Downscaled Climate Models and Public Engagement

What role could downscaled climate change impact models have on people’s perceptions towards action

on climate change? Submitted by Andrew Merton Bell to the University of Exeter as a dissertation towards the degree of Master

of Science by advanced study in Sustainable Development, September 2009

I certify that all material in this dissertation which is not my own work has been identified with appropriate acknowledgement and referencing and I also certify that no material is included for which a degree has previously been conferred upon me��..............................

Abstract

Climate change is a global issue but requires action at all levels from the individual to

intergovernmental conventions. The actions required at a national and international

level often require a personal response from the community. Although reported

emissions have dropped by 16.4% since 1990 Domestic energy consumption has

increased by 8% since 1990 (Committee on Climate Change, 2009) and there has

been an increase from 79MtCO2eq by 4Mt CO2eq by the residential sector between

1990 and 2005 (Lorenzoni et al 2007). Therefore, one might suggest that the

community at large is not behaving in a way that is conscious about its carbon

footprint. A combination of a downscaled climate model and a lumped catchment

model has been developed for the Taw Torridge catchments in North Devon that

combines 1 m resolution grid data of soil properties and land-use. The data are

presented as an interim model to stakeholders from the general public who advice on

scenarios for the future. The research explores the communication issues associated

with the uncertainty of combined climate and hydrological models and whether the

presentation of impacts that are closer in geography to the participants changes their

behaviour. Qualitative results suggest that the models provide a focus for the people

who become animated about the choices they can make for the future. The model

provides a decision support tool for the possible land-use changes associated with the

different choices the participants make.

ii

Table of Contents

Abstract .............................................................................................................................................. i

List of Tables ................................................................................................................................. iv

List of Figures ............................................................................................................................... iv

Chapter 1 Introduction .................................................................................................................... 1

Section 1.1 Aims and Objectives .......................................................................................... 3

Chapter 2 Literature Review ........................................................................................................ 4

Section 2.1 Communication of climate change ............................................................. 4

Section 2.2 Combined climate and hydrological models. ......................................... 6

2.2.1 A review of methods for downscaling Regional Climate Models (RCM) ........................................................................................................................................... 6

2.2.2 A review of methods for modelling catchment-wide hydrological behaviour (including soil erosion) .................................................................................... 8

2.2.3 A review of methods of combining RCMs and Hydrological catchment models ................................................................................................................... 9

2.2.4 Incorporating the social dimension into scenario modelling ................ 11

Chapter 3 Methodology .............................................................................................................. 13

3.1 Outline Approach; ............................................................................................................. 13

3.2 Detailed Methodology ..................................................................................................... 13

Area of study .......................................................................................................................... 13

3.2.1 Sampling from the general public ..................................................................... 14

3.2.2 Design of Public Questionnaire ......................................................................... 14

3.2.3 Sampling from the Landowning community ................................................ 15

3.2.4 Analysis of Questionnaires .................................................................................. 16

3.2.5 Focus Groups ............................................................................................................. 16

3.3 Development of the Impacts model. ........................................................................ 17

3.3.1 River catchment behaviour .................................................................................. 17

Chapter 4 RESULTS AND ANALYSIS ................................................................................ 23

4.1 The Description of the population: ............................................................................ 23

4.2 The Environmental attitudes of the North Devon survey ............................... 25

4.3 Reported Behaviours ...................................................................................................... 29

4.4 Communication issues: .................................................................................................. 31

4.4.1 Supporting information from the Public Focus Group analysis and open questions responses. ............................................................................................. 34

4.5 Farmers Questionnaire and Focus Group ............................................................ 37

Focus group description: .................................................................................................. 38

Section 4.6 Model ..................................................................................................................... 40

iii

4.6.1 Development of the hydrological model: ....................................................... 40

4.6.2 Results of Climate Model Downscaling ......................................................... 42

4.6.3 Focus group results for scenarios ........................................................................ 46

4.6.4 Secondary Modelling results .................................................................................. 48

4.6.5 Focus Group responses to the process ............................................................... 50

Chapter 5 Discussion ................................................................................................................... 52

Limitations .................................................................................................................................... 52

Summary of main findings ...................................................................................................... 52

The focus group and the model ............................................................................................ 55

The communication issues ..................................................................................................... 56

Chapter 6 CONCLUSIONS ....................................................................................................... 58

Acknowledgements ....................................................................................................................... 60

Bibliography ...................................................................................................................................... 61

Appendices ....................................................................................................................................... 64

Appendix 1 Public Questionnaire. ..................................................................................... 64

Appendix 2 Land owner Questionnaire (map images reduced to save space) ........................................................................................................................................................... 76

Appendix 3 Localised Climate change model presentation .................................. 84

Appendix 4 Daily Rainfall distribution for 20 worst floods (source MIDAS and interpolated using inverse distance weighting) ........................................................... 90

iv

List of Tables Table 3-1 Summary of data and what we expect to learn from it ............................... 17 Table 3-2 Catchments used in the analysis of hydrographs for rainfall responses. First order rivers are the larger main rivers, 2nd and 3rd order are the tributaries branching from the main rivers. (the terms “greater” and “upper” have been inserted to assist definition in the analysis) ........................................................................................... 20 Table 3-3 Data used in developing the catchment response model defining the source data and transformations of the data. ........................................................................ 22 Table 4-1 Mann-Whitney test results for comparison of Gender, age group, urban rural classification of dwelling and social grade. ........................................................ 23 Table 4-2 Mann Whitney statistic test results for the questions relating to reported climate change knowledge between the survey samples in the DEFRA 2007 Survey of Public Attitudes and Behaviours toward the environment and the North Devon Survey....................................................................................................................... 24 Table 4-3Distribution of scores for accuracy of explanation of “40% chance of rain” . 31 Table 4-4 Accuracy score for the response to interpretation of 1 in 10 year flood protection .................................................................................................................. 32 Table 4-5 Advice move from Place “X” v Advise move from Place “A” Cross-tabulation ................................................................................................................................. 33 Table 4-6 Responses and terminology used to identify risk to health, economy and environment of the area 25 years from now. ............................................................. 35 Table 4-7 Size distribution of farms responding to questionnaire............................... 37 4-8 Catchments and their distribution in the sample after filtering extreme values for the Flood curve ......................................................................................................... 40 Table 4-9 Catchments and associated percentage of land-use derived from 1 km grid ITE land cover data. .................................................................................................. 42 Table 4-10 Table of land-use scenario changes ........................................................ 48 Table 5-1 Suggested terminology to define bands of probability when communicating with the public. .......................................................................................................... 56

List of Figures Figure 3-1 The study area, showing the extent of the UNESCO Biosphere Reserve. 14 Figure 3-2 Location of gauging stations and catchment boundaries used in the model. ................................................................................................................................. 21 Figure 4-1 Comparative graphs of the distribution of percentage of the samples from the 2007 Survey of Public Attitudes and Behaviours toward the environment and the North Devon survey .................................................................................................. 24 Figure 4-2 Relative priorities of respondents to the North Devon survey. This is a cumulative score of those issues in the top 3 of people’s concerns. .......................... 25 Figure 4-3 Aggregated NEP score derived from summing re-coded NEP questions. 26 Figure 4-4 Histogram of Response of question “Climate change is more important than a dip in the economy” (1=strongly agree, 5 strongly disagree) (n=64) ............... 27 Figure 4-5 Histogram of degree of agreement about personal behaviour and climate change ...................................................................................................................... 28 Figure 4-6 Histogram of degree of agreement about government total responsibility and climate change. .................................................................................................. 28 Figure 4-7 Proportion of respondents and their reported behaviour for climate change mitigation. ................................................................................................................. 29 Figure 4-8 Qualitative representation of potential carbon saving based on proportion of people reporting a certain behaviour multiplied by an index of carbon saving potential for that behaviour. ....................................................................................... 30 Figure 4-9 Frequency of terms used to describe 40% chance of rain ........................ 32

v

Figure 4-10 Histogram of terms used in describing a 1 in 100 year flood protection. . 33 Figure 4-11 Personal priority for concerns about climate impacts .............................. 36 Figure 4-12 Comparison of mean measured response time and modelled response time for time from base to peak flow in a rainfall event. ............................................. 41 Figure 4-13 Current expected monthly winter (DJF) Rainfall based on 61-90 LTA .... 43 Figure 4-14 Expected monthly winter (DJF) rainfall 2040-2059 ................................. 43 Figure 4-15 Figure 4 15 Current expected rainfall summer (JJA) 61-90 LTA (mm/month) .............................................................................................................. 44 Figure 4-16 Expected rainfall summer 2040-2059 (mm/month) ................................. 44 Figure 4-17 Expected mean temp of warmest July day current 61-90 LTA ................ 45 Figure 4-18 Expected mean temp of warmest July day 2040-2059 ........................... 45 Figure 4-19 Change in flood response time of rivers under new land-use scenarios . 49 Figure 4-20 Soil sensitivity in 2050. The most vulnerable 20% of grids are in yellow . 50

What role could downscaled climate change impact models have on people’s perceptions towards action on climate change?

Chapter 1 Introduction

Climate change is a global issue but requires action at all levels from the individual to

intergovernmental conventions. The actions required at a national and international

level often require a personal response from the community, such as the choice to

change behaviour in the selection of travel, energy providers, food choice and

responses to various new developments and not forgetting the expression of their

concerns within a democratic system. Although reported emissions have dropped by

16.4% since 1990 (Jackson et al 2008, DEFRA 2008), much of this has been gained

through rapid industrial measures on potent greenhouse gasses (such as methane

and hydrofluorocarbons) and latterly through emissions trading in the EU emission

trading scheme. Domestic energy consumption has increased by 8% since 1990

(Committee on Climate Change, 2009) and there has been an increase from

79MtCO2eq by 4Mt CO2eq by the residential sector between 1990 and 2005

(Lorenzoni et al 2007). Therefore, one might suggest that the community at large is

not behaving in a way that is conscious about its carbon footprint.

What are the reasons for this? Some authors believe that the human cognitive

response to threats is based on the perception of risk severity and probability

(Grothmann and Patt; 2005). Some businesses and people will not contemplate

threats beyond a 5 year time frame (Stoll and Kleemann, 2001).Unless there are

imminent extreme events, those perceptions of risk will not be high, there is little too

offer any stimulus for adaptive behaviour.

The evidence and the communication of climate science has been very prolific but

expressed mainly in very large scales of time and geography terms. These scenarios

may be on too great a scale to illicit a response from the community. If this is true,

then possibly a series of credible information presented to the community which make

the expected impacts more “real and relevant” to them yet still presenting a scenario

that is below the threshold of panic or fear, lead to a more constructive debate and

action for climate change adaptation and mitigation. Therefore, if we wish to illicit a

positive cognitive response and change in behaviour on climate change we need to

2

identify ways to express the science and the risk increasing the relevance to the

audience. This will help us to understand what motivates people to make these

changes. Justus et al (2007) advocate useful and relevant scientific data as a key

component to stakeholder engagement in developing a climate change policy.

A key question is how to communicate these scientific data. Dessai et al, have sought

to indentify how “deep uncertainty” can be expressed consistently. To date this has

been focussed on the scientific community and policy makers but not to the public at

large or even key stakeholders. Therefore modes of communication and engagement

need to be developed and tested that impart the climate change information that is

still credible and understandable.

This dissertation seeks to develop a communication or decision support tool for

climate impacts at a local scale and test its use as an engagement and planning tool

with a range of people from the general public and from the farming sector in a

UNESCO designated area in North Devon, UK.

In this document I will refine the aims and objectives of the study, present a review of

the literature relating to communication of climate change issues to the public and the

development of local scale climate impacts models. The report then details the

methodology of the data collection and its subsequent analysis. The results are then

presented, followed by a discussion and conclusion.

3

Section 1.1 Aims and Objectives

This dissertation combines social and natural science approaches to understanding

the potential of downscaled climate impact modelling as an interactive mechanism to

engage and encourage a positive behaviour change in the community. The model

developed can be a platform for further work on the development of local ecosystem

service modelling and to put in place scientifically sound planning policies. (Jonsson,

2005). The developed model will be a decision support tool to stimulate a deeper

consideration by the stakeholders.

The aim of the dissertation is to ascertain whether the regionalised model information

will stimulate a deeper concern over climate change within a selected group and

increase the likelihood of positive behaviour change towards mitigation and

adaptation.

The objectives for this dissertation are:

1. Identify the current attitudes and behaviours of the community in the study

area towards climate change.

2. Explore the range of terminology and presentation methods that will assist the

public to engage them is discussions about climate change and their

behaviour using the model as a stimulus for that discussion.

3. Develop a representation of the climate change predictions translated to a

local scale.

4. Develop a model of the response of the river systems to climate change and

changes in land-use through societal decisions.

The structure of the dissertation is as follows:

• a literature review of the subject areas that guides the work that will ultimately

be undertaken,

• a detailed account of the methods used in gathering and handling the data for

the research question

• an analysis of the results from the exercises followed by

• a discussion of the results

• conclusions for wider consideration

4

Chapter 2 Literature Review

This proposal relates to modelling the impacts on a local scale and communicating

about climate change, including those impact models. This literature review will

address firstly the communication issues and the modelling aspects secondly.

Section 2.1 Communication of climate change

The communication of climate change, its impacts, how to mitigate and how to adapt

are critical to the success of any policy measures that might be put in place to counter

the damage that might happen to people and ecosystems that might arise. Many of

these policies and programmes require that the public understand the consequences

of climate change either as a result of their own actions or of the authorities they elect

and appoint in democratic systems.

The issues of communication centre on 3 areas;

• Psychology of communicating climate change and motivation of the audience

towards positive action

• Understanding of the science of climate change and global warming

• Communication of “deep uncertainty” about the models and the science of

climate change

When communicating climate issues to stakeholders and the public, a key objective is

to motivate towards pro-environmental or pro-climate behaviours and towards

adaptation. Kollmuss and Agyeman (2002) identify the mechanisms for behavioural

change and the barriers to those changes. The paper is a good review of the various

mechanisms and culminates in complex sociological models. They identify key

barriers to true environmental behaviour. These included lack of environmental

awareness due to complexity of environmental issues, time lag between behaviour

and impact and remoteness from the impact. Another key factor cited is lack of

emotional investment with the issue; this can be because the issue does not coincide

with the audience’s values and beliefs and the issue does not directly impact on their

personal well-being. A key motivational factor is whether the individual can do

anything about the problem; will their actions count? This is seems to be as true for

adaptation as it is about mitigation. (Grothmann and Patt, 2005).

5

Moving from awareness to action involves several stages according to various

authors reviewed in Kollmuss and Angyeman (2002). Crucially for this project

proposal, they acknowledge the work of various authors (Riordan & Burrgess, Owens

and Bloomfield) that “deliberative and inclusionary procedures (aka DIPS) are an

essential ingredient in the development of more responsive decision making and for

transformative potential for wider governance.

The emotional factor has an optimum level to which it can be taken. Pure

informational campaigns are limited in their impact and in particular the tactic of shock

has its constraints. O’Neill and Nicholson-Cole (2008) reviewed whether fear was a

good approach as motivating emotion. Whilst fear can be a method of getting the

attention of the audience, it proves not to be a persistent way of sustaining

environmental behaviour. As is the nature of the climate change, the changes can

take too long or happen too far away to sustain the positive behaviour, unless there

have been some meteorological events re-enforce the belief. Furthermore, the fear

can have a tendency to paralyse people into sensation of futility or apathy. It is

suggested that a little bit of fear helps, particularly where solutions are also presented.

The “bottleneck” that stifles adaptive behaviour can also be caused by agents’ and

stakeholders’ under-estimation of their capacity to adapt. (Grothmann and Patt, 2005).

Lorenzoni et al (2007) in their extensive survey and subsequent focus groups

identified a similar range of issues for UK laypeople. Featuring in the list was

information presented in formats that were too difficult for the general public. They

comment that the uncertainty surrounding climate change science is compounded by

the individual’s uncertainty or lack of knowledge about the science. This suggests that

education about climate science in its own right would assist a more rational response

to climate change issues.

