Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Study of small mammal populations within two Barn owl corridors at Folly Farm

Study of small mammal populations within two Barn owl corridors at Folly Farm Volume 2 † Number 2 † June 2009 10.1093/biohorizons/hzp018 ......................................................................................................................................................................................................................................... Research article Study of small mammal populations within two Barn owl corridors at Folly Farm Alex Keene* School of Science and the Environment, Bath Spa University, Newton St Loe, Bath BA2 9BN, UK. * Corresponding author: 62 Ashgrove, Peasedown St John, Bath BA2 8EF, UK. Tel: þ44 1761 420094. Email: alex.thewalkingcamera@virgin.net Supervisor: Ian Todd, School of Science and the Environment, Bath Spa University, Newton St Loe, Bath BA2 9BN, UK. ........................................................................................................................................................................................................................................ This study examines small mammal populations present within Barn owl corridors on Folly Farm, an Avon Wildlife Trust Reserve located near the village of Bishop Sutton in Bath and North East Somerset. Two corridors were chosen, the primary difference between the two sites being only one has undergone management (grazing). The focus of this study was the Microtus agrestis (Short-tailed field vole) population, the most frequently taken prey item by Barn owls. Apodemus sylvaticus (Wood mice) and Sorex araneus (Common shrew) populations are also discussed as they are frequently taken. Using Longworth live traps, 600 trap-nights data were collected from three sessions in November 2006, February and March 2007. Although M. agrestis was the most abundant species in both corridors, they were more prevalent in the un-grazed corridor (comprising 19 of the 31 individuals). In the corridor that had undergone manage- ment, fewer M. agrestis were captured (eight), although a higher species diversity and richness was recorded. Unusually for a grassland habitat, there were nearly as many A. sylvaticus as there were M. agrestis (seven compared with eight) in the grazed corridor. Some small mammal species not usually found in grassland habitats were captured; explanations for these seemingly anomalous results are discussed. Differences in population characteristics between the two corridors are discussed including: sex ratio, weights, seasonal variation and age structure. Pellet analysis from the nearby pair of Barn owls showed that they were preferentially hunting M. agrestis; the pellet data largely mirrored findings of the trapping data. Key words: Short-tailed field vole, small mammal, Barn owl corridor, pellet, Longworth trap. ........................................................................................................................................................................................................................................ Barn owls may consume as many as 5000 small mammals, Introduction with predominance of Microtus agrestis. Changes in small mammal abundance and species composition can have Barn Owls serious effects on the breeding success of Barn owls. Barn The Barn owl, Tyto alba, has suffered in recent decades as a 1 owls cast an average two pellets a day and the structure of result of intensive agriculture, although it is suggested that 2 small mammal communities can be deduced from the com- climatic deterioration may also be a significant cause. parative frequency of different species in the pellet sample. Between 1932 and 1998, Barn owl populations declined in 3 The Mammal Society National Owl Pellet Survey showed Britain from 12 000 pairs to ,3000. A census carried out that M. agrestis make up the bulk of prey. between 1995 and 1997 indicated that the decline has halted if not reversed in some areas. Unfortunately, their Small Mammals plight was once again highlighted in 2006, with national breeding failure rates as high as 70% (Andre Fournier, Changes in farming practice and the decline of other suitable personal communication). small mammal habitats, as discussed by Harris et al., have Barn owls are opportunists and will hunt over open led to a decrease in many small mammal species, particularly 5 3 countryside. Their ideal habitat is rough grassland with a M. agrestis. This has led to a variety of studies focused on thick sward and deep litter layer supporting high densities both seasonal and annual cycles within populations at a 10– 12 13 of small mammals. In an average year, a breeding pair of single location and nationally. Previous small ......................................................................................................................................................................................................................................... 2009 The Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 155 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... mammal surveys have been conducted locally looking specifi- primary aim of protecting and enhancing its wildlife 14, 15 32 cally at populations present in Barn owl feeding areas. value. Over 28 ha of the farm are designated as a biological This study looked at the overall composition of small Site of Special Scientific Interest (SSSI). mammal fauna within two corridors, three species were of Six Barn owl corridors were instated in 1999/2000. The particular interest: M. agrestis, Apodemus sylvaticus and pair of owls present on Folly Farm failed to breed in 2006 Sorex araneus. Found in rough grassland throughout the for the first time since monitoring began in 1992 (Andre UK mainland, the population dynamics of M. agrestis has Fournier, personal communication). There is no strict been studied in some detail. Densities can range from cutting/grazing rotation in place within the corridors; 9 17 1/ha in mixed farmland up to 250/ha in rough grassland, those with stock fencing are grazed, others are mown, with home ranges as large as 1.4 ha. A. sylvaticus is also some are even cut by hand. Management is accessed highly mobile with similar home ranges, 0.26–1.77 ha in annually and is dependent on stock and labour availability. farmland. It is widely distributed in woodland, field and Until now, no small mammal survey work has been con- hedge habitats, but also occurs in sand dunes, gardens and ducted in the corridors. urban areas. Densities of 0.5–17.5/ha are typical on Time and resources made it impossible to survey all six farmland. S. araneus are also found in a wide variety of corridors and therefore two were chosen, the primary differ- habitats, including grassland, arable and woodland. They ence between them being only one has undergone manage- occupy relatively small home ranges, often less than ment. Site 1, the ‘un-grazed’ corridor, has received no 0.1 ha with densities of 5–69/ha. management since being fenced off at the end of 1999, Populations of some small mammals fluctuate. The causes other than having ragwort pulled and thistles topped when of cyclic population changes, and the role that predators infestations occur. The result of this is a thick, deep sward play, have been largely unresolved, despite high volumes of and relatively high species diversity. The best NVC match literature. Specialist hunters, such as Barn owls, are par- is MG1b, an Arrhenathetum community, typical of ticularly reliant on small mammal populations, especially un-grazed grassland on neglected pastures and meadows. M. agrestis. Populations of these predators increase when There are two Lime trees, Tilia cordata, within the corridor. voles are abundant and contract when numbers are low. Site 2, the ‘grazed’ corridor, has been grazed twice, first in Numerous studies have tried to determine what influences August 2000 and again in 2004. This site is botanically small mammal community composition. Gause demon- less interesting, the sward less dense and vegetation height strated both empirically and theoretically that coexistence substantially lower. NVC results produced a classification of competing species is only possible if each have different, of MG10/MG9. These plant communities occur mainly in specialized abilities. Several authors have attempted to fields and associated boundaries, and are often associated find correlations between plant and animal species richness with poorly drained permanent pasture. There are two and diversity; some have shown correlations, others veteran Oaks, Quercus robur, present; the most southerly 25 26 have not. Rozenweig and Winakur found that spatial contains a Barn owl box. variations in the density of some species are responses to Trapping spatial characteristics of their environment. Dueser and Shugart both found that simple measurements of density Trapping was undertaken using Longworth live-capture and height of vegetation within different strata provided small mammal traps. Both corridors were trapped definite correlations with the densities of species present. simultaneously during 3 and 4 days, sessions in the A structurally diverse habitat is likely to contain a larger mid-November 2006, early February and late March 2007. variety of food sources and nesting sites, providing niches The traps were prepared with hay bedding, mixed birdseed 28, 29 for predatory species as well as protection for prey. If and blowfly puparia for bait. Due to the linear structure of mobility allows then mammals have been shown to distribute the habitats, and their effectiveness with assessing commu- 30 36 themselves between habitats in relation to habitat quality. nity structure, traps were laid in transects. Twenty-five The most important factors being food availability, food trapping points were marked at 5 m intervals within each quality, predation risk and competition. corridor; the Longworth traps were then located + m from the mark. On Site 2, the transect was located at the southern end of the corridor to avoid a regularly used foot- Methods path. Traps were deliberately placed in positions that indi- cated the presence of, or would be favoured by, small Study Area mammals. All field signs were noted and located by relevant Avon Wildlife Trust has owned Folly Farm since 1987. It is trap number. located near Bishop Sutton in the Chew Valley (Grid refer- There were 200 trap-nights in each session, a total of 600 ence ST607604). Much of the farm has escaped modern, trap-nights for the whole programme. The traps were laid intensive practices and is traditionally managed with the and set on the afternoon of the first day and checked in the ......................................................................................................................................................................................................................................... 