Listening to the bees – swarm prediction

Wouldn’t it be useful to know your colony was preparing to swarm?

With sufficient notice you could take preventative action.  You could conduct an artificial swarm or remove the queen to a nuc. With the major population of the hive now queenless the colony cannot swarm, though you will need to prevent casts by ‘discouraging’ too many virgin queens from emerging.

What is sufficient notice?

Certainly more than minutes or hours. You need to have time to collect the equipment you’ll need, you might need to wait until the rain clears, or you return from a long weekend in Rome (remember those?).

Inevitably, 2300 years ago Aristotle had both identified the problem (Are they about to swarm?) and found a way of answering the question.

When the flight of a swarm is imminent, a monotonous and quite peculiar sound made by all the bees is heard for several days 1.

Several days should be enough surely? The fact that you don’t see beekeepers hunched down by their hives listening intently suggests it’s not an entirely practical solution.

Depending where you live, the swarm season can extend from mid-April until mid-July. That’s a long time to be sat next to a hive with your ear glued to the brood box. 

Some sort of automated system is needed.

Woods’ Apidictor

In the 1960’s Edward Woods developed 2, patented and sold an electronic frequency analyser to “listen” to beehives. It incorporated analogue bandpass filters to screen out the background noise of the hive, focusing on the “monotonous and quite peculiar sound” (to quote Aristotle again) characteristic of a colony making swarm preparations.

Woods Apidictor

Woods sold about 300 Apidictors during the early 60’s. Analysing an individual colony took just minutes. In a 1965 article in Bee Craft Eddie Woods claimed that 60 colonies could be screened in 15 minutes.

In the right hands the Woods Apidictor was also accurate, predicting not only whether a colony was going to swarm, but also the particular stage it was at e.g. sealed queen cells present (presumably held back by bad weather), queenless colonies, superseding colonies, or even drone-laying queens 3.

One of the most impressive [demonstrations of accuracy] occurred when one purchaser complained that the instrument was “over-enthusiastic”. Our representative visited him and checked and predicted that seventeen out of twenty colonies in one apiary contained queen cells, even specifying the stage of development of the cells, a test which required only a few minutes. The manual examination, requiring hours, showed that indeed, the situation was as predicted. The three colonies which the instrument showed no swarm preparations were found to be clear. 4

Despite the fact that Woods only ever sold a limited number of Apidictors, the subsequent 60 years have seen continued interest in – literally – listening to the bees

The internet is littered with commercial and DIY solutions, all in various stages of development. I don’t intend to review them … Google is your friend. Make yourself a cuppa … there are over 2 million hits.

Importantly, the usefulness of this automation depends upon whether sounds in the hive are truly predictive of swarming, and our ability to identify the relevant sound that is predictive.

A monotonous and quite peculiar sound

When using sound as an indicator of swarming activity, false-positives will have us scurrying around performing artificial swarms when unnecessary, and false-negatives will mean the colony swarms.

Not the end of the world, but an irritation certainly.

So, what do we (or, more likely, our Raspberry Pi or Arduino-powered detectors) listen for? What is Aristotle’s “monotonous and quite peculiar sound”?

A recent paper by Martin Bencsik and colleagues addresses this and claims accuracy of greater than 90%, with successful swarming prediction up to 30 days prior to the event 5.

A beehive is a noisy environment. Some of the noises are constant, others vary during the day, or over longer periods. As the colony builds in strength during the early season the noise levels must increase. Similarly, the sound levels in the colony will be different at night than during a bright, sunny day with thousands of foragers working hard.

Late evening in the apiary

Late evening in the apiary

Many beekeepers will be aware of the sound a colony makes at night during a strong nectar flow. The entire hive hums as the workers drive off excess water prior to capping the cells. If you’ve not visited an apiary late on a calm, midsummer evening you should try; the smell of the honeysuckle in the hedgerow mingled with the scent of honey from the hive is a heady combination.

Toots and quacks and buzzes

In addition to the cumulative noises of tens of thousands of bees simply working in the hive, there are the noises individual bees make at certain times. 

