A Physics-Based Investigation of Termite Organization and Behaviour 

Rosalie Beaudin, Naia Kim, Mattias Sucher, Lucy Wiggers

Abstract

Though vastly considered pests outside of the scientific community, termites remain one of nature’s most formidable creatures. With colonies distributed across the globe, termite species display remarkably similar lifestyles despite the wide range of climates they inhabit. They have adopted distinct design strategies to navigate environmental pressures, which enable them to thrive as superorganisms. This paper highlights the physical basis of their internal organization and their interactions with the surrounding environment. In colony defense, soldier termites have evolved asymmetrical mandibles, which enable maximum elastic snapping potential and are optimal for use in tight spaces, such as termite mound corridors. Further, their vibrational processing capabilities allow them to forage wood of ideal density and moisture, identify surrounding predators, and create communication chains, with vibrational signals propagating at incredibly high speeds. Termite mounds are built via the cementation of saliva and faecal matter of optimized moisture content, allowing for peak structural stability and high porosity simultaneously. Fungi-growing termites optimize mound ventilation and heat transfer using diurnal, temperature-driven convection cycles and modify mound geometry for further thermoregulation and gas exchange. These innovative physical design strategies prove characteristic of termites’ eusocial success.

Introduction

Termites are known as a destructive species, particularly when they eat gardens or wooden structures valuable to humans. However, only 10 percent of the 2,750 known termite species are classified as pests (Fig. 1) (Kumar, 2024). In reality, the majority of termites are extremely beneficial to the ecosystem and this ecological success is mainly based on their complex social organization, such as in their termite mounds (Fig. 2). Termites are distributed all around the world, from tropical rainforests to savanna ecosystems. They are often confused with ants because of their similarities in appearance and size, but in fact, they belong to the same order as cockroaches, Blattodea. Termites are classified as infraorder Isoptera within Blattodea. They evolved from subsocial cockroaches, which exhibit temporary parental care and cooperation, becoming a eusocial species to increase survival.  

Fig. 1 Queen termite (genus Macrotermes) surrounded by workers (Kumar, 2024). 

Fig. 2 Ground mounds of Odontotermes termites strongly influence savanna productivity via ordered spatial distribution and modification of soil composition (Gamage, 2018). 

Termite colonies are highly organized social structures with a complex caste system and division of labor (Fig. 3). The number of termites is tightly regulated in each of the main three castes:  reproductives, workers, and soldiers. The king and queen are the primary reproductives who founded the colony. There are also supplementary reproductive termites that assist in egg production. Workers constitute the largest part of the colony and are responsible for foraging, building tunnels, and feeding the other colony members. Soldiers defend the colony with their large heads and powerful mandibles. Both workers and soldiers are sterile, can be male or female, and usually lack eyes. Pheromones have a significant role in caste differentiation, but the exact mechanism is still not fully known (Kumar, 2024).  

Fig. 3 Caste organization in termite colonies (Kumar, 2024). 

Termites obtain nutrients from plant matter like wood, grass, and leaves, but they cannot digest cellulose, a carbohydrate found in the cell walls of plants, on their own. Instead, they have microorganisms like protozoa and bacteria in their digestive system. These microbes produce enzymes which break down cellulose. Termites also act as crucial decomposers in ecosystems. They consume dead plants and return nutrients to the soil. 

Termites leverage the laws of physics in multiple ways. Soldier termites employ chemical and mechanical techniques in their defense strategies while developing specific adaptations to best defend their colony. The entire colony communicates through a complex vibrational system that travels in their remarkable environment. Termite mounds are architectural gems built with a combination of organic and inorganic materials providing both strength and durability. Ventilation and temperature regulation in these mounds are also based on thermodynamics and fluid dynamics.  

Soldier Termites Defense Mechanisms 

Termite’s eusocial lifestyle demands the division of tasks and specializations of casts within termite colonies. One of these roles is a soldier which protects the colony and its constituent workers, reproductive and nymphs. Different species have soldiers with various morphological adaptations that they use to best defend their colony. These adaptations are usually classified as chemical or mechanical. Species of termites have many ways of incorporating these defences and combining them to be as effective as possible. Mandibles are a primary method of defence used by soldier termites; however, not all mandibles are identical. They serve various purposes and thus have different constructions.  

Snapping Mandibles 

One type of mandible is the snapping mandible which comes in two forms: symmetrical or asymmetrical. They both take advantage of the many biological features within a termite, maximizing defense efficiency and power. Mandibles are comparable to latches and springs, creating power-amplifying systems which allow termites to produce incredibly strong strikes with little effort (Kuan et al., 2020). The snapping mechanism uses elastic potential energy stored in the termite’s mandibles that can be released on command by allowing them to slip past each other (Fig. 4).  By incorporating elastic potential energy, termites overcome the limitations of pure muscle fibers which require time to attain their maximum strength. Storing energy also allows the termite to strike quickly, powerfully and precisely by using their abductor muscle to create tension that when abruptly released becomes kinetic.  

Fig. 4. Snapping mandibles of termite soldiers. (a) In the symmetric snapping of Termes panamaensi, elastic energy is stored in both mandibles. (b) In the asymmetric snapping of Pericapritermes, elastic energy is stored in only the left mandible. (c) Morphology of the twisted left mandible of the Pericapritermes nitobei termite. The length of rotating section of the left mandible (L_A) and the mass of its subsections A₁ and A₂ (i.e., M_A1, M_A2) are required to estimate its moment of inertia (I_A) (Kuan et al., 2020). 

