Survey Results

Transcription

1 Survey Results

2 Introduction & Executive Summary 1 Polli:Nation Survey Method 2 Summary of Effort and Uptake 3 How many people took part and what experience did they have? 3 Making improvements 4 Where did the surveys take place? 5 Survey site information 6 Part A. Habitat Results 7 Which species of plant were present and were they in flower? 8 What habitat improvements were made? 10 Part B. Pollinator Results 12 Summary of the pollinators found 13 Factors influencing pollinator numbers 15 Habitats 16 Flower species 17 Weather 18 Pollinator response to habitat improvements 18 Whole data set 18 Groups who surveyed before and after changes were made 19 Species Quest 20 Key Findings 21 Acknowledgements 22

3 Polli:Nation is a Heritage Lottery funded UK-wide initiative supporting pupils from 260 schools to turn their school grounds and other local walk-to spaces into pollinator friendly habitats. As part of this programme, the OPAL Polli:Nation Survey was developed to: Survey their patch using the OPAL Polli:Nation survey Make improvements for pollinators on their patch See how well it has worked using the OPAL survey to see the impact the improvements have had. This cross-curricula project is giving participants (aka citizen scientists) direct hands-on experiences; from creating vertical green walls and night-blooming flower beds to lobbying to change school maintenance regimes and debating pesticide use. Citizen scientists are learning about the role pollinating insects play in ecosystem services and are able to contextualise this in the choices and actions they take. Alongside creating a network of knowledgeable and enthused young activists, the ambition of this programme is to utilise school grounds and other outdoor spaces to form local green corridors and stepping stones, enabling species to move between core areas thereby contributing to the overall aim of the project by increasing numbers and sightings of pollinating insects in the UK. The aim of the project is to engage and enthuse children and young people to protect pollinating insect species in the UK. The outcome of the project is to see an increase in the number and diversity of pollinating insect species in school grounds and local community spaces across the UK. This report summarises the key findings of the second survey season (2017) and makes comparisons to data collected since the survey began in Polli:Nation is a partnership project with the following organisations: 1

4 The Polli:Nation survey is a large-scale national survey that is providing answers to important research questions about the health and status of pollinating insects across the UK. A core part of the project is to engage pupils and the community in biological recording within school grounds and other local green spaces. The Polli:Nation Partnership designed the survey method to investigate the number of pollinators using different habitats at survey sites. For those registered schools within the Polli:Nation project the survey is also tailored to track the impact of positive changes made to school grounds to encourage more pollinating insects. The survey is structured in two parts: A) Habitats: This part of the survey looks at what is on offer for pollinators in survey sites (10m x 10m areas selected by participants) within the school ground in terms of providing feeding, nesting and sheltering resources. Participants were asked to map the site onto a grid (see below) and count the number of 1m x 1m squares containing different habitat types and also record which plant species are present. B) Pollinators: The second part of the survey involves counting the number of pollinating insects visiting flowers in quadrats within the survey sites. For each pollinator survey participants recorded the weather conditions, habitat type, floweriness and most common flowers within a quadrat. The pollinators were recorded in 9 different groups: Bumblebees, Honeybees, Solitary bees, Butterflies, Moths, Hoverflies, Other Flies, Beetles, Unidentified insects. Participants could also take part in the Species Quest by identifying a suite of pollinators from a list of 12 species (see below). 1

5 Highlights 807 sets of survey results were received from 291 groups over 2 years 54% of results are from Primary Schools; 17% from Secondary Schools & 29% from public groups (63% of all groups were Polli:Nation registered) The highest number of submissions was 28 from Bryncethin Primary School in Bridgend, Wales 107 groups have been introduced to identification and there has been a 5% rise in the number of primary schools with previous experience Data gaps are found around East Anglia, the South West, the North of England, Cumbria, the Highlands and Islands of Scotland How many people took part and what experience did they have? Polli:Nation is a rolling survey where participants continue to submit survey data in multiple years. Therefore the results are summarised by year of the survey (as opposed to year results submitted) and type of participant (Polli:Nation School or not). Over the whole survey so far, we have received 807 sets of results from surveys (Table 1). Of these surveys, over two thirds were submitted in The number of results submissions by Polli:Nation Schools has fallen by 63% between survey years, whereas non Polli:Nation group submissions have increased by 7.5%. Over the first two years of the survey, 74% of results were submitted by Polli:Nation registered schools. Table 1: Number of results submissions by type of group and year survey took place. Survey year Polli:Nation School Other groups (not Polli:Nation registered) Total Total Across the two survey periods of 2016 and 2017, 291 different groups have taken part (Table 2). The number of groups submitting data has halved between 2016 and The drop in submissions and groups taking part is likely due to the fact that half of the schools were in rest year following their 2