156 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... Table 1. Total number of captures and individuals caught during the morning and evening of the next 3 days, then checked and survey period, where individuals are first capture of an animal (i.e. no removed on the morning of the fourth day. visible fur clip) All small mammals caught, except shrews, were identified to species, weighed to the nearest gram using a 50  0.2 g Species Number of captures Individuals ................................................................................................................ Salter spring balance, aged, sexed and breeding condition M. agrestis 34 27 noted. All species captured were assigned to one of the fol- A. sylvaticus 14 11 lowing three age classes: adult, subadult or juvenile, accord- S. araneus 88 ing to Gurnell and Flowerdew. New captures were given a S. minutus 33 unique fur clip. All mammals were released at the point of capture. All shrews caught were identified to species, M. minutus 11 weighed and released immediately. Environmental data C. glareolus 11 were recorded every time the traps were checked, giving an N. fodiens 11 account of weather during the previous 12 h. Total 62 52 Vegetation of both corridors was largely homogenous throughout; therefore, detailed floral surveys around each case of M. agrestis and February in the case of A. sylvaticus trap were deemed unnecessary. Instead physiognomic and Sorex sp. (Fig. 1). vegetation classification was used; a popular method of habitat classification in researching habitat selection and utilization. Vegetation height was measured above the Sex ratio and reproductive condition entrance to each trap and recorded as short (,10 cm), The sex of shrews was not assessed. The overall sex ratio of medium (10–50 cm) or long (.50 cm). all other species caught in November and February was close to 1:1; however, by March, the ratio was skewed in favour of males (1.8:1) (Table 2). Between November and February, all Owl Pellets captures were non-breeding (condition of shrews was not Owl pellets were collected from below the Barn owl roost on assessed), all males having abdominal/small testes and all Folly Farm in May and June 2006. The roost was located in females imperforate. In March, all male captures were still the SSSI grassland 500 m from the survey sites, making non-breeding except for two A. sylvaticus caught in the both corridors well within hunting range of the owls. The un-grazed corridor, recorded as having testes large and pellets were soaked in warm soapy water, teased apart and scrotal, and all females were still non-breeding. all mammals remain cleaned and laid aside. Identification was carried out on skulls and lower jaw bones using a Age structure and weights 10 hand lens. Species were keyed out according to Juvenile mice were only captured in November, when they Yalden and Morris. made up 14% of the population (Fig. 2). At this time, the rest of the population was made up of a nearly even mix of adults and subadults. By February, the majority of the popu- Results lation is adult, 73%. The adult proportion has reached 100% Trapping A total of 62 small mammals were captured over the 600 trap-nights (summarized in Table 1). One Sorex minutus was found dead and one A. sylvaticus escaped before being processed. The vast majority of captures were recorded at dawn and therefore captured overnight. The two captures during the day were both S. araneus, one during the November session and the other in March. Both were adults and recorded in the un-grazed corridor. During all three sessions, there were no captures on the first day and highest number of captures were made on the final day. Small mammal catches were relatively evenly distributed across the un-grazed corridor, whereas the grazed corridor had a more patchy distribution pattern, with a clear lack of captures in the first seven traps. There are distinctly different Figure 1. Seasonal change in population structure of all species captured. patterns of seasonal population change in the M. agrestis and Total number of individuals captured in November ( ), in February (A) and A. sylvaticus populations. Numbers peak in November in the in March (B). ......................................................................................................................................................................................................................................... 157 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... Table 2. Number of male and female captures, of all species, in both corridors Trapping session M. agrestis A. sylvaticus Total (all species) Male Female Male Female Male Female Ratio ........................................................................................................................................................................................................................................ November 6 8 3 2 10 11 1.0:1.1 February 4 6 3 2 8 9 1.0:1.1 March 6 4 3 0 9 5 1.8:1.0 Table 3. Species composition, richness and diversity (Simpson’s index) in the grazed and un-grazed corridors Species Un-grazed corridor Grazed corridor ................................................................................................................ M. agrestis 19 8 A. sylvaticus 47 S. araneus 62 S. minutus 12 M. minutus 1— C. glareolus —1 N. fodiens —1 Total number of individuals 31 21 Figure 2. Age classes of all mammals caught in both corridors. Whole bar Species richness 5 6 is total population, is the number of adults, A the subadults and B the Species diversity 0.6 0.8 juveniles. was significantly higher in that corridor (X ¼ 4.48, df ¼ 1, p  0.05). The species diversity was higher in the grazed corridor (Simpson’s index of diversity, 1 2 D ¼ 0.8) than in the un-grazed (0.6). Five different species were recorded in the un-grazed corridor, the majority were M. agrestis (19) followed by S. araneus (6) then A. sylvaticus (4). Six species were recorded in the grazed corridor; the highest occurrence was still that of M. agrestis (eight) then A. sylvaticus (seven). Micromys minutus was only recorded in the un-grazed corridor, and Clethrionomys glareolus and Neomys fodiens were only recorded in the grazed corridor. Figure 3. Mean weight of individuals in all three sessions. The mean weight of M. agrestis ( ), A. sylvaticus (B), S. araneus (A) and S. minutus ( ). Movement of individuals by March. In Fig. 3, mean weights of S. minutus remained Ten small mammals were recaptured, eight of these in the unchanged between February and March. A. sylvaticus and un-grazed corridor. Seven were M. agrestis and three S. araneus weights remained unchanged between November A. sylvaticus. The mean distance travelled was 17.5 m in and February but increased in March. The mean weight of the un-grazed corridor, higher than that recorded in the M. agrestis fell, although not significantly, from November grazed corridor (5 m). The mean distance travelled by to February (t ¼ 1.54, df ¼ 20, p , 0.138). The mean M. agrestis was higher, 18.6 m, than A. sylvaticus, 6.7 m. weight of M. agrestis rose significantly between February These figures will however be skewed by the movement of and March (t ¼ 23.56, df ¼ 13, p , 0.003). an individual M. agrestis in the un-grazed corridor, travelling 65 m in a 24 h period. Only one long-term movement was Species recorded, between capture on the 9th February and recapture The total number of small mammals present (Table 3) on the 25th March the M. agrestis moved only 15 m. There was not significantly higher in the un-grazed corridor was no recorded movement between the two corridors, a (X ¼ 2.88, df ¼ 1, p  1). The proportion of M. agrestis minimum distance of 150 m. ......................................................................................................................................................................................................................................... 158 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... Effect of vegetation height on trapping success The proportion of mammals caught in the short and long grass is different between corridors (Fig. 4). More mammals were captured in the medium length vegetation (10–50 cm) in both corridors but probably this is an artefact of the greater availability of traps within this length. In fact, in the un-grazed corridor, a preference for long vegetation and an avoidance of short vegetation was shown. Although the differences were more subtle, the opposite was recorded in the grazed corridor, with more mammals preferring short vegetation. Figure 5 shows the un-grazed corridor to have the highest capture rates of M. agrestis in traps placed Figure 4. Vegetation length in which mammals were captured on both sites. L, long (.50 cm); M, medium (10–50 cm); S, short (,10 cm). under long vegetation, with 13% of traps making a Whole bar represents the number of traps placed in different vegetation capture. A. sylvaticus and Sorex sp. both showed a slight heights, area shaded ( ) is the total number of captures within that preference for the medium length vegetation. In the grazed length and the figure in white is the percentage of captures within each corridor, these trends are reversed for all three species. length. Figure 5. Vegetation length in which different species were captured on both sites. Bar represents the percentage of captures made within available traps at each different vegetation height, in both corridors. ......................................................................................................................................................................................................................................... 