Offshore Wind Sustained Observation Programme (OW-SOP): scoping report

5 Roadmap towards sustained observational programmes supporting Offshore Wind

This chapter provides recommendations in support of the delivery of impact assessment strategies and observational/monitoring programmes targeting the interaction between OW infrastructure and the marine water column environment. Recommendations build on the review provided by Chapters 1-4 of this report, and on the outcomes of the discussions held during the multistakeholder workshop as part of this project. Reflecting the structure of the review presented in Chapters 2 to 4, Chapter 5 focuses on three thematic points:

i. Data and variables (Section 5.1).

ii. Model predictions (Section 5.2).

iii. Observational/monitoring programmes (Section 5.3).

A summary of recommendations is then provided (Section 5.4).

5.1 Data and variables

As discussed in Chapter 2, there are several Essential Ocean Variables (EOVs) that can be measured to assess the baseline condition of an ocean system in the absence of man-made infrastructure (e.g. OWFs). These variables can also be used to evaluate the potential interaction between anthropogenic activity and the local (and surrounding) environment. Relevant EOVs need to be chosen that enable a fundamental description and understanding of the baseline conditions and can be used as indicators of potential changes in the system due to human-related activities (i.e. OWFs).

As discussed in Chapter 2, to be able to define EOVs we first need to clearly define:

i. The objectives: what are we trying to monitor/observe/establish via EOVs? E.g. reduce uncertainty in baseline conditions and/or quantify their changes due to the interaction between OW infrastructures and the water column structure.

ii. The current data coverage and data gaps: where are there sufficient data, in terms of spatial coverage and time series, and where are new data required?

iii. The scale of the challenge: what spatial and temporal resolution is required to ensure detection of change?

Regarding point (i), there are no substantial operational OWFs yet built in seasonally stratified waters, or at the scales expected in the next decades. Therefore, the definition of the objectives (what we need to measure/observe in order to detect the cumulative effects of OWFs on the water column) relies on expected/predicted changes only. Thus, the definition of the EOVs and their requirements for spatial and temporal resolution (point iii) should be guided by the predicted (and modelled) interaction between future OWFs and the water column structure within seasonally stratified waters. Our understanding heavily relies on the use of numerical modelling approaches, on existing data, and on the collation of data/parameters at the locations of present and future developments, which will help to validate model outputs and be used to improve model parametrisations and eventually future predictions of likely impacts expected from expanding OWFs (Section 5.2). Data collation is essential (point ii) to understand where/if gaps exist and to what extent (e.g. what are the missing EOVs). Efforts should therefore focus on data collation, prediction of change, validation of prediction and improved forecast (Figure 5.1).

Although there are few site-specific, wind farm focused, data focusing on the water column, existing data collation can be used to assess baseline conditions at a regional and/or local scale in the absence of OWF infrastructure. More specifically, a distinction should be made between ‘shelf-wide (regional) data’ (see sub-section 5.1.1) and ‘site-specific (OWF project) data’ (see sub-section 5.1.2) (Figure 5.1).

EOVs to be used for environmental impact assessment, as described in Section 5.1 of this report."/>

5.1.1 Shelf-wide (regional) data

Shelf-wide (regional) data are required to define the regional baseline conditions regarding water column mixing/stratification in absence of OWFs. Where these data exist, they should be made available to the developers for use in EIAs at future OWF development sites. The same data could be used to validate regional (coarse) 3D models which help define the expected, positive and negative, cumulative impacts of OWFs in the water column (see Section 5.2). Shelf-wide (regional) data include two fundamental sets of EOVs (Figure 5.1):

i. Water column variables, and

ii. Atmospheric and metocean variables.

Informed by literature review (Chapter 2) and by workshop discussions with stakeholders, this report identified the key water column variables that should be measured for a good understanding of the regional baseline conditions regarding water column mixing/stratification in absence of OWF; these key water column variables (point i) include full water profiles of (Figure 5.1 and Table 5.1):

• Salinity, temperature
• Chlorophyll-a (Chl-a) concentration
• Dissolved oxygen (DO) concentration
• Growth limiting nutrients (Nitrate/phosphate)
• Mixed layer depth (MLD) (calculated from temperature and salinity vertical profiles)

It is recommended that atmospheric and metocean data (point ii) are also collected, where possible (see sub-section 5.1.2); these data would be the same parameters outlined for site-specific (OWF project) data (see sub-section 5.1.2).