6 initial participation. Without a facilitator in this year, schools may have been less likely to undertake the surveys. Table 2: Number of individual groups that submitted results by type of group and year that they took part. Survey year Polli:Nation School Other groups (not Polli:Nation registered) Total Across the two survey seasons, the highest number of submissions were received from registered Polli:Nation Primary Schools (55%) and Secondary Schools (17%). The remaining 28% of submissions were received from volunteer groups, family/friends groups, youth groups, colleges/universities or other. Figure 1: Number of submissions by different groups taking part in Polli:Nation. Table 3: Number of groups taking part by type of group and whether or not they are registered as part of the Polli:Nation project. Type of group Polli:Nation School Other groups (not Polli:Nation registered) Total Primary School Secondary School Other (e.g. nursery &middle schools etc) Family or Friends Adult Volunteer Group Youth Group 8 8 College / University None given 3 3 Total Considering the number of results submissions per group across the two survey years, 50% of groups submitted one set of results, 19% of groups submitted 2 sets of results and a further 31% submitted 3

7 at least 3 sets of results. In total, 17 groups (6%) have submitted more than 7 sets of results. The highest number of submissions received so far is 28 (14 in each survey year) from Bryncethin Primary School in Bridgend and close behind with 25 submissions is Llanishen Fach Primary School in Cardiff. Figure 2: The number of results submissions per group A higher percentage of groups had some previous experience in identifying insects (62%) compared to 38% who had never identified insects before. This means that for 107 groups- Polli:Nation has introduced people to identifying insects for the first time. In the first survey year 55% of primary school groups and 62% of secondary school groups had some previous experience of identifying insects. Adding in the 2017 data, these percentages have switched: 60% of primary schools and 56% of secondary schools have experience of identifying insects. Figure 3: The number of different types of groups and the previous experience with identifying insects. Making improvements The aim of the Polli:Nation survey is to compare before and after any positive changes are made at a site to investigate how pollinators have benefited. 4

8 From the 291 groups that have participated in the first 2 years, 67% (N= 195) were proposing to make changes to the school grounds. Of these 195 groups, 167 were planning to make positive changes for pollinators through the Polli:Nation project. Figure 4: Number of groups planning improvements to their local patch at their involvement with the Polli:Nation project. As part of the survey, participants had to record which kinds of changes they made to their school grounds. In total, 56 groups (28% of those that planned to make changes) submitted responses after making improvements for pollinators- of these 43 were registered Polli:Nation schools (25% of all registered schools that planned changes) and the remaining 13 groups (46% of all other groups submitting results that planned) made their own changes. These 13 groups included schools but also universities, colleges, adult volunteer groups and groups of families and friends. There is a remaining season to run on the survey so these figures may rise. Where did the surveys take place? The Polli:Nation Survey is open to anyone to take part. Figure 5 shows where the different groups took part around the UK. The highest proportion of groups were located in England (46%) followed by 26% in Scotland and 16% in Wales and doubling of participation in Northern Ireland from 16 groups in 2016 to 32 (11% of all groups) by the second survey year (Table 4). Overall the coverage of the UK is good with notable improvement in Northern Ireland yet there still remains significant gaps in East Anglia, the south west, Cumbria, the north of England, and the Highlands and Islands of Scotland. Table 4: Number of groups participating from each of the UK countries and Ireland. Country Polli:Nation School Other groups (not Total Polli:Nation registered) England (46%) Scotland (26%) Wales (16%) Northern Ireland (11%) Ireland 1 1 (0.3%) Total

9 Figure 5: Location of the 291 groups across the British Isles that have submitted records since the survey began. Survey site information Before the actual surveys of the pollinators took place, participants had to choose a 10x10m survey site within their school grounds or local patch. Participants were asked to map the site onto a grid and count the number of 1x1m squares containing different habitat types and also record which plant species are present. This section describes what habitats and flower species were found in the 684 surveys at sites either before any improvements were done or in surveys where improvements weren t planned. 6