159 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... Table 4. Percentage of species trapped in both corridors, in owl features were vital corridors between preferred habitats. pellets from Folly Farm and in pellets analysed in The Mammal The presence of streams running either end of the corridor Society National Owl Pellet Survey suggests a transient capture. The majority of small mammals were captured at dawn. Species Trapping Owl pellet data Mammal Society data Owl pellets There were two exceptions, both S. araneus, caught in the Mean% May June Mean% Mean% un-grazed corridor. This is little surprise as A. slyvaticus ................................................................................................................ and M. agrestis are mainly nocturnal, whereas Sorex M. agrestis 53 26 28 54 43.4 species can be active night and day throughout the year. A. sylvaticus 24 17 4 21 16.6 During all three trapping sessions, no captures were made S. araneus 12 8 2 10 19.8 on the first day. There is no obvious reason for this other S. minutus 6 7 — 7 8.9 than the small mammals being ‘trap-shy’. This reluctance C. glareolus 4 4 1 5 5.5 to enter ‘new objects’, particularly by Microtus species, is M. minutus 3 — — 0 2.2 often found. Three nights of trapping is considered suffi- N. fodiens 4— — 0 ,1 cient to allow trap-shy mammals to accept the presence M. domesticus 02 1 3 ,1 of traps and therefore give an accurate picture of the population. In arable landscapes, seasonal variations in A. sylvaticus densities have been shown to vary dramatically. The For example, the long vegetation records the lowest capture 9 results in this study agree with Harris et al.; although rates for M. agrestis. there is seasonal variation, with a winter peak, these cycles are not as marked in marginal habitats such as grassland. Owl Pellets Similar trends were seen for the Sorex species, although Table 4 shows the results of the pellet analysis alongside a these figures should be examined with caution, the summary of the trapping data in Table 1 and the results February peak could purely be a reflection of the increased from the Mammal Society’s National Survey (1993– number of traps without ‘shrew-holes’ (an unavoidable con- 2005). The Barn owl pellets recovered in May contained sequence of having to borrow traps at short notice). an estimated 114 individual mammals and pellets collected M. agrestis show an opposite seasonal change in population in June contained 55; a total of 169. One hundred of these with numbers dropping off in February and increasing again were identifiable to species. M. agrestis formed the majority by March. As the March increase could not be down to of prey species followed by A. sylvaticus and then Sorex spp. breeding, it must be a result of immigration from an adjacent The rank of species in all three data sets is the same with two habitat or possibly an inaccurate assessment of the popu- exceptions: A. sylvaticus and Mus domesticus are ranked lation in February, with a percentage of M. agrestis alluding lower in the Mammal Society results than in the data sets capture. Low numbers meant no trends could be seen for the from Folly Farm. remaining species. If these seasonal fluctuations are accurate, the peak in A. sylvaticus numbers coinciding with a trough in M. agrestis numbers could benefit Barn owls. Meek et al. Discussion indicated that Barn owls are able to easily switch to Microtus agrestis in particular, A. sylvaticus and Sorex A. sylvaticus, if present, when M. agrestis numbers are low. species were the most common small mammals captured In March, there is a clear increase in the proportion of at Folly Farm. This was expected, as Barn owl corridors males within the population. It is recognized that such seaso- 9, 41, 42 should provide suitable habitat for these species. nal changes occur, with males being more numerous over 51 52 The single C. glareolus capture was unexpected as they are winter and spring. Krebs and Davies identify two main rarely found in grassland, although studies have shown mechanisms contributing to this seasonal variation, differen- those resident in nearby hedgerows or woodland to enter tial recruitment and differential survival rates. Of these two, 44, 45 field margins. The Micromys minutus capture in the it is likely that differential recruitment is the dominant of the un-grazed corridor was also unusual as field margins are two mechanisms as the increases coincide with the onset of not considered suitable habitat. Bence et al. did suggest the breeding season. It is also possible that the number of that, given suitable growth, they could support males has not increased; those present may simply be more M. minutus. The most unusual capture was a single active. During the breeding season, males are often noted N. fodiens in the grazed corridor, a species generally found as having increased activity and larger ranges. along watercourses. In 2004, a survey of the adjacent The overall trend of increase in mean weight will be watercourses recorded no indicators of their presence largely reflective of changes in age structure, as the pro- on the farm, making the find even more unusual. Tew portion of adults increases so does mean weight. However, has recorded them in hedgerows and suggested that linear the only long-term recapture (M. agrestis) did show an ......................................................................................................................................................................................................................................... 160 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... increase of 2 g between February and March, indicating that deeper thatch layer in the un-grazed corridor may provide individual weights had also increased. a wider range of food resources, microhabitats and increased M. agrestis were found to be the most abundant species in protection from avian hunters explaining the even distri- both corridors, and this mirrors the findings of Wilkinson bution. The distribution pattern in the grazed corridor is in rough grassland at nearby Weston Moor Nature reserve. more patchy. The peak in capture rates in some traps could Other studies by Lambin et al. also show relatively high be explained by their location under the Oak tree, abundances within this habitat. The proportion of Q. robur. Interestingly, in the un-grazed corridor, the Lime M. agrestis within the total population was significantly trees, Tilia cordata, appeared to have no effect on capture greater in the un-grazed corridor; this was unexpected. The success. There was also an absence of captures in traps Barn Owl Trust guidance suggests that low-intensity near a footpath in the grazed corridor. The presence of the grazing/cutting should be carried out at least every 2 years footpath could in part explain this; however, there was to maximize vole numbers. Even after 7 years with no also found to be a healthy population of weasels, Mustella cutting/grazing routine, there is still a higher population of nivalis which could have increased local predation pressure M. agrestis in this corridor. The grazed corridor, undergoing on the small mammal population. a grazing routine every 2/3 years, should have higher vole In the un-grazed corridor, as vegetation height increases so numbers; in fact, the opposite is the case. do the number of captures. This pattern might be expected. The grazed corridor supports a higher species richness and Hamback et al. showed that, especially over the winter diversity than the un-grazed one. Higher numbers of more months, increased vegetation height led to increased vole generalist species, in this case A. sylvaticus, were found in activity. They indicated that when selecting over-wintering the grazed corridor. These findings are similar to those of habitat vegetation height was important to reduce predation 14 53 Trump and Tattersal et al who were looking at field and freezing risk. Looking at the individual species, it is clear boundaries and set aside, respectively. The generalist nature that this trend is confined to M. agrestis, the other species of A. sylvaticus make it able to adapt to change, i.e. appear to avoid long vegetation. There are no perceivable grazing, and they are less dependent on ground cover than trends in the grazed corridor. voles. Another factor may be that they are able to travel The pellet data from Folly Farm are broadly similar to that considerable distances, so may not inhabit the area in which published from the Mammal Society’s National Survey. The they are trapped. Increased species richness and diversity Folly Farm pellets show M. agrestis to be the most popular in this corridor could be reflecting its diverse surroundings prey item, then a preference for A. sylvaticus over of woodland, streams, farm buildings and veteran trees. S. araneus, the opposite was the case in the national The use of transects and the low number of recaptures survey. This may be reflective of the relatively high pro- make the use of population estimates flawed in this study. portion of A. sylvaticus seen in the trapping data, particu- Although the Peterson-Lincoln method of population esti- larly in the grazed corridor. No M. minutus or N. fodiens mation will prove unreliable, it is interesting to speculate as were found in the pellet samples, indicating that the low this allows the population to be compared with documented capture numbers in the Longworth traps accurately depict densities. At their peak in February, M. agrestis densities the population. Optimal Foraging Theory would suggest reach 128/ha in the un-grazed corridor and 48/ha in the that a low number of less preferential species, such as grazed corridor. Although slightly on the low side, they do S. minutus, indicates that the proportion of more profitable 9, 17 fall within most expected densities for grasslands, and species is high. mixed farmland. There is a similar species ranking found between the trap- The small mammals captured in this study appear to be ping results and the pellet data. Similarly, Bonvicino and fairly sedentary. Despite recorded home ranges of up to Bezerra showed that regurgitated pellets from Barn owls 2 18 1000 m , the mean distance travelled by M. agrestis was were a good source of information for assessing species rich- only 18.8 m and the maximum being 65 m. Even more ness. They found strong correlations, when assessing total surprising was the 6.7 m mean distance covered by population, despite stressing the potential for bias in A. sylvaticus whose home ranges on farmland often exceed numbers due to the preferential feeding habits of such 1 ha (10 000 m ). This study did not extend to the breeding species. season, and Corbett and Harris noted that home ranges There is little information on the accessibility of prey items can remain small over winter until the onset of sexual matur- to Barn owls. There is an expected preference for M. agrestis ity in spring. Similar changes in movement patterns were in this study; a species that is more abundant and dominant recorded by Trump with few A. sylvaticus moving more in the un-grazed corridor. What is unknown is to what than 10 m in the winter months, but recorded movements degree the high population, in the deep thatch layer, of 30–50 m by May. is offset by their increased inaccessibility. Barn owl mor- The distribution patterns of small mammals in both corri- phology does suggest that their legs and talons should 1 56 dors are different. The higher floral species richness and enable them to penetrate dense grassland .Lack stated ......................................................................................................................................................................................................................................... 161 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... 