British Indian Ocean Territory (BIOT) MPA

30. The BIOT MPA holds about 281,000 breeding pairs of seabirds of 18 species (Hays et al. 2020). Tracking studies show that the vast majority of these seabirds forage within the BIOT MPA while breeding, and Hays et al. (2020) suggest that the lack of commercial fishing within the MPA may help ensure high availability of forage fish and reduce threats from fisheries bycatch of seabirds. However, that inference is based mainly on expert judgement, as the available evidence on this is very limited in that particular case study (Hays et al. 2020).

South Georgia and South Sandwich Islands (SGSSI) MPA and seabirds

31. South Georgia and South Sandwich Islands (SGSSI) MPA, an area of 1.07 million km2, was designated in 2012, as a multi-purpose MPA encompassing the entire EEZ. Impetus for the MPA came from the desire to conserve species and habitats under pressure from: climate variability and change; previously high levels of illegal, unregulated, and unreported fishing; and incidental mortality from fishery bycatch (Handley et al. 2020). Using extensive tracking data from 14 marine predator species within the MPA (over 1,400 tracks), Handley et al. (2020) evaluated the spatio-temporal overlap of these predators and the different management regimes of krill, demersal longline and pelagic trawl fisheries operating within the MPA. Their analyses show that current fishery management measures within the MPA contribute to protecting top predators, including a wide range of seabird species, and that resource harvesting within the MPA does not pose a major threat under current climate conditions. The authors concluded, however, that unregulated fisheries beyond the MPA pose a likely threat to some of these seabirds. A very similar conclusion was reached by Heerah et al. (2019) by analysing tracking data from four threatened seabird species breeding at Amsterdam Island, southern Indian Ocean, within Amsterdam Island Marine Protected Area. In both of these cases, conservation gains for seabirds were assessed in terms of likely reductions in fishery bycatch and likely benefits from anticipated higher mean biomasses of prey species. However, long-term data to test whether or not seabird numbers increased in response to MPA designation are not yet available for these sites.

Mediterranean MPAs and foraging Yelkouan shearwaters

32. The Yelkouan shearwater Puffinus yelkouan is an endemic seabird in the Mediterranean Sea and is listed as Vulnerable. No MPAs have been designated specifically to protect foraging habitat of this species, but Peron et al. (2013) considered that coastal MPAs in the western Mediterranean, that had been established to protect coastal fish, may provide suitable foraging habitat and good densities of small pelagic fish required by this seabird. They tracked adults from colonies, carried out ship-based and visual aerial surveys, and assessed the extent to which these birds foraged within the existing MPAs (which are areas of 152, 1,413 and 4,019 km2). They found that 38% of Yelkouan shearwater diving locations during the breeding season were within the three French MPAs, and therefore that this seabird most likely benefits from the existing management of those MPAs. They did not report what percentage of diving locations were expected to fall within these MPAs if the birds had shown no preference but concluded that 38% was dramatically higher than would have been expected if the birds were not responding to these sites as preferred foraging areas. While the MPAs were not designed to benefit shearwaters, they were established with constraints on fishing as part of their management, and the study of Peron et al. (2013) provides strong evidence that this policy has enhanced the foraging resource for shearwaters within the MPAs relative to unprotected (and heavily fished) waters outside the MPAs.

The Namibian Islands’ Marine Protected Area, designated specifically for its seabirds

