New Forest breeding Curlew survey: 2016 results

Transcription

1 New Forest breeding Curlew survey: 2016 results Co-ordinator: Prof Russell B Wynn (Wild New Forest) rbw1@noc.ac.uk Executive summary A volunteer-based survey of breeding Curlews in the New Forest was undertaken in 2016, in response to concerns about a perceived population decline and decreased productivity due to pressures such as nest predation and recreational disturbance. Territory mapping in 15 broad areas of suitable wet heathland, bog and mire habitat, covering 86 km 2, produced just 40 pairs. This overall total, and the results of repeat mapping of 31 one-kilometre squares covered during previous surveys in 1994 and 2004, appear to confirm a dramatic population reduction of up to two-thirds in the last decade. A total of 19 Curlew nests were located from late April onwards, equating to roughly half of the monitored territories. Limited information was obtained on productivity in 2016, with only about nine successful hatchings reported and just five observations of chicks. A wide range of potential pressures was recorded, although only avian predators and recreational users were seen to actively disturb Curlews during the incubation phase (when most recording of pressures was undertaken). Engagement with land managers has highlighted potential mitigation measures for recreational disturbance, such as targeted signage at selected nest sites. Other ground-nesting waders were only surveyed in detail in six areas, with totals of Lapwing and Snipe being slightly lower than but broadly comparable to the 2014 HLS survey. Breeding Redshanks now appear to be restricted to the southeast margins of the New Forest, with a total of just four pairs reported from two areas in 2016.

2 No Curlews were ringed in 2016 despite several attempts early in the season, but a colourringed bird reported in the north of the New Forest, that is thought to have been ringed in Southampton Water in late October 2011, indicates that at least some breeding birds may be wintering locally. A recent workshop on Curlews in southern England concluded that there are probably <200 pairs left in this broad area, highlighting the regional importance of the remaining New Forest population, both in terms of numbers and maintenance of UK range. However, if the recent decline continues at the same rate, the species could become locally extinct in the New Forest within the next few years. Further survey work in 2017 and beyond will better constrain the New Forest population level, while planned use of nest temperature loggers, colour ringing and GPS tagging will provide vital information on pressures, productivity and movement ecology. 1. Introduction and aims The Curlew is undergoing rapid national decline, and as a consequence has become one of the highest conservation-priority bird species in the UK. The valley mires and wet heathlands of the New Forest still support a regionally important Curlew population, at the southeast edge of the current national distribution. Based on published data, this population is thought to be in the region of pairs, however, there is recent anecdotal evidence that the current population is actually at a much lower level and that both predation and recreational disturbance may be leading to reduced productivity. This study, undertaken from 2016 onwards, has four main aims: To search for and map all Curlew territories and nests in known breeding habitats in the New Forest To monitor discovered nests as frequently as possible through the breeding season to assess productivity and, where possible, the cause of any breeding failure To record pressures on nesting birds, e.g. predators, recreational disturbance, to assess potential impacts on productivity. To ring adults and pulli (with colour rings) to investigate site fidelity and movements during the incubation and inter-breeding periods This report presents the initial results of the first (pilot) year of survey work in 2016, and presents initial plans for further fieldwork in Survey methods A total of 14 experienced volunteer surveyors were each allocated survey areas covering up to 10 km 2 (and totalling 86 km 2 ), which are known hotspots for breeding Curlews based on previous surveys and recent casual records provided to Hampshire Ornithological Society (see Fig. 1 and Appendix A). It should be noted that this is a volunteer-based survey, and there will inevitably have been some variation in observer coverage and ability. However, it is not thought that this will have significantly impacted the overall results Territory mapping Surveyors were asked to search as much of their survey area as possible once a month in the period from March to July, i.e. a minimum of five visits. This covered the period from birds

3 arrival on territory though to chick fledging. Further visits in the May-June period were encouraged (if time allowed), to improve the chance of locating nests and assessing productivity. Visits were ideally undertaken within three hours of dawn, but the evening period within two hours of dusk was also targeted. Surveys were only to be undertaken in clement weather conditions, avoiding periods of cold, wet and/or windy weather. All Curlew records were allocated a six-figure grid reference, and time, habitat and behaviour noted, e.g. displaying/feeding/incubating birds. Surveyors were encouraged to record other groundnesting waders, including Lapwing, Snipe and Redshank (especially in the vicinity of Curlew territories) to assess their population and productivity compared to Curlew. Figure 1: Location map showing position of the 15 broad areas surveyed in 2016 (numbered red polygons), overlain onto one-km core squares identified as suitable habitat in previous 1994 and 2004 surveys (grey shaded squares). Note that although the 2016 survey roughly covered the same broad areas as previous surveys, there are some discrepancies. However, all of the 31 one-kilometre squares directly surveyed in 1994 and 2004 (shown as grey numbered squares above) were resurveyed in Nest finding Curlew nests are hard to locate and the sitting birds are prone to disturbance, so surveyors were encouraged to observe birds from a distance, preferably from an elevated but unobtrusive viewpoint with a telescope. Close approach to the nest was only undertaken in order to make a final confirmation of nesting, as this can negatively impact the birds and may also attract opportunistic predators such as corvids and foxes to the nest location Pressures To ensure pressure data were effort-corrected, and therefore of value to the relevant authorities (e.g. Forestry Commission and New Forest National Park Authority), observers were requested to undertake timed one-hour observations of pressures at intervals during the season. These timed counts only commenced when nesting was confirmed, with the nest site or core of the territory clearly identified. The timed one-hour counts were used to record any recreational activity or predators (e.g. corvids, foxes) within ~500 m of the nest site. This will result in an effort-based pressures database, collected from similar-size areas centred on