10 Highlights Over 11,000m 2 of wildflowers were recorded as part of the Polli:Nation survey Short grass covered the largest area recorded (over 26,000m 2 ) and while playing fields are important, more nesting and shelter habitat could be created for pollinators Daisy, buttercup, dandelion and clover are still the most common flowers There was a 2% increase in the number of surveys containing at least one species of flower compared to now at 95% 168 positive habitat changes for pollinators were made to 55 school and community locations around the UK Across 684 surveys (excluding the surveys carried out after improvements were made) over the 2 years, the most commonly recorded resource type was short grass (Figure 6). The area of short grass surveyed increased by 26% between 2016 and 2017 (Table 5). This represents huge potential for improvements for pollinators. The second most common habitat type encountered was wildflowers with 91 survey sites at least 50% covered in wildflowers and at 21 sites the whole area was wildflowers. Another interesting result was that there has been an increase of 202m 2 of manmade pollinator habitat compared to last year (bee hotels and bee hives). The main difference between different groups was that schools surveyed a lower area of wildflower habitat that the other groups (universities, volunteer groups etc). Specifically 16% of primary school and 19% of secondary school survey areas were wildflowers compared to 27% of habitat area surveyed in non-school groups. This demonstrates the attractiveness of the Polli:Nation survey to other groups to monitor pollinators in areas where they might actually be seen. Figure 6: The proportional area of different habitats recorded in the survey sites 7

11 Table 5: The area recorded of each habitat resource and the number of sites at which each resource constituted at least 50% of the area of the site. Resource type Habitat type Total (m2) in 2016 Additional area surveyed in 2017 Total (m2) 2016 and 2017 Feeding habitat Flower beds or pots Wildflowers Trees Total Nesting and shelter Bare ground Man-made pollinator homes Damp places Total Other habitat Bare walls or fences Concrete or tarmac Short grass Total Which species of plant were present and were they in flower? The types of plants found in survey sites (excluding the surveys conducted after improvements had been made) are described here. From a list of 8 woody plants, 8 garden plants and 8 wild flowers the most frequently recorded species in flower within the survey sites was the daisy (64% of submissions) followed by buttercup (50%), dandelion (42%) and clover (42%) (Table 6). Interestingly there are some increases in the floweriness of survey locations between the two years the presence of flowering species such as lavender, buddleia and mint have been recorded on 3-4 times more surveys in 2017 compared to Given that this data set excludes surveys carried out after improvements, this increase in floweriness could be due to more interested groups focusing in on areas of interest which are already supporting pollinators. Table 6: The percentage of submissions which recorded each of the 28 pollinator-friendly plant species in flower and not in flower. Plant type In flower in 2016 (% submissions where recorded) In flower in 2017 (% submissions where recorded) Daisy Buttercup Dandelion Clover Nettle Bramble Lavender Umbelifers 13 9 Vetches 12 9 Dead-nettle 9 16 Buddleia

12 Plant type In flower in 2016 (% submissions where recorded) In flower in 2017 (% submissions where recorded) Thistle 9 11 Fruit tree 10 9 Wild Rose 7 13 Chives 7 15 Ragwort 7 13 Willowherb 7 12 Hawthorn 7 11 Ivy 7 7 Mint 4 17 Knapweed 6 7 Verbena 5 7 Willow 5 7 Heather 4 7 Marjoram 4 6 Blackthorn 2 7 Sedum 3 4 Sea Holly 1 3 In terms of the number of species recorded (diversity), only 5% of submissions across the two survey years (excluding those carried out after improvements) had no flowering plants present at all. This is good news as it means that 95% of quadrats had some flowers in them- this is great news for pollinators. For those species in flower the most common level of species diversity was 3 species (17% of submissions) followed by 4 species (15%) (Figure 7). Figure 7: The number of submissions recording different numbers of species flower in the survey sites 9

13 What habitat improvements were made? A key element of the Polli:Nation project was to make improvements to the outdoor areas where people live and work and the survey aimed to track these changes. In the previous sections we have looked at who has taken part in the Polli:Nation Survey and we have described the habitats and characteristics of the surveys that haven t been entered as the results following a positive change for pollinators. Here we look at data from those groups who indicated that they made changes. In total, 55 groups recorded that they had made at least one change to their site. At these 55 locations, 168 positive changes were made for pollinators at schools and community groups across the UK. Of these 55 groups, 42 were Polli:Nation registered schools and 13 were other groups. Among the groups that recorded positive changes, the most popular habitat to create has been to add pots, build raised beds or plant flower beds with 42 groups carrying out this work ( Figure 8). A further 36 groups created a wildflower meadow or verge. From the 55 groups the most popular number of changes made to a site was 4 changes (23.6% of groups made 4 changes) (Figure 9). From these 55 groups, 41 groups submitted 216 paired surveys that were carried out before and after improvements had been made. The impact of the changes is partly seen from the average area of habitat surveyed in the 10x10m survey sites (Figure 10). There has been a decrease in the amount of shortgrass surveyed and an increase in the amount of wildflower habitat and a slight increase in the amount of flower pot habitat. Figure 8: The number of different groups registering different positive pollinator improvements on their patch 10

14 Figure 9: The number of changes made to school/site grounds by the 55 groups Figure 10: The average area of habitat in the 41 survey sites before and after changes were made 11

15 Highlights The total number of individual pollinators recorded was 11,793: 6105 in 2016 and an additional 5688 in 2017 The most frequently recorded pollinator type was other flies (39%) followed by unidentified insects (16%) and bumblebees (11%) Fewer honeybees were recorded in 2017 as a proportion of all pollinators, this may not indicate population changes but rather improvements in identification skills (fewer people confusing them with hoverflies & wasps etc) Habitats with the highest average count of pollinators were damp places and trees with an average of 8.2 pollinators per 2 minute quadrat survey Quadrats containing Verbena plants attracted the highest number of pollinators (86 pollinators per 2 minute search) whereas Dead Nettle attracted the highest diversity (7 groups of pollinators on average) Across the whole data set, 4.24 pollinators were counted in quadrats before improvements had been made compared to 6.68 pollinators after In locations where surveys had been completed before and after improvements were made there was a statistically significant difference in the average number of pollinators and a significant correlation between the area of (pollinator friendly) habitat and the number of pollinators The pollinator part of the survey involves counting the number of pollinating insects visiting flowers within a 1m x 1m quadrat over a 2 minute period. In this section, the results are investigated at the quadrat level with respect to factors such as weather conditions during the survey, habitat type, floweriness and most common flower. Considering all the data provided from 807 surveys over 2 years we now have data from 1988 quadrat searches each lasting 2 minutes which equates to over 66 hours of survey time which is 11 days of school! Combining all the data collected so far, we can see that the most surveys and hence the most pollinators were observed in the months of June and July (Figure 11). The average number seen per quadrat search is less predictable across the season yet for total pollinators there were peaks in average abundance in the spring and summer (Figure 12). 12

16 Figure 11: Total pollinators recorded through the season combined across the 2 survey years Figure 12: Average number of pollinators recorded per quadrat survey across the survey season (2016 and 2017 combined) Summary of the pollinators found Across the 1988 quadrat surveys in , the most popular habitat for the pollinator surveys wildflower habitats (34.4% of quadrats) followed by short grass (28.5%) and flower beds or pots (21.1%) (Figure 13). 13

17 Figure 13: The proportion of different habitat types surveyed in the 1988 pollinator surveys Within each quadrat, participants recorded the level of floweriness in 3 categories: less than half, half and more than half. The most flowery quadrats were found within flower beds or pots and wildflower habitats (Figure 14). Similarly, the least flowery quadrats were surveyed on concrete/tarmac, bare walls/fences and bare ground. The most common flower recorded was a daisy (26% of quadrats) followed by buttercup (11%) and clover (9%). Figure 14: Percentage of quadrats within different habitats and the level of floweriness The total number of individual pollinators recorded was 11,793. In 2016 there were 6105 pollinators counted and an additional 5688 were recorded in Comparing the two survey years in terms of the percentages of pollinators in the different groups there are a few changes between the survey seasons. Fewer honeybees were recorded in 2017 as a proportion of all pollinators seen (Figure 16). A greater proportion of solitary bees and beetles were recorded in

18 Figure 15: Proportion of different pollinator groups recorded in the pollinator surveys Figure 16: The proportion of pollinator types making up the survey results between years Factors influencing pollinator numbers The next step to answering this question is to look across all the data collected in the two years to see if there is a relationship between the numbers of pollinators and the various factors that can influence their numbers. The results from a preliminary statistical analysis using a Generalised Linear Model showed that: The more flowery the quadrats, the more pollinators were seen There was a significant negative effect of habitats classified as concrete, bare walls fences/short grass) compared to feeding habitat Low temperature and partly sunny weather had a negative effect on the number of pollinators seen Partly sunny weather had a positive effect compared to cloudy weather The area of pollinator friendly habitat (feeding, roosting and nesting) within the survey site had a significant positive effect on the number of pollinators The amount of land use classed as urban within 1km of the survey location, did not have a significant effect on the number of pollinators seen The following sections look at all these factors in more detail. 15

