FARMER ATTITUDES TOWARDS DIFFUSE POLLUTION MITIGATION MEASURES IN ENGLAND

Transcription

1 FARMER ATTITUDES TOWARDS DIFFUSE POLLUTION MITIGATION MEASURES IN ENGLAND A Demonstration Test Catchments Report Emilie Vrain 1, Andrew Lovett 1, Lister Noble 2, Fiona Grant 3, Pete Blundell 3, Will Cleasby 4 1 University of East Anglia, 2 Farm Systems and Environment, 3 ADAS, 4 Eden Rivers Trust Contact s: e.vrain@uea.ac.uk, a.lovett@uea.ac.uk December

2 FARMER ATTITUDES TOWARDS DIFFUSE POLLUTION MITIGATION MEASURES IN ENGLAND Contents 1. Introduction Current developments in agri-environmental policy Knowledge of farmer behaviours and attitudes Research objectives Farm survey methodology Survey design Farmer sample Conducting the survey Survey data Farm survey results Characteristics of surveyed farms Farm type and land holding size Land tenure Participation in agri-environmental schemes and land in designated areas Farmer awareness of Catchment Sensitive Farming Categorisation of mitigation measures Current uptake of mitigation measures by farmers Uptake of highly applicable measures Influence of CSF and AES participation on measure uptake Influence of farm size Influence of land tenure Variations in uptake by farm type Overall trends Farmer attitudes to future uptake of measures Attitudes to land use change and farm infrastructure measures Attitudes of arable farmers Attitudes of lowland livestock farmers Attitudes of dairy farmers Attitudes of mixed farmers Farmer priorities The context of survey results and implications for agri-environmental policy Comparison of results with other surveys National survey results

3 4.1.2 Regional comparisons of surveys results Implications for agri-environmental policy Recommendations for future research Conclusion References Appendix A: DWPA mitigation measures included in the DTC farm baseline survey Appendix B: Farmer participation in agri-environmental schemes and designated areas Appendix C: Classifications of mitigation measures

4 1. Introduction Evaluating the effectiveness of land management measures that can help reduce diffuse pollution from agricultural activities is essential for future improvements to water quality and meeting the goals of the Water Framework Directive (WFD). Extensive research has been carried out to determine the best agricultural practices for pollution control (e.g. Deasy et al., 2010), however the implementation of such measures will only be effective with the cooperation of land owners and managers. Whilst many agricultural management options remain voluntary, stakeholder knowledge and engagement is increasingly seen as a necessary component of catchment management. The UK Demonstration Test Catchment (DTC) programme is a government-funded initiative, evaluating the effectiveness of an array of agricultural mitigation measures in terms of their ability to improve water quality whilst maintaining farm profitability (McGonigle et al., 2014). An important aim of the DTC programme is to scale up the results from the research in individual catchments to assess the implications of wider adoption. It is therefore vital to learn what farmer attitudes are towards different mitigation measures in order to evaluate the potential for implementation across the country. To assess such attitudes, the DTC teams carried out a farmer survey in three contrasting catchments (the Hampshire Avon, Eden and Wensum) during This report describes the background to the survey in Section 1 and outlines the methodology used to conduct it in Section 2. Key findings from the data obtained are presented in Section 3, with aspects of the results separated into the four dominant farming systems found within the catchments. An overall discussion and conclusion in Section 4 considers the implications of the results for policy makers. It is hoped that this research will also inform decision-making regarding the re-design of agri-environmental schemes (AES) and the revision of programmes of measures with respect to WFD implementation. 1.1 Current developments in agri-environmental policy The UK Government seeks to tackle the issue of diffuse water pollution from agriculture (DWPA) through various policy mechanisms: regulatory, financial and delivery of advice leading to voluntary adoption. Current examples in England include Nitrate Vulnerable Zones (NVZs), agri-environment schemes (AES) such as Entry Level Stewardship (ELS) and Higher Level Stewardship (HLS), and Catchment Sensitive Farming (CSF). At present, expenditure on AES in England alone is over 420 million per year (Natural England, 2014). With such costs, there is much debate regarding the effectiveness of schemes (NAO, 2010, p.24; Baker et al., 2012) and recommendations have been made that existing AES should be revised to ensure that policies are delivering the desired environmental objectives, including those set out by the WFD (e.g. Lawton et al., 2010). At the start of 2013, Natural England requested evidence and research to help inform the design of a new Rural Development Programme (Natural England, 2013). Evidence exists in the form of data on the current and historical uptake of measures included within AES, but for future uptake and for prospective measures, realistic assessments of the potential rates of implementation are required (ADAS, 2008 p.48). By attempting to understand farmers behaviours and attitudes regarding DWPA mitigation measures, it is hoped that more effective policy frameworks and mechanisms can be implemented. 1.2 Knowledge of farmer behaviours and attitudes Current knowledge of farmers behaviours and attitudes is largely derived through the use of farmer surveys, consultation with experts and the development of modelling tools. Farmers are regularly surveyed by researchers, the government and industry to document and assess the complexities of farming systems. Examples of such surveys include those carried out by initiatives such as Catchment Sensitive Farming, who provide farming advice to help reduce water pollution. They collect data on current DWPA mitigation measures uptake by recording which measures are adopted and recommended by their advisors (e.g. CSF Evidence Team, 2011). Farm practice surveys also provide ground truthing regarding adoption of different measures. Anthony (2011) highlights the importance of such surveys for checking modelling assumptions, whilst Pike (2008) notes that they provide business as usual snapshots which facilitate opportunities to observe behavioural change once new policy mechanisms have been enforced. Farm surveys are not only used to discover current farmer behaviour, but many researchers have used them to interpret an array of questions relating to environmental attitudes. Through understanding what do 4

5 farmers think? and why do farmers think that? it is hoped that there will be better understanding of how to encourage improvements in farming practices. Mills et al. (2013) and Fish (2014) provide recent reviews of the literature on such issues. Examples include research which has focussed on attitudes towards land use (Sutherland et al., 2011), the Single Farm Payment (Reading University, 2006), and AES uptake (Morris & Potter, 1995; Wilson & Hart, 2000, 2001; Wilson, 2014). Findings from these studies have also informed the development of conceptual frameworks such as that presented by Pike (2008) and shown in Figure 1 which encompasses a psychology-based approach to behaviours, a role for government intervention and considerations of behavioural economics. It also emphasises the need for a range of interventions described as the four E s (engage, exemplify, encourage, enable) in Defra (2005). In essence, such a model regards the adoption of a particular behaviour as a function of attitudes (practical expression of beliefs and values), surrounding norms (socially defined expectations of conduct), habits (frequency of past actions) and agency (real and imagined capacities to act). The latter, in turn, reflects a series of additional influences, which in an agricultural context can include those internal to the farm and farm household (e.g. size, tenure, age of decision makers) or external such as market conditions (Mills et al., 2013; Fish, 2014). However, the relative importance of these factors has been found to vary in different situations, so that, for instance, pro-change values may not translate into alterations in behaviour (Barr, 2002) and there can be feedback between elements (Pike, 2008). Some investigations have summarised their findings by developing typologies of farmers (e.g. Fish, 2014; Wilson et al., 2014), though others have suggested a need for caution when using a segmentation approach to enhance the design and implementation of policies (Burton, 2004; Wilson et al., 2013). A further caveat noted by Fish et al. (2003) is that AES uptake typologies are not mutually exclusive as land managers use different practices across their farms for a variety of reasons. The research presented in this report does not seek to create typologies, but aims to increase the knowledge of farmer attitudes towards a large variety of possible practices through the use of farm surveys. Figure 1. An integrated framework encompassing a psychology-based approach to behaviours, the role of government intervention and a consideration of behavioural economics (Source: Pike, 2008). The development of modelling tools and the use of expert guidance are other techniques employed to predict farmers behaviours and attitudes. To aid the design and implementation of DWPA polices in recent years, an inventory of possible measures (Newell-Price et al., 2011) and a decision support tool known as FARMSCOPER have been created for the Department for Environment, Food and Rural Affairs (Defra). The manual (hereafter the Defra User Guide) provides a detailed assessment of a wide variety of mitigation measures for DWPA, air pollution and greenhouse gas emissions. FARMSCOPER is a tool with the capability to model farm scenarios, providing outputs such as the amounts of pollution mitigated by changing various farm practices (Zhang et al., 2012). The data sets used to estimate the likely uptake of measures during the creation of the manual and the earlier versions of FARMSCOPER were however quite limited and not based on any formal auditing or monitoring scheme (ADAS, 2008). As many different factors influence adoption 5