9. Harris S, Morris P, Wray S et al. (1995) A Review of British Mammals: that sward heights of at least 25 cm were ideal for Barn owl Population Estimates and Conservation Status of British Mammals Other foraging. Meek et al. suggested that there is an optimal than Cetaceans. Peterborough: JNCC. corridor width if Barn owls are to hunt at maximum effi- 10. Tew TE (1994) Farmland hedgerows: habitat, corridors or irrelevant? A small ciency. This width is one where the greatest number of mammal’s perspective. In Watt TA, Buckley GP eds, Hedgerow Management prey can be intercepted on a single fly over. Flying most and Nature Conservation. Wye: Wye College Press, pp 80–94. commonly at a height of 3 m when hunting and having an 11. Gorman ML, Rogers LM (1995) The population dynamics of small mammals effective hearing range of 508 indicates that this margin living in set-aside and surrounding semi-natural and crop land. J Zool 236: width would need to exceed 7 m. Although not as wide as 451–464. the grazed corridor, the un-grazed corridor is up to 55 m 12. Kotzageorgis GC, Mason CF (1997) Small mammal populations in relation to hedgerow structure in an arable landscape. J Zool 242: 425–434. wide and no ,20 m wide at any point. 13. Mallorie HC, Flowerdew JR (1994) Woodland small mammal population This study is based on the premise that the strongest influ- ecology in Britain: a preliminary review of The Mammal Society survey of ence on community structure within these two corridors is wood mice Apodemus sylvaticus and bank voles Clethrionomys glareolus, habitat quality, the main influence on this being past and 1982–87. Mam Rev 24: 1–15. present management. Without a long-term monitoring 14. Trump DPC (2003) The Barn Owl Tyto alba, Field Margins and Small Mammals. programme and no specific historical data, the accuracy of PhD Thesis. Edinburgh: Napier University. this statement, and the degree to which un-measured external 15. Wilkinson SAJ (2002) Investigation of Invertebrate Diversity Associated with influences play a role on populations, has to remain Field Vole Habitat. PhD Thesis. Bristol: University of Bristol. un-quantifiable. Although M. agrestis was the most abundant 16. Corbett GB, Harris S eds (1991) The Handbook of British Mammals, 2nd ed. Oxford: Blackwell Scientific Publishing. species in both corridors, they were particularly dominant in 17. Lambin X, Petty SJ, Mackinnon JL (2000) Cyclical dynamics in field vole the un-grazed corridor. The total population is low, despite populations and generalist predation. J Anim Ecol 69: 106–118. the peak falling within the expected range. The un-grazed cor- 18. Macdonald DW, Tattersall F (2001) Britain’s Mammals: The Challenge for ridor has not received any management over the past 8 years Conservation. London: Peoples Trust for Conservation. and still supports a relatively healthy M. agrestis-dominated 19. Baker PJ, Ansell RJ, Dodds PAA et al. (2003) Factors affecting the distribution community of small mammals. The data presented in this of small mammals in an urban area. Mam Rev 33: 95–100. paper suggest that the current recommended rotation of 2/3 20. Churchfield S (1990) The Natural History of Shrews. London: Christopher years is unnecessary. The apparent M. agrestis densities in Helm, A & C Black. the grazed corridor are surprisingly low. The trapping 21. Tapper S (1979) The effect of fluctuating vole numbers on a population of results, NVC classifications and sward condition suggest weasels on farmland. J Anim Ecol 48: 603–617. that this corridor has been over-grazed. The use of a closely 22. Hanski I, Hansson L, Henttonen H (1991) Specialist predators, generalist pre- monitored system of cattle grazing would be recommended. dators, and the microtine rodent cycle. J Anim Ecol 60: 353–367. If not financially or practically possible, the corridors could be 23. Gause GF (1964) The Struggle for Existence. New York: Hafner. mechanically topped to a height of no ,20 cm. Although a 24. Evans FC, Murdoch WM, Peterson CH (1972) Diversity and pattern in plants and insects. Ecology 53: 819–829. valuable data set, the potential strength of this study lies in 25. MacArthur RH, MacArthur JW (1961) On bird species diversity. Ecology 42: it becoming baseline data for future monitoring. It is suggested 594–598. that a long-term monitoring programme be set up to give a 26. Rozenweig ML, Winakur J (1969) Population ecology of desert rodent more accurate picture of populations present within all the communities: habitats and environmental complexity. Ecology 59: corridors at Folly Farm. 558–572. 27. Dueser RD, Shugart HH (1978) Microhabitats in a forest-floor small mammal fauna. Ecology 59: 89–98. References 28. Bowers MA (1990) Exploitation of seed aggregations by Merriam’s kangaroo 1. Shawyer C (1998) The Barn Owl. Chelmsford: Arlequin Press. rat; harvesting rates and predatory risk. Ecology 71: 2334–2344. 2. Buckley J, Goldsmith JG (1975) The prey of the barn owl (Tyto alba) in East 29. Kotler BP, Brown JS, Mitchell WA (1993) Environmental factors affecting Norfolk. Mam Rev 5: 13–16. patch use in two species of gerbilline rodents. J Mamm 74: 614–620. 3. Love RA, Webon C, Glue DE et al. (2000) Changes in the food of British Barn 30. Hornfeldt B. Delayed density dependence as a determinant of vole cycles. Owls (Tyto alba) between 1974 and 1997. Mam Rev 30: 107–129. Ecology (1994) 75: 791–806. 4. Toms MP, Crick HQP, Shawyer CR (2001) The status of breeding barn owls 31. Hamback PA, Oksanen T, Schneider M (1998) Winter herbivory by voles Tyto alba in the United Kingdom 1995–1997. Bird Study 48: 23–37. during a population peak: the relative importance of local factors and land- scape pattern. J Anim Ecol 67: 544–553. 5. Bunn DS, Warburton AB, Wilson RDS (1982) The Barn Owl. London: Poyser. 32. Anon. (1999) Non-SSSI Grassland and Barn Owl Corridors. Folly Farm Nature 6. Webster JA (1973) Seasonal variation in mammal contents of barn owl cast- Reserve Management Plan 1999–2004. Bristol: Avon Wildlife Trust. ings. Bird Study 20: 185–196. 33. Rodwell JS, ed (1992) British Plant Communities, Vol. 3. Grasslands and 7. Glue DE (1967) Prey taken by the barn owl in England and Wales. Bird Study Montane Communities. Cambridge: Cambridge University Press. 14: 169–183. 34. Bunce RGH, Barr CJ, Gillespie MK et al. (1999) ECOFACT 1: Vegetation of the 8. Love RA (2005) Mammal Society National Owl Pellet Survey [online]. http:// www.abdn.ac.uk/mammal/owl_survey/index1.htm (accessed 2 February British Countryside—the Countryside Vegetation System. Rotherham: DETR 2007). Publications. ......................................................................................................................................................................................................................................... 162 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... 35. Chitty D, Kempston DA (1949) Prebaiting small mammal traps and a new 47. Carter PKL, Churchfield S (2006) Distribution and habitat occurrence of water shrews in Great Britain. Mamm Soc, Research Report Number 7. design in live trap. Ecology 6: 36–53. 48. Tanton MT (1965) Problems of live trapping and population estimation for 36. Pearson DE, Ruggiero LF (2003) Transect versus grid trapping arrangements the wood mouse. J Anim Ecol 34: 1–22. for sampling small mammal communities. Wildlife Soc Bull 31: 454–459. 49. Tattersall FH, MacDonald DW (2003) Woodmice in the arable system. In 37. Gurnell J, Flowerdew JR (2006) Live Trapping Small Mammals: A Practical Tattersal FH, Manley W eds, Conservation and Conflict: Mammals and Guide, 4th ed. Reading: The Mammal Society. Farming in Britain. London: The Linnean Society, pp 82–96. 38. Kent M, Coker P (1992) Vegetation Description and Analysis—A Practical 50. Meek WR, Burman PJ, Nowakowski M et al. (2002) Barn owl release in Approach. New York: John Wiley. lowland Southern Britain—a twenty year study. Biol Conserv 109: 271–282. 39. Yalden DW, Morris PA (1990) The Analysis of Owl Pellets. Occasional publi- 51. Bryja J, Heroldova M, Janova E et al. (2005) Common vole (Microtus agrestis) cation of the Mammal Society No. 13. Bristol: The Mammal Society. population sex ratio: biomass and process variation. Can J Zool 83: 40. Simpson EH (1949) Measurement of diversity. Nature 163: 688. 1391–1399. 41. Taylor IR (1994) Barn Owls: Predator–Prey Relationships and Conservation. 52. Krebs CJ, Davies NB (1993) An Introduction to Behavioral Ecology, 3rd ed. Cambridge: Cambridge University Press. Oxford: Blackwell Scientific Publications. 42. Barn Owl Trust (2006) Habitat Management—Leaflet No. 1. Ashburton: Barn 53. Tattersall FH, MacDonald DW, Hart BJ et al. (2002) Is habitat linearity Owl Trust. important for small mammal communities on farmland? J Appl Ecol 39: 43. Fitzgibbon CD (1997) Small mammals in farm woodlands: the effects of 643–652. habitat isolation and surrounding land-use patterns. J Appl Ecol 34: 530–539. 54. Sibbald S, Carter P, Poulton S (2006) The Mammal Society Research Report 44. Gurnell J (1985) Woodland rodent communities. Symp Zool Soc Lond 55: No. 6. Proposal for a National Monitoring Scheme for Small Mammals in the 377–411. UK and the Republic of Eire. London: The Mammal Society. 45. Alibhai SK, Gipps JHW (1991) The bank vole. In Harris S, Corbett GB eds, The 55. Bonvicino CR, Bezerra AMR (2003) Use of regurgitated pellets of Barn owl Handbook of British Mammals. Oxford: Blackwell Scientific Publications. (Tyto alba) for inventorying small mammals in the Cerrado of Central Brazil. Stud Neotrop Fauna Environ 38: 1–5. 46. Bence S, Griffiths M, Stander K (2003) Habitat characteristics of Harvest mouse nests on arable farmland. Agric Ecosyst Environ 99: 179–186. 56. Lack P (1992) Birds on Lowland Farms. London: HMSO. Author Biography Alex Keene is currently studying Environmental Biology at Bath Spa University. On completion of the course he will be seeking work involving both the practical and educational side of environmental/conservation management. Alex has been working as a freelance photographer for the past ten years and is keen to integrate this into his new career at some point in the future. ........................................................................................................................................................................................................................................ Submitted on 30 September 2008; accepted on 12 February 2009; advance access publication 17 April 2009 ......................................................................................................................................................................................................................................... http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bioscience Horizons Oxford University Press

Study of small mammal populations within two Barn owl corridors at Folly Farm

Keene, Alex
Bioscience Horizons , Volume 2 (2) – Jun 17, 2009
Download PDF
9 pages

Loading next page...
 
/lp/oxford-university-press/study-of-small-mammal-populations-within-two-barn-owl-corridors-at-h7VFaHALLx

References (64)

Publisher
Oxford University Press
Copyright
© Published by Oxford University Press.