Symmetrical mandibles  

Symmetrical snap mandibles are thought to be the more primitive of snap mandibles (Deligne & Blum, 1981). They are an evolutionary trait that later evolved into asymmetrical snap mandibles; however, they still exist and are incredibly effective. The symmetric mandible attack consists of 3 parts: deformation, strike, and return. The mandible and its attack process are shown in Fig. 5. During the deformation, mandible-closing muscles are used to create deformation on both sides of the mandible (Fig. 5A). This stores energy from muscle contraction that can be released instantaneously. In the termite species Termes panamaensis, the contraction process takes 0.78 ± 0.25 seconds, so unlike traditional mandibles that use only direct muscle contractions, T. panamaensis can then use their snap mandibles to release the summation of 0.78 ± 0.25 seconds of muscle contractions into a 0.025 millisecond strike (Seid et al., 2008). When the termite is ready to strike, the mandibles will slip past each other in a scissor-like motion rapidly releasing the energy built up from the deformation step (Fig. 5B). Once released, the tips of the mandibles travel 53.8 ± 8.53° and as mandibles are around 1.484 ± 0.063 mm long, this gives a mean velocity of 56.0 ± 8.87 ms⁻¹ throughout the strike. The strike creates 1.5 W, has a mean force of 54mN and requires only 0.15 mJ to execute (Seid et al., 2008). If a human were to flick at the same velocity as the symmetrical mandible snap, it would generate a force of 1600N, a power of 89,600 W, requiring only 313.6 J of input energy. This flick would have the force to pick up a 160 kg mass, and the power to run 75 American homes simultaneously, while only requiring 10% of an AA battery’s energy. Between the mandibles is a modified mouthpart, the labrum (Fig 5A), that serves a crucial role in the defense mechanism by making the strike reusable. As the strike occurs, the labrum is compressed, preventing the mandibles from over-rotating and jamming, as would occur if the mandible were removed (Seid et al., 2008). Allowing the soldier termite to reset its defense mechanism is crucial in making this adaptation viable.  

Fig 5. (A) A scanning electron micrograph of the head of a soldier termite. (B) A schematic of mandible movements before and during a strike. Before the strike, the mandibles bend, their point of contact (blue arrows) moving towards the head. The strike is executed when the mandibles cross (red arrows). The region captured by the high-speed video camera in (D) is indicated by a pale blue box. (C) A serial reconstruction of the head including the muscles, brain and frontal gland. The position of the mandibles is indicated by a grey dashed outline. Scale bars indicate 0.5 mm (Seid et al., 2008).  

While the mandible’s strike velocity is very impressive, the force and power are not, at least without accounting for scale. Considering the termites’ mandible-closing muscles weigh 0.15 ± 0.08 mg and have a cross-sectional area of 0.17 mm², we can calculate their specific power and muscle stress (Seid et al., 2008). Specific power is found by dividing the power (P) by the mass of the muscle (m), and is given by: 

This provides an understanding of how well mass is translated into power. T. panamaensis has a specific power output of 11 MW per kg of muscle. The specific power output of rocket engines can range from 2 MW/kg to 10 MW/kg during launch. In addition, muscle stress is found by dividing the force (in Newtons) by the cross-sectional area (in m²) of a muscle: 

This provides an understanding of the efficiency of a muscle by evaluating how much force a muscle can exert relative to its size. In this case, T. panamaensis generates 3 MN per m² of muscle, 6 times the force of a saltwater crocodile bite (Seid et al., 2008). 

Asymmetrical Mandibles 

Pericapritermes nitobei is a termite species with soldiers possessing asymmetrical mandibles. They use the elastic properties of deformed mandibles to produce strikes that exceed the power that the termites’ muscles could produce directly. They have a stiff and curved right mandible and a flexible and twisted left one so that all the elastic energy is stored in a single mandible instead of both. This is done by pressing the left mandible against the right one and contracting the mandible-closing muscles together (Deligne & Blum, 1981). The rigid mandible bends the flexible one, building energy in the flexible one. Its shape creates a pivot point (Fig. 6c) that the mandible bends about. Then, when released, the termite must raise its antenna to protect itself from its own strike (Fig 6a). The curve of the right and twist of the left reduce friction forces to maximize the stored energy transferred into kinetic energy. This loading faze takes 261.3 ± 43.1 ms (Kuan et al., 2020). When striking, the left mandible sweeps right in 21.7–43.4 μs attaining an average peak velocity of 111.1 m/s. The swinging part of the mandible has an average length of 28.0 mm and weight of 30 μg, and these strikes had an average peak force of 131.0 mN and require an average energy exertion of 30.2 μJ. (Kuan et al., 2020). If a human were to flick at the same velocity as the asymmetrical mandible snap, it would generate a force of 3,174 N, a power of 352,663 W, and only requiring 1234.32 J. This flick would have the force to pick up a 324 kg mass, the power to run 294 American homes simultaneously, while only requiring 40% of a AA battery’s energy. 

Fig. 6. Behavioural phases of Pericapritermes nitobei mandibular snaps. (a) Stage 1: right mandible pressed against the left mandible for 261 ± 43 ms before snaping. Stage 2: termites raised their antennae 47.5 ± 22.5 ms before the snap. Stage 3-4: Mandibular snaps were performed over 21.7–43.4 μs. The peak linear velocity (V_MT) of two soldiers at 8.68 μs were 132.4 and 89.7 m/s. (b, c) are some of the images used to create the snapping processes in (a) (Kuan et al., 2020).  

The average peak energy of the asymmetrical snap mandible is much lower than that of the symmetrical, revealing that significantly less elastic energy is stored. However, the speeds attained by the asymmetrical mandibles result in higher acceleration and therefore greater force, making them considerably more efficient.   

This equation is used to calculate the force of the strike by using the collected data from the study by Kuan et al.   

