Hedgerow Management in Pastoral Landscapes

Hedgerow Management in Pastoral LandscapesAbstractHedgerows ar an serious part of the British landscape, providing some(prenominal) food and shelter for a amount of taxa. As part of the UK g overnments Environmental Stewardship (ES) Scheme, farmers are granted subsidies for, amongst other things, enhanced overr separatelyrow precaution. Although hedge counseling under(a) ES is expected to film beneficial personal effects for taxa such as invertebrates and birds, slight is known about the effects ES management will postulate on petty mammalianianian communities. The aim of this study was to investigate whether this enhanced hedgerow management is affecting hedgerow fountistics in idyl landscapes and whether thin mammal teemingnesss are make upd under ES managed hedgerows. Conservation lover strips (2m+ unimproved grassy margins) were investigated as a possible improvement to ES hedgerow management. employ live ho apply methods, I investigated teensy-weensy mammal abundances in ES managed hedgerows compared with non-ES managed hedgerows. Wood mice Apodemus sylvaticus and shore fieldmouses Clethrionomys glareolus were the most abundant species, with some captures of field of operation fieldmouses genus Microtus agrestis and common shrews Sorex araneus. weeny mammal abundances were increased in ES managed hedgerows, however, the posture of a preservation buffer strip was more(prenominal) epochal in increasing olive-sized mammal densities. ES management showed no definite effect on the hedgerows characteristics.IntroductionAgricultural intensification since the 1940s has led to general and significant reductions in the bio variation of many agricultural areas. This drive for greater yields has been linked with the population decreases seen in many species of farmland specialists and non-specialists who oftentimes inhabit farmland (Robinson and Sutherland, 2002). cultivated land habitats can be categorised into non-linear habitats such as set-aside, cropped field and wood areas, and linear habitats, generally field boundaries, such as ditches, banks, streams and hedgerows. These field boundaries remain relatively undisturbed areas and are therefore significant wildlife corridors in spite of appearance otherwise inhospitable agricultural landscapes (Tew, 1994).Although there continued to be a reduction in hail area of hedgerows in spite of appearance the UK during the 80s and early 90s, the last decade has seen olive-sized increases in the area of hedgerow as their conservation significance became more documented (Barr and Gillespie, 2000). This increase in the way out of hedgerows has been driven by government backed grants. Countryside Steward (CS), set up in 1991 encouraged selected farmers to enhance and conserve the wildlife in spite of appearance their farms, a turgid part of this involved the set of new hedgerows. The CS objects have now been superseded by the Environmental Stewardship Schem es. More recently, hedgerow grant pilot schemes have been set up in a number of regions to encourage landowners, both farmers and non-farmers to manage their hedgerows more effectively these grants are available to pay for gapping up, hedge laying or coppicing.Small mammals in pastoral land are badly confined to hedgerows or other non-crop features and are therefore particularly vulnerable to intensification (Bates and Harris, 2009). Small mammal species constitute the main prey biomass for a number of species of mammals and birds, and therefore itty-bitty mammal abundance directly influences the abundance and diversity of predator species add to the complexity of local food webs (Korpimaki and Norrdahl, 1991). at that place remains some debate on the importance of linear habitats for small mammals, with some declare oneselfing that they cannot support viable populations, that those prepare in hedgerows are sink populations (Tattersall et al. 2004). However, there is evidence t hat small mammal abundance and diversity does not depend on the linear or non-linear character of the habitat and that linear habitats can support viable populations (Gelling et al. 2007). Thus, in large expanses of uninhabitable grassland, field boundary hedgerows are of great importance for maintaining small mammal populations in an agricultural landscape, but differing farming practices can lead to a huge variety in the quality of these habitatsAs the emphasis of farming has shifted there have been a number of agri-environment schemes introduced across Europe with the aim of reversing the effects of previous intensification and enhancing agricultural land for wildlife (Kleijn and Sutherland, 2003). The UK introduced a new set of farming standards in 2005 with farmers now guaranteed subsidy payments, known as cross-compliance, as long as they follow a set of prescribed conditions aimed at improving the environmental pass judgment of their farms. A compulsory code of