Scout bees produce piping noises when doing their buzz runs as they encourage a swarm to leave the hive. Queens, in particular virgin queens, produce a variety of noises including duck-like quacks and piping before, during or after emergence. I’ll return to these in a future post as this paper also includes a lot of information on the timing and relevance of queen piping.

Marked queen surrounded by a retinue of workers.

If it quacks like a duck …

And if all that wasn’t sufficiently complicated there are additional acoustic signals such as the intermittent, but extensively repeated,’whooping‘ noise (and others), which may be similar to the ‘stop signals’ workers direct at dancing foragers advertising less than favourable locations.

All of which means you cannot just stick a microphone at the hive entrance and instantaneously determine whether they’re about to swarm.

In particular, the variable nature of sound over time needs to be taken into account. The steadily increasing background noise of a hive building up through the early spring and summer is probably not significant.

In contrast, a spike in the signal averaged over several hours or days is probably important, but there may also be characteristic sounds that – if present – indicate swarm preparation.

In the Ramsey paper the authors have used two machine learning techniques to analyse the sound from accelerometers (something that measures vibration) inside the hive.

Training data

Woods had identified a 250 Hz signal that he considered characteristic of a colony making swarm preparations. Modulation of this signal, which Woods termed “the warble”, in three test colonies was associated with swarming.

Ramsey and colleagues analysed the full spectrum using highly accurate accelerometers in about 25 hives, sampling continuously at 3 minute intervals, over a two year period. About 50% of these colonies swarmed during this time, generating 11 prime swarms and 19 casts.

This dataset was then analysed to find acoustic features characteristic of hives that did or did not swarm. Essentially the authors trained the algorithms to detect particular acoustic patterns that were – through empirical observation – associated with swarming (or not).

To do this they used two separate approaches:

The instantaneous alarm procedure.

In the first (the instantaneous alarm procedure) they took a one hour reading from the hive and then compared it to the trained data. By computationally analysing discriminant functions (i.e. acoustic features characteristic of swarming or non-swarming hives) they could determine whether the test colony fell within the “swarming” or “non-swarming” category.

In the diagram above they show the application of two discriminant functions, but the actual study used three. 

The second method used a much more complicated sounding three dimensional Fourier transform (conveniently abbreviated to 3DFT). In contrast to the first approach this involved analysis of the acoustic spectra collected over a ten day period.

3DFT sounds more complicated because it is more complicated. A Fourier transform converts a complex mix of signals into its individual components – for example, determining the individual volumes and frequencies (notes) in a musical chord. The diagram for this is a more colourful version of the one shown above, but is unlikely to help understand the process. If you insist you can view the original.

Application

Both methods were similar in that they used sound profiles collected from colonies known to have swarmed (or not) to define patterns characteristic of swarm preparation.

Interestingly, neither appear to show significant differences in the region of the spectrum Woods’ ‘warble’ occupied. 

Having trained the software they went on to analyse colonies in an unknown state in an attempt to predict swarming.

Using the instantaneous analysis 6 15 of 18 colonies that swarmed were correctly predicted, with no false positives in the colonies that did not swarm. Of those that swarmed, the prediction could be made an average of 22 days in advance of the first swarm leaving the colony. 

That sounds like a pretty convincing 90% prediction rate. However, looking at the primary data – all 33 Mb of supplementary figures – it is clear that many of the swarmed colonies produced “swarm-like” signals well after swarming, without repeated swarming. As the authors state “false positives are still triggered occasionally on an hourly basis, and this becomes exacerbated when the time duration of the season under scrutiny is extended to the rest of the summer”

So, it works OK for the first swarm of the year, but after that gives a lot of false positives.

It’s not clear from the figures what the range (or standard deviation) on the “22 day average warning” is. If it’s a range of 20-24 days that could be really useful, if it’s 3-45 days, less so.

Using the 3DFT methodology the authors could predict swarming in ~80% of colonies an average of 10±2 days before the swarm issued. Although this is a lower prediction rate, the clearly tighter time window might be more useful for practical beekeeping.

