APPENDIX 3
Objective 3:the nature of secondary foraging habitats and extent of use by birds
(i) Use of boundary versus in-field habitats with respect to field type.
Methods
The 2008-09 winter bird surveys described under Objective 2 (surveys of all tetrads in each of December or January and February or March, plus further visits to seven tetrads in the first two weeks of April), incorporated coverage of all fields in each tetrad and recording of all birds encountered as being in the field itself, in boundary vegetation or in flight. In order to investigate the extent to which field boundary habitats were used and how this varied with respect to crop type, fields consisting of (a priori) seed-rich vegetation, i.e. crop stubbles and wild bird covers, were compared to other field types in terms of the proportion of birds of the key seed-eating species found in boundaries versus in-field. Birds present in the boundaries of seed-rich fields are most likely to be resting or loafing between foraging bouts, whereas there is no obvious reason why these species should use the boundaries of non-seed-rich fields unless they are finding food there.
Percentages of the local populations of each species in the boundaries and field centres of seed-rich and other field types were estimated from the winter survey results and the significance of apparently non-random distributions of individuals was assessed using 2 tests.
Results
Summaries of the count results for the principal seed-eating species found during the surveys are summarized by visit in Table 1. Large proportions of the populations of all these species were found in field habitats not usually considered “seed-rich”, i.e. fields under active cropping and bare plough/till, and these proportions rose between successive survey visits through the winter as the available areas of seed-rich habitat (chiefly stubble and wild bird cover) fell. Note, however, that the use of stubble and wild bird cover was still far higher for most species than up to the c. 10% (Appendix 2: Tables 1 & 2) that would have been expected if seed-rich habitats were not selected actively at all (Table 1). This pattern mirrors the results of BD1639 and suggests that granivorous birds are finding food resources outside stubbles, covers and fallow fields, as well as within these accepted seed-rich habitats. The results are also similar to those of BD1639 in that bird counts were more biased towards records in boundary vegetation that in field centres in non-seed-rich field types than in stubbles and covers for chaffinch, yellowhammer, grey partridge and (less clearly) goldfinch (Table 1). This suggests, as might be expected, that whatever resources these birds are finding in cropped fields, they are more concentrated in the boundary vegetation than in the cropped area than is the case in stubbles and covers. However, it should be noted that the opposite pattern was found for linnet and (less clearly) skylark (Table 1): both species tend to avoid field boundary habitats in winter and this result seems primarily to reflect a reduced avoidance of boundary habitats in seed-rich fields, i.e. linnets perhaps using hedges as foraging cover in a way that they tend not to do in more open, cropped fields and skylarks perhaps being more tolerant of vertical structure in fields containing a richer food resource. It is unknown whether the use of boundary habitats has increased since set-aside disappeared and the data collected under BD1628 do not allow this to be tested.
Table 1. Selection of field and boundary habitats by key granivorous species in seed-rich (SRH) and other field types. Preferences are directions of relative selection effects identified by significant 2 tests.