33. The northern Benguela Upwelling System supports an important seabird community, including several globally and locally endangered species. Threats to these species include a shortage of food, interactions with fisheries, human disturbance, habitat destruction and severe weather events possibly exacerbated by climate change (Ludynia et al. 2012). Populations of African penguins Spheniscus demersus, Cape gannets Morus capensis and bank cormorants Phalacrocorax neglectus are in rapid decline, and consequently are listed as globally ‘‘Endangered’’ (African penguin and bank cormorant), or globally ‘‘Vulnerable’’/ ‘‘Endangered’’ in Namibia (Cape gannet). These declines have been related to the absence of small pelagic fish (especially sardines Sardinops sagax and anchovies Engraulis encrasicolus) in the Benguela after many years of intense exploitation of resource by fisheries (Ludynia et al. 2012). Namibia designated its first MPA, the Namibian Islands’ Marine Protected Area (NIMPA) in 2009, stretching 400 km along the southern Namibian coast, covering almost 10,000 km2. One of the NIMPA’s three key objectives is to protect the breeding sites as well as main foraging areas of the three threatened seabirds breeding and feeding along Namibia’s coast. Using a zoned approach, with a large buffer zone connecting areas of higher protection status, the NIMPA places restrictions on human activities, including fishing, mining, guano harvesting and recreational activities in the area (Ludynia et al. 2012). Ludynia et al. (2012) used tracking data to confirm that the boundaries of the NIMPA were appropriate for the seabirds, but concluded that the almost complete absence small pelagic fish stocks in the northern Benguela now, will require additional measures to achieve tangible long-term restoration of the system and an improved conservation status for seabirds breeding in Namibia; the destruction of the pelagic fish stocks appears to have resulted in a change in ecosystem structure and function that might not be reversible, even if the MPA is protected from further fishing. As of 2020, there seems no reason to alter that pessimistic conclusion yet (Jean-Paul Roux, pers. comm.).