A major body of work on communicating climate change models and scenarios has

been done by Dessai, Hulme and various co-authors since 2005. In their 2008 paper,

they propose that downscaled models and scenarios such as UKCIP02 are the

product of a negotiation between policy makers, communicators, stakeholders and

(restricted part of) the science community. The scenarios presented are not

necessarily a scientific produce but a social construct. Such a product can only

contribute to the uncertainty and the public doubt over such models.

“Deep Uncertainty” is a strong characteristic of climate models due to the chaotic

nature of the earth’s atmospheric/ocean/earth interactions. Kandlikar et al (2004)

6

proposed a schema for communicating the deep uncertainty for outputs that can be

supported by a full probability density function that quantifies the certainty of any

statement being made and another approach for which a probability density function

(PDF) cannot be provided. This particular review of communicating uncertainty was

targeted towards the scientific community and not towards the general public, where

terms such as “orders of magnitude” are unlikely to be used correctly if at all. The next

step should be to define a methodology for communicating the uncertainty to the

public.

To summarise the communication and risk perception issues from the literature; the

overall problem for motivating people to behavioural change and communicating

climate change centres on the technical nature of the subject bewildering the

community. There are key issues associated with the deep uncertainty and the scale

(time and space) of the impacts and changes. The audience needs a consistent and

comprehendible framework for communicating the levels of uncertainty. There is a

need to engage people in understanding the issues, understand the possible impacts

(including their likelihoods) and present them with achievable actions to provide a

sense of control and empowerment. A strategy for increasing the understanding and

deepening the engagement is through a participatory planning process whereby the

science can be explained, the consequences of decisions can be reviewed and

optimal solutions identified that are ultimately enacted.

Section 2.2 Combined climate and hydrological models.

In this section I will review the literature of combining climate change and hydrological

models. From the outset it must be understood that the time and resource constraints

on this project will not permit the development of a brand new methodology. The

objective of this review is to identify the optimum methodology to develop a model that

will reflect the climate change on the currently observed hydrology and how that might

be used as an interactive decision support tool as part of the communication exercise.

Consideration will be given towards a methodology that might be transferable to other

areas.

2.2.1 A review of methods for downscaling Regional Climate Models (RCM)

The development of impact models has been very prolific over the last 10 years. The

range of models, from the large scale circulation models developed by CSIRO in

Australia, Hadley Centre in the UK amongst others, have dominated the scientific

7

press and have been the basis for much of the published information from the

Intergovernmental Panel on Climate Change (IPCC).

The ability to refine the models to a higher resolution has been achieved through

methods of downscaling based on methodologies such as dynamic, statistical

downscaling and change factor. Much of the published work on the methodology has

been done by Wilby. Guidelines for downscaling techniques have been produced for

the IPCC by (Wilby et al 2004).

Essentially the statistical downscaling will produce climate change parameters based

on the statistical point data provided from individual meteorological stations that

reflect the variations caused by local factors (such as topography, land-use and sea

proximity) and linked to a single General Circulation Model (GCM). Change factor is

done by applying the coarse output of a GCM to finer resolution historical

meteorological data. Again this is based on a single GCM but can be repeated for the

various possible outcomes and probabilities for a given GCM or indeed repeated with

other GCMs.

Referring to the application of modelling methodologies, Diaz-Nieto and Wilby (2005)

suggest the complementary application of change factor and statistical downscale

modelling techniques. These papers suggest methods to produce finer scale

resolutions whilst retaining scientific validity by defining probabilities and ranges of

potential outcomes.

Regional climate models are treated with great caution, in that they are derived from

the larger scale models that are known to have wide error margins associated with

them (Lahsen, 2005). The resulting model might appear to be regarded by the non-

specialist to have a degree of accuracy that cannot stand to scientific rigour.

However, the models have been generally performing within parameters over the last

decade, with the exception of precipitation on downscaled models (Dessai and Hulme

2008). Bell et al (2007) suggest that the models are capable of predicting extreme

rainfall events and their return periods. (Fowler et al 2005).The value of the model is

determined by its utility; if temperature is the main factor of interest, then the models

have proven to very worthwhile. Conversely, if one was mainly interested in

precipitation, the models will attract a degree of public criticism.

The explanation of this variation and the public interpretation of these outputs is

advocated by Dessai and Hulme (2008). For example the weakness of UKCIP02 data

8

is that it is based on the one large scale model output, and from one emission

scenario, that has then been interpolated for different emission scenarios. Therefore

what might be taken as a credible model, does not use the full range of data and other

outputs to support its validity. However, it moves towards a presentation of

information at geographical and temporal scales that stakeholders and policy makers

are seeking. UKCIP09 addresses this particular issue by offering a range of 3

emission scenarios (A1F1, A1, B1). Furthermore, the range of outcomes are

supported by a stated probability. These are generated by a system that has been

called “Perturbed Physics Ensemble” (PPE). In this methodology, the important (but

small scale parameters but whose scale of impact is not known precisely that can

determine the outcome of a model run), are altered to produce a range of possible

outcomes that build up a probability distribution of a certain climate variable.

(UKCIP09, August 09) The net result is a large matrix of possible outcomes based on

possible emission scenarios and each output grid having a range of probabilities for

each parameter. The result, although much more technically correct, is bewildering for

a general member of the public.

Further to the climate predictions, the UKCIP09 models include a weather generator,

which can produce up to 1000 (minimum 100) weather outputs for a given future year

for a particular 5 km grid square. These outputs include derived variables such as

Potential evapo-transpiration (PET) and are varied by the PPE set of variables as

described above. It is possible to state which of these variables can be altered. I

suggest that such a generation of weather possibilities will help to explain to the public

how a climate model incorporates variability and uncertainty.

2.2.2 A review of methods for modelling catchment-wide hydrological behaviour (including soil erosion)

Olsson and Pilesjo (2002) define methodologies to produce hydrological models

based on catchments categorised as stochastic and deterministic. The former will

produce, eventually, a probability distribution function for the behaviour of the

catchment given the statistical factors put into the model. The deterministic model

must be consistent in its results and can be classified as empirical models based on

statistical analysis of the phenomena and identifying correlations. This “black box

approach” does not require a full understanding of the physical processes. This

approach can be applied at the individual cell level or it can be done at an aggregated

level by looking at catchment behaviour as a whole. Spatially distributed data to

create a catchment model will include; climate parameters (such as precipitation,

9

cloud cover, daytime temperature, wind-speed, precipitation intensity & relative

humidity), soil properties (texture, permeability, hydraulic conductivity, infiltration

capacity and velocity), vegetation cover, land-use and topography. From these

parameters it is possible to identify the hydrologically effective rainfall (HER), overland

flows, sub-surface flow, groundwater flow and channel flow in the rivers. With a lack of

data and the time to gather what there is, the development of a model starting form

basic processes is not going to be achieved in the time permitted for this study. Later

in this document it will be shown that this has been done for the UK though various

projects.

Olsson and Pilesjo (2002) define a difference between distributed hydrological models

and those that aggregate the data for each catchment known as lumped models. The

former are used to identify changes within any particular catchment in particular land-

use, the latter tend to be used for rainfall and run-off models.

TOPMODEL is one of many models that have been produced but has been

successful because of its relative simplicity using a slope index based on the tangent

of the gradient Beven (1997). Mendoza et al (2002) classify spatial hydrology models

into 4 categories; these are: 1. Integrate hydrology and GIS, 2. Integrate hydrology

and remote sensing, 3. Integrate hydrology, remote sensing and GIS, and 4

Integration of geomorphological knowledge with remote sensing and GIS. This

demonstrates a number of routes to providing an hydrological prediction depending

on the data available. The benefit of these varied approaches is that it can help to

support management decision for catchments where there may be a lack of direct

data to guide decision makers. Brazier (2004) points out that modelling should be no

substitute for real data and, indeed, Mendoza suggests that each model developed

requires calibration sites and therefore transferral of model to an unknown catchment

must be treated with caution.

2.2.3 A review of methods of combining RCMs and Hydrological catchment models

Various authors have now begun to incorporate climate models into river basin

hydrological models.Yu-Pin(2007), Bronstert et al, 2002, Bell et al 2007 offer some

good examples. A degree of the focus of this work has been targeted towards water

resource management and disaster prediction, such as floods and how to handle

droughts. Dessai and Hulme, 2008, suggest that only very recently has the

anthropogenic signal been detected in rainfall at a global level, the remaining rainfall

10

patterns are attributed to normal variation, or possibly through other climate

interactions not yet included in the models. Bell et al (2007) suggest that the

precipitation predictions fromUKCIP02 data provides reasonable accuracy for return

periods between 2 and 20 years, and reports accuracy of up to 50 years.

Furthermore the literature suggests that the development of a pareto distribution of

return periods for floods can only provide reasonable information for return periods up

to 15 years when incorporating the RCM model outputs. (Kay et al, 2006)

To date there has not been a local impacts model for the catchment of the Taw

Torridge catchment basin. Bell et al (2007) incorporated the Taw as a suite of

catchments to assess the increase in flood risk arising from increased precipitation

identified in UKCIP02. In this circumstance long-term average data was used to

compile the catchment response to rainfall events without consideration to the large

variation of rainfall across the catchment.

The model development that most closely parallels the proposed project is that

devised by Bell at al (2007). This is known as the grid to grid (G2G) model which

incorporates downscaled region climate model data for rainfall, potential evapo-

transpiration (calculated from MORECS but closely matching the Penman-Monteith

equation), topographic digital terrain model to provide average slope and maximum

slop within the 1 Km grids.

The G2G model is relatively simple with only a few parameters that need to be

included in the input stage. The variation within each 1km grid is accommodated by a

variation factor determined by a weighting of mean gradient and maximum gradient

within each of the grid squares.

Fortunately there has been a body of work on modelling and to account for land-use

and rainfall for UK and their catchments to consider not only the hydrology but also

the associated processes, such as soil erosion. Two particular approaches are

MAGPIE and NEAP. Within MAGPIE (Lord & Anthony, 2000) the inputs ot the model

on a 1km grid distribution include data from the agricultural census; (crops, livestock

numbers) land-use types from Institute of Terrestrial Ecology, climate data from the

UK Met Office and soil physical properties. The MAGPIE framework is a GIS model

that was developed through the incorporation of lessons from field scale models

developed between 1971 and 1999. NEAP-N incorporates rainfall, hydrologically

effective rainfall (HER), soil properties as well as the land-use, stocking and

11

application of nitrates. This grid still provides a useful database for the application of

grid based GIS modelling. However the variation of soil properties across a 1Km grid

can be quite significant and will not be shown at this level.

Closely aligned to modelling hydrology and a climate impact is nitrogen leaching and

soil erosion potential of the water movement over land. The weakness of the NEAP-N

model is that it caters for current land management and farming practices. Fawcett,

Anthony and Silgram (2004) comment that as management practices change, the

model will need to be re-calibrated. Given that much of the nitrogen leaching is

coming from application of artificial fertilisers based on oil products, the practices will

and indeed already are changing significantly due to the price and ultimately the

availability of chemical fertilisers.

There have been reports produced for the Taw Torridge estuary on the impacts of

sea-level rise on the geomorphology of the estuary and including a simple change

factor (100% increase of maximum flow) for the fluvial component for the shape of the

residual channels (Pethick 2006). This has been used to inform the current production

of Shoreline Management Plans, as a tool for engagement with stakeholders and

planners.

2.2.4 Incorporating the social dimension into scenario modelling

Scenario development is not just a physical climate science discipline but also social

process that is influenced by the adaptive behaviour of society (Hulme & Dessai

2008). Therefore scenario planning is an iterative and feedback process that requires

clear communication with the audience, about probabilities, ranges and uncertainty

that is confounded by shifting goal posts as models improve and as society reacts

(positively and negatively) with policy and action.

The recommendation for engaging people about the impacts demands better

modelling of low probability, high impact events, such as floods, storm surges and

heat waves (Goodess et al 2003). These have the effect of presenting the risks that

have a scale large enough to warrant a response from the audience. However, this

can build in the fear factor to too high a degree as highlighted already in preceding

text. (Lorenzoni et al, 2007).

Models have a strong potential for stakeholder dialogue. Once developed, they are

useful tools in their own right but have a stronger role in developing participatory

12

planning scenarios and exploring integrated assessments (Tansey et al 2002, Olssen

et al 2005, Jonsson 2005). This has also been referred to earlier in this text with

regard to interactive models being part of a DIPS. However, implicit in this is the

development of models on a temporal and geographic scale that meet the needs of

the stakeholder, and include variables within the model that the stakeholders will have

some control over. These might include land-use, derived through a societal choice of

policies such as local purchasing, food security policies, development and land-use

policies, mitigation options for flood attenuation.

To summarise, I have reviewed an evolution of modelling in the recent years of the

very recent downscaled climate models and how all but the very recent UKCIP09 has

been incorporated into hydrological and catchment scale modelling. Each of these

models has been devised for specific technical purposes to assist management or

assessment of specific issues rather than a tool for public engagement. The literature

provides us with insights into the variables and how they are treated in such modelling

within a range of types of modelling suited for this exercise.

The development of the climate change impact model will focus on the response of

the catchment and its associated hydrographs and what this might mean for flooding

within the catchment area. This project will assist our understanding of communicating

climate change with a group of people who might not regard themselves as obvious

decision makers. The resulting work and methodology might be applicable to any area

in Europe or the developing world as a participatory planning tool and a method of

implementing the Ecosystem Approach as defined in the Convention on Biological

Diversity. (CBD,2009)

13

Chapter 3 Methodology The dissertation works in both social and natural science sectors, through quantitative

analysis response of climate data and physical data of a region, quantitative analysis

of questionnaires. Because the research will also investigate language, terminology

and a deeper understanding of values, the work will also involve a qualitative analysis

of transcripts from focus groups and sections from the questionnaire interviews.

3.1 Outline Approach;

The model of climate change impacts was developed that interpolates the recent

UKCIP09 predictions and will attempt to incorporate these factors under the range of

land-use scenarios developed.

The approach to the human sciences part of this project will involve recruiting 2

samples of population of the Taw Torridge basin. One cohort represents the wider

public and the second represents the landowning public along a transect descending

the River Taw from the upper catchment to the estuary. A focus group from each of

these samples was created. Each group was;

• Analysed further for their attitudes to risk and uncertainty

• Asked for a collective view on future scenarios supported by interim outputs of

the climate change impacts model.

3.2 Detailed Methodology

Area of study

The area for this study is the catchments of the rivers Taw and Torridge in northern

Devon, which forms a substantial part of the UNESCO Biosphere Reserve (see figure

3.1). This area has not benefited from such a study in the past, but the presence of

such as designation lends itself to providing a framework for such a programme and

any follow on work that might ensue. The total population in the Biosphere Reserve

area is around 160,000 people.

14

Figure 3-1 The study area, showing the extent of the UNESCO Biosphere Reserve.

3.2.1 Sampling from the general public

A target number of 100 responses were sought from the community population in the

Barnstaple area. Sampling methods included an online survey and a series of face to

face surveys. The latter was used to target areas to ensure a sample that was

representative of the local population could be found. Face to face surveys were

selected by door to door surveys in postal areas of Barnstaple that were likely to yield

“hard to reach sectors” of the community who may not normally respond to

questionnaires. Further face to face surveys were completed through opportunistic

sampling at events during the summer period, such as the North Devon Show.

3.2.2 Design of Public Questionnaire

This questionnaire was designed to ascertain attitudes towards climate change,

environmental disposition, and aspects of risk communication and assessment and

social grouping.

The survey on Public Attitudes to Environmental Issues (DEFRA, 2007) provided a

comparative data set from which to correlate the findings. Therefore questions that

helped to describe the sample were copied from that survey. These questions include

background information on age, social grouping, house ownership, household

15

income, and climate change behaviour in the household with regard to energy

conservation, water conservation and waste. These responses were recorded as tick

boxes or through Likert scale responses.