This equation describes the kinetic energy in the strike using measurements collected in the study. This is used to find the elastic energy from the distortion portion of the strike by the law of conservation of energy. Despite the velocity and acceleration of the asymmetrical mandible being significantly higher than those of symmetrical mandibles (111.1 m/s vs. 56 m/s), its moment of inertia is so low that it reduces the energy required to reach those speeds beyond what a symmetrical mandible can achieve. Asymmetrical mandibles require 30.2μJ of energy input while symmetrical mandibles require 0.15mJ, almost 4 times more. Snap mandibles are practical in close quarters, like the tunnels of termite mounds, because they attack predators without requiring the opening of the mandibles (Deligne & Brum, 1981). This makes them ideal for drywood termites that do not need to scavenge and simply need to keep predators out of the colony. Between the two forms of snap mandibles, asymmetric mandibles are proven to be much more efficient than symmetrical ones (Kuan et al., 2020).  

Vibrations and Communication 

Termites Foraging Decisions 

Termites are often perceived as ravenous feeders that eat all the wood that they find. However, termites are selective when it comes to wood choice. Drywood termite species use vibrations generated during foraging to determine the size of a wood block (Hager & Kirchner, 2013). The vibrations of a wooden block are influenced by many factors. Its shape and how it is supported play a crucial role, along with the block’s material properties. These properties include mass, density, and internal damping. A study on foraging decisions was conducted on the basal Australian termite species Coptotermes acinaciformis (Oberst et al., 2018). They found that these termites were more attracted to higher density and early wood (Fig. 7). Termites can process vibrational signals by detecting waves and their attenuation. The reflection, scattering, and dispersion of waves, along with variations in their amplitude, provide critical information related to wood choice. It enables termites to detect tree knots, identify other insects (including predators), and assess wood density or moisture levels (Oberst et al., 2018). Moisturized wood has a lower resonance frequency, which termites can detect and choose over dry wood (Fig. 7). 

Fig. 7 Results of permutation analysis on Coptotermes acinaciformis. Preference for a certain permutation of wood properties; circles indicate a relative preference, a value of one indicates relative aversion, and a value of less than one indicates a relative aversion; triangle markers represent lie in a zone of indifference (Oberst et al., 2018). 

Vibrational Long-Distance Communication 

Termites also use vibrations to communicate. They use body movements to produce vibrations that can be transmitted through the substrate and detected by other colony members (Kumar, 2024). These vibrations can convey diverse types of information depending on the context. When disturbed or threatened, termite soldiers tend to drum their heads against the substrate to create a pulsed vibration (Hager & Kirchner, 2014). Each of the pulses originates from a single tap of the head against the ground. A study on vibrational alarm communication was done with the South African termite species Macrotermes natalensis (Hager & Kirchner, 2013). One goal of the study was to analyze the propagation, velocity, and attenuation of drumming signals in the nest. To determine the attenuation properties, they compared the highest vibration levels at two different points and confirmed that the amplitude of vibration waves decreases exponentially with distance. The study concluded that Macrotermes natalensis produce vibrations with a pulse repetition rate of 11 Hz. They also observed that most energy is found between 1 and 5 kHz and that the amplitude of signals produced by a single termite decreases with distance (Figure 8.C). 

Fig. 8 Typical drumming signals of Macrotermes natalensis (Hager & Kirchner, 2013). (A) Oscillogram of several pulse groups produced by several termites. (B) Oscillogram of a pulse group produced by a single termite. (C) Frequency spectrogram of the pulse group shown in B. 

However, M. natalensis were able to spread the alarm throughout the colony in impressive time (Fig. 9). The study observed a transmission velocity of 1.3 m s⁻¹ on this long-distance alarm (Hager & Kirchner, 2013). This is because soldiers respond to the drumming signals of their nearby colony members by producing similar drumming signals themselves, creating a chain of communication within the colony that prevents the amplitude from decreasing. 

Fig. 9 Long-distance communication of M. natalensis (Hager & Kirchner, 2013). 

Chordotonal Organ 

To receive the alarm signal and participate in the communication chain, termites use a vibratory sensing organ in their legs called the chordotonal organ (5). This subgenual organ is in the tibia and contains chordotonal sensilla, which are sensory structures (Fig 10.A) (Shaw, 1994). The organ is suspended in hemolymph (insect blood) between the tibia walls. It can sense vibrations transmitted through the hemolymph (Shaw, 1994). This overly sensitive nervous system allows termites to react to different short-time vibrations coming at their legs and decide how to react when threatened. 

Fig. 10 Schematic diagram of three major steps envisaged in the evolutionary transformation of a vibration-sensitive subgenual organ into an auditory organ (Shaw, 1994). 

Overall Design and Stability, Material Properties of Termite Mounds 

Not all termite species build mounds, but those who do create one of the most impressive architectural structures in the insect world. Some are as tall as 8-10 m. These mounds are mostly found in regions such as Asia, Australia, Africa, and South America (Fagundes et al., 2021). The primary reason for this mound-dwelling behaviour is to regulate and stabilize the living environment of the colony, protecting termites from variant weather and external threats. The outer structure of the mound is partially influenced by solar heating, local wind, and other factors, so mounds require specific microclimate control functions to maintain optimal living conditions for the termites (Fagundes et al., 2021). This is why the shape of each mound is unique depending on the local climate and environmental conditions. Understanding their structure, durability, and materials used can provide insights into the surrounding environment. 