Again, the 3DFT approach produced signals that indicated swarming was imminent throughout the remainder of the season, often during periods of intense foraging. To exclude these the authors used averaged the night time values (midnight to 5am), rather than day-long assessments.

Take a deep breath

Overall, taking account of the false prediction rate and the false-positive triggering rate (the former being an overall incorrect prediction as to whether a colony will swarm, the second being ‘noise’ in the analysis when the threshold is reached), the authors favour the simpler “instantaneous” measurement method.

However, don’t be misled, this is not like Eddie Woods rocking up with his Apidictor, putting the stethoscope-like microphone on the side of the brood box and saying “take a deep breath”

These colonies are being constantly monitored, with accelerometers embedded in the frames, and the associated wiring dangling out of the brood box.

I have run colonies with embedded hive temperature and humidity monitors 7. The cabling is usually run under the crownboard and down between the frames. Along with 20,000 bees, it’s another thing that gets in the way during weekly inspections. In this paper Ramsey shows that the accelerometers can be fitted to any frame in the hive and still provide valid predictions. This offers the opportunity to perhaps use one of the ‘edge’ frames which would be more convenient than temperature monitors which have to be embedded in the centre of the brood nest.

I’m sure Woods’ Apidictor was not inexpensive in its day 8. This current implementation of in-hive technology, despite the advances in microelectronics and computing, uses accurate and sensitive accelerometers which are also not inexpensive 9. The apiary would need a power supply, computers and a way of transmitting a signal to the beekeeper (who is currently quaffing Barolo in a fancy trattoria anyway).

All of this is achievable.

But is it worth it?

And is it really needed?

Analogue beekeeping

The digital revolution and, most recently, the internet of things (IoT) has made monitoring “stuff” (like beehives, the house temperature, your fridge or coffee machine) inexpensive and relatively straightforward.

With smartphone apps you can be “in when you’re out” and get a warning that your colony is going to swarm … just as you sit down to lunch at Pierluigis.

The internet is littered with commercial and DIY hive monitoring equipment. Most of it is advertised, or at least promoted, as making beekeeping “easier”.

There’s the implication, stated or otherwise, that this type of automation reduces the need to conduct those pesky hive inspections.

But is that desirable? 

What about all the other things you check when inspecting a colony?

Nectar collection … how heavy are the supers? Yes, you can monitor this electronically as well with hive scales. But what about …

  • colony build up – how much space does the queen have to lay?
  • sufficient stores – are they going to starve if it rains for a week?
  • laying pattern of the queen – is she failing, is she a drone layer?
  • signs of disease
  • robbing etc.

I’m enthusiastic about technology but I’m not sure I’m enthusiastic about this technology. 

Beekeeping is in many ways, already ‘easy’. It’s also an intensely practical discipline.

A thorough hive inspection tells you a whole lot more about the colony than its likelihood of swarming.

I’d actually argue that the easiest thing to determine qualitatively is whether a colony is thinking of swarming. All of those other things listed above – and lots that aren’t – are both important and only acquired by standing hunched over the hive.

If your hive monitors discouraged you from checking colonies so often how would be ever learn, or know, these other important things?

Finally, in closing, I reckon I could open a colony in mid-April and – based upon its strength and knowing a little bit about the local nectar flows – predict with 90% confidence whether it will swarm later in the season 😉

What do you think? Are you in favour of automating some aspects of beekeeping? 