19 Habitats Looking at all that data, the habitats with the highest average count of pollinators were damp places and trees with an average of 8.2 pollinators per 2 minute quadrat survey (Figure 17). Interestingly, quadrats classed as bare walls and fences came third with an average of 7.6 pollinators per quadrat and then wildflowers came 4 th with 7.2 pollinators per quadrat. The habitat supporting the lowest numbers of pollinators was short grass (3.3) and concrete or tarmac (2.6) (Figure 17). Considering pollinator diversity across the whole survey of 807 responses, damp places consistently return the most different types of pollinators followed by wildflowers and then trees (Figure 18). Figure 17: Average number of pollinators per quadrat survey for each habitat type (+/- Standard Error) Figure 18: Average pollinator diversity per quadrat survey (+/- Standard error) 16

20 Flower species We can also look at the average number and diversity of pollinators seen across all surveys with respect to the most common flower present in the quadrat. In terms of numbers of pollinators, the data are particularly skewed by some very popular Verbena plants which attracted huge numbers of pollinators- an average of 86 pollinators per 2 minute quadrat search was recorded on Verbena (from a sample of 7 quadrats containing Verbena). In second place was bramble which, from a sample of 51 quadrats, supported on average, 27 pollinators per 2 minute search (Figure 19). Considering the effect of dominant flower present and the diversity of pollinators (number of different groups recorded) the most diverse flowers were dead nettle (an average of 7 groups recorded), thistle (5.8) and verbena (5.3) (Figure 20). Figure 19: Average number of pollinators per quadrat survey for each flower type (+/- Standard Error) Figure 20: Average diversity of pollinators per quadrat survey for each flower type (+/- Standard Error) 17

21 Weather The chances of seeing pollinators actively pollinating flowers is largely affected by the weather and therefore for each 2 minute survey, participants recorded the weather (sunny, partially sunny or cloudy) the temperature and the wind (leaves on trees still, leaves moving gently all the time, leaves moving strongly). The surveys were carried out in good weather conditions. 77% of the surveys took place in sunny or partly sunny conditions and only 9% of the surveys took place when the wind was strong enough to make leaves move strongly. 76% of surveys took place when the air temp was above 15 C which is a good temperature for pollinating insects to be active (Figure 21). When comparing the total number of pollinators seen with temperature, there is a significant correlation such that more pollinators are seen in warmer temperatures (Pearson's product-moment correlation, t = , df = 1911, p<0.01). Figure 21: Number of quadrat surveys undertaken at different temperatures Pollinator response to habitat improvements As mentioned in Part B, a key aim of the Polli:Nation Survey was to look at the impact of positive changes to a site to increase the number of pollinators. Whole data set In the tables below the total number of pollinators and total numbers of quadrats surveyed in each year are shown (before and after improvements were made). Using these figures, on average 4.24 pollinators were counted in the before quadrats compared to 6.68 pollinators per quadrat in the after surveys. Table 7: The total number of pollinators recorded each year as before or after improvements were made at the site Year Before After Timing not Total recorded Total

22 Table 8: The number of quadrats surveyed each year before and after improvements Year Before After Timing not Total recorded Total From this impressive number of pollinating insects, 16% were recorded as unidentified and the remaining 84% were categorised into the 8 pollinator groups. The most frequently recorded pollinator type was other flies (39%) followed by unidentified insects (16%) and bumblebees (11%). Hoverflies, honeybees, butterflies, moths and solitary bees made up 23% of the pollinators counted in the quadrats. Groups who surveyed before and after changes were made As mentioned in Part A., 42 groups made changes and conducted the survey both before and after the changes. For those sites where changes have been made, there is a significant difference in the average number of pollinators counted in 2 minute searches carried out before and after positive pollinator improvements have been made (t=2.35, df 40, P<0.05) (Figure 22). When comparing the total area of feeding, nesting and sheltering habitats (excluding bare walls and fences, concrete/tarmac and short grass) and the total pollinators seen there is a significant correlation between the area of habitat and the number of pollinators (Pearson's product-moment correlation, t = , df = 1986, p<0.05). Figure 22: The average number of pollinators per survey carried out before and after positive changes for pollinators were made at each of the 41 locations (+/- Standard error). Species Quest Species quest is an additional part of the survey which requires participants to record whether or not they recorded at least one individual from a list of 12 easily identified pollinators. Across the two survey years, Species Quest species were encountered on 297 surveys (37% up from 33% on 2016 only data). From these 297 encounters there were 597 individual encounters from the species quest list. Of the 37% recording species, most surveys (54%) recorded only one species and the maximum number of species quest species recorded was 11 out of 12 by Wells Park School, Essex. The most frequently recorded species was the Honeybee (seen on 135 surveys, 81 locations) followed by the 19