Species / Visit / Field type / Observed / Expected / 2 P-value / Preference for in-field (F) or boundary (B) in SRH / % total count in seed-rich fieldsField / Boundary / Field / Boundary
Corn Bunting / 1 / Other / 9 / 63 / 8.9 / 63.1
1 / SRH / 3 / 22 / 3.1 / 21.9 / 0.948 / 25.8
2 / Other / 3 / 11 / 1.6 / 12.4
2 / SRH / 0 / 12 / 1.4 / 10.6 / 0.088 / 46.2
3 / Other / 27 / 13 / 27.3 / 12.7
3 / SRH / 1 / 0 / 0.7 / 0.3 / 0.690 / 2.4
Chaffinch / 1 / Other / 369 / 1254 / 443.0 / 1180.0
1 / SRH / 279 / 472 / 205.0 / 546.0 / 0.000 / F / 31.6
2 / Other / 179 / 1859 / 188.0 / 1850.0
2 / SRH / 36 / 257 / 27.0 / 266.0 / 0.053 / 12.6
3 / Other / 49 / 447 / 55.2 / 440.8
3 / SRH / 8 / 8 / 1.8 / 14.2 / 0.000 / F / 3.1
Goldfinch / 1 / Other / 25 / 334 / 29.4 / 329.6
1 / SRH / 10 / 58 / 5.6 / 62.4 / 0.033 / F / 15.9
2 / Other / 31 / 340 / 29.5 / 341.5
2 / SRH / 19 / 1.5 / 17.5 / 0.643 / 4.9
3 / Other / 67 / 0.0 / 67.0
3 / SRH / 3 / 0.0 / 3.0 / 1.000 / 4.3
Greenfinch / 1 / Other / 111 / 273 / 114.7 / 269.3
1 / SRH / 27 / 51 / 23.3 / 54.7 / 0.315 / 16.9
2 / Other / 37 / 318 / 35.0 / 320.0
2 / SRH / 20 / 2.0 / 18.0 / 0.560 / 5.3
3 / Other / 1 / 128 / 1.0 / 128.0
3 / SRH / 0 / 0 / 0.0 / 0.0 / #DIV/0! / 0.0
Linnet / 1 / Other / 446 / 312 / 404.4 / 353.6
1 / SRH / 78 / 41.6 / 36.4 / 0.000 / B / 9.3
2 / Other / 554 / 174 / 526.1 / 201.9
2 / SRH / 27 / 49 / 54.9 / 21.1 / 0.000 / B / 9.5
3 / Other / 129 / 69 / 118.3 / 79.8
3 / SRH / 18 / 10.8 / 7.3 / 0.000 / B / 8.3
Grey Partridge / 1 / Other / 19 / 23 / 25.1 / 16.9
1 / SRH / 15 / 8.9 / 6.1 / 0.005 / F / 26.3
2 / Other / 3 / 44 / 6.1 / 40.9
2 / SRH / 4 / 3 / 0.9 / 6.1 / 0.000 / F / 13.0
3 / Other / 2 / 4 / 2.0 / 4.0
3 / SRH / 0 / 0 / 0.0 / 0.0 / #DIV/0! / 0.0
Reed Bunting / 1 / Other / 1 / 10 / 1.9 / 9.1
1 / SRH / 7 / 28 / 6.1 / 28.9 / 0.405 / 76.1
2 / Other / 6 / 23 / 6.0 / 23.0
2 / SRH / 4 / 15 / 4.0 / 15.0 / 0.976 / 39.6
3 / Other / 7 / 24 / 8.2 / 22.8
3 / SRH / 2 / 1 / 0.8 / 2.2 / 0.098 / 8.8
Table 1, continued.
Species / Visit / Field type / Observed / Expected / 2 P-value / Preference for in-field (F) or boundary (B) in SRH / % total count in seed-rich fieldsField / Boundary / Field / Boundary
Skylark / 1 / Other / 835 / 9 / 837.7 / 6.3
1 / SRH / 496 / 1 / 493.3 / 3.7 / 0.075 / 37.1
2 / Other / 970 / 29 / 957.3 / 41.7
2 / SRH / 200 / 22 / 212.7 / 9.3 / 0.000 / B / 18.2
3 / Other / 473 / 22 / 473.1 / 21.9
3 / SRH / 3 / 2.9 / 0.1 / 0.933 / 0.6
Yellowhammer / 1 / Other / 88 / 438 / 160.3 / 365.7
1 / SRH / 232 / 292 / 159.7 / 364.3 / 0.000 / F / 49.9
2 / Other / 137 / 800 / 169.7 / 767.3
2 / SRH / 59 / 86 / 26.3 / 118.7 / 0.000 / F / 13.4
3 / Other / 113 / 315 / 127.1 / 300.9
3 / SRH / 23 / 7 / 8.9 / 21.1 / 0.000 / F / 6.6
(ii) Assessment of habitat features defining bird-use “hotspots”
The result (above and in BD1639) that farmland birds appear to be usingfield boundary habitats more than had previously been suspected leads to the question of what features of field boundaries are proving attractive. Answering this question will indicate which boundary habitats might beneficially be protected and which boundary features might be promoted in order to replace the benefits of the high-quality foraging habitat provided by set-aside, without resorting to major changes to cropping patterns or field centre management. Data on the structure and vegetation content of all field boundaries in the 20 study tetrads in East Anglia were collected under BD1628, allowing the features of areas of boundary being used by farmland birds (“hotspots”) during winter 2008-09 to be assessed with respect to the range of habitats available.