A section of the questionnaire was dedicated to gaining a level of comprehension

about climate change terminology, also replicated from the DEFRA 2007

Questionnaire. A standard set of questions was included to measure environmental

attitudes using the New Environmental Paradigm (NEP) standard questions and other

possible priority areas of concern to the individual.

The communication of risk questions included open ended questions that required

people to define the level of risk in their own words the impacts that climate change

might have on the region’s and their own financial/economic, health and

environmental well-being. This question was intended to draw out the terminology that

people use and give an indication of the comparative levels of importance between

these three factors to the respondent.

On understanding geographic representation of risk, the face to face respondents

were asked to advise the questioner whether he or she should move from a house

identified on 2 maps that showed flood risk represented in two ways. One map

showed the probability of flooding on a continuous spectrum of colour shading, the

second showed the probability divided by contour and contrasting shading. The

locations of the fictitious house on both maps were placed at exactly the same level of

probability of flooding.

The questionnaire concluded with further questions on respondent’s background and

willingness to participate in a focus group.

The public questionnaire can be found in Appendix 1.

3.2.3 Sampling from the Landowning community

Landowners were contacted within the catchment with a different self-completion

postal questionnaire. 250 self completion questionnaires were sent to farmers in three

areas of the Taw catchment. The database used was a register of livestock holders,

which was 5 years old. As a database of this type, the term “stock holders” also

includes households that had pet sheep or goats for example. Addresses that

appeared to be of this type were filtered out of the data base, prior to selection.

16

Questions similar to those used in the public questionnaire were used to gauge the

farming community’s disposition towards climate change and their behaviour either

primarily as a business and their communication and understanding of risk. It also

asked what changes the landowners had seen that had arisen from what they felt was

climate change.

Their perceptions on climate change and current and future land-use were sought

through a set of questions that were scored through a Likert scale of level of

agreement with certain statements. Individuals from this sample were asked to join a

second focus group to consider specifically the landownership and practical land-use

issues.

The landowners’ questionnaire and geographic sampling area can be found in

Appendix 2

3.2.4 Analysis of Questionnaires

The questionnaire results were compiled in an SPSS database and tested against the

national survey for a comparison of the populations using Student’s test.

Quantitative data unique to the survey was tested with appropriate tests for

nonparametric data (Mann Whitney, Chi-Squared) and correlation of parametric data

to establish trends within the community and their attitudes and actions towards

climate change. (Wheeler et al, 2004). Of particular interest are the comparisons

between reported behaviour and environmental attitudes set against the social and

economic background of the respondents and the identification of the potential for

more carbon reduction behaviours within the sample.

The qualitative responses within the questionnaire are given a score for the accuracy

of the response but also examined for frequency of certain terms and for any

particular underlying themes of the response. These were considered thematically

and for some consideration of the scale of impacts that people thought would be

apparent.

3.2.5 Focus Groups

Participants in the group were offered a minor payment as compensation. The group

size should be in the order of 10 individuals as recommended in the published works

(Goss 1996, Zeigler, 1996 and Burgess 1996). Each focus group was held in 2 parts;

17

the first part to deepen the questioning about the understanding of risk and

uncertainty and how they are communicated. The second to consider the downscaled

model of the UKCIP09 outputs to a local catchment scale, climate change and to

explore a number of scenarios of the type of community they think might happen in

the epochs of the 2020s, the 2050s and the 2080s years. The groups were asked to

give some opinion on the future land-uses they thought would be needed for their low

carbon future, in terms of food production and food security as well as energy

production.

Each focus group meeting was digitally recorded with the consent of participants and

the transcripts were analysed thematically according to the questions. Each group

meeting was approximately one and a half hours long. Data and its uses are

summarised in table 3-1.

Data Type Information expected

Questionnaire results Quantitative (Likert scale, ranking, scale data)

Current attitudes and behaviours, priority concerns for people, demographic background, social background.

Questionnaire results Qualitative data. (open questions)

Terminology for communicating risk and probability. A sense of current knowledge, particular topic areas of concern.

Focus Groups Qualitative data Deeper understanding of the terminology used and what people prefer. Understanding of risk perception associated with climate change.

Focus group Qualitative data for modelling

Direction that the group thinks scenarios should be taking.

Whether this work will influence their thinking on climate change

Table 3-1 Summary of data and what we expect to learn from it

3.3 Development of the Impacts model.

The area to be covered by the model will be the Taw Torridge catchment in North

Devon, consistent with the Biosphere Reserve. The modelling will be based on the

catchments upstream of the river gauges at Umberleigh on the Taw and Torrington

Sewage Treatment Works on the Torridge.(see Figure 3.2)

3.3.1 River catchment behaviour

River flow data was provided by the Environment Agency in the form of gauge data

18

from 7 gauging stations. Data for the 20 worst flood events was provided for flow rate

against time recorded at hourly intervals in the early data sets and at 15 minute

intervals for the more recent years.

Hydrographs were isolated for the 2 main rivers and the 5 sub-catchments using the

straight line method (Blume et al, 2007). This hydrograph was defined by the flood

response time, the recession time, baseline flow and peak flow for each of the flood

events.

The rainfall events causing these flood events were retrieved from the MIDAS

database, accessed through the British Atmospheric Data Centre

(http://badc.nerc.ac.uk/home/). Over 8000 daily rainfall records from this network of

rain gauge stations for the day and the day before each event were compiled and

interpolated across the catchments. Inverse distance weighting model was used so

that total rainfall was not overestimated for any event.

UK Climate Impacts Partnership (UKCIP) has produced smaller scale model outputs

based on the HADCM3 (300Km2) called UKCIP02 that is presented in a series of

smaller scale outputs at the 50 x 50 km grid scale. Very recently a series of climate

parameters have been published as UKCIP09 which down-scales the large-scale

model to 25 x25 km. These recent UKCIP09 outputs are incorporated into the model.

UKCIP09 presents the scenarios with a probability distribution function. The scenario

adopted for the model is the 50% range which depicts the most common result of the

downscaled model and used the high emissions scenario (SRES A1FI, based on 810

ppm CO2). Other scenarios were not likely to yield a greater change in the factors

being modelled and the target audience is one whose behaviour indicates a high

emissions future.

The UKCIP09 25km grid outputs for the SW river basin region were interpolated using

a Kriging method. It is acknowledged that this takes the UKCIP model beyond its

intended level of geographic accuracy; the justification was to develop a model as a

communication tool that had a reasonable degree of credibility. The UKCIP09 data is

presented on a grid that has a “rotated pole” projection. The majority of the other data

is presented as either a grid format or point data format with an OSGB projection. To

enable cross-interrogation and combination of the grids requires all data is presented

in the same orientation (at least for MapInfo ® and Vertical Mapper ®). Therefore the

grids are re-projected to a common format.

19

For change factor modelling, the known climatic parameter is increased (or

decreased) by the factor of change from the UKCIP09 data for each part of the grid.

E.g. If the model suggests a rainfall increase by 30% in winter months, the entire

rainfall winter grid of the long-term average data of 1961 to 1990 ( the baseline for the

climate models) is multiplied by 1.3 to give absolute numbers. For interval data, such

as temperature in degrees centigrade, the change was added to the whole grid. The

same change factor approach was done for relative humidity.

Each of these change factors vary across the grid according to the gridded

interpolation and therefore any transformations are executed as “grid to grid”.

It was intended to incorporate a flood risk map based on known historical food events

and their extent. However, the data is not available from the usual sources (South

West Water, EA or the local authorities) The development of flood risk assessment

currently being undertaken by the Environment Agency is not yet available. Level1 of

the flood risk study (North Devon Council and Torridge District Council, 2009

(prepared by the Environment Agency)) highlights the same lack of data.

The data provided for land use, aggregated farm census data, soil texture, nitrogen

leaching and erosion potential information was provided in 1 Km grid format.

Topographical data was provided by the Ordnance Survey though EDINA to give a

100m grid. Coefficients for the overland flows were adopted from Engel (2004).

A GIS layer of unknown origin was provided by the University of Exeter which, after

investigation and consideration was taken to be a cumulative flow path of the

catchments i.e. any point on the map indicated the sum of all lengths of travel for

water in the catchment to reach that point.

The gridded data was imported into “MapInfo ®” geographical information system

(GIS) software which included “Vertical Mapper ®” add-in to provide interpolation and

grid software.

The gridded outputs and feature outputs such as river flows, land-use were analysed

individually using SPSS for correlations and multivariate regression equations to build

the model equations. The 7 gauging station data sets allowed the analysis of 7

catchments as though they were separate because the sub-catchments were small in

comparison to the main.

20

The model was developed with incremental complexity. A “lumped model” (as

described by Olsson& Pilesjo 2002) was the initial target, with a follow on to identify

the land use for individual grids where time permitted.

Ist order river

catchment

2nd order river

catchment

3rd order river

catchment

Area (km2)

“Greater” Taw 84,618

“Upper” Taw 7,521

Yeo and Mole 33,007

Yeo 5597

“Greater” Torridge 66,542

Lew 6783

“Upper” Torridge 25,813

Table 3-2 Catchments used in the analysis of hydrographs for rainfall responses. First order rivers are the larger main rivers, 2

nd and 3

rd order are the tributaries branching from the main

rivers. (the terms “greater” and “upper” have been inserted to assist definition in the analysis)

21

Figure 3-2 Location of gauging stations and catchment boundaries used in the model.

Secondary modelling of these data was done to take account of land-use changes

that might be expected and indicated by the focus groups. For example, more local

food security resulting in more mixed farming including arable, which implies a land-

use change and therefore a change in the hydrological properties of the catchment.

The impact on the hydrographs for the rivers was the primary output.

A summary of the database used for the modelling is shown in table 3.3.

22

Input Current data Process Output

Current LTA data for rainfall, humidity, temperatures

Long term average 61-90 UK Met-Office

Download gridded dataset

Rainfall change UKCIP09

Grid of rainfall UK Met-Office

Change Factor Expected rainfall

Temperature change

UKCIP09

Factor change from Downscaled model

Change factor

Temperatures expected;

Relative Humidity

UKCIP09

Gridded Met office data

Change Factor Indication of water demands for crops

River Gauge data

30 year data set for 20 worst flood events Environment Agency

Hydrograph extraction Flood response, recession curve, peak and base flows

Associated rainfall data for events.

MIDAS (BADC) Correlate and regression with rainfall and rainfall change.

Daily rainfall for selected dates

Land-use

Aggregated DEFRA Farm census (dates) 1 km Grid

Catchment analysis Impact of land-use on hydrograph

Soils MAGPIE data base (source Brazier, Anthony) 1 km grid

Catchment analysis Erosion potential

Secondary Modelling Changes in the river hydrographs

ITE data on land cover attributed to arable, agricultural grassland, rough grassland, woodland, urban, open water.

Change factor Revised flood response time of the rivers

MAGPIE and HOST,

new Climate Change downscale

Topography

Combined factor selection

Soil vulnerability mapping

Table 3-3 Data used in developing the catchment response model defining the source data and transformations of the data.

23

Chapter 4 RESULTS AND ANALYSIS

4.1 The Description of the population:

A total of 64 samples were obtained from the general public, 24 of which were on-line.

The respondents took on average 28 minutes to reply on-line (median value was 19

minutes) Face to face interviews generally took 20 to 25 minutes.

The sample responding to the public questionnaire was tested against the DEFRA

2007 Survey of Public Attitudes and Behaviours toward the environment. The data in

the DEFRA survey had been categorized, and therefore the data from the North

Devon survey was re-coded to matching categories. Mann-Whitney tests were applied

to Social Grading, Age group, Gender and urban/rural classification. Of these social

grade and urban classification were significantly (95% confidence) different from the

DEFRA 2007 Survey of Public Attitudes and Behaviours toward the environment. The

North Devon survey sample was much higher in professional grades of respondents

and, un-surprisingly, more rural. (see tables 4.1 and Figure 4.1)

To briefly describe the sample; ages were between 20 and over 70. The highest age

group represented was 41-54 years representing 46 percent of the sample. The 20-29

years age group represented 19% of the sample. Forty-eight percent of the sample

was male.

.

Respondent's

gender

Please could you tell me your age last birthday?

Urban/Rural classification Social Grade

Mann-Whitney U 107559.000 106647.000 64462.500 79881.000 Wilcoxon W 109389.000 6653418.000 6611233.500 81772.000 Z -.139 -1.106 -8.529 -3.782 Asymp. Sig. (2-tailed) .889 .269 .000 .000

Table 4-1 Mann-Whitney test results for comparison of Gender, age group, urban rural classification of dwelling and social grade.

24

Figure 4-1 Comparative graphs of the distribution of percentage of the samples from the 2007 Survey of Public Attitudes and Behaviours toward the environment and the North Devon survey

This report focuses specifically on changing understanding in climate behaviour and therefore the reported understanding of climate change terms is also compared between the national and local samples in the table 4.2.

F1.1 How much would you say you know about this term? -

Climate change


Global warming


Carbon footprint


CO2 or Carbon Dioxide

emissions


Carbon offsetting

Mann-Whitney U 103900.500 109567.500 64418.500 108485.500 72286.500 Wilcoxon W 105791.500 6656338.500 66434.500 110501.500 74302.500 Z -.840 -.565 -6.089 -.690 -5.158 Asymp. Sig. (2-tailed) .401 .572 .000 .490 .000

Table 4-2 Mann Whitney statistic test results for the questions relating to reported climate change knowledge between the survey samples in the DEFRA 2007 Survey of Public Attitudes and Behaviours toward the environment and the North Devon Survey

The survey suggests that the North Devon population is more aware of the climate

change terminology relating to “Carbon footprint” and “carbon offsetting” that the

national survey suggests two years prior to this. Given that social grade was

significantly different between the two samples, the knowledge of carbon footprint and

offsetting were tested for a correlation between social grade and reported knowledge

of carbon footprint and carbon offsetting. Whist a correlation was identified (significant

at the 95% confidence level) in the national datasets (R2=0.09), no significant

correlation was found in the North Devon dataset. It is not likely therefore that the

2 3 4 5

Social Grade

0%

10%

20%

30%

Percent

National North Devon

2 3 4 5

Social Grade

25

difference in understanding of these terms is related to social grading. We therefore

conclude that the North Devon sample of people reports to be more climate

terminology aware than the national dataset taken in 2007. The reasons for this might

be related to publicity, information campaigns or policy activity in the intervening two

years. These may include the development of the Climate Change Act 2008, local

planning applications and decisions relating to energy installations as well as national

media and local campaigns.

4.2 The Environmental attitudes of the North Devon survey

From this point forward the analysis of the questionnaire will relate to the North Devon

survey only. The North Devon sample was tested for differences in the gender, age

and social grading between those who completed the survey online form those who

completed face to face, through a “t” test. No significant difference in the means was

highlighted at the 95% confidence limit. We can therefore regard the 2 samples as

representative of the same population.

Figure 4-2 Relative priorities of respondents to the North Devon survey. This is a cumulative score of those issues in the top 3 of people’s concerns.

The sample was asked about the relative priorities or concerns they have about

society and the environment at the moment. The results regarding the ranking of

0

5

10

15

20

25

30

35

26

concerns can be seen in figure 4.2. Climate change appears as ranked jointly 3rd in

the set of concerns.

For efficiency reasons the scores for the NEP questions have been re-coded to be

consistent in scoring low for a pro- environmental attitude and high for anti-

environmental attitude. The scores were summed; the most environmentally

conscious score would be 7, the least 35. The results of these questions suggest that

the sample is sympathetic towards the environment, (Mean score 14.1, Kurtosis

0.806, skewness 0.98). See Figure 4.3.

Figure 4-3 Aggregated NEP score derived from summing re-coded NEP questions.

We therefore have a sample of people who are broadly representative of the national

population in terms of age and gender, are more confident about the terminology of

climate change and show a high degree of environmental sympathy.