6 (Blackstock et al., 2010) any data collected by surveying farmers on the likely uptake of measures would help improve the reliability of information included in such decision support resources (Anthony, 2011). Overall, the argument for the importance of social science, surveys and other qualitative methodologies within catchment management is increasingly being recognised. Anthony et al. (2009) and Zhang et al. (2012) highlight the need for more information on the realistic farmer uptake of measures and Macleod et al. (2008) suggested that acceptability of proposed future mitigation methods could be addressed through social research. Numerous datasets exist regarding river quality, with Collins (2013) listing fourteen different surveys which are conducted predominantly by the Environment Agency (EA). However, rather less attention has been given to the social dimensions of farm activity that can influence water quality. Many papers on catchment management and diffuse pollution (Collins & McGonigle, 2008; Gerrits & Edelenbos, 2004; Hewett et al., 2009; Macleod et al., 2007) also stress the need for improved inter-disciplinary collaboration amongst researchers and their interaction with stakeholders, especially policy makers. Combining knowledge gained from social science research, such as that presented here, with research from the physical sciences, will therefore help to provide a more robust evidence for agri-environmental policy makers. 1.3 Research objectives Given the above context, as part of a baseline farm survey in the DTC catchments it was decided that a set of questions on current practice and attitudes towards mitigation measures would be useful. The main objectives of the survey were to: Determine the nature of the farm businesses in the three catchments. Ascertain the current uptake of mitigation measures by farmers. Investigate the attitudes of farmers towards future adoption of measures. Identify any farm characteristics which influenced current or likely future uptake of measures. Discover which measures farmers prioritise for implementation. It was anticipated that the research would i) help decrease the data uncertainties within tools such as FARMSCOPER ii) provide a clearer understanding of the land management within the catchments being monitored and iii) help identify the mechanisms that may be required to encourage uptake of particular measures. It was also hoped that through the process of data collection and interpretation, an integrated and collaborative research community (including stakeholders) would be further developed (as discussed by McGonigle et al., 2014), as a step towards the shared understanding necessary for successful catchment management. 2. Farm survey methodology A farm baseline survey was designed to collect data on current farm practice and attitudes towards future uptake of DWPA mitigation measures. The survey design and collection of the farmer sample in each of the three main DTC catchments is described below. 2.1 Survey design The initial survey comprised a structured questionnaire covering such issues as AES participation, business structure and general farm attributes, and was based on a standard survey ADAS (an agricultural and environmental consultancy) use during farm visits. Questions were added regarding the current uptake of DWPA mitigation measures and attitudes to their future adoption. Many of the questions were in a closed response format, designed to aid comparability across farms and timely completion of the survey. As innovative mitigation measures are continually being developed, trialled and tested, it was challenging to design a methodology which would stay in date with the ever advancing literature. Numerous manuals and reports exist from various countries which list a multitude of potential measures - many being crop or region specific (e.g. Schoumans et al., 2011; Holsten et al., 2012). It was concluded that it would be most appropriate 6

7 to focus on the list of measures within the Defra User Guide (Newell-Price et al., 2011). This was the most comprehensive list of measures related to DWPA and relevant across UK farming, so allowing the research to provide a broader overview of current farm practices and attitudes than most previous studies. The complete list of mitigation measures surveyed can be found in Appendix A and a full description and assessment of the potential environmental and economic impacts of each measure is given in Newell-Price et al. (2011). During the survey farmers were asked Do you do x mitigation measure? If not, would you be very likely, likely, unlikely or never to consider doing it in the future? An example of the question format is shown in Table 1. A follow-up question after the list of mitigation measures asked farmers to state which three measures they would consider a priority to implement on their farm. Table 1. Example of the question format regarding infrastructure change on arable farms. Mitigation measure Re-site gateways away from high-risk areas Farm track management Establish new hedges Present use Yes No Would you consider doing this in the future? very likely, likely, unlikely, never As the majority of questions originated from a survey ADAS frequently use, it was deemed unnecessary to carry out an extensive pilot test. The survey questions were circulated to each of the DTC teams and structured into an acceptable format for interviews and postal surveys, as the need to gather information had to be balanced with the time-to-complete tolerance of survey participants. In the Wensum catchment, the survey was pre-tested on two Suffolk farmers to assess suitability of wording and timing. Layout and format changes were made so that the face-to-face interview section was as interactive as possible. For instance, a folder was given to farmers requiring them to tick and rank boxes to answer a range of questions as well as asking them to draw the farm boundaries on a map. These methods were included to keep the farmer engaged and to make the interview process interesting. A sample of the entire Wensum survey is provided in a separate online document available at Farmer sample As the survey was conducted as part of the DTC programme, the farmer sample was drawn from within the three DTC catchments, which were originally selected due to their differences in agricultural and environmental characteristics. Where the business address of a farm fell within the catchment, it was considered as a potential respondent for the survey, irrespective of whether some land was outside of the catchment itself. Farm businesses rather than holdings, were considered the most appropriate unit of study for this survey as clusters of holdings in a farm business are likely to be managed in a relatively uniform manner. However, during the survey the respondent was requested to only consider the land farmed within the study catchments. Various strategies were used to help recruit a representative cross-section of farmers within each catchment. Eighty eight surveys in total were carried out in the three catchments between February 2012 and February Frontier Agriculture and the Catchment Sensitive Farming Officer aided the Wensum catchment by suggesting possible farm contacts, whilst several of the participants were identified from previous activities by the Wensum DTC team e.g. attendees at meetings. Farmers were initially contacted by telephone to arrange meetings and recommendations from the initial group of participants provided further contacts to approach. In the Avon catchment, a questionnaire including the mitigation measures section of the survey was posted to all 86 farmers in the focus sub-catchments, along with a letter requesting a face-to-face interview to conduct the remaining farm business structure questions. The option of opting out of being contacted was also provided, but farmers who responded were then phoned to arrange an interview. In the Eden catchment, participants were targeted from within the Morland study sub-catchment where the majority of mitigation measures will be implemented as part of the DTC research. A handful of representative 7