Methods
BD1628 boundary data
All field boundaries in every tetrad were mapped during bird survey fieldwork in Autumn 2005. Gaps in data coverage and gross changes in boundary habitat structure were accounted for by additional data collection during winter 2009-10. Boundary recording was conducted on the basis of discontinuities in structure delimiting sections in which habitat was constant. Each section was then numbered and the habitat recorded as per the variables described in Table 2. The limits and number of each section were also recorded on a field map and, subsequently, digitized using ArcMap 9.2 (ESRI 2006) so that section lengths could be extracted. All boundaries of agricultural habitats were considered, i.e. all linear features and edges of habitat blocks adjacent to farmland. Boundaries within areas of human habitat and visible linear features within woodland were ignored, as were boundaries between crops where there is no permanent uncropped area. Double hedges, i.e. where two hedges run parallel on either side of a road, path or track, were recorded as separate sections, unless they met to form a single canopy (generally only across footpaths), when they were treated as a single section. All heights and widths were estimated by eye in the field.
Table 2. Data recorded to characterize field boundaries
Category / Variable / DescriptionGeneral features / Type / Nine-level factor: solid hedge, gappy hedge, isolated bushes, treeline (line of trees with no hedge beneath), bank, wood, scrub (including new plantations), rough (non-woody vegetation only) and “narrow” (strips of uncropped land of less than 1m in width).
Garden / Binary variable, present or absent on non-agricultural side of boundary.
River/ stream / Binary variable, present or absent along boundary.
RTP / Four-level factor: road, track, path or none.
Ditch / Binary variable, present or absent.
Fence / Binary variable: “yes” if the fence was the only vertical structure and not when it was buried in hedge.
Vegetation / Character of (woody) vegetation in the boundary: deciduous, coniferous, mixed or herbaceous vegetation only (no woody vegetation present).
Dimensions / Number of trees / Trees standing above the rest of the boundary.
Boundary width / Width of entire boundary, i.e. the uncropped area, but excluding any field margin.
Hedge width / Width of any hedge present, i.e. of the woody hedge vegetation only, in metres.
Hedge height / Height of any hedge present in metres, excluding trees.
Tree height / Average height of any trees present in metres, including woods and scrub. N.B. hedge height and tree height categories basically allowed for two heights in a single hedgerow section; trees were >5m tall, but various combinations were possible, e.g. 10m trees in a 7m hedge or 6m trees in a 2m hedge.
Margin width / Width of any field margin present in metres, excluding any other boundary vegetation. Margins on both sides of the boundary were recorded separately (e.g. “5+4” for a 5m boundary on one side and a 4m one on the other) and converted to a mean width and binary variable, “double margin”, for analysis.
Field data on bird occurrence
The locations of all chaffinches, yellowhammers, greenfinches, goldfinches, reed buntings, linnets, tree sparrows and corn buntings seen or heard within tetrads during the winter field surveys described under Objective 2 were recorded on field maps. All records of groups of more than two individuals of each of these species that were not in open field areas were used to define the associated boundary habitat section as a “hotspot” for the species concerned. Boundary sections adjacent to known seed-rich habitats (stubbles or wild bird covers) were identified as such; they are likely to have been used as cover rather than as foraging habitat in their own right. Separate sets of hotspots were defined for each survey visit (December-January, February-March and April, the latter involving just seven tetrads).