 

Footnotes

  1. Aristotle. The Works of Aristotle the Famous Philosopher Containing his Complete Masterpiece and Family Physician; his Experienced Midwife, his Book of Problems and his Remarks on Physiognomy. (J. A. Publishing, 2018).
  2. Woods, E. Electronic prediction of swarming in bees. Nature 184, 842–844 (1959).
  3. Frankly that’s better than some beekeepers can manage when they’re inspecting the colony!
  4. E.F. Woods, Bee Craft, August 1965.
  5. Ramsey et al., (2020) The prediction of swarming in honeybee colonies using vibrational spectra. Scientific Reports 10: 9789. The paper is freely available under Open Access.
  6. Which confusingly isn’t instantaneous … it really means data collected for only one hour, ignoring everything that has gone before. The monitors are permanently fitted in the hive and record a rolling one hour window.
  7. And am currently building a homemade solution which I’ll discuss in a smorgasbord of linux, perl, RRD and DS18B20’s at some point in the future. This has nothing to do with swarming, but everything to do with rational and optimal Varroa control.
  8. If anyone knows how much they were please contact me, or add it to the comments below.
  9. I couldn’t find a purchase price for the Brüel and Kjær Type 4508 piezoelectric accelerometers used, but did find somewhere that would rent them for £4 a day. You’d need them for a few months per season. You’d have to sell a lot of honey to recoup that outlay.

20 thoughts on “Listening to the bees – swarm prediction

  1. Paul H

    I designed scientific instrumentation for decades and I agree, it is unnecessary in beekeeping. Particularly with digital instrumentation, you spend a vast amount of time fighting software, which is invariably built on a shaky pyramid of operating systems that no longer work with your kit after 2 years.

    The one stand-out function I’ve seen for added electronics is theft detection for remote apiaries, but people who’ve used it and lost hives say they did not manage to get there in time when it went off, because in 15 minutes a thief can load hives on a truck and be off.

    Some people are experimenting with frequency analysis apps on mobile phones, but very early days yet and no convincing results.

    Reply
    1. David Post author

      Hi Paul

      I know exactly what you mean about the “shaky pyramid” from my own forays in coding. I think we’re a very long way from ‘plug and play’ solutions and am not sure they’re needed or desirable anyway.

      Interesting comment about theft detection/prevention. It’s an increasing problem. The embedded GPS tracker would certainly work (or at least alert) for the theft of a hive, but not for the ‘sneak thief’ that pops the lid, steals the queen (perhaps plus a couple of frames of brood) and then puts the box back together again. Security through obscurity, whilst not particularly dependable, must be a benefit here.

      Cheers
      David

      Reply
  2. Nigel Hurst

    Hi David, very interesting article, some years ago I took out a copy of the book or article on the Apidictor from the Moir library I can’t say I really understood all that I read, I have heard of people more recently with their up to date equipment doing various monitoring tasks to try and predict swarming etc.
    At the end of the day though one is probably on a hiding to nothing trying to reduce such an amazing thing as beekeeping to the monitoring of various signals by computers, long may it remain so!

    Reply
    1. David Post author

      > long may it remain so!

      Hear hear!

      I love the practical aspects of beekeeping and the tactile experience of working with the bees. I’d be a lot less interested in just having my phone send me messages … SWARM ALERT, Time to harvest spring honey or The beekeeping robot has added Apivar strips today (though it might be good to get a reminder when the strips are due to be removed!).

      One of the great things about producing good quality local honey, for sale or as a gift, is that it takes effort and commitment. It’s a fusion of hard graft, art and science.

      As you say … long may it remain so.

      Cheers
      David

      Reply
  3. Martyn

    Great article David, now I must resist the temptation to take out the Raspi and go down a project rabbit hole. You reminded me of the German beekeeper video where they were familiar enough with swarm behaviour to catch them on exiting the hive!

    https://youtu.be/hn5OxMXCSz8?t=1737

    Reply
    1. David Post author

      Hi Martyn

      Those iwf.de videos are excellent viewing. I guess with skep beekeeping you cannot (easily) conduct any sort of artificial swarm so are forced to let them swarm and capture the swarm as it emerges. That really is high maintenance beekeeping.

      Temperature monitoring with the RPi is a doddle. I’ve got one hooked up to monitors in a shed which – in due course – I’m going to extend for some very targeted hive monitoring. I’m really interested in the timing of the broodless period in “midwinter” (which usually isn’t in midwinter of course) as it is an ideal time to treat for Varroa. The winter is the one time I can see benefit in semi-automated hive monitoring – it’s a time when you don’t want to be rummaging around in the brood box.

