POLLINATORS POLLINATION AND FOOD PRODUCTION

THE ASSESSMENT REPORT ON POLLINATORS, POLLINATION AND FOOD PRODUCTION
support tools). This would be a valuable development,
as some of the other modelling platforms place more
emphasis on non-economic values and different groups
of beneficiaries. For example, the Artificial Intelligence for
Ecosystem Services modelling framework (ARIES; http://
www.ariesonline.org) maps ecosystem service flows
with an emphasis on the beneficiaries of each service.
Pollination is suggested as a service suitable for ARIES
modelling (Villa et al., 2014), but to our knowledge this has
not been developed.
Spatially-explicit modelling of bee nesting and foraging
resources in agricultural landscapes was used by Rands
and Whitney (2011) to show that increasing the width of field
margins would provide more food resources to wild bees
whatever their foraging range.
6.5.1.10.2 Other modelling techniques
Various modelling techniques have been used to predict
effects of future land-use change and climate change and
on pollinators or pollination demand (see sections 2.1.1
and 2.5.2.3 respectively). These could provide information
to inform crop management or conservation decisions,
but we know of no specific examples where they have.
For example Giannini et al. (2013) showed a substantial
reduction and northward shift in the areas suitable for
passion fruit pollinators in mid-Western Brazil by 2050. This
information could be used by the passion fruit industry to
target conservation effort for these pollinators and their food
plants, although there is no evidence it has been used for
this purpose.
Population dynamic models have been built for honey
bees (for example, DeGrandi Hoffman et al., 1989). An
integrated model of honey bee colony dynamics that
includes interactions with external influences such as
landscape-scale forage provision has recently been
developed (Becher et al., 2014), which accurately generates
results of previous honey bee experiments. Bryden et
al. (2013) used a dynamic bumble bee colony model to
demonstrate multiple possible outcomes (success or
failure) in response to sublethal stress from exposure to
neonicotinoids, while a spatially-explicit model of individual
solitary bee foraging behaviour has recently been developed
(Everaars and Dormann, 2015). All these models have great
potential to be used for testing effects on bees of different
mitigation options, such as enhancing floral resources in the
landscape, or reducing pesticide exposures.
A stochastic economic model was employed to quantify the
potential cost of Varroa mites arriving in Australia, in terms
of lost crop yields to due reduced pollination (Cook et al.,
2007). This model has been used as a guide to how much
the Government should spend trying to delay the arrival of
Varroa (Commonwealth of Australia, 2011).
6.5.1.11 Participatory integrated
assessment and scenario building
Participatory Integrated Assessment involves a range
of stakeholders in scenario building or use of models to
consider and decide on complex environmental problems.
Its techniques have been extensively used in climate-change
policy development at local and regional levels (Salter et
al., 2010) and are sometimes used to develop scenarios
for multi-criteria analysis. The underlying assumption is that
participation improves the assessment, and the final decision.
Salter et al. (2010) provide a review of methods and issues.
Future scenarios were built using a deliberative approach
by the Millennium Ecosystem Assessment and UK National
Ecosystem Assessment (Haines-Young et al., 2011). Those
from the UK NEA were used to develop pollination futures
to 2025 in a recent assessment of evidence for the UK
Government (Vanbergen et al., 2014).
6.5.1.12 Decision support tools
Decision support tools are increasingly being used in
environmental management to help decision-making
(Laniak et al., 2013). They are distinct from the analytical
mapping and modelling tools discussed above because
they are designed around a particular decision or decisionmaking
context, and ideally developed collaboratively
with end-users. Most decision support tools are software
based, and assist with decisions by illustrating possible
outcomes visually or numerically, or leading users through
logical decision steps (see section 4.6.3 for an example of
stepwise decision trees). Some rely on complex models,
only operable by their developers (see Modelling pollinators
and pollination). Others have simple interfaces designed
to be used by non-experts. Costs are variable, but can be
relatively high (Dicks et al., 2014a).
A variety of decision support tools have emerged for
systematic assessment of ecosystem services, in order to
examine trade-offs and assist policy decisions. Bagstad et
al. (2013) identified 17 different tools, ranging from detailed
modelling and mapping tools (including InVEST, discussed
in Models for mapping the pollination above) to low-cost
qualitative screening tools developed for business, such
as the Ecosystem Services Review (Hanson et al., 2012),
and others have been developed since then. Many include
carbon storage, sediment deposition, water supply and the
scenic beauty of landscapes, among other services. Only
a few such tools currently include pollination (for example,
InVEST, Envision [using the InVEST pollination module (Guzy
et al., 2008)] Ecometrix and the Ecosystem Services Review).
The Ecosystem Services Review includes pollination as one
of a list of 31 possible goods and services, and business
425
6. RESPONSES TO RISKS AND OPPORTUNITIES ASSOCIATED
WITH POLLINATORS AND POLLINATION