8 farming types within the three other focus sub-catchments were also selected. Farmers who had previously engaged with the Eden Rivers Trust (ERT) were phoned by ERT employees to arrange a convenient time to conduct a face-to-face interview for the entire questionnaire. 2.3 Conducting the survey The surveys were conducted using face-to-face interviews and self-completion postal surveys, but the methods used for different sections varied between catchments. This was a pragmatic response to the survey resources available in each catchment. Ideally an identical approach would have been used in all three catchments, but all participants were still asked the same questions and the differences do not appear to have substantially impacted on or biased the information obtained. However, not all questions were relevant to all farmers (e.g. crop management mitigation measures for those only with livestock). During the interviews, irrelevant sections were omitted once the farm type had been determined at the start of questioning. In self-completion cases the farmers were provided with the entire list of measure questions and asked to leave inapplicable sections blank. The key topics covered in the survey are shown in Table 2, along with the number of responses received in each catchment and the methods through which data were obtained. Table 2. Topics in the DTC farm survey and the numbers of responses obtained through the use of face-to-face interviews and self-completed surveys. Wensum Avon Eden Face to Face Face to Face Face to Face Farm Type Selfcompletion Selfcompletion Selfcompletion Farm size 13 * 19 38** 18 Soils, drainage and waterways Land tenure Environmental schemes Mitigation measures Farm business and operational data * obtained from follow up telephone calls. ** 10 of which were obtained through Rural Land Register datasets The duration of face-to-face interviews ranged from forty minutes to three hours (with an average of sixty minutes), depending upon how much of the survey was covered in this manner. This represents a considerable time commitment by participants, for which the researchers are very grateful. The Avon and Eden catchment teams used DTC and Rivers Trust employees from either a farming background or with a substantial amount of field knowledge to conduct the face-to-face interviews. In the Wensum, two people conducted the interviews together, one a researcher, the other the farm liaison adviser in the DTC team (a former East Anglian farmer and agricultural consultant). Being highly knowledgeable in farming practices, the farm liaison adviser acted as a form of translator. Working as an interviewing duo proved highly successful. The use of experienced people from a farming background is considered a key factor by Blackstock et al. (2010), who report that such qualities convince farmers of the credibility of the survey and encourage an exchange of information. The self-completed surveys were either handed to farmers at the end of an interview or posted to them separately. Sometimes this approach was used because of the amount of information requested (e.g. regarding machinery and cropping practices on large arable farms). A disadvantage of this method was that the true meanings of some answers were difficult to ascertain. For example, a question mark in response to Would you consider doing this measure in the future? could be interpreted as meaning that they did not know what the measure was or were unsure whether they would adopt it in the future. If any such answers were ambivalent they were coded as missing data. 8

9 2.4 Survey data Information collected from the sections of the survey discussed in this report is listed below. The key farm attributes and involvement in environmental schemes were included to allow assessment of the effect these variables might have on behaviours and attitudes to mitigation measures. The farm business and operational data collected during the survey is not considered further in this report, but is being used as part of ongoing DTC research to help interpret water quality monitoring data and assess the cost implications of adopting different measures. i) Key farm attributes - Farm type, farm size and land tenure details. ii) Environmental schemes - Involvement in environmental schemes and farming on land in designated areas, as well as farmer awareness of and involvement with the Catchment Sensitive Farming (CSF) initiative. iii) Mitigation measures - Measures investigated during the survey were grouped using six categories from the Defra User Guide (Newell-Price et al., 2011, p.4) 1 : Land use change Livestock management Farm infrastructure Soil management Manure management Fertiliser management With respect to the mitigation measures, it was not the intention to further investigate the reasons as to why the activities had been undertaken or were likely/ unlikely to be considered in the future. Nor was it planned to ascertain if a measure had been implemented with or without external funding, or if an action considered possible in the future would depend upon funding. However, without prompting, many farmers willingly provided such reasons or explanations during interviews. These comments have been considered during analysis to provide context and depth. The following section presents the results and describes the key findings of the farmer survey. 3. Farm survey results The characteristics of the surveyed farms are described in Section 3.1, followed by classifications of the different mitigation measures in Section 3.2. The results regarding farmer behaviours and attitudes towards DWPA mitigation measures are then discussed in Sections 3.3 to 3.5. These sections particularly seek to: Ascertain the current uptake of mitigation measures by farmers. Evaluate farm characteristics which may influence the uptake of measures. Identify the attitudes of farmers towards future uptake of measures given the current economic and political environment. Investigate any associations between farm characteristics and attitudes to future uptake. Discover which measures farmers prioritise for implementation. 3.1 Characteristics of surveyed farms Farm type and land holding size Eighty farms provided details regarding their farm type. Defra s Robust Farm Type classification system was used, but cereal and general cropping were grouped together as arable, because many of the arable farmers surveyed did not distinguish which of the two provided the greater income. The June Census data for the main counties and unitary authorities encompassing each catchment (Defra, 2010a) was used to ascertain how representative the survey sample was in terms of farm type and size. Table 3 below calculates the percentage of farmers in each farm type from 1) the administrative areas as a whole, using data from the 1 The Defra User Guide and the original DTC proposal excluded analysis of pesticides due to the variety used within agriculture and the costs of sample analysis. Therefore pesticide management measures were not considered in the survey. Other research (e.g. as part of the Voluntary Initiative, has examined pesticide management issues. 9

10 June Census and 2) the surveyed farmers in each catchment. The farm type categories not included in Table 3 were under represented by surveyed farms and so have been omitted from analysis. For this reason the percentages for each catchment do not add to 100%. Table 3. The percentage of farmers 1) in administrative areas 2) in the survey, within each farm type category. County / Unitary Authority DTC catchment % of June Census data Arable % of survey sample % of June Census data Dairy % of survey sample Farm Type Grazing Livestock (lowland) % of June Census data % of survey sample % of June Census data Mixed % of survey sample Norfolk Wensum 56% 59% 1% 3% 16% 13% 8% 19% Portsmouth, Southampton & Hampshire CC Avon 35% 3% 4% 21% 45% 16% 7% 37% East Cumbria Eden 8% 0% 14% 28% 23% 39% 4% 33% Table 3 provides an insight into the different mixes of farming systems which occur across the three catchments and indicates that overall a fairly good representation of farming types were surveyed. There were, however, certain types of farming in particular catchments which were over or under represented. For example, the survey sample from the Avon was biased towards mixed and dairy rather than arable farmers. One explanation is that the area covered by Portsmouth, Southampton & Hampshire CC encompasses a large region in which a great variety of geology and soil types exist. As the Avon DTC has focussed efforts on predominantly chalk sub-catchments, it is not surprising that that was a different balance of farm types. Several other discrepancies also exist, but for the purposes of interpretation are considered less important than the fact that the sample composition clearly varies between the three catchments. In the Wensum the dominant farming system is arable; in the Avon sample mixed and dairy are most common; and in the Eden respondents lowland grazing livestock, mixed and dairy farms are well represented. In terms of farm size, Table 4 summarises how representative survey participants were, and for context provides detail of the proportion of land covered by the respondents within each catchment. The average size of sampled farm is greater in each catchment than the overall administrative area average. This is a reflection of the way in which survey participants were recruited. It is also apparent that average size was much larger in the Wensum than the other two catchments, with the respondents in the former accounting for just over 20% of the catchment area. Table 4. Farm size characteristics and the area (ha) of land managed by survey participants in each catchment. Area of land managed by survey participants (ha) Total area of catchment (ha) % of total catchment area covered by survey Average farm size of participants (ha) Average farm size in June census data 2012 (ha) Farm size range of participants (ha) Wensum 13,091 65, Avon 6, , Eden 2, ,

11 3.1.2 Land tenure Tenure data were collected from 61 farmers within the three catchments. The surveyed Wensum farms were predominantly owned, whilst the majority in the Eden were tenanted. In the Avon it was quite common for respondents to own some of the land they farmed but also rent additional land. In order to simplify analysis, farms were categorised according to the dominant type of ownership, resulting in Avon farms being predominantly classed as owned (see Table 5). Table 5. The number of farmers in each land tenure category within each catchment. Owned Number of farmers Grazing Tenancy license Contract Total Wensum Avon Eden Participation in agri-environmental schemes and land in designated areas Questions regarding participation in government-financed AES and farming in designated areas were answered by 77 respondents. Details of precise numbers of participants in each scheme can be found in Appendix B, however it is important to highlight that a higher proportion of surveyed farms participated in an AES compared to all farms in the administrative areas within which the catchments are located 2. Of the three catchments, the Eden had the greatest percentage of farmers participating only in the ELS scheme, but also the lowest share with HLS agreements. In the Avon a third of the respondents had a SSSI on their farm, reflecting the focus on chalk stream catchments. These statistics suggest that the survey respondents were more engaged with environmental schemes than the wider farming community, potentially influencing their responses. However, a benefit to such bias is that if such environmentally-minded farmers are reluctant to adopt certain measures then this suggests that there would be even greater challenges to increase uptake in the overall farming population Farmer awareness of Catchment Sensitive Farming Seventy-seven farmers were asked whether they had engaged with their local Catchment Sensitive Farming Officer (CSFO) as this was anticipated to be a potential influence on their responses (because of the advice given on DWPA measures by CSFOs). Forty-five participants responded in the affirmative, the percentages being 44% in the Wensum, 56% in the Avon and 89% in the Eden. To gain a greater understanding of the degree of engagement, the Wensum survey included further questions which revealed that 91% of farmers had heard of the initiative and 38% provided the correct name of the CSFO. 3.2 Categorisation of mitigation measures The measures studied were grouped into a number of categorisations to facilitate comparisons of current and future uptake. Although categorisation of measures into: mitigating pollution at source; slowing the pathway or protecting the receptor has a degree of fuzziness, experts such as local agronomists and authors of the Defra User Guide were consulted to provide validation of appropriate classifications. Similarly, measures were categorised according to the location on farm in which they would be implemented. A number of measures that do not occur in a particular location were described as all farm. Some measures are more effective than others, particularly when considering different target pollutants and the risks of pollution swapping. The Defra User Guide summarises the efficacy of individual measures by assigning up to three arrows to depict the anticipated degree of reduction or increase for each of twelve water and atmospheric pollutants. Working from this assessment, each measure was allocated an overall 2 Of Wensum farmers surveyed, 88% participated in AES compared to the 59% uptake in Norfolk, 78% of surveyed farmers in the Avon compared to 44% in Hampshire, and 100% of Eden farmers surveyed compared to 68% of Cumbrian farmers. Further details of AES uptake in each county are available at: 11

