the raw sounding data logged by the AS. Depths and positions are associated with various confidence metrics.
• Data validation, checking and approval. Surveyors validate the calculated soundings on a run by run basis editing soundings as appropriate. The validated data is checked by a more senior surveyor and finally approved by the Field Party Leader.
• Data output. Approved data is output to the client in digital form along with hardcopy fair sheets.
In addition, the GS provides facilities for the generation of survey management plots and reports.

B.10.1 Mission planning
At the commencement of a survey one or more databases are established on the GS. Each database contains spheroid and grid data, tide data and survey objectives.
Sub-areas are defined covering the specific areas to be surveyed. Survey lines are then generated within each sub-area at an operator specified line spacing. Other survey lines can be specified by entering start and end coordinates.

B.10.4 Data organisation
Data within the GS database is held on a line by line basis. Within lines, data is grouped into one second frames made up of 18 scans of 48 sounding pulses ie. 864 pulses per frame. (The outer two laser pulses are not used for sounding purposes.)

B.10.5 Primary and secondary soundings
All soundings comprise the primary sounding set. Where data set reduction is required a shoal biased subset of the primary soundings called secondary soundings is created. Secondary soundings form a shoal biased sub-set based on operator selected confidence and secondary selection radius criteria. Only secondary soundings are validated, checked, approved and output. For this survey a secondary sounding reduction radial of one metre has been used which means all soundings have been hydrographically reviewed and all valid soundings have been provided in the final data set.

B.10.6 Automatic data processing
Automatic processing is completed in two stages:

1.
Sortie Tape Processing (STP). STP reads the data on the tape and stores it in the internal GS database for further processing. The data is line based, and consists of raw waveform data, navigation data, platform data, system data, and error and event logs. This process also includes producing a backup of the Raw Data Tape on DAT or DLT.

2.
Sortie Run Processing (SRP). SRP is the second and major processing phase during which sounding depths and positions are calculated on a line by line basis. The process is normally triggered automatically by STP as each line becomes available, but may be invoked later by the operator if reprocessing of lines with different processing parameters is required.
The major processing steps of SRP are: • Apply post-processed KGPS positions to the raw data and digital images from the downward looking camera.
• Process the Raw Waveform to identify surface reflections.
• Process the Raw Waveform to identify and calculate initial depths for the two most likely bottom return pulses.
• Classify each of the identified bottom return pulses by signal noise ratio, agreement with near neighbours and a maximum likelihood estimator.
• Select the most likely bottom return pulse based on the above classification and a shoal weighting function.
• Model the sea surface from the available surface pulses.
• Correct the bottom depths for sea surface datum including tide, slant range, optical propagation and early/late entry. Tidal corrections may be reapplied later if required.
• Calculate position of each sounding on the seabed. This algorithm uses corrected GPS fixes, aircraft track and speed, antenna offsets, platform attitude (heading, roll and pitch), beam scan angles and sounding depth. Where the GS is unable to determine a depth from the raw data the sounding is classified as "No Bottom Detected" (NBD).
Tenix LADS Corporation Pty Ltd. B-12 Report of Survey

B.10.7 Bottom Object Detection (BOD)
A particular feature in the SRP improves the ability of the LADS Mk II GS to detect small objects on the seabed.
The BOD algorithm proceeds in two phases, each phase can be independently enabled/disabled and tuned via a series of BOD processing parameters set by the operator prior to SRP processing.
Phase one of the algorithm is designed to detect objects 2-3m in height while phase two is only invoked if phase one fails. Phase two is more sensitive and intended to find objects less than 2m in height.

B.10.8 Line reprocessing and segmentation
It may be necessary to reprocess the same raw sounding data with different processing parameters. The run identification scheme adopted in LADS Mk II provides a mechanism to manage the reprocessing of survey line data a number of times.
After a line is reprocessed the required segment can then be set to accepted, and the remaining data can be set to anomalous or rejected and is therefore ignored by the system.

