Why do birds sing in the morning?

Ruppell's Robin-chat: an impressive mimic. Lake Duluti

I enjoyed a walk around Lake Duluti yesterday morning and came across a couple of wonderfully singing Ruppell's Robin-chats. These are great birds, with an amazingly varied song hat's gull of mimicry (of you want to hear one, listen here!). For me, one of the best things about camping in the bush is being able to lie in bed and listen to the birds waking up while it's still too dim to see them properly. The dawn chorus is a worldwide phenomenon and I'm often asked about bird song, so I thought it would be worth exploring some of the theories behind bird song, and - particularly - why birds sing in the morning. It's something that's interested me since I was introduced to the question by a friend of mine who did a PhD on the subject some years ago, and I know he reads the blog so I'm hoping he'll make sure I get the answers right!


There are two parts to the question, of course: why do birds sing? And why do they sing in the morning more than at other times of day? The first I think we've had a pretty good idea about for a long time, and there are two main reasons: to attract mates, and to claim their territory. In general the two actions aren't mutually exclusive - as a male bird you can sing to both let the other males around know that you're still in your territory, and you can at the same time let females know that you're around and looking for a mate. For some species by listening to the song you can tell what the bird is actually trying to say  - Nightingales are well known in Europe for their beautiful song, but it's not uncommon to hear them singing down here during the non-breeding season too (they seem to be more territorial than many migrant species). The difference is that here, I've almost never hear the typical long whistle notes that are so frequent in the breeding areas during the early breeding season (listen to them here). They also use even more of these whistles during the middle of the night before they find a mate, and it seems that these whistles are particularly for attracting females, which obviously isn't relevant when not breeding. As females might be flying over at night, singing in the middle of the night and whistling a lot, might be a very good way to attract a female to your territory.

Yellow-vented Bulbul - nesting all year round in my garden (March 2012),
both sexes sing far too early in the morning for comfort...

For some species song is probably more important for attracting females than in other species - particularly, I would suggest, for species where mimicry is very important. Like my Ruppell's Robin-chat, many birds are accomplished mimics, and observations suggest that the more varied an individual male's repertoire, the more attractive the male. The most impressive example of this that I'm aware of is the Marsh Warbler, a species that we mainly see on passage here in Tanzania from time to time - though it is another than sometimes sings during the non-breeding season. In what's become one of the classic studies of song mimicry, 30 individual males were recorded during the breeding season and the songs then analysed and mimicry identified. The total diversity of song mimicked by these 30 males was 212 species, 113 of which were African species! On average, each male knew the songs of 76 different species, mostly African - one male included identifiable fragments of 84 species - an astonishing variety of songs and calls! Why do they do this? Well, it seems most likely that females prefer the males that have the most varied songs - males that can learn lots of different songs have clearly got some good mental abilities, whilst - the hypothesis goes - must be tied to their ability to find food, avoid predators or otherwise help around the nest. So there's not only a chance to let a female know that you're around like the nightingales do, but there's a real competition going on to be the best singer of all.

As I said before, there's also plenty of evidence that it's not only about attracting a female: even after finding a female and her sitting on the nest many species keep singing for a while - this is clearly about territory defence. And the first example of how effective this is comes from a nice experiment where birds were removed from their territories and the time it took other birds to come and occupy the patches was recorded: it's pretty quick, but if you play a tape recording of the song of the resident male even though he's been removed, it takes a lot longer. Clearly birds looking for a territory recognise individual songs and as long as the territory holder is still around and still singing they won't bother trying to take over - an efficient system of defence if ever there was one.

Usambiro Barbet, a typical African species with duets. Serengeti, Dec 2011.

Now, that's the general answer to why birds sing. But here in Africa there's an additional complication we should consider: here, many female birds sing too. This isn't unknown in the north, but it's fairly unusual. And we've already seen that a whole load of African birds engage in complex duets too, with both male and female taking part. Why is it that many more females take part in song here in Africa and the rest of the tropics than they do elsewhere? There's no good answer to this one yet, but a couple of interesting theories are being developed that are worth mentioning. Firstly, as we've already talked about, birds in the tropics tend to be rather long lived - they do things slowly, and they have a rather tougher time raising young in many ways than in more temperate regions. So here they're often also with their partner longer, and often holding territories year-round - lots more reason for females to be involved in territory defence as well as males, so that's one possible reason. The other theory is that here in Africa the seasons might be harder to define (at least in the forests), with breeding possible much of the year (certainly the bulbuls in my garden think it's fine to nest all year around). But birds don't tend to keep their ovaries or other bits of breeding equipment large and ready to breed year round - it would be a waste energy to carry around fully developed genitalia if you're not going to use them, and flight is tricky. So in a less seasonal environment, how do both male and female make sure they're in breeding condition at the same time? Well, perhaps, by singing to each other - there's certainly some evidence that song can help bring on breeding condition in some birds. Interesting thoughts though, and (yet) another area of tropical ornithology where we're well behind on theory and experiment.