12 score by totalling the numbers of decrease and increase arrows to represent its effect on water pollutants. Measures with a more negative score were associated with the greatest reduction in pollution and are listed in Table 6. It was also important to consider whether a measure would be of benefit to other ecosystem services. Emphasis has recently been placed on AES options that provide a range of benefits (Lawton et al., 2010) as they provide greater environmental rewards to society for the funding invested, creating a win-win scenario. Certain DWPA measures are considered to help reduce soil erosion or greenhouse gas emissions, maintain food security, or improve flood management and biodiversity. A simple method was used to rank individual measures by assigning a point for each additional benefit it supplies. Points were again totalled and the highest scoring measures are listed in Table 6. Table 6. DWPA measures with the greatest provision of ecosystem benefits and reductions in water pollutants. Score Greatest number of ecosystem benefits Score Greatest reduction in water pollutants 5 Leave Autumn seedbed rough -16 Establish permanent woodlands 5 Convert arable land to unfertilised grass -14 Convert arable land to unfertilised grass 5 Establish permanent woodlands -14 Establish and maintain artificial wetlands 5 Establish in-field grass buffer strips -14 Grow biomass crops 5 Establish riparian buffer strips -14 Transport manure to neighbouring farms 5 Incorporate manure into the soil -13 Do not spread slurry/manure at high-risk times 5 Establish new hedges -12 Change from slurry to solid manure handling 5 Establish tree shelter belts -11 Incinerate poultry litter -10 Establish riparian buffer strips 4 Make use of improved genetic resources -10 Establish new hedges 4 Arable reversion to extensive grazing -10 Cultivate land for crops in Spring 4 Establish cover crops in Autumn -9 Establish cover crops in Autumn 4 Cultivate compacted tillage soils -9 Reduce overall stocking rates for livestock 4 Cultivate and drill across slope -8 Establish in-field grass buffer strips 4 Manage over-winter tramlines -8 Move feeders at regular intervals 4 Loosen compacted soil layers in grassland -8 Construct troughs with permeable base 4 Move feeders at regular intervals -8 Adopt reduced cultivation systems 4 Construct troughs with permeable base -8 Re-site gateways away from high-risk areas 4 Reduce overall stocking rates -8 Farm track management 4 Establish and maintain artificial wetlands -8 Arable reversion to extensive grazing -8 Fence off rivers/streams from livestock -8 Increase the capacity of slurry stores to improve timing of applications -8 Reduce length of grazing day/season -8 Reduce stocking rates when soils are wet -8 Construct bridges for livestock crossing 12

13 To assess the relationship between the two sets of scores the values for individual measures were plotted as shown in Figure 2. On this graph the measures have been colour coded into six different management types, and the centroid for each of these is highlighted with a star to indicate the average score. The centroid point for land use change measures scored the highest in ecosystem benefits and is most effective at reducing water pollutants. The scoring system suggests that farm infrastructure measures are better for reducing pollutants than soil management measures, but the latter provide greater benefits to the wider environment. Categorising measures in such a way can help inform the design of agri-environmental programmes and the identification of priority measures to help achieve objectives such as those of the WFD. Figure 2. The ecosystem benefit and pollutant reduction score for each mitigation measure. 13

14 3.3 Current uptake of mitigation measures by farmers Questions regarding their current uptake of measures were completed by 73 farmers. For each measure, participant s answered yes or no to the question Do you currently do (mitigation measure)? The numbers of farmers adopting each of 86 measures from the Defra User Guide are displayed in Figure 3 in descending order of frequency of yes responses. Not all measures were applicable to all farmers surveyed, resulting in fewer responses for such measures. Overall, current uptake greatly varied across the 86 measures. Measures which are compulsory for farmers to implement as part of cross compliance for the Single Farm Payments are highlighted in Figure 3 along with measures which, according to the Defra User Guide, have no substantial benefit to water quality. It is clear that these two sets of measures cluster at opposite ends of the graph, with compulsory measures related to manure and fertiliser management, not surprisingly, having the highest uptake. One other widely adopted measure - fertiliser spreader calibration stands out by not being highlighted. Consultation with agronomists confirmed that although this measure is not part of cross compliance there has been a significant drive for farmers to practice fertiliser calibration in Nitrate Vulnerable Zones (NVZs). Of the measures within the survey, those which are compulsory within NVZ also fall within cross compliance and so have not been identified separately. The results depicted previously in Figure 2 indicate that the fertiliser and manure management measures within cross compliance and NVZ regulations do help to reduce water pollution, but not as much as some other measures can achieve. The current uptake of measures requiring significant changes to land use also tended to be low. As Figure 2 implies that such measures can have the greatest benefits for ecosystems and in reducing water pollution, then this suggests that increasing their adoption would be one way of improving environmental quality. Some classifications of measures do not show any obvious clustering in Figure 3. For instance, this applies to the categories of source preventing, pathway slowing or receptor protecting measures (see Appendix C). Such an outcome is significant in itself because it suggests that the participating farmers had no particular preferences regarding adoption of these types of measures. It is acknowledged that the characteristics of the survey participants will have influenced the results obtained and chi-square tests were performed to examine associations between uptake of particular measures and variables such as participation in AES, farm size, tenure and type. The results are discussed in subsequent sections of this report, but it is worth noting here that the sample size did not permit more complex statistical analysis of multiple influences on uptake. 14