Subject
Research articles
eISSN
1754-7431
DOI
10.1093/biohorizons/hzp018
Publisher site
See Article on Publisher Site

Abstract

Volume 2 † Number 2 † June 2009 10.1093/biohorizons/hzp018 ......................................................................................................................................................................................................................................... Research article Study of small mammal populations within two Barn owl corridors at Folly Farm Alex Keene* School of Science and the Environment, Bath Spa University, Newton St Loe, Bath BA2 9BN, UK. * Corresponding author: 62 Ashgrove, Peasedown St John, Bath BA2 8EF, UK. Tel: þ44 1761 420094. Email: alex.thewalkingcamera@virgin.net Supervisor: Ian Todd, School of Science and the Environment, Bath Spa University, Newton St Loe, Bath BA2 9BN, UK. ........................................................................................................................................................................................................................................ This study examines small mammal populations present within Barn owl corridors on Folly Farm, an Avon Wildlife Trust Reserve located near the village of Bishop Sutton in Bath and North East Somerset. Two corridors were chosen, the primary difference between the two sites being only one has undergone management (grazing). The focus of this study was the Microtus agrestis (Short-tailed field vole) population, the most frequently taken prey item by Barn owls. Apodemus sylvaticus (Wood mice) and Sorex araneus (Common shrew) populations are also discussed as they are frequently taken. Using Longworth live traps, 600 trap-nights data were collected from three sessions in November 2006, February and March 2007. Although M. agrestis was the most abundant species in both corridors, they were more prevalent in the un-grazed corridor (comprising 19 of the 31 individuals). In the corridor that had undergone manage- ment, fewer M. agrestis were captured (eight), although a higher species diversity and richness was recorded. Unusually for a grassland habitat, there were nearly as many A. sylvaticus as there were M. agrestis (seven compared with eight) in the grazed corridor. Some small mammal species not usually found in grassland habitats were captured; explanations for these seemingly anomalous results are discussed. Differences in population characteristics between the two corridors are discussed including: sex ratio, weights, seasonal variation and age structure. Pellet analysis from the nearby pair of Barn owls showed that they were preferentially hunting M. agrestis; the pellet data largely mirrored findings of the trapping data. Key words: Short-tailed field vole, small mammal, Barn owl corridor, pellet, Longworth trap. ........................................................................................................................................................................................................................................ Barn owls may consume as many as 5000 small mammals, Introduction with predominance of Microtus agrestis. Changes in small mammal abundance and species composition can have Barn Owls serious effects on the breeding success of Barn owls. Barn The Barn owl, Tyto alba, has suffered in recent decades as a 1 owls cast an average two pellets a day and the structure of result of intensive agriculture, although it is suggested that 2 small mammal communities can be deduced from the com- climatic deterioration may also be a significant cause. parative frequency of different species in the pellet sample. Between 1932 and 1998, Barn owl populations declined in 3 The Mammal Society National Owl Pellet Survey showed Britain from 12 000 pairs to ,3000. A census carried out that M. agrestis make up the bulk of prey. between 1995 and 1997 indicated that the decline has halted if not reversed in some areas. Unfortunately, their Small Mammals plight was once again highlighted in 2006, with national breeding failure rates as high as 70% (Andre Fournier, Changes in farming practice and the decline of other suitable personal communication). small mammal habitats, as discussed by Harris et al., have Barn owls are opportunists and will hunt over open led to a decrease in many small mammal species, particularly 5 3 countryside. Their ideal habitat is rough grassland with a M. agrestis. This has led to a variety of studies focused on thick sward and deep litter layer supporting high densities both seasonal and annual cycles within populations at a 10– 12 13 of small mammals. In an average year, a breeding pair of single location and nationally. Previous small ......................................................................................................................................................................................................................................... 2009 The Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 155 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... mammal surveys have been conducted locally looking specifi- primary aim of protecting and enhancing its wildlife 14, 15 32 cally at populations present in Barn owl feeding areas. value. Over 28 ha of the farm are designated as a biological This study looked at the overall composition of small Site of Special Scientific Interest (SSSI). mammal fauna within two corridors, three species were of Six Barn owl corridors were instated in 1999/2000. The particular interest: M. agrestis, Apodemus sylvaticus and pair of owls present on Folly Farm failed to breed in 2006 Sorex araneus. Found in rough grassland throughout the for the first time since monitoring began in 1992 (Andre UK mainland, the population dynamics of M. agrestis has Fournier, personal communication). There is no strict been studied in some detail. Densities can range from cutting/grazing rotation in place within the corridors; 9 17 1/ha in mixed farmland up to 250/ha in rough grassland, those with stock fencing are grazed, others are mown, with home ranges as large as 1.4 ha. A. sylvaticus is also some are even cut by hand. Management is accessed highly mobile with similar home ranges, 0.26–1.77 ha in annually and is dependent on stock and labour availability. farmland. It is widely distributed in woodland, field and Until now, no small mammal survey work has been con- hedge habitats, but also occurs in sand dunes, gardens and ducted in the corridors. urban areas. Densities of 0.5–17.5/ha are typical on Time and resources made it impossible to survey all six farmland. S. araneus are also found in a wide variety of corridors and therefore two were chosen, the primary differ- habitats, including grassland, arable and woodland. They ence between them being only one has undergone manage- occupy relatively small home ranges, often less than ment. Site 1, the ‘un-grazed’ corridor, has received no 0.1 ha with densities of 5–69/ha. management since being fenced off at the end of 1999, Populations of some small mammals fluctuate. The causes other than having ragwort pulled and thistles topped when of cyclic population changes, and the role that predators infestations occur. The result of this is a thick, deep sward play, have been largely unresolved, despite high volumes of and relatively high species diversity. The best NVC match literature. Specialist hunters, such as Barn owls, are par- is MG1b, an Arrhenathetum community, typical of ticularly reliant on small mammal populations, especially un-grazed grassland on neglected pastures and meadows. M. agrestis. Populations of these predators increase when There are two Lime trees, Tilia cordata, within the corridor. voles are abundant and contract when numbers are low. Site 2, the ‘grazed’ corridor, has been grazed twice, first in Numerous studies have tried to determine what influences August 2000 and again in 2004. This site is botanically small mammal community composition. Gause demon- less interesting, the sward less dense and vegetation height strated both empirically and theoretically that coexistence substantially lower. NVC results produced a classification of competing species is only possible if each have different, of MG10/MG9. These plant communities occur mainly in specialized abilities. Several authors have attempted to fields and associated boundaries, and are often associated find correlations between plant and animal species richness with poorly drained permanent pasture. There are two and diversity; some have shown correlations, others veteran Oaks, Quercus robur, present; the most southerly 25 26 have not. Rozenweig and Winakur found that spatial contains a Barn owl box. variations in the density of some species are responses to Trapping spatial characteristics of their environment. Dueser and Shugart both found that simple measurements of density Trapping was undertaken using Longworth live-capture and height of vegetation within different strata provided small mammal traps. Both corridors were trapped definite correlations with the densities of species present. simultaneously during 3 and 4 days, sessions in the A structurally diverse habitat is likely to contain a larger mid-November 2006, early February and late March 2007. variety of food sources and nesting sites, providing niches The traps were prepared with hay bedding, mixed birdseed 28, 29 for predatory species as well as protection for prey. If and blowfly puparia for bait. Due to the linear structure of mobility allows then mammals have been shown to distribute the habitats, and their effectiveness with assessing commu- 30 36 themselves between habitats in relation to habitat quality. nity structure, traps were laid in transects. Twenty-five The most important factors being food availability, food trapping points were marked at 5 m intervals within each quality, predation risk and competition. corridor; the Longworth traps were then located + m from the mark. On Site 2, the transect was located at the southern end of the corridor to avoid a regularly used foot- Methods path. Traps were deliberately placed in positions that indi- cated the presence of, or would be favoured by, small Study Area mammals. All field signs were noted and located by relevant Avon Wildlife Trust has owned Folly Farm since 1987. It is trap number. located near Bishop Sutton in the Chew Valley (Grid refer- There were 200 trap-nights in each session, a total of 600 ence ST607604). Much of the farm has escaped modern, trap-nights for the whole programme. The traps were laid intensive practices and is traditionally managed with the and set on the afternoon of the first day and checked in the ......................................................................................................................................................................................................................................... 