good practice will preclude farming land within 2 m of the centre of a hedge (DEFRA, 2005a). Beyond cross-compliance subsidies, farmers can also apply to put their farmland into Environmental Stewardship (ES). ES is a tiered system, with Entry-Level ES designed to allow most farmers glide slope to the payments by compiling a farm management plan that aims to improve the features of their farm for wildlife and to maintain/improve the scenic value of the British countryside. The enhanced hedgerow management option within ES requires that the farmer skip over the hedge no more than once every 2 years, that hedgerows are cut during the wintertime and that cutting be staggered across the farm. The combined aim of these prescriptions is to ensure that at least some of the hedgerow is allowed to flower every summer (Defra, 2005b).Properly managed hedgerows are valuable features, playing a key role in enhancing the wildlife value of farmland. Flowering hedgerows are an important source of food and sh elter for a number of birds (Hinsley and Bellamy, 2000). Studies suggest that the ES schemes will have a beneficial effect, mainly for taxa such as invertebrates and birds (Kleijn and Sutherland 2003), Whittingham (2007) emphasizes the importance of monitoring the effects of ES to ensure that the schemes prescriptions meet the exacts of a greater range of species. It is a good deal less nearly understood how the changes to hedgerow management will effect small mammal abundance, and it is important that there is greater understanding of the factors that influence small mammal populations since small mammals provide the major source of prey biomass for many big predators ( respect et al., 2000). Small mammals also play a role in a range of important ecosystem processes (Hayward and Phillipson, 1979).Previous studies have established the main effects of change hedgerow management within arable landscapes (Shore et al. 2005). Arable environments provide cover for small mammals due to the height and density of the crop. Small mammals have been shown to make essential use of the field at certain times of the year (Tattersall et al. 2001 Tew et al. 2000 Todd et al. 2000). However, no small mammal species have been shown to make use of agriculturally improved pastoral fields at any time of year (Montgomery and Dowie 1993). Grazed pastoral land provides very little cover, restricting the movements of resident small mammal communities. Therefore, hedgerow management in predominantly dairy and cattle areas will likely have a large influence on the success of small mammal populations (Gelling et al. 2007). In particular, the level of solid realm cover vegetation along the hedgerow and the presence of some form of non-farmed margin can significantly affect the small mammal abundance (Bates and Harris 2009, Gelling et al. 2007). The 2m margin prescribed by cross compliance is irrelevant in terms of providing cover within pastoral landscapes. Although the 2m margin remains jolty and clear of interference from the farmer (no fertilisers), year round range will mean that little cover is offered right up to the base of the hedgerow. Therefore, whereas ES management whitethorn boost small mammal numbers within arable areas (Shore et al. 2005), the value of ES hedgerow management within pastoral landscapes is less well understood. I utilised a number of hedgerow sends to compare hedgerow structure and small mammal communities on ES farms versus non-ES farms. For each farm, one site was selected to be representative and one to include a significant (2m plus) conservation buffer strip of unimproved, non-grazed grass/shrubland. I aimed to investigate (i) how ES management effects the hedgerow characteristics, in particular the level of ground cover for small mammals (ii) whether these ES prescriptions are providing any significant benefit for small mammal densities and (iii) as the movements of small mammals within pastoral landscapes are so restr icted, could small mammal assemblages in hedgerows be significantly improved by including an unimproved, non-grazed, grassy margin or conservation buffer strip (2+m from the edge of the hedgerow).MethodsSitesThe study was conducted over 20 different farms spread across County Durham and Northumberland. The farms were selected due to their suitability for this study, each farm containing both a hedgerow site with a conservation buffer strip and at least one without. All farms selected were representative in terms of habitat of those within the local area. A hedgerow was defined as a continuous line of woody vegetation no more than 3m tall.Hedgerow SurveyThe farms were mated, with one ES farm neighbouring a non-ES farm, making 10 farm pairs and 20 farms in total. Hedgerow surveys were carried out throughout June 2009. 