2021 Counting Bumblebees

Some of you saw me over the spring and summer walking along the Coastal Path and recording counts of each of the bumblebees seen and which flower they were visiting. You may well ask – Why count bumblebees? - How do you count them? - And what has the counting shown? This note will provide some answers those questions.

Counting wildlife is now quite a familiar activity, with many taking part in the Big Garden Bird Survey or some of the many other counts of different wildlife groups. The aim of all these counts is to keep an eye on the abundance of the different sets of species and whether they are going up or down. Apart from often producing a lot of anguish and hand wringing, the results are valuable to help with information on actions that are needed to halt further declines or pass on information of successes and how they can be repeated.

Bumblebees as a group are easy to count as they are familiar and easy to recognise as they move from flower to flower.Counting the individual species that are seen is rather more difficult.After leaving my insect ecology studies in the late 60s to find a paying job, I was able to return to insects again when I started volunteering with SNH counting bumblebees and butterflies with a group of other interested and enthusiastic people in 2008.Thanks to guidance from SNH staff and others in the group I joined, I learned to identify the species that are most commonly found in our area and also improved my identification of the flowers they were using.The technique used at Loch Leven was to steadily walk one way along a set route on a specified day each week from April, when the queens emerge from their winter sleep, to September when the new queens stock up before they hibernate ready to repeat the cycle the following year.Bumblebees on both sides of the path were counted and, in some years, (2013 – 2015) we also recorded which flowers each individual bumblebee was visiting to seek either nectar or pollen.The same method was applied here in 2015 when I did my first assessment of Dalgety Bay bumblebees on a path through the areas we had started to manage for people and wildlife as part of the newly created DBCWG.My 2021 count was a repeat along the same route and this is shown in Fig 1.To help with counting and to look at any changes in specific areas, the route, as in 2015, was divided into 5 sections.Unlike the Loch Leven counts of one day each week I was able to do my counts on up to 4 days each week.This allowed more flexibility in choosing suitable weather conditions when bumblebee activity would be high and so build up a greater body of data to help with analysis. Each survey took a little over one hour. One difference from the previous survey was the use of a hand held dictating machine to enable faster and more detailed recording in the field, though the data still had to be transcribed onto paper for input into Excel spreadsheets for the analysis that is used in this note to show the information gleaned from the counts.The data presented in the graphs show the maximum number recorded in any of the counts each week.

Fig 1

Which Bumblebees were Counted?

In the list of the bumblebees counted I will give both the common name and the specific scientific name. The latter is important as this universally accepted name allows for them to be clearly identified by anyone else despite any variations in what they might commonly be called in different areas. There were 6 species being counted in 2015, but they have since been joined by a 7th, the Tree Bumblebee (Bombus hypnorum), which has been steadily colonising the UK from first appearing in the south in 2001.

Of these 7, three are fairly straightforward to identify from their colour patterns Red-tailed (Bombus lapidarius); Common Carder (B.pascuorum) and Tree Bumblebee (B. hypnorum), but the other four require particular care to distinguish between them as they can have very similar bands of black, white, and yellow.

These four species are: Early (B. pratorum) distinguished by a pink/buff tail; Garden (B. hortorum) with two yellow bands in the middle; Buff-tailed (B. terrestris); and White-tailed (B.lucorum). Examples of workers of all these species are shown in the accompanying photographs. The last two can be very difficult to tell apart as workers so, although a separate identity was attempted and recorded for each one seen, many researchers often count them together as terrestris/lucorum (Buff-/White-tailed), and this combination will be also used in some of the figures discussed here.

Workers of the different bumblebees are illustrated in the photos in the gallery below.

Altogether the 2021 survey logged 2677 individual bumblebees from 53 walks, along with 59 Honey bees (Apis mellifera) and 95 Solitary bees of a range of species. This compares to 2152 individual bumblebees from 43 walks counted in 2015.

There is a wealth of information in the two surveys, but for now I will just give the main conclusions regarding abundance.

Overall Abundance

To get a good idea of the maximum population size a Peak Count was calculated separately for each species an the combined population by taking an average of the 3 highest count numbers. These “Peak counts” of all the species together were broadly similar for the two years, so it appears that within our managed area, bumblebees are faring quite well. (Fig 2) However, given the increased abundance and diversity of flowers as a result of our further expansion of the wildflower area by the seed sowing at the Heritage Viewpoint, I would have expected higher overall peak numbers in 2021. This year was however a difficult one for bumblebees and other pollinating insects. The see-sawing weather in the early part of the year with early very warm conditions on some days followed by plummeting temperatures; the cold wet May; and long periods with no rain at all during some of the summer months has made it hard both on the bumblebees themselves and the flowers on which they depend for pollen and nectar over all of the year.