Water column variables are generally detected using sensors deployed by vessel-based multicomponent surveys, MAS (including gliders), CTD on moorings, and to some extent also satellite remote sensing, at selected sites (see Section 5.3). A variety of organisations can play a role in providing collation, access and storage (i.e. recognised repositories) of these data such as research institutes, public bodies and government (Figure 5.1). Salinity and temperature are particularly important in the detection of tidal mixing fronts and in the definition of stratification duration (start/end) and strength. To ensure accurate detection, CTD samples should ideally be acquired at weekly resolution, therefore resolving different stages of the tidal cycle (spring and neap tides) and thus tidal mixing; this sampling should be undertaken during times of the year in which the water column is stratified, or is expected to transition from mixed to stratified conditions (spring and summer), as OWFs are likely to impact the marine environment via alteration of stratification and mixing. Detection of inter-annual/decadal variability is also required in order to isolate the OWF-induced impacts from natural variability and climate change. In terms of spatial resolution, vertical sampling across various water column depths is required to resolve the pycnocline. The spatial extent and horizontal resolution of sampling would likely be dependent on the site and scale of the individual OWF though, as highlighted within this scoping study, the footprint of OWFs may impact 100s of kilometres from the structures. MAS such as gliders are capable of travelling and collecting data over 20 km/day and therefore could adequately resolve the horizontal sampling resolution needed to cover potential footprints of OWFs.

Although we may already have a good spatial coverage for the EOVs described above, extra-measurements are needed in the areas with largest predicted/modelled changes to stratification due to OWFs/climate change (see discussion on model prediction in Section 5.2). Finally, sufficient observational data with appropriate temporal coverage in Scottish shelf seas currently does not exist to capture:

• Duration and strength of stratification, and
• Start/end of stratification over several years.

Therefore, appropriate data collation and additional data are needed to make sure we appropriately assess the water column structure in seasonally stratified settings before and after OWF development at the predicted scale. To inform future data acquisition, suggestions on recommended parameters to be collected for shelf-wide (regional) scale assessments, the platforms and sensors to be used, the sampling duration and recommended temporal and spatial (vertical) resolutions are presented in Table 5.1.

5.1.2 Site-specific (OWF project) data

Site-specific (OWF project) data will be required, and should be used (and provided), by developers to assess OWF-induced variability in the system (Figure 5.1). Site-specific data collected by developers are expected to be at higher resolution compared to shelf-wide data. Therefore, these can be used to refine/validate the 3D models that are used to assess the magnitude of impacts due to (cumulative) OWF effects at specific water column settings (see Section 5.2). Thus, site-specific data can aid the understanding of how OWF-induced changes compare in magnitude with natural variability and expected climate change effects predicted by coarse 3D models.

Given the absence of existing infrastructure to provide data, in the case of future FLOW developments, this assessment must be supported by model predictions (Section 5.2); thus, it is recommended that the necessary data are made publicly available (where they are not otherwise available) for environmental assessment studies (e.g., a data repository which is accessible for EIA purposes). Developers could access these data for their EIAs, integrating them with data at a higher resolution than regional-scale data, potentially obtained by wider sustained observation programmes, helping to validate prediction tools used to assess how water column variables can be observed to detect near- to far-field changes. The development of modelling guidelines will be essential for future EIA modelling strategies.

Informed by literature review (Chapter 2) and by workshop discussions with stakeholders, this report identified the key water column variables at a site-specific (OWF project) scale; these include (Figure 5.1 and Table 5.2):

• Salinity, temperature
• Chlorophyll-a (Chl-a) concentration
• Dissolved oxygen (DO) concentration
• Current speeds
• Suspended Particulate Matter (SPM)
• Mixed Layer Depth (MLD)

Similarly, this report identified the key atmospheric and metocean variables at a site-specific (OWF project) scale (and also at shelf-wide, regional scale, where possible; see sub-section 5.1.1); these include (Figure 5.1 and Table 5.2):

• Near Surface Atmospheric Temperature
• Near Surface Wind Velocity
• Waves

Measurements of the system’s hydrodynamics (in three dimensions) from ADCP, surface waves capturing wind-wake, SPM and Chl-a, combined with available site bathymetry and surficial sediments information, are key at future development sites. Furthermore, anything related to wind and currents is likely to be assessed by the developer before OWF construction, as this impacts profit, site’s risk and engineering design considerations. Ideally, all these measurements should continue during construction and operation. Near surface wind and related lower atmospheric parameters are relevant information for developers, and they are specific to operating conditions; thus, those parameters are typically continuously monitored during operation. Sharing of such data with stakeholders should be encouraged, through a mechanism that protects proprietary information.