10 hedgerows were randomly selected on each farm. All hedgerows on each farm were surveyed using an edited version of the Defra Hedgerow Survey Form and handbook (DEFRA, 2007). for each one hedgerow was measured to determine its cross-section(a) area. The character of the hedgerow was scored by reference to a series of standard diagrams, noting the level of available ground level cover for small mammals (1=little or no vegetation cover at ground level, 2=gappy cover at ground level, 3= incessant vegetation cover from hedgerows at ground level). Additional variables were recorded, including whether the hedge had been flailed (mechanically cut) recently, i.e. during the previous winter, the number of standard and veteran trees and the number of woody species within the hedgerow. The data sets for cross-sectional area, level of ground vegetation cover and the number of woody species were averaged to cause an overall mean value for each farm. The number of flailed hedgerows was summed to give an overall voice of hedgerows flailed on each farm.Trapping ProcedurePrevious living accommodations studies have shown that, unlike in arable land, small mamm als within pastoral land brook almost entirely within the hedgerows and therefore hedgerows can be treated as linear habitats (Gelling et al. 2007). Trapping was carried out in two major trapping sessions, mid-April to June and mid-July to August, 2009. Within each of the 20 farm sites I selected a representative hedgerow and a hedgerow flanked by an unimproved 2m+ grassy margin, designated a conservation buffer strip, making a total of 40 trapping sites. Where possible the hedgerow sites were selected randomly, however, each ES site was required to have been managed according to the prescriptions of Stewardship farming, i.e. the hedgerows were cut not more than once every two years and the farmers adhered to the prescribed 2m margin of non-interference (2m from the centre of the hedge) (DEFRA 2005a, DEFRA 2005b). Every hedgerow selected was flanked by improved or semi-improved grassland for the grazing of dairy cattle and/or the production of silage. At each site, a 104m section o f isolated hedgerow (not directly connected to woodland) was selected.13 Longworth traps were placed at ground level within the hedgerow, at 8m intervals. Traps were provisioned with hay, apple, oat grains and dried mealworm. The traps were set at dusk and checked at dawn and dusk for three days. All targeted animals that were captured were fur-clipped to help locate re-captures. Species, sex and weight were recorded for each animal forrader release at the point of capture.AnalysisHedgerow characteristics were recorded and analysed using a paired measures multivariate analysis of variance (MANOVA) (SPSS 17.0.2). I had multiple dependent variables that I wished to analyse, however, using multiple one-way ANOVAs to try to do this would have raised the chance of a Type I error (Gibson et al. 2007). Therefore the data was investigated using a MANOVA which controls the experiment-wide error rate. Multiple dependent variables that were related (e.g. Cross sectional area of hedge and a mount of ground cover, etc.) were analysed in one test, with the hedgerow management (ES managed or non-ES managed) being treated as the two levels of the treatment factor (Gibson et al. 2007). There was a total of 4 dependent variables the mean cross-sectional area, the percentage of flailed hedgerows, the average number of woody species and the mean level of ground cover.For each trapping session the relative density was estimated as the minimum number alive (MNA), or the total number of individuals caught over the three days. Species richness was calculated as the number of different species caught. Using General Linear Modelling (GLM Minitab 15), I examined the kinships between small mammal densities and a number of predictor variables. The dependent variables I investigated were the overall total small mammal density (MNA) and the total biomass of all small mammals caught within 104m. I also investigated the density of each individual species, constructing similar models for the number of captures and biomass for each individual species. I foc utilize on wood mice Apodemus sylvaticus and bank voles Clethrionomys glareolus. There were also some captures of field voles Microtus agrestis and common shrew Sorex araneus, these data were not investigated individually but were included in the total density of small mammals and the total biomass. The predictor variables considered were the presence/absence of ES management, the presence/absence of a conservation buffer strip and the number of standard and veteran trees within the hedgerow. The relationships were analysed using a reversed stepwise GLM, with all main predictors and their first order interactions initially included within the model. The insignificant interactions were then removed. Each trapping session was carried out over 3 days on 4 sites on neighbouring farms, the variation between trapping locations and times was taken into account by including the variable block