The HypHPMA for the case-study was located in South Devon, just south-east of the mouth of the river Exe, within the 6nm boundary of the D&S IFCA ( Figure 1 ). The Natural England (NE)/Joint Nature Conservation Committee (JNCC) criteria for choosing a location of a HypHPMA was reviewed (JNCC, 2021), to ensure that the location of the HypHPMA was appropriate. The HypHPMA was a square of 5km by 5km, covering a total an area of 25km². The size of the HypHPMA was based on the NE/JNCC recommendation of a minimum HPMA size of 5km diameter. In addition, the HypHPMA was bounded by; a 10km by 10km square (100km²) and a 20km by 20km square (400km²) ( Figure 1 ). These areas were considered within the study as smaller fishing vessels are often displaced to areas within a close proximity of the protected area or demonstrate ‘fishing the line’ behaviour (ABPmer, 2017). Of the two assessment types developed in ABPMer 2017 the comprehensive assessment was chosen due to the district’s large number of mobile vessels, as it is likely multiple vessels operate within the HypHPMA. Bathymetry data used in plot of HypHPMA was acquired from GEBCO (GEBCO, 2024).

For use of mobile fishing vessels within the D&S IFCA’s District a permit holder or representative must have a fully functioning, remotely accessed electronic reporting device on board the vessel at all times. This applies to vessels between 6.99 m and 15.24 m, which must transmit information at least every 3 minutes when in a restricted access area, based on the D&S IFCA’s Mobile Fishing Permit Byelaw, correct as of April 2024 (D&S-IFCA, 2013, 2024). For VMS reporting outside of the District it is an EU requirement that all fishing vessels 12 m or more in length must transmit their position every 2 hours when at sea (MMO, 2019).

The data supplied by the MMO were I-VMS and VMS pings for all 180 D&S IFCA mobile fishing permit holders operating during 2019 (hereafter referred to as the mobile fleet). Variables recorded for the VMS/I-VMS data included: vessel ID, latitude, longitude, date, time, speed and heading of vessel. These variables were all required for analysis within VMStools.

Landings (logbook) data were provided for 180 vessels by the MMO. The iFISH2 landings dataset was requested as this dataset is commonly used in other studies analysing fishing effort in UK waters (Dignan et al., 2014; Enever et al., 2017; MacNab and Nimmo, 2022), allowing comparison and the development of a standardised method. The datasets were based on all D&S IFCA mobile fishing permit holders operating during 2019. The variables present within the dataset included: date of landing, species, value (£), live weight (kg), gear type, vessel power (kWh), vessel size (m), landing port and vessel ID. Data were then combined with I-VMS and VMS data for VMStools analysis, if I-VMS/VMS data were available.

As only mobile gear vessels >6.99m are required to provide I-VMS data, D&S IFCA potting and netting activity (fixed gear vessels) surveys and permits were used to give an idea of the potential number of potting and netting vessels fishing within the HypHPMA and surrounding areas. D&S IFCA undertook potting and netting activity surveys in 2014. These survey data enabled the number of potters/netters that fished within the HypHPMA to be quantified along with the port they operated from. Noting however, the overall response rate of the 2014 survey was low (potters 28.7%, netters 26.6%). Potting and netting permit numbers issued by the D&S IFCA in 2021 were analysed to give an idea of the total number of potters and netters fishing in the surrounding area.

Benthic habitats within the HypHPMA were identified using the Natural England Marine Habitats and Species Map (MHabS), a spatial database comprised of the UK combined habitat map and Natural England data (v. 2024) (NE, 2024). Habitats are classified using the EUNIS (European Nature Information System) classification system, where all habitat types are listed via a EUNIS code (Tyler-Walters et al., 2018, Tyler-Walters et al., 2023). This hierarchical classification system describes more detail with increasing levels of the hierarchy; for example, A1 is a marine habitat of littoral rock and other hard substrata, whilst A1.111 is a marine habitat of Mytilus edulis and barnacles on very exposed eulittoral rock (MARLIN, 2022). For the EUNIS habitat maps layer, the EUNIS code for each area is as detailed as the habitat survey data provided. For areas with overlapping survey data a MESH confidence assessment is used to decide which layer is shown. A MESH confidence assessment was used to assess the confidence in each of the mapping studies that make up the overall habitat map (EMODnet, 2021; JNCC, 2016b).