To inform future data acquisition, suggestions on recommended parameters to be collected for site-specific (OWF) scale assessments, the platforms and sensors to be used, the sampling duration and recommended temporal and spatial (vertical) resolutions are presented in Table 5.2.

5.1.3 Data management

A central, publicly available resource for managing regional-scale data as well as non-commercially sensitive project specific data could be a key tool. This would allow stakeholders to improve their understanding at both near-field site-specific and far-field scales within shelf seas. These data are currently missing as there are no substantial OWFs (in particular FLOWFs) yet built within deep, stratified water settings. Nevertheless:

• Data may be available for fixed structures in mixed waters such as O&G infrastructures, that could be relevant and useful.
• Some data may soon be available from fixed OWF structures in stratified waters, e.g. from the ECOWind programme, which could then be used to assess the wind farm-wake scale.
• There is a lack of data from FLOW infrastructures, with only a few demonstrator farms currently present. Data from the ECOFLOW programme will be available in the future, with data collection beginning in 2024-2025.
• There is an opportunity to set-up data sharing mechanisms now, in order to be ready for data as they become available.
• Furthermore, to make sure these data are collected and made available before, during and after OWFs are built in seasonally stratified waters, observational/monitoring procedures and stipulation of some specific action regarding data collection or data sharing should be considered at the stages of consenting or even leasing of sites.

Due to the current lack of empirical data available, modelling approaches should be utilised to understand potential water column impacts (see further information in Section 5.2). However, models also rely on the iterative process of predicting change and validating model results, the latter ultimately comes from targeted data acquisition activities. In the absence of direct sampling at FLOWFs, the impact of floating structures, cables etc. could be assessed using 3D models within an EIA. It is recommended that data for model parametrisation (where available) be made available by all parties (e.g. developers, research institutions and government) to improve the confidence in model predictions.

As discussed in Section 5.2 of this report, standard industry-accepted 3D models could be developed, that suitably resolve and model perspective OWFs, for use by developers at the scoping and EIA stages. These models, and model outputs, could be made available to developers for use in scoping studies, enabling them to scope in/out certain processes/impacts with a higher degree of confidence. Such models would most likely be ‘precautionary’, e.g. consider ‘worst case scenarios’ with simplistic parameterizations of OWFs (Figure 5.2). Nonetheless this approach could standardise the scoping stage and make EIAs more efficient. These same precautionary models could be used by developers within EIAs, either as they are, or refined, or used in conjunction with additional observational data and enhanced modelling. The approach recommended above should be taken into consideration for the future definition of guidelines around the implementation of modelling procedures.

The uploading of environmental data collected by industry into a public portal should be considered as a lease condition. Thus, portals like the TCE open data portal could be used to collate and make OWF data publicly available. However, as discussed during the stakeholder engagement workshop, there may be commercially sensitive data. For instance, developers might be interested in wind-wake issues due to the potential impacts on profitability; thus, data on atmospheric variables are well studied by industry but are also commercially sensitive and therefore hard to access. The ‘Dutch’ model for baseline/pre-consent data gathering may be an effective solution, consisting of the use of public funds to gather baseline information to (i) de-risk investments for industry, (ii) increase bidding for leases, (iii) maintain data ownership in the public domain.

To overcome data sensitivity and ensure support from developers, careful engagement is required with industry; it is recommended that a follow-up to this report includes a workshop on data sharing, including representatives from developers, CES/TCE, research institutions and government.

5.1.4 Data for baseline conditions

Existing data/surveys and new technologies can be used to define baseline parameters, validate models for predicting changes, assess water column change, and design new data collection to improve prediction at future OWFs. These existing data/surveys and new technologies include:

• Developer’s metocean/geophysical surveys can provide a spatial picture of ‘habitat’ types, levels of stratification, areas of high SCM Chl-a.
• Some data from offshore O&G may be relevant as these existing infrastructures sit in deep and stratified water settings - see INSITE programme. Government-led engagement with the O&G industry to identify data sources that can be shared is recommended.
• Any existing long-term moorings could potentially be equipped with additional sensors (fluorescence, O₂, temperature, etc.) which could support these preliminary data: locating them in contrasting sites (with and without OWFs, and potentially at sites which models indicate are at risk of changes, from stratified to not stratified) for longer term data collection. Note that many of the latest ADCP instruments have integrated echo sounders enabling bio-physical studies and can be integrated on autonomous vehicles (i.e. gliders) for wider scale data collection and assessment of impact.
• Industry data from surveys used before selecting a site/construction e.g. lidar, wave buoys, wind speeds, ADCPs (which are left offshore for 1-2 years collecting data).
• CTDs and water samples, which are often part of semi-permanent offshore monitoring stations.
• Sensors could be located on the OWF structures themselves, potentially providing real time data from instruments powered from the structures.

Further information on data, surveys and new technologies are presented in Section 5.3 and a summary in Figure 5.1.

5.1.5 A pragmatic approach looking forward

To ensure that a better understanding of the potential environmental impact is developed in a pragmatic manner (e.g. minimising monitoring effort), measurements need to be fit for the purpose of detecting changes at the right time of year. Substantial observational programmes should be conducted on a few (potentially early) large OWFs, especially at sites which models indicate are at risk of OWF-induced water column changes, in order to gather a strong evidence base to support future development and future (more minimal) monitoring effort. Thus, a Joint Industry Project (JIP) could prove beneficial, sharing costs across stakeholders. Such a project would need to focus on an early deepwater (FL)OWF gathering evidence to help the consenting of subsequent OWFs across the industry. It is also important for government, industry and research institutes to cooperate in data sharing. These efforts should aim to:

• Identify a suitable data portal for sharing data between all parties.
• Use industry baseline data for model validation and direct future sustained monitoring programmes.
• Encourage developers to share their (non-commercially sensitive) data in order to help de-risk future development sites. Data sharing from a health and safety (H&S) perspective could also be explored, e.g. G+ global health and safety organisation (further information available in Global Offshore Wind Health and Safety Organisation (2024).
• Sensitive data from developers could be collected and managed by a non-commercial body, which would ideally make data from environmental acquisitional programmes publicly available after a certain period of time (embargo period).

5.2 Model predictions

Modelling approaches are necessary tools for predicting the interaction between offshore infrastructures and the marine environment, essential to disentangle infrastructure-induced changes from natural variability and climate change. As such, modelling techniques, informed by direct information and data from OWF sites and other locations, are increasingly likely to be key in decision making. As stated in the previous section, models can be used: (i) for the definition of the EOVs and their requirements for spatial and temporal resolution; (ii) for selecting time of the year and location when/where largest OWF-induced changes will be happening; (iii) as early ‘screening’ for site selection. Model predictions can be used to quantify present and future environmental changes due to OWFs, in the absence of direct information and data from existing sites or complementing existing information with future predictions. Fundamentally, the cumulative impacts due to OWFs or due to other long-term processes such as climate change cannot be assessed a priori without numerical predictions and models can be used to optimise the planning of future OWF sites (e.g. minimising environmental change). However, there remain some fundamental improvements that need investment in research to increase the accuracy of models’ predictions:

• Shelf-wide (regional) scale models can be used to effectively assess baseline conditions as well as the natural seasonal, inter-annual and decadal variability of the system, and can be designed to also account for the effects of climate change (Figure 5.2) (Holt et al., 2022, 2018, 2016; De Dominicis et al., 2018; Tinker et al., 2016; Mathis et al., 2018; Schrum et al., 2016; Wakelin et al., 2015; Mathis & Pohlmann, 2014). However, existing models (e.g. NEMO, FVCOM) need to be correctly parametrised and validated, to predict changes due to OWFs.
• Site-specific (OWF project) scale models used by developers, are usually high-resolution, but cover a limited area. In order to correctly parametrise both site-specific (OWF project) scale models and shelf-wide (regional) scale models, site-specific/regional observational data taken in the vicinity of existing OWFs need to be integrated to calibrate models (e.g. atmospheric parameters, drag coefficients, turbulence), and used to validate their prediction (Figure 5.2).
• Model simulations can be used to optimise observational programmes at existing and future OWFs sites. This needs to be an iterative process: (i) data collection can be informed by model predictions, and (ii) the models will be improved/validated by using the data collected in the observational programme. Although models may have a dynamic role (as they are built by feeding in new observations), using them could significantly reduce monitoring requirements. Figure 5.3 shows an example study from PELAgIO at an OWF where models have been used to drive data acquisition.
• There is a lack of theoretical understanding of water flows around a structure in a stratified body of water, which needs to be addressed to improve parameterizations within ocean circulation models. Data from laboratory environments are currently the only data available to understand stratified water flows, e.g. by trialling different densities of water alongside surface winds. Jointly funded projects (across government, industry and academia) could represent a good opportunity to develop lab-scale projects complemented by field measurements and modelling work, as well as favour future cooperation and data acquisition at specific OWFs, when these are developed and become operational.
• It is important to define how often the models would need to be run and updated (e.g. does this need to be done operationally, as a one-off or associated with a planning or statutory reporting cycle). Who would be responsible for running the models, will depend upon the aim(s) of the models/what the models are used for, e.g. developers may run them for licence applications, government could run them for planning purposes, academia may run them for research purposes. Ensemble modelling approaches/model intercomparison should also be considered, where feasible, as it could help reduce the uncertainty associated with modelled parameters such as PEA and change in PEA due to OWFs (Figure 5.3). Ensemble modelling outputs could be used to: (i) target data collection (instead of using just one model) and (ii) compare the OWFs-induced changes obtained from different models to understand models’ sensitivity and uncertainties.