External Structure 

Termite mounds have different forms depending on the local climate and available materials: domes, cone-shaped, cathedral-shaped, and wedge-shaped (Fig. 11). However, there is no correlation between termite species and mound structure defined. For example, a study done by Korb & Linsenmair shows that two different structures of mounds of Macrotermes bellicosus termites were found in the same geographical region (Zachariah et al., 2020). Still, there is a relationship between the shape of mounds and the weather conditions observed. For example, according to Fagundes’ study about termite mound structures, mounds that are exposed to strong solar irradiance are cone-shaped and pointed towards the sun, while mounds under shade are vertical domes. While mounds may have different appearances, they generally share a common internal structure. Typically, a mound consists of a solid outer wall, multiple chambers, tunnels, and a nest. The outer wall acts as a shield, protecting the interior against predators, fires, and large animals. It is solid when dry, but easily taken apart when wet. The nest includes nursery rooms, fungus chambers, the royal cell, and connecting tunnels (Van Thuyne et al., 2023). (Fig. 12) 

Fig. 11 Variation in natural termite mound morphologies. (left) Cathedral-shaped Nasutitermes triodiae mound; (top middle) cone-shaped Macrotermes michaelseni mound with tilted spire; (top right) wedge-shaped Amitermes laurensis mound; (bottom middle) mushroom-shaped Nasutitermes triodiae mound; (bottom right) cone-shaped Macrotermes michaelseni mound. 

Fig. 12 Schematic representation compiling field observations and literature models of the internal layout of the mound compartments. The dashed white line denotes the level of the surrounding ground. The fungus chambers are found within the nest area (Van Thuyne et al., 2023). 

Materials Used and Their Characteristics 

The mound is built with two types of materials: exogenous and faecal material. Exogenous material includes soil and woods transported by the strong mandibles of termites and faecal material includes rectal content, deposited by termites. This faecal material holds a certain level of moisture and fluidity which contribute to the strength and cohesion of mound structure, especially in the construction of walls. The solidity is determined by saliva, which helps moisten exogenous materials and the moisture contained in the faecal material, as well as the chemical structures of saliva and fecal material. This proportion varies from termite to termite (Noirot & Darlington, 2000). Varied species in the same habitat, therefore, can produce nests of quite different solidity. 

Structure Stability 

Termite workers have the ability to make “boluses”. Boluses are small clumps of moist soil. The moisture level can be between 15% to 60%. This is important because boluses act as basic building blocks to the mound. Termites mix moist soil and their faecal materials, and this mixture acts as a biocement, which is binding material produced by activity of microorganisms (Ariyanti et al., 2011). However, if the soil has no organic matter, and is too dry or even too moist, it is hard for termites to work with. This implies that organic matter is important to build durable mounds (Zachariah et al., 2020). Also, mounds can get wet and dry as they are exposed to weather fluctuations, and this also contributes to the stability of the structure. This concept is explored more in the next section. The size of the bolus depends on the caste of the termites. Major workers tend to make larger boluses than those of minor workers. A study on the mound of Odontotermes obesus, a fungus-growing termite common in India, compares the physical properties of mound soil and control soil, focusing on density, porosity, particle size distribution and Atterberg limits. The Atterberg limits are fundamental measures of the consistency or behavior of fine-grained soils, defined by specific values of moisture content. It consists of liquid and plastic limits. This study found that the soil of the termite mound was denser at the bottom due to gravity causing the consolidation of soil over time. The variation of density was from 1.42 g/cm³, at the top, to 1.68 g/cm³, at the bottom (Fig. 13). This is 9–23% lower than the maximum density achievable in the lab, suggesting that termites construct their nests on the wetter side of the optimum moisture content (Kandasami et al., 2016). 

Fig. 13 (a) Variation of dry density in mound with elevation. The abscissa shows the height of the termite mound from ground level (GL). (b) Variation of porosity at different levels of the mound. The abscissa shows the height of the termite mound from ground level (Kandasami et al., 2016) 

There was one more component that varied depending on the parts of mounds: porosity. The porosity was increasing, from 37% to 47%, as the mound’s height increased (Fig. 13). Lastly, both the mound and control soils had consistent liquid and plastic limits of 33% and 17%, respectively, across different sections of the mound. The liquid limit is the moisture content at which a fine-grained soil stops flowing like a liquid, and the plastic limit is the moisture content at which a fine-grained soil can no longer be molded without cracking (Kaliakin, 2017). These limits were consistent across various mounds in the study area (Kandasami et al., 2016). Termite mounds also show an interesting characteristic when interacting with water. Zachariah studied strength and cementation in a termite mound. Termite mound samples fell apart by about 20% when soaked in water for 100 minutes, and rapidly collapsed soon afterward. Erosion occurred more quickly at the top of the mound, where the soil density was lower, and porosity was higher. When the mound soil was disintegrated, and reconstituted to its original dry density, it collapsed much faster, completely collapsing in 60 minutes, when compared to the original termite mound (Fig 14), which showed greater resistance to weathering due to termite-induced cementation (Zachariah et al., 2019). This study demonstrates the importance of termite construction in improving the soil’s stability against natural weathering processes. 

Fig. 14 Stability analysis of termite mound soil using alternating wetting and drying and experiments. Samples were collected at different heights from the ground level (GL) of a termite mound and were subjected to alternative cycles of wetting and drying. The results were compared with the weathering of reconstituted termite mound soil (Zachariah et al., 2020). 

Material strength  

Termite mounds can last for several years to even centuries and they are 10 times stronger than the surrounding soil. This strength can be explained by multiple factors including bio-cement, matrix suction, clay minerals, termite salivary amylase and polysaccharides (Zachariah et al., 2020). A study by Zachariah explores specifically the effect of drying on soil strength by mimicking the soil drying of the outer wall of termite mounds. The strength of the soil increased as wet soil particles settled, and water evaporated from the spaces between them. When water evaporates, the smaller particles get pulled toward larger particles, so larger ones end up surrounded by smaller ones. This process repeats and helps the soil particles to lock together and create “capillary bridges”. The capillary bridge is a little amount of liquid between two particles, and it generates adhesion forces at the micro-scale (Yang et al., 2021). This process allows the particles to interact more effectively due to van der Waals forces. They stay stable even when they get wet again, indicating that this process is resilient. The key to strength is a variety of particle sizes. If small particles are removed, the remaining break apart when wet. Having a mix of particle sizes is crucial for forming strong connections between particles. The following graph (Fig. 15) shows the percentage reduction in the weight of different type of soil in function of time. Samples taken from abandoned termite mounds showed the highest resistance to weathering by water, followed by samples from occupied mounds, and then controlled soil samples. 