Benthic habitat sensitivity was also assessed for demersal trawling and dredging for two different pressures: ‘Penetration and/or disturbance of the substratum below the surface of the seabed’ and ‘Abrasion/disturbance of the substrate on the surface of the seabed’ (MARLIN, 2024), during this paper these will be referred to as subsurface abrasion (SAB) and surface abrasion (AB) respectively. These pressures were assessed for the HypHPMA and surrounding areas. The assessment was carried out using the Natural England Spatial Seabed Sensitivity Tool (NESSST), (v. 2023); a map that spatially presents the sensitivity of benthic habitats to standardised pressures caused by human activities. The sensitivity assessment is based on the Marine Evidence based Sensitivity Assessment (MarESA) (Tyler-Walters et al., 2018, Tyler-Walters et al., 2023), whilst the habitat assessment uses the NE MHabS map. The assessment considers the resistance (the likelihood of damage from a pressure) and the resilience (the speed of recovery, assuming the pressure has stopped), the product of these defines the overall sensitivity. The sensitivity is assessed based on a pressure of particular threshold or benchmark and each pressure type is assessed individually. The sensitivity ranges from low to high, with other assessments included such as data deficient and not sensitive, depending on the data and literature available. Confidence scores for pressures were also presented; these considered the quality of evidence for resistance and resilience, alongside the MESH confidence score of the confidence in habitat type. A Biotope assignment confidence estimate was also generated; this is a score (0 – 1) representing a quantitative estimation of the certainty associated with the biotope assigned to mapped habitat. A more detailed explanation of the NESSST can be found within the Supplementary Material .

The blue carbon habitats were identified using the inshore layer generated from the Blue Carbon Natural England Research Report (Swale et al., 2022). This identifies important marine and coastal habitats for carbon storage and sequestration. Habitats include kelp, saltmarsh, seagrass, oyster beds, subtidal and intertidal sand and mud.

The Essential Fish Habitat (EFH) map (project MMO1133) was used to identify adult, juvenile and spawning hotspots (Katara et al., 2021) within the study area. The EFH layer is based on individual species’ hotspots, identified using the Optimised Hot Spot Analysis tool, clipped to areas of >99% hotspot confidence and given a value of 1. The EFH layer is comprised of 29 commercially and ecologically important fish species overlapped with each other. If there is an overlap the score is summed, so three different species would have a score of 3.

A large proportion of the data analysis was carried out using the R package “VMStools”; a package that focuses on analysing and visualising logbook (eflalo) and VMS/I-VMS (tacsat) data (Hintzen et al., 2012). Firstly, the data were converted into the correct format for use in VMS tools. Eflalo and tacsat data were then cleaned so: a) points on land; b) points in harbour; c) points with incorrect latitude or longitude; d) vessels travelling too fast to be considered fishing and; e) duplicate records were removed from the tacsat dataset. Whilst vessels with arrival date before departure date, duplicate records, and records with significantly higher catch weights or values (approximately thirty times larger on a log10 scale compared to the next highest value) were removed from the eflalo dataset. Due to the differences in ping frequency between VMS (ping every 2hrs) and I-VMS (ping every 3 minutes) data, the ping rate was standardised using the interpolateTacsat() function, using a Cubic Hermite Spline interpolation, with the recommended fm value of 0.2 to give a more detailed resolution for the VMS data, adding 38 extra points in between pings (Hintzen et al., 2012).

To identify when the vessels were fishing, rather than drifting or steaming, tacsat data were then analysed using the function activityTacsat(), this function fits a normal density curve based on speed profile. Tacsat and eflalo data were linked using the mergeEflalo2Tacsat function based on date, time and vessel ref. The number of vessels using the HypHPMA and surrounding areas were identified using the st_over() function, which identifies which points are within a polygon. Heatmaps of total time spent fishing (hrs), catch value (£) and catch weight (kg) and fishing effort (kWh) were plotted for the HypHPMA and surrounding areas for 0.025 by 0.025 decimal degree cells using “VMStools” and “ggplot2” (Wickham, 2011), histograms of these variables based on month were also created. Any squares containing fewer than three vessels were not plotted due to data sensitivity, to prevent individual vessel activity being discerned. To identify differences in catch value, weight and fishing effort, depending on vessel and month Kruskal Wallis tests were performed, as data were not normally distributed. A Dunn’s test was then applied for post hoc comparisons, coupled with a Bonferroni correction.