Of particular interest are the responses to the priorities for action on climate change in

relation to other issues. With regard to action on climate change being more important

than biodiversity or than landscape, the responses are normally distributed around the

neither agree nor disagree category. However, there is a significant (95% confidence)

bias towards a preference of action on climate change being more important than the

27

dip in the economy.(Figure 4.4). The mean value as 2.47 skewness of the distribution

was 0.636.

Figure 4-4 Histogram of Response of question “Climate change is more important than a dip in the economy” (1=strongly agree, 5 strongly disagree) (n=64)

Government and Individual responsibilities

There were two questions that related to the degree of responsibility of mitigating

climate change relating to either the individual or to the government. The Question “I

don’t believe my everyday behaviour and lifestyle contributes to climate change”

occasionally drew ambiguous responses. The activity of better “energy management”

in the home was proven to be a more reliable predictor of climate sensitive behaviour.

People with low carbon lifestyles and a high environmental conscience suggested

that, whatever they did, they may still contribute to global warming and others would

acknowledge that their practice of low carbon lifestyle contributed less to global

warming. However the same histogram plot for the question “The government should

take complete responsibility for tackling climate change” suggests a correlation with

level of action on climate change. i.e those who think the responsibility is not totally

the governments are more likely to act. The correlation was tested and found to be

significant at the 95% level of confidence. The results are contrasted graphically in

Figures 4.5 and 4.6.

1 2 3 4

climate change more important than economy

0

5

10

15

20

Count

28

Figure 4-5 Histogram of degree of agreement about personal behaviour and climate change

(x axis; 3=neither agree nor disagree, 4=tend to disagree, 5=definitely disagree. “Not likely”= not likely toreduce carbon this way, “May do”=possibly do this action to reduce carbon, “Definitely”= already acting to reduce carbon).

Figure 4-6 Histogram of degree of agreement about government total responsibility and climate change.

(x axis; 3=neither agree nor disagree, 4=tend to disagree, 5=definitely disagree. “Not likely”= not likely to reduce carbon this way, “May do”=possibly will do this action to reduce carbon, “Definitely”= already acting to reduce carbon).

Something we can draw from this result is that the more people see that they as

individuals have a responsibility (and ability) to help reduce greenhouse gas

emissions (GHG) they are more likely to do so. This should be borne in mind for any

Not likely

May do

Definitely

Energy management Re-code

Bars show counts

3 4 5 6

my behaviour does not contribute to climate change

5

10

15

20

25

Count

n=4

n=3

n=1

n=22

n=4

n=1

n=19

n=3

n=3

n=1

Not likely

May do

Definitely

Energy management Re-code

Bars show counts

3.0 3.5 4.0 4.5 5.0

government total responsibility

0

10

20

30

Count

n=7

n=4

n=1

n=13

n=5

n=2

n=26

n=1n=2

29

information and campaigns on climate change behaviour.

The survey was designed to test for a relationship between a sense of futility and

reported actions. However, none of the respondents strongly agree or tend to agree

that the “climate is beyond control”. The same response applies to the question

whether the “effects of climate change are too far in the future to affect them”. This

can be a reflection of the fact that this is a coastal community, where the impacts of

sea-level rise are already evident.

4.3 Reported Behaviours The respondents were asked to report their behaviours against certain climate

mitigation actions. The results are summarised in the bar chart (Figure 4.7).

Figure 4-7 Proportion of respondents and their reported behaviour for climate change mitigation.

Taking the analysis a little further, the current and potential for carbon reduction has

been portrayed by adopting a “carbon saving index”. This index has been derived

from graphical representation in the Defra 2008 document “Framework for Pro-

environmental behaviours.”

The resulting Figure 4.8 indicates where the best gains can be made for carbon

saving beyond the high degree of activity reported by many of the respondents. The

obvious areas where there are low scores for “doing it now” are in the installation of

renewable energy systems around the home. Whilst it would seem a priori that

affordability might be the issue, analysis for correlation of income-bracket against the

measures proved to be inconclusive with a Spearman’s Ranked correlation with the

0 0.2 0.4 0.6 0.8 1 1.2

Use car less

Fewer flights

Improved energy management in home

Reduce water use

Recycle

Waste less food

Buy local

Fuel efficient vehicle

Solar PV

Solar thermal

Wind turbine

Low energy bulbs

Energy efficient boiler

Insulation

Secondary glazing

Don’t want to

Haven't thought about it

Thought but wont

Thinking about it

Doing it but won't keep it up

Tried but not doing it now

Doing it and keep it up

Don’t know

NA

practice of installing solar thermal heating (solar thermal was chos

requires an investment, but can pay for itself quite quickly).

demand the question

heating?” The survey indicates that richer people h

not going to do it.

Figure 4-8 Qualitative representation of potential carbon saving based on proportion of people reporting a certain behaviour multiplied by an index of c

The extra potential is the proportion of respondents who are thinking about, trying or have tried each behaviour type multiplied by the carbon saving index. The proportion showing a barrier to a particular action is included.

The use of such an index for carbon saving can suggest where policy activity can be

focussed and where best carbon savings can be made by both individuals and

authorities. However,

might be made. For example

household makes in any year.

there was no correlation at 95% confidence between income

taken. Therefore the carbon footprint reduction by reduced flights

any one sector of society.

To summarise this section

climate aware than the national average, they are on the whole envi

sympathetic and demonstrate a good reported behaviour for climate change.

30

practice of installing solar thermal heating (solar thermal was chos

requires an investment, but can pay for itself quite quickly). The response does

demand the question “Why don’t more affluent people have the solar thermal

The survey indicates that richer people have considered the option but are

Qualitative representation of potential carbon saving based on proportion of people reporting a certain behaviour multiplied by an index of carbon saving potential for that behaviour.

The extra potential is the proportion of respondents who are thinking about, trying or have tried each behaviour type multiplied by the carbon saving index. The proportion showing a barrier to a

is included.



, from the data in the survey we cannot tell the true savings th

might be made. For example, this survey does not indicate how many flights a

household makes in any year. It is noted that according to the DEFRA 2007

there was no correlation at 95% confidence between income and the number of flights

erefore the carbon footprint reduction by reduced flights

sector of society.

section of the survey results, the north Devon community is more

climate aware than the national average, they are on the whole envi


practice of installing solar thermal heating (solar thermal was chosen because it

The response does

Why don’t more affluent people have the solar thermal

ave considered the option but are

Qualitative representation of potential carbon saving based on proportion of people arbon saving potential for that behaviour.

The extra potential is the proportion of respondents who are thinking about, trying or have tried each behaviour type multiplied by the carbon saving index. The proportion showing a barrier to a



from the data in the survey we cannot tell the true savings that

this survey does not indicate how many flights a

DEFRA 2007 data set

and the number of flights

erefore the carbon footprint reduction by reduced flights is not restricted to

community is more

climate aware than the national average, they are on the whole environmentally


31

However, the affluent section of the community that could afford to do more do not

seem to be. The group has a high sense of individual responsibility for solving climate

change problems. Indeed working with such a pro-environmental group will be difficult

to prove that involvement in scenario modelling increases individual environmental

practice.

4.4 Communication issues:

This section of analysis will focus on the qualitative analysis of the communications

and risk perception backed up by some quantitative information of the questionnaire.

There were 2 questions that related to the understanding of risk and likelihood. In

response to the question regarding “40% chance of rain”, six of the 53 definite replies

exposed a low regard for the apparent ability to accurately forecast. The answers

were given a score out of 10 for the accuracy of their statement (Table 4.3).

Table 4-3Distribution of scores for accuracy of explanation of “40% chance of rain”

The answers ranged from a very accurate “statistically 4 days out of every 10 with

these conditions, there will be rain” as the high scoring end. Scores in the middle of

range were given to statements such as “the weather will be fine probably won't rain”

or “slightly more likely not to rain”. Some respondents assumed the 40% referred to a

coverage value i.e “40% of the UK will have rain”. Others it was a definition of time;

i.e. “it will rain for 40% of the day” or the intensity of rain such as “light showers”.

The language used to describe the probability of “40%” by the respondents is

noteworthy. The most common term (11 out of 53) used was “less likely” (or more

likely when offering the counter proposal of no rain). The respective terms are

captured in Figure 4.9.

Score for answer 40% chance

6 9.4 11.3 11.3

4 6.3 7.5 18.9

11 17.2 20.8 39.6

21 32.8 39.6 79.2

7 10.9 13.2 92.5

3 4.7 5.7 98.1

1 1.6 1.9 100.0

53 82.8 100.0

11 17.2

64 100.0

1

2

3

4

5

6

8

Total

Valid

SystemMissing

Total

Frequency Percent Valid PercentCumulative

Percent

32

Figure 4-9 Frequency of terms used to describe 40% chance of rain

The following question requested the interpretation of a “1 in 100 year flood

protection”. This particular phrase was used because it is still in common parlance in

Environment Agency documents. As with the 40% chance of rain question the answer

was scored on accuracy of the answer and a record for the terminology used. The

resulting score are seen in table 4.4

Frequency Percent Valid Percent Cumulative

Percent Valid 1.00 10 15.6 19.6 19.6

2.00 6 9.4 11.8 31.4 3.00 16 25.0 31.4 62.7 4.00 8 12.5 15.7 78.4 5.00 9 14.1 17.6 96.1 6.00 1 1.6 2.0 98.0 7.00 1 1.6 2.0 100.0 Total 51 79.7 100.0

Missing System 13 20.3 Total 64 100.0

Table 4-4 Accuracy score for the response to interpretation of 1 in 10 year flood protection

terminology used for 40%

low chance

marginalsignificantmore chance

mightnegligablestrong60% counter

probablepossiblelikely

Frequ

ency

12

10

8

6

4

2

0

terminology used for 40%

Figure 4-10 Histogram of terms used in describing a 1 in 100 year flood protection.

Table 4.10 shows the frequency of the terminology used to describe the probability

associated with the “1 in 1

quantified the probability correctly as 1%. The terminology used shows little overlap

with that used for the 40% probability. This suggests that a derivation of a standard

language for public consum

The question regarding the

was any preference to how people regard or risk or even interpret it better. Below

table 4.5, is the cross tabulation of those interviewed f

the question.

Count

Place X move

yes

no

don't know

Total

Table 4-5 Advice move from Place “X” v Advise move from Place “A” Cross

Although this distribution was tested with Chi Squared which returned a 99%

33

Histogram of terms used in describing a 1 in 100 year flood protection.

shows the frequency of the terminology used to describe the probability

associated with the “1 in 100 year” phrase. Out of the 15 scalable responses, only 2



consumption could be developed.

question regarding the understanding of risk on maps was to test

was any preference to how people regard or risk or even interpret it better. Below

is the cross tabulation of those interviewed face to face that responded to

Place A move Total

yes no don't know yes

10 11 4 25

1 9 0 10

0 2 6 8

11 22 10 43

from Place “X” v Advise move from Place “A” Cross


Histogram of terms used in describing a 1 in 100 year flood protection.

shows the frequency of the terminology used to describe the probability

00 year” phrase. Out of the 15 scalable responses, only 2



understanding of risk on maps was to test whether there

was any preference to how people regard or risk or even interpret it better. Below, in

ace to face that responded to

from Place “X” v Advise move from Place “A” Cross-tabulation


34

probability that this distribution is not expected by chance, the test had to be rejected

because too many cells had an expected value of less than 5. Both of the points lay

on exactly the same probability of inundation on the maps, therefore the expected

answers would be internally consistent. (There is no right or wrong answer). Where

the probability of flooding is shown on a coloured continuum, people are more likely

(twice as likely) to advise to move than on the distinct contour basis. The frequencies

for the various permutations of the “don’t knows” are too small to draw any

conclusions. The landowners responses had a very similar distribution. When these

are added in to the results, the 2x2 contingency table for yes/no, the result is

significant at the 90% level. From this we can tentatively deduce that people make

more confident decisions when presented with more absolute data such as contour

data rather than relying on judgement though a continuum, or conversely there is an

aversion to taking a risk when information is presented as continuous variables.

4.4.1 Supporting information from the Public Focus Group analysis and open questions responses.

In this part I will review the results of the open questions regarding risk and the

discussions from the focus group made of members of the general public.

Returning to the questionnaire, the set of questions asked the interviewee to express

their perception of climate change on the area’s wellbeing for public health, economy

and environment. This was done mainly to explore the terminology that people would

use and get a sense for any magnification of terms as an indication of where their

sensitivities lay. The question was deliberately set with a 25 year time frame to place

it quite possibly within the individual’s lifetime.

The responses varied from “No chance” to a range of defined impacts that will

definitely happen. The scalable responses have synthesised in the table 4.6 along

with the associated terminology given the respondents.

35

Probability

scale

Terms commonly used Health Economy Environment

None None, no 5 2 1

Negligible Negligible, very small,

slight

12 4 3

Small Small, not huge 7 8 6

Moderate Significant, moderate,

medium, quite likely

7 6 9

Very likely Big, highest, high risk,

huge, major

6 14 10

Definite Definite(ly), will, (if

impact is defined using

term will)

8 12 15

Total 45 46 44

Table 4-6 Responses and terminology used to identify risk to health, economy and environment of the area 25 years from now.

These results indicate that the respondents perceived greater risks for the

environment and economy than they did for the area’s population’s health. Most

health concerns, when identified, were expressed in relation to impacts of flooding,

though other comments included inter alia skin cancer, malaria, lack of food and

strains on health infrastructure due to inward migration of people from the east.

Economic impacts were regarded as both positive and negative and highlighted better

tourism, green energy industries on the positive side. On the negative side the cost of

infrastructure such as flood and coastal defence and health infrastructure were cited.

The ranked priorities for the questionnaire sample’s personal concerns are shown in

Figure 4.11. There is an apparent disparity between the respondents’ personal

concerns and the impacts that they expect. For example the low scale and likelihood

of regional health impacts expressed are not reflected in the personal response where

it features highly in priority ranking one and two.

Figure 4-11 Personal priority for concerns about climate impacts

The focus group consisted of

A: female, 40’s, community

50’s, resident. D; Female; early 50s, company director. E; Male; mid 50’s, retired

teacher. F; female late 50’s;

part-time media sales person.

The focus group session consisted of 2 parts, the first to deal with the communication

questions, the second to respond to the interim model and discuss scenarios.

The focus group was drawn mainly from a coastal/estuarine community and the

session topic became very focussed on flooding issues despite introducing the

concepts of risk for the health sector and campaigns such as

On conveying information about climate change, the

impact definition would

were keen to express that people do care about what happens in Africa and

developing nations with regard to climate change. “The government campaign against

smoking worked because it was done over a long time and only after gr

being put on cigarette packages

areas where a person can smoke coupled with p

36

Personal priority for concerns about climate impacts

focus group consisted of:

A: female, 40’s, community web journalist. B; female, 50’s, shop owner. C; female


teacher. F; female late 50’s; field botanist. G; Male; 20; labourer. H; Male, late 50’s.,

media sales person.





concepts of risk for the health sector and campaigns such as anti-

On conveying information about climate change, the focus group suggested that local

impact definition would be more effective in motivating people. However, the group



smoking worked because it was done over a long time and only after gr

being put on cigarette packages”. The group added that legislation to reduce the

areas where a person can smoke coupled with policy measures to help one quit

. B; female, 50’s, shop owner. C; female


rer. H; Male, late 50’s.,





-smoking.

group suggested that local

ng people. However, the group



smoking worked because it was done over a long time and only after graphic images

The group added that legislation to reduce the

olicy measures to help one quit the

37

habit. “The risks were never expressly quantified in the publicity material other than

“can seriously damage your heath””.

The terminology used to define flood risk is better expressed as a percentage; i.e. 1 in

100 years is better expressed as 1% chance in any year. It would convey the

probability on a “shorter timescale”. However, the expression of risk to many has the 2

elements of probability and impact. Focus group members said that they would accept

a 1% risk if the water “lapped up around the floor boards” once in a rare occasion, but

if it “destroyed the contents of the house” on the same frequency, they would not

suffer the risk. Another said that “I would buy a house with a 1% risk but never buy a

one with a history of flooding”. Another point of consideration was that if the house

was in a beautiful location, then this would make the risk more bearable too.