15 Figure 3. Current uptake of all mitigation measures from the Defra User Guide (Newell-Price et al., 2011). 15

16 3.3.1 Uptake of highly applicable measures The measures of greatest interest in this research were those related to DWPA mitigation which farmers have a choice to adopt (predominantly found in the mid-section of Figure 3). To assess these in more detail the measures which were applicable to 75% or more of the farmers surveyed were defined as high applicability and their current adoption is summarised in Figure 4. Currently adopted Not currently adopted Adopt field heap storage of solid manure Fertiliser spreader calibration Cultivate compacted tillage soils Maintain field drainage systems Incorporate manure into the soil Reduce fertiliser applications rates Cultivate and drill across slope Leave Autumn seedbed rough Establish riparian buffer strips Farm track management Early harvesting/establishment in Autumn Cultivate land for crops in Spring rather than Autumn Adopt reduced cultivation systems Establish new hedges Manage over-winter tramlines to reduce run-off Manure spreader calibration Compost solid manure Use clover in place of grass Use plants with improved nitrogen use efficiency Re-site gateways away from high-risk areas Establish permanent woodlands Convert arable land to unfertilised grass Use fertiliser placement technologies Loosen compacted soil layers in grassland fields Store solid manure heaps on concrete and collect Establish cover crops in Autumn Arable reversion to low fertiliser input extensive Establish and maintain artificial wetlands Cover solid manure stores with sheeting Grow biomass crops 0% 20% 40% 60% 80% 100% Percentage of farmers Figure 4. Current uptake of highly applicable DWPA measures (defined as applicable to 75% of farmers in the baseline survey). Some of the 30 measures in Figure 3 are supported by AES or other incentives, but a number of those towards the top of the list also provide an insight into what is considered as general good farming practice. Examples include cultivating compacted tillage soils and maintaining field drainage systems. It is also important to recognise that what is regarded as the norm is likely to vary between catchments. For instance, reduced tillage methods were relatively common amongst among Wensum arable farmers, but not in the other two catchments. To investigate current uptake further, the highly applicable measures were split into five equal-sized groups according their level of adoption. Mean scores were then calculated for each of these groups with respect to ecosystem benefits and reduction of nitrate (N), phosphate (P), sedimentation and all water pollutants, (using the approach described in Section 3.2). Analysis of Variance revealed no significant difference in ability to reduce water pollutants or provide ecosystem benefits between the quintiles of uptake. However the variations in mean scores shown in Table 7 do highlight certain contrasts. For example, the two groups with the lowest uptake had the best mean scores in terms of reducing overall water pollution - agreeing with the 16

17 observations regarding land use change. However the scores for ecosystem benefits showed less variation, with the lowest value corresponding to the least common level of uptake. Table 7. Mean scores for ecosystem benefits and pollutant reductions for each quintile of measures by current uptake. Grouping of measures Ecosystem benefit Total N impact Total P impact Sediment impact Total reduction in water pollution 1 (highest uptake) (lowest uptake) The group in Table 7 representing the highest uptake of measures had relatively low mean scores in terms of reductions in water pollutants. None of these measures featured in the list of those generating the greatest pollutant reductions (Table 6) and one (the maintenance of field drainage systems) was assessed in the Defra User Guide as actually increasing water pollution Influence of CSF and AES participation on measure uptake The CSF initiative offers advice and small grants to assist farmers in reducing water pollution. Measures encouraged through the CSF Capital Grant Scheme are predominantly farm infrastructure or manure management focussed, occurring within farmyards or around field boundaries. Re-siting gateways and farm track management were the measures from the CSF capital scheme list with the highest level of adoption in the survey. Engagement with a CSFO had a weak positive association with a farmer s adoption of storing solid manure heaps on concrete (x² = 2.75, d.f. = 1, p 0.1). As this measure is within the grant scheme, during a CSFO visit a recommendation could have been made about manure storage, raising awareness of the issue, and possibly resulting in a funding application to partially support uptake of the measure. No particularly strong trends were apparent between measure uptake and whether they were currently supported within AES. The seven ELS measures included in the survey encompassed different types of land management or location within the farm and had no obvious bias towards being source, pathway or receptor focussed. Of the twelve HLS measures, ten were in-field, many focusing on soil management or land use change. Certain AES measures such as establishing riparian buffer strips were found to be quite widely adopted, whilst others (e.g. cover crops or in-field buffer strips) were not. Farmers who participate in AES could be regarded as environmentally minded and so more inclined to adopt additional measures. As the great majority (87%) of surveyed farmers participated in ELS it was impractical to assess if this influenced current uptake. HLS participation was lower at 40% and was found to be significantly associated with four measures. Farmers in HLS were more likely to have established permanent woodlands (x² = 4.58, d.f. = 1, p 0.05), riparian buffer strips (x² = 3.22, d.f. = 1, p = 0.07), new hedges (x² =7.0, d.f. = 1, p = <0.01) and compost solid manure (x² = 2.86, d.f = 1, p 0.1). Even though reducing fertiliser application rates is currently within HLS options, many farmers who were not participating in the scheme were currently adopting such a measure, the same also being true regarding use of reduced cultivation systems in the Wensum catchment. Having an SSSI on the farm was found to positively influence the uptake of two measures. The 21% of farmers with a SSSI were more likely to have established permanent woodland (x² = 6.86, d.f. = 1 p = <0.01) and new hedges (x² = 8.15, d.f. = 1, p = <0.01) than those without. This may well relate to HLS participation, since the presence of an SSSI can be an important factor in the success of an HLS application. 17

18 3.3.3 Influence of farm size Variations in farm size were strongly associated with the type of farm and catchment, for example, the larger farms (>200ha) tended to be arable and in the Wensum. These farms were more likely to have undertaken farm track management and established permanent woodland, riparian buffer strips or (most significantly at p < 0.01) new hedges Influence of land tenure There are several types of farm tenure, with the vast majority of surveyed farmers classifying themselves as owners or tenants. This factor can substantially influence 1) the willingness of a farmer to invest time or finances into adopting particular mitigation measures and 2) the number of people involved in making such decisions (e.g. agents or landlords may need to be consulted). Infrastructure measures were more likely to have been implemented by farmers who owned their farm, but the contrasts were not statistically significant. The measures where there were significant differences, with owners being more likely to adopt them, all belonged to the soil management category and were relevant to arable farms. This result may therefore also reflect the influence of farm type, as the majority of arable farms were owned Variations in uptake by farm type Responses from farmers within each of four systems are considered in the following subsections. Measures found at the extremities of current uptake are highlighted, with those towards the middle (25% to 75% current uptake) being discussed in Section 3.4 concerned with future uptake. Arable Farms Almost all the 20 arable farmers who participated in the survey came from the Wensum catchment, with the exception of one from the Avon. The most popular measures with 100% current implementation amongst arable farmers included a number which are regarded as good farm practice such as fertiliser spreader calibration. Other measures which were carried out by 100% of arable farmers, but by a much smaller percentage of mixed farmers growing crops, included incorporating manure into the soil and reducing fertiliser application rates. Lowland Livestock Farms The distribution of livestock farmers was more evenly spread between catchments compared to the arable category (four in the Wensum, four in the Avon and six in the Eden). Measures with high uptake included reducing stocking rates when fields are wet as well as farm track management which 70% stated they had carried out. Moving feeders at regular intervals was implemented by 64%. Relevant measures with low uptake included covering manure with sheeting and only 14% had established new hedges (compared with 54% of all surveyed farms). Dairy Farms Several measures in the Defra User Guide are targeted at dairy farms but are not considered in this report as they focus on reducing air pollution and greenhouse gases. Ten dairy farmers in total were surveyed, with only one located in the Wensum. With such a small sample, interpretation of results must be treated with caution. Many farmers claimed to currently minimise the volume of dirty water and slurry produced, but few responded yes to other measures which would help to do this. The one exception was extension of the grazing season which nearly three-quarters of those who claimed to minimise volume of dirty water also adopted. Other measures which can help reduce volume - such as covering slurry stores and using liquid/solid separation techniques - had a very low uptake rate. Another uncommon measure was use of an anaerobic digester, with none of the dairy farmers currently operating one. 18

19 Mixed farms The 19 mixed farms surveyed came from all three catchments. As there were at least five in each catchment some comparisons can be made between the three areas. A higher uptake of fencing rivers and streams was found in the Avon compared to the Wensum. Re-siting gateways was much less common in the Wensum than the other two catchments with only one of six farmers stating they had made such a change. Reducing overall stocking rates and using clover in place of grass was far more common in the Avon than in the other two catchments. Only one of six Eden mixed farmers had established riparian buffer strips and yet all were in an ELS agreement. No mixed famers in the Eden had adopted reduced cultivation systems compared with high uptake in the Wensum Overall trends The current uptake of individual mitigation measures was quite varied and appeared to reflect a variety of factors. The limited sample size restricts the scope for statistical analysis, but some particular differences in practice by farm type and catchment were apparent. Wensum arable farmers acted relatively uniformly compared to mixed farmers growing crops in the other two catchments. Similarly, uptake of infrastructure measures for livestock farming differed amongst catchments. Assumptions regarding farmer behaviour cannot be made solely on the basis of farm type, but some consistency was evident within catchments. 19