The total number of potting and netting permits for each port within the D&S IFCA district, in close proximity to the HypHPMA were calculated. These ports were plotted using “ggplot2”, to give an indication of how close ports were to the HypHPMA, with the size of points representing number of permits present. The total number of vessels that responded to the potting and netting surveys was used, as well as the number identified that would have fished within the HypHPMA boundary. A brief analysis occurred identifying the total landings weights and values and most common species caught, for potting and netting vessels with D&S IFCA permit.

For the environmental analysis Swept Area Ratio (SAR) was calculated using the equation:

The SAR was calculated for 0.025 by 0.025 decimal degree grid cells using “VMStools” and plotted using “ggplot2” (Wickham, 2011). Any squares containing fewer than three vessels were not plotted due to data sensitivity. This occurred for both surface abrasion (AB), width of gear that came into contact with the surface of the seabed and subsurface abrasion (SAB), width of gear that penetrated >2cm into the seabed, this definition is used in multiple marine studies (Eigaard et al., 2016, Eigaard et al., 2017; Rijnsdorp et al., 2018). Gear width was assigned based on gear type, vessel length and most common species type caught. This was based on expert opinion from Seafish ( Supplementary Tables S1 , S2 ).

The swept sensitivity ratio (SSR) was also calculated using the following equation:

NESSST sensitivity was linked to VMS/I-VMS data using the st_over() function and used within the equation above, in order to present both swept area (AB and SAB) and habitat sensitivity (abrasion and penetration) to demersal trawling and dredging. Sensitivities were rescored with a value between 0 and 3 (from not relevant to high sensitivity), for the calculation of a composite score for swept sensitivity ratios, rescoring used a precautionary approach described in Supplementary Table S3 . A composite scoring approach is often used for risk indexes and appropriate for this study as the composite variable has similar meaning to the original variable (Dolge and Blumberga, 2021; Song et al., 2013). For benthic habitat analysis the NE MHabS was linked to VMS/I-VMS data using the st_over() function, fishing effort based on habitat type was plotted in “ggplot2” (Wickham, 2011), stacked bar charts were also created for fishing effort depending on habitat type. The percentage coverage of each habitat type within the HypHPMA and surrounding areas were calculated using the st_area() function within the “sf” R package (Pebesma, 2018), with the crs set at ETRS89 LAEA, the UK standard for geospatial statistics, as part of retained EU legislation. The percentage area coverage of each habitat type was plotted on a stacked bar chart for the HypHPMA and surrounding areas. The MESH confidence scores for habitat type, the biotope assignment confidence estimate and confidence scores for quality of evidence for resistance and resilience, were also plotted to indicate the confidence in sensitivity and habitat type there is for each area. This method was also used for essential fish habitat and blue carbon maps.

This study used three simple methods to predict fishing displacement, based on known fishing behaviours:

Method 1: Fishing the line

Vessels are likely to concentrate fishing effort at the boundary area of the closed area, to exploit the benefits of the spillover effect (ABPmer, 2017). This was achieved by identifying pings within the HypHPMA and moving them to the closest point on the boundary line, via the st_nearest_points() function within the “sf” R package (Pebesma, 2018).

Method 2: Habitat similarity

Fishing displacement may also be dependent on habitat type; as displaced fishers are likely to visit preferred habitats of target species in the surrounding area (Cabral et al., 2017; Forcada et al., 2010; Horta e Costa et al., 2013). Therefore, the habitat similarity method moves points found within a certain habitat type in the HypHPMA to the closest similar habitat polygon outside of the HypHPMA, also using the st_nearest_points() function (Pebesma, 2018).

Method 3: Equal distribution

The percentage increase in fishing effort in areas outside of the HypHPMA was calculated using the formula:

Where R is the percentage of total fishing effort currently occurring in the HypHPMA and y is the percentage of total fishing effort in the area of interest (ABPmer, 2017). Fishing effort, SAR (AB and SAB) and SSR (AB and SAB) increase was calculated for the 100km² and 400km² areas surrounding the HypHPMA as well as inside and outside of the rest of the D&S IFCA. A map of the change in fishing activity before and after the HypHPMA closure was used by using the formula on ICES rectangles.