Therefore, with regard to issues such as flooding, risk should not be expressed in

pure terms of probability of getting wet, but more in terms of a “damage index and

probability”. The aesthetic values are too personal or subjective to communicate.

4.5 Farmers Questionnaire and Focus Group

Eighteen responses were received from the landowners’ questionnaire (7.6%

response rate to the postal questionnaire). The size distribution of the farms and type

is shown in the table 4.7;

Frequency Percent Valid Percent Cumulative

Percent Valid 0 to 5 1 5.6 5.9 5.9

6 to 19 3 16.7 17.6 23.5 20 to 50 6 33.3 35.3 58.8 50 to 100 1 5.6 5.9 64.7 over 100 ha 5 27.8 29.4 94.1 over 200 ha 1 5.6 5.9 100.0 Total 17 94.4 100.0

Missing System 1 5.6 Total 18 100.0

Table 4-7 Size distribution of farms responding to questionnaire

Of the 19, 10 mixed beef and sheep, 2 beef only, 2 sheep only. The remainder

included deer farming, vineyard and game bird rearing. For the respondents, the

minimum tenure of the land was 10 years to maximum of 50 years, with a mean of

26.25 years. The total agricultural experience of each of the respondents was

between 10 and 60 years with an average of 35 years. We therefore have a

reasonable body of experience to draw on for perceived changes in local climate and

38

adaptations made over a time-span of 25 years.

Focus group description:

Five people from this pool joined the landowner focus group from the nine that offered

to join from the questionnaire responses. The five comprised:

A: Male, Mid forties, organic farmer, B: Female, Mid Forties, Organic farmer and

retailer, C: Male, Mid fifties, beef and sheep farmer high in the Culm grassland area,

D: Male; 60’s small holder with cattle. E: Male late 60’s part time farmer assisting

son/daughter, mixed farm.

A further 3 people were expected to attend. However, unfortunately the meeting was

scheduled for a time that coincided with a rare weather opportunity to harvest.

The main function of this focus group was to serve as a panel of community experts

on what land-uses would be realistic for the area in future years with expected

impacts. It was also an opportunity for them to see the interim model and comment on

its findings and its application. The opportunity was also taken to explore their

attitudes to risk and climate change.

The changes that the farmers mainly reported were the milder winters. The main

concern for them was the persistence of pests and diseases over the winter months

that would not die out. There was a strong interest in “Blue Tongue” which is a

disease that is thought to have gradually moved north through warming temperatures

and mainly transmitted by flies. Therefore the year-round existence of flies was a

concern.

The group were asked to consider what adaptations they had made over recent years

to climate change. Examples included drainage enhancements and also technology.

The advent of silage meant that the opportunity to harvest a winter feed with smaller

weather windows, another adaptation was wrapping bales in plastic. The other

adaptation mentioned was the growing of crimped corn as a way of gathering grain

based crops when not fully dry and treating the same way as silage. On dealing with

risk and adaptation, the farming life is seldom planned very far into the future

therefore the farms have to be very adaptive. A crop plan may be made 1 year in

advance, but the day to day activity and the harvest is determined by the weather of

the day. Naturally, a longer term approach to a type of farming determines the

investment in machinery and buildings etc which does carry a risk. One of the

39

participants said “It’s difficult to know which way to go sometimes; hanging on in hope

for another year.” Another quote “Farmers constantly have to adapt”.

From the questionnaire to the farmers most adaptation actions that can be attributed

to climate change were driven primarily by cost as opposed to directly for climate

change.

40

Section 4.6 Model

4.6.1 Development of the hydrological model:

The model has been developed by isolating hydrographs from the data provide by the

Environment Agency. The storm events supplied were compounded flood events with

multiple flood peaks. Therefore it was difficult to isolate clean hydrographs to scale

the essential elements as identified in the methodology. A total of 124 hydrographs

were analysed from the 7 gauging stations. Filtering out anomalies and extremes

reduced the sample size to 65.

Catchment Name Frequency Percent

Area (Ha)

Greater Taw 15 23.1 84618

Greater Torridge 10 15.4 66542

Lew 4 6.2 6783

Upper Taw 14 21.5 7521

Upper Torridge 2 3.1 25813

Yeo 9 13.8 5597

Yeo and Mole 11 16.9 33007

Total 65 100.0 229881

4-8 Catchments and their distribution in the sample after filtering extreme values for the Flood curve

Only daily rainfall data was available for the events associated with the hydrographs.

The nearest historical record for rainfall intensity was for a station near Plymouth.

Given the variation across the catchments, Plymouth readings would not provide an

accurate source of rainfall intensity. The interpolations of the daily rainfall data from

the 33 stations within the catchment can be found in appendix 4 for illustrative

purposes.

The data captured in the 1km grids along with the river response times and the daily

rainfall data were exported to SPSS for analysis. The data were re-coded or re-

calculated to estimate best-fit. For the lumped model, each sub-catchment is treated

as one unit with the characteristics from the grids aggregated to the sub-catchment

level.

Due to the focus group’s collective concerns about flooding, the flashiness or the time

for the river to respond to a rainfall event was the dependent variable that was

analysed. The justification being that if the peak of the hydrograph can be reduced by

slowing the response time, flooding events will be reduced.

The hydrograph data were aggregated again into mean seasonal response times for

each catchment. After

best possible fit, the st

R2 value of 0.881. Ie. The model can account for

The determinants accepted were: Cumulative flow length (

catchment (G); seasonal factor a

class for the catchment unit

catchment unit (g).

The Flood response time (F) can be determined from the model by:

F= 0.244- 6.03S+-5.4

Figure 4-12 Comparison of mean measured response time and modelled response time for time from

base to peak flow in a rainfall event.

For the model to be an interactive planning tool, it has t

adjusted by the stakeholders, and in this case a policy response to climate change.

41


After curve fitting for the identifiable factors from the data

best possible fit, the stepwise multivariate regression analysis yielded a model with an

. Ie. The model can account for 88% of the variation.

The determinants accepted were: Cumulative flow length (L), √Ov

; seasonal factor adjustment (S), and rainfall (P), mean SNH Drainage

class for the catchment unit (D), mean of Maximum gradient from each 1km cell in the


5.4x10-6L+10.57G-0.02P-4.94D+0.083g

Comparison of mean measured response time and modelled response time for time from

base to peak flow in a rainfall event.

For the model to be an interactive planning tool, it has to include factors that can



for the identifiable factors from the data to seek the

epwise multivariate regression analysis yielded a model with an

% of the variation.

√Overall gradient of

t (S), and rainfall (P), mean SNH Drainage

gradient from each 1km cell in the


Comparison of mean measured response time and modelled response time for time from

o include factors that can be


42

The statistical model correlated with a coefficient for land-use runoff in the direction

that countered the direction had this been a process model. Therefore the land-use

run off has not been included in the regression but is used in the alteration of the

model as part of the decision support tool. The secondary modelling will use the

existing land-use mixes as a starting point (as in table 4.9) and by altering the

proportions will alter the expected run-off time from the land to the rivers.

River Basin

Taw Torridge

Catchment unit Catchment unit

Yeo Mole Upper Taw Lew

Upper Torridge

Mean Mean Mean Mean Mean Percentage Arable Land

6.572 9.769 14.392 4.840 11.125

Percentage Agricultural Grassland

66.345 71.027 52.844 68.312 64.443

Percentage Rough & Amenity Grassland

14.137 8.313 24.004 13.432 10.096

Percentage Woodland & Forest

12.925 10.727 8.230 13.411 12.388

Percentage Sub / Urban Development

.000 .126 .437 .000 1.731

Table 4-9 Catchments and associated percentage of land-use derived from 1 km grid ITE land cover data.

4.6.2 Results of Climate Model Downscaling

Outputs from the model can be seen in figures 4:12 to 4:17. More results of the

downscaled model are displayed graphically in Appendix 3. These were presented to

the focus groups to initiate the discussions about the changes that may take place

under the high emission scenario. The set presented also include the 2080-2099 set.

The presentation of the model was accompanied by a verbal explanation of the range

of emission scenarios and the PDF that results in a central tendency, which has been

ultimately illustrated on the figures.

43

Figure 4-13 Current expected monthly winter (DJF) Rainfall based on 61-90 LTA

Figure 4-14 Expected monthly winter (DJF) rainfall 2040-2059

44

Figure 4-15 Figure 4 15 Current expected rainfall summer (JJA) 61-90 LTA (mm/month)

Figure 4-16 Expected rainfall summer 2040-2059 (mm/month)

45

Figure 4-17 Expected mean temp of warmest July day current 61-90 LTA

Figure 4-18 Expected mean temp of warmest July day 2040-2059

46

4.6.3 Focus group results for scenarios

The groups were encouraged to explore a range of policy areas for scenarios. They

included food supply policy, rural/urban population balance, energy policy and

resources such as availability of fertilisers.

Both groups anticipated an increase in local food being more important. Implicit in this

is a land-use change to include more arable. The farmers commented that the area

could never be totally self-reliant for all types of food.

The challenges were the physical property of the ground:

• “Soils and topography work against you. We have either clays, steep or thin stony

soils.”

The crop husbandry required to produce food in a “post peak-oil” world:

• “long fallow periods or plenty of organic manure; otherwise only one third of

potential arable land could be in production at any one time.”

• “Would like to see more variation in agriculture and produce more fruit plums,

cherries etc, but also root crops in the right areas.”

• “This emphasises a regional approach for more localised production, but some

trans-national ideas. Such as grow more vegetables in the east and meat in the

west.”

• “Organic farming is often said that it won’t feed the world. However an allotment

the size of the [meeting] room will produce a huge amount; multi-cropping. People

grow food, not the machines!”

• “Mixing Barley Oats and peas seems to work well in Brittany. Oats and peas and

barley and peas mix has been really good.”

• “Farm size will reach a maximum; there are no more efficiency gains to be had.

Zero grazing is a time bomb waiting to go off; big welfare problems and pollution.”

On the societal aspects of the future rural economy

• “Food may not be as cheap. Need to be more efficient with the way we use food.

Subtle and sub-conscious education.”

• “Co-ordination and co-operative between farmers mean that production and

variety is shared, people will need to lower their standards of expectation about

standard sizes.”

Both groups were attracted to the multiple farm use for production of energy, food,

fibre, supported safeguarded biodiversity and water management.

47

• “Are there the skills to manage all of these “ecosystem services”, how do we get

the skills back into the rural areas. Such a diversity of skills is not found in the one

place. The way the UK has gone does not help. There is a step back needed.”

Energy production for the areas was “eagerly” discussed in the public group, mainly

around the issues of wind energy. The group conceded that some wind was inevitable

in the mix of energy in the future along with solar in towns, biomass and tidal energy.

The groups individually highlighted a rural re-population in order to produce the food.

• “Removal of workforce has reduced the productivity of farmland. Developing

nations have lessons to teach the west.”

• “Local community farming and community farm shares might be a growing idea.”

The focus groups were very energetic when it came to exploring the future, based on

the climate scenarios. There was a degree of disbelief about the summer rain

predictions, coloured by the recent 3 years of wet summers. Set in the context of long-

term changes the downscaled climate models were accepted as credible.

Based on the discussion with both groups, the scenario put into the secondary

modelling:

• An increase in the proportion of arable land in areas where the topography

and soils are suited to it.

• Maintenance of grazing land. With some extensification and allowing year

round grazing.

• Biomass production on areas of land that may flood or even help to attenuate

floods.

• Increase in woodland cover (for multi-use) on steeper slopes

• Food production systems that support biodiversity

• No target to be totally self-reliant

• An increase in developed land to accommodate the increased food producing/

land management sector and in line with population growth as presented.

These scenarios do not conflict with the attitudes reported in the landowner

questionnaire.

48

4.6.4 Secondary Modelling results

For simplicity at this point, 2 land-use scenarios have been produced.

They are expressed in table 4.10

Land-use Scenario1 Scenario 2

Land for grains Maintained the same to

supplementary feed cattle.

Maintained the same to

supplementary feed the cattle

and other above ground

crops for people.

Land for root crops Doubled to account for

greater local independence

for food.

Increase by 60%, more food

coming. Buy extra food from

EU etc.

Land for agricultural

grassland

Land reduced to move to

permanent pasture, foresty,

conversion to arable

Land reduced to move to

permanent pasture, foresty,

conversion arable to

Land for

permanent/rough

pasture

Increase by 10% to move

from temporary grassland

and reduced fertiliser

Increase by 20% to reduce

fertiliser inputs and increase

biodiversity

Land for

Woodland/Forestry/

biomass

Increase by 20% to

develop extra

Increase by 50% for more

carbon sequestration and

biomass production

Urban development Increased by 20% to

account for accommodating

population growth

Increased by 20% to account

for accommodating

population growth

Table 4-10 Table of land-use scenario changes

The impacts on the river responses are as follows, according to the model results.

Figure 4-19 Change in flood response

The land-use changes do not result in large c

rivers. This is mainly due to the positive and negative impacts almost balancing out.

Moving from this basic step is it possible to identify the soils that will be most at risk

from erosion due to the more intensive winter rains predicted in the climate model.

Increases in arable areas should only be accommodated on flatter areas of lan

free draining soils, and even relocated from the high risk

not sensitive enough to

suggest might be the method to safely increase the arable coverage.

combined climate model the rainfall, run

MAGPIE data) and slope gradient combine to give an index of vulnerability.

49

flood response time of rivers under new land-use scenarios

use changes do not result in large changes in the response time of the




Increases in arable areas should only be accommodated on flatter areas of lan

free draining soils, and even relocated from the high risk areas. The data provided are

sensitive enough to identify a field or sub-field level as the farmers focus groups

suggest might be the method to safely increase the arable coverage.

combined climate model the rainfall, run-off percent, soil texture class (from the


use scenarios

hanges in the response time of the




Increases in arable areas should only be accommodated on flatter areas of land with

areas. The data provided are

field level as the farmers focus groups

suggest might be the method to safely increase the arable coverage. From the

off percent, soil texture class (from the


50

Figure 4-20 Soil sensitivity in 2050. The most vulnerable 20% of grids are in yellow

4.6.5 Focus Group responses to the process

The public group were asked to give their reflection to the process and the model so

far and what it meant to them. There was total support for the group to continue to

work with the model. The group expressed that they felt animated and empowered.

Most agreed that they had learnt more.

Typical statements were:

• “I have been glad to learn more from other opinions. I would like to make a

proper action plan that can be implemented.”

• “Have not changed opinions but do want to know more about how planning is

being impacted by the models and whose responsibility is it. Can local people

actually do this?”

• “Farmers also want to do more. Would be nice to meet at the same time and

get the variety of opinions.”

On questioning about how to make effective communication (leading to action) for

climate change, the group made the following typical responses.

• “Planning should start at the local level, and then communicated upwards so

local people can have some power over their own destiny. There is a lack of

motivation in people.”

• “Make communities stronger!”.

• “Paint the picture for them and get some results soon after to show the goods”.

51

Although the farmers were not part of the test group, they also stated that they would

like the work to continue and for them to be part of the programme to move the local

climate response plan further as a community initiative.

These qualitative results do not offer any definitive proof of full engagement as

defined by Lorenonzi (2007). Time constraints did not allow a follow up study to see if

the enthusiasm expressed at the focus group meetings is sustained and translated

into continued action.

The following discussion section will explore the meaning of these results with regard

to the aims of the dissertation.

52

Chapter 5 Discussion

Limitations The limitations to the work have been on time and availability of the data. This work

has been fitted around a busy full-time career and within 3 months.

The resolution of the land-use and soil data along with the lack of hourly rainfall data

precluded the development of a more sensitive hydrological model. The model is still

in a relatively unrefined form and is based on one set of scenarios from one climate

model. Prudhomme et al (2009) advocate the use of not just a range of scenarios but

also a suite of different models for any climate change impact on river flows.