Habitat management plans to conserve African penguins in South Africa

34. Through the early 20th Century, very large populations of seabirds bred along the South African and Namibian coasts, where they feed predominantly on sardine (pilchard) and anchovy within the Benguela ecosystem. One of the key seabird species in that region is the African penguin. Because they are flightless, African penguins are resident within the ecosystem and are highly dependent on sardine and anchovy to survive, as well as to breed successfully. Off South Africa, anchovy and sardine contributed 50–90% by mass of African penguin diet in six studies conducted between 1953 and 1992, and 83–85% by number of prey items eaten in two studies between 1977 and 1985 (Crawford et al. 2006). Although there were 1.5 million African penguins in this ecosystem early in the 20th Century, the population declined to fewer than 200,000 individuals by the latter part of the century and was then classified as “Vulnerable” (Crawford 1998). Trends in regional populations of African penguins are related to long-term changes in the abundance and distribution of sardine and anchovy (Crawford 1998). Purse-seine fisheries developed rapidly off South Africa and Namibia after 1945. They reduced availability of food to penguins, especially off Namibia, following the collapse of the stock of sardine. Numbers of penguins breeding in southern Namibia dropped from 40,000 pairs in 1956 to 1,000 pairs in 2000 (Crawford et al. 2006). From 1984 to 1989, during a period when sardine fishing was increasing off South Africa, the breeding success of African penguins in Saldanha Bay was significantly related to the contribution of sardine to the diet (Adams et al. 1992). From 1989 to 1995, breeding success at Robben Island was significantly related to the biomass of anchovy (Crawford et al. 1999). From 1989 to 2004, the breeding success of African penguins at Robben Island, South Africa was significantly related to estimates of the abundance of both their main prey species, anchovy and sardine, and to the combined biomass of these species. When the combined spawner biomass of fish prey was less than two million tonnes, pairs fledged an average of 0.46 chicks annually. When it was above two million tonnes, annual breeding success had a mean value of 0.73 chicks per pair (Crawford et al. 2006). Crawford et al. (2006) concluded that in order to conserve the penguin population, management of the purse-seine fishery should ensure adequate escapement of fish to maintain the combined biomass of anchovy and sardine above two million tonnes. This research showed clearly that African penguin breeding success and population trend are driven to a considerable extent by forage fish stock status, and that forage fish stock status was strongly influenced by fishing pressures from the directed sardine and anchovy fisheries. Between 2004 and 2008, the African penguin population fell to just 26,000 pairs, the lowest value on record (Pichegru et al. 2010). Rather than constraining the economically important fisheries on sardine and anchovy stocks throughout their geographic distribution, an alternative that was tested was the establishment of MPAs for breeding African penguins by creating fishery no take zones (NTZs) close to islands supporting major colonies of these penguins. That was done as a scientific experiment using appropriate experimental and control areas with defined manipulations, with two objectives; firstly, to restore African penguin breeding success and numbers, and secondly, to test the efficacy of NTZs around penguin colonies as a tool to restore depleted populations. The study and associated research led to a large literature on African penguin demography in relation to MPAs, fisheries and forage-fish stocks (Pichegru et al. 2010, 2012, Sherley et al. 2013, 2015, 2017, 2018, 2020, Ludynia et al. 2014, Weller et al. 2014, Robinson et al. 2015, McInnes et al. 2017, 2019, Campbell et al. 2019, Crawford et al. 2019).
35. Because penguins are flightless, they forage while breeding in waters close to their colony, usually within 20 km of the nest site (Pichegru et al. 2010). In January 2009, a 20 km radius area was closed to purse-seine fishing around the world’s largest African penguin colony at St Croix Island, Algoa Bay (the ‘experimental treatment’). The waters around Bird Island, another penguin colony 50 km away within the same bay, remained open to fishing (the ‘control area’). By studying the foraging behaviour of adult penguins raising chicks at both sites before and after the closure to fishing, Pichegru et al. (2010) tested whether a relatively small no-take zone could benefit breeding penguins relying on pelagic prey. The foraging behaviour of adult penguins raising chicks of one to three weeks old was studied at St Croix Island (the ‘treatment colony’) and at Bird Island (the ‘control colony’), in May–June 2008 before, and in April–May 2009 after closure to fishing. The positions of purse-seine vessels were monitored via satellite telemetry, ensuring compliance within the experimental closure. African penguins share the care of their brood of one or two chicks between March and August, with typically one adult attending the nest when the partner is at sea. Birds were equipped with GPS-TD loggers which record latitude and longitude at 1 min intervals to an accuracy of less than 10 m, and depth at one second intervals to the nearest 0.1 m. In 2008, the average foraging path travelled for birds from the treatment island was 70 km (maximum 150 km), at 18–45 km away from the colony for an average of 22 hours per trip. After the fishery closure in 2009, penguins reduced their effort by 25–30%, travelling 50 km (maximum 80 km) to forage for on average 17 hours, within 5–30 km of the island. By contrast, from 2008 to 2009, penguins from the control island increased their time spent foraging (from 15.6 to 17.8 hours on average), potentially as a result of reduced marine productivity and/or increased fishing pressure around the island in 2009 (Pichegru et al. 2010).
36. Despite having data from just a single season before and a single season after fishery closure, Pichegru et al. (2010) concluded that their study provided strong evidence that even the relatively small MPA designated around the penguin colony can benefit penguins. However, Pichegru et al. (2012) describe the foraging behaviour of adult penguins raising chicks at both sites in the second year of fishing exclusion around St. Croix Island (2010). In that second paper they compared the penguins’ at-sea behaviour in the year before closure and the two years after closure with the distribution and abundance of purse-seine fish catches. They also compared the penguins’ breeding success and chick growth at the two colonies in the first two breeding seasons after the local fishery closure. Birds from St. Croix Island (the ‘treatment colony’) spent more time feeding within the boundaries of the closure after the ban (75% and 55% of their dives within the closure in 2009 and 2010, respectively) than when fishing was allowed in that area, and significantly reduced foraging effort, whereas effort increased at the control colony, but there was no clear evidence that the NTZ resulted in higher breeding success or survival of penguins. Pichegru et al. (2012) found that the local fishing fleet increased fishing effort in the area immediately outside the closed area and concluded that in order to avoid such problems the closed area needed to be larger, or to have a buffer area around it with limited fishing. They concluded ‘The collapse of Africa’s only breeding penguin species adds urgency to the wider implementation of such measures, which are likely to also benefit the important biomass of endemic predators of the Benguela upwelling ecosystem’.