156 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... Table 1. Total number of captures and individuals caught during the morning and evening of the next 3 days, then checked and survey period, where individuals are first capture of an animal (i.e. no removed on the morning of the fourth day. visible fur clip) All small mammals caught, except shrews, were identified to species, weighed to the nearest gram using a 50  0.2 g Species Number of captures Individuals ................................................................................................................ Salter spring balance, aged, sexed and breeding condition M. agrestis 34 27 noted. All species captured were assigned to one of the fol- A. sylvaticus 14 11 lowing three age classes: adult, subadult or juvenile, accord- S. araneus 88 ing to Gurnell and Flowerdew. New captures were given a S. minutus 33 unique fur clip. All mammals were released at the point of capture. All shrews caught were identified to species, M. minutus 11 weighed and released immediately. Environmental data C. glareolus 11 were recorded every time the traps were checked, giving an N. fodiens 11 account of weather during the previous 12 h. Total 62 52 Vegetation of both corridors was largely homogenous throughout; therefore, detailed floral surveys around each case of M. agrestis and February in the case of A. sylvaticus trap were deemed unnecessary. Instead physiognomic and Sorex sp. (Fig. 1). vegetation classification was used; a popular method of habitat classification in researching habitat selection and utilization. Vegetation height was measured above the Sex ratio and reproductive condition entrance to each trap and recorded as short (,10 cm), The sex of shrews was not assessed. The overall sex ratio of medium (10–50 cm) or long (.50 cm). all other species caught in November and February was close to 1:1; however, by March, the ratio was skewed in favour of males (1.8:1) (Table 2). Between November and February, all Owl Pellets captures were non-breeding (condition of shrews was not Owl pellets were collected from below the Barn owl roost on assessed), all males having abdominal/small testes and all Folly Farm in May and June 2006. The roost was located in females imperforate. In March, all male captures were still the SSSI grassland 500 m from the survey sites, making non-breeding except for two A. sylvaticus caught in the both corridors well within hunting range of the owls. The un-grazed corridor, recorded as having testes large and pellets were soaked in warm soapy water, teased apart and scrotal, and all females were still non-breeding. all mammals remain cleaned and laid aside. Identification was carried out on skulls and lower jaw bones using a Age structure and weights 10 hand lens. Species were keyed out according to Juvenile mice were only captured in November, when they Yalden and Morris. made up 14% of the population (Fig. 2). At this time, the rest of the population was made up of a nearly even mix of adults and subadults. By February, the majority of the popu- Results lation is adult, 73%. The adult proportion has reached 100% Trapping A total of 62 small mammals were captured over the 600 trap-nights (summarized in Table 1). One Sorex minutus was found dead and one A. sylvaticus escaped before being processed. The vast majority of captures were recorded at dawn and therefore captured overnight. The two captures during the day were both S. araneus, one during the November session and the other in March. Both were adults and recorded in the un-grazed corridor. During all three sessions, there were no captures on the first day and highest number of captures were made on the final day. Small mammal catches were relatively evenly distributed across the un-grazed corridor, whereas the grazed corridor had a more patchy distribution pattern, with a clear lack of captures in the first seven traps. There are distinctly different Figure 1. Seasonal change in population structure of all species captured. patterns of seasonal population change in the M. agrestis and Total number of individuals captured in November ( ), in February (A) and A. sylvaticus populations. Numbers peak in November in the in March (B). ......................................................................................................................................................................................................................................... 157 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... Table 2. Number of male and female captures, of all species, in both corridors Trapping session M. agrestis A. sylvaticus Total (all species) Male Female Male Female Male Female Ratio ........................................................................................................................................................................................................................................ November 6 8 3 2 10 11 1.0:1.1 February 4 6 3 2 8 9 1.0:1.1 March 6 4 3 0 9 5 1.8:1.0 Table 3. Species composition, richness and diversity (Simpson’s index) in the grazed and un-grazed corridors Species Un-grazed corridor Grazed corridor ................................................................................................................ M. agrestis 19 8 A. sylvaticus 47 S. araneus 62 S. minutus 12 M. minutus 1— C. glareolus —1 N. fodiens —1 Total number of individuals 31 21 Figure 2. Age classes of all mammals caught in both corridors. Whole bar Species richness 5 6 is total population, is the number of adults, A the subadults and B the Species diversity 0.6 0.8 juveniles. was significantly higher in that corridor (X ¼ 4.48, df ¼ 1, p  0.05). The species diversity was higher in the grazed corridor (Simpson’s index of diversity, 1 2 D ¼ 0.8) than in the un-grazed (0.6). Five different species were recorded in the un-grazed corridor, the majority were M. agrestis (19) followed by S. araneus (6) then A. sylvaticus (4). Six species were recorded in the grazed corridor; the highest occurrence was still that of M. agrestis (eight) then A. sylvaticus (seven). Micromys minutus was only recorded in the un-grazed corridor, and Clethrionomys glareolus and Neomys fodiens were only recorded in the grazed corridor. Figure 3. Mean weight of individuals in all three sessions. The mean weight of M. agrestis ( ), A. sylvaticus (B), S. araneus (A) and S. minutus ( ). Movement of individuals by March. In Fig. 3, mean weights of S. minutus remained Ten small mammals were recaptured, eight of these in the unchanged between February and March. A. sylvaticus and un-grazed corridor. Seven were M. agrestis and three S. araneus weights remained unchanged between November A. sylvaticus. The mean distance travelled was 17.5 m in and February but increased in March. The mean weight of the un-grazed corridor, higher than that recorded in the M. agrestis fell, although not significantly, from November grazed corridor (5 m). The mean distance travelled by to February (t ¼ 1.54, df ¼ 20, p , 0.138). The mean M. agrestis was higher, 18.6 m, than A. sylvaticus, 6.7 m. weight of M. agrestis rose significantly between February These figures will however be skewed by the movement of and March (t ¼ 23.56, df ¼ 13, p , 0.003). an individual M. agrestis in the un-grazed corridor, travelling 65 m in a 24 h period. Only one long-term movement was Species recorded, between capture on the 9th February and recapture The total number of small mammals present (Table 3) on the 25th March the M. agrestis moved only 15 m. There was not significantly higher in the un-grazed corridor was no recorded movement between the two corridors, a (X ¼ 2.88, df ¼ 1, p  1). The proportion of M. agrestis minimum distance of 150 m. ......................................................................................................................................................................................................................................... 158 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... Effect of vegetation height on trapping success The proportion of mammals caught in the short and long grass is different between corridors (Fig. 4). More mammals were captured in the medium length vegetation (10–50 cm) in both corridors but probably this is an artefact of the greater availability of traps within this length. In fact, in the un-grazed corridor, a preference for long vegetation and an avoidance of short vegetation was shown. Although the differences were more subtle, the opposite was recorded in the grazed corridor, with more mammals preferring short vegetation. Figure 5 shows the un-grazed corridor to have the highest capture rates of M. agrestis in traps placed Figure 4. Vegetation length in which mammals were captured on both sites. L, long (.50 cm); M, medium (10–50 cm); S, short (,10 cm). under long vegetation, with 13% of traps making a Whole bar represents the number of traps placed in different vegetation capture. A. sylvaticus and Sorex sp. both showed a slight heights, area shaded ( ) is the total number of captures within that preference for the medium length vegetation. In the grazed length and the figure in white is the percentage of captures within each corridor, these trends are reversed for all three species. length. Figure 5. Vegetation length in which different species were captured on both sites. Bar represents the percentage of captures made within available traps at each different vegetation height, in both corridors. ......................................................................................................................................................................................................................................... 