For the socio-economic analysis, results indicated that the average vessel using the HypHPMA could lose 2.98% ± 4.49 SD of their catch value due to closure assuming no redistribution of fishing effort. Furthermore, three vessels using dredges earnt over 10% of their income within the HypHPMA; a >10% loss in earnings is deemed a high risk for fishing vessels (Hintzen, 2023). 15% of the overall D&S IFCA mobile fleet used the HypHPMA at some point during the year, these were a mixture of<12m and ≥12m vessel lengths with the majority deploying scallop dredges. Assessing both individual and fleet wide impacts of displacement is recommended (ABPmer, 2017; Bastardie et al., 2015), especially for<12m vessels, as these may have a smaller operating range and may be more adversely impacted by loss in income (ABPmer, 2017; Benyon et al., 2020). In this case, two of the three of vessels with >10% of earnings from the HypHPMA were<12m in length, highlighting the importance of consultation with fishers before site designation.

There was significant variation in fishing activity throughout 2019; with the highest fishing activity in October and significantly lower activity over the summer and during the winter months in the HypHPMA. This could be due to a) the seasonality of fish caught (Bell et al., 2017; Respondek et al., 2014); b) holidays during those months (Guiet et al., 2019) c) fishing opportunities both within the study area and elsewhere, due to area closures or restrictions. In addition, 85.29% of the total catch value within the HypHPMA was Great Atlantic scallop; scallops have seasonally variable yield and are highly seasonal fisheries (Bell et al., 2017). Thus, it is likely that fishing effort may have decreased over the summer due to seasonal closures during the spawning season. Scallop closure occurs within this area between 1st July – 30th September, which would explain the low fishing effort over the summer months. There was also variation in fishing effort and landings weights and values spatially; this was much higher in the northeast of the 100km² surrounding area, however the average effort and landings in the HypHPMA and 100km² per km² were similar. This could suggest that they are both similar quality fishing grounds, however it also highlights the potential for competition for displaced vessels and vessels currently fishing in the 100km² area. Once again the need for multiple years of data is emphasised, to see if this spatial trend is consistent.

Swept Area Ratios (SAR) are an important tool for benthic impact assessments (Gerritsen et al., 2013; Hiddink et al., 2017; OSPAR, 2023), as they convert fishing activity data into a metric that is representative of the pressure on the seabed, that considers gear width, vessel speed and time (JNCC, 2016a). The SAR of the HypHPMA was 1.96; this means on average the area of HypHPMA was trawled 1.96 times a year. This is supported by the ICES data outputs on mobile bottom fishing disturbance (ICES, 2021); based on the interactive ICES pressure map a large proportion of the area within the D&S IFCA had an average SAR of between 1-5. Although, this only included >12m vessels so is likely to be underestimated. Furthermore, the SAR for<12m was higher than ≥12m length vessels within the HypHPMA, emphasizing the amount of fishing activity in inshore areas by smaller vessels, currently overlooked by VMS data.

Although the total time spent fishing by dredges within the HypHPMA was significantly higher than bottom otter trawling, the average SAR of bottom otter trawling was higher. This is because the gear width for OTB is much higher than DRB ( Supplementary Table S1 ). This is also evident in the JNCC report on Pressure Mapping Methodology (JNCC, 2016a); where the average width of OTB was five times that of DRB (for ≥ 15m vessels), leading to a larger SAR for that gear type. However, this is only for surface abrasion, for subsurface abrasion (deeper than 2cm), the gear width used in the JNCC report was 12m for DRB and 2m for OTB, due to the fact only the trawl doors of OTB gear types penetrate >2cm into the seabed. Subsurface abrasion or penetration is important for assessing the sensitivity of different benthic habitats, especially those with burrowing infauna (Collie et al., 1997; JNCC, 2016a). Our findings indicate that the subsurface abrasion SAR for dredges within the HypHPMA (0.684) was far higher than OTB (0.0181). In addition, penetration depth should also be considered, this is important for calculating Swept Volume Ratio (SVR), which can be used to assess blue carbon disturbance (Epstein and Roberts, 2022); for example DRBs are estimated to penetrate 5.47cm into the seabed whilst OTBs penetrate 2.44cm (Hiddink et al., 2017). For this assessment SVR would be particularly useful, as 94.01% of the HypHPMA was blue carbon habitat, however penetration depths for smaller vessels are under researched, highlighting an area of further research.