37. Sherley et al. (2013) showed that breeding numbers, and the fledging period, of African penguins at Robben Island increased and decreased in relation to local abundance of sardines in that area prior to breeding. Breeding success and chick-fledging rates also increased with increasing biomass of forage fish (indexed through the industrial catch of anchovy within 56 km of the colony). They concluded that the local abundance of forage fish rather than the total abundance throughout the Benguela Upwelling Ecosystem, is the key driver of penguin breeding success at this colony. They concluded that management needs to ensure adequate biomass of forage fish close to the colony during the breeding season and also at a regional level in the nonbreeding period when birds are more dispersed than while breeding. Using a population modelling approach to assess trends in African penguin numbers at Robben Island, Weller et al. (2014) concluded ‘The modelled population was found to be strongly driven by food availability and to a lesser degree by oiling and marine predation, while climate events and terrestrial predation had low impacts. Food biomass levels (small pelagic fish) in the penguins’ foraging area around the island (used during nesting) and further afield (used during the rest of the year) had an equal influence in driving population development in the short and long term. The impact of short-term (three years) fishing restrictions currently being trialled around the island was found to be generally beneficial to the modelled population, but easily masked by food-driven variability in population growth’. Their results suggested that improving food availability and mitigating the impact of oiling would have the highest beneficial impact on this penguin population. Robinson et al. (2015) developed a population dynamics model for African penguin at Robben Island and concluded that the predator–prey interaction was best explained by a sardine–penguin mortality relationship with average penguin survival decreasing only when the local sardine biomass was less than approximately one-quarter of the maximum observed. From that, they inferred that declines in penguin numbers would be most likely when forage fish biomass was severely reduced but would be unlikely if forage fish biomass was maintained at moderate to high levels. From 2011 to 2013, a 20 km radius around Robben Island was closed to purse-seine fishing. Sherley et al. (2015) examined how African penguin chick survival responded to that experimental closure. Chick survival is heavily influenced by the rate and amount of food delivered to the nest, so should respond if closure increases prey availability above baseline levels. Sherley et al. (2015) examined whether penguin chick survival varied between years with (2011–2013) and without (2001–2010) fisheries closure and used a demographic model to examine the impact on population growth. Crucially, they used biomass estimates to account for variation in prey availability, penguin population estimates to control for density-dependent effects and catch data from outside the closure to control for changes in fishing activity over larger spatial scales. Although the closure was relatively small, and catches continued at its boundary, chick survival increased by 18% after the closure was initiated compared with when fishing had occurred close to the island, which alone led to a predicted 27% higher population compared with a scenario with continued fishing. However, the modelled population continued to decline, probably because of high adult mortality linked to poor prey availability over larger spatial scales. Sherley et al. (2015) concluded that the results illustrate that small no-take zones can have bottom-up benefits for highly mobile marine predators, but are only one component of holistic, ecosystem-based management regimes.
38. Campbell et al. (2019) provided a detailed study of African penguin foraging from Robben Island. They tracked 75 chick‐provisioning penguins with GPS–time–depth devices, measured body condition of 569 chicks, quantified the diet of 83 breeding penguins and conducted 12 forage fish hydro‐acoustic surveys within a 20 km radius of Robben Island over three years (2011–2013), during the MPA/NTZ period at that colony. Local forage fish abundance explained 60% of the variation in time spent diving. Penguin foraging effort (time spent diving, number of wiggles per trip, number of foraging dives and the maximum distance travelled) increased and offspring body condition decreased as forage fish abundance declined. In addition, quantile regression revealed that variation in foraging effort increased as prey abundance around the colony declined. Their results demonstrate that local forage fish abundance influences seabird foraging and offspring fitness. They also highlight the potential for offspring condition and the mean–variance relationship in foraging behaviour to act as leading indicators of poor prey abundance. Those metrics were suggested as possible ones to use to manage MPAs for these breeding seabirds.
39. Ludynia et al. (2014) studied the population trend and demography of African penguins at Dyer Island, where there has been a very large decrease in breeding numbers. They concluded that numbers breeding on the island were negatively correlated with purse-seine fishery catches from within 20 nautical miles around the island, but that once the colony had declined below 3,500 breeding pairs, the impact of fishing became less evident and other factors took over. Those particularly included predation impacts from fur seals and kelp gulls, which appear to show an increase when penguin numbers were depleted.
40. Sherley et al. (2018) used Bayesian inference to examine changes in chick survival, body condition and population growth rate of African penguins in response to eight years of alternating time-area closures around two pairs of colonies. Their results demonstrate that fishing closures improved chick survival and condition, after controlling for changing prey availability. However, this effect was inconsistent across sites and years, highlighting the difficulty of assessing management interventions in marine ecosystems. Nevertheless, predicted increase in penguin population growth rate as a consequence of local fishery closure exceeded 1% at one colony. Fishing closures evidently can improve the population trend of a forage-fish dependent predator. Sherley et al. (2018) therefore recommended that they continue in South Africa and support their application elsewhere. However, they also noted that detecting demographic gains for mobile marine predators from small no-take zones requires experimental time frames and scales that will often exceed those desired by decision makers.
41. Crawford et al. (2019) used Principal Component Analysis of seabird diets in the Benguela Upwelling Ecosystem to assess the effect of variation in forage fish abundance on these birds. They found that PC2 provided a Forage Availability Index that correlated with African penguin annual survival, emphasizing that while local MPA/NTZs may help to improve penguin breeding success, the abundance of forage fish in the wider ecosystem is important in influencing survival of penguins from year to year.
42. This case study, based on a large amount of detailed research and experimental manipulation, provides strong evidence that a NTZ around African penguin colonies can reduce the foraging effort and increase breeding success of the birds, but also that this measure may not be enough on its own if the population decline is also caused by low forage fish abundance throughout the region that results in poor survival of birds when dispersed away from the breeding site.