So. Having tried to answer why birds sing, let's tackle why we think they sing most in the morning. There are a number of (not mutually exclusive) theories here too. They all start with the assumption that singing is good (for the reasons we've covered above) - but it's costly in terms of time and energy (another reason why females might like males that sing more: they're showing honestly how strong they area). So the challenge birds face is in deciding when and how much to sing. One of the oldest ideas is that they sing in the morning because it's still too dark to be out and about finding food, so you might as well sing. That's a pretty solid idea - but it doesn't really explain why they sing more in the morning than in the evening, when the light also fades - or even in the middle of the night. Another theory is that the conditions early in the morning - often cool and with lower humidity than later in the day - might be particularly good for letting the sounds of the song carry further, though recent experiments suggest that actually the middle of the day might be the best time to sing if acoustic conditions are what's important so I don't think that's a winning idea. The third and most interesting theory is that birds sing most in the morning because that's when, most days, they've got spare energy to use up [Rob, surely there's a short snappy label for this theory? The McNamara/Mace/Houston hypothesis seems a bit of a mouthful? The stochastic environment hypothesis?].

Many species, like this Cisticola are best identified by song.
More drinks for the first correct ID! Nr Arusha, March 2012.

This is a much more complex argument that's absolutely fascinating in it's detail and worth looking into in some detail. It starts with some basic assumptions - like the others it says singing is good, but other activities are also important - feeding, sleeping, etc. And most important of all, it requires that the world is variable: the conditions you find today such as how easy it is to find food, what the temperature is, how long you have to sit out a rain storm, are all variable day to day. When you go to bed at night, you don't know how cold the night will be, and when you wake in the morning, you don't know how long you'll have to spend finding the food you need. Given these initial conditions, this theory says that birds are essentially playing a game of survival. Each day they need to find enough food to survive the following night (when they can't feed because it's dark) and they'll fit singing around these activities when they have energy to spare. Now, birds live very close to the edge much of the time and each night they loose a significant amount of weight just keeping warm: if they don't start the night with enough extra fat to burn the could easily die overnight. So because they don't know exactly how cold each night might be, the only way to be sure of surviving is to have enough fat to survive the coldest night they can imagine might realistically happen. Most times, of course, it turns out not to be so cold and when they wake in the morning, they've got some spare energy to use before they need to start feeding again: the perfect time to sing. During the day too they don't know how hard it will be to find enough food to have the required amount of fat by the evening, so once they start feeding they're wise to feed like mad for a bit, just in case it's tougher to find food than they thought, but then if they keep finding food all day, by the evening they might have actually got fatter than they really need to be (it's unwise to have too much extra fat, as it makes you less manoeuvrable and more vulnerable to predation), so again you might want to sing a bit in the evening too - which is exactly what we see. These predictions and assumptions were tested in a nice series of experiments by Rob Thomas (for example, in one experiment he fed individual robins the same amount of food during the day, but some were persuaded that there was a regular and constant supply of food throughout the day, others thought it much more variable getting lots of food at one point, but much less later in the day, for example, and then listening to how long they sang that evening or the following morning: more variable gives more song), with the general conclusion that the theory is fairly robust. Which also explains why the dawn chorus here in Tanzania, where the weather from day to day is fairly predictable, is much shorter and less impressive than, say, in a British woodland where it's quite possible one day will be warm and sunny and the next day cold and snowy! Brief as it is, the dawn chorus here is still impressive in many parts of Tanzania and well worth listening to when you're out in the bush - the best way to start the day I think!

Main References:
Slater, P., & Mann, N. (2004). Why do the females of many bird species sing in the tropics? Journal of Avian Biology, 35 (4), 289-294 DOI: 10.1111/j.0908-8857.2004.03392.x

Thomas, R., & Cuthill, I. (2002). Body mass regulation and the daily singing routines of European robins Animal Behaviour, 63 (2), 285-295 DOI: 10.1006/anbe.2001.1926