However, the information provided and the model developed functions as a decision

support tool and a communication tool about the impacts of climate change and

meets the requirements, if not the author’s aspirations, for the project.

The survey of the public was intended to achieve 100 responses; in effect only 65

were produced. Although short of the target, the length of the interview and the

numbers achieved provided a great deal of quantitative and qualitative data. A

longitudinal study of the behaviour of the focus groups at least 1 month after the

working with the model would provide more conclusive evidence of change in

awareness, individual actions and ultimately advocacy for policy implementation.

Summary of main findings The thrust of the work looked into the communication issues around public

understanding climate change and how the derivation of a localised model animates

discussion about the climate issues. I will start with a discussion about the model, its

derivation, results it indicates and its use as a communication or decision support tool.

The discussion will then link that model to the work with the community responses

and the focus group.

The catchment model has provided us with an understanding of the factors that

impact on the hydrographs of the main rivers. The model internalises the catchment

characteristics in a “black box” method, where the processes are assumed to

demonstrate the cause and effect through the multivariate analysis. This is not always

the case, and therefore the land-use run off coefficient had to be externalised from the

equation. It is suggested that the negative correlation between the land-use and the

response time of the river in the catchment is because there is less arable and less

53

urban areas in the faster catchments, so the relationship is not causal in terms of

hydrological processes but purely of land-use and settlement. The scaling effect of the

rainfall is very minor; this is generally expected. The classic hydrograph described by

time to peak (or response curve) and regression or decay curve is broadly

independent of the intensity of the rainfall event but more on the duration of the event.

The size of the peak is rainfall dependant.

Examining the land-use data in the 1km grid against the vulnerability of soils suggests

that the choice of arable fields is not ideally suited to the conditions of the soil and

topography. The distribution of land-use throughout the area has been the result of

overlays of government agricultural policy over the years, coupled with economic

drivers to the extent that the initial concept of land-use matching land capability has

been somewhat blurred. These distortions have resulted in fields losing soil and

nutrients to the extent that the Taw catchment is a Nitrate Vulnerable Zone as a

requirement under the EC Nitrates Directive (91/676/EEC). It will be possible through

the analysis to theoretically re-locate arable areas to soils that have good drainage

capacity, on lower gradients and lower run-off coefficients, so that the increase in

demand for locally grown vegetables does not exacerbate the risk of flooding.

The conservation of mass leads us to assume an inverse relationship between the

value of peak flow in the river and the response time of the flood curve for any given

rainfall event. Therefore, even if the extension of the flood response curve is small,

due to the asymptotic nature of the return periods and flood peaks, the extension in

return periods can be high. Analysis of return curves (Figure 5.1) for the Taw at

Umberleigh from CEH data suggests that a 7% extension in response time will alter a

flooding probability of 0.025 to 0.0111. However, these will be countered by the

increases in rainfall intensity that are anticipated during the winter months in the area.

The application of the climate model data incorporated to date does not permit a

credible calculation of the expected change in rainfall intensity. We can qualitatively

assume that the intensity of winter rains will increase and therefore the return periods

for any given rainfall event will decrease. This will counter any gains from the changes

in land-use that have been modelled.

Figure 5-1 Probability returns for volumes recorded at Umberleigh.

The model methodology can be replicated for other catchments, provided that the

same input data are available. However the model equation will b

of catchments. The catchment model is sensitive to changes in land

in daily rainfall and therefore as that data becomes available in a format suitable from

the UKCIP09 or other models, it can be p

The people in the sample

appear to show this with

indications of some choice not being

obvious one is not taking as many flights

is often the largest contribution to his or her carbon footprint.

At the moment the investment in low carbon technology alternatives has been weak.

Several members of the community with

installation of renewable energy but have not followed this through. The survey does

not identify the barriers to these changes

planning permission or technical

as reviewed in Kollmuss and Agyeman (2002)

and psychological.

54

Probability returns for volumes recorded at Umberleigh.


same input data are available. However the model equation will be unique to this suite

of catchments. The catchment model is sensitive to changes in land


or other models, it can be put into the catchment equation.

in the sample appear to be quite well versed in climate change issues and

appear to show this with activity that is easier to do. However, there have been

indications of some choice not being made towards their own low

taking as many flights: this is a cost saving for the individual and it

is often the largest contribution to his or her carbon footprint.


eral members of the community with higher incomes have considered the


not identify the barriers to these changes. This can due to the practicalities

ssion or technical challenges with retro-fitting equipment

Kollmuss and Agyeman (2002) the reasons for this are sociological


e unique to this suite

of catchments. The catchment model is sensitive to changes in land-use and changes


equation.

to be quite well versed in climate change issues and

there have been

towards their own low carbon future. An

cost saving for the individual and it


incomes have considered the


. This can due to the practicalities such as

fitting equipment. Alternatively,

the reasons for this are sociological

55

As Dessai et al (2008) have already pointed out that decisions made within the

current levels of uncertainty, fortunately have for the greater part been correct. The

predictive skill of this model is dependent on the climate models used as its basis. As

a consequence, it would not yet be wise to use this as a planning and adaptation tool.

Counter to this, the over-riding concern of the coastal communities is flooding from

the both the rivers and the seas. Therefore as a model it has two functions; first to

provide planners with the possible impacts of land-use decisions on the trends of

catchment responses and secondly as a communication tool to engage the local

community.

The focus group and the model

The evidence of what climatic changes have already been happening along with the

localised predictions of what might happen in the future has provided a good stimulus

for debate with the groups. The climate model is not the be-all and end-all for the

community focus group. It has offered a sense of direction and established a

framework for adaptive management and the stimulus for discussions about the

stakeholder’s future. There is still a sense within the community members of the focus

group to treat the models as absolute truth, despite the warnings that accompanied

the presentation.

The model’s apparent function with the focus group has been to establish a visual tool

for them to understand the impacts and how their decisions can impact on the future.

This is a form of empowerment has given them a status of “agency” (or the start of it).

Agency gives them the greater freedom to act differently, enabled by the knowledge

they have gained, not only from the model and their interactions with it but from the

dynamics of the group itself and the fact that they had something to say and that they

were being listened to.

Could this have happened without the catchment and climate model? As already

stated it was the focus of the second half of the focus group meetings and akin to the

processes of “backcasting” (Tansey et al, 2002) or scenario planning to incorporate

social and moral judgements rather than purely physical or natural science

phenomena.

56

The communication issues

The variability in the responses reflects the deep uncertainty in the science of climate

change and the general uncertainty in the general public about how science is

communicated.

Paradoxically, communicating to the public about climate change should not be done

entirely in the manner they would prefer to hear or see it. The Met Office

announcement in April 2009 of a “65% chance of a barbeque summer” is a clear

example of where the science was presented with levels of uncertainty but was

ignored by the media. From the survey for this document, although it would seem that

many would understand probability expressed as a percentage, the terms they chose

to use are more subjective and qualitative rather than numerically descriptive. I

suggest that the key to bridging the communication gap but still relaying the

uncertainty lies in establishing a standard set of terminology and rubrics for talking

about climate change to the public as Kandlikar et al (2005) had proposed for the

scientific community. However, the terminology offered would need to be in terms that

the public can understand. Based on the survey responses, a suggested framework

is shown in table 5.1

Qualitative scale Probability

scale (%)

None 0

Negligible 1-10

Small 10-35

Moderate 35-60

Likely 60-70

Very Likely 70-80

Almost Definite 80-90

Definite 98-100

Table 5-1 Suggested terminology to define bands of probability when communicating with the public.

Again following the format suggested by Kandlikar et al (2005), such a qualitative

scale could then use the bounds of the probability density function.

For example “under the high emissions scenario, it is very likely that summer between

2040 and 2060 temperatures will be 2.5 degrees centigrade higher than they were on

average between 1961 and 1990”.

57

The focus group confirmed the statement made by Grothmann and Patt (2005) about

risk being a combination of probability of an event and the damage that might occur.

For anyone in the public to make a valued judgement on any risk, the probability

should be accompanied by some scale of damage that may be associated.

A degree of scaling damage has already been done by Association of British Insurers.

The insurance industry has an accurate figure for this scale, but expresses the

combined risk and hazard in financial terms (McCarthy, 2007) and only to its

customers in the form of an insurance premium. According to McCarthy (2007), the

data is commercially sensitive and only used in aggregate form within the industry and

ultimately used to lobby for policy support or change. Whilst this data might not cover

all of the aspects of an integrated assessment on climate change, the data will refer to

events and losses that are very pertinent to the local community and individuals.

Using this following example; a house in a 1 in 100 year flood plain that when floods

will ruin everything on the ground-floor might be expressed (using the terms of

probability mentioned in table 5.1.)

“There is a negligible (but real) risk that in any year this house might suffer

50% contents damage”.

Although the focus group suggested that 1% was negligible, the problem might arise

that “negligible” is interpreted as “no chance”. In which case, a resident might place

themselves at risk (however small) without knowing; hence the addition of “but real” in

parenthesis.

The results of the advice choice on the map question in the survey suggested that

public have a tendency for risk aversion when faced with a continuum on which to

make a decision. This has an impact on how we might consider communicating

climate change related information. Whilst there is uncertainty in the underpinning

information, it would be irresponsible in terms of adaptation planning for the

uncertainly to be ignored. Referring back to Dessai et al (2008) the reminder of the

uncertainty to the end-user is essential. The presentation of geographic data on

climate change in clearly defined contours conveys an apparent precision that does

not exist. Therefore, based on the outcomes of this work, geographic, or even

graphic information should be presented as part of a continuum.

58

Chapter 6 CONCLUSIONS

In this project I have identified the attitudes and awareness of and action for climate

change within the northern Devon population. The research has identified some the

problems and some solutions related the communication of climate change and its

associated uncertainty with the public. The key challenges are to communicate

climate change science in a way that describes the projections and the uncertainty

associated with them in terms that are easily digestible.

The research suggests that effective communication of the issues with the uncertainty

embedded in could be achieved through acceptable framework of terminology and

through the graphic portrayal of the range of possible outcomes through a continuum.

Further research on communication should centre on the derivation of a suitable

framework for describing uncertainty to the lay community with graphics as well as

language and how this ultimately sways their decisions for adaptation.

Through the derivation of a downscaled climate impact model, the inputs and the

outputs have been a focal point for debate with focus groups for them to suggest

scenarios for the future based on societal as well as physical parameters. These

scenarios have been incorporated into a lumped catchment model to explore the

impacts on the flood response curve of the river hydrographs as recorded at the

gauging stations. The inclusion of the model in the focus groups was a clear stimulus

to exploring possible scenarios for the future.

Outputs from the model suggest that in order to slow down run-off after a rain event to

reduce the risk of flooding, acute and chronic soil erosion, increases in arable and

urban land-cover must be countered by increased woodland and permanent or rough

grassland. Such measures applied to the current land cover regime may reduce the

return periods of flooding considerably between the 2% and 1% (1 in 50 and 1 in 100

return periods).

The key conclusions are that although the methodology can be applied in any

catchment with the same limited data, the catchment model should be refined further

towards a more transferable process based model that can accept the data from a

range of outcomes from a single climate model, or other models. The variation in the

outcomes should be relayed as suggested in the communication recommendations.

59

To assist the development of the process model, monitoring stations giving hourly

rainfall records and up to date field level data of land-use will be an added advantage.

Exploring these topics with the two focus groups suggests that the area has features

that warrant special study. For example the predominance of grazing land and the

challenges of producing arable goods, puts the area quite distinct from the typical

mixed farm approach where land capability might offer more choice than here.

Coupling the evidence of a very high environmental quality of the catchment

recognised as UNESCO Biosphere Reserve with the expectation that environmental

regulation will increase for the protection of water courses through the Water

Framework Directive and the protection of soils, constraints will be placed on the land-

use choices of the future. The further development of this model can assist in

considering the alternatives for reduced emissions from farming and food production,

strengthened local economies and sustained ecosystem services. Such a model must

be backed up by appropriate field measurements and monitoring.

60

Acknowledgements Data sources:

UKCIP; For outputs from the UKCIP09 (c) Crown Copyright 2009.

The UK Climate Projections data have been made available by the

Department for Environment, Food and Rural Affairs (Defra) and

Department for Energy and Climate Change (DECC) under licence

from the Met Office, Newcastle University, University of East

Anglia and Proudman Oceanographic Laboratory. These organisations

accept no responsibility for any inaccuracies or omissions in

the data, nor for any loss or damage directly or indirectly

caused to any person or body by reason of, or arising out of,

any use of this data.

MIDAS (BADC): For rain gauge data and Long-term Average climate data for the

climate model.

EDINA Ordnance Survey for Digital Elevation Model

===========================================================

© Crown Copyright/database right 2009. An Ordnance Survey/EDINA supplied

service.

===========================================================

My thanks go to Dr R E Brazier University of Exeter and Dr. Suraje Dessai for their

support and advice for this dissertation and the provision of data and extra reading

material associated with this project.

My deepest thanks go to Sarah Felgate who helped to gather the face to face

interviews with the questionnaires. I would also like to express gratitude to my house-

mate, Peter Yeo, who provided logistic support of nourishment and refreshments.

Finally, I am extremely grateful to the focus group members from both of the groups

contributed willingly and freely with their time and opinions. Long may it continue!

61

Bibliography

Bathurst et al. Validation of catchment models for predicting land-use and climate change. Journal of Hydrology (2004) vol. 287 pp. 74-94

Bell, V A, Kay, A L, Jones, R G, Moore, R J(2007) Development of a high resolution grid-based river flow model for use with regional climate model �. Hydrology and Earth System Sciences (2007)

Blume T, Zehe E, Bronstert A; 2007 Rainfall–runoff response, event-based runoff coefficients and hydrograph separation. Hydrological Sciences–Journal–des Sciences Hydrologique, 52,5 843-862

Bronstert et al. Effects of climate and land-use change on storm runoff generation: present knowledge and modelling capabilities. Hydrol. Process. (2002) vol. 16 (2) pp. 509-529

Burgess. Focusing on Fear: The Use of Focus Groups in a Project for the Community Forest Unit, Countryside Commission. (1996) Area vol. 28 (2) pp. 130-135

Committee on Climate Change, 2009, http://www.theccc.org.uk/sectors/residential-property/emissions; Sourced 02/05/09

Convention on Biological Diversity, Ecosystem Approach, sourced 2009 http://www.cbd.int/ecosystem/

DEFRA, 2008, http://www.defra.gov.uk/news/2008/080131a.htm sourced 02/05/09

DEFRA, 2008, A framework for Pro-environmental behaviours http://www.defra.gov.uk/evidence/social/behaviour/pdf/behaviours-jan08-report.pdf

Defra, (2007) Survey of Public Attitudes and Behaviours toward the Environment, 2007 http://www.defra.gov.uk/environment/statistics/pubatt/download/pubattsum2007.pdf

Dessai and Hulme. (2008) How do UK climate scenarios compare with recent observations? Atmos. Sci. Lett. vol. 9 (4) pp. 189-195

Diaz-Nieto and Wilby. (2005) Statistical downscaling and climate change factor methods: impacts on low flows in the River Thames �. Climatic Change

Engel, B. 2004 SEDSPEC, Purdue Research Foundation http://cobweb.ecn.purdue.edu/~engelb/abe526/Runoff/C_table.html. Accessed August 09 Fawcett L, Anthony S, Silgram M, 2004 Application of the EveNFlow model to the Vansjø-Hobøl Catchment, Norway, EUROHARP WP3-2004 (http://www.euroharp.org/toolbox/catchment%20reports/EveNflow/ADAS_CatchmentResportNorwayV2.doc.)

62

Fowler et al. New estimates of future changes in extreme rainfall across the UK using regional climate model integrations. 1 Assessment of control climate. Journal of Hydrology 300 (2005) 212–233 (2005) pp. 1-22

Goodess, C, Osborn, T, Hulme, M, 2003, The identification and evaluation of suitable scenario development methods for the estimation of future probabilities of extreme weather events. . Tyndall Centre for Climate Change Research, Technical Report 4. (2003) pp. 1-69

Goss. J. (1996) Introduction to Focus groups. Area vol. 28 (2) pp. 113-114

Grothmann T, and Patt A; 2005; Adaptive capacity and human cognition: The process of individual adaptation to climate change; Global Environmental Change Part A Volume 15, Issue 3, October 2005, Pages 199-213

Hulme M and Dessai S (2008) Predicting, deciding, learning: can one evaluate the 'success' of national climate scenarios? . Environmental Research Letters (2008) pp. 1-14

Jackson J, Li Y, Passant N, Thomas J, Thistlethwaite G, Thomson A and Cardenas L; 2008, Greenhouse Gas Inventories for England, Scotland, Wales and Northern Ireland: 1990 - 2006 (http://www.airquality.co.uk/reports/cat07/0809291432_DA_GHGI_report_2006_main_text_Issue_1r.pdf ; Sourced 02/05/09)

Jonsson and Alkan-Olsson. (2005) Participatory Modelling–(how) can computer generated information affect the” room of action” of local stakeholders ACSIS nationella forskarkonferens för kulturstudier

Jonsson A. 2005 Public Participation in Water Resources Management: Stakeholder Voices on Degree, Scale, Potential, and Methods in Future Water Management. AMBIO: A Journal of the Human Environment (2005) vol. 34 (7) pp. 1-6

Justus, M., P. Kinrade and B.L. Preston; 2007; Translating Scientific Understanding of Climate Change Impacts into Effective Policy Response – The Critical Role of Stakeholder Engagement

Kandlikar M, Risbey J, Dessai S, (2005) Representing and communicating deep uncertainty in climate change assessments. C R Geoscience 337, 443,455 Kollmuss and Agyeman (2002). Mind the gap: why do people act environmentally and what are the barriers to pro-environmental �. Environmental Education Research

Lahsen. M (2005) Seductive Simulations? Uncertainty Distribution Around Climate Models. Social Studies of Science vol. 35 (6) pp. 895-922

Lorenzoni I, Nicholson-Cole S & Whitmarsh L, (2007). Barriers perceived to engaging with climate change among the public and their policy implications. UK Global Environmental Change 17, 445–459 (2007) pp. 1-15 McCarthy S (2007) Contextual influences on national level flood risk communication. Environmental Hazards, 7,2,128-140

63

Nicholson-Cole and ONeill. Promoting Positive Engagement With Climate Change Through Visual and Iconic Representations. Science Communication In print (2008) pp. 1-25

Olsson AJ and Karin Berg K 2005 Local Stakeholders' Acceptance of Model-generated Data Used as a Communication Tool in Water Management: The Rönneå Study; AMBIO: A Journal of the Human Environment 34(7):507-512.

Olsson L & Pilesjo P; Approaches to spatially distributed hydrological modelling in a GIS environment; Environmental Modelling with GIS and remote sensing, Edited Andrew Skidmore 2002 isbn 0415241707

Pethick, J. (2006) Taw Torridge Final Report final version. pp. 1-87

Pidgeon, N.F. et al. , 2006, Public Risk Perceptions, Climate Change and the Reframing of UK Energy Policy in Britain, 2005 Colchester, Essex: UK Data Archive May 2006. SN: 5357.

Prudhomme and Davies. (2009) Assessing uncertainties in climate change impact analyses on the river flow regimes in the UK. Part 2: future climate. Climatic Change vol. 93 (1-2) pp. 197-222

Tansey, James, Carmichael, Jeff, VanWynsberghe, Rob, Robinson, John. (2002) The future is not what it used to be-participatory integrated assessment in the Georgia Basin. Global Environmental Change (2002) vol. 12 pp. 1-8

Stoll-Kleemann et al. The psychology of denial concerning climate mitigation measures: evidence from Swiss focus groups. Global Environmental Change (2001)

Wheeler, D., Shaw, G. and Barr, S. (2004) Statistical Techniques in Geographical Analysis. (Fulton, London).

Wilby et al. (2004) Guidelines Use of Climate Scenarios from statistical downscaling methods. IPCC pp. 1-27

Yu-Pin Lin a, Nien-Ming Hong , Pei-Jung Wu , Chen-Fa Wu , Peter H. Verburg; Impacts of land use change scenarios on hydrology and land use patterns in the Wu-Tu watershed in Northern Taiwan. Landscape and Urban Planning 80 (2007) 111–126 (2007) pp. 1-16

Zeigler et al. (1996) Focusing on Hurricane Andrew through the Eyes of the Victims . Area vol. 28 (2) pp. 124-129

64

Appendices

Appendix 1 Public Questionnaire. (face to face interviews were supported by answer cards to give options to the

respondent. Page breaks have been removed)

NDCCS

Bideford Station East the Water

Bideford Devon

EX39 4BB Dear Sir or madam, I am carrying out research in the North Devon’s UNESCO Biosphere Reserve and part of a Master in Science (sustainable development) research at University of Exeter on people’s perceptions about climate change. I expect the Biosphere Reserve use the data to help to develop suitable action and how to present our information on climate change. After helping us with this questionnaire, we are seeking a small group of people (around 8 to 10) to help us to describe a range of possible outcomes of the future for this area such as landscape, types of food production. To do this the group will need to meet all together twice for just about an hour each time. Participants will be paid a small fee for their participation. If you wish to participate we will ask for your details at the end of this questionnaire. We hope the meetings will be fun as well as interesting for the participants. The meeting is likely to be in Barnstaple Library early in the evening. The following questions will help us to understand the general feeling by people in the area from a range of backgrounds about climate change and their actions and attitudes. I guarantee that any personal data is solely for the purpose of this research and will not be passed on to any other organisation in any form that individuals will be identifiable, nor will any individual be identifiable within work. Thank you in advance for participating. If you wish to follow up the progress of this study or more information on the UNESCO Biosphere Reserve in North Devon, please refer to the website: www.northdevonbiosphere.org.uk

65

A About the respondent HOUSEHOLD AND RESPONDENT CHARACTERISTICS, WELLBEING Household and respondent characteristics First of all, I’d just like to ask some questions about you and your household. ASK ALL SHOW CARD A1 A1 Which of the statements on this card applies to you?

1. Working full time 2. Working part time 3. On a local or government training scheme (GTS) 4. On an Apprenticeship 5. Registered unemployed/signing on for jobseekers allowance 6. Not registered unemployed but seeking work 7. At home/not seeking work 8. Long-term sick or disabled 9. Retired 10. In full-time education 11. Other (specify) 12. Don’t know 13. Refused

A2 Including yourself, how many people usually live in your home? Please include all adults and children. ENTER NUMBER SHOW CARD A2 A4 First of all, which of these best describes your home?

1. Detached house 2. Semi-detached house 3. Terraced house 4. Bungalow 5. Flat (in a block of flats) 6. Flat (in a house) 7. Maisonette 8. Other (specify) 9. Don’t know

Refused SHOW CARD A3 A5 In which of these ways do you occupy this accommodation?

1. Own it outright 2. Buying it with the help of a mortgage or loan 3. Pay part rent and part mortgage (shared ownership) 4. Rent it 5. Live here rent free (inc. rent free in relative/friend's property, 6. excluding squatting) 7. Squatting 8. Don’t know

A7 And how long have you lived in your current home?

1. Up to 1 year 2. More than 1 year, up to 2 years 3. More than 2 years, up to 5 years 4. More than 5 years, up to 10 years 5. More than 10 years, up to 20 years

66

6. More than 20 years 7. Don’t know 8. Refused

SHOW CARD A4 A8 Which phrase on this card best describes the area where you live?

1. A big city 2. The suburbs or outskirts of a big city 3. A town or a small city 4. A country village 5. A farm or home in the countryside 6. Other (specify) 7. Don’t know

Wellbeing SHOW CARD A5 A9 The next few questions ask about some different aspects of your life. First, all things considered, how satisfied are you with your life as a whole nowadays? Please answer using this card, where 0 means extremely dissatisfied and 10 means extremely satisfied. 0 (Extremely dissatisfied) - 10 (Extremely satisfied) Don’t know Refused Wellbeing Value statements A10 Card A6 ASK RESPONDENT TO READ BACK NUMBERS IN EACH ANSWER CATEGORY, STARTING WITH ‘DEFINITELY AGREE’

1. In general I feel very positive about myself 2. I spend a lot of time worrying about things

A6 Value responses

1. Definitely agree 2. Tend to agree 3. Neither agree nor disagree 4. Tend to disagree 5. Definitely disagree 6. Not applicable 7. Don’t know

A11 Priorities Card A7 Which 3 of the following issues concern you the most?

1. Air pollution 2. Pollution of rivers and seas 3. Flooding 4. Crime 5. Litter 6. Poor waste management (e.g. overuse of landfills) 7. Traffic/ congestion 8. War in the Middle East and Afghanistan 9. GM food

67

10. Climate change 11. Economic recession 12. The hole in the ozone layer 13. Using up the earth's resources 14. Extinction of species 15. Radioactive waste 16. Overpopulation (of the earth by humans

B About climate change awareness

ATTITUDES AND KNOWLEDGE IN RELATION TO THE ENVIRONMENT/CLIMATE CHANGE Next some more general questions to do with the environment SHOW CARD B1 B1 How much, if anything, would you say you know about the following terms? READ OUT � (RANDOMISE ORDER OF STATEMENTS) 1. Climate Change (TO ALWAYS APPEAR FIRST IN LIST) 2. Global Warming 3. Carbon footprint 4. C02 or Carbon dioxide emissions 5. Carbon offsetting Responses

1. A lot 2. A fair amount 3. Just a little 4. Nothing – have only heard of the name 5. Nothing – have never heard of it 6. Don’t know

ASK ALL SHOW CARD B2 B2 I am going to read out some changes that people might make to their lifestyles/homes. For each one, please tell me which answer on the card applies to you personally at the moment. There are no right or wrong answers – we’re just interested in what you do at the moment, not what you think you should or shouldn’t be doing.

68

1 I don't re

ally want to do th

is

2 I haven't re

ally thought ab

out doing this

3 I've thoug

ht abou

t doin

g this, but pro

bably w

on't do

it

4 I'm thinking

about doing this

5 I'm a

lready do

ing th

is, but I prob

ably won't m

anag

e to

keep it up

6 I've tried doing this, b

ut I've given up

7 I'm

already do

ing th

is and in

tend to ke

ep it up

8 Don

’t know

9 Not app

licable

1. Use a car less

2. Take fewer flights

3. Reduce the use of gas/oil/electricity

4. Cut down on water usage

5. Recycle more waste 6. Waste less food

7. Buy food produced locally rather than food produced abroad

8. Use a more fuel-efficient vehicle

9. Install Solar panels for electricity

10. Install Solar panels for heating

11. Install Wind turbine

12. Install Low energy bulbs

13. Install Energy efficient boiler

14. Install Increased insulation

15. Install Secondary Glazing

69

Show card B3 B3 Which of these best describes how you feel about your current lifestyle and the environment?

1. I’m happy with what I do at the moment 2. I’d like to do a bit more to help the environment 3. I’d like to do a lot more to help to environment 4. Don’t know

B4 And which of these would you say best describes your current lifestyle? (tick one)

1. I don’t really do anything that is environmentally-friendly 2. I do one or two things that are environmentally-friendly 3. I do quite a few things that are environmentally-friendly 4. I’m environmentally-friendly in most things I do 5. I’m environmentally-friendly in everything I do 6. Don’t know

Show Card B5 B5

How much do you agree or disagree with the following statements? 1. Strongly agree 2. Tend to agree 3. Neither agree nor disagree 4. Tend to disagree 5. Strongly disagree 6. Don't Know

1. We are close to the limit of the number of people the earth can support 2. Humans are severely abusing the environment 3. The Earth has very limited room and resources 4. The so-called ‘environmental crisis’ facing humanity has been greatly exaggerated 5. Reducing carbon footprint is more important than saving biodiversity 6. Scientists will find a solution to global warming without people having to make big

changes to their lifestyles 7. Reducing carbon footprint is more important than preserving landscape 8. Humans were meant to rule over the rest of nature 9. Dealing with climate change is more important than a dip in the economy 10. The government should defend houses and property from erosion and flooding

regardless of their value and cost of the defences. 11. The environment is a low priority for me compared with a lot of other things in my life 12. Climate Change is beyond control – it’s too late to do anything about it 13. So many people are environmentally-friendly these days, it does make a difference 14. The effects of climate change are too far in the future to really worry me 15. I do worry about the changes to the countryside in the UK and the loss of native

animals and plants 16. I don’t believe my behaviour and everyday lifestyle contribute to climate change 17. The government should take complete responsibility for tackling climate change

B6 Where do you mainly get your information on environmental issues? Tick no more than 3.

1. News Paper 2. Radio 3. TV

70

4. Internet 5. Friends/family 6. Local authority leaflet/government leaflet 7. Environmental group publications

C1 About understanding uncertainty and understanding the areas of

perceived risk.

Show card C1

1. What does is mean when weather forecast says “40% chance of rain?”

2. What does a “1 in 100 year flood protection” area mean to you?

Show card Map Card

1. If I live in the yellow box next to X on this map, would you advise me to move?

2. If I live in the If I live in the black box next to A on this map, would you advise me to move?

C2 Show card C2 (In this question I wish to find out how people express risk and what risk is perceived to be the most significant to the respondent.) Using your own words, give a value to the following;

1. Describe the risk of global warming and climate change forcing a significant impact on public health in Devon in 25 years from now.

2. Describe the risk of global warming and climate change forcing a significant impact on economic development in Devon in 25 years from now?

3. Describe the risk of global warming and climate change forcing a significant impact on the environment in Devon in 25 years from now?

C3 In terms of being impacted by climate change which subject area concerns you most? Place them in order (1 being first)

a. Your health

b. Your environment

c. Your financial circumstances

d. none

71

D About climate change behaviour ENERGY AND WATER EFFICIENCY IN THE HOME Energy efficiency Next some questions about your home D1 Do you have a room thermostat for your heating system?

1. Yes 2. No 3. Don’t know

D2 Do you know what temperature your thermostat is set at most of the time? IF YES: What temperature is it set at? Temperature in degrees C/F Varies too much to say Don’t know SHOW CARD D5 D6 Please could you tell me how often you personally do each of the following 1. Leave your TV on standby overnight 2. Leave the lights on in rooms that aren’t being used 3. Leave a mobile phone charger switched on at the socket when not in use 4. Fill the kettle with more water than you are going to use 5. Keep the tap running while you brush your teeth 6. Leave the heating on when you go out for a few hours 7. When there is a choice, have a bath rather than a shower 8. Put more clothes on when you feel cold, rather than putting the heating on or turning it up 9. Throw away food because it has gone off 10. Decide not to buy something because you feel it has too much packaging 11. Take your own shopping bag when shopping 12. Re-use things like empty bottles, tubs or jars, envelopes or paper

1. Always/Very often 2. Quite often 3. Sometimes 4. Rarely 5. Never 6. Not applicable/cannot do this 7. Don’t know

E. Final personal details Finally, I’d just like to ask you a few more questions about your circumstances E4

1. Have you ever heard of the Biosphere Reserve? i. Yes No

2. Do you know where it is? i. Yes No

E8 CODE RESPONDENT’S GENDER

1. Male 2. Female

E9 Please could you tell me your age last birthday?

1. Numeric range (16 – 99) or age group? 2. Refused

72

E10 To which of these groups do you consider you belong?

A. White - British B. White – Irish C. White – other white background D. Mixed – White and Black Caribbean E. Mixed – White and Black African F. Mixed – White and Asian G. Mixed – any other Mixed background H. Asian or Asian British – Indian I. Asian or Asian British – Pakistani J. Asian or Asian British – Bangladeshi K. Asian or Asian British – other Asian background L. Black or Black British – Caribbean M. Black or Black British – African N. Black or Black British – other Black background O. Chinese P. Other (specify) Don’t know Refused

SHOW CARD E2 E13 I am now going to ask you about your household income. I only need to know an approximate amount, to see if this influences people’s views and experiences. Please can you tell me your overall HOUSEHOLD income from all sources in the last year? This includes earnings from employment or self-employment, income from benefits and pensions, and income from other sources such as interest and savings. Please look at this card and tell me which letter represents your TOTAL HOUSEHOLD INCOME in the last year from all sources BEFORE tax and other deductions.