5.2.1 A proposed cumulative/combined impact assessment approach

As discussed in sub-section 5.1.3, coarse, regional 3D models could be used to account for the cumulative effect of OWFs in an area, thus providing ‘precautionary’ model outputs for use during scoping and EIA (Figure 5.2). Such models could be used to predict the magnitude of cumulative and/or combined OWF-related changes pre-construction. An example is the Scottish aquaculture industry, where SEPA performs modelling for screening using an industry accepted model. Another example is the aggregate industry, where collective industry-wide modelling of a region is performed (Regional Impact Assessment, RIA), which assesses cumulative impacts and can be used for individual EIAs for specific developments.

Similarly to the aquaculture and aggregate industries, standard industry-accepted 3D models and model outputs could be developed to define the expected cumulative/combined effects of OWFs in Scottish stratified waters. These outputs can be used to drive the understanding of how the infrastructure cumulatively affects the environment, as well as providing precautionary models to be used by developers for their EIAs.

It is therefore recommended that standard industry-accepted 3D models should be developed to inform scoping and EIA; these models should include cumulative and combined effects due to multiple OWF developments. Creation of model data repositories where developers can access data needed for scoping and EIA would also be beneficial.

Thus, the following are recommended:

• Developing a screening/scoping procedure based on far-field modelling at an early stage of project development.
• Produce 3D far-field hydrodynamic models that take a precautionary approach to representing the mixing effects of OWFs. These models could then be used at project scoping and EIA to assess the level of predicted mixing caused by the infrastructures. There is a need to include both benefits as well as disadvantages related to the presence of OWFs, and the need to consider other effects, e.g. wind deficit due to the presence of the turbines, which may re-balance the infrastructure-induced mixing, reducing the underwater infrastructure effect. Therefore, the parameters used for these 3D models, the selected boundary conditions, and their outputs, should be carefully explained.
• Share 3D far-field model results with developers, which would help with risk-management of developments whilst also providing guidance on pre-construction data acquisition needs.

The precautionary 3D model could be augmented by site-specific modelling performed by developers. Developer modelling could be conducted using models of their choice and be refined by site-specific (OWF project) data, which should be used to demonstrate if their project falls within the coarse model-predicted magnitude of OWF-induced change (precautionary model). Thus, the developers should use precautionary model outputs to scope in/out processes and impacts from EIA, plan pre-construction data acquisition to ensure environmental impacts (both positive and negative) are fully understood and consider refined modelling approaches for EIA.

5.2.2 Modelling approaches

There are several shelf-wide (regional) scale models (including both physics and biogeochemical models) which could be used to inform an initial impact assessment in early planning, site selection and pre-construction. As mentioned in the previous sub-section, this approach could help developers to de-risk future developments at an early stage through an initial screening of a site. Developers could then perform their 3D models (potentially using refined versions of the 3D models) using their data, thus improving the impact assessment. However, this would be at the discretion of the developer.

Although not a straightforward task, it is important to define impact modelling guidelines and a threshold of change to be used; the latter, should be defined and agreed before using these models and tested within the impact assessment. Definition of those thresholds, as well as the definition of modelling guidelines, will require further work beyond the scope of this report and is recommended as a follow-on project/workshop.

OWF project) scale predictions, as described in Section 5.2 of this report."/>

PELAgIO project on how models can be used to assess present climate, climate change, natural variability and the effects of offshore wind infrastructures. Models' outputs can guide future observational programmes."/>

5.3 Observational/monitoring programmes

Due to the absence, at present, of extensive OWF (especially FLOW) in Scottish seasonally stratified waters, data mining is essential to inform impact assessments using model predictions and to guide future observational programmes. Therefore, we should aim to utilise as much of the existing data as possible from, for example, managed repositories storing data from regional scale acquisitional programmes as well as data from developers before further data collection is undertaken. Developers’ data should include data from the O&G industry, as they have been operating for decades with structures in stratified water settings.