159 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... Table 4. Percentage of species trapped in both corridors, in owl features were vital corridors between preferred habitats. pellets from Folly Farm and in pellets analysed in The Mammal The presence of streams running either end of the corridor Society National Owl Pellet Survey suggests a transient capture. The majority of small mammals were captured at dawn. Species Trapping Owl pellet data Mammal Society data Owl pellets There were two exceptions, both S. araneus, caught in the Mean% May June Mean% Mean% un-grazed corridor. This is little surprise as A. slyvaticus ................................................................................................................ and M. agrestis are mainly nocturnal, whereas Sorex M. agrestis 53 26 28 54 43.4 species can be active night and day throughout the year. A. sylvaticus 24 17 4 21 16.6 During all three trapping sessions, no captures were made S. araneus 12 8 2 10 19.8 on the first day. There is no obvious reason for this other S. minutus 6 7 — 7 8.9 than the small mammals being ‘trap-shy’. This reluctance C. glareolus 4 4 1 5 5.5 to enter ‘new objects’, particularly by Microtus species, is M. minutus 3 — — 0 2.2 often found. Three nights of trapping is considered suffi- N. fodiens 4— — 0 ,1 cient to allow trap-shy mammals to accept the presence M. domesticus 02 1 3 ,1 of traps and therefore give an accurate picture of the population. In arable landscapes, seasonal variations in A. sylvaticus densities have been shown to vary dramatically. The For example, the long vegetation records the lowest capture 9 results in this study agree with Harris et al.; although rates for M. agrestis. there is seasonal variation, with a winter peak, these cycles are not as marked in marginal habitats such as grassland. Owl Pellets Similar trends were seen for the Sorex species, although Table 4 shows the results of the pellet analysis alongside a these figures should be examined with caution, the summary of the trapping data in Table 1 and the results February peak could purely be a reflection of the increased from the Mammal Society’s National Survey (1993– number of traps without ‘shrew-holes’ (an unavoidable con- 2005). The Barn owl pellets recovered in May contained sequence of having to borrow traps at short notice). an estimated 114 individual mammals and pellets collected M. agrestis show an opposite seasonal change in population in June contained 55; a total of 169. One hundred of these with numbers dropping off in February and increasing again were identifiable to species. M. agrestis formed the majority by March. As the March increase could not be down to of prey species followed by A. sylvaticus and then Sorex spp. breeding, it must be a result of immigration from an adjacent The rank of species in all three data sets is the same with two habitat or possibly an inaccurate assessment of the popu- exceptions: A. sylvaticus and Mus domesticus are ranked lation in February, with a percentage of M. agrestis alluding lower in the Mammal Society results than in the data sets capture. Low numbers meant no trends could be seen for the from Folly Farm. remaining species. If these seasonal fluctuations are accurate, the peak in A. sylvaticus numbers coinciding with a trough in M. agrestis numbers could benefit Barn owls. Meek et al. Discussion indicated that Barn owls are able to easily switch to Microtus agrestis in particular, A. sylvaticus and Sorex A. sylvaticus, if present, when M. agrestis numbers are low. species were the most common small mammals captured In March, there is a clear increase in the proportion of at Folly Farm. This was expected, as Barn owl corridors males within the population. It is recognized that such seaso- 9, 41, 42 should provide suitable habitat for these species. nal changes occur, with males being more numerous over 51 52 The single C. glareolus capture was unexpected as they are winter and spring. Krebs and Davies identify two main rarely found in grassland, although studies have shown mechanisms contributing to this seasonal variation, differen- those resident in nearby hedgerows or woodland to enter tial recruitment and differential survival rates. Of these two, 44, 45 field margins. The Micromys minutus capture in the it is likely that differential recruitment is the dominant of the un-grazed corridor was also unusual as field margins are two mechanisms as the increases coincide with the onset of not considered suitable habitat. Bence et al. did suggest the breeding season. It is also possible that the number of that, given suitable growth, they could support males has not increased; those present may simply be more M. minutus. The most unusual capture was a single active. During the breeding season, males are often noted N. fodiens in the grazed corridor, a species generally found as having increased activity and larger ranges. along watercourses. In 2004, a survey of the adjacent The overall trend of increase in mean weight will be watercourses recorded no indicators of their presence largely reflective of changes in age structure, as the pro- on the farm, making the find even more unusual. Tew portion of adults increases so does mean weight. However, has recorded them in hedgerows and suggested that linear the only long-term recapture (M. agrestis) did show an ......................................................................................................................................................................................................................................... 160 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... increase of 2 g between February and March, indicating that deeper thatch layer in the un-grazed corridor may provide individual weights had also increased. a wider range of food resources, microhabitats and increased M. agrestis were found to be the most abundant species in protection from avian hunters explaining the even distri- both corridors, and this mirrors the findings of Wilkinson bution. The distribution pattern in the grazed corridor is in rough grassland at nearby Weston Moor Nature reserve. more patchy. The peak in capture rates in some traps could Other studies by Lambin et al. also show relatively high be explained by their location under the Oak tree, abundances within this habitat. The proportion of Q. robur. Interestingly, in the un-grazed corridor, the Lime M. agrestis within the total population was significantly trees, Tilia cordata, appeared to have no effect on capture greater in the un-grazed corridor; this was unexpected. The success. There was also an absence of captures in traps Barn Owl Trust guidance suggests that low-intensity near a footpath in the grazed corridor. The presence of the grazing/cutting should be carried out at least every 2 years footpath could in part explain this; however, there was to maximize vole numbers. Even after 7 years with no also found to be a healthy population of weasels, Mustella cutting/grazing routine, there is still a higher population of nivalis which could have increased local predation pressure M. agrestis in this corridor. The grazed corridor, undergoing on the small mammal population. a grazing routine every 2/3 years, should have higher vole In the un-grazed corridor, as vegetation height increases so numbers; in fact, the opposite is the case. do the number of captures. This pattern might be expected. The grazed corridor supports a higher species richness and Hamback et al. showed that, especially over the winter diversity than the un-grazed one. Higher numbers of more months, increased vegetation height led to increased vole generalist species, in this case A. sylvaticus, were found in activity. They indicated that when selecting over-wintering the grazed corridor. These findings are similar to those of habitat vegetation height was important to reduce predation 14 53 Trump and Tattersal et al who were looking at field and freezing risk. Looking at the individual species, it is clear boundaries and set aside, respectively. The generalist nature that this trend is confined to M. agrestis, the other species of A. sylvaticus make it able to adapt to change, i.e. appear to avoid long vegetation. There are no perceivable grazing, and they are less dependent on ground cover than trends in the grazed corridor. voles. Another factor may be that they are able to travel The pellet data from Folly Farm are broadly similar to that considerable distances, so may not inhabit the area in which published from the Mammal Society’s National Survey. The they are trapped. Increased species richness and diversity Folly Farm pellets show M. agrestis to be the most popular in this corridor could be reflecting its diverse surroundings prey item, then a preference for A. sylvaticus over of woodland, streams, farm buildings and veteran trees. S. araneus, the opposite was the case in the national The use of transects and the low number of recaptures survey. This may be reflective of the relatively high pro- make the use of population estimates flawed in this study. portion of A. sylvaticus seen in the trapping data, particu- Although the Peterson-Lincoln method of population esti- larly in the grazed corridor. No M. minutus or N. fodiens mation will prove unreliable, it is interesting to speculate as were found in the pellet samples, indicating that the low this allows the population to be compared with documented capture numbers in the Longworth traps accurately depict densities. At their peak in February, M. agrestis densities the population. Optimal Foraging Theory would suggest reach 128/ha in the un-grazed corridor and 48/ha in the that a low number of less preferential species, such as grazed corridor. Although slightly on the low side, they do S. minutus, indicates that the proportion of more profitable 9, 17 fall within most expected densities for grasslands, and species is high. mixed farmland. There is a similar species ranking found between the trap- The small mammals captured in this study appear to be ping results and the pellet data. Similarly, Bonvicino and fairly sedentary. Despite recorded home ranges of up to Bezerra showed that regurgitated pellets from Barn owls 2 18 1000 m , the mean distance travelled by M. agrestis was were a good source of information for assessing species rich- only 18.8 m and the maximum being 65 m. Even more ness. They found strong correlations, when assessing total surprising was the 6.7 m mean distance covered by population, despite stressing the potential for bias in A. sylvaticus whose home ranges on farmland often exceed numbers due to the preferential feeding habits of such 1 ha (10 000 m ). This study did not extend to the breeding species. season, and Corbett and Harris noted that home ranges There is little information on the accessibility of prey items can remain small over winter until the onset of sexual matur- to Barn owls. There is an expected preference for M. agrestis ity in spring. Similar changes in movement patterns were in this study; a species that is more abundant and dominant recorded by Trump with few A. sylvaticus moving more in the un-grazed corridor. What is unknown is to what than 10 m in the winter months, but recorded movements degree the high population, in the deep thatch layer, of 30–50 m by May. is offset by their increased inaccessibility. Barn owl mor- The distribution patterns of small mammals in both corri- phology does suggest that their legs and talons should 1 56 dors are different. The higher floral species richness and enable them to penetrate dense grassland .Lack stated ......................................................................................................................................................................................................................................... 161 Research article Bioscience Horizons † Volume 2 † Number 2 † June 2009 ......................................................................................................................................................................................................................................... 