The effectiveness of fisheries management changes to seabird populations

43. The available evidence indicates that habitat management zones that are small can be effective for sedentary animals. But for seabirds, habitat management zones would be intended to enhance food supply, and the mobility of seabirds means that to be effective any such zones would need to be large. Outside the breeding season, seabirds tend either to migrate substantial distances to overwinter in areas with good food supplies and benign conditions, or in those species that do not migrate, the birds disperse from the breeding area.
44. In the breeding season, seabirds become central-place foragers, commuting from their nest site to foraging habitat. While this constrains their spatial distribution, the maximum foraging ranges of many breeding seabirds are large.
45. This review found numerous examples that changes to fisheries management (through MPAs with NTZs) improved prey species populations resulting in positive demographic change for seabird population that foraged on those prey. This was across a wide variety of seabird taxa, fish taxa, fisheries type and locations around the world. These results suggest that changes to the fisheries of key prey fish populations could provide suitable compensation measures for impacts predicted to affect protected seabird populations.
46. There is therefore good evidence in general that compensation measures that have positive effects on sandeel stocks in the North Sea are likely to positively benefit seabird populations that forage on those sandeel stocks.

ANNEX B. Residual plots for each relationhip between species demographic rates and sandeel tsb in SA4

Figure 1.27 Residual plot for kittiwake return rates

Figure 1.28 Residual plot for guillemot return rates

Figure 1.29 Residual plot for razorbill return rates

Figure 1.30 Residual plot for puffin return rates

Figure 1.31 Residual plot for kittiwake productivity

Figure 1.32 Residual plot for guillemot productivity

Figure 1.33 Residual plot for puffin productivity

ANNEX C. Consultation with Stakeholders

47. Throughout the development of the sandeel fisheries compensation plan the Applicant has consulted with statutory and non-statutory stakeholders on the approaches used. These consultations have resulting in useful feedback and improvements to this report. These are summarised below. Additional information regarding consultation undertaken by the Applicant is presented in Appendix 1 of the Derogation Case.

Questions asked of consultees

48. During the consultations with key stakeholders four important questions were asked:
    1. Are the sandeel stock biomass scenarios a suitable basis for assessing efficacy of compensation measures?
    2. Are CGR and growth rates the most appropriate basis for assessing efficacy of compensation measures from PVA results?
    3. Are there other SPAs that should be assessed? and
    4. How do we estimate the compensation ratios based on SPAs that benefit from the proposed measures but are not impacted?
49. Consultee responses for each of these questions are summarised below, with responses provided. Note that for clarity “stock biomass” refers to “total stock biomass” and is referred to as “TSB” hereafter. It is the estimated biomass of sandeels of all age classes, as presented in annual stock assessments by ICES. For sandeel, spawning stock biomass is defined as the biomass of fish 2 years old and older as few 1 year old sandeels spawn. Because a high proportion of the stock of this short-lived fish comprises 1 year old fish, TSB is usually considerably larger than SSB but the ratio between these can be variable.