New data collection should be informed by impact assessments (models) and performed pragmatically so that acquisition is done at the minimum required specification allowing cost-effective collection of the necessary data while fulfilling our understanding of the potential impact(s). This report provides recommendations for the minimum data collection to inform data collection programmes.

5.3.1 Existing data for initial impact assessments

Existing data sampling EOVs which are considered fundamental for the definition of water column baseline conditions as well as for enabling the detection of any possible environmental change (as discussed in Section 5.1), should be used to answer the following questions:

• What do we want to measure/observe?
• Define the main parameters needed to be able to parametrise models and detect any OWF-induced variation in the water column. For this it is important to:
  • o Define the strength of stratification.
  • o Define the threshold to alter the stratification (based on existing OWFs and model’s predictions).
  • o Measure through onset and breakdown of stratification and use models to confirm when this happens.

To define the above parameters, and to support the initial impact assessment, which should then drive the programme’s design, it is therefore recommended that the following should be done at a strategic level:

• Collate existing data including data from OW and O&G, as well as data acquired for different purposes and accessible from public repositories e.g., environmental assessment, fisheries, etc. into a single managed data repository, located in an official and recognised webpage, e.g. The Crown Estate Scotland.
• Use the data above to define baseline conditions and to assess natural variability (Section 5.1).
• Use numerical modelling to identify where additional data are necessary to detect OWF effects, providing guidance on who will need to collect those data and how (Section 5.2).
• Encourage developers to share their data highlighting the advantages of data sharing (e.g., early-stage de-risking of developments and improving health and safety).
• As discussed in sub-section 5.1.3, provide coarse, industry-recognised models that can be used to simulate the cumulative impacts of OWFs. Results from these models should provide a ‘precautionary model’ that developers could then compare against refined fine-scale models based on OWF project parameters including e.g. loads from currents and waves based on drag coefficients, wind deficit, foundation types, parameters to simulate atmospheric wake, etc. (Figure 5.2).

Research/academic institutions should also contribute to data gathering, sharing cooperative efforts through e.g. Marine Data Exchange (see ECOWind programme). To aid this the following could be undertaken:

• Co-development of best practice guidelines with experts from government, academia, and industry, to help researchers and OWF developers collaborate effectively when access to OWF sites is requested for research purposes.
• Ensure evidence from research is applicable to integrate or update guidance for environmental impact mitigation and supports strategic observational programmes and decision-making.
• Propose multidisciplinary approaches (e.g. combining remote satellite sensing, fixed sensors acquisition, improvements to existing site surveys) that can foster innovation, improve industry performance, and pave the way for efficiently meeting national climate targets.
• Adherence to these best practices enabling the establishment of effective communication channels, mutual understanding, and safety considerations for accessing OWF sites for research.

The aim is to ultimately encourage the development of a collaborative offshore environmental observational network (Figure 5.4).

5.3.2 Future programmes

When it comes to future observational programmes in areas with/without OWF, their location and spatial-temporal resolution should be data and models’ results-driven (Figure 5.3; Figure 5.4). Furthermore, to enable timely and cost-effective acquisition while ensuring fundamental parameters are collected, it is recommended to integrate various approaches. Three types of observational strategies can be used:

i. Fixed (long-term) measurements.

ii. Dynamic (MAS/ship-based) measurements.

iii. Remote (satellite) observations.

It is important to keep existing, fixed (long-term) measurements (e.g. moorings, buoys) at selected locations without OWFs, as well as placing them at future OWF development sites. These measurements can provide information about the natural variability of the system, as well as helping to estimate the magnitude of the OWF-induced variability (in terms of contribution to the natural/climate change driven variability). Fixed upstream and downstream monitoring points can be used at development sites included as part of the development process. Multiple fixed locations should also be used to collect atmospheric data. It is also important to select measurement locations that are informed by model outputs, and by where there are uncertainties within and between models (i.e. identify sites at which large-scale deployments are expected to change stratification away from the baseline). Monitoring of base ecology data (e.g. CTD/O₂/visual/eDNA) would also be of value (Figure 5.4).

However, in order to optimise fixed sensor’s acquisition at future infrastructures such as floating platforms, sensors could potentially be integrated into the structure themselves at a design stage (e.g. the integration of sensors into foundations). The selection of the sensors and/or provisions for maintainable data collection would need to be considered to ensure these approaches are economically and technically practical.

