Estimating colony sizes

[Neils]

Found this while looking to answer a question elsewhere and thought it might be useful. I dare say it is a very rough estimate but what the heck:



The number of workers found in a honey bee hive can be estimated using the following information:

About one-third of the worker bee in a hive forage every day. Based on average number of flights per day by a single bee and the amount of time spent foraging, the following formula can be used to calculate the number of bees in a hive:

N = 3 x (f/0.0138)
N = number of bees in the hive
f = number of bees seen leaving the nest in one minute
If Joe observes 35 bees leaving a hive in one minute, how many bees are inside? Round off the answer to the nearest whole number.

Answer: 35/0.0138 = 2,536 bees foraging per day. This is about one-third of the hive, so 2,536 x 3 = 7,608 bees in the hive.

Note: The value 0.0138 is based on average amount of time spent foraging for an average honey bee colony on an average day. This value will actually change considerably with amount of food available, weather conditions, etc.


The assumption that there's only 1/3 of the colony foraging seems low to me and it seems a bit odd that they disclaim the 0.0138 assumption but ignore completely the assertion that a third of the bees in a colony are foragers as a constant.

[edit] Actually I think it depends on how you interpret that third. A significant quantity more bees might be, technically, foragers but are they out on flowers foraging or sat around inside the hive waiting for other returning foragers to tell them where to go?

[gavin]

Don't know when I last saw an average day. Is the average day a Bristolian or S English day, even an American one?

The foraging activity varies hugely through your average day so I suggest that about 50 random samples of foraging activity are made for each hive to improve the estimate.

Off now to squelch about in the UK's biggest outdoor potato event, in a very muddy field not 20 min drive from here ...

[Adam]

Not convinced the estimate works. When a flow is on you see a lot more bees flying than otherwise - like today when a few drift in and out of the hive. Same number of bees in the colony though!

[Neils]

I think the clue is in the words "rough estimate."

Maybe they explained it better than i did and perhaps the 0.0138 variable is "tuned" for Arizona and you'd need to do a whole bunch of different research to come up with that.

While I was checking that out and looking into the assertion that 1/3 of the hive only are active foragers I did come across an implementation of the travelling salesman problem using a bee colony specifically as Its example.

So that would potentially allow you to work out in more accuracy the proportion of your colony that are giraffes (phone auto correct that I can't being myself to change), but you need to know the colony size to work that out :D

Someone was suggesting weighing a bee, tipping every single bee out then weighing the hive and frames. But riddle me this batman, how many of those bees are empty, full of unstored nectar, carrying pollen, and how many drones do you have compared to workers? So I don't think that's a desperately accurate measure either.

[Rosie]

I think the clue is in the words "rough estimate."

I would say it's far worse than a "rough estimate" - it's worse than useless. Dave Cushman has a page somewhere on his site giving guidance on how to estimate the number of bees on a frame. If you were to follow his guidelines on a day when they weren't flying you might have a clue.

Another alternative is to count the brood and double it for the number of adults. That only works well when the colony is neither growing nor shrinking but an allowance can be made for that by consulting colony records.

Rosie

[Neils]

Give some context please as to why it's worse than a rough estimate.

Just for fun, while I was pottering around looking into this, I found a program dealing with an apparently well known maths conundrum called the "travelling salesman problem" that attempted to deal with it in the context of a beehive. Is the underlying maths any good in terms of real world bee colonies? No idea.

You should be able to compile it using visual studio express, if there's any interest in it but people don't want to go to the trouble of downloading it and compiling it themselves I'll sort it out and put the executable up.

In the meantime you can find the article and source code from Microsoft: http://msdn.microsoft.com/en-gb/magazine/gg983491.aspx

[Rosie]

Give some context please as to why it's worse than a rough estimate.

I think it's worse than a rough estimate because the number of bees flying at any one time bears no relationship to the number of bees in a hive. Bees will pour out of quite a small hive at certain times of day when their particular plant is yielding. Another enormous colony might have nothing flying then because they are waiting for their own favourite source to start yielding. Then there's the weather, variations in time of day when individual colonies start, age distribution inside the colony - the list is endless.

Rosie