Are the sandeel TSB scenarios a suitable basis for assessing efficacy of compensation measures?

50. Marine Scotland stated that, “To confidently predict the effect of compensation measures would require data at a much finer scale than SA4. For a resident species like sandeel, showing limited movement and dispersal, data at the bank level are required. This would then need to be related to sandeel availability and accessibility to the breeding seabirds of interest (considering inter alia foraging ranges). Ideally time series of age structured data at the bank level would allow the estimation of mortality for each age class of sandeel and compensation measures should result in a decrease in mortality rate of the age classes impacted by the fishery and benefiting from the compensation measures. Crucially to understand how seabirds could benefit, sandeel abundance is not sufficient to estimate efficacy, the spatial aspects and temporal availability need to be considered.”
51. These data are not likely to be suitable for these analyses for a variety of reasons:

• these data are not available from ICES (these are commercially sensitive data);
• these data will not be available across all of the relevant sand banks as a time series that can be linked to species’ demography on the Isle of May or at other SPA colonies;
• the seabirds being assessed here have sufficiently long foraging ranges that they are able to visit multiple sandbanks, and there is likely to be variation in this within and between years; and
• analysis (see 1.7) showed that there is no important difference in the response of kittiwake productivity between colonies likely foraging in the sandeel box and those foraging outside the sandeel box.

52. ICES will hold the catch information on sandeels in SA4 at the bank scale across multiple years, However, since this is highly sensitive commercial data it is not publicly available. For the reasons listed below, it is also unlikely to provide a useful evidence base for assessing the effects on seabird demography.
53. It is clear from the catch reporting from ICES (2022) that between 2006 and 2021 fishing effort on the east of Scotland has increased ( Figure 1.34   Open ▸ ).

Figure 1.34 Sandeel fishing in ICES SA and Division 3.a. Catch by ICES rectangles from 2006–2021. The area of the circles is proportional to catch by rectangle. From ICES 2022.

54. The breeding season foraging ranges of the species being assessed here are sufficiently long it is likely that they are foraging across multiple sandbanks within any one year and it is likely that they forage across multiple sandbanks in both within and between years. The available data on sandeel stock sizes that combines spatially and temporally with seabird demographic parameters occurs most usefully at the SA4 level. At smaller spatial scales the catch or stock estimation data is not available at the temporal scales needed to understand the relationships between species demography and sandeel prey availability.
55. Further analyses was undertaken on the relationship between sandeel TSB in SA4 and kittiwake productivity from colonies on the east coast from the Cromarty Firth to the Farne Islands This showed that the relationship between sandeel TSB in SA4 and kittiwake productivity in SA4 occurred down the whole of the east coast  and did not appears to be substantially different for colonies inside and outside the sandeel box (see 1.7).
56. NatureScot noted that the range of TSB scenarios tested was useful but noted that the historic maximum biomass (about 900,000 tonnes) “should be achievable again”. We agree that this may possibly occur with suitable fisheries management in place and sufficient time for stock recovery, but there is a possibility that the historic maximum can no longer be achieved as a result of impacts of climate change and ecosystem change, as suggested by Lindegren et al. (2018) in relation to SA1r, but the same argument applies to SA4. The maximum TSB used in the scenarios was 800,000 tonnes, which is similar to the historic maximum value. It is important to note that the stock is likely to fluctuate through natural processes, which is one reason why the scenario-based approach is likely to be useful when the range of responses are considered. So, while a maximum TSB may occur in the future due to changes in fisheries management, it is unlikely to occur in every year. The reported historic TSB from the 2022 ICES report (ICES 2022) occurred in 1997 at 779,492 tonnes.
57. The scenario testing approach has been used for a variety of important reasons. The future TSB in SA4 is uncertain either under the current fisheries management or future changes to fisheries management. The baseline TSB used and the predicted increases in TSB as a result of compensation measures in each scenario covers a range of plausible changes. The aim of testing a wide range of scenarios is to present of potential outcomes from most precautionary to most optimistic based on the past information on sandeel TSB in SA4. The values used have been based on the historic TSB for SA4 ( Figure 1.35   Open ▸ ). The scenarios include all of the distribution, except values below 100,000 tonnes TSB. It is important to note that the large changes in species demography occur at the lower end of the scale of sandeel TSB, and that the choice to limit the baseline scenarios to 300,000 tonnes was based on the approximate “one third for the birds” rules of thumb described by Cury et al. (2011).