20 3.4 Farmer attitudes to future uptake of measures For each question a farmer responded no to current uptake, they were then asked would you be very likely, likely, unlikely or never to consider doing it in the future? Attitudes to the measures which were applicable to 75% of farmers surveyed are shown in Figure 5, along with the rate of current uptake to place the measures into context. The measures of most interest are not those which are already highly adopted in the current economic and political environment, but those with a mid to low uptake. Knowledge of whether attitudes are more inclined towards positive or negative future adoption can help inform the use of appropriate policy mechanisms and the effort that may be required to encourage uptake. Results for measures which are applicable to all farmers are discussed first, followed by subsections presenting the results for measures relevant to each of four farming types. Currently done Future very likely Future likely Future unlikely Future never Adopt field heap storage of solid manure Fertiliser spreader calibration Cultivate compacted tillage soils Maintain field drainage systems Incorporate manure into the soil Reduce fertiliser applications rates Cultivate and drill across slope Leave autumn seedbed rough Farm track management Establish riparian buffer strips Early harvesting/establishment in autumn Cultivate land for crops in spring rather than autumn Adopt reduced cultivation systems Establish new hedges Manure spreader calibration Manage over-winter tramlines to reduce run-off Compost solid manure Use clover in place of grass Use plants with improved nitrogen use efficiency Re-site gateways away from high-risk areas Establish permanent woodlands Convert arable land to unfertilised grass Use fertiliser placement technologies Loosen compacted soil layers in grassland fields Establish cover crops in autumn Store solid manure heaps on concrete and collect effluent Arable reversion to low fertiliser input extensive grazing Establish and maintain artificial wetlands Cover solid manure stores with sheeting Grow biomass crops 0% 20% 40% 60% 80% 100% Figure 5. Current uptake and attitudes to future uptake of highly applicable DWPA measures (defined as applicable to 75% of farmers in the baseline survey) Attitudes to land use change and farm infrastructure measures Many of the land use change and farm infrastructure measures are applicable to all farm types. When considering land use change the measure most likely to be adopted in the future by the farmers surveyed was the establishment of woodland. Overall, however, land use change measures appeared to be among the least popular for future adoption (see Figure 5). Changes to land use could be perceived as too radical for a farming business, thus resulting in negative attitudes. Similar to land use change, farm infrastructure options may involve large commitments on the part of the farmer. Despite this, several measures such as farm track management, establishing new hedges and re-siting gateways all generally gained positive responses from farmers who had not already adopted them (Figure 5). 20

21 As with current uptake, attitudes varied between the different farm types. Table 8 summarises the current uptake and attitudes towards future adoption for measures which were applicable to over 70% of the farmers within each farming type. The measures are divided into four groups, separating current high uptake with little scope for future uptake from those with medium to low uptake. The latter are in three categories according to attitudes regarding future adoption, namely generally positive, mixed opinions and those which farmers would be unlikely to consider implementing. The key points from Table 8 are described within the following subsections. 21

22 High current uptake ( 75%) Table 8. Summary of surveyed farmers current uptake and attitudes to future adoption of DWPA mitigation measures. Medium to low current uptake with positive future attitudes Medium to low current uptake with mixed future attitudes Medium to low current uptake with negative future attitudes Arable Cultivate and drill cross slope Establish riparian buffer strips Early harvesting/establishment in Autumn Cultivate compacted tillage soils Reduce fertiliser applications rates Fertiliser spreader calibration Adopt field heap storage of solid manure Incorporate manure into the soil Adopt reduced cultivation systems Maintain field drainage systems Farm track management Establish new hedges Leave Autumn seedbed rough Use fertiliser placement technologies Re-site gateways Manage over-winter tramlines Establish permanent woodlands Use plants with improved nitrogen use efficiency Establish cover crops in Autumn Loosen compacted soil layers in grassland fields Grow biomass crops Store solid manure heaps on concrete and collect effluent Cultivate land for crops in Spring rather than Autumn Use clover in place of grass Irrigate crops to achieve maximum yield Replace urea fertiliser with another nitrogen form (e.g. ammonium Convert arable land to unfertilised grass Cover solid manure stores with sheeting Arable reversion to low fertiliser input extensive grazing Establish and maintain artificial wetlands Lowland Livestock Reduce field stocking rates if soils are wet Adopt field heap storage of solid manure Re-site gateways Move feeders at regular intervals Farm track management Establish new hedges Establish permanent woodlands Construct troughs with a firm but permeable base Fence off rivers and streams Compost solid manure Manure spreader calibration Cover solid manure stores with sheeting Establish and maintain artificial wetlands Grow biomass crops Reduce overall stocking rates Store solid manure heaps on concrete and collect effluent Construct bridges for livestock Establish tree shelter belts around livestock housing and slurry storage 22

23 High current uptake ( 75%) Medium to low current uptake with positive future attitudes Medium to low current uptake with mixed future attitudes Medium to low current uptake with negative future attitudes 23

24 Dairy Reduce field stocking rates if soils are wet Maintain field drainage systems Fertiliser spreader calibration Use anaerobic digestion for farm manures Reduce fertiliser applications rates Minimise volume of dirty water and slurry produced Construct bridges for livestock Use fertiliser placement technologies Install covers on slurry stores Use slurry injection application techniques Additional targeted straw-bedding for cattle housing Fence off rivers and streams Adopt reduced cultivation systems Store solid manure heaps on concrete & collect effluent Re-site gateways Use clover in place of grass Increase the capacity of slurry stores Use nitrification inhibitors Reduce dietary N and P intakes Establish new hedges Farm track management Loosen compacted soil layers in grassland fields Cultivate compacted tillage soils Make use of improved genetic resources Use plants with improved nitrogen use efficiency Ditch management Incorporate manure into the soil Cover solid manure stores with sheeting Establish tree shelter belts around livestock housing and slurry storage Transport manure to neighbouring farms Establish & maintain artificial wetlands Manure spreader Calibration Establish riparian buffer strips Compost solid manure Allow field drainage systems to deteriorate Grow biomass crops Establish permanent woodlands Out-wintering of cattle on woodchip stand-off pads Reduce length of grazing day/grazing season Reduce overall stocking rates Construct troughs with a firm but permeable base 24

25 High current uptake ( 75%) Medium to low current uptake with positive future attitudes Medium to low current uptake with mixed future attitudes Medium to low current uptake with negative future attitudes Mixed Cultivate land for crops in Spring rather than Autumn Cultivate and drill across slope Incorporate manure into the soil Farm track management Fertiliser spreader calibration Reduce field stocking rates if soils are wet Cultivate compacted tillage soils Adopt field heap storage of solid manure Adopt reduced cultivation systems Use plants with improved nitrogen use efficiency Make use of improved genetic resources Establish new hedges Maintain field drainage systems Establish cover crops in Autumn Use fertiliser placement technologies Move feeders at regular intervals Manage over-winter tramlines Reduce fertiliser applications rates Establish tree shelter belts around livestock housing and slurry storage Establish permanent woodlands Fence off rivers and streams Manure spreader Calibration Establish riparian buffer strips Loosen compacted soil layers in grassland fields Re-site gateways Compost solid manure Early harvesting/establishment in Autumn Grow biomass crops Arable reversion to low fertiliser input extensive grazing Establish and maintain artificial wetlands Reduce length of grazing day/grazing season Convert arable land to unfertilised grass Store solid manure heaps on concrete & collect effluent Use clover in place of grass Cover solid manure stores with sheeting Reduce overall stocking rates 25