within the initial model, ho wever, it was found to have no significance and was therefore removed from the final model. There are well documented timeal variations in small mammal abundance (Alibhai and Gipps 1985 Flowerdew 1985 Butet et al. 2006), therefore, as there were two major trapping seasons (mid-April to May and Mid-June to July) I included the variable season in all models. The number of captures of field voles and common shrew were too low to allow thorough analysis however, the number of captures for each species was investigated using a Kruskal-Wallis test (Minitab 15) to determine the relationship between the presence of a buffer strip and their individual abundance.ResultsThe total number of catches was 276 individual small mammals of four different species, during 240 trap sessions (dusk till dawn and dawn till dusk). The most abundant species were wood mice, making up 45% of the captures, 11% of which were juveniles, with a total capture of 122 individuals (61 in the first season of trapping and 61 in the second season). 32% (89 individuals) of captures were bank voles, none of which were juveniles, with 26 captures in season 1 and 53 captures in season 2. 17% of captures (48 individuals) were common shrews and 6% (17 individuals) were field voles.Table 1. Summary of the number of captures for each speciesTotal Wood mice Captured date 1 (juveniles) / time 2 (juveniles) Bank vole date 1 / Season 2 Field vole Season 1 / Season 2 Common shrew Season 1 / Season 2 Total Season 1 / Season 2 Total N trapped throughout study 122 61 (2) / 61 (11) 89 36 / 53 17 4 / 13 48 28 / 20 276 129 / 147 character of total 44 33 6 17 100 circumstances of hedgerows present 93 46 23 45 Effect of ES Management and Buffer stripsA total of 40 hedgerows were surveyed with 20 hedgerows under ES hedgerow management and 20 hedgerows under non-ES management. ES sites had been under ES hedgerow management for 2 years or more. The measured dimensions of the hedgerow were used to estimate the hedgerow cross sectional area. Analysis using a paired measures MANOVA found no significant difference in the size of ES managed hedgerows to the size of non-ES managed hedgerows (F(1,9)=0.847, P=0.381). ES management also had no significant effect on the percentage of flailed hedgerows within the farm (F(1,9)=0.019, P=0.889). The woody species diversity within hedgerows was not significantly different between ES managed hedgerows and non-ES managed hedgerows (F(1,9)=3.047, P=0.115). There was a significant positive association of the presence of ES hedgerow management with the level of woody vegetation cover at ground level (F(1,9)=10.613, P=0.010).Table 2. Comparisons of hedgerow characteristics on ES managed farms versus non-ES managed farms. Data were analysed using a paired MANOVA.Mean (SE) Hedgerow characteristic Description of measurement ES Non-ES F(1,9) P Area Average cross sectional area/m2 2.99 (0.12) 2.83 (0.14) 0.847 0.381 Flailed Percentage of hedgerows that had b een recently flailed (flailed during previous Winter) 26.00 (2.21) 25.00 (6.54) 0.019 0.893 Species diversity Number of woody species 3.16 (0.24) 2.73 (0.27) 3.047 0.115 Small mammal cover Average Area of small Mammal cover (1=little or no vegetation cover at ground level, 2=gappy cover at ground level, 3=constant vegetation cover from hedgerows at ground level) 2.63 (0.87) 2.13 (0.11) 10.613 0.010Small Mammal AssemblagesBackward stepwise general linear modelling was used to analyse the data. The results showed that buffer strips have a significant effect on the total number caught within the hedgerow (F(1,35)= 16.29, PA GLM for total biomass showed similar results with Season (F(1,34)=0.83, P=0.369) and the number of standard trees (F(1,34)=1.12, P=0.298) both having no significant effect on the total biomass. ES management had a positive association with total biomass (F(1,34)=4.92, P=0.033), as did the presence of a buffer strip (F(1,34)=27.62, PWood mice were the most common spe cies trapped, contributing 45% of the captures. The factors affecting wood mice captures were analysed using a backward stepwise GLM. Season had no significant effect (F(1,34)=2.36, P=0.134). Unlike the model involving total captures, ES management (F(1,34)=0.07, P=0.798) and Buffer flight strip (F(1,34)A backward stepwise GLM was constructed for both bank vole captures and the total bank vole mass, both models produced similar results. Season had no effect on bank vole captures (F(1,35)=2.06, P=0.160) and total bank vole mass (F(1,35)=1.66, P=0.206). The presence of ES management on the hedgerow had a significant positive effect on the number of bank vole captures (F(1,35)=7.15, P=0.011) and on the total bank vole mass (F(1,35)=5.91, P=0.020). The presence of a buffer also had a significant effect, increasing the number of bank vole captures (F(1,35)=34.90, PTable 3. Summary statistics from general linear modelsModel Variables F P Adj. R2 Total Captures