Termite mounds

Bologonja termite mounds, September 2011

Mostly dead termite mounds visible from the air, Northern Serengeti Sep 2011

One of the most impressive things I saw whilst up in northern Serenget a few weeks ago was the incredible density of termite mounds around the Bologonja River. They were extraordinarily dense, with a mound every few metres. More generally the whole of northern Serengeti seemed very well endowed with termite mounds, both old and new. In fact, flying over I was first puzzled by the obvious bare patches visible across many of the mlains up there, and only once I'd been on the ground for a day or so did I finally convince myself that they really were the remains of old termite mounds. So this set me wondering just what determines the densities of mound-building termites, and when I got back I was able to look things up.

Dead termite mound from the ground. I'm sure that's what it was though...

The most interesting paper I found is this one, which is about the Kruger in South Africa, but I'm sure the same processes are at work in Serengeti. Now, as a bit of background it're worth recapping that (a) not all termites are mound building (in fact, most of them aren't), (b) there's a tremendous number of them out there, having a massive impact on nutrient cycling - three genera (Odontotermes, Macrotermes and Trinervitermes) of mound building termites in Serengeti, and total biomass very similar to that of ungulates or mega-herbivores, (c) they really are fascinating, and deserve more attention on this blog! So, massive densities of termite mounds are quite interesting to understand. Termite mounds are shaped by three main, interacting processes: behaviour of the builders (different species build different mounds); soil properties (the soil used is from the below ground resource); and climate (both how much they have to battle with heat, but also the water table and degree to which they might wash away in heavy rain). So to explain the very high densities in northern Serengeti we both need to be aware that there's obviously a lot of food in this very high rainfall savannah area, but also the soil properties might just be perfect.

Interestingly, in both the South African work and this work in Serengeti termite mounds were found to be at their highest density in general on the tops of hills. The South African study explains this very nicely in terms of soil and water - over very long time-frames, rain washed clay out of the soils on the hilltops, and deposits it lower down the slopes. This mean that when it rains, the soils on the hill tops are fairly free draining - which is important to termites as they have underground chambers that mustn't flood. But it also means that where the clay is washed down to, there's a layer around the hills where the clay content suddenly increases, and when it rains the water flows through the upper soil layers, then hits the clay layer and flows horizontally until it reaches the surface as a seep line - we've all seen them (and probably driven into them!) - the point on the hill slopes where there's a line of tiny springs. However, whilst the hill-top is very well drained, it's also rather poor in clay content, which the termites need if they're going to build a good mound - sand doesn't work very well. So although the top of the hills are the best areas, in some place the termite densities increase towards the seep line - but then below the seep line the soil is too wet, so they suddenly disappear. And I guess that these very high hillside densities in Bologonja must be just the ideal location where there's plenty of clay, but just above the seep line - certainly there were seeps very close to some of these mounds a couple of weeks ago.

Bologonja termite mounds, happily functioning.

There's one more interesting thing that struck me when I was reading up on all this though - in South Africa, the highest densities of termite mounds above the seep lines indicate the areas of (broad-leaved) savannah. Below the seep lines lie grasslands. The authors of that paper found a relationship between local rainfall and the relative position of the seep line (more rain had washed the clay further down the slopes and closer to the rivers, which makes sense), and suggested that termite mounds could therefore be used to predict where woodlands might be if rainfally patterns change. Now, the really high densities on the Bologonja slopes are definitely on grasslands - but there's a good chance these were wooded not so long ago. Are they a relict of that time, and can we take their presence to even indicate areas where there were woods not so long ago? Or not? And  if not, is something different happening in Tanzania to South Africa? Also, there seemed to me to be an huge number of dead termite mounds, now visible only as a bare patch in other parts of the Northern Serengeti. Has something happened? Is this, too, an indication of major change in the savannah? Or does it just take so long for a termite mound to be washed away after the colony dies that we'd expect this many dead mounds? Hmmm.... All interesting stuff I think, any ideas?