26 3.4.2 Attitudes of arable farmers Several of the soil management options are very reliant on being suitable for particular soil types, e.g. leaving autumn seedbeds rough. Others are dependent upon field size, crop rotation or topography. Many in-field measures which would require a change in the current crop rotation or overall farm management received negative responses for future adoption. For example, the results suggest that altering the timing of cultivation or crop type through the establishment of cover crops would be unlikely to occur on arable farms. However, managing overwinter tramlines is an in-field measure which received positive responses for future uptake. Therefore potential exists for more farmers to adopt this measure. For riparian buffer strips, it is interesting that although uptake was high amongst arable farmers surveyed, those who do not currently have them are unlikely or would never consider introducing them. Many of the fertiliser management measures can be categorised as reducing the source of pollution and attitudes to future adoption were highly positive. Fertiliser inputs have increased postwar, but in recent years the cost of fuel and fertiliser along with unpredictable weather has resulted in a more cautious approach to usage. Many farmers surveyed did not want to waste fertiliser and stated they are likely to reduce application rates. New technologies such as variable rate fertiliser placement and improved genetics of N efficiency in crops received positive attitudes for future adoption across the board; however several comments were made during interviews regarding the difficulty of justifying the expense of machinery when it was believed the strength of evidence regarding effectiveness was weak Attitudes of lowland livestock farmers Overall there were more negative responses regarding the future uptake of livestock measures. Changes to farm practices may not be as popular for the livestock farmers surveyed as the majority were tenant farmers making it more difficult to implement change. The uncertainty of the economic environment for stock prices also possibly resulted in caution regarding measure uptake. Farm infrastructure measures related to keeping livestock out of rivers e.g. through fencing, received polarised responses for future uptake along with the measure having troughs with a firm but permeable base. Measures which could provide substantial improvements to reduce soil erosion received positive attitudes, such as moving feeders at regular intervals, farm track management and re-siting gateways. Reducing overall stocking rates is very effective in reducing many target pollutants if the land is too intensively farmed. The issue of food security was raised by farmers during face-to-face interviews as the increasing demand for local British meat exists. Nearly 40% have already reduced their stock, however all but two responded negatively regarding future reductions in stock Attitudes of dairy farmers The results indicated that the dairy farmers surveyed wanted to improve their current manure management as many of the measures in this category gained positive attitudes for future uptake. Covering slurry stores and increasing storage capacity were considered likely actions in the future and the majority were considering the use of anaerobic digesters. Establishing woodland received the most negative responses from dairy farms compared to any other farming type, but they gave more positive responses for other measures such as fencing off watercourses and using clover in place of grass. 26

27 3.4.5 Attitudes of mixed farmers The responses from the mixed farmers were quite mixed as illustrated in Table 8. Some measures which arable farmers rated negatively received positive answers from mixed farmers e.g. cover crops in autumn, and some measures received similar positive results to arable farmers, such as the use of technology. An example of a measure which predominantly gained negative responses from the third of farmers not currently doing it is using clover in place of grass. Unlike dairy farmers, the mixed farmers predominantly provided negative responses regarding future uptake of manure management measures. One example was the use of an anaerobic digestion for farm manures as the majority of mixed farmers stated they would be unlikely or never to consider using one. Comparing mixed farms between catchments, Avon farmers currently practice more measures overall than the other two catchments and provided more positive responses for future uptake. All mixed farmers in the Wensum were unlikely to adopt manure spreader calibration in the future, but all those in the Eden stated they would be likely to do so. Such differences emphasises the point made in Section regarding the importance of not categorising farmers merely by farm type. The attitudes to future uptake of mitigation measures amongst farmers showed consistencies amongst farm types when considering each catchment individually and identify a number of measures where there may be considerable potential for policy measures to encourage future uptake. Nevertheless, the limited sizes of sub samples means that some caution is needed when interpreting the findings. To complement these results, those regarding farmer priorities are presented in Section Farmer priorities To gain further insight into attitudes towards DWPA mitigation measures, the participating farmers were asked to prioritise three measures they would like to implement within their business, considering the list from the survey and any they could think of that were not included. Sixty-five farmers provided responses, with a fifth being content with their current farming practices and providing no priorities. The remaining farmers listed between one and three measures, resulting in 105 individual priorities stated. The majority of farmers with no priorities came from the Wensum catchment, had engaged with a CSFO and were in ELS, but only 55% were in HLS. Measures which had already been adopted by farmers with no priorities included the establishment of woodland and adoption of reduced cultivation systems. It was interesting to discover that even some of those who believed they had nothing further to change on their farm did not cover their manure, raising the possible need for greater advice on such a topic. The priorities mentioned were grouped into various categories to identify the most common characteristics. When classed into source, pathway and receptor measures, the greatest number of priorities can be described as slowing the pathway (47%), followed by source reduction (37%). Figure 6 categorises the priorities by a) management type and b) location on farm, showing that over half the priorities involved changing part of the farm infrastructure, with measures predominantly being within the farmyard. 27

28 Number of measures a) b) Farm yard Field boundary All farm Infield Farm infrastructure Manure management Fertiliser management Soil management Land use change Livestock management Agri-env scheme Management Type Figure 6. The frequency of priority measures mentioned by surveyed farmers. Categorised by a) management type and b) location. Table 9 lists the 10 most commonly cited priorities by individual farmers. The focus of responses was on additional concreting on farms, with a variety of uses raised. For example, concrete for manure heaps, diverting dirty water and track repair. Improved fertiliser and manure management ranked second, encompassing options related to correct timing and application efficiency. Cover and storage capacity for manure and slurry was also of high importance for farmers surveyed. Table 9. The most commonly cited priority mitigation measures. Mitigation measure Overall Frequency Concreting 17 Fertiliser and manure management 14 Increase manure/slurry storage 11 Manure/ slurry storage cover 11 Roofing in farm yard 9 Biobed 8 New Machinery or buildings 8 Fencing/repair walls 6 Collect rainwater 4 Plant trees/hedges/grass strips 4 Responses from farmers within the three catchments varied and are shown in Figure 7. There was interest, predominantly in the Wensum, for the management of waste water and establishment of reed biobeds to collect dirty water from farmyard wash-down areas (an option not included in the inventory manual). In the Avon, management of fertiliser/manure use along with covering manure were seen as a priority, compared to the common requests for farmyard roofing and increased storage facilities for manure and slurry in the Eden. 28

29 Number of farmers Wensum Avon Eden Fencing/repair walls New machinary or stores Biobed Farmyard roofing Increase muck storage Cover for manure/slurry storage Management of fertiliser/manure use Concrete No priority Figure 7. Farmer priority mitigation measures in the Wensum, Avon and Eden catchments. 4. The context of survey results and implications for agri-environmental policy The results from the DTC baseline farm survey presented in this report provide an insight into: the existing uptake of DWPA mitigation measures amongst farmers in the three DTC catchments; their attitudes towards future adoption of measures, whether positive or negative and what they prioritise implementing on their farms. In the following section, the findings are placed into a wider context through comparisons with other surveys and suggestions are made regarding the implications of the results for agri-environmental scheme design and policy. 4.1 Comparison of results with other surveys As one of the ultimate objectives of the DTC programme is to scale up research findings nationally, it was deemed important to compare the results from the baseline survey with those from the wider literature. Focusing on national and regional farm surveys allowed assessment of similarities and the nature of any contrasts. Caution must be taken when making such comparisons given possible variations in survey timing, sample composition and terminology used. Key DWPA mitigation measures from Section 3 which have the potential for wider adoption have been selected for particular attention, with similarities and differences between surveys discussed below National survey results Two annual national farmer surveys can be compared to the DTC survey. These are the Farm Business Survey (FBS) which provides information on the financial, physical and environmental performance of farm businesses in England 3, and the Farm Practice Survey (FPS) which looks at how English farming practices are affected by current agricultural and environmental issues 4. Neither of these surveys assesses a list of DWPA mitigation measures which is as comprehensive as the one included in this study, however they do offer an insight into behaviour and attitudes regarding particular measures. Both the FBS and FPS for the years corroborate the message highlighted within this report