Season F(1,35)=1.09 0.305 5 3.79% ES Managed F(1,35)=5.23 0.028a Buffer Strip F(1,35)=16.29 Standard Trees F(1,35)=0.91 0.346 Total Biomassc Season F(1,34)=0.83 0.369 65.32% ES Managed F(1,34)=4.92 0.033a Buffer Strip F(1,34)=27.62 Standard Trees F(1,34)=1.12 0.298 Season*Buffer Strip F(1,34)=3.18 0.083b Wood Mice Captures Season F(1,34)=2.36 0.134 79.72% ES Managed F(1,34)=0.07 0.798 Buffer Strip F(1,34) Standard Trees F(1,34)=79.65 Season*Standard Trees F(1,34)=4.81 0.035a Total Wood Mice Massd Season F(1,35)=1.36 0.252 69.06% ES Managed F(1,35)=0.26 0.616 Buffer Strip F(1,35)=0.05 0.831 Standard Trees F(1,35)=49.03 0.003a Bank vole Captures Season F(1,35)=2.06 0.160 54.76% ES Managed F(1,35)=7.15 0.011a Buffer Strip F(1,35)=34.90 Standard Trees F(1,35)=4.41 0.043a Total Bank Vole Masse Season F(1,35)=1.66 0.206 50.74% ES Managed F(1,35)=5.91 0.020a Buffer Strip F(1,35)=28.11 Standard Trees F(1,35)=2.32 0.137a Significant to the 95% confidence level b Significant to the 90% confidence lev el c Total Biomass was square root transformed sooner analysis. d Wood Mice Mass was square root transformed before analysis. e Bank Vole Mass was square root transformed before analysis.A total of 17 field voles were captured, with all 17 trapped in hedgerows flanked by a conservation buffer strip. A total of 48 Common shrews were trapped, 81% of which were caught in hedgerows not flanked by a buffer stripTable 4. Non-target species captures. Effect of buffer strip, analysed using Kruskal-Wallis test.Total Captures Species Buffer Strip Present No Buffer Strip H P (adjusted for ties) Field vole 17 0 8.30 0.004 Common shrew 9 38 12.73 handlingHedgerow characteristics are known to affect small mammal numbers. Hedgerows with many gaps and a lack of ground cover support significantly lower small mammal populations (Gelling et al. 2007). Small mammals will select against hedgerows with a lack of vegetative cover due to the increased risk of predation (Orrock et al. 2004). Our resul ts suggest that ES farms produce denser hedgerows with more cover at the ground level than non-ES farms. This is reflected in the small mammal survey which shows a somewhat strong association between small mammal numbers and ES hedgerows. However, having surveyed the farms and the farmers, I acknowledge that a wide number of variables affect the characteristics of the hedgerow. I suggest that the state of the hedgerows for small mammals is more significantly affected by the mindset of the farmer. Those farmers who have moved onto the Entry level ES scheme are generally those who most actively manage their farm. iodine supporting piece of data for this theory, is the number of flailed hedgerows on ES farms compared to non-ES farms. The hedgerow survey found that there were no differences in the number of recently flailed hedgerows within ES farms compared to non-ES farms, therefore, even though the cutting of hedgerows on ES farms is restricted, it still occurs as often on the ES fa rms within this survey than on the non-ES farms. The touch is that those farmers who are on the ES scheme are more actively involved in managing their farm, including their hedgerows, therefore hedgerows on ES farms commonly provide denser vegetation, less gaps and more cover at ground level. The typical ES farmer is more actively managing the hedge as a boundary or barricade to cattle than the typical non-ES farmer. The author suggests this conclusion having discussed hedgerow management with the farmers as part of the hedgerow survey and having a background in agriculture, however, it is also value that this topic goes beyond the scope and available data of this investigation.Hedgerows can be thought of as corridors linking woodland habitat, allowing small mammal migration (Soule and Terbough 1999), however, within the British pastoral landscape, hedgerows are often acting as the sole habitat for small mammals (Fitzgibbon 1997). My investigation found that the ratio of juvenile to adult wood mice increased during the season, with greater numbers present later in the summer, this is consistent with the observations of others (Alibhai and Gipps 1991, Flowerdew 1991). The breeding season for most small mammals begins in spring and ends in late summer, therefore it is rude(a) that more juveniles are present in hedgerows as the summer progresses and they travel outward to establish their own home ranges. The presence of fully grown, breeding adults in both seasons of trapping indicates that animals are resident within the hedgerows, providing support for the argument that linear habitats can provide suitable habitat to support viable populations of small mammals.My results show that the total small mammal abundance and therefore the availability of prey biomass for predators is increased in hedgerows under ES management. The results of the hedgerow survey suggest that there is greater ground level vegetation cover in ES hedgerows. An