Haines-Young, Roy and Tratalos, Jamie and Birkinshaw, Stephen J. and Butler, Simon and Gosling, Simon N. and Hull, Stephen and Kass, Gary and Lewis, Elizabeth and Lum, Richard and Norris, Ken and Potschin, Marion and Walmsley, Suzannah
UK National Ecosystem Assessment Follow-on. Work package 7: Operationalising scenarios in the UK National Ecosystem Assessment Follow-on.
Study aims and approach
An aim of the UK NEA Follow-on (UK NEAFO) is to develop and communicate the evidence base of
the UK NEA and make it relevant to decision and policy making. It also provides an important
opportunity for those working on scenario methods and concepts to scrutinise the role of futures
thinking in the management of ecosystem services and so develop their effectiveness as decision
support tools. In this study we have therefore asked: how can the UK NEA scenarios help us to
understand, manage and communicate the consequences of changes in ecosystem services across all
There are many different understandings about what scenarios are, and what they should be used
for. To clarify the issues surrounding the role of scenarios, we have approached this work from two
angles. We have firstly looked at the way the storylines can support decision making processes.
Secondly, we have looked at the content of the scenarios themselves and explored how through the
use of models the UK NEA scenarios as products might be refined to enhance their value as
Scenarios in Action
We used the opportunity of a series of meetings with stakeholders to develop the UK NEA scenarios
from a process perspective. These meetings took various forms, but throughout the main aim was to
find out whether people found the scenarios sufficiently believable, challenging and relevant.
In workshops organised by the scenario team in Leeds, Edinburgh and Belfast, we worked with
participants on a series of tasks designed to help them immerse themselves in the scenarios and
reflect on them critically. While those we worked with had many comments about the scenarios in
detail, the evidence we collected from these meetings suggests that the majority of people found
the scenarios to be plausible and the projections consistent. The majority also agreed with the
proposition that the suite of scenarios as a whole addressed a relevant ranges of issues.
We explored with the workshop participants several ways in which the storylines could be enriched,
by: developing the narrative about the way people might live in the different scenario worlds;
developing time-lines for the scenarios; thinking more deeply about regional and local differences;
and, exploring how the scenarios would frustrate or facilitate the embedding of the ecosystem
approach in decision making. We found that while all of these elements had value in terms of
stimulating discussion and understanding of the scenarios, they were not needed in order to address
deficiencies in the original storylines in terms of plausibility or credibility.
The evidence we collected therefore suggests that the existing narratives are probably sufficient as
an entry point for discussions about the future of ecosystem services in the UK. What was apparent
from the observations that we made in the workshops was that it would probably be a mistake to
‘over-engineer’ or ‘over-specify’ the narratives because there needs to be room for discussion and
probing. We were struck how people took the existing scenarios and found new features and ideas
in them than had not been identified by in the original work. For example, in one session National
Security, with its emphasis on resource efficiency, was found to be ‘greener’ than it initially looks. In
another Local Stewardship was discovered to need some degree of central control and regulation to
work efficiently. These kinds of discussion are evidence of the reflection, deliberation and social
learning that can be promoted by using the UK NEA scenarios.
UK NEAFO Work Package 7: Operationalising scenarios
In the workshop we organised in Belfast we found that the presentation of the scenarios could be
tailored to a specific region (i.e. Northern Ireland) and, through area-specific breakout groups during
the workshop, to specific localities within it. However, our experiences here emphasised the need
for considerable preparation, consultation with the stakeholder community, and changing of the
workshop format to make the scenarios intelligible and engaging to local stakeholders.
Work on the use of the scenarios in a more explicit decision support role will be reported via the
work on response options (WP8), which considered how they could be used to ‘stress-test’ policy
response options. The experience gained from the work undertaken in the early stages of UK NEAFO
was that the scenarios appeared to provide a suitable platform for the work, but that the stresstesting
methodology needed to be refined. During the follow-on we have also interviewed policy
leads in Defra, for example, to gain a better picture of policy needs, and the way scenarios might
usefully serve them. Apart from the challenge of ‘relevance’ it is clear that the time needed for
people to work with scenarios probably means that they are less useful to policy customers in the
context of their everyday work but can be useful at a very broad and strategic level. However, there
is clearly an opportunity for scenarios to be used more extensively through commissioned work. The
importance of commissioned work has been emphasised during the follow-on phase by invitations
to observe the work of the CAMERAS1 work in Scotland, and the Noise Study being undertaken for
Defra. Both are actively using the UK NEA scenarios. The outcomes of these on-going studies will be
reported elsewhere by others. Nevertheless, even though these projects are at a preliminary stage
they help us better understand how scenarios can be used to communicate the consequences of
changes in ecosystem services to different groups and individuals.