23 Marmalade Hoverfly (80 surveys) closely followed by the Common Carder Bee (79 surveys) (Figure 23). The earliest species recorded was a Honeybee on the 9 th March 2017 followed by a Red Admiral, a Marmalade Hoverfly and a Honey Bee all in one survey on the 27 th March The most northerly Common Carder Bee was seen at Canisbay Primary School near John o Groats, Caithness. Similarly, there is a record of the Ashy Mining Bee from the Glasgow region- this record will need to be verified as at present this species is confirmed as far north as Penrith although there are confirmed records north of Oban and out on to the Isle of Mull. Figure 23: The number of survey sites where each species quest species was recorded. 20

24 Recorders Increasing numbers of Citizen Scientists are studying pollinators: Almost 300 groups took part in the first two years; 63% of these were schools registered in the Polli:Nation project and the rest were public participants. Over 800 sets of survey results were received Awareness about pollinators is spreading: Polli:Nation provided 38% of groups (107) with their first experience of identifying insects and the proportion of primary schools with some previous experience of insect identification has risen by 5% Pollinator protectors are needed in new areas: Pollinators need your help in East Anglia, the South West, the North of England, Cumbria, the Highlands and Islands of Scotland Habitats Schools and gardens have immense natural capital: Over 11,000m 2 of wildflowers were recorded as part of the Polli:Nation Survey. Over 95% of quadrats had some flowers in them: daisies, dandelions and clovers remain the main plant resources found Pollinator friendly improvements are starting to be made: Over 150 positive habitat changes for pollinators were made to 55 school and community locations around the UK More habitat creation is needed: Short grass covered the largest area (>26,000 m2) and there is huge potential to support nesting and feeding while keeping precious playing fields Pollinators Small habitat improvements can have a big impact on pollinators: Across all sites, a 58% increase in the average pollinator number per quadrat was found after habitat changes and there was a significant positive correlation between the area of (pollinator friendly) habitat and the number of pollinators (t = , df = 1986, p<0.05) Certain habitats and plants attract more pollinators: Verbena plants attracted an average of 86 pollinators per 2 minute quadrat search, damp places attracted an average of 8.2 pollinators Flies and Honeybees are topping the charts Flies are essential pollinators and they were the most common type of pollinator found (39%). The Honeybee continues to be the most frequently recorded Quest Species seen in 135 surveys at 81 locations 21

25 Report Reference Cruickshanks, K., Lakeman Fraser, P., Bulman, C., Burgess, S., Jackson, M., Townsend, S., Evans, L., Fradera, R., Slawson, D. (2018) Polli:Nation: Survey Results p22. Available at: Acknowledgements We would like to thank all of the Polli:Nation programme and survey participants. The original team who developed the survey included: Vanessa Barber 1, Catherine Bertrand 2, Caroline Bulman 2, Claire Carvell 3, Helene Colman 1, Richard Comont 4, Katie Cruickshanks 2, Laurence Evans 1, Roger Fradera 1, Dominic Greves 1, Mary Jackson 5, Vicky Kindemba previously 6, Poppy Lakeman Fraser 1, Simon Norman 7, Ruth Staples-Rolfe 5, David Slawson 1, Sue Townsend 7. Illustrations by: Richard Burkmar 7, Chris Shields, Steven Falk, Simon Norman 7, Dreamstime. Editing by: Poppy Lakeman Fraser 1, Simon Norman 7. 1 Imperial College London. 2 Butterfly Conservation. 3 Centre for Ecology and Hydrology 4 Bumblebee Conservation Trust. 5 Learning through Landscapes. 6 Buglife. 7 Field Studies Council. 22