|
Abstract from BTO
Research Report No. 485:
Vickery J., ChamberlainD.,
Evans A., Ewing S., Boatman N., Pietravalle S., Norris K. & Butler
S. (2008)
Predicting the impact of future agricultural change and uptake of Entry
Level Stewardship on farmland birds. ISBN 978-1-906204-29-7
INTRODUCTION
- Environmental Stewardship (ES) was launched in England in 2005 with
three components; Higher Level Stewardship (HLS), Entry Level Stewardship
(ELS) and Organic Entry Level Stewardship (OELS). The work reported
here is intended to inform the ongoing review of ES but focuses on
ELS for which uptake data are more extensive. From the outset ELS
had four broad objectives (improving water and soil quality, enhancing
farmland wildlife, maintaining landscape character and protecting
historic environment). The conservation on farmland birds is central
to one of these objectives and the scheme is a key tool for the delivery
of the farmland bird PSA target. The work reported here has seven
distinct objectives: (i) to update the evidence base and identify
remaining knowledge gaps of resource requirements and causes of decline
of farmland birds; (ii) to review knowledge on practical measures
to improve quality of ELS options; (iii) to review knowledge on optimal
ways to deploy options in the landscape; (iv) to review knowledge
for past value of set-aside for farmland birds and potential impact
of its loss; (v) to assess the proportion of landscape required under
specific management to produce a population increase in relation to
current availability of such habitat under ELS; (vi) to predict likely
future trends in the farmland bird index in relation to a small number
of future scenarios; and, (vii) to make recommendations for future
research and highlight implications for the design and operational
delivery of the scheme. In the following summary we highlight, in
bold, key findings with research or policy (operational or practical)
implications.
- The first objective was to update the evidence base and identify
remaining knowledge gaps of resource requirements and causes of decline
of farmland birds. Comprehensive searches of literature databases
revealed over 70 new references of direct relevance (since a previous
review in 2001). Many of these related to five species; Lapwing Vanellus
vanellus, Skylark Alauda arvensis, Yellow Wagtail Motacilla
flava, Reed Bunting Emberiza schoeniclus and Corn
Bunting Milaria calandra, and most reinforced findings of
previous studies. The exception to this was Yellow Wagtail for which
a great deal more is now known about its ecology in arable farmland.
In particular, a mid-season shift in the habitat associations, from
cereal to potatoes, suggests a lack of suitable breeding habitat in
late summer may reduce productivity and contribute to the observed
population decline. There may be a need to develop agri-environment
scheme options that provide late summer nesting habitat for yellow
wagtails in a similar way to the in-field plot approach for Lapwing
and Skylark. In addition to single species studies, there
has been considerable research on birds in grassland habitats. This
has highlighted the importance of interactions between food abundance
(high in tall complex swards) and food accessibility (high in short
even swards) in grasslands. This work suggests the optimal foraging
habitat in grassland would be a mosaic of different sward heights.
The principle questions remaining to be resolved are how to
best deliver this sward heterogeneity in grassland through agri-environment
options, and at what spatial scale
- The aim of objective 2 was to identify possible modifications to
existing prescriptions that may improve their measurable benefits
for farmland birds. As for objective 1, this was addressed through
a literature search and expert opinion. The two most important
option modifications suggested are: (i) an extension of winter food
resource options (stubbles EF6, EG4, EG5 and wild bird cover crop
EF2, EF3, EG2) in to late winter (i.e. beyond February 15th); and,
(ii) management of grass margin options to increase structural heterogeneity,
particularly opening up the sward to improve accessibility of prey.
- Objective 3 reviewed current knowledge on the extent to which three
aspects of option deployment can influence the effectiveness with
which they deliver resources to birds: (i) extent or scale, (ii) configuration
(size and shape of options) and (iii) context (local and landscape
effects). There were very few studies to draw on that addressed
the issues of quantity, size, shape or the influence of local or landscape
context of habitat patches on effectiveness of resource delivery for
birds. The most valuable studies on scale (quantity or size
of patches) relate to providing winter food and suggest that, for
skylark, approximately 10ha of stubble per 1-km2 is needed to stem
the population decline and winter use will be maximised by creating
these resource patches more than 1km apart. The main effect of shape
is likely to be through influencing edge: area ratios on patches which
may be important in determining predation rates. The effect of context
on habitat use is complex; margins near good quality hedgerows may
offer higher value for birds, particularly in summer but hedgerows
and margins may also be sources of predators for ground nesters. At
the landscape scale, the value of a given habitat may be enhanced
by creating it where it also serves to enhance heterogeneity e.g.
arable areas in grassland. However, areas of high quality habitat
in poor quality landscapes may also serve as predator traps. Issues
such as the impact of scale, configuration and context on the effectiveness
of agri-environment scheme options could be addressed by linking the
monitoring of ELS with targeted research.
- Objective 4 was designed to investigate the potential impact the
loss of set aside on birds, as a habitat that provides nest sites
for some species and food in summer and winter for many species. We
related data on the population trends of the nineteen Farmland Bird
Index species to temporal changes in the availability of set-aside
for the period 1994-2005. The population trend for all Farmland Bird
Index species was significantly positively correlated with the availability
of set-aside in the same year, although individual trends were significant
for only two species, Lapwing and Linnet Carduelis cannabina.
A suite of other factors are known to affect farmland bird populations
and these preliminary analyses should be viewed only as indicative
of an effect of area of set aside. They merit further investigation
and more rigorous analyses, for example, to derive how extending the
time series back to 1988 and/or considering birds in different taxonomic,
or functional groups affects the direction and significance of correlations.
However, the results do provide the first suggestion of a potential
positive effect of set aside at the population level for widespread
farmland birds. The predicted loss of set aside in the near
future could therefore have a significant negative impact on the population
trends of several key farmland bird species and hence the Farmland
Bird Index as a whole. There are several options within ELS designed
to provide the winter food benefits of set-aside but few options would
provide the equivalent foraging and nesting habitat in summer. Consideration
should be given to options such as summer fallow to mitigate some
of the potential negative impacts of the future loss of set-aside.
The scheme design needs to be altered to encourage greater take up
of in-field options if ELS is going to provide winter foraging habitat
on the same scale as set-aside.
- The success with which ELS delivers the Farmland Bird PSA target
will be largely determined by the extent of uptake of key options.
However, few empirical studies have attempted to quantify the area
of a given option required to have an effect at the population level.
Under objective 5, we used a theoretical approach, with three stages,
to attempt to address this issue. First, population models were used
to determine: (i) the magnitude of change required in the key demographic
rate (i.e. the factor limiting population growth) to result in a small
(1%) annual population increase; (ii) how the magnitude of these parameters
varied if only a proportion of the population is affected. The smaller
the proportion of the population affected, the greater the increase
in the demographic rate required. The lower limit to this proportion
is set by the maximum possible value of the demographic parameter
concerned. Assuming the ‘proportion of the population’
to be roughly equivalent to the ‘proportion of farmland’,
this provides an estimate of the minimum area of farmland required
under relevant options to effect a population increase. Next, each
ELS option was scored according to whether they would definitely (good
scientific evidence) or potentially (less evidence) deliver summer
food, winter food or nest sites for each species. Using ELS uptake
data for November 2006, the total area of farmland that had options
affecting the key parameter for each species was then calculated.
Finally, this information was combined to ask (i) is current ELS uptake
sufficient and, if not, (ii) will it be sufficient under the predicted
future 70% uptake of ELS?
- The procedure outlined in (6) was carried out for 12 species with
an annual population growth rate of <1% (Kestrel Falco tinnunclus,
Grey Partridge Perdix perdix, Lapwing, Turtle Dove Streptopelia
turtur, Starling Sturnus vulgaris, Skylark, Yellow Wagtail,
Tree Sparrow Passer montanus, Linnet, Yellowhammer Emberiza
citrinella, Reed Bunting, Corn Bunting) and suggests mixed results.
For seven species – Lapwing, Turtle Dove, Skylark, Yellow
Wagtail, Starling, Linnet and Yellowhammer, the current uptake of
ELS options is insufficient to effect a population growth; for one,
Grey Partridge it is close to sufficient; for two, Kestrel and Tree
Sparrow, it is sufficient; and, for two, Reed Bunting and Corn Bunting,
results are mixed (The latter are limited by more than one
demographic factor and uptake was insufficient or sufficient/approaching
sufficient depending on the factor concerned).
- Under a predicted 70% uptake rate for ELS the area under
relevant options would be insufficient for five species - Lapwing,
Turtle Dove, Yellow Wagtail, Linnet and Yellowhammer, but
it is predicted to become sufficient or nearly sufficient for Grey
Partridge, Skylark and Starling (with a mixed outcome for Corn Bunting
and Reed Bunting).
- Ranking species in terms of the likelihood of ELS achieving 1% growth
suggested Turtle Dove may be the species least likely to benefit
from ELS options. Yellow Wagtail, Lapwing and Linnet were also predicted
to be relatively poorly served by ELS. For two species, Corn
Bunting and Reed Bunting, there were two limiting factors and for
each species ELS option uptake was sufficient or nearly sufficient
for one but not for the other. In general the shortfalls are
caused by a lack of certainty about the extent to which grassland
options deliver food and/or nest sites and low uptake of in-field
options.
- There are a number of important caveats associated with this modelling
procedure. For example, it may have ‘under estimated’
the effectiveness of ELS in several ways. First, ELS options were
considered to have an effect on the key parameter only where such
evidence existed in the literature. If potential effects, based on
general ecological knowledge rather than published evidence, were
also included, the proportion of farm area under a relevant ELS options
increased substantially, although they were still insufficient for
Corn Bunting and, to a lesser extent, Grey Partridge, Turtle Dove,
Skylark, Yellow Wagtail, Starling. Second, the models assume only
a single key parameter will be affected and other parameters stay
constant, whereas it is likely that that management options that have
a primary effect on the key parameter are also likely to affect other
demographic parameters. Adjusting models slightly for this again resulted
in more achievable targets for some species, but not for Lapwing,
Turtle Dove, Yellow Wagtail, Skylark, Linnet, Yellowhammer and Corn
Bunting. The effectiveness of ELS could also be over estimated by
assuming that ELS options deliver increases in the limiting demographic
rate up to the maximum rate. If this is not true (which seems likely
at least in some cases) then the area of land required is greater
than predicted here and fewer species targets may be achievable.