4 Curlew nests at a range of sites across the New Forest. Even if each observer can only commit to doing one or two of these counts per month, this will still yield a valuable dataset that can be added to in subsequent years Ringing Ringing activities were led by experienced, BTO-licensed A ringers (e.g. Graham Giddens and Tony Davis), supported by trainees as appropriate. Mist nets and tape lures were used in attempts to capture adults prior to breeding. Approximately 50 colour ring combinations were available, with slightly shorter rings allocated for any pulli that were ringed. A BTO metal ring and combinations of four colour rings were to be used Permissions and data All survey and ringing activities were undertaken with permission of the Forestry Commission (contact Andy Page) and Natural England (contact Jenni Thomas). Ringing activities were to be conducted under a BTO license, and the survey will contribute to national survey work being conducted by BTO and RSPB. Survey data will be provided to FC, NE, HOS and BTO. 3. Initial results 3.1. Territory mapping A total of 40 Curlew territories were identified during the 2016 survey (Table 1), which is considerably lower than previous survey total, including the 2014 HLS survey. Most birds appeared to arrive on territory in the second half of March. Area Size (km 2 ) Territories HLS Core squares Curlew in core squares (1 nest) (2 nests) (4 nests) (4 nests) (2 nests) (2 nests) (2 nests) (2 nests) Total (19 nests) Table 1: Number of Curlew territories (and nests) by area during the 2016 survey, and also in the same areas during the 2014 HLS survey. Number of core squares in each area (from 1994 and 2004 surveys) and numbers of Curlews in those core squares in 2016, also shown. Detailed data for other ground-nesting waders were also received for six areas (Table 2). Totals for these three species were more comparable to the 2014 HLS survey, although consistently lower, possibly due to these species not being the focus of the 2016 survey. A total of four Redshank territories were noted in two areas in the southeast of the New Forest (compared to 13 in three areas during the 2014 HLS survey), indicating that this species is very much on the brink as a breeding bird in the study area.

5 Area Lapwing HLS Snipe HLS Redshank HLS Total Table 2: Number of Lapwing, Snipe and Redshank territories in six areas where detailed records were received during the 2016 survey, and also in the same areas during the 2014 HLS survey. Note that 2016 data from these areas should be taken as minima, as these three species were not the focus of the survey Nest finding and initial productivity data A total of 19 nests were found from 24 April onwards, equating to about half of the mapped territories (most of these nests were found by Andy Page, an experienced wader fieldworker, e.g. Fig. 2). Figure 2: Curlew nest photographed at Longwater Lawn by Andy Page (left) and Tara Dempsey (right), showing the clutch of four eggs before and after hatching. The right-hand image generated some debate as to whether this represented successful hatching or predation. Although confirmation of a sitting bird is relatively straightforward, confirmation of hatching and subsequent fledging is harder to establish (e.g. Fig. 3), especially in the New Forest where birds with chicks have extensive areas of wet mire/bog habitat to disappear into. Consequently, observations of birds with chicks were made at just five sites (Brogenslade Bottom, Fir Pound, Leaden Hall, Strodgemoor Bottom and Hill Top), and a further four nests were located where hatching was thought to have occurred but was not confirmed.