Dynamic measurements at locations where OWF impacts are predicted (as identified by models) are also needed. Dynamic measurements can include the combined use of ships and MAS to observe seasonal cycles like stratification (getting locations from model results), build upon existing monitoring programmes and use of existing infrastructure or maintenance operations – e.g. utilising an asset inspection ROV to collect other data, collection of aerial data, the expanded use of technology like gliders/USVs augmented by more targeted and precisely piloted MAS deployments (e.g. Autosub Long Range, ALR) (Figure 5.4). Ship-based surveys should also be considered, as they can be convenient (especially when utilising already planned observational programmes) for assessing the extent of stratification, a key variable for the type of assessments discussed within the remit of this programme. The extent of stratification can be estimated from MLD which can be measured using echo sounders exploiting, for example, boats performing regular work to and from the OWFs.

Remote observations (satellite data) are also key, and they represent a low-cost option for monitoring parameters such as sea-surface temperature (SST), Chl-a concentrations and wind speed (Figure 5.4).

Recommendations for parameter acquisition at future observational programmes, and the fixed, dynamic and remote strategies in which those parameters should be collected, are presented in Table 5.1 and Table 5.2.

Future observation/monitoring efforts should focus on obtaining measurements at the right time and location, by adhering to the following key principles when programmes are designed:

• Focus on the transition between mixed and stratified water columns e.g. spring/autumn, as well as throughout the stratified period (summer).
• Monitor the transition zones where there is marginal water column stability or intermittent stratification (e.g. changes through spring-neap cycle) and use this to parameterise and validate models.
• It would be preferrable to have multi-year data, thus the need to access sustained data collection which is publicly available on top of new data acquisitions. However, some locations may need longer duration and sparse data collections; other locations may instead need shorter duration and more detailed data. Programme design and data requirements should be informed by models.
• Prioritise sustained observations from fixed platforms at/near early OWFs in stratified waters.
• Consider the addition of e.g. CTD sensors to other planned, or existing, moorings used for marine mammal/passive acoustic detection and eDNA, providing valuable ecological data that can complement the enhanced understanding of stratification.
• Ensure monitoring plans look beyond the project level and consider strategic sustained observations and methods to fund these e.g. governments and/or industry funded “sentinel” monitoring stations collecting long time series data. Lease funds could be used for this purpose, and they could provide valuable data to validate and parametrise models.
• Ensure future programmes link to existing sustained observation programmes (both national and international). It would be useful to consider whether these existing programmes could be altered slightly to gain data that is valuable for OWF EIA/SEA.
• Ensure consideration of the latest technologies for ongoing data collection. If instruments are to be added to OWF infrastructure, then developers and manufacturers should engage early to ensure this is included at engineering design.

Observational programmes should provide data that follow accepted international standards (FAIR). Data sharing (now and future) is fundamental, and a project to collate industry data could be a useful starting point for this.

5.4 Summary of recommendations

Table 5.3 summarises some key recommendations from this literature review and stakeholder engagement process, providing an indicative assessment of ‘cost’, ‘delivery difficultly’, ‘lead time’ and ‘impact’ resulting from undertaking the work/task. Each category has been qualitatively scored ‘low’, ‘medium’, and ‘high’ based on the multistakeholder feedback from the workshop and review within the Scottish Government Offshore Wind Directorate. The objective of this table is to allow easy review of recommendations to inform decisions on which to take forwards, depending on the time and resources available. For example, defining key parameters for data collection has a low cost, low difficulty and short lead time with a good impact versus a development of a regional monitoring programme which is considered to have a high impact potential, but is more challenging to deliver in the near term. Lastly, it should be considered that many of these recommendations will have inter-dependencies or mutual benefits: using the previous example, defining key water parameters to measure feeds into requirements for site and regional modelling.

The categories and qualitative scores are broadly defined as:

• Cost – direct cost of undertaking an action. A high-level assessment has placed each in the following bands: low <£100,000, medium £100,000 – £250,000, high >£250,000.
• Difficulty – qualitative assessment of the challenges associated with each recommendation. This has considered technical difficulties and/or other challenges such as stakeholder engagement. For example, multistakeholder negotiations around legal agreements have been considered of high difficulty, whereas defining parameters for measurement (based on published sources) has been considered low difficulty.
• Lead Time –the following broad scales have been used for the purposes of assigning a qualitative score: low <6 months, medium 6-24 months, high >24 months.
• Impact – refers to the predicted influence/relevance expected from undertaking the recommended task. This impact has been assessed on a case-by-case basis using the workshop output, author’s expertise, and discussion with Scottish Government Officials.