Statistical analyses
Selection of a boundary section as a “hotspot” by a species, on a given survey visit, was tested using logistic regression models of the binary variable for selection as a function of boundary habitat features, controlling for boundary section length. Data were pooled across study tetrads. Each of the “general feature” variables listed in Table 2, together with a binary variable for field margin presence (“margin”: zero versus non-zero margin widths), was tested for influences on boundary selection. Hedge and tree dimension variables were tested among the hedgerow and remnant hedgerow habitats only, i.e. solid hedges, gappy hedges and isolated bushes, because they would not have consistent meanings biologically across the full range of habitat types considered.
The tests were repeated using a second definition of habitat selection, considering only flocks of more than ten birds to indicate selection. This was intended to reveal whether particular features were associated with larger feeding flocks, but had the cost that the data included fewer “selected” locations, increasing the potential for apparent patterns of selection merely to reflect chance associations.
All analyses were conducted using SAS 9.2 (SAS Institute, 2008); significance was tested using likelihood-ratio tests.
Results
Tests on hotspot selection with respect to habitat were conducted for all the target species except reed bunting, tree sparrow and corn bunting, for which none were possible because no groups, or only one or two groups, respectively, of more than two individuals were found away from seed-rich habitats (Table 3). The occurrence of groups of more than three individuals of the rarer species away from seed-rich habitats also became rarer as the winter progressed (Figure 1), so tests could not be conducted for tree sparrow, corn bunting and linnet for one or both of the later two visits. In general, larger groups were rarer on later surveys for species like chaffinch, as more birds were recorded as pairs or singles, presumably often on breeding territories. Nevertheless, there was clear pattern for the five most common species considered for birds to be found in a combination of small and large groups (Figure 1). Small groups (fewer than five individuals) tended to be most common for the commoner species, but a number of large flocks also occurred, presumably associated with particularly valuable food resources. More individual greenfinches and linnets were found in larger flocks. The proportion of the birds recorded found in hotspots (i.e. groups of three or more away from seed-rich habitat fell rapidly after December-January for all species (Table 3), perhaps reflecting a combination of increased territorial behaviour promoting the occurrence of smaller groups and decreased food densities that only supported small groups.
Table 3. Summary of bird counts in hotspot (non-seed-rich) habitats. Data from all 19 study tetrads are pooled. Note that only seven of the tetrads were covered during the April visit, so total counts were, inevitably, smaller.
Species / Visit / Counts per hotspot / No. of hotspots / Total across hotspots / Total across all habitats / Percentage in hotspotsMean / Median / Min. / Max.
Chaffinch / Dec-Jan / 13.0 / 5 / 3 / 120 / 73 / 951 / 2455 / 38.7
Feb-Mar / 6.7 / 4 / 3 / 26 / 18 / 121 / 2429 / 5.0
Apr / 4.5 / 4.5 / 3 / 6 / 4 / 18 / 553 / 3.3
Yellowhammer / Dec-Jan / 13.2 / 4.5 / 3 / 75 / 26 / 344 / 1056 / 32.6
Feb-Mar / 16.7 / 16 / 3 / 75 / 11 / 184 / 1113 / 16.5
Apr / 5.0 / 5 / 5 / 5 / 2 / 10 / 462 / 2.2
Goldfinch / Dec-Jan / 6.4 / 4 / 3 / 15 / 27 / 173 / 483 / 35.8
Feb-Mar / 5.0 / 5 / 4 / 6 / 4 / 20 / 429 / 4.7
Apr / 4.0 / 4 / 4 / 4 / 1 / 4 / 74 / 5.4
Greenfinch / Dec-Jan / 18.5 / 7 / 3 / 100 / 12 / 222 / 486 / 45.7
Feb-Mar / 4.3 / 5 / 3 / 5 / 3 / 13 / 406 / 3.2
Apr / 5.0 / 5 / 4 / 6 / 2 / 10 / 152 / 6.6
Linnet / Dec-Jan / 35.8 / 12 / 3 / 140 / 10 / 358 / 841 / 42.6
Feb-Mar / 25.5 / 25.5 / 13 / 38 / 2 / 51 / 845 / 6.0
Tree Sparrow / Dec-Jan / 7.0 / 7 / 7 / 7 / 1 / 7 / 20 / 35.0
Corn Bunting / Dec-Jan / 24.0 / 24 / 8 / 40 / 2 / 48 / 98 / 49.0
Figure 1. Frequency of occurrence of different flock sizes at hotspot locations for each species. Data are separated by visit. Visit 1 was December-January, visit 2 February-March and visit 3 April. Note that only seven tetrads were visited in April.