6 Figure 3: Curlew nests photographed at Duckhole Bog by Russell Wynn on 03 July (right) and at Strodgemoor Bottom by Simon Currie on 12 June. In both situations, despite the presence of fragments of eggshell in the nest, the fate of the clutch could not be confirmed Pressures Timed pressure counts were made at several sites as follows: A nest at Duckhole Bog was watched on eight dates during incubation, between 14 May (nest found) and 03 July (nest confirmed empty success unknown), totalling 9.5 hrs. A variety of potential pressures were noted, including avian and mammalian predators (Fox, Grey Heron, Buzzard, Kestrel, Carrion Crow), recreational users (runners, walkers, dog-walkers, horseriders), light/commercial/military aircraft, and grazing animals (ponies and deer). Most of the observed pressures instigated no response from Curlews. However, a Curlew response was noted on 28 May when a runner in a neon top with a spaniel off the lead, passing ~50 m from the nest, instigated an alarm call from the sitting bird. In addition, one or both Curlews were seen acting aggressively towards single Carrion Crows in the vicinity of the nest site on 24 May and 12 June. At the Brogenslade Bottom nest, 5.5 hrs of observation were undertaken during incubation between 05 May (nest located) and 28 May (nest empty successful hatching). Buzzards, Carrion Crows and dog-walkers (with dogs under control) were all noted, but no disturbances were seen within that period. Interestingly, the placement of Dogs under Control signs by the National Trust at either end of a track passing close to the nest site, at the start of incubation, was felt by the site observer to have had a positive effect in reducing disturbance, with an estimated 90% reduction in dog-walker traffic. Prior to the signs being emplaced, disturbance of Curlews by dogs off the lead had been noted on two dates in late April/early May (Fig. 4).

7 Figure 4: Photo by Martin Bennett, showing two dogs ranging extensively off the lead (and off the path, marked by dashed red line) in the vicinity of a Curlew territory at Brogenslade Bottom in late April. At Black Gutter Bottom, although a nest could not be located, pressures noted throughout the breeding season included corvids (up to 40 Carrion Crows and four Magpies), Buzzards, and recreational users (particularly dog-walkers). There was also an interesting observation on 23 May of a Curlew being pursued by a female Peregrine over this area, about 1 km from a presumed nest site. Despite these disturbances, at least one nest produced chicks in this area. Another report of a pair of Curlew alarm calling and chasing a Peregrine was received from the Cranes Moor area on 12 June Ringing Attempts were made to catch adult birds early in the season at two locations from 20 March onwards, using tape lures and mist nets at sites where up to eight birds were known to be communally roosting near open water. Despite some near misses, no birds were captured, but a lot of useful lessons were learned about techniques. There were also plans to colour ring pulli later in the season, but due to methodological concerns expressed by some project partners, and a delay in obtaining required permissions from NE, this activity was pended until Of note was a colour-ringed Curlew reported in the north of the NF in April, that is tentatively thought to be a bird ringed by Pete Potts on 29 Oct 2011 at Weston Shore in Southampton Water, hinting at local wintering of our breeding Curlews. In addition, a colourringed Lapwing photographed on Beaulieu Heath West had been ringed as a chick in the Avon Valley in Discussion The total of 40 Curlew territories (Table 1) located during the 2016 survey is much lower than that obtained during the 2014 HLS survey. Although some of the surveyed areas received incomplete coverage during 2016, there is high confidence amongst experienced field observers that the overall NF population is now <50 pairs. In some areas, it was clear that the HLS figures significantly over-estimated the number of territories, likely due to methodological

8 differences (for example, in Area 7, five territories were located in 2016, compared to 20 during the HLS survey). Curlews are highly mobile, particularly in the early stages of the breeding season, and one of the main challenges for 2016 observers was fully understanding the areal range occupied by each territorial pair (as well as recognising incursions by wandering non-breeding birds). This may be a reason for the discrepancy between the 2016 data presented here and the 2014 HLS survey data. Curlew have been comprehensively surveyed twice previously in the NF, in 1994 and The estimated numbers of Curlew territories identified in 93 one-kilometre squares containing >15 ha of wet heath and/or bog were 132 in 1994 and 99 in These numbers are likely to be slight under-estimates, as a small number of territories may have occurred in other habitats. However, overall, it would suggest that the 2016 total of 40 territories (with a comparable survey area of 86 km 2 ) represents a loss of up to two-thirds of the NF Curlew population over the last decade, which is an acceleration of the reduction noted in the previous decade. The 1994 and 2004 population estimates provided above were derived from extrapolation from 31 directly surveyed one-kilometre squares of suitable habitat. The number of Curlew territories identified in these 31 core squares during the 1994 and 2004 surveys was 44 and 33, respectively. The number identified in these same squares during fieldwork in 2016 was just 10 (Table 1), further supporting the likely reduction in the breeding population to roughly one-third of that in the early 2000 s. Nest locations were found in about half of the 40 territories (Fig. 5), but observations of chicks were only made at five sites and successful hatching was thought to have occurred at just four additional sites. It should be noted that confident identification of successful hatching is challenging when undertaking sporadic site visits, but these tentative indications of low productivity are perhaps unsurprising given the large number of potential pressures observed at some sites, both during and pre-incubation. These mostly involved generalist predators (particularly foxes and corvids) and recreational disturbance, although other direct pressures such as Peregrine Falcons were also noted. Figure 5: Photos by Andy Page showing examples of occupied Curlew breeding habitat at Fir Pound (left) and Holmsley Ridge (right). There was some evidence that birds are relatively tolerant of human and animal traffic when this is confined to marked trails, but quickly become disturbed when nest sites away from the trail are approached. This does provide hope that more targeted signage, as was utilised effectively by the National Trust at Brogenslade Bottom, may be a useful mitigation method.