- This novel approach to assessing how much habitat is required to
deliver population recovery in relation to current and predicted uptake
of ELS should be viewed as a guide. It may, therefore, be more informative
to consider the broad patterns, rather than the results for each individual
species. The two key findings in this respect are: (i) the
lack of certainty of delivery through grassland options – this
reflects a lack of research which is to some extent already being
addressed through a number of Defra-NE funded projects; and, (ii)
the lack of uptake of in-field options and perhaps to a lesser extent,
the more complex/demanding field margin and boundary options. Over
and above these broad patterns, we consider four species, Turtle Dove,
Yellow Wagtail, Lapwing and Linnet, to be poorly served by ELS. Turtle
Dove requires seed-rich foraging habitats in the breeding season (e.g.
conservation headlands) which have a poor uptake. Yellow Wagtail and
Lapwing both require similar in-field nesting habitat options that
are unpopular in terms of uptake. For Linnet however, poor forecast
delivery by ELS is largely a result of the uncertainty of the value
of grassland options.
- Objective 6 was designed to explore the potential value of a new
and novel risk assessment model in predicting the impact, on farmland
birds, of policy reforms likely to cause land-use change. Two land-use
change scenarios were explored: a) predicted declines in the area
sown to (spring) barley and sugar beet in favour of winter wheat and
oil seed rape, with increased block cropping of these simplified rotations
and b) increased growth of biofuel crops on set-aside land. The risk
assessment process has three stages. Forty-two of the species included
in these analyses, including 16 of the 19 species in the farmland
bird index, either forage and/or nest in the cropped area of arable
fields and are susceptible to exposure to these changes. If either
of these land-use changes occurred, three species, Meadow Pipit Anthus
pratensis, Wood Pigeon Columba palumbus and Kestrel
are predicted to be re-classified to a less favourable conservation
status. If both an increase in block cropping and an increase in the
growth of biofuel crops took place, seven species are predicted to
be re-classified to a less favourable conservation status, with Meadow
Pipit, Kestrel and Barn Owl Tyto alba all predicted to become
red-listed. It should be highlighted that these predictions assume
UK-wide land-use change and therefore exposure of the whole population
of vulnerable species to these hazards.
- Implications and recommendations for policy recommendations.
• the overall quantity of farmland entering the scheme
is broadly on track and is not the main reason why the scheme should
not be potentially successful
• There is scope for improving the quality
of some individual options, eg extending duration of options delivering
seed food in winter and increasing structural heterogeneity of grass
margin options
• The biggest problem is the pattern of uptake,
i.e. the mix of options chosen. The popularity of some and unpopularity
of others results in gaps in the resource provision necessary to help
certain species sufficiently. This is already evident and seems unlikely
to improve with increased overall uptake. There is a need to find
ways of ‘rebalancing’ option uptake. The most important
options to ‘promote’ are in field options such as Skylark
plots, conservation headlands and stubbles and the more ‘complex’
field edge options such as enhanced hedgerow management. The number
of farms, and hence area of land required, with these options may
be reduced by targeting them towards certain options in particular
geographical areas or habitats. This is particularly true for species
with more restricted distributions such as Turtle Dove and Yellow
Wagtail.
• The loss of set-aside is potentially a major
risk factor to the PSA target and at present it is hard to see how
ELS options, however well designed, could provide sufficient quantity
and distribution of resource to replace it. This is particularly true
for the summer fallow nesting and foraging habitat it provides, but
stubble options may replace the winter food value of set aside. It
should be noted that if the ELS options designed to ‘replace’
set-aside deliver high quality resources the area required is likely
to be considerably less than that of set-aside.
- If several species are not well served by the ELS, it is almost
certainly not succeeding in its underlying biodiversity objective
to ‘improve the health of the whole ecosystem(s)’, In
this context, it is important to stress that there are several
species that are (a) still declining and (b) unlikely to be helped
sufficiently by the present pattern of uptake even at higher volume.
The level of benefits depends on option quality, option quantity and
delivery, but the option mix is probably the most important variable
with the current uptake patterns. Overall, quantity seems to be on
track in terms of numbers of farms entering the scheme and although
there is some scope for improving the quality (i.e. effectiveness)
of individual options, this is secondary compared to the overall imbalance
in terms of options chosen. In particular it is vital to promote higher
uptake of in-field and more complex margin/boundary options and to
continue to enhance knowledge of delivery in grassland landscapes.
 Back
to Research Reports 455 onwards
|