Fig. 15 Weathering of mound soil, reconstituted mound soil and control soil during repeated cycles of wetting and drying. Open circles represent averages of six values (Zachariah et al., 2020). 

In the same study, it was found that termites prefer working with soil that has 30% moisture content. It is difficult for termites to work with dry soil because they cannot achieve best moisture levels by their faecal material alone. Further, soil that was too wet affected the movement of the termites (Zachariah et al., 2017). Interestingly, maximum self-weight consolidation was achieved at this level, and therefore the peak compressive stress. The author defined self-weight consolidation as follows: densification of soil under its own weight without application of any external force (Zachariah et al., 2020). At 30% moisture content, the peak compressive stress was around 1600-1800kPa (Fig. 16), and it was close to the peak compressive stress found in actual termite mound soil. This was significantly different from the stresses recorded in soil modeled at 40%, 50%, and 60% moisture contents, indicating that the optimal moisture level for strength is 30%. Another study by Zachariah measured the compressive strength of non-termite mound soil (control soil) and it was about 1/10th of the termite mound soil (Zachariah et al., 2019). 

Fig. 16 Role of moisture in soil strength. (a) Peak compressive strength of soil after compaction or self-weight consolidation at different moisture contents and drying. (Zachariah et al., 2020). 

Thermoregulation and fluid dynamics in termite mounds 

Termites are well-recognized for their characteristic mound structures. Reaching several metres high, these massive mounds are an impressive display of collaborative ability, exhibiting termites’ eusocial and colonial lifestyle. Termite mounds are constructed to optimize their specific thermoregulatory and ventilatory needs, facilitating the transfer and exchange of heat and fluids through the mound’s porous walls (Ocko et al., 2017). This becomes especially relevant in termite species of the subfamily Macrotermintinae, which cultivate symbiotic fungi in combs deep within the mound, acting as a nutrient source for the colony. Hence, they require effective mechanisms to maintain thermal and respiratory homeostasis, ensuring the survival of their brood and symbiotic counterparts (King et al., 2015). Further, the vast range of environments that termites inhabit provides specific challenges in individual termite colonies; they must come up with distinct design strategies to optimize the internal microclimate of their mounds.  

Mechanisms of ventilation and heat transfer 

With colonies ranging up to a few million individuals and coupled with fungal gardens of high biomass, termite mounds are sources of high metabolic activity; they produce significant amounts of CO₂ and CH₄, which must be mitigated (Räsänen et al., 2023). Additionally, Macrotermintinae mounds must maintain a narrow temperature range (around 30 °C) to optimize fungi growth. Hence, they require effective ventilation to regulate gas exchange, humidity levels and heat (Ndlovu & Perez-Rodriguez, 2018). While non-fungi-growing mounds may use open shafts for heat transfer and airflow, this mechanism is too variable for fungi-growing termites due to the impact of external temperature fluctuations (Räsänen et al., 2023). Instead, they employ effective internal ventilation, a two-step mechanism. First, the bulk movement of gas from the underground nest, where most termites reside, to the mound surface must be facilitated. Second, gas exchange must occur across the surface mound walls. While gas exchange at the surface occurs primarily via passive diffusion across the porous material, the bulk transport of gas from the nest to the surface requires more active means (King et al., 2015). The mechanism of this transport is a highly debated topic amongst the scientific community. Experts propose temperature-driven convection, external wind flow, buoyant forces driven by metabolic heat, or a combination of those, as the main driving source (Singh et al., 2019). However, most scholars agree that thermal convection is the primary ventilation mechanism, which is explored below.  

Ventilation via thermal convection 

Thermal convection in fungi-growing termite mounds is driven by external oscillations in diurnal (daily) ambient temperatures. When coupled with internal mound geometry, this allows for the formation of a closed convection cell (King et al., 2015; Räsänen et al., 2023; Hariyanto et al., 2024). This ventilation method is common across several fungi-growing termite species, although the efficiency can differ in mounds of different temporal climates. In the termite species Odontotermes obesus (Macrotermitinae), found in Southern Asia, mounds are characterized by radial, flute-like shafts surrounding a central chimney, which extends deep into the underground nest. These flutes contain an internal network of conduits which form a continuum into the chimney, resulting in a distinct macroporous interior for gas and heat flow (Fig. 17). Importantly, O. obesus mounds are found in shaded, temperate regions of low wind activity.  

Fig. 17 Termite mound of O. obesus. A), B), and C) show the side, top and cross-section of the mound, respectively. D) and E) reveal the internal network of conduits (cast in white gypsum) (King et al., 2015).  

The convection mechanism is as follows. During the day, when the sun’s influence is strong, the surface conduits warm comparatively to the chimney; warm air within the conduits rises, pushing cooler air down the chimney in a convection loop. The opposite occurs when the thermal gradient is reversed at night: the cooler exterior pushes air down the conduits while warm air flows up the chimney (Fig 18) (King et al., 2015). Not only does this enable gas exchange, but thermoregulation. While temperatures fluctuate within the conduits and central shaft, the nest temperature remains relatively constant. The convection cycle allows for thermal stability, dissipating extra heat and drawing in cool air where needed. Paired with the convection cell, conductive processes involved in heat transfer between deep underground soil and the nest create a two-phased thermoregulatory mechanism (Hariyanto et al., 2024).  