Figure 1.35 Frequency distribution of sandeel TSB in SA4 from 1993 to 2021

58. NatureScot also noted that the there was a discrepancy between the “current” TSB and the statement that the TSB had not been above 500,000 tonnes since 2006. This is due to the availability of data on both the SA4 TSB and return rates and productivity of the relevant species in order to study relationships between these. There were no seabird demographic data available from the Isle of May in 2020 or 2021. During the period from 2004 to 2019 the TSB did not exceed 500,000 tonnes. The ICES (2022) predictions of sandeel TSB in SA4 is provided below ( Table 1.34   Open ▸ ) to provide a complete account of predictions at the stock level.

Table 1.34 Total Stock Biomass (TSB) in SA4 from 1993 to 2021 from ICES (2022).

Are CGR and growth rates the most appropriate basis for assessing efficacy of compensation measures from PVA results?

59. While NatureScot stated that, “we agree that the ratio metrics are the most sensible approach to assessing predicted change with the different scenarios” the RSPB noted that, “Greater clarity is required as to why PVAs outputs only focus on projections of growth rate and the Counterfactual of Population Growth Rate, and do not include Counterfactuals of Populations Size, which is considered the metric easiest to interpret by non-specialists”.
60. The population models run are density independent, as empirical information on density dependence in the populations being modelled are considered insufficient to inform a PVA assessment. There is therefore no mechanism within the model to prevent the population size from increasing without limits. Since the comparisons being made are between the two population scenarios (impacted but no compensation vs. impacted with compensation) it is likely that this will result in very large projected population sizes from the scenarios that include compensation measures, particularly when these are based on increases in both adult survival and productivity. Thus, the counterfactual of population size will be much higher as the models are density Independence. Since the models are being projected over 50 years, this difference will be much larger than at shorter time scales. The counterfactual of growth rate is much less affected by the density independence assumptions in the model and are therefore more useful in comparing scenarios.

• Are there other SPAs that should be assessed?

61. It is clear that this question required further explanation. The question was whether there are any additional SPAs, that are either predicted to be impacted or not impacted, that should be included in the assessment of the compensation measures. The question of whether impacts are not adverse will be addressed in the Report to Inform the Appropriate Assessment. A complete list of the SPAs, and features, that will be assessed, is shown in Table 1.35   Open ▸ .

Table 1.35 SPAs and the relevant qualifying features including in the sandeel fisheries compensation measures assessment.

How do we estimate the compensation ratios based on SPAs that benefit from the proposed measures but are not impacted?

62. It is typical for compensation measures to provide some level of surplus compensation to account for either uncertainty in the assessment or a gap in timing between impacts occurring on European site features and compensation measures occurring, or both. This surplus compensation is measured as a compensation ratio. This is the ratio of the predicted level of impact to the Natura network to the level of compensation secured. This ratio is commonly used for measures being applied to habitats in Special Areas of Conservation (SAC), where the ratio of habitat area lost due to a project compared to the area gained through compensation measures is a simple calculation. For the combination of impacts and compensation measures on populations of mobile species this approach is unlikely to be directly replicable.
63. Since the sandeel fisheries compensation measures being proposed would be very likely to have positive effects on all of the SPAs that border ICES SA4 and will be much greater that is required to offset the impacts of the proposed Development  only, it is likely that there will be surplus compensation to the SPA network for the features predicted to be impacted by the Proposed Development. However, the approach to assessing the combined effects of predicted impacts and sandeel compensation measures is to compare the predicted changes in population growth rate due to impacts only and the combined impacts and compensation measures. This is also based on a range of possible future changes in sandeel TSB in SA4, as it is uncertain what the possible future change in sandeel TSB in SA4 as result of compensation measures may be.
64. The simplest solution is to compare the increase in estimated number of additional birds in each SPA with connectivity to SA4 that would occur as a result of the predicted changes in sandeel TSB. This is a simple calculation based on the change in adult survival, change in adult productivity, and both of these combined (where appropriate). This should give sufficient information to estimate an approximate predicted compensation ratio.
65. We do not think that the predicted difference in end population size would be a meaningful approach to take, as the density independent assumptions made in the population models are known to be violated, so modelled population sizes are able to increase with no limitations which is clearly unrealistic.
66. An effective approach would be to use the calculated number of additional adult birds per annum in the population and/or the additional number of adult birds per annum in the population as a result of the increase in productivity combined with the survival to age at first breeding. This approach has the advantage that the total number of additional adult birds in the population in one year as a result of the compensation measures across the SPA network can be compared with the predicted annual losses of adult birds in the SPA network as a result of the Proposed Development. From these numbers an approximation of the compensation ratio can be calculated and therefore compared in a meaningful way between compensation scenarios. This information, combined with the CGR information across the scenario it will be possible to characterise the over-compensation even if the “compensation ratio” value is not identical to the typical approach used for habitats.