9. Harris S, Morris P, Wray S et al. (1995) A Review of British Mammals: that sward heights of at least 25 cm were ideal for Barn owl Population Estimates and Conservation Status of British Mammals Other foraging. Meek et al. suggested that there is an optimal than Cetaceans. Peterborough: JNCC. corridor width if Barn owls are to hunt at maximum effi- 10. Tew TE (1994) Farmland hedgerows: habitat, corridors or irrelevant? A small ciency. This width is one where the greatest number of mammal’s perspective. In Watt TA, Buckley GP eds, Hedgerow Management prey can be intercepted on a single fly over. Flying most and Nature Conservation. Wye: Wye College Press, pp 80–94. commonly at a height of 3 m when hunting and having an 11. Gorman ML, Rogers LM (1995) The population dynamics of small mammals effective hearing range of 508 indicates that this margin living in set-aside and surrounding semi-natural and crop land. J Zool 236: width would need to exceed 7 m. Although not as wide as 451–464. the grazed corridor, the un-grazed corridor is up to 55 m 12. Kotzageorgis GC, Mason CF (1997) Small mammal populations in relation to hedgerow structure in an arable landscape. J Zool 242: 425–434. wide and no ,20 m wide at any point. 13. Mallorie HC, Flowerdew JR (1994) Woodland small mammal population This study is based on the premise that the strongest influ- ecology in Britain: a preliminary review of The Mammal Society survey of ence on community structure within these two corridors is wood mice Apodemus sylvaticus and bank voles Clethrionomys glareolus, habitat quality, the main influence on this being past and 1982–87. Mam Rev 24: 1–15. present management. Without a long-term monitoring 14. Trump DPC (2003) The Barn Owl Tyto alba, Field Margins and Small Mammals. programme and no specific historical data, the accuracy of PhD Thesis. Edinburgh: Napier University. this statement, and the degree to which un-measured external 15. Wilkinson SAJ (2002) Investigation of Invertebrate Diversity Associated with influences play a role on populations, has to remain Field Vole Habitat. PhD Thesis. Bristol: University of Bristol. un-quantifiable. Although M. agrestis was the most abundant 16. Corbett GB, Harris S eds (1991) The Handbook of British Mammals, 2nd ed. Oxford: Blackwell Scientific Publishing. species in both corridors, they were particularly dominant in 17. Lambin X, Petty SJ, Mackinnon JL (2000) Cyclical dynamics in field vole the un-grazed corridor. The total population is low, despite populations and generalist predation. J Anim Ecol 69: 106–118. the peak falling within the expected range. The un-grazed cor- 18. Macdonald DW, Tattersall F (2001) Britain’s Mammals: The Challenge for ridor has not received any management over the past 8 years Conservation. London: Peoples Trust for Conservation. and still supports a relatively healthy M. agrestis-dominated 19. Baker PJ, Ansell RJ, Dodds PAA et al. (2003) Factors affecting the distribution community of small mammals. The data presented in this of small mammals in an urban area. Mam Rev 33: 95–100. paper suggest that the current recommended rotation of 2/3 20. Churchfield S (1990) The Natural History of Shrews. London: Christopher years is unnecessary. The apparent M. agrestis densities in Helm, A & C Black. the grazed corridor are surprisingly low. The trapping 21. Tapper S (1979) The effect of fluctuating vole numbers on a population of results, NVC classifications and sward condition suggest weasels on farmland. J Anim Ecol 48: 603–617. that this corridor has been over-grazed. The use of a closely 22. Hanski I, Hansson L, Henttonen H (1991) Specialist predators, generalist pre- monitored system of cattle grazing would be recommended. dators, and the microtine rodent cycle. J Anim Ecol 60: 353–367. If not financially or practically possible, the corridors could be 23. Gause GF (1964) The Struggle for Existence. New York: Hafner. mechanically topped to a height of no ,20 cm. Although a 24. Evans FC, Murdoch WM, Peterson CH (1972) Diversity and pattern in plants and insects. Ecology 53: 819–829. valuable data set, the potential strength of this study lies in 25. MacArthur RH, MacArthur JW (1961) On bird species diversity. Ecology 42: it becoming baseline data for future monitoring. It is suggested 594–598. that a long-term monitoring programme be set up to give a 26. Rozenweig ML, Winakur J (1969) Population ecology of desert rodent more accurate picture of populations present within all the communities: habitats and environmental complexity. Ecology 59: corridors at Folly Farm. 558–572. 27. Dueser RD, Shugart HH (1978) Microhabitats in a forest-floor small mammal fauna. Ecology 59: 89–98. References 28. Bowers MA (1990) Exploitation of seed aggregations by Merriam’s kangaroo 1. Shawyer C (1998) The Barn Owl. Chelmsford: Arlequin Press. rat; harvesting rates and predatory risk. Ecology 71: 2334–2344. 2. Buckley J, Goldsmith JG (1975) The prey of the barn owl (Tyto alba) in East 29. Kotler BP, Brown JS, Mitchell WA (1993) Environmental factors affecting Norfolk. Mam Rev 5: 13–16. patch use in two species of gerbilline rodents. J Mamm 74: 614–620. 3. Love RA, Webon C, Glue DE et al. (2000) Changes in the food of British Barn 30. Hornfeldt B. Delayed density dependence as a determinant of vole cycles. Owls (Tyto alba) between 1974 and 1997. Mam Rev 30: 107–129. Ecology (1994) 75: 791–806. 4. Toms MP, Crick HQP, Shawyer CR (2001) The status of breeding barn owls 31. Hamback PA, Oksanen T, Schneider M (1998) Winter herbivory by voles Tyto alba in the United Kingdom 1995–1997. Bird Study 48: 23–37. during a population peak: the relative importance of local factors and land- scape pattern. J Anim Ecol 67: 544–553. 5. Bunn DS, Warburton AB, Wilson RDS (1982) The Barn Owl. London: Poyser. 32. Anon. (1999) Non-SSSI Grassland and Barn Owl Corridors. Folly Farm Nature 6. Webster JA (1973) Seasonal variation in mammal contents of barn owl cast- Reserve Management Plan 1999–2004. Bristol: Avon Wildlife Trust. ings. Bird Study 20: 185–196. 33. Rodwell JS, ed (1992) British Plant Communities, Vol. 3. Grasslands and 7. Glue DE (1967) Prey taken by the barn owl in England and Wales. Bird Study Montane Communities. Cambridge: Cambridge University Press. 14: 169–183. 34. Bunce RGH, Barr CJ, Gillespie MK et al. (1999) ECOFACT 1: Vegetation of the 8. Love RA (2005) Mammal Society National Owl Pellet Survey [online]. http:// www.abdn.ac.uk/mammal/owl_survey/index1.htm (accessed 2 February British Countryside—the Countryside Vegetation System. Rotherham: DETR 2007). Publications. ......................................................................................................................................................................................................................................... 162 Bioscience Horizons † Volume 2 † Number 2 † June 2009 Research article ......................................................................................................................................................................................................................................... 35. Chitty D, Kempston DA (1949) Prebaiting small mammal traps and a new 47. Carter PKL, Churchfield S (2006) Distribution and habitat occurrence of water shrews in Great Britain. Mamm Soc, Research Report Number 7. design in live trap. Ecology 6: 36–53. 48. Tanton MT (1965) Problems of live trapping and population estimation for 36. Pearson DE, Ruggiero LF (2003) Transect versus grid trapping arrangements the wood mouse. J Anim Ecol 34: 1–22. for sampling small mammal communities. Wildlife Soc Bull 31: 454–459. 49. Tattersall FH, MacDonald DW (2003) Woodmice in the arable system. In 37. Gurnell J, Flowerdew JR (2006) Live Trapping Small Mammals: A Practical Tattersal FH, Manley W eds, Conservation and Conflict: Mammals and Guide, 4th ed. Reading: The Mammal Society. Farming in Britain. London: The Linnean Society, pp 82–96. 38. Kent M, Coker P (1992) Vegetation Description and Analysis—A Practical 50. Meek WR, Burman PJ, Nowakowski M et al. (2002) Barn owl release in Approach. New York: John Wiley. lowland Southern Britain—a twenty year study. Biol Conserv 109: 271–282. 39. Yalden DW, Morris PA (1990) The Analysis of Owl Pellets. Occasional publi- 51. Bryja J, Heroldova M, Janova E et al. (2005) Common vole (Microtus agrestis) cation of the Mammal Society No. 13. Bristol: The Mammal Society. population sex ratio: biomass and process variation. Can J Zool 83: 40. Simpson EH (1949) Measurement of diversity. Nature 163: 688. 1391–1399. 41. Taylor IR (1994) Barn Owls: Predator–Prey Relationships and Conservation. 52. Krebs CJ, Davies NB (1993) An Introduction to Behavioral Ecology, 3rd ed. Cambridge: Cambridge University Press. Oxford: Blackwell Scientific Publications. 42. Barn Owl Trust (2006) Habitat Management—Leaflet No. 1. Ashburton: Barn 53. Tattersall FH, MacDonald DW, Hart BJ et al. (2002) Is habitat linearity Owl Trust. important for small mammal communities on farmland? J Appl Ecol 39: 43. Fitzgibbon CD (1997) Small mammals in farm woodlands: the effects of 643–652. habitat isolation and surrounding land-use patterns. J Appl Ecol 34: 530–539. 54. Sibbald S, Carter P, Poulton S (2006) The Mammal Society Research Report 44. Gurnell J (1985) Woodland rodent communities. Symp Zool Soc Lond 55: No. 6. Proposal for a National Monitoring Scheme for Small Mammals in the 377–411. UK and the Republic of Eire. London: The Mammal Society. 45. Alibhai SK, Gipps JHW (1991) The bank vole. In Harris S, Corbett GB eds, The 55. Bonvicino CR, Bezerra AMR (2003) Use of regurgitated pellets of Barn owl Handbook of British Mammals. Oxford: Blackwell Scientific Publications. (Tyto alba) for inventorying small mammals in the Cerrado of Central Brazil. Stud Neotrop Fauna Environ 38: 1–5. 46. Bence S, Griffiths M, Stander K (2003) Habitat characteristics of Harvest mouse nests on arable farmland. Agric Ecosyst Environ 99: 179–186. 56. Lack P (1992) Birds on Lowland Farms. London: HMSO. Author Biography Alex Keene is currently studying Environmental Biology at Bath Spa University. On completion of the course he will be seeking work involving both the practical and educational side of environmental/conservation management. Alex has been working as a freelance photographer for the past ten years and is keen to integrate this into his new career at some point in the future. ........................................................................................................................................................................................................................................ Submitted on 30 September 2008; accepted on 12 February 2009; advance access publication 17 April 2009 .........................................................................................................................................................................................................................................

Journal

Bioscience Horizons – Oxford University Press

Published: Jun 17, 2009

Keywords: Key words Short-tailed field vole small mammal Barn owl corridor pellet Longworth trap

There are no references for this article.