Fig. 18 Simplified geometry of an O. obesus mound showing the central chimney and outer surface conduits (Left). Thin surface conduits heat and cool rapidly in an alternating diurnal cycle, driving convective flows in a cyclical, reversible manner from the conduits to the chimney (Right). Flow cycles enable bulk gas movement and heat transfer. [Adapted from King et al., 2015].  

This ventilation mechanism is relatively simple but effective given the low-stress environment in which O. obesus reside. In contrast, Macrotermes michaelseni, a close relative to O. obesus, is found in the southern African savanna. Hence, their mounds are far more subject to solar irradiance, heating the mound non-uniformly as the day goes on (Ocko et al., 2017). Ventilation primarily occurs through the same convection mechanism, where gases circulate between the conduits and the chimney driven by daily thermal gradients. However, direct solar heating introduces variability by causing uneven warming of the mound. At certain times of the day and positions of the sun, different conduits will heat up relative to the rest of the mound. This creates non-uniform temperature differences between the mound conduits and the central chimney, and these are predictable as a function of time and an azimuth angle, f, which describes the orientation of a conduit relative to the centre of the mound. In the morning, the east side of the mound is directly heated by the sun, and convection occurs mainly between the conduits on either side of the mound, with less airflow in the chimney. This cycle flips in the afternoon when the west side is primarily heated. At night and midafternoon, the solar heating is more uniform, and convection cells mirror those seen in O. obesus (Fig. 19). In general, the convection cells in M. michaelseni are still able to provide effective ventilation despite non-uniform solar flux.  

Fig. 19 Side view of M. michaelseni mound (Left). Convection cells in M. michaelseni mounds following daily thermal fluctuations coupled with angular solar gradients (Right). Cell cycles are affected by non-uniform heating as the sun rises and sets, heating the east and west sides of the mound at different times of the day. Further, the mounds are in the southern hemisphere and the north side of the mound consistently receives more sun than the south side [Adapted from Ocko et al., 2017].  

Internal gas flow speeds 

Given the conduit-chimney temperature differences in both M. michaelseni and O. obesus mounds, the gas flow speed can be modelled as a result of the convection cells. The uniform temperature difference between conduits and central chimney in the shaded O. obesus mounds is given by: 

(1)  

The non-uniform temperature difference between conduit and chimney in the exposed M. michaelseni mounds is given by:  

(2)  

King et al. and Ocko et al. propose a simple physical model for the mean flow speed observed, approximating the conduit-centre convection loop as a pipe of radius r in a closed vertical loop of height h. The driving pressure in the pipe follows: 

where 𝜌 is air density, 𝛼 is the coefficient of thermal expansion and ΔT is given by Equation 1 or 2 respectively. Poiseuille’s Law, which describes the laminar volumetric flow rate, Q, of fluid in a tube, then gives: 

where is the dynamic viscosity. Note that the factor of 2 accounts for resistance on both sides of the loop. Finally, internal flow velocity, , is derived from a related equation for volumetric flow rate, given by: 

While this model has geometrical limitations, it produces flow velocity speeds within the order of what is observed in both O. obesus mounds and M. michaelseni mounds. Importantly, it is clear how the temperature difference between the conduits and the central shaft is proportional to the speed of fluid flow, even while those temperature differences fluctuate from mound to mound. As the conduits heat and cool in diurnal cycles, the transfer of heat and gas is facilitated, promoting effective ventilation.   

CO₂  exchange across porous mound walls  

Inevitably, gas flow rates and velocity fields are driven by ventilation, which contributes to the efficacy of CO₂ diffusion. In O. obesus mounds, ventilation is slower during the day, as the external temperature more closely matches the mound temperature and the heat difference between conduit and chimney is less extreme. Hence, flow rates are slower, and CO₂ levels climb to around 5% in the nest, though this is well within the termites’ tolerance. At night, the flow rate increases with stronger convection cycles, and CO₂ can be exchanged more effectively, leading to overall lower recorded levels (around 1%) (Fig. 20).  

Fig. 20 CO₂ levels in O. obesus mound over a one-day cycle. Distinction made between the nest and the top of the chimney, 1.5 meters above [Adapted from King et al., 2015].  

Interestingly, CO₂ fluctuation patterns within M. michaelseni mounds differ from the daily cycle of buildup and release, as in O. obesus mounds. Instead, CO₂ levels remain constant throughout the day. Still, they are higher on average at around 5% in the nest and chimney (Fig. 21). This indicates that gas exchange is more limited by transport across the mound surface than internal mixing and bulk flow, contrary to fluid behaviour observed in O. obesus mounds (Ocko et al., 2017).  

Fig. 21 CO₂ levels in the mound of M. michaelseni are steady and uniform, indicating that gas exchange is more limited by transport across the surface rather than bulk flow and mixing, as compared to the mound of O. obesus (inset) (Ocko et al., 2017; King et al., 2015).  

These gas exchange trends can be explained by the hotter and generally windier environment in which M. michaelseni reside. In fact, the homogeneity of the gas concentration within the mound can be attributed to transient mixing, which is caused by greater thermal fluctuations and the impact of higher windspeeds on the porous mound surface (Okco et al., 2017). Further, the limited CO₂ diffusion in M. michaelseni mounds suggests a trade-off between gas exchange and thermoregulation. Experts agree that termite mound design prioritizes either gas exchange or thermoregulation depending on the thermal environment, a distinction largely reflected in wall thickness and the presence of complex surface features (Ocko et al., 2017; Fagundes et al., 2021). In low-stress, temperate climates, such as where O. obesus colonies are located, the need for thermoregulation is less of a priority. Walls become thinner, and external surface complexity increases to optimize gas exchange. In more extreme environments, thermoregulation becomes increasingly important. Walls are often thicker, and surface complexities decrease, either to avoid high solar exposure (in hotter environments, such as where M. michaelseni reside) or to provide more uniform, effective heating (in colder environments). This is further observed in Macrotermes bellicosus, a termite species that builds their mounds in forests and savannas alike. Studies found that forest mounds tended to have thinner walls, prioritizing gas exchange, while savanna mounds had thicker walls for enhanced thermoregulation (Korb, 2003). The relationship between thermoregulation and mound structure is further explored in the next section.  