B.11.1 Data processing
Data processing involves the following stages: • Automatic Data Processing, described earlier

B.11.2 Validation
Validation proceeds through the following steps: Examining the Depth Profile for the correct processing of each expected Survey Run.
Examining a range of position, coverage and system performance confidences to ensure only good data is accepted.
Resolving anomalous soundings by examining data points in the Survey Run by checking: All operator interactions during the validation phase are logged so that complete traceability is maintained.
The imagery, collected by the downward-looking digital camera in real-time, is processed along with the raw data. These images are geo-referenced and can be either manually or automatically displayed alongside of the Raw Data Display, the Waveform Display or the Local Area Display. The images are automatically rotated to fit the current display and are used during all phases of data processing.
These images are displayed in the GS Digital Image Window on the second dual screen monitor. This display is automatically linked to all of the GS displays mentioned previously and the selected sounding is highlighted in the downward-looking image with a yellow circle of 5m diameter.

Figure 2
The GS Digital Image Window enables the operator to easily correlate features such as coastline, islands, islets, drying rocks, rocks awash, shallow rocks, kelp, beacons, buoys, boats, jetties, buildings and trees in the image with the data presented in the different GS displays. The quality of imagery and zoom functionality of the window even enables discernment of biological data artefacts, such as bird strikes and whale returns.

B.11.3 Checking
When a line has been validated it is passed to a checker. All edits made by the validator are marked on the line and logged in a validation log. The checker independently assesses the line and checks the validation edits.

B.11.4 Data visualisation
All validated and approved data is exported from the GS in a defined ASCII format for spatial presentation and checking. The position, depth, run and other relevant information are extracted from the line-based data for use in the generation of TIN models and gridded data sets. Both of these are used to produce contour plots, sun-illuminated colour banded images and coverage Tenix LADS Corporation Pty Ltd. B-15 Report of Survey check plots. Anomalies found in these plots are reported back to the checkers for remedial action in the GS.
A number of software packages are used to produce these QC products namely:

B.11.5 Approval
In the final phase an IHO Category A qualified Hydrographic Surveyor reviews each line and approves the data for delivery. All actions in validation, checking and approval are logged on appropriate forms and the procedures used have been certified as conforming to ISO-9001 Quality Assurance standards.

B.11.6 Audit trail
All actions in validation, checking and approval are logged on appropriate forms and the procedures used have been certified as conforming to ISO-9001 Quality Assurance standards. In addition, all operator actions are logged by the GS.

B.11.7 Tagging of Soundings in the GS
During data processing on the GS, the operators have the ability to assign S-57 and user-defined tags to gaps and features in the data. This enables accurate delineation and attribution of features using the S-57 feature file.
All actions in validation, checking and approval are logged on appropriate forms and the procedures used have been certified as conforming to ISO-9001 Quality Assurance standards. In addition, all operator actions are logged by the GS.  Child -Range 1..9 This field denotes the segment (or child section) of a <LineNumber> .<Section>.<Sequence>.
Hydrographic surveyors divide lines into ACCEPTED, REJECTED or ANOMALOUS segments during the Line Validation process, these segments are given sequential child numbers. Thus: 498.0.1.1 -is the first child (segment) of the first processing of the original line. This provides the mechanism of ensuring only ACCEPTED data is output for products.

B.11.10 Software versions
The following software versions were used during survey operations.

System
Version Remarks

B.11.11 Processing parameters
Each survey line is processed with a specific set of processing parameters, with the set used for the line recorded on the Survey Line History Sheet for the line. Full details are recorded in the Survey Data Management Folder held by TLC.

B.12 Data output
The data is delivered in XYZ, ASCII format based on the sheet limits clashed at a 3 metre radial range. All files have been written to DVD (Juliet format)

B.12.1 File Naming Conventions
All file prefixes are named using the following convention.

C.1 Geodetic Parameters
The following are the parameters for the Geodetic Datum used for data delivery. Throughout the survey, the real-time position of the LADS Mk II system was derived from an Ashtech GG24 GPS receiver. WADGPS corrections from the Fugro OmniStar Virtual Base Station (VBS) service were received using an OmniStar 3510LR system and applied to the raw GPS position as received by the Ashtech GG24 GPS receiver.

C.2.2 LADS Local GPS Base Station -Radisson SAS Hotel, Galway
A local GPS base station was established by LADS surveyors on 19 May at the Radisson SAS Hotel in Galway. This station was coordinated using static GPS methods. A check of the base station coordination was performed by Arrigan Geo-Surveyors using rapid-static GPS methods. The static GPS data was processed using Waypoint GrafNet software and constraining the station using IGS continually operating GPS stations nearby. Refer to enclosure 2 for the processing and adjustment report.
The derived ETRS89 coordinates for the local GPS base station are:

C.2.3 Static Position Check Control Marks -Galway International Airport
Arrigan Geo-Surveyors undertook the coordination of the static check marks at Galway International Airport using a combination of rapid static GPS and terrestrial methods on the 19 May 2008. The report of survey from Arrigan Geo-Surveyors is enclosed in Enclosure 2. Table 3 gives the coordinates of the static marks. Enclosure 3 shows the layout of the static position check marks and the laser source mark.