30 that a great variation in measure uptake exists. In terms of uptake for groups of measures (e.g. fertiliser management being highly adopted compared to radical land use changes), direct comparisons prove difficult between surveys but generalisations can be made through consideration of particular measures within groups. The 2010 FPS (Defra, 2011) surveyed over 10,000 farms with the results consistent with those of the DTC survey regarding high uptake of fertiliser management measures (for example, fertiliser calibration), and low uptake of manure storage measures (for example, storage cover or plans to enlarge, upgrade or reconstruct storage facilities). The low popularity of manure storage measures also reflects similar results in a Scottish farmer survey (Barnes et al., 2009). In a similar manner to the DTC survey, the FPS and FBS results indicate that measures which are compatible with current farm practices are more widely adopted compared to those which are perceived as more radical changes. This is illustrated by livestock farmers having a high uptake rate for reducing stocking rates when soils are wet (Defra, 2010b) and a very low uptake rate for the measure of using an anaerobic digester (Defra, 2011). It is also apparent that the perception of what constitutes as an environmental measure or normal practice may influence farmers behaviour and attitudes. For instance, a survey investigating East Anglian arable farmers attitudes to environmental management found that participants believed that environmental activities should take place at field boundaries (Mills et al., 2013). However, many of the measures in the DTC survey that arable farmers were found to be currently adopting involved in-field management. These encompassed several that could be classed as normal practice and included cultivating compacted tillage soils or across slope. This difference implies that if environmental management is perceived as somehow separate from farming then measures involving field boundaries are more likely to be favoured, whereas if something is considered normal practice then it would be more likely to be adopted. In terms of surveys investigating the likelihood of future measure uptake and farmer priorities, a limited amount of research exists for measures discussed in this report. However, two examples are worth highlighting. Firstly, results from a CSF survey in 2012 supported the DTC finding that improving farm infrastructure is a future key priority amongst farmers (Environment Agency, 2013). The second example illustrates changes over time with the 2011 FPS reporting that only 3% of farmers planned to have an anaerobic digestion plant in the future (Defra, 2011), but the DTC survey found that nearly a third would be likely to - with the greatest interest coming from the sampled dairy farmers. This difference may well reflect differences in sample composition, but it is undoubtedly also influenced by the rapid changes in policy regarding renewable energy incentives which have stimulated a substantial recent expansion in use of anaerobic digestion (Biogas-Info, 2014) Regional comparisons of surveys results In terms of overall measure uptake, the DTC survey found farmers in the Eden (North West) to have the lowest rate, whilst those in the Wensum (East Anglia) had the highest. These differences are in line with regional variations in the CSF survey on recommendation uptake (Environment Agency, 2011). Not surprisingly, certain types of measure vary in adoption across the country, with some appearing to be normal farm practice only in specific regions. The DTC survey and Defra FPS survey (Defra, 2012a) both identified East Anglian arable farmers as having a far higher rate of uptake, and more positive attitudes, towards technological measures than the corresponding national averages. Examples of such measures include computer controlled, variable-rate, fertiliser spreaders and reduced cultivation methods (Defra, 2013). Such common practices in East Anglia reflect the nature of the farming businesses and the favourable financial circumstances for many farmers compared to elsewhere (Defra, 2012b). 30

31 Equally unsurprisingly, the uptake of livestock specific measures was more common outside East Anglia. One example was the adoption of clover mixes. The 2012 FPS identified the South West and West Midlands as the English regions with the greatest proportion of farmers sowing 100% clover mixes (Defra, 2012c), with the DTC survey also finding the South West famers had the highest rate of adoption. However, the DTC results also suggested that there could also be potential to increase uptake elsewhere, with dairy farmers in the Eden having positive attitudes towards sowing clover in the future. Overall it is positive to note that where comparisons can be made the DTC baseline results generally agree with those from other surveys. This provides confidence in a distinctive feature of the DTC survey regarding the investigation of uptake and attitudes towards a larger range of potential mitigation measures than has been previously researched. Comparison with the wider literature has provided some verification of findings and also identified potential opportunities for the increased uptake of measures in other regions. 4.2 Implications for agri-environmental policy Policy makers face the challenge of re-designing mechanisms to effectively reduce agricultural diffuse water pollution whilst ensuring policies deliver consistently across a range of other desired outcomes or societal needs (e.g. ecosystem services and national food security) (McGonigle et al., 2014). The findings of this study improve the evidence base regarding current farmer behaviours and attitudes towards possible future changes that could contribute to the mitigation of diffuse water pollution. Interpretation of the survey results also provides insights relevant for recent and ongoing discussions occurring within government regarding the reformulation of agri-environmental policy in the context of CAP reform and WFD implementation. To assist in this process the mitigation measures investigated in this report have been categorised into the following four groups (see Table 8): i. High current uptake with little scope for future uptake ii. Mid to low current uptake with positive attitudes for future uptake iii. Mid to low current uptake with mixed attitudes regarding future uptake iv. Mid to low current uptake with negative attitudes for future uptake These categories can be linked to the policy mechanisms (the 4 E s) shown in Figure 8, a framework described in Pike (2008) and originally taken from the UK Sustainable Development Strategy (Defra, 2005). Making such associations helps to inform decisions as to which mechanisms may be most appropriate to address internal and external barriers or social factors preventing greater uptake of particular mitigation measures. Pike (2008) describes how regulatory and market-based instruments should focus on external factors making desirable behaviours easier/cheaper and those less desirable harder/more expensive. He then discusses how internal barriers can be addressed through communication, advice and other engagement options to influence attitudes and social norms. To increase adoption of individual mitigation measures it is likely that varying proportions of the four mechanisms will be needed and deciding upon an effective balance of emphasis is considered to be crucial for policy success. 31

32 Figure 8. Mechanisms to increase the adoption of mitigation measures (source Defra, 2005). Measures in category (i) which already have a high uptake rate, such as riparian buffer strips, may have reached an upper limit in terms of adoption. For instance, many of the Wensum arable farmers who did not have them did not intend to introduce them. This potential saturation implies that the most effective policy options are probably those which lead the measure to become a social norm. Consequently it is questionable as to whether efforts through incentives (enable) or advice delivery (engage) should be pursued to change the attitudes of the relatively small percentage of farmers involved. Measures in this category would probably benefit most from a high proportional use of encouragement, for example through inclusion in the greening options required under the CAP. Adequate evidence would need to be provided to justify the change of policy, as resistance or dissatisfaction can occur when using a regulatory approach (Barnes et al., 2009). Certain mitigation measures had positive attitudes regarding future adoption (category ii) and were also named as priorities by survey participants. These are measures where the government could increase uptake through relatively simple mechanisms. Leading by example through providing good demonstrations (exemplify), raising awareness of benefits through initiatives such as CSF (engage), and small incentives (enable) would be anticipated to improve adoption. Measures in this category with positive attitudes included using plants with improved nitrogen efficiency, reduced cultivation systems and re-siting gateways, whilst those which farmers prioritised encompassed many forms of other infrastructure improvements. These measures deliver multiple benefits and reduce the source or slow the pathway of pollution rather than protect the receptor. If offered as part of an AES, it could be anticipated that many farmers would choose to adopt such measures. However, it is likely that there would still be variations in uptake by farm type and location. One example is reduced cultivation methods, since although these received positive responses from arable farmers in the survey, there were comments that their suitability will vary according to soil type or weather conditions and also depends on having the financial resources to purchase additional equipment. This raises the question is AES the most effective mechanism to increase uptake? and shows how important it is to consider each measure individually and the differing balance of the four E s needed to increase uptake. Measures with mixed and polarised views regarding future uptake (category iii) represent an opportunity for wider adoption but imply that higher levels of effort may be required to achieve the desired outcomes. A greater use and emphasis on engagement and exemplify mechanisms could be of benefit for such measures. One example is the use of cover crops which were included in ELS options from 2010, but have featured in relatively few agreements. However, providing advice about the long- term benefits of such crops for soil and nutrient management, as well as the wide variety of mixtures now available (e.g. Kings, 2014), through CSF and other sources of agronomic advice has been an important factor behind a recent increase in farmer interest and the inclusion of cover crops as an Ecological Focus Area option under the new CAP arrangements (Defra, 2014) is also likely to enhance their use. 32