Annual Weekly Monthly 1 Under £10, 000 Under £200 Under £830 4 £10,000 - £14,999 £200 - £289 £830 - £1,249 5 £15,000 - £19,999 £290 - £389 £1,250 - £1,649 6 £20,000 - £24,999 £390 - £489 £1,650 - £2,099 7 £25,000 - £29,999 £490 - £579 £2,100 - £2,499 8 £30,000 - £40,000 £580 - £770 £2,500 - £3, 350 10 £40,000 - £50, 000 £770 - £969 £3,350 - £4149 12 £50,000 -£59,999 £970 - £1,149 £4,150 - £4,999 13 More than £60,000 More than £1149 More than £6250 E15 Do you have any qualifications...?

1. Yes 2. No 3. refuse

E16 What is the highest qualification you have? DO NOT READ OUT, BUT PROMPT AS NECESSARY. CODE THE ANSWER WHICH IS HIGHER UP THE LIST.

1. Degree level qualification including foundation degrees, graduate membership of a professional institute, PGCE or higher

2. Diploma in higher education 3. HNC/HND 4. ONC/OND 5. BTEC/BEC/TEC/EdExcel 6. Teaching qualifications (excluding PGCE) 7. Nursing or other medical qualification not covered above

73

8. Other higher education qualification below degree level 9. A-level/vocational A-level or equivalent 10. International Baccalaureate 11. NVQ/SVQ 12. GNVQ/GSVQ 13. AS-level/vocational AS-level or equivalent 14. Access to HE 15. O-level or equivalent 16. GCSE/vocational GCSE 17. CSE 18. RSA/OCR 19. City and Guilds 20. YT certificate 21. Key skills 22. Basic skills 23. Don't Know 24. Any other professional/vocational/foreign/other qualification (SPECIFY)

Could you tell me your current or most recent job��.. E18 INTERVIEWER: CODE RESPONDENT’S SOCIAL GRADE.

• A • B • C1 • C2 and r • D • E

F Willingness to participate As part of this project we are asking a small group of people meeting on 2 occasions to discuss their thoughts on climate change and the future of this area. Participants will be paid small amount for their time. All data will be kept anonymous. Name ADDRESS POST CODE CONTACT NUMBER EMAIL ADDRESS

M

ap

in

dic

ati

ng

th

eo

reti

cal

flo

od

ris

k. R

an

gin

g f

rom

on

ce e

ve

ry 5

ye

ars

on

av

era

ge

in

da

rk b

lue

, to

on

ce e

ve

ry 2

00

ye

ars

on

av

era

ge

, th

e

lig

hte

st s

ha

de

.

79.0m

79.5m

80.1m

80.2m

BM

80.85m

65.9m

7.7m

7.6m

7.5m

7.8m

7.6m

15.7m

11.0m

19.9m

11.2m

10.9m

15.6m

BM 17.14m

9.23m

BM

52.2m

56.3m

BM 22.66m

37.2m

27.4m

BM 8.24m

7.5m

15.1m

13.7m

BM 13.30m

16.5m

11.3m

BM 13.69m

16.7m

7.6m

9.6m

BM 10.49m

12.0m

BM 13.19m

8.7m

18.2m

16.0m

15.3m

X

75

M

ap

sh

ow

ing

th

eo

reti

cal

flo

od

ris

k. A

rea

sh

ad

ed

blu

e w

ill

flo

od

at

lea

st o

nce

ev

ery

10

0 y

ea

rs o

n a

ve

rag

e, t

he

ye

llo

w/

gre

en

wil

l fl

oo

d

on

ce e

ve

ry 2

00

ye

ars

on

av

era

ge

.

9.3m

12.1m

45.2m

BM 54.1m

54.32m

41.6m

BM 28.21m

BM 16.71m

29.8m

20.9m

10.5m

11.3m

12.4m

9.9m

10.7m

BM

14.6m

14.8m

14.50m

18.1m

9.6m

BM 10.28m

13.5m

11.0m

11.8m

9.8m

10.4m

BM 11.87m

10.8m

11.3m

13.9m

17.1m

21.0m

BM 16.61m

11.5m

17.20m

BM

13.8m

BM 15.97m

16.9m

BM 16.14m

12.4m

BM 14.18m

18.9m

BM 51.50m

42.5m

54.5m

A

Appendix 2 Land owner Questionnaire (map images reduced to save space)

Bideford Station

East the Water

Bideford

Devon

EX39 4BB

Dear Sir or madam I am carrying out research in the North Devon’s UNESCO Biosphere Reserve on people’s perceptions about climate change as part of a Masters Research Project at University of Exeter. I expect that the Biosphere Reserve will use the data to help to develop suitable action for climate change and how to present that type of information. We are firstly asking a large group of landowners to complete this questionnaire to get a broad understanding of attitudes. After helping us with this questionnaire, we are seeking a small group of people (around 8 to 10) to help us to describe a range of possible outcomes of the future for this area. To do this, the group will need to meet all together once or possibly twice for just over an hour each time. Participants will be paid a small fee for their participation. The first meeting will be on 12 August in the late afternoon in the Eggesford/Umberleigh area. If you wish to participate we will ask for your details at the end of this questionnaire. I guarantee that this data is solely for the purpose of this research and will not be passed on to any other organisation, nor will any individual be identifiable within the report. When you have completed the questionnaire, please post it back as soon as possible using the pre-paid envelope provided. Thank you in advance for participating. If you wish to follow up the progress of this study or more information on the UNESCO Biosphere Reserve in North Devon, please refer to the website: www.northdevonbiosphere.org.uk Kind regards

Andrew Bell

77

Questionnaire: Section A

1. In order to consider issues of farming and climate change for a range of farms we need some basic details about the farm you have and the land-use on it.

i. Total Size (ha),

2. Type of topography; what are the main features of the farm? i. Gently sloping hillside, ii. steeply sloping, iii. large flood plain, iv. hill top plain, v. gentle rolling or undulation

3. Type of farm: please indicate the type(s) of farming you do. Tick as many as apply i. Dairy ii. Beef iii. Pigs iv. Sheep v. Poultry vi. Other

4. Landcover; What area (hectares please) of land comes under the following

categories? i. Set-aside ii. Arable iii. Permanent grassland iv. Temporary ley v. Spring Cereal (not including maize) vi. Winter cereal (not including maize) vii. Maize viii. Root crops ix. Woodland x. Orchard xi. other

5. We would like to identify river catchments where we will do some modelling of future scenarios. We will be able to locate the catchment from your post-code and you will still have reasonable anonymity. Please enter your post code��

6. Type of tenancy/occupancy on the majority of your holding. Please enter the number of hectares under each category

i. Freehold ii. Farm Business Tenancy iii. Full agricultural tenancy iv. Annual rent/grass keep

7. How long have you been farming on that holding? ��..Years

8. What is the total number of years farming experience you have?............years

9. Is your farm under any Agri-environment schemes? (Higher level Stewardship, Entry

level Stewardship, Countryside Stewardship, ESA, Organic Stewardship i. Current

1. Type 2. Date entered

ii. Applied for (type)

78

Section B ATTITUDES AND KNOWLEDGE IN RELATION TO THE ENVIRONMENT/CLIMATE CHANGE Next some more general questions to do with the environment B1 How much, if anything, would you say you know about the following terms? Use the following scores for the replies.

a. A lot b. A fair amount c. Just a little d. Nothing – have only heard of the name e. Nothing – have never heard of it, f Don’t know

1. Climate Change Score��.. 2. Global Warming Score��.. 3. Carbon footprint Score��.. 4. C02 or Carbon dioxide emissions Score��.. 5. Carbon offsetting Score��.. B2 Below are some changes that people might make to their lifestyles and or businesses. For each one, please tell me which answer on the card applies to you personally at the moment. There are no right or wrong answers – we’re just interested in what you do at the moment, not what you think you should or shouldn’t be doing. Please tick one box in each row for the response in each of the behaviours.

I don't really w

ant to do

this

I haven't really th

ought a

bout do

ing th

is

I've thought ab

out doing this, but probab

ly wo

n't do it

I'm th

inking abou

t doin

g this

I'm a

lready doing this, but I proba

bly wo

n't mana

ge to keep

it up

I'm a

lready doing this a

nd intend to ke

ep it up

I've tried do

ing th

is, but I've given

up

Don’t know

Not a

pplicable

Use a car less

Take fewer flights

Cut down on the use of gas, oil and electricity

Cut down on water usage

Recycle more rather than throwing things away Waste less food

Buy food produced locally rather than food produced abroad

Use a more fuel-efficient machinery

79

Climate Change awareness Do you think climate change is having an effect on your farm/land now? Yes No Don’t know If so, how? Do you think climate change will have an effect on your farm/land in the next 25 years? Yes No Don’t know If so, how? Do you think climate change presents more opportunities or risks to your farm business?

Section C About understanding uncertainty and understanding the areas

of perceived risk.

3. In your own words, what does is mean when the weather forecast says “40% chance of rain?”

4. In your own words, what does a 1 in 100 year flood protection area mean to you?

5. If I live in the yellow box at place X on this map, would you advise me to move?

Map indicating theoretical flood risk. Ranging from one flood event in any 5 years on average in very dark blue, to once in any 200 years on average, the lightest shade.

79.0m

79.5m

80.1m

80.2m

BM80.85m

65.9m

7.7m

7.6m

7.5m

7.8m

7.6m

15.7m

11.0m

19.9m

11.2m

10.9m

15.6m

BM 17.14m

9.23m

BM

52.2m

56.3m

BM 22.66m

37.2m

27.4m

BM 8.24m

7.5m

15.1m

13.7m

BM 13.30m

16.5m

11.3m

BM 13.69m

16.7m

7.6m

9.6m

BM 10.49m

12.0m

BM 13.19m

8.7m

18.2m

16.0m

15.3m

X X

80

6. If I live in the black box next to A on this map, would you advise me to move?

Map showing theoretical flood risk. Area shaded blue will flood at least one event in any 100 years on average, the yellow/green will flood once in any 200 years on average. Using your own words, give a value to the following statements

1. In your opinion, what is the risk of global warming and climate change forcing a significant impact on public health in Devon in 25 years from now?

2. In your opinion, what is the risk of global warming and climate change forcing a

significant impact on economic development in Devon in 25 years from now?

3. In your opinion, what is the risk of global warming and climate change forcing a significant impact on the environment in Devon in 25 years from now?

9.3m

12.1m

45.2m

BM

54.1m

54.32m

41.6m

BM 28.21m

BM 16.71m

29.8m

20.9m

10.5m

11.3m

12.4m

9.9m

10.7m

BM

14.6m

14.8m

14.50m

18.1m

9.6m

BM 10.28m

13.5m

11.0m

11.8m9.8m

10.4m

BM 11.87m

10.8m11.3m

13.9m

17.1m

21.0m

BM 16.61m

11.5m

17.20m

BM

13.8m

BM 15.97m

16.9m

BM 16.14m 12.4m

BM 14.18m

18.9m

BM 51.50m

42.5m

54.5m

A A

81

State to what degree you agree or disagree with the following statements (1 is totally disagree, 5 is totally agree)

1. Global warming and climate change will have a noticeably negative impact on my health in the next 25 years.

Score ��. 2. Global warming and climate change will have a noticeably negative impact on my

economic and financial situation in the next 25 years.

Score��. 3. Global warming and climate change will have a noticeably negative impact on the

environment in which my family and I live.

Score��

Current Climate behaviour 1. Are you taking any action to adapt to the impacts (past or expected) of climate change

on your farm? (e.g. reduce water usage, different crops, different tillage mechanisms). If so, what are they?

2. Are you taking any action to reduce greenhouse gas emissions from your farm? (biogas digestion, energy efficiency, machinery efficiency, soil restoration, carbon fixing, solar heating). If so, what are they

3. Are these actions driven by cost saving or environmental factors?

a. Only environment b. Only cost saving c. Both

4. Do you or are you interested in measuring your on-farm carbon emissions? Yes b. No

82

General Questions about how think about the environment

1. Which of these best describes how you feel about your current lifestyle and farming ethic and the environment? a) I’m happy with what I do at the moment b) I’d like to do a bit more to help the environment c) I’d like to do a lot more to help to environment d) Don’t know

2. Which 3 of the following issues concern you the most? Please tick the three greatest

concerns. a. Air pollution b. Pollution of rivers and seas c. Flooding d. Crime e. Litter f. Poor waste management (e.g. overuse of landfills) g. Traffic/ congestion h. War in the Middle East and Afghanistan i. GM food j. Climate change k. Economic recession l. The hole in the ozone layer m. Using up the earth's resources n. Extinction of species o. Radioactive waste p. Overpopulation (of the earth by humans)

On a scale of 1 to 5 (1=totally disagree to 5=totally agree)

1. I feel optimistic about the future of farming in the UK for the next 25 years. ��.

2. Food production within the UK will become increasingly important. ��.

3. Environmental factors will become increasingly important over the next 25 years. �.

4. Food security and a wildlife rich environment are incompatible concepts. ��..

5. The UK will increasingly import its food from EU and beyond. ��

6. The way I manage my land has an impact on communities some distance away. ��..

7. Agricultural land provides more public benefits than food alone. ��..

83

1. Which, if any, of these statements would you say applies to you? You can choose as

many or as few as you like. Please tick against the numbers.

a) I know a lot about environmental issues, such as climate change b) I often talk to friends or family about things they can do to help the environment c) I try to persuade people I know in farming to become more environmentally-friendly d) I’ve suggested improvements at my farm to make it more environmentally-friendly e) I’ve told relatives or friends to avoid buying from a particular company because I feel

they are damaging the environment f) Don’t know g) None of these

1. Where do you normally go for your farm business/financial advice?

2. Where do you go for your land management advice?

3. Do you ever receive wildlife and environment advice?

a. If so from where/who?

4. What other sources of information do you use?

a. Trade publications (Farmers weekly etc)

b. Internet

c. Newspapers

Personal information To strengthen the findings of this questionnaire, we wish to hold 1 possibly 2 focus groups involving a selection of landowners. The purpose of these is to explore what perceptions of risk there are or lack of risk relating to climate change and land-use. We also need your experience to contribute to the model of a range of scenarios for land-use in the future given expected climate change impacts. We will pay a small fee for attending the groups, each will be about 1.5 hours long. Are you willing to help with these groups? Yes No If so, Please can I have a contact email address and phone number to confirm the venue and time? Name

Address Phone Number Email address Thank you again for your help.

84

Appendix 3 Localised Climate change model presentation Climate Scenarios Most frequent model result. (50%): Based on High emissions from UKCIP09 25Km model.

Plots of UKCIP changes in mean temperature and precipitation to give the participants a context for the provenance of the localised model.(source UKCIP)

85

Localised interpretation of the UKCIP results

(Sample values for meteorological data given are for Kings Nympton)

Rainfall Autumn.

Present: 104 mm

2040-59: 109mm

2080-99:110mm

86

Summer Rain

Present: 68mm

2040-59: 41mm

2080-99: 26mm

87

Winter Rain

Present: 114mm

2040-59: 129mm

2080-99: 139mm

88

Summer Temperatures

Present: 20 degrees

2040-59: 23.4 degrees

2080-99: 24.7 degrees

89

Relative Humidity in July

Present: 78%

2040-99: 75%

2050-99: 73%

90

Appendix 4 Daily Rainfall distribution for 20 worst floods (source MIDAS and interpolated using inverse distance weighting) Sample dates given only.

04/01/83 09/02/74

06/10/88 31/10/94

27/09/74

25/12/68

Downscaled Climate Models and Public Engagement

Documents