C.2.4 Derived Antenna Position for Static Position Check
For the static position check the aircraft was parked in the centre of the three control points. The position of the laser source was then plumbed down from the aircraft to the tarmac and marked with a nail. This position, the laser source mark, was derived by measuring the distances between it and the static position check marks and elementary trigonometry was used to derive the coordinate. To derive the antenna position the aircraft heading was determined using the AHRS Gyro Compass Alignment routine, this was found to be 165º.

D.2 GPS Base Station Coordinate Validation
Arrigan Geo-Surveyors of Galway provided a check on the LADS GPS Base station at the Radisson SAS Hotel. A short static session of 1 hour was used by Arrigan Geo-Surveyors to obtain the coordinates. The differences in the coordinates calculated by LADS and Arrigan are shown in table 1.

D.3.1 Session 1 -WADGPS
Session 1 used real-time differential corrections from the Fugro OmniStar VBS WADGPS service combined with stand-alone GPS from an Ashtech GG24 GPS receiver to provide a differentially corrected real-time position for the Airborne System. The local GPS base station and the roving receiver commenced recording GPS data a short time before AS logging commenced. The Fugro OmniStar VBS WADGPS corrected position was recorded on tape using the GPS manual logging function on the AS. This position was recorded for two hours.

D.3.2 Session 2 -GPS Only
Session 2 used no real-time differential corrections. The local GPS base station and the roving receiver began recording GPS data a short time before AS logging commenced. The AS was set to receive no differential corrections and this resulted in a stand-alone GPS position. The position data was recorded on tape using the manual logging function on the AS. This position was recorded for two hours.

D.3.3 Observations
The observation periods were as follows: The AS GPS observables were recorded manually every ten minutes. The easting, northing, height, PDOP, EHE, EVE and number of GPS satellites used were observed and recorded.

D.3.4 Processing
The KGPS positions were produced by processing the reference station file and the aircraft file with GrafNAV software. A KGPS position is produced by solving for the carrier phase ambiguity and using double differencing and forward and backward processing techniques. Both the standalone GPS file and the WADGPS file are produced in real-time on the AS and the solution was logged directly to tape. The files were then processed using Position Analysis Software on the GS.
Tenix LADS Corporation Pty Ltd. D-3 Report of Survey

D.3.5 Results
The final positions were exported to a commercial spreadsheet/graphical based software package where calculations of means and standard deviations were completed and scatter plots produced. The calculation of these results was completed in a commercial spreadsheet application, this sheet in enclosed in Enclosure 2. 1: WADGPS corrections supplied by the Fugro OmniStar VBS service were used in realtime to control the aircraft navigation during the flight. Accuracies achieved by this service will be similar to those achieved in the static position check.
A compilation of graphs illustrating the spread of solved positions for each positioning system is provided at Enclosure 1. These graphs show the mean point of recorded positions and the position of the actual antenna as determined by coordination of the laser source using resection techniques and applying the laser source -antenna offset.

D.3.6 Conclusion
The absolute accuracy of the logged WADGPS position solution based on a virtual base station within the survey area was consistent with previous results and was sufficient for the real-time positioning of the aircraft.
The KGPS position yielded a more accurate result and this positioning solution was subsequently applied to all survey data.
The position check of the three systems shows that there are no gross errors.

D.4 Dynamic Position Check
During each sortie, GPS data was logged both on the aircraft and at the base station which enabled a KGPS position solution to be determined. These positions were then compared to the position as determined by the real-time positioning system. For each survey line the mean difference and standard deviation have been calculated. Table 5 shows the mean and standard deviation of the difference in position between the real-time positioning system and the postprocessed KGPS for each data collection sortie.

Lines Flown
Max. Difference

AS -KGPS (m)
Overall These results show good agreement between the real-time position and the post-processed KGPS position. An extract from the dynamic GPS position check report for Sortie 5 is provided in Enclosure 3.
Tenix LADS Corporation Pty Ltd. D-6 Report of Survey

D.5 Navigation Check
Navigation checks were conducted over the GPS Base Station on the rooftop of the Radisson SAS Hotel in Galway on five occasions during the survey.