Scenarios as products: developing the model base
The UK NEA scenarios were initially used to make both qualitative and quantitative projections. The
quantitative work mainly involved modelling how land cover would change under the different
scenarios (Haines-Young et al. 2011). Although these data were used to make an analysis of the
changes in marginal economic values for some ecosystem services during the initial phase of the UK
NEA, they have not been fully exploited. At the time it was recognised that there were many gaps in
our understanding of the links between land cover and ecosystem services; UK NEAFO has provided
the opportunity to address some of these deficiencies.
Thus in the follow-on work we have sought to extend the range of models that can be used to
explore the UK NEA scenarios. The modelling work has not sought to change the scenarios
fundamentally, but to enrich the insights that can be derived from exploring the differences
between them in a systematic, and quantitative way. The goal, has been to extend the analysis that
can be built up around the narratives and hence enrich the scenarios as ‘products’. Four topic areas
were selected as the focus for this work: flood and drought risk (based on an analysis of changes in
river flows), biodiversity (farmland birds), marine and cultural ecosystem services.
We looked at the effects of land-use change on river flows under each of the UK NEA scenarios. We
modelled hydrological discharge within 34 UK catchments and calculated four hydrological indicators
for each catchment: average annual discharge, flood hazard, and Q5 and Q95 (measures of the
magnitude of unusually high (Q5) and unusually low (Q95) flows). For our flood hazard indicator we
calculated the interval between floods of a size currently occurring every 30 years. Although we kept
climate constant in the models, as we wanted to isolate the effects of land cover change, we ran
them for both the high and low climate change land cover variants for each scenario.
1 A Coordinated Agenda for Marine, Environment and Rural Affairs Science, 2011-2016.
In general, the ‘green’ scenarios, Nature@Work and Green and Pleasant Land, as well as National
Security, were associated with lower flows than currently occur (when measured using any of the
four indicators). However, for a given scenario there was a great deal of variability between
catchments in terms of the size and statistical significance of the differences. The magnitude of
change across all scenarios and catchments ranged from -13% to 6% for average annual discharge,
-14 to 7% for Q5, -24 to 27 % for Q95 and -16 to 36 years for flood hazard. Differences were
particularly evident between Nature@Work and World Markets, with the latter associated with
higher flows than occur currently, and the majority of the statistically significant increased flows.
Some catchments showed significant changes that were different in sign between these two
Taken together, our results indicate that that in managing change a balance needs to be struck
between alleviating the likelihoods of increased drought and increased flooding, depending on the
likely effects of these phenomena in the catchment.
We looked at the relationship between land use data produced during the first phase of the UK NEA
and models of farmland bird populations, in 1kmx1km squares covered by the Breeding Bird Survey
(BBS) and Winter Farmland Bird Survey (WFBS).
We used Functional Space Models to estimate the annual population growth rate under each
scenario of each of the 19 farmland bird species used to calculate the farmland bird index (Gregory
et al. 2004). We used this to look at the relationship between land use under the scenarios and: i)
the average population growth rate for all 19 species, and ii) a subset of 11 species showing
declining population trends under current land use. Overall we found that land use change across
the scenarios had relatively little impact. However, the only statistically significant change was for
declining species under Green and Pleasant Land, where population growth rates became
significantly more negative.
We used Mechanistic Models to estimate the number of over-winter ‘bird-days’ for two types of
seed-eating farmland birds, a yellowhammer-type and linnet-type These species were chosen
because they differ in their food preferences with respect to cereal, oil and weed seeds, but
between them are representative of the diversity of seed-eating farmland birds as a whole. We
found a significant decline in the ecological value of lowland agricultural areas for these species
across all UK NEA scenarios, but the greatest impact was for scenarios with the highest monetised
values for ecosystem services, as measured by the first phase of the UK NEA (Nature@Work, Green
and Pleasant Land). This appears to be due to the fact that, compared with the baseline, the area of
arable crops declines most sharply under these scenarios, due partly to changes in land use but also
because of conversion of arable land to other habitats important for ecosystem services (e.g.
Taken together these results imply a trade-off between overall value for ecosystem services and
conservation of farmland birds, and highlight the need to consider the specific impacts of land use
change on biodiversity, alongside other ecosystem services.
