My blog
Suppose Martin Lüscher was correct in his assertion that Macrotermes colonies represent the culmination of a long evolutionary trend among the termites toward ever more precise social regulation of nest temperature. The data he offered in support of this claim included traces of daily variation of temperature among a variety of termite nests, taken over several days. The Macrotermes nest indeed showed the steadiest temperature.
Is this sufficient evidence to claim that temperature of Macrotermes nests are therefore remarkably well regulated? This is a problematic conclusion on many fronts. This page outlines some of the arguments, for and against, this claim. Following pages outline new data which lead strongly to the conclusion that nest temperature is not a regulated property of termite colonies.
This and subsequent pages outline the results of a study of temperature and moisture distributions in mounds and nests of Macrotermes michaelseni in the arid savanna of northern Namibia. These results undermine the prevailing idea that temperature is regulated in Macrotermes nests. They also point to a surprising aspect of social homeostasis in termite colonies: water balance.
The driver-follower problem
Does steady nest temperature indicate regulation of nest temperature?
There is a wrinkle
The driver/follower problem
In any regulated system, work must be done to offset the tendency of the system toward equilibrium. To keep a house warm when it is cold outside, for example, energy must be expended to offset the house's losses of heat to the environment. In living systems, this is called the cost of homeostasis. In a variable environment, the cost of homeostasis varies with how wide the disequilibrium is that must be maintained: the colder it is outside, the more it costs to keep a house (or a body) warm.
Among other things, this points to a precise definition of a homeostatic system. It is not steadiness of a property that defines it, it is the work that is done to maintain the steadiness. This work is done by effectors, machines that mobilize energy and direct work in a particular way. In a standard homeostatic system, the work done by the effectors is managed by a negative feedback system, such that the effector works to offsets perturbations of a physical property P.
A negative feedback system always operates in an environmental context. Perturbations in P arise ultimately from energetic interaction with environments: perturbation of temperature, for example, arises from a flux of heat energy between the body and environment. This has implications for the ability of such systems to self-regulate. For example, perturbations can arise from environmental drivers of energy flow. Heat in sunlight, for example, is a very powerful driver of body temperature. To regulate a body's temperature, the effector must mobilize sufficient energy to work against the energy flows mediated through the driver. If the driver is very powerful compared to the effector's work capacity, the property cannot be regulated: there is insufficient capacity to meet the costs of homeostasis.
A similar problem arises from so-called resistance-capacitance (or RC) constraints. A fish living in water, for example, has a difficult time regulating its temperature. In this instance, the fish's gills present a low resistance pathway for heat to flow from the body to the water. In addition, the fish is embedded in a body of water that has a thermal capacity much larger than the fish's body. In this instance, the fish's body temperature willl tend to follow variation of water temperature. Following also has energetic implications, because the property is determined in large part by a much larger external capacity. Changing body temperature, for example, therefore involves not just work to heat the body, but the environment as well. Fishes cannot regulate body temperature because doing so would also mean regulating the ocean temperature. Regulation of body temperature can only occur if the effector can mobilize sufficient energy to bring the environmental temperature along as well.
Follower limitations are particularly relevant to the question of whether Macrotermes colonies can regulate their nest temperatures. The nest is embedded in a massive thermal sink: the surrounding soil. Can they manage the task? TOP
Does steady nest temperature signify regulation of nest temperature?
In a word, no. The reason lies in the driver-follower problem.
For a termite nest in a tree (right), for example, the nest is surrounded on all sides by insulating air, and the nest is exposed to heat input from the sun. Consequently, nest temperature is very strongly driven by the daily variation of incident sunlight. Heat absorbed by the nest tends therefore to be retained by the nest, and nest temperature will ringe widely through the day, limited by the nest's own thermal capacity. Nest temperature of arboreal nests therefore sits strongly to the driver side of the driver-follower problem.
A subterranean nest, like that found in Macrotermes sits at the other end of the driver-follower problem. In this instance, the nest is embedded in an immense thermal sink: the surrounding soil. Among other things, this ensures that deep soil temperatures will vary little through the day, even if soil surface temperatures are driven strongly by daily variation of insolation. Furthermore, the nest is coupled strongly to this immense thermal sink. This means, among other things, that nest temperature will follow closely the soil temperature at the depth of the nest.
A few calculations show just how daunting the problem is. One can estimate a nest's thermal capacity from the volume of the nest, the specific heat of the parent material and the void space in the mound. By our estimates, this amounts to about 2,500-3,500 kJ K-1, meaning that roughly 2.5-3.5 megajoules of energy are needed to raise the nest temperature by one degree Celsius. The metabolic rate of a typical Macrotermes colonyihas been estimated at 50-100 W, which amounts to an hourly metabolic heat production of 200-400 kilojoules. If the entire hourly heat production was concentrated into an instantaneous flash, this would be sufficient to raise nest temperature by about a tenth of a degree. In reality, the temperature elevation would be considerably smaller. It is then sobering to consider that the nest is coupled to the surrounding soil. The thermal capacity of a plug of soil 5 m in radius and 2 m deep (excluding the volume occupied by the nest is roughly 120 MJ K-1, about three orders of magnitude larger than the nest's. A "flash" of the metabolic heat produced in an hour would alter this temperature by less than a hundred microkelvins.
The conclusion is hard to avoid: there is simply not enough metabolic energy flowing through this system to budge temperature much. The temperature of the Macrotermes nest will be strongly driven by the surrounding soil. TOP
There is a wrinkle ...
N
o matter how problematic regulating nest temperature is, there is a wrinkle that complicates the issue: termites clearly respond to local temperature and modify their structures in ways that seemingly indicate regulation. The most dramatic example of this is the construction of the northward tilt of the mound, introduced here, and with video here.
This phenomenon should produce relatively uniform temperatures within the mound's spire. This does appear to be the case. The diagram below shows temperatures (a) and insolation (b) on the north (sunny) side of a spire and the south (shady) side of several mound spires.
In several of the mounds, we sawed off the spires and turned them around. We then measured temperature and insolation at several locations, and compared them against values measured on spires that had been left alone. Turning the spire around dramatically increases the insolation, and hence temperature, of the north-facing side of the mound. In other words, perturbing the architecture (rotating the mound) perturbs the distribution of mound temperature. This seemingly elicits rebuilding to undo the perturbation: we found evidence of more intense building on the north side of the rotated spires.TOP