Based on the NESSST, a high proportion of the HypHPMA was highly sensitive to penetration from demersal trawling or dredging, with 99.6% of fishing activity over a highly sensitive area in the HypHPMA, whilst 100% of the HypHPMA had a medium sensitivity to abrasion. As the 27 vessels using the HypHPMA are dredgers or demersal trawlers this poses a risk to the area. However, it is worth noting that a large proportion of the surrounding area also has a high or medium sensitivity to these pressures, thus a proportion of displaced effort is likely to be displaced into highly sensitive areas. This is also relevant for blue carbon habitat due to the high proportion of subtidal mud in the 100km² and 400km² surrounding areas. However, this displacement is likely to allow recovery within the HypHPMA itself, which could allow fishers to benefit from a potential spillover effect; that protected areas support larger densities of individuals and there will eventually be a net movement from these areas to outside of the boundaries (Buxton et al., 2014; Rowley, 1994). Furthermore, the predicted increase in fishing activity is below 15hrs a year for all c-squares based on the equal distribution model, which is a minimal increase compared to the amount of fishing activity already present. SSRs should also be considered, as they account for both habitat sensitivity and SAR; surface abrasion SSR was highest to the northeast, outside of the HypHPMA, whilst subsurface abrasion SSR was highest to the east within the 100km² surrounding area.

For both the fishing the line and the habitat similarity models 100% of displaced fishing effort was displaced to the 100km² area, whereas in the equal distribution model 15.67% was. Although fishing the line can be common for displaced effort (ABPmer, 2017; Benyon et al., 2020; Kellner et al., 2007) it is unlikely that all displaced effort will be shifted to that area (Murawski et al., 2005). Thus, the equal distribution model is recommended for a quick assessment, with perhaps a certain percentage distributed for fishing the line. One study by Murawski et al. (2005) found that about 10% of effort targeting groundfish was deployed within 1km of MPA boundaries and 25% to the nearest 5km, so perhaps 10% of effort could be re-distributed as fishing the line in predictions. In addition, the displacement assessment does not consider potential removal of effort from the system, due to competition with vessels in the surrounding area, or gear switching to reduce competition with those of the same gear type (ABPmer, 2017; Vaughan, 2017), thus the increase in effort in surrounding areas may be reduced. For a more detailed prediction of fishing displacement, the DISPLACE model may be more beneficial (Bastardie et al., 2015, Bastardie et al., 2014), this platform runs Monte-Carlo simulations that project scenarios of different spatial management scenarios including area closures. However, this platform requires a large amount of computing power and is currently based on larger vessels.

Smaller vessels are often overlooked, with benthic disturbance studies often concentrating on ≥12m or ≥15m vessels (ICES, 2021; JNCC, 2016a). This underrepresentation of smaller vessels is probably due to the fact VMS is not currently a requirement for<12m vessels (MMO, 2019; Vaughan, 2017). This is a problem for management of the English fleet, especially in inshore waters; in 2021, 81.8% of the English fleet were<10m in length (MMO, 2021). This study has provided guidance on bridging this gap, looking at both<12m and ≥12m vessels, as well as highlighting the importance of inshore areas to smaller vessels. As of December 2023, 80% of the<12m English fleet had purchased an I-VMS device, with an I-VMS legislation that has recently come into force as of May 12^th 2025 (MMO, 2025). Thus, the underrepresentation of smaller inshore vessels is likely to be mitigated by this act (ABPmer, 2017), this includes the potting and netting vessels that were underrepresented in this study. Local knowledge of D&S IFCA authors indicates that potting and netting may be prevalent in the north-west of the area, where trawling and dredging is less common. However, with the lack of I-VMS data for smaller vessels and no standardised method for identifying when a >12m potting and netting vessel is fishing, it is difficult to assess these activities. With I-VMS legislation now in force (as of 12^th May 2025), it is essential that these vessels are represented in future studies using a standardised method.

The fact that only one year of data was used within this study, at one site was also a limitation. I-VMS data was previously only available for a small number of sites in English waters, with COVID disrupting fishing activity for a few years after 2019. Thus, these years were not included, as they were unlikely to be representative of fishing activity in the HypHPMA and surrounding areas. Once a long-term dataset of IVMS data is available across English waters, it would be beneficial to analyse multiple sites for HypHPMAs, considering the variability in fishing activity across years. Looking at a designated HPMA would also be valuable to inform the displacement assessment, when predicting fisher behaviour. Assessing the proportion of effort relocated or removed from the system and assessing behaviours such as fishing the line would be vital to consider. Thus, this area of research is recommended as a future case study.