9 Although data on other ground-nesting waders were not consistently collected at all sites during the 2016 survey, initial analysis of six sites where detailed data were collected indicates they are more comparable to previous surveys (Table 2). It should be noted that these findings represent a single (pilot) year only and should there be treated as tentative; it is hoped that ongoing fieldwork, as described below, will help to further constrain the NF Curlew population figure and allow more accurate quantification of the level of population decline, as well as provide more insight into productivity and potential drivers of low productivity. It should also be noted that this survey provides no information on other potential drivers of the observed population reduction, e.g. changes to grazing pressure and water levels at breeding sites, and food/habitat availability at wintering sites. 5. Future survey plans It is anticipated that the current programme of Curlew survey work in the New Forest will continue for at last three to five years, to ensure collected data are increasingly robust and to overcome vagaries of inter-annual variability, e.g. spring weather conditions. In addition to a continuation of territory mapping to further constrain population numbers, additional tools will be implemented to aid understanding of productivity and movement ecology. For example, given the labour intensive nature of direct nest observation, it is planned to use miniaturised temperature loggers in the nest to identify periods when incubating birds leave the clutch temporarily or permanently, due to predators, disturbance, inclement weather or hatching (additional to routine changeovers). There are also plans to colour-ring adult birds at early season roosts and during incubation, and also potentially to deploy a small number of GPS loggers, under the co-ordination of Pete Potts and a team of qualified ringers. 6. Acknowledgements Survey work was supported by a number of organisations, including Hampshire Ornithological Society, Forestry Commission, Natural England, British Trust for Ornithology, Royal Society for the Protection of Birds and New Forest National Park Authority, particularly relating to permissions, advice and recruitment of volunteers. Huge thanks are due to the following surveyors who were responsible for collecting field data in 2016: Mike Baker, Martin Bennett, Andy Broadhurst, Simon Buckell, Bob Chapman, Simon Currie, Tony Davis, Tara Dempsey, Olly Frampton, Graham Giddens, Andy Page, Ade Parker, Richard Reeves, Marcus Ward, Russell Wynn. Thanks also to everyone who provided additional records and advice that contributed to the survey, including several participants at the southern Curlew workshop in Feb 2017 (Fig. 6 below).

10 Appendix A: Survey areas and surveyors Grid refs in bold represent 31 randomly selected squares covered during previous surveys in 1994 and Setley/Widden/Hincheslea (Russell Wynn) Key squares = SU2700; SU2800; SU2900; SZ2899; SZ South Weirs/Holmhill (Russell Wynn) Key squares = SU2503; SU2502; SU2602; SU2601; SU2702; SU2801; SU2701; SU Holmsley/Clayhill (Olly Frampton) Key squares = SU2001; SU2101; SU2201; SU2301; SU2000; SU Bisterne/Kingston/Strodgemoor (Simon Currie) Key squares = SU1801; SU1802; SU1902; SU1803; SU Backley/Bratley (Ade Parker/Mike Baker) Key squares = SU2106; SU2207; SU2208; SU2308; SU Milkham/Broomy/Ocknell (Andy Page) Key squares = SU2109; SU2110; SU2211; SU Latchmoor/Sloden/Fritham (Andy Page/Martin Bennett) Key squares = SU1811; SU1812; SU1813; SU1912; SU1913; SU2012; SU2112; SU2212; SU Godshill/Cockley/Black Gutter (Martin Bennett/Andy Page) Key squares = SU1816; SU1916; SU1915; SU2015; SU2016; SU Howen/Longcross/Pipers Wait (Simon Buckell) Key squares = SU2315; SU2415; SU Beaulieu Heath West (Marcus Ward) Key squares = SZ3499; SZ3599; SZ3699; SU3400; SU3500; SU3600; SU3401; SU3501; SU Beaulieu Heath East (Bob Chapman) Key squares = SU4102; SU4003; SU4103; SU4004; SU Dibden Bottom/Gurnetfields (Bob Chapman) Key squares = SU3905; SU3906; SU3805; SU3806; SU3705; SU Woodfidley/Yew Tree/Beaulieu Road Sta (Tony Davis) Key squares = SU3504; SU3405; SU3505; SU3605; SU3406; SU3506; SU Matley Holms/Ipley (Tony Davis/Richard Reeves) Key squares = SU3407; SU3507; SU Longwater/Matley (Richard Reeves/Tara Dempsey) Key squares = SU3207; SU3208; SU3209; SU3308