Scallop fishing and sandeels

67. Both NatureScot and RSPB requested further information on how this information may be taken forward as a compensation measure. At present we do not think that there is sufficient information to include this as a compensation measure. There is considerable uncertainty in the efficacy of restricting scallop dredging in order to benefit sandeels and it would not be possible to quantitatively predict the benefits of such a measure to the qualifying features of SPAs. However, this may be developed as part of the adaptive management, allowing gathering of the relevant information to predict the benefits to SPA seabirds should the proposed measures not be sufficient, as outlined within the Implementation and Monitoring Plan.

Climate change

68. In relation to the effects of climate change on the proposed compensation measures NatureScot noted, “The proposed compensation measures should cover the full operational phase of the project. Within this time, we might expect to see potential changes to sandeel distribution associated with warmer/stormier seas. If sandeel biomass reduces over time, other fisheries measures might need to be considered to make up the short fall of available prey in the later years of operation”.
69. It is important to note that climate change may cause changes to the baseline conditions as well as the compensation measures. The scenarios used for assessing compensation measures is based on a wide range of baseline conditions, from a TSB of 100,000 to 300,000 tonnes. The available information on SA4 TSB shows only four years with the TSB below 100,000 tonnes between 1991 and 2021. The available evidence therefore suggests that the future sandeel TSB in SA4 is unlikely to remain this low for long periods. At this sandeel TSB the change in demographic rates is relatively large across relatively small changes in TSB.  As such, relatively small increases in TSB as a result of the proposed fisheries compensation measures would have a more important difference in the demographic response compared to the baseline condition.
70. The response of sandeel populations to the effects of climate change are hard to predict at the level of accuracy and precision required to enumerate the effect. Clearly if climate change were to cause a large scale reduction in sandeel abundance in the North Sea this could have an effect on kittiwake populations. However, kittiwakes do nest much further south in Europe than the North Sea and forage on other species of fish in other areas. While there may be predictions of the negative effects of climate change on sandeels, there is also a possibility of positive effects on more southerly prey species that move north with changing climate (e.g. sardines, anchovies). This is speculative and it is unknown what the potential affect this might have on kittiwake populations in the North Sea.
71. The recent BTO report predicting the effect of climate change on seabirds concluded that there was a high risk of a decline by 2050 to kittiwake and puffin (a predicted decline in population size of 54% and 89% respectively) and a medium risk to guillemot and razorbill. It is important to note that as populations decline, so will their predicted impacts. This highlights the importance of climate change mitigation plans and the value of projects, including the Proposed Development, to this plan.
72. Further information on climate change effects on the assessment of the baseline conditions and impact assessment for the Proposed Development are provided in Chapter 20 (Inter-related Effects) of the EIA.

[1] https://www.ceh.ac.uk/isle-may-breeding-season-summaries

[2] https://www.dw.com/en/north-sea-warming-twice-as-fast-as-worlds-oceans/a-40427339

[3] https://www.ices.dk/community/groups/pages/hawg.aspx