Another potential drawback of this method is the limited effectiveness in areas where geolocation data is limited. A large number of countries now use satellite-based monitoring for tracking fishing vessels (Dunn et al., 2018; Orofino et al., 2023). However, there are still areas where satellite coverage is poor and the use of VMS and AIS among small vessels or small-scale fisheries is quite limited (Orofino et al., 2023). Lightweight and more cost-efficient alternatives have been developed for monitoring smaller vessels, including multiple GPS tracking devices. These devices can be solar or battery powered, using cellular networks or a mixture of cellular networks and satellites to transmit data (Fujita et al., 2018; Orofino et al., 2023). As these tracking devices can transmit similar information as VMS, such as speed, heading, latitude and longitude, they could be used as potential alternatives within the analysis.

The swept sensitivity ratios used in this study were a novel approach that considered both habitat sensitivity and swept area, allowing areas of high sensitivity which are trawled frequently to be highlighted for the displacement assessment. These heatmaps could be very effective when predicting displacement impacts, however they should always be presented alongside their confidence levels. As a large proportion of the confidence for the HypHPMA and surrounding areas were low, especially the MESH confidence and confidence in biotope simulation, highlighting that further surveys and consultations may be required when assessing the sensitivity of a HPMA. In addition, the gear widths for smaller vessels were also novel, as few studies consider<12m vessels when assessing SAR. These were based on expert opinion for average vessels within the area of interest, however further research into surface and subsurface abrasion is highly recommended, as gear widths can be location specific.

Deciding on suitable c-square size was one challenge of this study, many other studies use 0.05° by 0.05° grid cells (ICES, 2021; JNCC, 2016a), however due to the size of the HypHPMA smaller grid cells were required to observe fishing activity within it. One argument is that the grid cell should not be less than the maximum distance a vessel can travel between two pings whilst fishing (Fock, 2008), as I-VMS pings are every 3 mins the vessel is highly unlikely to go undetected within a grid squareOne caveat to this study is that it is only based on one year’s worth of data, so it is difficult to assess how variable seasonal and spatial trends are for different years. This study recommends multiple years of data for HPMA assessments and other spatial closures, although as I-VMS data will only be available from 2025 at the earliest this may be more challenging for<12m vessels. In addition to this assessment this study recommends consultation with local fishers and fishing authorities, to help account for seasonality in fishing activity, gauge how important the area is as a fishing ground and explore whether gear switching or using alternative fishing grounds will minimize impacts of displacement.

Although fishers’ behaviour (fishing the line) was somewhat considered with respect to the spillover effect, the socio-economic and environmental benefits of spillover were not fully assessed due to lack of dispersal and biomass data. The spillover effect is the net movement fish from protected areas (as they support larger population densities) to areas outside of the boundaries (Buxton et al., 2014; Rowley, 1994). This movement can increase fish populations in areas surrounding the protected area and catch rates in adjacent fisheries. For example, a study by Lynham and Villaseñor-Derbez (2024) on marine protected areas found a 12-18% percent increase in catch-per-unit-of-fishing-effort in the waters near protected area boundaries, for tuna purse seine fisheries. Therefore, if a similar increase was seen for the HypHPMA, this could reduce the effects of landings loss due to area closure. Spillover has the potential to result in positive net-benefits for fishers over time; due to increased recruitment of eggs and larvae, alongside spillover of adults into adjacent fishing grounds (Goñi et al., 2010; Medoff et al., 2022). Goñi et al. (2010) demonstrated a mean annual net benefit of 10% of the catch in weight for lobster (Palinurus elephas) from the Columbretes Islands Marine Reserve, the mean size of the lobsters emigrating from the reserve was larger than those outside. Estimating the potential benefits of spillover would be a useful addition to the displacement assessment but is challenging due to the variability of the effect depending on protected area size, duration of protection, initial fish biomass, connectivity to surrounding areas and the fishing intensity on the border of the area (Da Silva et al., 2015). This analysis is not currently suited to predict or evaluate the potential benefits of spillover, but it should still be considered when instating management measures or designations.

In addition, other costs with respect to changes in fishers’ behaviour are not fully considered within the analysis, one of these is changes in fuel costs and journey time due to area closure. Fishers may spend more time and fuel travelling to alternative fishing grounds, further increasing the socio-economic impact of fishing displacement (Bastardie et al., 2014; Vaughan, 2017). Thus, calculating current fuel consumption for each fishing trip and predicted fuel consumption after displacement, would be important to consider in future analyses. The DISPLACE model takes into account The fuel consumption rate (FCR in litres per hour), based on vessel engine power (kW) and operational conditions (Bastardie et al., 2014), which may be useful to consider in future analyses. However, predicting this after displacement would be more challenging, as the route taken from port to alternative fishing grounds and time spent travelling would need to be estimated.