Abstract
Over the period October 2003 to August 2004 E.ON UK (formerly PowerGen) Renewables Offshore Wind Ltd. constructed a wind farm comprised of 30 high capacity turbines on Scroby Sands, a dynamic sand bar system approximately 3km offshore of Great Yarmouth, Norfolk.
The development is located directly offshore from the Great Yarmouth North Denes Special Protection Area (SPA), designated as a result of the presence of the largest colony of Little Tern Sterna albifrons in the UK, protected and managed by the Royal Society for the Protection of Birds (RSPB). Little terns are an endangered species and are in long-term chronic decline in the UK, reducing by some 27% between 1985-87 and 2000. From 1983 to 2001, North Denes regularly held over 200 breeding pairs, >10% of the UK total and around 2% of the European population.
An Appropriate Assessment of the likely impact of the proposed wind farm concluded that although Little terns used Scroby Sands when feeding, the impact of the wind farm on local bird populations was likely to be of moderate significance at most. Such is the importance of the site and its species that Department for Environment, Food and Rural Affairs (DEFRA), after discussion with English Nature (EN), instructed that monitoring of Little terns be undertaken to validate these conclusions. Determination of any impact relied on the suitability and rigour of the monitoring techniques employed. Suitable methodologies were developed after further consultation with the Royal Society for the Protection of Birds (RSPB).
The primary aim of this project was thus to monitor the impact of the proposed wind farm upon Little terns, although impacts on other bird species, of which 38 species have been recorded during the study thus far, were to be included as a matter of course.
The work was to be divided into several areas before and after construction:
• Feeding studies i.e. spatial and temporal distribution of foraging birds;
• Breeding colony studies − focusing on chick feeding ecology;
Prey studies i.e. spatial and temporal distribution of prey at sea Bird strike studies; which were to be added after construction.
The two years (2002 and 2003) prior to construction were to form a baseline against which future change relative to the presence of the wind farm could be evaluated. Monitoring following piling and during turbine construction was undertaken in 2004. Post-construction impacts may be evaluated in monitoring conducted in 2005 and 2006.
Reversing the pattern of the previous 20 years, in all years of the study thus far Little terns have not breed in numbers at North Denes. In 2002, the colony was destroyed by a single act of vandalism, although a small number of pairs (c. 7) managed to persist and fledge chicks (c.5). In 2003, helicopter patrols were thought to displace birds before breeding was attempted, although 10 pairs did eventually nest, fledging just 2 chicks. In 2004, 40 nests were put down over the course of the protracted season but no chicks fledged. With 1996, this became the worst year on record for the colony. In all years, Little terns established at Winterton some 12km to the north, where they had formerly bred and which is included in the SPA. In 2002, a minimum of 124 pairs raised a minimum of 43 chicks, whereas in 2003, 233 pairs fledged 447 chicks, the greatest number of chicks raised from a single colony in the UK since records began in 1969. In 2004, 150 nests were put down, but all failed, which was mainly attributed to the lack of prey, although disturbance and predation were also thought to play a role. A comparison between the two colonies was made to promote further understanding of Little tern breeding and feeding ecology, which was deemed essential to ultimately assess the impact of the proposed wind farm.
In all years monitoring was conducted throughout the breeding season (May/June-August) of Little terns, with the following recorded approximately every two weeks:
• Numbers of birds at different sampling stations across both study sites;
• Parameters of foraging activity including dive and fish capture rate;
• Provisioning rate to chicks;
• Density and population dynamics of available prey, particularly fish.
A small surface tow net, sampling the upper 30cm of water was specifically developed to sample the invertebrate and fish prey available to Little terns. This net has been towed 340 km during the course of the study and a total of 46 potential prey species, including 14 fish species, have been captured. Of these, clupeid fish – Herring Clupea harengus and Sprat Sprattus sprattus – the crustacean Idotea linearis and the Ghost shrimp Schistomysis spiritus were by far the most numerous. Young-of-the-year (YOY) clupeid fish were overwhelmingly the most important dietary item nd form the mainstay of tern breeding success.