Marine ecosystem services
Only a limited attempt was made to model marine ecosystem services during the first phase of the
UK NEA. In the follow-on we have conducted preliminary work to produce spatially explicit models
for three important marine ecosystem services: fisheries landings, aquaculture production and
carbon sequestration. We made comparisons between baseline data and time slices for 2015, 2030
UK NEAFO Work Package 7: Operationalising scenarios
and 2060 under four of the UK NEA scenarios that were considered most relevant for the sector, and
mapped these across UK territorial waters.
There is a high degree of uncertainty associated with the models, mainly due to a lack of suitable
data and poor knowledge of the drivers of change. In many cases, in the absence of robust
quantitative models, we needed to take the qualitative descriptions of the UK NEA scenarios and
combine these with expert knowledge to estimate changes in the three types of ecosystem service.
We estimated that in three of the four scenarios: Nature@Work, Local Stewardship and National
Security, fisheries landings would be, by 2060 only slightly lower or at higher levels than they are
today. Under World Markets, however, projected landings would decline significantly by 2060, due
to a lack of regulation combined with high levels of investment from private capital. In the light of
this, it was interesting that aquaculture was at higher levels under World Markets than under any of
the other scenarios, although all of them showed higher levels than the baseline. This was because
under this scenario more investment capital would be available to invest in fish farms.
We believe that carbon sequestration would be most likely to be impacted by the World Markets
and Natural Security, due to higher CO2 emissions causing an increase in ocean acidification.
Our results, although tentative, mark a significant first step in attempts to map and project the
impact of possible future change on marine ecosystem services.
Cultural Ecosystem Services
In the first phase of the UK NEA, the relationship between the drivers of change and cultural
ecosystem services (CES) was mainly explored through the impact they had on land cover. For UK
NEAFO, we additionally used the Monitor of Engagement for the Natural Environment (MENE)
dataset. We examined how the UK NEA scenarios can be used as a framework to explore the
relationship between the supply of cultural spaces in the landscape and peoples preferences for
different types of natural spaces and practices in them. We have developed a Bayesian Belief
Network (BBN) that allows users to explore these relationships interactively and look at the potential
impacts of changes socio-demographic structure of the kind described by the UK NEA scenarios.
Our spatial analysis of the MENE data showed that people tend to select locations with higher
woodland cover than the average for the surroundings, when they travel intermediate distances
from their home, but that this tendency declines when they travel longer distances. Woodland cover
is projected to double under both Nature@Work and Green and Pleasant Land, and both provide
more opportunities to visit woodland close to home than under scenarios such as World Markets.
However, our analysis shows that on the basis of the current geography of people and woodlands,
the way planting is targeting under Green and Pleasant Land has the potential to deliver greater joint
benefits from biodiversity change and cultural ecosystem services than Nature@Work.
The BBN we have developed using the HUGIN Expert software allows the relationships within the
MENE data to be explored interactively; it is hosted on a prototype website that is open to the wider
community. By examining the relationships between socio-demographic characteristics of the MENE
respondents, the types of natural spaces they visit and the activities they do in them, this BBN tool
allows users to explore the impacts of possible future change on the supply and demand of CES.
How can plausible future scenarios help understand, manage and communicate the consequences of
changes in ecosystem services across all scales? In this work we have shown that they can be used to
promote understanding by the deliberative processes that they engender. The UK NEA scenarios
appear to be sufficiently rich and comprehensive to support debate across a wide range of topic
areas relevant to current policy concerns. The scenarios can also help understanding by providing a
framework in which current models can be applied and the outcome used both to test the
plausibility of the scenarios themselves and to deepen the insights that can be derived from them.
These analytical ‘scenario products’ can be equally important both in terms of deepening our
understanding of the assumptions on which the scenarios are built and in stimulating debate about
We have shown that the distinction between the ‘process’ and ‘product’ dimensions of scenario
thinking is a useful one, given the many ways scenarios can be used. The distinction clarifies some of
the different purposes and problems that scenarios work seeks to address. However, our work also
demonstrates that both components have their strengths, and neither can be taken isolation. If we
are to use scenarios to understand, manage and communicate the consequences of changes in
ecosystem services across different scales and in different contexts, then targeted analytical studies
developed within the qualitative framework of the UK NEA scenarios, can enrich our understanding
of today’s issues and how we might respond to them.
Actions (Archive Staff Only)