Test results for the general habitat features of selected versus non-selected boundary sections are given in Table 4. Tests for corn bunting and tree sparrow could only be conducted for the December-January visit and were all non-significant, although the generality of this result may be limited because these species were only found in small numbers and a few sites (Table 3). There were at least two significant results across the range of tests conducted for each of the other species, but there was no evidence for consistent patterns across species (Table 4). Chaffinches in December-January tended to select boundary sections without field margins and showed some, relative avoidance of isolated bushes and “narrow” boundaries, while never being found in treelines or banks (Table 5). In February-March, goldfinches showed preferences for boundaries without ditches and with adjacent tracks or, to a lesser extent, no human thoroughfare (Tables 4 and 5). Greenfinches showed preferences for boundaries with margins and adjacent to gardens in December-January (Tables 4 and 5). Linnets selected boundaries with ditches in December-January and without margins in February-March (Tables 4 and 5). Yellowhammers showed significant habitat selection only in December-January, when they tended to select boundaries with ditches but to avoid those along roads and with adjacent gardens (Tables 4 and 5).
Among linear boundary features with woody structures that were likely, a priori, to be attractive to the species of interest (solid and gappy hedges, and isolated bushes), there were rather few significant associations with structural variables, especially given the likely occurrence of apparently significant effects given the number of tests conducted, and there were no clear, consistent patterns across or within species (Table 6). Chaffinches showed a weakly negative association with margin width during April visits, while greenfinches showed a positive association in February-March. Greenfinches, (February-March) and yellowhammers (April) showed positive associations with boundary width, but greenfinches also showed a weakly negative association in December-January (Table 6). Linnets showed a tendency to select wider hedges with taller trees in February-March, but both patterns were weak (Table 6).
Considering “selected” locations only to be those with flocks of more than ten birds, fewer tests were possible, especially later in the winter, because large flocks were rarer or absent entirely (Figure 1). Chaffinches showed the same pattern of avoidance of isolated bushes, “narrow” boundaries, treelines and banks in December-January, but the avoidance of field margins disappeared (Tables 7 and 8). In February-March, they appeared to select boundaries with streams (Tables 7 and 8). Goldfinches showed a weak pattern for selection of boundaries with gardens and greenfinches a negative association with ditches, neither of which patterns were apparent in the results for all hotspot locations (Tables 7 and 8). The results for linnet may be the least reliable because the sample size was smallest (Table 3, Figure 1), but the patterns for a selection of ditches in December-January and an avoidance of margins in February-March were still detected, along with a weak association with boundaries with streams in December-January (Tables 7 and 8). There was also evidence for a further, weak association with field margins in December-January (Table 7), but this was in the opposite direction to the pattern later in the winter (Table 8). All the significant patterns for yellowhammer found previously were no longer apparent considering flocks of ten or more, but these flocks occurred more often in boundaries with field margins than would have been expected by chance (Tables 7 and 8). All the patterns for selection of boundaries with respect to hedgerow structure, except for the weak associations of linnets with taller trees and wider hedges, were no longer detectable when flocks of ten or more alone were considered (Table 9). One additional effect was apparent, for goldfinches to avoid boundaries with wider margins, but this too was weak (Table 9).