Figure 5.5 illustrates the summary information from Table 5.3, highlighting the relative qualitive scores as per the description above.

5.4.1 Primary recommendations for implementation of Offshore Wind sustained observation programme(s) for physical processes

Based on the discussions in Sections 5.1 to 5.3 and the summary in Table 5.3 above, the following recommendations are proposed as key for implementation of an Offshore Wind sustained observation programme(s) for physical processes:

• Water column parameters that should be measured for impact assessments to be well defined, and guidance produced (Item 1 from Table 5.3). Suggested parameters, durations, sensors and resolutions are listed in Table 5.1 and Table 5.2.
• Fixed monitoring stations should be established at/near OWFs which will be built in stratified waters. Ideally, these stations should be set up before construction of the OWFs and their placement informed by model outputs predicting the most likely location for impacts. Ideally this could form part of the consenting process (Items 3 and 5 from Table 5.3).
• Standard industry-accepted, far-field, precautionary 3D hydrodynamic models should be developed to (i) scope in/out potential water column impacts from EIAs, (ii) support EIAs, and (iii) enable cumulative water column impact assessments. Developers should then be able to propose the use of refined models to compare evidence provided by the standard precautionary models as part of the consenting process (Item 8 from Table 5.3):
  • Initial results from models should be used to guide/inform future observational/monitoring programmes, to optimise model validation and improve forecast, providing a more reliable parameterization of OWFs in models.
  • Models should also guide the selection of EOVs, the spatial-temporal scale at which those should be monitored and should be used to define their error margins.
• To facilitate a screening approach, a regional monitoring programme should be considered (Item 11 from Table 5.3). As per item 4 from Table 5.3, this regional monitoring should include:
  • Collection of the recommended minimum parameters, durations, sensors and resolutions as listed in Table 5.1 and Table 5.2.
  • Detection of inter-annual/decadal variability.
• Sharing of data between stakeholders. Sharing of data is considered key to allow accurate assessment of the interaction between the physical water column and the impact at higher trophic levels (Items 9 and 10 from Table 5.3). Regulatory bodies are recommended to engage with industry to:
  • Identify which datasets from industry are needed and why, e.g. specifications about individual operational turbines, specifications about certain turbine-types and/or foundation structure details. This information may be necessary to build and/or parameterise models.
  • Agree data access – this may be via an existing or new data portal (Item 10 from Table 5.3) and may be addressed through a workshop programme similar to this project (Item 2 from Table 5.3)
• Existing modelling techniques should be compared to assess the margin of uncertainty in modelling OWFs due to differences in model setups and parameterizations and to decide what models to use (and to make those models industry-recognised models). This should take the form of a programme reviewing options and opportunities and workshop suggestions with a diversity of stakeholders.

5.4.2 Secondary recommendations to support implementation of Offshore Wind sustained observation programme(s)

A suite of additional recommended approaches to support implementation of sustained observation programme(s) are listed below:

• Any projects where data are to be used in models should utilise an iterative approach of observation-model-observation-model.
• When designing post-construction monitoring programmes, cost-effective and low-carbon solutions should include integration of a diversity of approaches including remote satellite sensing, use of autonomous vehicles (MAS), existing vessel-based surveys and fixed-point observations (e.g., moorings). These should also consider methods of data collection not currently used as standard such as:
  • Sensors on fixed structure in the field (Item 5 from Table 5.3).
  • Use of vessels of opportunity (e.g. crew transfer vessel/maintenance vessels) for data collection at sites (Item 6 from Table 5.3).
  • Use of MAS as a low-carbon solution helping to optimise data collection continuity and coverage (Item 7 from Table 5.3).
• Continual engagement with stakeholders including industry, government and academia is vital as OW infrastructure increases in stratified water and resultant observation programmes develop. Regular discussion should be held by relevant experts and stakeholders to identify key needs and further development in observational and monitoring strategies (e.g., organising workshops) (Item 2 from Table 5.3).
• Public bodies should coordinate data collation and collection from various sources/stakeholders, including developers’ data.
• Joint Industry Projects (JIP) should be encouraged, e.g. to provide evidence from early deep water OWFs/FLOWFs to support a sustainable development of the industry or fund laboratory studies.
• To reduce the cost burdens on developers and/or public bodies from additional data collection and modelling, alternative funding approaches should be reviewed: this could be a JIP complementing publicly funded activities focusing on other spatial data gathering methods (e.g. MAS or ship-based surveys).
• The O&G industry should be engaged, to explore if they have data for offshore assets in stratified waters that may be shared.