Fig 2

Species Differences and Consistency.

Al the bumblebees counted do not follow identical patterns of presence and abundance.The counts show that the biology and ecology of the 7 species counted can be divided into three broad categories.Early, Garden and Tree have a short colony cycle and low peak numbers early in the summer. Three others, Red-tailed, and Buff/White-tailed, have moderate to high peak numbers later in the year and one, Common Carder, also has a high peak number, but even later in the season.These differences between species are consistent across the two surveys both in number and timing, though the later peaking species were slightly earlier to reach their maximum in 2021 (Fig 3 & Table 1

Fig 3

Table 1

Specific sections of the survey

Results from Section 2, which includes the Heritage Viewpoint with its array of different wildflowers which were sown in 2018, shows the benefit of this work with numbers being 2 to 3 times higher in 2021 compared to 2015 (Fig 4). In 2015 it was an area of rough grassland and thistles, but by 2021, while the Cornfield Annuals brightened up the path-sides, there were many different perennial species along the path, on the bank, and the area below providing good sources of pollen and nectar (figs 5 & 6).

Fig 4

Fig 5 Part of Section 2 before sowing of wildflowers

Fig 6 Part of Section 2 in 2021

In contrast, in Section 3, while 2021 counts were higher than in 2015 up to June, they were then lower for the rest of the year, with peak numbers being about 1/3 of those in 2015 (Fig 7). While there may be various reasons for this disappointing change, one that needs to be examined in greater depth is the well-meant but excessive path-side strimming in early June that went well beyond the path edge and cut down flowers that were being used by the bumblebees at that time but also plants that, if left, were going to flower later in the summer.

Fig 7

In the other sections, Sections, 1, 4 & 5 numbers were very similar over the two years. As an example, data for Section 1 are shown in Fig 8.

Fig 8

End of Season Strimming

From the start of our wildflower growing in 2014, we cut back the wildflower areas and removed the strimmed material in October, or sometimes November. This was a decision based on the observation during the Loch Leven surveys of foraging bumblebees that were still active through September. The counts here in both years demonstrate, as seen by looking at Figs 2 4 & 8, that to strim any earlier than into October would remove the food supply from the bumblebees, principally Common Carder, that are still abundant in September, and make it especially hard for any foraging new queen bumblebees at this time. They need to lay down fat reserves to see them over their winter hibernation.

Our strategy for managing the wildflower areas makes sense when viewed with the charts of population changes in our bumblebees but it is very different from advice that can be found elsewhere, often given by well-regarded gardeners, that recommends cutting very much earlier, sometimes between the end of June and the end of July. As can be seen in the graphs of counts, strimming then would cut right across the main foraging time for many of the bumblebees that we have monitored, with serious consequences for the survival of the colonies. The advice for July strimming seems to be based on the old farming practice of cutting the grass for hay in that month in what were, in those far off days, flower rich meadows. This approach may have the benefit of a greater reduction of the soil nutrient levels which would in turn help to reduce the growth of some of the more vigorous competitor plants and, from a gardener’s point of view look tidier, but it would not help these important insects to thrive. There would also be a serious impact on the many other, often unseen, species that rely on the plants for food and shelter.

There is much more still to be gleaned from the data on the biology of these seven species, especially in relation to the flowers being utilised over the colony cycle of each species and how this might have changed as the wildflowers in all the areas have developed. A further and more detailed analysis of the data will follow, looking especially at the selection of flowers being used by different species at different times of the year, but for now we can take heart from the fact shown by these counts that the two areas of Scottish origin wildflowers that we have established, Sections 1 and 2, have the highest counts of bumblebees on the circuit. This shows what can be achieved on even a relatively small area for this one group of insects. The benefits are not limited just to bumblebee populations, increasing the diversity and abundance of wild plants not only in the areas with sown wildflower seed, but also where ivy and nettle clearing has been done by the Donibristle P7 pupils, from their first efforts in 2015 until the Covid lockdown in 2020, all helps to support a much bigger range of less visible plant eating wildlife and through them the birds that can be seen and heard on any walk through our woodlands.

For anyone interested in learning more about these fascinating insects I recommend going to the website of the Bumblebee Conservation Trust https://www.bumblebeeconservation.org . For an excellent and often amusing book Dave Goulson’s “A Sting in the Tale” is a good read that gives an insight into bumblebee biology and why they are so fascinating and important, and if you want to read about the detailed research work that has been done to establish what is known about bumblebees, Goulson’s “Bumblebees - Behaviour, Ecology and Conservation” is a very good place to start.