increase in the amount of physical habitat creates greater run opportunities and can increase small mammal abundance (Gelling et al. 2007). Small mammals prefer hedgerows with greater ground level cover as they provide better mental institution from predators (Orrock et al. 2004).Whereas the benefits of ES management for small mammal abundance remain unclear, this investigation highlights the importance of buffer strips. The value of unimproved grassy margins, in arable landscapes, for small mammal numbers has already been shown (Shore et al. 2005). This study suggests that the presence of a buffer strip along a hedgerow can provide a much improved habitat to support larger small mammal numbers in hedgerows within pastoral landscapes. Grassy margins are a refuge for small mammals beyond the hedgerow they allow increased safety for foraging and greater shelter (Orrock et al. 2004).To understand the variation in the numbers trapped of each species, we need to establish an understanding of the differing ecol ogical requirements for each species. The two most abundant species were the wood mouse and the bank vole. The results show that wood mice are found in greater numbers in hedgerows containing standard/veteran trees. This conclusion is supported by previous studies which have shown that trees within hedgerows are beneficial for wood mice (Montgomerie and Dowie, 1993). Mice often take shelter in burrows formed beneath trees/within tree roots which may suggest why this species was found more commonly within hedgerows containing standard/veteran trees (Montgomerie and Dowie, 1993). Wood mice are a generalist species occupying a wide variety of habitat (Flowerdew 1993). They general occupy a relatively large home range and travel extensively, consuming a wide range of food sources depending upon season and availability (Flowerdew 1993). This is reflected in the results, with wood mice having been trapped in 93% of all the hedgerows. The results also show that wood mice abundance is not a ffected by ES management for hedgerows, nor is it significantly improved by the presence of a buffer strip. Wood mice have been shown to avoid hedgerows with major gaps, and wood mouse captures have been shown to increase with proximity to woodland (Gelling et al. 2007). Wood mice have relatively large home ranges and the suggestion is that individuals rarely stay long within any one hedgerow rather they travel through, utilising hedgerows for foraging and shelter between woodland (Montgomery and Dowie 1993 Gelling et al 2007 Todd et al 2000 Tew et al. 2000). Therefore, ES management and the presence of buffer strips have little effect on the number of wood mouse captures more important is the proximity to woodland or the presence of trees within a hedgerow which provide the preferred shelter for the wood mouse (Todd et al. 2000 Tew et al. 2000).Bank voles are a more specialist species, and generally occupy much smaller home ranges than do wood mice. They are burrowers, using ground vegetation to create runs and pathways in deciduous habitats (Morris 1982 Alibhai and Gipps 1985). Bank voles are a major prey resource for a number of raptorial birds and bank vole abundance has been shown to significantly affect raptor populations (Korpimaki and Norrdahl, 1991). Other studies have found that bank vole numbers are positively associated with the size of hedgerows (Pollard Relton, 1970 Tew, 1994 Bellamy et al., 2000). Grassy margins of 2m plus have been shown to significantly increase bank vole numbers in arable fields (Shore et al. 2005), my results show that this conclusion extends to pastoral landscapes with bank vole numbers being significantly increased by the presence of an unimproved grassy margin or conservation buffer strip. The results also suggest that ES management improves hedgerows for bank voles, with bank vole abundance found to be significantly higher on ES hedgerow sites. Bank voles are found in much greater abundance in areas which provide thick ground vegetation and suffer little disturbance (Tew 1994), my results suggest that this is partially provided by ES management, however, the creation of grassy margins along hedgerows could significantly improve bank vole abundance in pastoral landscapes.The creation of margins could also be significant in the conservation of field voles. Field vole numbers in the UK are in decline believed to be due to the loss of rough grass habitat in intensively managed arable regions (Harris et al., 1995 Love et al., 2000). Field voles are specialists and depend upon rough, ungrazed grassland within woodland and hedgerows. Field voles are generally only found within areas of long grass (Alibhai and Gipps, 1991b). Very a couple of(prenominal) captures of field voles were recorded within this experiment, however all field voles captures occurred within hedgerows flanked by conservation buffer strips. The presence of a buffer strip may provide the field voles