Mound shape and thermoregulation 

Alongside ventilation as a heat regulation mechanism, termite colonies adjust their mound shape to optimize internal mound temperature (Fagundes et al., 2021; Korb, 2003). There are two well-defined ends of the termite mound structural spectrum: the dome-shaped mound and the cone-shaped mound. The shape determination is influenced heavily by the amount of external solar irradiance. This is modelled by Fagundes et al. in Fig. 22. Mounds built in colder areas have shorter, dome-like structures with a uniformly exposed surface area and a smaller distance between the nest and mound surface. In this way, they can take in more heat closer to the nest where they primarily reside. In contrast, mounds with high sun exposure prove much taller and leaner, to reduce the surface area exposed to the sun. The distance between the outer surface and the nest is also much larger so that surface conduits can dissipate heat via convective cooling before it reaches the termites in the nest (Fagundes et al., 2021).  

Fig. 22 Temperature distribution of optimized mound shapes for varying solar irradiance ratios, , and constant wind speed and solar zenith angle. A clear trend between irradiance and cone shape is observed. Note that temperature scales are individualized for clarity (Fagundes et al., 2021).  

Additionally, cone-shaped domes found in hotter climates also show distinct inclinations related to their geographic location. Specifically, inclinations mirror the solar zenith angle of their location, limited only by the gravitational constraints on structural stability. The zenith angle, which is the angle between the sun’s rays and the vertical direction, varies with geographical latitude (Fig. 23) Directed toward the incoming solar irradiance, the mounds concentrate solar heat on their highest peak, and, consequently, the part of the mound furthest away from the nest. In this way, the external heat is mitigated. 

Fig. 23 Temperature distribution of optimized mound shape for varying zenith angles, , and constant solar irradiance and wind speed. Notice the clear trend between zenith angle (), mound inclination () and heat distribution. (Fagundes et al., 2021). 

Conclusion

The organization of termite colonies is complex, employing a plethora of physics-based design strategies to ensure their survival. First, colonies require a caste of soldier termites, who use their unique snap mandibles for defense. Snap mandibles build up elastic potential energy via deformation and then release this energy rapidly in a scissor-like motion, allowing them to reach incredible speeds. They can be symmetrical, storing energy in both mandibles and asymmetrical, storing energy in only one, though asymmetrical mandibles provide greater snap efficiency. Further, termites have a distinct communication system, using vibrations to alert colony members in impressively brief time frames. They perceive vibration signals via a sensory organ in their legs, allowing them to relay information rapidly throughout the mound. Termites also use vibrations to source specific wood in foraging, based on its mass, density, and damping. Remarkably, termites are notable ecosystem engineers, building some of the most extraordinary structures in the world. The unique shapes of termite mounds are easily distinguishable, though more so by the surrounding environment than by the species inhabiting the mound. The mounds serve as protective shields for termite colonies and are built with both organic and inorganic materials, which contribute to their strength and stability. Soil moisture and organic composition play an important role in helping mounds endure erosion while maintaining their strength and shape over time. Finally, mounds contain specific regulative abilities. In fungi-growing termites, the use of a simple, but effective, ventilation system driven by external temperature fluctuations helps to mitigate excessive gas concentrations within the mound and maintain suitable nest temperatures. Termite mound shape is designed to optimize these thermoregulatory and ventilatory processes, allowing for the creation of distinct mound microclimates. Understanding the physical organization and architectural abilities of termites provides valuable insights into their adaptability as superorganisms.  

References

Ariyanti, D., Handayani, N. A., & Hadiyanto, H. (2011). An Overview of Biocement Production from Microalgae. DOAJ (DOAJ: Directory of Open Access Journals). https://doi.org/10.12777/ijse.2.2.31-33 

Deligne, J., Quennedey, A., & Blum, M. S. (1981). The Enemies and Defense Mechanisms of Termites. Elsevier EBooks, 1–76. https://doi.org/10.1016/b978-0-12-342202-6.50008-3 

Fagundes, T. M., Ordonez, J. C., & Yaghoobian, N. (2021). The role of mound functions and local environment in the diversity of termite mound structures. Journal of Theoretical  Biology, 527, 110823. https://doi.org/10.1016/j.jtbi.2021.110823  

Hager, F. A., & Kirchner, W. H. (2014). Directional vibration sensing in the termite Macrotermes natalensis. Journal of Experimental Biology, 217(14), 2526-2530. https://doi.org/10.1242/jeb.103184  

Hager, F. A., & Kirchner, W. H. (2013). Vibrational long-distance communication in the termites Macrotermes natalensis and Odontotermes sp. Journal of Experimental Biology, 216(17), 3249-3256. https://doi.org/10.1242/jeb.086991  

Inta, R., Lai, J. C. S., Fu, E. W., & Evans, T. A. (2007). Termites live in a material world: exploration of their ability to differentiate between food sources. Journal of the Royal Society Interface, 4(15), 735-744. https://doi.org/doi:10.1098/rsif.2007.0223  

Kaliakin, V. N. (2017). Example problems related to soil identification and classification. In Elsevier eBooks (pp. 51–92). https://doi.org/10.1016/b978-0-12-804491-9.00002-1 

Kandasami, R. K., Borges, R. M., & Murthy, T. G. (2016). Effect of biocementation on the strength and stability of Termite Mounds. Environmental Geotechnics, 3(2), 99–113. https://doi.org/10.1680/jenge.15.00036 