In accordance with the known distribution of spawning and nursery areas of Herring and Sprat, Scroby appears to be by far the most important nursery area for clupeids along the stretch of coast sampled (including into North Norfolk). What are thought to be locally born Herring appear in the first samples in May at about 30mm in length. Peak numbers of Herring are recorded in June, before numbers rapidly decline perhaps as these fish move further offshore. Little tern breeding is thought to be closely tied in with the seasonal pattern of Herring, with chick development occurring in the peak phase, with fledging prior to decline of fish density. Sprat spawn offshore and larvae appear to be transported into the area through residual drift. Sprat appear in samples at about 15 mm in June, reaching a smaller peak of abundance than Herring by late July before again disappearing almost completely from samples in August. Late or re-nesting terns, particularly if these have moved colony may rely on this later peak in Sprat although they may still experience difficulty in finding enough food for hicks.
During the period of occurrence, the fish are patchily distributed and are considerably more abundant inshore (up to 2 individuals per m-2), with the best sites immediately adjacent to the North Denes colony and Caister. Coupled with less inter-annual fluctuation than other sites – Winterton for example appears to be dependent on overspill from Scroby – this makes North Denes the colony location of choice. Within Scroby, the concentration of fish at North Denes and Caister is almost certainly because these sites tend to have more turbid water, which is thought to bring the fish closer to the surface and within reach of the terns. Terns as well as fish are thus significantly associated with more turbid water. Consequently, during surveys in 2002 and 2003 Little terns were encountered in largest numbers immediately adjacent to both colonies and were only sporadically recorded in small numbers over Scroby Sands themselves, typically early and late in the season before and after breeding.
number of birds yet encountered at a site recorded in the southern part of Scroby in early season. Moreover, birds were recorded on several occasions on the outer edge of Scroby near the wind farm, where they had never been recorded before. Birds did therefore not appear to be displaced from using Scroby as a result of turbine construction. Changes in the nature of Scroby including the formation of a subsidiary sand bar through the wind farm appear to have offered birds additional foraging grounds, perhaps especially for invertebrates. An increase in quantity and/or quality of foraging area may thus be tempered by the potential for greater risk of collision of birds with turbines. The continued use of radio telemetry, developed for use on Little terns for the first time in the UK during this study, is seen as a key tool in the assessment of this risk during any post construction monitoring. Combined with further data gathering on the flight height of Little terns, the relative amount of time spent at risk may be determined.
Colony location and foraging success are closely tied in with fish abundance. Whilst the latter is an essential prerequisite of successful breeding, breeding performance may also be constrained by disturbance and predation of eggs and chicks. When fish prey is abundant, birds forage significantly closer to shore and enjoy a significantly higher rate of dives producing fish. By far the lowest rates of the latter (c. 30% of peak rates) were recorded in 2004 in accordance with a virtual failure of recruitment of young-of-the-year Herring. The shortage of prey was so severe as to mean metabolic constraints came into operation and nesting birds which had already adapted to the conditions by laying a reduced clutch size (0.5 egg less than average) were forced to abandon their nests. Radio telemetry showed that birds in 2004 travelled over twice as far in forging bouts lasting twice as long in 2003, with the maximum distance traveled in a single foraging bout a staggering 25 km. Radio-tagged birds spent an average of 72% of their time foraging compared to 56% in 2003 even when they had chicks to feed. When nesting, birds appear to be tied to home ranges of around 4 km2. In 2004, after failure, some birds ranged widely within average ranges of 25 km2 incorporating the entire stretch of coast between North Denes and Winterton, seemingly in an attempt to exploit any available food supply.
Although inter-annual variation in recruitment of YOY clupeids is a known phenomenon and to be expected, particularly according to the North Atlantic Oscillation (NAO), virtual failure of Herring in 2004 was thought to be exceptional, if only because of the unprecedented failure of the Little terns, which appear to depend on Herring recruitment. Without further data, an analysis of possible explanations of the lack of YOY could only be speculative, although this did reveal potential for removal of adults perhaps through commercial activity and a NAO-temperature-phytoplankton-zooplankton-Herring larvae link. Moreover, recent research on the impact of underwater noise from pile driving on fish with both avoidance and mortality, indicated a potential for a short-term impact of piling of the turbines conducted in November-December 2003, in what is the documented critical spawning and initial development period for Herring in the area.
With no reason to suggest that successful fish spawning and recruitment has not occurred over the winter/spring of 2004/05, coupled with the prospect of a return of chicks fledged from Winterton in 2003 to the area for the first time, pro-active wardening and protection using all available means has been planned at both the prospective North Denes and Winterton colonies.