D.5.1 Navigation Check Results
The logged aircraft position over the base station during the position check was processed against the downward looking digital camera record to determine the difference in position at the time of overflight. This provided a gross error check of the aircraft positioning.
A position for the coordinated mark was placed on the digital camera image. The X, Y pixel values numbers for the coordinated mark on the image were then entered into the Ground System which combined them with the platform pitch and roll, aircraft position, aircraft heading and time over the mark to compute the actual offsets in eastings and northings in metres.
All offsets computed were assigned a confidence of 1 by the Ground System and given a hydrographic confidence of 1 by the System Operator responsible for the navigation position check. The     A second set of digital data was generated reduced to Lowest Astronomical Tide (LAT) for all areas for subsequent delivery to UKHO for charting purposes. LAT connections were calculated using geodetic information and software supplied by the GSI. LAT connections at each gauge location are highlighted in Enclosure 4.

E.2 Tidal Models
The survey was carried out in five distinct regions. Areas surveyed included Tralee Bay, Galway Bay, Donegal Bay, Blacksod Bay and Lough Foyle. Separate tide models were created for each area. Reduction of tidal data for each of the survey areas is discussed separately below.
To enable an effective tide model to be established, a total of twelve tide stations were incorporated into twenty four tide areas which were used for the vertical control of soundings in each of the survey areas. Tide areas created within Donegal Bay have been excluded from this annex as LADS data collected in the region was not accepted due to water clarity issues.
The tidal model adopted for each area is discussed separately below:

GALWAY BAY
Tide Area 1: soundings were reduced using observed tides from the Galway Port tide gauge.
Tide Area 2: soundings were reduced using a linear interpolation between observed tides at the Galway Port and Inishmore tide gauges.
Tide Areas 3 & 4: soundings were reduced using observed tides from the Inishmore tide gauge.
Tide Area 5: soundings were reduced using a linear interpolation between observed tides at the Rossaveel and Inishmore tide gauges.
Tide Area 6: soundings were reduced using observed tides from the Rossaveel tide gauge.
Tide Area 7: soundings were reduced using a linear interpolation between observed tides at the Galway Port and Rossaveel tide gauges.
Tide Area 8: soundings were reduced using a planar interpolation between observed tides at the Galway Port, Inishmore and Rossaveel tide gauges.
Tide Area 21: soundings were reduced using observed tides from the Inishmore Ocean Buoy tide gauge. Note no data was actually collected in this area.

TRALEE BAY
Tide Area 18: soundings were reduced using observed tides from the Maheree's Pier tide gauge.
Tide Area 19&20: soundings were reduced using a linear interpolation between observed tides at the Maheree's Pier and Fenit Harbour tide gauges.

BLACKSOD BAY
Tide Area 22: soundings were reduced using observed tides from the Blacksod tide gauge.

LOUGH FOYLE
Tide Area 23: soundings were reduced using a linear interpolation between observed tides at the Greencastle and Derry Port tide gauges.
Tide Area 24: soundings were reduced using observed tides from the Greencastle tide gauge.
Chartlets of all areas and lists of coordinates of the tide stations and areas are presented in Enclosures 1 and 2.

E.3 Tidal Reduction
The reduction of soundings within the Ground System is undertaken by means of linear interpolation within tide areas containing two tide stations, planar interpolation within areas containing three stations, and where only one station is incorporated within an area the value from that single station is applied across the entire tidal area.

E.4 Provision of Observed Tide Data
Observed tidal data for the survey was sourced from permanently installed gauges of the Irish National Tide Gauge Network, temporary tide gauges installed by the Geological Survey of Ireland and five temporary tide gauges established by Tenix LADS Corporation surveyors.
Data from tide gauges located at Galway Port, Inishmore and Killybegs were sourced from the Irish National Tide Gauge Network. Tides were provided in UTC time format and reduced to the Malin Head Ordnance Datum (MHOD). The Inishmore Ocean buoy gauge was created within the Ground System to satisfy tide area requirements, for potential data collection in this area, however no data was collected and so this gauge site was not used.
Observed tides for Maheree's Pier, Blacksod Bay and Greencastle tide gauges were supplied by the Geological Survey of Ireland. Data was provided in UTC time and reduced to MHOD.

Enclosure 3 -LADS tide gauge comparisons and datum relationships MULLAGHMORE
Tenix LADS surveyors installed a Valport 740c tide gauge and tide pole inside Mullaghmore Harbour on 16 May 2008. A 25-hour tide pole / gauge comparison was conducted and the pole was conducted via a closed levelling run between to a GSI Survey mark to enable MH datum to be recovered. Pole readings were taken every 30 minutes during rising and falling tide periods and at 10 minute intervals over the high and low water periods. These observed values were compared to the Valeport 10 minute logged readings and the height difference between the pole and gauge determined. The analysis also proved that the gauge was logging correctly in both time and range. Tenix LADS surveyors installed a tide pole and gauge on the north western harbour wall and conducted a closed leveling run between the tide pole and a GSI Survey Mark (MO 125). Pole readings were taken every 30 minutes during rising and falling tide periods and at 10 minute intervals over the high and low water periods.

Location of LADS gauge at Mullaghmore Harbour
These observed values were compared to the Valeport 10 minute logged readings and the height difference between the pole and gauge determined. The analysis also proved that the gauge was logging correctly in both time and range. LADS Gauge observed to be reading 0.02m deep 3.

Location of LADS Gauge in Rossaveel
Therefore subtract 0.02m from all gauge depths to get true reading.

FENIT HARBOUR
The Valeport 740c tide gauge established at Mullaghmore Harbour was recovered and placed at Fenit Harbour on 29 March 2008. A spot check was conducted to ensure the gauge was operating correctly. LADS surveyors installed a tide pole and gauge on the southern arm of the harbour and conducted a closed leveling run between the tide pole and a GSI Survey Mark (MO 112). Pole readings were taken every 10 minutes over a 3 hour period. These observed values were compared to the Valeport 10 minute logged readings and the height difference between the pole and gauge determined. This analysis confirmed the gauge was operating correctly after being removed from Mullaghmore Harbour.

F.1 General
Depth benchmarks and cross lines are used as the primary means of checking the quality of the depth data. A depth benchmark is typically a flat area of seabed which is re-surveyed on each sortie to check for correct system operation and correct application of tides in that area; the depth benchmark results can also be used in the assessment of the precision of the survey.
Cross lines are compared against main survey lines, and are used to check correct system operation and the tidal model throughout the survey area.

F.2 Benchmark Comparison Function
The LADS data is compared against the gridded benchmark surface in the Ground System and statistics are generated which include the number of points compared, the Mean Depth Difference (MDD) and the Standard Deviation (SD) between the data sets. The comparison compares the secondary data against the benchmark surface, and as this data is unedited it may contain noise normally removed during the validation process which is flagged as the shoalest and deepest differences.

F.3 Benchmarks
One benchmark areas were selected north of the Aran Islands, near the aircraft's base at Galway airport, at the start of the survey; however the highly variable water and weather conditions meant that no meaningful data was collected over these benchmarks on subsequent sorites. In addition, the wide spread of the geographical locations of each of the 5 survey areas undertaken meant that there was little opportunity to obtain repeat observations at the one location.
As a result no meaningful benchmark comparisons were calculated during this survey and a historical values of 0.15 metres (68% Confidence), gained from over 10 years of trials and survey operations, has been adopted for the system accuracy and been used in the calculation of the overall vertical accuracy of the depths.
The observed overlap between lines indicated that the LADS Mk II system operated correctly, and that the tidal model was consistent, and that tides were correctly applied.

F.4 Cross Line Comparisons
Fifteen cross lines were flown across the survey areas. Comparisons were made with main survey lines for analysis. Details are tabled below and the comparison summaries are provided at Enclosure 2.
Tenix LADS Corporation Pty Ltd. F-2 Report of Survey

F.4.1 Mean Depth Differences (MDD) and Standard Deviation (SD)
The averages of the Mean Depth Differences and Standard Deviation for each cross line run are as follows:

Run
No.

F.4.2 Cross Line Comparison Results
The mean of the average mean depth differences of 0.03 metres and the mean of the average standard deviations of 0.18 metres was generated from the intersection of 290 runs and the comparison of 447615 individual depths.
The final number of accepted comparisons was affected by the rejection of unacceptable values from this statistical analysis as a result of three main poor quality indications: • Insufficient comparisons -intersections with less than 500 comparisons were rejected.
• Poor C0 confidence value -comparisons with a Subsurface Confidence less than 6 were rejected.
• Poor C1 confidence value -comparisons with a Near Neighbour Confidence of less than 4 were rejected.
• All comparisons containing topographic data was also rejected as this data contains artefacts such as trees and man made structures which influence the results.
The cross line comparisons are generally good in depths < 20 metres, however there is a increase in noise in the steep coastlines with large kelp forests; this was particularly true along the north shore of Galway Bay. This does not reflect deterioration in system performance, rather the complex nature of the seabed along these coastlines.
Details are attached as Enclosure 1. The theoretical accuracy of the positioning systems is related to the distance of the roving GPS receiver from the base station. The relationship between baseline distance and theoretical accuracy was provided by Thales GeoSolutions UK for the Ashtech GG24 WADGPS receiver and is based on empirical data using LandStar corrections. The Thales LandStar correction service was withdrawn by Fugro in late 2004 and replaced with Fugro OmniStar corrections using the Virtual Base Station service.

G.1.2 Fugro OmniStar Wide Area Real-Time DGPS solution
Empirical tests undertaken by Thales GeoSolutions U.K. have detailed the horizontal standard deviation of positions obtained by the Ashtech GG24 WADGPS receiver with LandStar DGPS corrections at varying baseline distances. The expected error has been determined to be 1.01m + 1.4ppm. This standard deviation defines the theoretical repeatability of position fixes at various ranges. In 2004 the LandStar WADGPS service was withdrawn and replaced with the Fugro OmniStar system, which provides for a virtual station based on the real-time raw GPS feed that is then used in the generation of a solution from the nearest five base stations. Previous Thales advice also stated that the user should also be aware of the following: "DGPS systems are single frequency systems: the DGPS corrections are corrections to the user's pseudo-ranges, which are only available on the L1 frequency. Due to this, ionospheric delays cannot be corrected for, and this will cause a bias in the position. This bias is typically about 20-25 centimetres per 100 kilometres (2.5ppm), and will be in the same direction as the 'baseline' between reference station and user. Therefore if corrections are taken from a DGPS station due north, the position will be consistently out in latitude. This is a general limitation of DGPS systems, since they are single frequency systems". By adopting a virtual base station in the position of the aircraft and generating a wide area solution for this position based on up to five fixed base stations, much of this error is reduced.
No accuracy figures for the modelled Fugro OmniStar corrections are available; however given that the modelled solution is calculated for the real-time position based on observations at up to five base stations, the achieved accuracies for the real-time system during operations should be similar to the observed absolute accuracy obtained during the static position check conducted under similar conditions.
The accuracy of the WADGPS position was found to be 2.73m (95% confidence) as determined by the static position check on 21 May 2008. This is consistent with previous results and is sufficient for the real-time positioning of the aircraft. In any event, more accurate post-processed KGPS positions were applied to soundings during post-processing.

G.1.3 Post-Processed Novatel GrafNav Dual Frequency KGPS Solution
The theoretical accuracy of the post-processed GrafNav positional data has been determined from the GrafNav Software User's Manual and through consultation with Novatel. For a PDOP of less than 5 the following GrafNav data processing accuracy has been quoted:

G.1.6 Dynamic Position Monitoring
During the survey, GPS data was logged on the aircraft and at the local base station, which enabled post-processing to produce KGPS result files (off-line). These result files were then compared to the position as determined by the real-time WADGPS on the AS. For each survey line, the mean difference and standard deviation have been calculated. The dynamic position check results for each sortie are referenced in the paragraph 'WADGPS -Fugro OmniStar' in Annex D.

G.1.7 Accuracy of Position
The total expected error of the LADS Mk II system is a combination of the following errors:

LADS Mk II System Accuracy
A standard deviation of 0.15 metres (68% confidence) is the historical average for the LADS system obtained from historical benchmark data collected during previous surveys and during trials. This value has been adopted as the LADS MkII system accuracy for this survey.

G.2.2 Tides and Tidal Models
Tides were observed from multiple tide gauges throughout all survey areas, and the range of the observed tides was small. Thus the residual error due to tides is small and an accuracy value of 0.10m (68% confidence) was given to the observed tides used for the survey.

G.2.3 Swell
Swell had a small affect on survey operations in all the survey areas. The swell never exceeded 1 metres in most areas and 2 metres in the exposed parts of the Tralee Bay and Galway Bay areas. An allowance of 0.15 metres (68% confidence) has been allowed for the residual affects of swell and sea state.

G.2.4 Water Clarity
Water clarity was variable throughout the survey area and where possible data affected by varying water clarity has been removed and re-flown. An allowance of 0.15 metres (68% confidence) for the affects of degraded water clarity has been included in the accuracy model.

G.2.5 Accuracy of Soundings
An assessment of the total survey accuracy can be determined by combining the errors due to the LADS Mk II system, tidal model, swell and water clarity. These are combined using a Gaussian model as follows: σ 2 Survey = σ 2 LADS Mk II System + σ 2 Water Clarity + σ 2 Residual Swell + σ 2 Tidal Model

K.1 Beacons
The following beacons were detected by the system and have been retained in the dataset.

K.2 Buoy
The following buoys were observed in the video, but have been removed from the datasets.

Easting
Northing The following description of the Hawkeye II system is taken from the BLOM Aerofilms tender documents.
Two lasers are used in the Hawk Eye II system. These are a topographic laser system (1064nm) running at 64kHz and a hydrographic laser system running between 1 -4kHz with light of two different wavelengths being transmitted; green (532nm) and infrared (1064nm). In the hydrographic system the infrared laser light is reflected at the water surface whereas the green laser light proceeds into the water column. The green laser light is then reflected in the water column and at the seabed. A fraction of the reflected light reaches the Hawk Eye II receivers. The signal from the receiver is shown as a pulse response graph inside the "Pulse Response" boxes of the figure below.

Figure 1: Dual Laser System
In the topographic system, the second infrared laser running at 64 kHz, scans the land, foreshore and any exposed land of the inter-tidal zone of the coastal margin. All land objects above the water reflect the laser beam including vegetation, structures and other features. A fraction of the reflected light also reaches the Hawk Eye II receivers. Each hit from the beam will be recorded as an object and the position will be stored together with information on its characteristics.

N.2 Hawkeye II Flight operations
During the first survey sortie to each area, the cross tie lines were surveyed and main survey lines were sounded. During the subsequent sorties the cross ties were resurveyed and main survey lines sounded.
Automatic data processing to generate point cloud data occurs immediately following data collection. Data processing was initially conducted using no tidal model; observed tides were reapplied when received.

N.3 Hawkeye II Processing
The following description of the Hawkeye II Processing methodology is taken from the Hawkeye II tender documents.

Automatic data processing -generate point cloud
The first stage in the processing is the production of the aircraft trajectory from the airborne GPS data and the GPS data from the OSi Active Network base stations. The trajectory is processed using Applanix POSPac software. POSGPS is used for the production of the GPS positions and a second module POSProc is used to blend the inertial measurement using data and produce a final smoothed trajectory.
The next stage is the production of the laser point cloud using Coastal Survey Studio (CSS). This is a bespoke survey software package for the Hawk Eye II system.
In the post processing mode the trajectory data is combined with the scanning mirror data, timing information and laser ranges to generate a point cloud for each laser system. The result is a topographic laser point cloud and a bathymetric laser point cloud.

Ground survey data processing
The ground survey observations for the additional control points will be processed with the RTK GCA surveys. The data processing is carried out using Leica Geomatics Office (LGO) software. The observed static GPS data will be processed as a control framework to the OSi Active Network base stations. The RTK data will then be adjusted to the precise ETRF89 coordinates of the newly established base stations. The GCAs will be output in ASCII format suitable for input to the TerraSolid LiDAR processing software for QC of the land DTM data.

Vessel survey data processing
The SCA survey data from the single beam echosounder surveys is processed using Sonar XP software. The crossing lines are used to ensure data is well matched precise in the ETRF89 coordinate system. This results in soundings referenced in time and to the same precise ETRF89 surface of the bathymetric LIDAR point cloud. The soundings will also be output in ASCII format suitable for input to TerraSolid LiDAR processing software for QC of the seabed DTM data.

Laser data processing and checking
During the laser data processing the hydrographic and topographic laser data are kept separate. Once completed, the datasets can be combined together if required.
Having processed the data through CSS the next stage is to ensure that the laser point cloud is correctly aligned. This is achieved by identifying and then correcting for the heading, roll and pitch parameters present in the system. Using the TerraMatch software which runs within the MicroStation environment it is possible to identify the misalignments within a calibration area (using cross lines) and then apply these results to the entire data set. The final point cloud is then checked against the ground and sea control areas (GCA's and SCA's). After comparison it is then possible to apply a further shift to the data to ensure specification is met.