King, H., Ocko, S., & Mahadevan, L. (2015). Termite mounds harness diurnal temperature oscillations for ventilation. Proceedings of the National Academy of Sciences USA, 112(37), 11589-11593. https://doi.org/10.1073/pnas.1423242112  

Knight, K. (2014). Termite legs sense super-short time differences. Journal of Experimental Biology, 217(14), 2433-2433. https://doi.org/10.1242/jeb.110338 

Korb, J. (2003). Thermoregulation and ventilation of termite mounds. Naturwissenschaften, 90(5), 212-219. https://doi.org/10.1007/s00114-002-0401-4 

Kuan, K.-C., Chiu, C.-I., Shih, M.-C., Chi, K.-J., & Li, H.-F. (2020). Termite’s Twisted Mandible Presents Fast, Powerful, and Precise Strikes. Scientific Reports, 10(1). https://doi.org/10.1038/s41598-020-66294-1 

Lívia Fonseca, N., Paulo Fellipe, C., Silva, P. S., Leonardo Bonato, F., Danilo Miranda, R., & DeSouza, O. (2019). Dataset on Substrate-Borne Vibrations of Constrictotermes cyphergaster (Blattodea: Isoptera) Termites. Data, 4(2), 87. https://doi.org/https://doi.org/10.3390/data4020087  

Ndlovu, M., & Perez-Rodriguez, A. (2018). Temperature fluctuations inside savanna termite mounds: Do size and plant shade matter? Journal of Thermal Biology, 74, 23-28. https://doi.org/10.1016/j.jtherbio.2018.03.004  

Noirot, C., & Darlington, J. P. (2000). Termite nests: Architecture, regulation and Defence. Termites: Evolution, Sociality, Symbioses, Ecology, 121–139. https://doi.org/10.1007/978-94-017-3223-9_6 

Oberst, S., Lai, J. C. S., & Evans, T. A. (2018). Key physical wood properties in termite foraging decisions [Article]. Journal of the Royal Society Interface, 15(149), Article 20180505. https://doi.org/10.1098/rsif.2018.0505  

Ocko, S. A., King, H., Andreen, D., Bardunias, P., Turner, J. S., Soar, R., & Mahadevan, L. (2017). Solar-powered ventilation of African termite mounds. Journal of Experimental Biology, 220(Pt 18), 3260-3269. https://doi.org/10.1242/jeb.160895  

Hariyanto, S., Sumanto, Y. D., Aji, B. W., Nastangini, Chasanah, S. N., & Raharjo, A. A. (2024). Two phase heat transfer on porous media base termite nest structure with finite element method. Mathematical Modelling of Engineering Problems, 11(6), 1520–1528. https://doi.org/10.18280/mmep.110612  

Räsänen, M., Vesala, R., Rönnholm, P., Arppe, L., Manninen, P., Jylhä, M., Rikkinen, J., Pellikka, P., & Rinne, J. (2023). Carbon dioxide and methane fluxes from mounds of African fungus-growing termites. Biogeosciences, 20(19), 4029–4042. https://doi.org/10.5194/bg-20-4029-2023  

Scheffrahn, R. H., Bourguignon, T., Akama, P. D., Sillam-Dussès, D., & Šobotník, J. (2018). Roisinitermes ebogoensis gen. & sp. n., an outstanding drywood termite with snapping soldiers from Cameroon (Isoptera, Kalotermitidae). ZooKeys, 787, 91–105. https://doi.org/10.3897/zookeys.787.28195 

Seid, M. A., Scheffrahn, R. H., & Niven, J. E. (2008). The rapid mandible strike of a termite soldier. Current Biology, 18(22), R1049–R1050. https://doi.org/10.1016/j.cub.2008.09.033 

Shaw, S. R. (1994). Detection of Airborne Sound by a Cockroach ‘Vibration Detector’: A Possible Missing Link in Insect Auditory Evolution. Journal of Experimental Biology, 193(1), 13-47. https://doi.org/10.1242/jeb.193.1.13   

Singh, K., Muljadi, B. P., Raeini, A. Q., Jost, C., Vandeginste, V., Blunt, M. J., Theraulaz, G., & Degond, P. (2019). The architectural design of smart ventilation and drainage systems in termite nests. Science Advances, 5(3), eaat8520. https://doi.org/10.1126/sciadv.aat8520  

Van Thuyne, J., Irving, J., & Verrecchia, E. P. (2023). A multi-method approach for the investigation of Termite Mound Structures (kalahari basin, Botswana). CATENA, 228, 107158. https://doi.org/10.1016/j.catena.2023.107158 

Yang, L., Sega, M., & Harting, J. (2021). Capillary‐bridge forces between solid particles: Insights from lattice Boltzmann simulations. AIChE Journal, 67(9). https://doi.org/10.1002/aic.17350 

Zachariah, N., Das, A., Murthy, T. G., & Borges, R. M. (2017). Building Mud Castles: A perspective from brick-laying termites. Scientific Reports, 7(1). https://doi.org/10.1038/s41598-017-04295-3 

Zachariah, N., Kandasami, R.K., Das, A., Murthy, T.G., Borges, R.M. (2019). Strength and Cementation in a Termite Mound. In: Reddy, B., Mani, M., Walker, P. (eds) Earthen Dwellings and Structures. Springer Transactions in Civil and Environmental Engineering. Springer, Singapore. https://doi.org/10.1007/978-981-13-5883-8_12 

Zachariah, N., Murthy, T. G., & Borges, R. M. (2020). Moisture alone is sufficient to impart strength but not weathering resistance to termite mound soil. Royal Society Open Science, 7(8), 200485. https://doi.org/10.1098/rsos.200485