      Cheers
      David

      Reply
  4. John Eaden

    Thank you for a fascinating review of both the analogue Woods Apidictor and the recent digital technology for listening to hive sounds.
    Both systems are evidence of a techie mindset in some folk which looks to electronics for solutions to problems that arise out of complex biological systems such as honeybee colonies.
    It’s the very complexity that means that the technology can only provide partial information which leaves the beekeeper out of pocket but not really much better off than if he/she used their own eyes, ears and brain to evaluate their colonies.
    My own scientific background led me initially to try to approach beekeeping as just a set of logical puzzles to solve. Eight years later I realise that the numerous hive inspections I have completed have been building a body of experiential knowledge that allows quick assessment of a colony.
    I suppose one could take technology to the limit and build a robot coupled to an AI system to carry out all of the other hive observations to deal with the various issues you mention at the end of your post- but what would be the point? I imagine that the reason most hobbyists keep bees is for the sheer pleasure that bees provide us in managing colonies.

    Reply
    1. David Post author

      Hello John

      I agree – the management of colonies, the ‘working with the bees’ rather than simply using them as a means of production, is what attracts me to beekeeping. I can’t think of anything worse than a robot doing all the work for me (though my back this morning tells me that some help with the lifting would be welcome!).

      I think beekeeping does present a series of logical puzzles to solve. The learning experience, from trying to find the queen to realising you don’t ever need to see her to know things are OK, to being able to determine the development of the colony from the brood pattern etc. is what makes it so constantly challenging and interesting. I don’t think I ever visit the apiary without learning something new … even if it’s that I need some sort of of automated reminder system to make sure I take the smoker with me 😉

      Cheers
      David

      Reply
  5. John Bolger

    I always look forward to the Friday blog.
    But this was the most tense drama of them all. By the end of the blog will he say there IS an inexpensive device that will simply make an accurate prediction?
    Or not.
    I’m tired of losing swarms, due largely to my inability to find the Q and then sometimes not spotting the QCs.
    However I’m going to carry on. There’s more to it than honey.

    Reply
    1. David Post author

      Hello John

      There is more to it than honey, but you will learn to reduce swarms lost.

      You don’t need to see the queen. If there are eggs present there was a queen there within the last 3 days. It’s likely she’s still there. If there are a reduced number of eggs present and there’s still nectar coming in, it’s possible she’s slowing down laying in preparation for swarming. In that case, now is the time to look for queen cells. Since there are eggs present you can safely shake all the bees off the frame and look at it without the confusing, wriggling, mass of bees hiding the detail. Even if there are queen cells (which might be damaged by shaking), there are fresh eggs for them to work with if needed. Keep good notes and inspect regularly.

      The more hives you open the more experience you will gain. I probably look in 500+ colonies a year and, after a few years, have a reasonable idea of what’s (usually!) going on. Even then, I’m sometimes completely flummoxed 🙁

      If you get a chance, tag along with a bee inspector. They inspect thousands of colonies and are often willing to talk you through an inspection. It’s an enlightening experience.

      Cheers
      David

      Reply
  6. Kevin Barron

    Hi David,
    Interesting observations.
    I only have my bees a week now and I am by no means any type of an expert.
    I do however like to see what modern technology has to offer.
    To that end I actually purchased an item that apparently kills varroa by heating the hive to 40°c. Yet to try this.
    However the item comes with a “hive heart that they say can predict swarming up to 21 days in advance.
    It does lots of other things also.
    I only installed it yesterday. I do find it interesting.
    However being a part time farmer I really enjoy getting into the hive, I have been given a swarming hive and it is proving interesting hence the more than once a week inspection. Anyway as you say nothing beats getting your hands dirty so to speak but I do think technology can help us.

    Reply
    1. David Post author

      Hello Kevin

      Welcome to The Apiarist … I’ve seen one or two ‘solutions’ to the Varroa problem that involve heat. Whether those solutions actually solve the problem is less clear. I’d advise you not to rely on either the heater or the “hive heart” for your swarming prediction. Learn how to read the hives yourself and then see if the results the technology produces is dependable.

