NatureScot Research Report 1418 - Loch Ailort serpulids: diving, ROV, and multibeam survey 27th February – 3rd March 2023

Year of publication: 2026

Authors: Sayer-Mitchell, J.D., Roberts, K.C., Schulz, P.H., Brown, H (Tritonia Scientific Ltd)

Cite as:  Sayer-Mitchell, J.D., Roberts, K.C., Schulz, P.H., Brown, H (Tritonia Scientific Ltd) Loch Ailort serpulids: diving, ROV, and multibeam survey. NatureScot Research Report 1418. 

Keywords

Loch Ailort; serpulid worm; Serpula vermicularis; diving; ROV; multibeam sonar; photogrammetry; priority marine feature

Background

NatureScot had a requirement to repeat a series of surveys undertaken in 2017 of the serpulid reef communities found in Loch Ailort and reported in the document:

Kamphausen, L., Henderson, S., Kent, F., MacMinn, R., Beaton, J., Dodd, J., McGrath, O., James, B. 2018.  Loch Ailort serpulid diving survey and drop-down video survey April 24-28, 2017.  Scottish Natural Heritage post-survey summary.

Following discussions with NatureScot, SAMS Enterprise, and Tritonia Scientific, it was decided that the repeat surveys, to be conducted before the end of February 2023, would comprise a hybrid approach that firstly repeated the methods of the 2017 surveys (diving-based transects and manual measurement), and then secondly  employed more advanced methodologies to enhance the data quality (multibeam  echo sounding and georeferenced photogrammetry).

Main findings

• The survey window encompassed 27^th February to 3^rd March 2023.
• Three priority areas were surveyed using multibeam sonar techniques to an accuracy of 0.5m.
• The diver transects from the 2017 report were identified and repeated, with georeferenced video transects taken.
• The diver transects were repeated and extended using an ROV, and a fully georeferenced orthorectified “digital twin” of the transects was produced.
• No serpulid aggregations, live or dead, were identified on either the diver or ROV transects. 

Acknowledgements

We would like to thank Ronnie and all the staff at MOWI Ardnish for access to their facilities and slipway.

Abbreviations

GPS – Global Positioning System

MBES – Multibeam Echo Sounder

PMF – Priority Marine Feature

ROV – Remotely Operated Vehicle 

UGPS – Underwater Global Positioning System

UK BAP - UK Biodiversity Action Plan

Introduction

The serpulid worm (Serpula vermicularis) is a marine tubeworm that forms complex reef structures that grow up to 75cm high and 1m across, and at depth ranges of 6m – 10m. These serpulid reefs are designated as Priority Marine Features (PMFs) in Scotland, and as a UK Biodiversity Action Plan (UK BAP) habitat. The health of serpulid reefs appears to go through cycles of growth and collapse, and there has been evidence from other west coast sea lochs of a decline in health and population of this habitat (Moore et al., 2020). 

Loch Ailort is a sea loch in Lochaber on Scotland’s west coast. There are a number of aquaculture developments within the loch, including a salmon fish farm, and a mussel farm. Serpulid reefs have been observed in this loch as recently as 2017 (Kamphausen et al., 2018). In 2023, Tritonia Scientific Ltd. were approached about repeating and expanding on the diver transects carried out by Kamphausen et al., 2018. It was proposed that the diver transects in 2023 would repeat the 2017 surveys using the same diver-based methodologies.  The transects would be augmented using video, and the area covered by diving would be extended using ROV surveys. All video data collected would be fully georeferenced, and 3D digital twins of the seabed would be created. These surveys would be further supplemented by a MBES survey in the three priority zones shown in Figure 1a and 1b.

Methods

The survey design used three methods to collect data – diving, ROV, and MBES. The wider area was to be mapped using MBES, confined to three priority zones. Next, the divers repeated the 2017 survey transects. Finally, the ROV repeated the diver transects and extended the area of coverage. All video footage was used to create a 3D digital twin of the survey area. The three priority MBES areas, ROV survey area, and diver transects are detailed in Figure 1a and 1b.

Diver survey

The diver transects were carried out based on the locations outlined in Kamphausen et al., 2018. Two transects were videoed and any serpulid aggregations recorded and measured. Transect 1 began at the marker rock from 2017 and extended in the water at 045 degrees. This transect extended for 80 m and a depth of 9 m. Transect 2 was taken perpendicular from 60 m on transect 1, along the depth contour. Transect 2 extended for 30 m. The video transects were taken using a GoPro 9 camera and covered the transect plus 2 m either side. 

MBES-based Bathymetry

WASSP S3Pr System: All bathymetry data were collected using a WASSP S3Pr multibeam echosounder (MBES). The wideband fairing transducer head and Ellipse Inertial Measurement Unit (IMU) Sensor were vertically side mounted to the 6 metre Rigid Inflatable Boat (RIB), Glaucus. Pings were collected using a 120-degree swath at a default centre frequency of 160 kHz. A WASSP IP66 DRX-32 was used for computation of the MBES data stream. These data were transferred to acquisition software using an RPM module on a fixed local IP. The Ellipse Dual Antenna Real-Time Kinematic (RTK) Global Navigation Satellite System (GNSS)/Inertial Navigation System (INS) was used to provide 0.2° heading and attain 1 cm RTK GNSS positioning. The Ellipse sensor box computes and corrects for pitch, roll and heave motion. To achieve RTK corrections, a ‘SmartNet’ license was attained from a third-party provider called ”The Survey Equipment Company”. A Valeport Sound Velocity Sensor (SVS) (max theoretical error ±0.017 m/s) was used to measure sound velocity at the transducer head and a Sontek castaway sound velocity profiler was used to obtain sound velocity profiles, which were subsequently used to correct the bathymetry data. 

Correction for Sound Velocity and Transducer Offsets:  After launching and berthing the Glaucus at MOWI Ardnish fish farm, the MBES system was set up, Sound Velocity Profiles (SVPs) were collected, and patch testing was carried out. Patch tests were caried out using the BeamworX software suite. The results of this are displayed in the calibration sections in Annex 1. Graphical displays of roll, heading and pitch calibrations are displayed in Annex 2. The SVP and calculated offsets were then applied to the survey configuration, meaning subsequently collected data were corrected during acquisition.  For each of the designated survey areas, one SVP was acquired and applied at the deepest point prior to data acquisition. Since the sound velocity’s recorded at the transducer head throughout the rest of the survey were no more than ± 2 m/s different to the sound velocity recorded at the corresponding depth on the SVP, no additional SVPs were required. Real time corrections for tidal height were performed automatically by RTK positioning upon acquisition.

Data Acquisition: To maintain safe depths whilst operating the MBES, survey lines were first sailed along deeper contours (parallel to the shore). Subsequent lines were then sailed at increasingly shallower depths. Lines were also sailed to achieve a minimum of 50% swath overlap with previous lines to allow for the clipping of the outer extents of the swath during post-processing. The vessel speed varied with the sea state and depth: shallower areas sailing with the tide were faster (between 2 to 3 m/s) and deeper areas against the tide were slower (between 1.5 to 2.5 m/s). All data acquisition was carried out using the BeamworX software suite. Data were collected in the Ordnance Survey Great Britain 36 (OSGB36), ESPG:27700 Coordinate Reference System (CRS). A GPS Quality Indicator (QI) of 4 (RTK fix <2 cm) was retained throughout the duration of the survey.

Post-Processing: All bathymetry data were post-processed using the BeamworX software suite. All Z data were transformed from ordnance datum (OSGB36) to chart datum using a z-shift of ‘-2.39’ (recommended for location). All data displayed in this report are presented in chart datum. A comparison with corresponding admiralty depths was used as a gross error check. 

The following filters were applied to the data to compensate for inevitable increased error at the outer extent of swaths:

• Slope filter (rejects points based on its mean slope to its neighbours)
• Clip outer beams (rejects 10% of the outer beams of the swath)
• Overlap fixer (corrects for refraction on file overlaps) 

All presented data fits well within the Total Horizontal Uncertainty (THU) and Total Vertical Uncertainty (TVU) criteria for Special order S-44 compliance as stated in the 2020 IHO Standards for Hydrographic Surveys (I.E. THU must not exceed 2 m and TVU must not exceed 0.25 m).

Data for mean bathymetry and intensity (backscatter) were exported as geotiff files then imported and presented in Quantum Geographic Information System (QGIS).

ROV videography and photogrammetry

ROV, GPS and UGPS:  A Blue Robotics ROV 2 with an 8 thruster Retrofit Heavy Configuration was used to perform all ROV transects. The ROV was connected to a 150 m Fathom Tether and Fathom-X Tether Interface junction box, which was subsequently connected to the control laptop for integration with Q Ground Control software. The ROV was fitted with a D1 Ping Sonar for safe navigation, a Water Linked DVL-A50 for improved stability and an A1 Locator for enhanced positioning.

For topside positioning and heading, a Simrad HS60 Satellite Compass was vertically side-mounted to the port side of Glaucus. Directly below, a Water Linked UGPS G2 antenna with four acoustic receivers was mounted below the waterline. Global positions of the locators are then computed using both the locators relative under water position and the Simrad HS60 topside position.

ROV Imagery:  For each diving transect, the ROV was driven along the complete transect length followed by parallel passes in the surrounding area. The ROV was driven at a speed of 1 m/s with a distance of one metre from the seabed. 

Source imagery for the construction of orthomosaics was captured using two downward facing GoPro Hero 10s. These images were captured at a rate of 1 per second. All imagery was georeferenced using the location of the Water Linked UGPS A1 locator which was mounted to the port side of the ROV. These images were then used to construct georeferenced orthomosaics using Agisoft Metashape v2.0. 

Results

The charts, pictures and 3D digital twin from the survey are displayed in the section below. During the survey, no evidence of serpulid reefs, alive or dead, was found, by diver, ROV, or MBES. Of note were multiple large aggregations and individuals of the parchment worm (Chaetopterus variopedatus), and a species of large nudibranch (possibly Doris pseudoargus). Examples of this are displayed in Annex 4. 

MBES bathymetry charts and backscatter plots

All MBES bathymetry charts and backscatter plots for priority zones 1 to 3, are shown in Figures 2 to 7.

Diver video transects

The total lack of serpulid aggregations observed means that no diver-based maps were made, and no measurement data obtained.

The diver video transects can be viewed at the following YouTube Links. These are unlisted videos that are only visible by invitation.

• Transect 1
• Download Link
• Transect 2
• Download Link
• Transect 2 (georeferenced)
• Download Link

3D digital twinning

A fully interactive georeferenced three-dimensional orthomosaic of the total diver and ROV survey area, showing both transects, can be accessed. 

ROV Transect video download links:

• Transect 1
• Transect 2

Note that to display telemetry on the ROV nav videos the .ass file should be in the accompanying folder of the mkv file and the mkv file should be opened with VLC media player.

The georeferenced combination of the orthomosaic with the priority zone1 MBES results is shown in Figure 8. The main seabed types identified from the orthomosaic are  shown in Figure 9.

All UGPS, depth and time data from ROV and diver transects are attached in the appendix at the end of the report

Discussion

Combining the ROV and diver transects would have provided data to ground truth the MBES survey area. However, as no evidence of serpulids was found, it reduced the clarity provided by the MBES survey. No definitive evidence of live or dead serpulids were found by ROV, diver or MBES. More expansive ROV and diver surveying could be done in the bay to check if the population centres have shifted. A downward trend in population fits in with other evidence of serpulid populations in west coast sea lochs (Moore et al., 2020).

Conclusion 

The population of Serpula vermicularis that was previously observed in Loch Ailort in 2017 was no longer evident over the same survey area in 2023. 

References

Kamphausen, L., Henderson, S.,  Kent,  F., MacMinn,  R.,  Beaton, J.,  Dodd,  J.,   McGrath,  O.,  James, B.  2018.  Loch Ailort serpulid  diving  survey and  drop-down  video survey  April  24-28, 2017.  Scottish Natural Heritage post-survey summary.

Moore, C.G., Harries, D.B., Tulbure, K.W., Cook, R.L., Saunders, G.R., Lyndon, A.R., Kamphausen, L., & James, B. 2020. The current status of serpulid reefs, horse mussel beds and flame shell beds in Loch Creran SAC and MPA. Scottish Natural Heritage Research Report No. 1156.