      And if it is … please post a follow-up comment 🙂

      Cheers
      David

      Reply
  7. Daire Einston

    Thanks David, excellent post. I’m all in favour of learning as much from observing the bees at inspection time as possible. Sure that’s part of the fun!

    Reply
    1. David Post author

      Hi Daire

      It certainly is. And it’s a never ending process – firstly because there’s so much to learn and (in my case) because I seem to forget things more easily these days 🙁

      Cheers
      David

      Reply
  8. Alex

    There is an app out there already that works along similar lines by listening to bees and using AI – Bee Health Guru – https://www.kickstarter.com/projects/beehealthguru/bee-health-guru-a-smartphone-app-for-beekeepers . I haven’t used it but maybe someone out there might comment.

    Your question is interesting. Firstly, you can’t stop progress. New tech will become a part of beekeeping whether you like it or not. Secondly, new tech often brings both advantage and disadvantage. In beekeeping tech might allow us to know what colonies have what disease and then response to outbreaks of EFB could be better managed, for example. Conversely it may create lazy beekeepers who blindly do what the tech tells them without really knowing what’s going on – dangerous in the sense that chaos could ensue and sad in missing out on the wonder of fully appreciating what the bees are doing.

    The question for me is how do maximise the advantages and minimise the disadvantages?

    PS if there was a robot to do all the honey extraction I’d be okay with that 😉

    Reply
    1. David Post author

      Hi Alex

      $28k committed to the project … not bad. I’d be really interested to understand the accuracy with which they can predict foulbroods or Varroa infestation. And how they trained the software to identify these particular things.

      I’m fully aware that progress cannot be stopped. I wouldn’t want it to be. Hive materials are evolving and someone, somewhere, will achieve a really ‘good’ poly nuc design. Please? However, I agree with the sentiment you express in your comment. The magic of beekeeping involves working the bees. If I just wanted some honey I could go to the local farmstore and buy a few jars – it would cost me less, I wouldn’t have a bad back and I could have spent the last 48 hours walking, cycling, canoeing, sleeping or whatever.

      But that honey would be so much less satisfying. And giving someone else that jar of honey would mean so much less.

      I also agree … after the first spin of the extractor the novelty wears off really fast 😉

      Cheers
      David

      Reply
      1. Alex

        http://beekeepingtodaypodcast.com/dr-jerry-bromenshenkdr-david-firth-bee-health-guru-uom-master-beekeeper-program-026

        It was an old Beekeeping Today podcast episode. I’m not sure of the accuracy but remember they seemed confident. They fed the AI sound bites of different colonies in different states and used machine learning to get it to recognise different things.

        I remember wondering about the subspecies they may have used to calibrate/teach the AI, and if different subspecies would have different dialects. Would AI trained with American bees be as accurate with British bees?

        Reply
        1. David Post author

          Hi Alex

          “but remember they seemed confident” … a phrase that could be applied to every Kickstarter project, ever 😉

          If there were marked subspecies differences you might expect hybrids to exhibit oddities in swarming activity. Perhaps. I don’t think this is seen, but would need to look into it more. The study referred to in this post used hives in two apiaries in France and one in Nottingham. The strain of bee isn’t mentioned.

          Cheers
          David

          Reply
  9. Alan Deeley

    Dear David,
    I run an Arnia hive set up for the Warwick & Leamington branch which collates weight, temperature and sound data. All of this is sent daily to a server and is then available to view in graphical form. Alerts are sent if various parameters are breached. Reading your article has encouraged me to get in touch to see if they can use the data to predict swarming rather than just react to passed events. Thank you.

    Reply
    1. David Post author

      Hello Alan

      I’ve used an Arnia system as well. It’s not currently connected.

      I suspect the monitors they currently have are unsuitable for prediction, though flight noise and weight loss will work fine to tell you that you’ve just lost a swarm 🙁

      Cheers
      David

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *